Results of an experiment
zoe_althrop
week or so, and I finally found time tonight.
Consider this a mop-up operation on the isochron threads (drum
roll...adjust my scientific beret here).
Preamble (or postramble): The issue finally came down to whether or
not a linear relationship between D/Di and P/Di is retained during a
melt, causing a premelt slope to remain in evidence at time of
solidification. Right, Jon? Was that where we left off? I believe
so.
In order for the linear relationship between D/Di and P/Di to be
erased, melted rock would have to be reduced to a homogenous state.
The question is: what does it take for a liquid to become homogenous?
Pulling up a few sources that deal with homogenization, this is what I
see:
http://www.apecs.com.au/alloyinginsmal/page6.htm
"As the alloy is approaching it's pouring temperature, the metal needs
stirring to be sure of a homogeneous mix."
http://cmk.yz.yamagata-u.ac.jp/Diels.DA-eg.html
"The mixture was then stirred 24 hrs to obtain a homogenous mixture."
http://www.fao.org/WAICENT/FAOINFO/AGRICULT/AGP/AGPP/Pesticid/specs/docs/BLSC.doc
"Shake the mixture vigorously (or use the blender) to produce a
homogeneous suspension."
http://www.baaqmd.gov/permit/handbook/s11c02pd.htm
"A batch mix facility has a plant tower which includes the vibrating
screens ... and mixer....The liquid asphalt is then dropped into the
pug mill where it is wet-mixed until homogeneous."
Apparently, some of the actions needed to produce a homogenous mixture
are:
stirring
shaking
vibrating
wet mixing
None of the above actions are considered regular underground activity
in the earth. Heat is not sufficient to produce a homogenous mixture
(see first link above) and, evidently, pressure will not do it,
either, judging from my latest experiment in my kitchen-lab.
The Experiment:
Chocolate chip morsels were sealed into three saran wrap (thin
plastic) "crystals" in the following ratios of:
Crystal 1 -- 8D, 4Di, 8P (or 8 pieces of brown chocolate, 4 pieces of
white chocolate, and 8 pieces of caramel morsels)
Crystal 2 -- 16D, 4Di, 24P (or 16 pieces of brown chocolate, 4 pieces
of white chocolate, and 24 pieces of caramel morsels)
Crystal 3 -- 40D, 8Di, 32P (or 40 pieces of brown chocolate, 8 pieces
of white chocolate, and 32 pieces of caramel morsels)
Age: 0.5 half lives (if I haven't messed up my math once again)
All three saran-wrapped "crystals" were wrapped tightly in a single
sheet of saran wrap plastic to simulate a cogenetic rock source. The
plastic wrapping represented the regular elements of the mineral, and
the chocolate morsels represented the isotopes that supposedly were
incorporated in same ratios at one time in earth's history. This
"rock source" was placed in an uncovered drinking glass, and the glass
was placed in the microwave oven to test the effect of heat on the
rock source.
Results of the melt:
The chocolate melted but maintained its roughly same position for a
while. At this point, the "magma" was definitely not homogenous. As
the heat continued to be applied, the saran wrap melted and was no
longer observable in the mixture. At this point, homogeneity would be
predicted (don't I sound scientific...I'm impressed), but instead
burning began to take place. The white chocolate turned brown, the
caramel bits turned browner, and the brown chocolate turned black.
When smoke began to pour out of the microwave, I had to snatch the
experiment out and abort the process.
Question: What prevents magma from burning up underground?
The experiment was then repeated, this time in a sealed bottle (magma
chamber). A small glass jar that originally held imported Lindsay
capers, was filled with chocolate morsels of brown, white, and caramel
pieces. Not in any particular ratios this time, since the experiment
now was only to observe the effects of heat on homogenization. The
glass jar was tightly sealed and began to be heated in the microwave.
The process was more frightening this time. Nah, I can't be a
scientist in a lab, I don't think. The threats of possible explosions
hover like the sword of Damocles.
This time, the chocolate melted and gave convincing evidence that heat
under pressure will not cause the different pieces to become
homogenous. But suddenly the sealed cover began to act like the
beginning of a volcanic explosion, and chocolate began to pour out
from the sides of the seal and run down, lava-like, burning and
turning black as it went. I guess this is the point where pressure
builds up underground to where, eventually, a volcanic eruption takes
place?
At any rate, once the glass jar cooled down, it could be observed that
the pattern of white, caramel, and brown chocolate pieces remained in
approximately the same positions they were in when they were placed in
the jar in solid form.
CONCLUSION:
The linear relationship between D/Di and P/Di is retained throughout a
melt, with some fuzzying of the borders (scatter), sufficient to
reduce the apparent age of the rock with each succeeding remelt, but
insufficient to bring the slope back to zero at time of
solidification.
-----------
okay, Jon & Company, I have a lot more chocolate morsels on hand. Are
there any other or better experiments that can be done with
them?...quickly, please, before they disappear down my gullet.
----
zoe
Eric Rowley
> http://www.apecs.com.au/alloyinginsmal/page6.htm
> http://cmk.yz.yamagata-u.ac.jp/Diels.DA-eg.html
>
http://www.fao.org/WAICENT/FAOINFO/AGRICULT/AGP/AGPP/Pesticid/specs/d
ocs/BLSC.doc
> http://www.baaqmd.gov/permit/handbook/s11c02pd.htm
Stirring is, I believe, by definition mechanical or manual using an
implement such as a spoon or mixer and therefore unlikely in a
natural setting.
Shaking and vibration is unlikely to be much of a factor.
But mixing by convection _is_ considered a regular activity in
underground magma chambers.
> Heat is not sufficient to produce a
> homogenous mixture (see first link above)
Have you considered the posibility that heat alone just isn't
fast enough for their purposes?
Any liquid that is heated from below and cooled at the top,
like molten rock in a magma chamber or soup on your stove,
will tend to form convection cells where hot liquid wells
up in the center and (slightly) cooler liquid sinks down
around the edges.
> and, evidently, pressure will not do it, either,
No, I expect pressure is largely irrelevant, it affects the
melting and boiling temperatures but it doesn't do any actual
mixing.
> judging from my latest
> experiment in my kitchen-lab.
> The Experiment:
> Chocolate chip morsels were sealed into three saran wrap (thin
> plastic) "crystals" in the following ratios of:
> Crystal 1 --8D, 4Di, 8P (or 8 pieces of brown chocolate, 4 pieces
> of white chocolate, and 8 pieces of caramel morsels)
> Crystal 2 --16D, 4Di, 24P (or 16 pieces of brown chocolate, 4
> pieces of white chocolate, and 24 pieces of caramel morsels)
> Crystal 3 --40D, 8Di, 32P (or 40 pieces of brown chocolate, 8
> pieces of white chocolate, and 32 pieces of caramel morsels)
> Age: 0.5 half lives (if I haven't messed up my math once again)
> All three saran-wrapped "crystals" were wrapped tightly in a
> single sheet of saran wrap plastic to simulate a cogenetic rock
> source. The plastic wrapping represented the regular elements of
> the mineral, and the chocolate morsels represented the isotopes
> that supposedly were incorporated in same ratios at one time in
> earth's history. This "rock source" was placed in an uncovered
> drinking glass, and the glass was placed in the microwave oven to
> test the effect of heat on the rock source.
> Results of the melt:
> The chocolate melted but maintained its roughly same position for
> a while. At this point, the "magma" was definitely not
> homogenous. As the heat continued to be applied, the saran wrap
> melted and was no longer observable in the mixture. At this
> point, homogeneity would be predicted (don't I sound
> scientific...I'm impressed), but instead burning began to take
> place. The white chocolate turned brown, the caramel bits turned
> browner, and the brown chocolate turned black.
> When smoke began to pour out of the microwave, I had to snatch
> the experiment out and abort the process.
> Question: What prevents magma from burning up underground?
Lack of burnable substances!
Anything burnable would have been converted to oil, or natural
gas way before it reached the melting point of the rest of the rock.
> The experiment was then repeated, this time in a sealed bottle
> (magma chamber). A small glass jar that originally held imported
> Lindsay capers, was filled with chocolate morsels of brown,
> white, and caramel pieces. Not in any particular ratios this
> time, since the experiment now was only to observe the effects of
> heat on homogenization. The glass jar was tightly sealed and
> began to be heated in the microwave.
> The process was more frightening this time. Nah, I can't be a
> scientist in a lab, I don't think. The threats of possible
> explosions hover like the sword of Damocles.
> This time, the chocolate melted and gave convincing evidence that
> heat under pressure will not cause the different pieces to become
> homogenous. But suddenly the sealed cover began to act like the
> beginning of a volcanic explosion, and chocolate began to pour
> out from the sides of the seal and run down, lava-like, burning
> and turning black as it went. I guess this is the point where
> pressure builds up underground to where, eventually, a volcanic
> eruption takes place?
> At any rate, once the glass jar cooled down, it could be observed
> that the pattern of white, caramel, and brown chocolate pieces
> remained in approximately the same positions they were in when
> they were placed in the jar in solid form.
> CONCLUSION:
An ounce or so of chocolate in a microwave oven for a minute or
two doesn't act the same as a million tons of magma in a magma
chamber over millions of years?
Your sample melts may well have been too small to form convection
currents.
Besides convection currents are caused by _uneven_ heating, a
microwave oven is designed to heat things evenly from all sides
(even the inside).
How much time did they have for difusion to occur before
they started to burn or leak?
> The linear relationship between D/Di and P/Di is retained
> throughout a melt,
Where does the original linear relationship come from in your
scenario?
And how is it retained? Are you suggesting that _no_ mixing
takes place in magma chambers or volcanic eruptions?
It seems to me that any mixing at all would destroy the original
relationship and only total mixing would create a new linear
relationship.
> with some fuzzying of the borders (scatter),
> sufficient to reduce the apparent age of the rock with each
> succeeding remelt, but insufficient to bring the slope back to
> zero at time of solidification.
So where are all the measurements of nonzero isochrons from fresh
lava?
> -----------
> okay, Jon & Company, I have a lot more chocolate morsels on hand.
> Are there any other or better experiments that can be done with
> them?...quickly, please, before they disappear down my gullet.
Yes, take a larger quantity (at least a pound) and heat _gently_
in a pot, the best way is in a waterbath (a small pot in a big pot
with a little water in between), so that the chocolate is melted but
not burned.
After a while you should get convection currents, the chocolate
should well up in the center of the pot and sink down at the edges
causing mixing.
Eric
Frank J
Correct me if I'm wrong, but this sounds like a subtle way to suggest
to those who do not understand the dynamics of solidification that
"intelligent design" is necessary to achieve homogeneity. But since
intelligent design is unfalsifiable, we're back to square one.
(snip)
Jon Fleming
wrote:
As you said, those are _SOME_ of the actions that can be used to
produce a homogeneous mixture. They are not _ALL_ of the actions or
effects that can produce a homogeneous mixture.
For example, your list is missing "diffusion (given sufficient time at
high temperature)".
>None of the above actions are considered regular underground activity
>in the earth. Heat is not sufficient to produce a homogenous mixture
>(see first link above
Sorry, the link above does not say that heat is not sufficient to
produce a homogeneous mixture. It implicitly says that heat is not
sufficient to produce a homogeneous mixture _in_ _the_ _amount_ _of_
_time_ _that_ _is_ _allotted_.
If they waited a few years, the allow would be well mixed without
stirring. They just don't want to wait that long.
>and, evidently, pressure will not do it,
>either, judging from my latest experiment in my kitchen-lab.
Pressure doesn't do much for mixing. Diffusion does (given sufficient
time at sufficiently high temperature).
Your experiment is irrelevant. The temperature is too low by an order
of magnitude or two, the time is too short by _many_ _many_ orders of
magnitude, the constituents are inappropriate because they cannot
stand up to temperatures that are realistic for underground magma.
>Question: What prevents magma from burning up underground?
1. Chemistry. The constituents of magma do not burn (combine
somewhat rapidly with oxygen) at the temperatures encountered
underground, even though these temperatures are much much higher than
encountered in your experiment.
2. Even if they did burn at those temperatures, there's not much
oxygen there to combine with.
Put a rock in your microwave or conventional oven. It doesn't burn.
>The experiment was then repeated, this time in a sealed bottle (magma
>chamber). A small glass jar that originally held imported Lindsay
>capers, was filled with chocolate morsels of brown, white, and caramel
>pieces. Not in any particular ratios this time, since the experiment
>now was only to observe the effects of heat on homogenization. The
>glass jar was tightly sealed and began to be heated in the microwave.
>
>
>The process was more frightening this time. Nah, I can't be a
>scientist in a lab, I don't think. The threats of possible explosions
>hover like the sword of Damocles.
>
>This time, the chocolate melted and gave convincing evidence that heat
>under pressure will not cause the different pieces to become
>homogenous.
Again an irrelevant experiment. The constituents are inappropriate
because they can't stand up to the temperatures that magma encounters
underground, the time is many many orders of magnitude too short, the
temperature is way too low.
>Suddenly the sealed cover began to act like the
>beginning of a volcanic explosion, and chocolate began to pour out
>from the sides of the seal and run down, lava-like, burning and
>turning black as it went. I guess this is the point where pressure
>builds up underground to where, eventually, a volcanic eruption takes
>place?
Not really ... the source of the pressure on the magma is due to many
sources. Your "explosion" was probably due to water turning to steam,
which is significant in some but not all volcanic eruptions.
>At any rate, once the glass jar cooled down, it could be observed that
>the pattern of white, caramel, and brown chocolate pieces remained in
>approximately the same positions they were in when they were placed in
>the jar in solid form.
>
>CONCLUSION:
>
>The linear relationship between D/Di and P/Di is retained throughout a
>melt, with some fuzzying of the borders (scatter), sufficient to
>reduce the apparent age of the rock with each succeeding remelt, but
>insufficient to bring the slope back to zero at time of
>solidification.
Your conclusion is not justified.
You _can_ conclude that chocolate is not mixed well by diffusion at
the temperatures you investigated (maybe a few hundred degrees F) over
the times that you investigated (a few minutes, or about 0.00005
years).
You _cannot_ extrapolate this to magma and P/Di or D/Di until you
establish that chocolate is a reasonable model of magma (it isn't),
that the temperatures you used are reasonable temperatures for magma
(they aren't, magma underground is at a thousand degrees F or
thousands of degrees F) and that the time you used is reasonable (it
isn't, you used 0.00005 year and magma is underground for tens of
thousands to millions of years; your times are about 50 BILLION times
too short.
Of course, you could try to use the known mathematical laws of
diffusion to extrapolate your experiment to realistic conditions ...
but that's probably too much of an extrapolation.
We also observe a very strong, almost universal, correlation between
the position of a rock in the geological column and the radiometric
age ... deeper rocks measure as older. How does your proposed
non-mixing account for this observed fact?
>okay, Jon & Company, I have a lot more chocolate morsels on hand. Are
>there any other or better experiments that can be done with
>them?
Nope. Eat 'em.
The only materials suitable for a useful experiment are molten rocks.
You don't have the facilities to run a realistic experiment. You
can't get temperatures high enough, you can't wait long enough, you
don't have the instrumentation required to make accurate measurements
of what happens over shorter times.
--
Replace nospam with group to email
Jon Fleming
Rowley) wrote:
>From: muz...@aol.com (zoe_althrop):
<snip>
>> At any rate, once the glass jar cooled down, it could be observed
>> that the pattern of white, caramel, and brown chocolate pieces
>> remained in approximately the same positions they were in when
>> they were placed in the jar in solid form.
>
>> CONCLUSION:
>
>An ounce or so of chocolate in a microwave oven for a minute or
>two doesn't act the same as a million tons of magma in a magma
>chamber over millions of years?
Why the question mark? I would have used an exclamation point!
{grin}
I doubt that convection currents would thoroughly mix chocolate in a
reasonable time ... but I could be wrong.
However, Zoe's definitely using inappropriate materials and
conditions.
Richard McBane
Mechanically stirring speeds the homogenous mixing process. None of
these processes would be economical if the manufacturer heated the
mixtures for years to allow for mixing by convection. But magma
chambers do heat for many years allowing for mixing.
When your chocolate burns, you are combining hydrocarbons with oxygen to
produce CO2, H2O and ash. Burning is a rapid oxidation process. In a
magma chamber you have elements like silicon, aluminum, potassium. Many
of these elements will oxidize if oxygen is present. But they form
minerals such as quartz SiO2 or other silicates depending upon the
chemistry of the mix.
As Eric stated, heat the chocolate in a double boiler. One caution
however, don't allow the water to boil off or the chocolate will burn as
the temperature increases. Add water to the water bath as the water
evaporates.
Another experiment which I think may have been suggested before:
Put food coloring in water and use it to make ice cubes. Then put the
colored ice cubes in a pitcher of water and see how the food coloring
diffuses through the water as the ice cubes melt. Given a couple of
days you should see the food coloring totally mixed through the pitcher
of water with no mixing or heating.
--
Richard McBane
Seppo Pietikainen
zoe_althrop wrote:
> This is an experiment I've been trying to get around to for the last
> week or so, and I finally found time tonight.
First of all, *DO NOT EAT* you saran wrap+chocolate mixture!!!
(please, tell your next of kin when you're going to do a "scientific
experiment").
>
> Consider this a mop-up operation on the isochron threads (drum
> roll...adjust my scientific beret here).
>
> Preamble (or postramble): The issue finally came down to whether or
> not a linear relationship between D/Di and P/Di is retained during a
> melt, causing a premelt slope to remain in evidence at time of
> solidification. Right, Jon? Was that where we left off? I believe
> so.
>
> In order for the linear relationship between D/Di and P/Di to be
> erased, melted rock would have to be reduced to a homogenous state.
> The question is: what does it take for a liquid to become homogenous?
> Pulling up a few sources that deal with homogenization, this is what I
> see:
>
> http://www.apecs.com.au/alloyinginsmal/page6.htm
>
There *are* some time constraints associated with industrial processes.
Hence, every now and then things go awfully wrong, especially when large
amount of reagents are mixed... Volcanoes behave "a little"
different.(I'd prefer studying volcanoes to stratification problems in
chemical industry...[I do neither, but I've been involved with the
latter, in a fashion, at least])
> "As the alloy is approaching it's pouring temperature, the metal needs
> stirring to be sure of a homogeneous mix."
>
> http://cmk.yz.yamagata-u.ac.jp/Diels.DA-eg.html
>
> "The mixture was then stirred 24 hrs to obtain a homogenous mixture."
>
> http://www.fao.org/WAICENT/FAOINFO/AGRICULT/AGP/AGPP/Pesticid/specs/docs/BLSC.doc
>
> "Shake the mixture vigorously (or use the blender) to produce a
> homogeneous suspension."
>
> http://www.baaqmd.gov/permit/handbook/s11c02pd.htm
>
> "A batch mix facility has a plant tower which includes the vibrating
> screens ... and mixer....The liquid asphalt is then dropped into the
> pug mill where it is wet-mixed until homogeneous."
>
> Apparently, some of the actions needed to produce a homogenous mixture
> are:
>
> stirring
> shaking
> vibrating
> wet mixing
>
> None of the above actions are considered regular underground activity
> in the earth. Heat is not sufficient to produce a homogenous mixture
> (see first link above) and, evidently, pressure will not do it,
> either, judging from my latest experiment in my kitchen-lab.
Add convection to the list and you might get it somewhat closer.
May I suggest a far simpler (and safer for minors and pets...) experiment:
1) Take a kettle (greater than 1 litre capacity [about 1/4 Gallons])
2) Fill it with approximately 1 litre (~ 1 pint) of cold water
3) Add about 30 grams (~1 oz) of ordinary kitchen salt.
4) Heat it to, say, 70 degrees C (~160F), continue for 5 minutes
without stirring.
5) Report non-uniformities in the mixture.
Meanwhile, you might want to do a search of "thermal convection" in the
web or your local library.
>
> The Experiment:
>
> Chocolate chip morsels were sealed into three saran wrap (thin
> plastic) "crystals" in the following ratios of:
>
> Crystal 1 -- 8D, 4Di, 8P (or 8 pieces of brown chocolate, 4 pieces of
> white chocolate, and 8 pieces of caramel morsels)
>
> Crystal 2 -- 16D, 4Di, 24P (or 16 pieces of brown chocolate, 4 pieces
> of white chocolate, and 24 pieces of caramel morsels)
>
> Crystal 3 -- 40D, 8Di, 32P (or 40 pieces of brown chocolate, 8 pieces
> of white chocolate, and 32 pieces of caramel morsels)
>
> Age: 0.5 half lives (if I haven't messed up my math once again)
>
> All three saran-wrapped "crystals" were wrapped tightly in a single
> sheet of saran wrap plastic to simulate a cogenetic rock source. The
> plastic wrapping represented the regular elements of the mineral, and
> the chocolate morsels represented the isotopes that supposedly were
> incorporated in same ratios at one time in earth's history. This
> "rock source" was placed in an uncovered drinking glass, and the glass
> was placed in the microwave oven to test the effect of heat on the
> rock source.
I don't even *begin* to understand your interpretation of half-life in
the above. Suffice it to say, that crystals are not wrapped to "saran
wrap" (well, *sometimes* they are encased, in a fashion...).
>
> Results of the melt:
>
> The chocolate melted but maintained its roughly same position for a
> while. At this point, the "magma" was definitely not homogenous. As
> the heat continued to be applied, the saran wrap melted and was no
> longer observable in the mixture. At this point, homogeneity would be
> predicted (don't I sound scientific...I'm impressed), but instead
> burning began to take place. The white chocolate turned brown, the
> caramel bits turned browner, and the brown chocolate turned black.
>
> When smoke began to pour out of the microwave, I had to snatch the
> experiment out and abort the process.
>
> Question: What prevents magma from burning up underground?
Ummm. Maybe you haven't heard, but *free* oxygen is not terribly
abundant underground.
Another thing is, that there *does* happen chemical oxidization (aka.
"burning") in such environments. To simplify things (a lot); it means
exchanging oxygen between oxygen rich and oxygen poor compounds in such
environments. The mechanisms are *very* well known and keep the titanium
and steel industries afloat (not to speak of computer industry).
>
> The experiment was then repeated, this time in a sealed bottle (magma
> chamber). A small glass jar that originally held imported Lindsay
> capers, was filled with chocolate morsels of brown, white, and caramel
> pieces. Not in any particular ratios this time, since the experiment
> now was only to observe the effects of heat on homogenization. The
> glass jar was tightly sealed and began to be heated in the microwave.
>
>
> The process was more frightening this time. Nah, I can't be a
> scientist in a lab, I don't think. The threats of possible explosions
> hover like the sword of Damocles.
Yep, *do* be careful with your experiments (some people even go as far
to say: "Don't do this at home!") !
>
> This time, the chocolate melted and gave convincing evidence that heat
> under pressure will not cause the different pieces to become
> homogenous. But suddenly the sealed cover began to act like the
> beginning of a volcanic explosion, and chocolate began to pour out
> from the sides of the seal and run down, lava-like, burning and
> turning black as it went. I guess this is the point where pressure
> builds up underground to where, eventually, a volcanic eruption takes
> place?
I think you made a rather decent replica of a volcano eruption in your
kitchen. The outpouring of chocolate in your experiment was due to
expanding gas bubbles (steam) in your mix. The burning and color change
was due to the energy input of your microwave oven.
You may not have noticed, but it was the *gas* pressure within your
weird (and potentially dangerous) experiment that caused all this
outpouring of "lava-like" substance (creating further uniformism of
within your treacle).
>
> At any rate, once the glass jar cooled down, it could be observed that
> the pattern of white, caramel, and brown chocolate pieces remained in
> approximately the same positions they were in when they were placed in
> the jar in solid form.
>
> CONCLUSION:
>
> The linear relationship between D/Di and P/Di is retained throughout a
> melt, with some fuzzying of the borders (scatter), sufficient to
> reduce the apparent age of the rock with each succeeding remelt, but
> insufficient to bring the slope back to zero at time of
> solidification.
Another CONCLUSION: You still continue to be *totally* ignorant of
basic physics.
>
> -----------
>
> okay, Jon & Company, I have a lot more chocolate morsels on hand. Are
> there any other or better experiments that can be done with
> them?...quickly, please, before they disappear down my gullet.
>
> ----
> zoe
>
Seppo P.
P. Venkman
> week or so, and I finally found time tonight.
>
> Consider this a mop-up operation on the isochron threads (drum
> roll...adjust my scientific beret here).
>
> Preamble (or postramble): The issue finally came down to whether or
> not a linear relationship between D/Di and P/Di is retained during a
> melt, causing a premelt slope to remain in evidence at time of
> solidification. Right, Jon? Was that where we left off? I believe
> so.
Actually I have a question about this. You're questioning whether a
linear relationship between D/Di and P/Di can be retained during a
melt. That pre-supposes a linear relationship exists pre-melt. Take
that back to its logical conclusion and that means a linear
relationship had to exist at the formation of the earth. Is there any
reason to believe this is true?
<SNIP some references>
> Apparently, some of the actions needed to produce a homogenous mixture
> are:
>
> stirring
> shaking
> vibrating
> wet mixing
>
> None of the above actions are considered regular underground activity
> in the earth. Heat is not sufficient to produce a homogenous mixture
> (see first link above) and, evidently, pressure will not do it,
> either, judging from my latest experiment in my kitchen-lab.
Well, it depends. I know that uneven heating can lead to convection
currents which can cause stirring. For instance, salt gets pretty
evenly mixed in boiling water.
The other element missing from the references you gave is time. Think
about stirring for a moment. Obviously just swishing a spoon through
cake batter once isn't enough to mix the batter evenly, you have to
keep mixing for some length of time. The same thing is true with
convection, it takes time. 24 hours might not be enough when creating
metal alloys, but maybe 24 million years would be enough.
<SNIP some experiments>
-----------
>
> okay, Jon & Company, I have a lot more chocolate morsels on hand. Are
> there any other or better experiments that can be done with
> them?...quickly, please, before they disappear down my gullet.
>
> ----
> zoe
I don't know that I'm qualified to speak about this, but maybe putting
your 'crystals' in a double-boiler over relatively low heat for a few
weeks? I realize that may not be practical. Maybe try it for 2
hours, then try another batch for 6 hours, and another for 12 and see
if the homogenity increases as time increases?
I do have one question for you. How do you account for the fact that
in new flows that have been measured we do see that D/Di is the same
for all samples?
Jon Fleming
<s.pietika...@kolumbus.fi> wrote:
>Zoe,
>
>zoe_althrop wrote:
<snip>
>> Chocolate chip morsels were sealed into three saran wrap (thin
>> plastic) "crystals" in the following ratios of:
>>
>> Crystal 1 -- 8D, 4Di, 8P (or 8 pieces of brown chocolate, 4 pieces of
>> white chocolate, and 8 pieces of caramel morsels)
>>
>> Crystal 2 -- 16D, 4Di, 24P (or 16 pieces of brown chocolate, 4 pieces
>> of white chocolate, and 24 pieces of caramel morsels)
>>
>> Crystal 3 -- 40D, 8Di, 32P (or 40 pieces of brown chocolate, 8 pieces
>> of white chocolate, and 32 pieces of caramel morsels)
>>
>> Age: 0.5 half lives (if I haven't messed up my math once again)
>>
>> All three saran-wrapped "crystals" were wrapped tightly in a single
>> sheet of saran wrap plastic to simulate a cogenetic rock source. The
>> plastic wrapping represented the regular elements of the mineral, and
>> the chocolate morsels represented the isotopes that supposedly were
>> incorporated in same ratios at one time in earth's history. This
>> "rock source" was placed in an uncovered drinking glass, and the glass
>> was placed in the microwave oven to test the effect of heat on the
>> rock source.
>
> I don't even *begin* to understand your interpretation of half-life in
>the above. Suffice it to say, that crystals are not wrapped to "saran
>wrap" (well, *sometimes* they are encased, in a fashion...).
It's not really important, but ..
She's attempting to model a rock containing three crystals, in which
Parent atoms (or some number thereof) are represented by caramel
morsels, Daughter atoms (to which Parent atoms decay) are represented
by brown chocolate pieces, Daughter Isotope (to which Parent atoms do
_not_ decay) atoms are represented by white chocolate morsels, and the
Saran wrap is something like the forces holding each crystal together.
Of course, the first step in the experiment should have been removing
the Saran wrap to simulate melting, but there are so many other major
problems with the experiment that this is a minor issue.
She attempted to choose mixtures of P, D, and Di that would indicate
that the rock is 0.5 half-lives old if those three crystals were
analyzed by an isochron method. Of course, she _did_ mess up the
math; the three crystals she specified have the following ratios:
P/Di D/Di
2 2
6 4
4 5
which don't line up on a straight line on an isochron plot ... but the
numbers do suggest that she intended the P/Di ratio of the third
crystal to be 8 (64 caramel morsels instead of 32). This seems to be
the minimum change that yields a straight line using integer numbers
of P, D, and Di and a slope of 0.5 (which Zoe keeps confusing with the
age in half-lives).
She never seems to remember (despite an incredible number of posted
explanations and reminders and formulas and tables) that the slope has
to be plugged into another formula to give the age in half-lives (a
slope of 0.5 is an age of 0.585 half lives).
<snip>
John Stockwell
(snip)
>
>Apparently, some of the actions needed to produce a homogenous mixture
>are:
>
>stirring
>shaking
>vibrating
>wet mixing
>
>None of the above actions are considered regular underground activity
>in the earth. Heat is not sufficient to produce a homogenous mixture
>(see first link above) and, evidently, pressure will not do it,
>either, judging from my latest experiment in my kitchen-lab.
You forgot "diffusion". Look up something called "Brownian motion".
John Stockwell | jo...@dix.Mines.EDU
Center for Wave Phenomena (The Home of Seismic Un*x)
Colorado School of Mines
Golden, CO 80401 | http://www.cwp.mines.edu/cwpcodes
voice: (303) 273-3049
Our book:
Norman Bleistein, Jack K. Cohen, John W. Stockwell Jr., [2001],
Mathematics of multidimensional seismic imaging, migration, and inversion,
(Interdisciplinary Applied Mathematics, V. 13.), Springer-Verlag, New York.
P. Venkman
> okay, Jon & Company, I have a lot more chocolate morsels on hand. Are
> there any other or better experiments that can be done with
> them?...quickly, please, before they disappear down my gullet.
>
OK, here's one for you. Get a large glass and fill it with water.
Add 3 drops of red ink to the water on one side of the glass, then add
three drops of blue ink to the water on the opposite side of the
glass. Cover the glass with some plastic wrap (just to avoid
evaporation problems) and leave it for a week.
Now pour the contents of the glass into an ice cube tray, stick the
tray in the freezer, and make ice cubes. After they're frozen, pull
them out. Are the ice cubes all the same color, or do you have some
mostly red and some mostly blue cubes?
What do you conclude from this experiment?
R. Baldwin
news:3ecd8ced....@news-server.cfl.rr.com...
Simple pressure can cause a liquid or colloid to extrude through a
medium or a constricted opening, causing mixing.
[snip]
Mike Dworetsky
"zoe_althrop" <muz...@aol.com> wrote in message
news:3ecd8ced....@news-server.cfl.rr.com...
Zoe, I think it would be far safer for you to enrol in mathematics and
geology classes at your local university to resolve issues regarding
isochrons, than to risk continuing to experiment, and perhaps burning down
your house.
--
Mike Dworetsky
(Remove "pants" spamblock to send e-mail)
Forest Ghost
Just an idea if you want to try your experiments again- use a double
boiler set up instead of a microwave. This is easily done by placing a
preferably metal bowl on top of a pot of water and letting it boil.
You'll find that you can melt the chocolate and probably saran wrap this
way without burning it, and if your volcano jar does overflow again,
it'll be a lot easier to clean out the bowl than your microwave.
--
-Forest Ghost, official baker of T.O.
Bigdakine
>From: pven...@hotmail.com (P. Venkman)
>Date: 5/23/03 10:56 AM Hawaiian Standard Time
>Message-id: <791aa995.03052...@posting.google.com>
Zoe would conclude intelligent design.
Stuart
Dr. Stuart A. Weinstein
Ewa Beach Institute of Tectonics
"To err is human, but to really foul things up
requires a creationist"
Andrew Arensburger
> Simple pressure can cause a liquid or colloid to extrude through a
> medium or a constricted opening, causing mixing.
How, then, do you explain striped toothpaste?
I'm only being semi-facetious: obviously the toothpaste is not
thoroughly mixed after being forced by pressure through a constricted
opening. Either your statement is untrue, or else it doesn't apply to
toothpaste. If the latter, why?
--
Andrew Arensburger, Systems guy University of Maryland
arensb.no-...@glue.umd.edu Office of Information Technology
Insufficient mug space. Sysadmin failure.
zoe_althrop
Rowley) wrote:
snip>
>But mixing by convection _is_ considered a regular activity in
>underground magma chambers.
true, but the results of this kind of mixing won't do the job. At
least not according to this site:
http://www.gps.caltech.edu/~gurnis/Movies/Science_Captions/stirring.html
"Even after a few billion years of stirring by mantle convection, the
fluid is still quite heterogneous."
>> Heat is not sufficient to produce a
>> homogenous mixture (see first link above)
>
>Have you considered the posibility that heat alone just isn't
>fast enough for their purposes?
prolonged heat results in burning, dehydration, or pressure buildup.
It does not contribute to homogeneity, not even by convection currents
within magma. At least this is what I am learning so far.
>Any liquid that is heated from below and cooled at the top,
>like molten rock in a magma chamber or soup on your stove,
>will tend to form convection cells where hot liquid wells
>up in the center and (slightly) cooler liquid sinks down
>around the edges.
boiling, you mean? A magma chamber and soup on a stove carry entirely
different variables, so I don't think it is fair to compare the
boiling of soup in an open pot on a stove to the pressurized contents
of a magma chamber. That is why I have, in the past, used a pressure
cooker for comparison, because that would have more in common with a
magma chamber.
snip>
>> Question: What prevents magma from burning up underground?
>
>Lack of burnable substances!
>Anything burnable would have been converted to oil, or natural
>gas way before it reached the melting point of the rest of the rock.
that's interesting. Let me follow up your reply with some research on
your pointers. Maybe you're right, maybe not.
snip>
>An ounce or so of chocolate in a microwave oven for a minute or
>two doesn't act the same as a million tons of magma in a magma
>chamber over millions of years?
the parallel is a relative one. Once the substance (rock or
chocolate) reaches the point of melting (5000 degrees or 250 degrees),
the characteristics and behavior of the liquefied rock/chocolate
should be the same. Boiling is boiling, regardless of what
temperature it takes to bring a substance to a boil. Same for
melting.
>Your sample melts may well have been too small to form convection
>currents.
if by "convection currents" you mean small bubbles percolating up
through the melted chocolate, then I did see convection currents. The
chocolate, however, remained pretty much in place, with a probably
somewhat wider zone.
>Besides convection currents are caused by _uneven_ heating, a
>microwave oven is designed to heat things evenly from all sides
>(even the inside).
actually, no, the microwave heats from outside in. If you take the
item out too soon, the edges will be hotter than the middle.
>How much time did they have for difusion to occur before
>they started to burn or leak?
not much, granted that.
>> The linear relationship between D/Di and P/Di is retained
>> throughout a melt,
>
>Where does the original linear relationship come from in your
>scenario?
the original linear relationship of D/Di to P/Di would come from an
original formation in which there was no oldD present. Since D
originates from P, there is no reason to imagine that there was a time
when D originated all by itself, without the aid of P. Therefore, the
moment that P begins to decay to D, the linear relationship would be
established, and it is this linear relationship that becomes more and
more fuzzied by remelts. The more remelts, the younger the rock
appears to be. The less remelts, the older the rock shows up to be.
>And how is it retained? Are you suggesting that _no_ mixing
>takes place in magma chambers or volcanic eruptions?
the linear relationship is retained because very little mixing occurs
in magma chambers. What serves to fuzzy the relationship is continual
remelts, leaching, contamination, and other environmental variables
that interfere with the original linear relationship. Given sufficient
time, I'm betting that there won't be any rocks left that would
produce an acceptable isochron.
>It seems to me that any mixing at all would destroy the original
>relationship and only total mixing would create a new linear
>relationship.
if the mixing is negligible, the original relationship only becomes
fuzzied (scatter). It is only total mixing that would create a new
linear relationship, but I am submitting that total mixing is not a
reality.
>> with some fuzzying of the borders (scatter),
>> sufficient to reduce the apparent age of the rock with each
>> succeeding remelt, but insufficient to bring the slope back to
>> zero at time of solidification.
>
>So where are all the measurements of nonzero isochrons from fresh
>lava?
if freshly extruded lava is found to be nonzero via Rb/Sr isochron, it
would mean that the premelt linear relationship is still visible. But
why would anybody use Rb/Sr to date new lava? It is useless to use
such long-lived isotopes (48 billion years) to check for recent age.
>> okay, Jon & Company, I have a lot more chocolate morsels on hand.
>> Are there any other or better experiments that can be done with
>> them?...quickly, please, before they disappear down my gullet.
>
>Yes, take a larger quantity (at least a pound) and heat _gently_
>in a pot, the best way is in a waterbath (a small pot in a big pot
>with a little water in between), so that the chocolate is melted but
>not burned.
>
>After a while you should get convection currents, the chocolate
>should well up in the center of the pot and sink down at the edges
>causing mixing.
thanks for the suggestion, Eric, but this experiment would not match
up the variables between melted chocolate and magma. One is in an
open pot where boiling can take place. The other is in a pressurized
magma chamber where boiling does not take place.
----
zoe
zoe_althrop
wrote:
snip>
>Correct me if I'm wrong, but this sounds like a subtle way to suggest
>to those who do not understand the dynamics of solidification that
>"intelligent design" is necessary to achieve homogeneity. But since
>intelligent design is unfalsifiable, we're back to square one.
it would be helpful if you could see past your one all-consuming
prejudice, Frank; i.e., that if a creationist raises any question at
all, it is because the one objective of such a question is to say
"Goddidit." Your preconceptions are getting in the way of
communication. For instance, see what's happened here. You're back
to square one, talking to yourself, while I'm out here on square 1+X,
trying to talk to you. How can we have a discussion when you insist
on boomeranging like this?
----
zoe
zoe_althrop
<jo...@fleming-nospam.com> wrote:
snip>
>As you said, those are _SOME_ of the actions that can be used to
>produce a homogeneous mixture. They are not _ALL_ of the actions or
>effects that can produce a homogeneous mixture.
>
>For example, your list is missing "diffusion (given sufficient time at
>high temperature)".
viscosity (low versus high) is a critical factor in diffusion, I
think. Why would you compare a watery solution to a magmatic
solution? See:
http://www.uic.edu/classes/phys/phys450/MARKO/N004.html
"Water, and solutions which are mostly water, have a viscosity close
to hwater = 0.01 Poise or 0.001 Pa·sec."
or better yet, this link:
http://www.brookes.ac.uk/geology/sedstruc/viscosit/vis.htm
"Material Poise
Water 1
Engine Oil 10
Plasticine 1000
Asphalt 10,000
Basalts (lava) 100 - 1000
Granites (magma) 1,000,000 - 1,000,000,000,000"
diffusion in water is going to be at a far higher rate than diffusion
in molten rock. Indeed, I would expect that the concentration in
magma is such that there is no need for exchange of particles from
areas of higher concentration to areas of lower concentration.
Density matters, as appears to be the case in this link:
http://chemed.chem.purdue.edu/analyticalreview/chromatography/ratetheory.htm
"2) flow velocities will vary throughout a packed column and cause
zone spreading,"
You need to have two or more different species in a solution in order
for diffusion to occur, right? What are the different species in
molten rock? I mean, diffusion of salt or sugar molecules into water
would result in eventual homogeneity -- well, no, not even then,
unless you stir the solution; otherwise, the crystals would settle out
instead of dissolve. But in more dense materials, the rate of
diffusion should slow down. And in the case of a magma chamber, all
you have is liquid rock, so you won't have magma dissolving into water
as you would with salt/sugar/red dye in water.
>>None of the above actions are considered regular underground activity
>>in the earth. Heat is not sufficient to produce a homogenous mixture
>>(see first link above
>
>Sorry, the link above does not say that heat is not sufficient to
>produce a homogeneous mixture. It implicitly says that heat is not
>sufficient to produce a homogeneous mixture _in_ _the_ _amount_ _of_
>_time_ _that_ _is_ _allotted_.
if you are invoking billions of years for homogeneity to take place,
then your hypothesis of magma being homogenous becomes an
unfalsifiable hypothesis.
>If they waited a few years, the allow would be well mixed without
>stirring. They just don't want to wait that long.
please for support for the idea that the only reason for mixing is
because it would take too long otherwise. And also give support for
the idea that the alloy would become well mixed if allowed to sit for
a few years.
snip>
>Your experiment is irrelevant. The temperature is too low by an order
>of magnitude or two, the time is too short by _many_ _many_ orders of
>magnitude, the constituents are inappropriate because they cannot
>stand up to temperatures that are realistic for underground magma.
are you saying that one cannot do an experiment on a lower level and
extrapolate the results out to a higher level -- that the laws of
physics do change with time?
>>Question: What prevents magma from burning up underground?
>
>1. Chemistry. The constituents of magma do not burn (combine
>somewhat rapidly with oxygen) at the temperatures encountered
>underground, even though these temperatures are much much higher than
>encountered in your experiment.
okay, lack of oxygen.
>
>2. Even if they did burn at those temperatures, there's not much
>oxygen there to combine with.
>
>Put a rock in your microwave or conventional oven. It doesn't burn.
it's all relative. The characteristics of melting should be the same
once the rock melts. It shouldn't matter that it takes a higher
temperature to melt a rock. What I am interested in is what happens
when the material (be it rock or chocolate) finally does melt.
snip>
>Again an irrelevant experiment. The constituents are inappropriate
>because they can't stand up to the temperatures that magma encounters
>underground, the time is many many orders of magnitude too short, the
>temperature is way too low.
again, Jon, are you saying that the characteristics and behavior of a
melt differs strictly because it took Item A longer to melt than Item
B and because it took more heat to melt Item A than Item B?
>>Suddenly the sealed cover began to act like the
>>beginning of a volcanic explosion, and chocolate began to pour out
>>from the sides of the seal and run down, lava-like, burning and
>>turning black as it went. I guess this is the point where pressure
>>builds up underground to where, eventually, a volcanic eruption takes
>>place?
>
>Not really ... the source of the pressure on the magma is due to many
>sources. Your "explosion" was probably due to water turning to steam,
>which is significant in some but not all volcanic eruptions.
water in chocolate? Chocolate is a mixture of cocoa paste, cocoa
butter, and sugar. What water?
snip>
>The only materials suitable for a useful experiment are molten rocks.
>You don't have the facilities to run a realistic experiment. You
>can't get temperatures high enough, you can't wait long enough, you
>don't have the instrumentation required to make accurate measurements
>of what happens over shorter times.
well, so much for extrapolation in science; it is a useless tool -- as
far as Jon is concerned, anyway.
----
zoe
zoe_althrop
<rmc...@attbi.com> wrote:
snip>
zoe asked:
>> Question: What prevents magma from burning up underground?
>
>When your chocolate burns, you are combining hydrocarbons with oxygen to
>produce CO2, H2O and ash. Burning is a rapid oxidation process. In a
>magma chamber you have elements like silicon, aluminum, potassium. Many
>of these elements will oxidize if oxygen is present. But they form
>minerals such as quartz SiO2 or other silicates depending upon the
>chemistry of the mix.
so after a short while, there would be no oxygen left in a magma
chamber, right?
snip>
>As Eric stated, heat the chocolate in a double boiler. One caution
>however, don't allow the water to boil off or the chocolate will burn as
>the temperature increases. Add water to the water bath as the water
>evaporates.
how would you suggest that I eliminate oxygen from this experiment so
that it would more closely resemble the conditions in a magma chamber?
>Another experiment which I think may have been suggested before:
>Put food coloring in water and use it to make ice cubes. Then put the
>colored ice cubes in a pitcher of water and see how the food coloring
>diffuses through the water as the ice cubes melt. Given a couple of
>days you should see the food coloring totally mixed through the pitcher
>of water with no mixing or heating.
the viscosity of water and magma are so different that I don't think
this experiment will throw much light on mixing in a magma chamber,
would it?
----
zoe
zoe_althrop
<s.pietika...@kolumbus.fi> wrote:
snip>
> First of all, *DO NOT EAT* you saran wrap+chocolate mixture!!!
> (please, tell your next of kin when you're going to do a "scientific
>experiment").
good advice, Seppo. I think I have sufficient chocolate morsels on
hand that I won't be driven to eat the plastic+chocolate mixture.
snip>
> There *are* some time constraints associated with industrial processes.
>Hence, every now and then things go awfully wrong, especially when large
>amount of reagents are mixed... Volcanoes behave "a little"
>different.(I'd prefer studying volcanoes to stratification problems in
>chemical industry...[I do neither, but I've been involved with the
>latter, in a fashion, at least])
in what fashion?
snip>
> Add convection to the list and you might get it somewhat closer.
>
> May I suggest a far simpler (and safer for minors and pets...) experiment:
> 1) Take a kettle (greater than 1 litre capacity [about 1/4 Gallons])
> 2) Fill it with approximately 1 litre (~ 1 pint) of cold water
> 3) Add about 30 grams (~1 oz) of ordinary kitchen salt.
> 4) Heat it to, say, 70 degrees C (~160F), continue for 5 minutes
>without stirring.
> 5) Report non-uniformities in the mixture.
convection in water that is boiled in an open pot, with access to all
the oxygen available, is much more rapid than convection in a far more
dense material in a non-oxygenated chamber, so this experiment would
not be much of a commentary on the behavior of magma, would it?
>Meanwhile, you might want to do a search of "thermal convection" in the
>web or your local library.
if by "thermal" convection you mean heat convection, then, yes, I've
been doing a bit of reading up on it.
snip>
>> Question: What prevents magma from burning up underground?
>
> Ummm. Maybe you haven't heard, but *free* oxygen is not terribly
>abundant underground.
yes, I'm hearing about that in this thread. Which would be another
factor that would rule out an open-pot comparison.
snip>
----
zoe
zoe_althrop
Venkman) wrote:
snip>
>Actually I have a question about this. You're questioning whether a
>linear relationship between D/Di and P/Di can be retained during a
>melt. That pre-supposes a linear relationship exists pre-melt. Take
>that back to its logical conclusion and that means a linear
>relationship had to exist at the formation of the earth. Is there any
>reason to believe this is true?
the reason to believe this is true is that the laws of physics is the
same today as it always was. And since the only observation of these
laws that holds true today is the one that say that D originates from
decay of P, then there could be no original oldD at the first
formation of the earth -- unless you want to posit some miraculous
appearance of oldD that did not originate from P. Occam would not
approve of that, now would he?
><SNIP some references>
>
>> Apparently, some of the actions needed to produce a homogenous mixture
>> are:
>>
>> stirring
>> shaking
>> vibrating
>> wet mixing
>>
>> None of the above actions are considered regular underground activity
>> in the earth. Heat is not sufficient to produce a homogenous mixture
>> (see first link above) and, evidently, pressure will not do it,
>> either, judging from my latest experiment in my kitchen-lab.
>
>Well, it depends. I know that uneven heating can lead to convection
>currents which can cause stirring. For instance, salt gets pretty
>evenly mixed in boiling water.
again, you're equating the viscosity of boiling water with the
viscosity of magma.
>The other element missing from the references you gave is time. Think
>about stirring for a moment. Obviously just swishing a spoon through
>cake batter once isn't enough to mix the batter evenly, you have to
>keep mixing for some length of time. The same thing is true with
>convection, it takes time. 24 hours might not be enough when creating
>metal alloys, but maybe 24 million years would be enough.
I don't think that my cake batter would be any better mixed if I left
the ingredients to sit for several months or years. Ask any woman in
her kitchen if she thought her cake batter would become nicely mixed
if she let it sit for a few months or years, and she would question
your sanity.
I don't think you should invoke time as the solution. To do so only
makes your theory unfalsifiable, since millions of years is not
available to test your theory.
snip>
>I don't know that I'm qualified to speak about this, but maybe putting
>your 'crystals' in a double-boiler over relatively low heat for a few
>weeks? I realize that may not be practical. Maybe try it for 2
>hours, then try another batch for 6 hours, and another for 12 and see
>if the homogenity increases as time increases?
I would need to eliminate the presence of oxygen, too. Any
suggestions?
>I do have one question for you. How do you account for the fact that
>in new flows that have been measured we do see that D/Di is the same
>for all samples?
using the Rb/Sr measure?
----
zoe
zoe_althrop
<jo...@dix.Mines.EDU> wrote:
snip>
>You forgot "diffusion". Look up something called "Brownian motion".
John, I did look it up. It seems, from what little I've covered so
far, that diffusion rates depend on temperature of the suspension,
viscosity of the solvent, and radius of the colloid particle.
Therefore, to loosely apply diffusion of particles in water to
diffusion of isotopes in magma would be like comparing apples to
oranges.
----
zoe
zoe_althrop
<res0...@nozirevBACKWARDS.net> wrote:
snip>
>Simple pressure can cause a liquid or colloid to extrude through a
>medium or a constricted opening, causing mixing.
where is there place for extrusion in a magma chamber? The only time
extrusion would occur would be in an eruption (either internally or
externally). Would that eruption be sufficient to cause
homogenization? I'm thinking, no.
----
zoe
zoe_althrop
without even conducting this experiment, I can conclude that it does
not in any way simulate magma mixing.
----
zoe
zoe_althrop
appreciate the suggestion, Forest. Can you suggest a way to add the
variable of "no oxygen" to this experiment? I thought a pressure
cooker was getting closer to resembling a magma chamber, but maybe
there is still oxygen to deal with in there.
----
zoe
zoe_althrop
<plati...@pants.btinternet.com> wrote:
snip>
>Zoe, I think it would be far safer for you to enrol in mathematics and
>geology classes at your local university to resolve issues regarding
>isochrons, than to risk continuing to experiment, and perhaps burning down
>your house.
you are probably right, Mike. Except how else to obtain fresh data if
not through actual experimentation?
----
zoe
Richard McBane
>
> On Fri, 23 May 2003 14:08:27 +0000 (UTC), Richard McBane
> <rmc...@attbi.com> wrote:
>
> snip>
>
> zoe asked:
>
> >> Question: What prevents magma from burning up underground?
> >
> >When your chocolate burns, you are combining hydrocarbons with oxygen to
> >produce CO2, H2O and ash. Burning is a rapid oxidation process. In a
> >magma chamber you have elements like silicon, aluminum, potassium. Many
> >of these elements will oxidize if oxygen is present. But they form
> >minerals such as quartz SiO2 or other silicates depending upon the
> >chemistry of the mix.
>
> so after a short while, there would be no oxygen left in a magma
> chamber, right?
would be combined with other elements.
> <snip>
>
> >As Eric stated, heat the chocolate in a double boiler. One caution
> >however, don't allow the water to boil off or the chocolate will burn as
> >the temperature increases. Add water to the water bath as the water
> >evaporates.
>
> how would you suggest that I eliminate oxygen from this experiment so
> that it would more closely resemble the conditions in a magma chamber?
I'm not certain that you can eliminate the oxygen using kitchen
utensils.
One way might be to put a lit candle in the pressure cooker with the
chocolate before heating. If the candle is large enough and if the
pressure cooker is sealed well enough, the candle would consume all the
oxygen in the pressure cooker. Then you could heat up the chocolate,
but the candle wax would melt as well.
If you are willing to drill some holes in the pressure cooker and
install appropriate fittings, you might be able to flood the pressure
cooker with CO2 from a CO2 fire extinguisher. You would need two
fittings with valves. Connect the fire extinguisher to one valve. Open
both valves, then turn on the extinguisher, let it flow for a while,
then close both valve, the one not connected to the extinguisher first.
This would remove most of the oxygen, but probably not all.
These approaches aren't really practical at home, which is why the
double boiler approach was suggested. The chocolate will melt in a
double boiler but the water prevents chocolate from reaching a
temperature where it will start to burn. I thought that a double boiler
was the standard approach for melting chocolate in cooking.
> >Another experiment which I think may have been suggested before:
> >Put food coloring in water and use it to make ice cubes. Then put the
> >colored ice cubes in a pitcher of water and see how the food coloring
> >diffuses through the water as the ice cubes melt. Given a couple of
> >days you should see the food coloring totally mixed through the pitcher
> >of water with no mixing or heating.
>
> the viscosity of water and magma are so different that I don't think
> this experiment will throw much light on mixing in a magma chamber,
> would it?
>
> ----
> zoe
Well there are a lot of other differences in your experiment than
viscosity.
--
Richard McBane
Richard McBane
With a double boiler you don't need to worry about the oxygen. The
water in the boiler keeps the chocolate from reaching its ignition
point. So the chocolate doesn't burn as long as there is water in the
double boiler.
--
Richard McBane
Jon Fleming
wrote:
>On Fri, 23 May 2003 13:27:48 +0000 (UTC), Jon Fleming
><jo...@fleming-nospam.com> wrote:
>
>snip>
>
>>As you said, those are _SOME_ of the actions that can be used to
>>produce a homogeneous mixture. They are not _ALL_ of the actions or
>>effects that can produce a homogeneous mixture.
>>
>>For example, your list is missing "diffusion (given sufficient time at
>>high temperature)".
>
>viscosity (low versus high) is a critical factor in diffusion, I
>think.
You are wrong. Viscosity is not a critical factor (in diffusion of
atoms and molecules). Density is not a critical factor. The major
critical factors are temperature, concentration gradient, and mobility
within the material. Although viscosity slightly affects mobility
within the material, the material form (solid, liquid, or gas) is the
major effect on mobility.
>Why would you compare a watery solution to a magmatic
>solution?
Because they are both liquids, and that's much more important than
their viscosity.
> See:
>
>http://www.uic.edu/classes/phys/phys450/MARKO/N004.html
>
>"Water, and solutions which are mostly water, have a viscosity close
>to hwater = 0.01 Poise or 0.001 Pa·sec."
Note that this link talks about viscosity in relation to "small
particles" in liquids, _not_ in relation to atoms (which are millions
of times smaller than the "small particles"). Viscosity is important
in "diffusion" of particles that are much much larger than the atoms
or molecules of the fluid; viscosity is _not_ very important in
diffusion of atoms that are about the same size as the atoms or
molecules of the fluid.
>or better yet, this link:
>
>http://www.brookes.ac.uk/geology/sedstruc/viscosit/vis.htm
>
>"Material Poise
>
>Water 1
>Engine Oil 10
>Plasticine 1000
>Asphalt 10,000
>Basalts (lava) 100 - 1000
>Granites (magma) 1,000,000 - 1,000,000,000,000"
>
>diffusion in water is going to be at a far higher rate than diffusion
>in molten rock.
Interesting. You _ASSUMED_ that viscosity is important, then you
_ASSERTED_ on that basis that diffusion is faster in water.
Sorry, Zoe, just making it up as you go along doesn't cut it.
Diffusion in high temperature magma may well be much faster than
diffusion in low-temperature water, I don't know off-hand. I do know
that temperature is much more important than diffusivity.
>Indeed, I would expect that the concentration in
>magma is such that there is no need for exchange of particles from
>areas of higher concentration to areas of lower concentration.
Indeed, that's an incredibly silly expectation.
Whenever there is a difference in concentration, there is exchange of
particles from areas of higher concentration to areas of lower
concentration. The only question is how fast it happens.
The only way that there can be "no need for exchange of particles from
areas of higher concentration to areas of lower concentration" is if
the concentration is uniform, and you are trying to argue that the
concentration is _not_ uniform.
>Density matters, as appears to be the case in this link:
>
>http://chemed.chem.purdue.edu/analyticalreview/chromatography/ratetheory.htm
>
>"2) flow velocities will vary throughout a packed column and cause
>zone spreading,"
This does not refer to density ... it does not relate to diffusion in
magma or convection or any kind of magmatic mixing.
>You need to have two or more different species in a solution in order
>for diffusion to occur, right?
Wrong. All you need is one species with different concentrations in
different places in the solution.
>What are the different species in
>molten rock?
For example:
P atoms ... if there is more P in one part of the magma than another,
then P atoms will move from the part with more P to the part with less
P.
Or D atoms ... if there is more D in one part of the magma than
another, then D atoms will move from the part with more D to the part
with less D.
And this will destroy the straight line relationship on the isochron
diagram ... until the process has gone far enough that the melt is
almost homogeneous, at which time the points representing the samples
will "home in on" a horizontal straight line
>I mean, diffusion of salt or sugar molecules into water
>would result in eventual homogeneity -- well, no, not even then,
>unless you stir the solution; otherwise, the crystals would settle out
>instead of dissolve.
Er, no, unless there is so much sugar or salt that the solution is
supersaturated .. which is not realistic for magma.
Diffusion of sugar or salt molecules into water will result in
eventual homogeneity. This will be slowed down if the sugar or salt
starts out solid and has to dissolve ... but when a rock melts,the P
and D gets dissolved immediately, so this factor doesn't have an
effect on magma.
>But in more dense materials, the rate of
>diffusion should slow down.
Not necessarily. Higher density can slow diffusion somewhat. Higher
temperature increases the rate of diffusion terrifically.
I really wish you would stop making claims about diffusion that are
based only on your fantasies. Diffusion has been studied for many
years, and is pretty well understood. There's no reason for making
stuff up as you are doing.
>And in the case of a magma chamber, all
>you have is liquid rock, so you won't have magma dissolving into water
>as you would with salt/sugar/red dye in water.
You have liquid rock with P atoms and D atoms dissolved in it. You
are claiming that there will be different amounts of P and D in
different parts of the molten rock (and that will be true for some
period of time after the rock melts; the question is for how long it
will be true). Under those conditions, P and D atoms will diffuse.
The only question is how fast.
>>>None of the above actions are considered regular underground activity
>>>in the earth. Heat is not sufficient to produce a homogenous mixture
>>>(see first link above
>>
>>Sorry, the link above does not say that heat is not sufficient to
>>produce a homogeneous mixture. It implicitly says that heat is not
>>sufficient to produce a homogeneous mixture _in_ _the_ _amount_ _of_
>>_time_ _that_ _is_ _allotted_.
>
>if you are invoking billions of years for homogeneity to take place,
>then your hypothesis of magma being homogenous becomes an
>unfalsifiable hypothesis.
Nope. Testable predictions can be made, such as recently solidified
rock should be homogeneous. It almost always is. Or, samples from a
rock should often plot as a straight line on an isochron diagram (your
proposed "memory" would not, and you haven't addressed this). They
do.
>>If they waited a few years, the allow would be well mixed without
>>stirring. They just don't want to wait that long.
>
>please for support for the idea that the only reason for mixing is
>because it would take too long otherwise. And also give support for
>the idea that the alloy would become well mixed if allowed to sit for
>a few years.
EVERYTHING becomes well mixed if allowed to sit for long enough.
Diffusion happens in EVERY inhomogeneous mixture. The only question
is how long it takes.
Plug the numbers for your molten metal into Fick's law and see what
happens. If you don't know Fick's law, you don't know enough to
understand the support for the claim.
>snip>
>
>>Your experiment is irrelevant. The temperature is too low by an order
>>of magnitude or two, the time is too short by _many_ _many_ orders of
>>magnitude, the constituents are inappropriate because they cannot
>>stand up to temperatures that are realistic for underground magma.
>
>are you saying that one cannot do an experiment on a lower level and
>extrapolate the results out to a higher level -- that the laws of
>physics do change with time?
Certainly you can extrapolate, and certainly the laws of physics
appear to be the same over time. But you have to _do_ the
extrapolation, and you have to do it using the laws of physics, and
you have to do it using physically valid formulas.
You didn't do any extrapolation. You didn't make any use of the laws
of physics (the effect of temperature on diffusion rate, the effect of
time on the amount of mixing by diffusion).
Go ahead, do your extrapolation. Show your calculations. Remember the
exponential dependence of diffusion rate on temperature.
Doing an experiment under one set of conditions and claiming that the
result applies directly to wildly different conditions is not
extrapolation, it's fantasizing.
>>>Question: What prevents magma from burning up underground?
>>
>>1. Chemistry. The constituents of magma do not burn (combine
>>somewhat rapidly with oxygen) at the temperatures encountered
>>underground, even though these temperatures are much much higher than
>>encountered in your experiment.
>
>okay, lack of oxygen.
No, even with lots of oxygen present, they don't burn.
>>
>>2. Even if they did burn at those temperatures, there's not much
>>oxygen there to combine with.
>>
>>Put a rock in your microwave or conventional oven. It doesn't burn.
>
>it's all relative. The characteristics of melting should be the same
>once the rock melts. It shouldn't matter that it takes a higher
>temperature to melt a rock. What I am interested in is what happens
>when the material (be it rock or chocolate) finally does melt.
Molten rock doesn't burn (in the sense of rapid reaction producing
ash), even with plenty of oxygen.
It does oxidize, just not fast. And the product isn't ash. For
example quartz is oxidized (sort of "burned") silicon.
>snip>
>
>>Again an irrelevant experiment. The constituents are inappropriate
>>because they can't stand up to the temperatures that magma encounters
>>underground, the time is many many orders of magnitude too short, the
>>temperature is way too low.
>
>again, Jon, are you saying that the characteristics and behavior of a
>melt differs strictly because it took Item A longer to melt than Item
>B and because it took more heat to melt Item A than Item B?
No. I'm saying that the characteristics and behavior of melts differ
when the constituents differ, and the characteristics and behavior of
melts differ when the temperature differs. I'm also saying that the
effects of a slow process aren't seen over short periods of time but
are seen over long periods of time.
>>>Suddenly the sealed cover began to act like the
>>>beginning of a volcanic explosion, and chocolate began to pour out
>>>from the sides of the seal and run down, lava-like, burning and
>>>turning black as it went. I guess this is the point where pressure
>>>builds up underground to where, eventually, a volcanic eruption takes
>>>place?
>>
>>Not really ... the source of the pressure on the magma is due to many
>>sources. Your "explosion" was probably due to water turning to steam,
>>which is significant in some but not all volcanic eruptions.
>
>water in chocolate? Chocolate is a mixture of cocoa paste, cocoa
>butter, and sugar. What water?
There's water in there, 0.5 to 1.6% of the chocolate, assuming you
have stored it under very dry conditions. See
<http://www.tis-gdv.de/tis_e/ware/lebensmi/schoko/schoko.htm#feuchte>.
You have an unfortunate tendency to assume you know something about
many areas in which you are totally ignorant.
>
>snip>
>
>>The only materials suitable for a useful experiment are molten rocks.
>>You don't have the facilities to run a realistic experiment. You
>>can't get temperatures high enough, you can't wait long enough, you
>>don't have the instrumentation required to make accurate measurements
>>of what happens over shorter times.
>
>well, so much for extrapolation in science; it is a useless tool -- as
>far as Jon is concerned, anyway.
You haven't even attempted an extrapolation.
Extrapolation is a very dangerous tool, because it's prone to error,
and the farther you extrapolate the more likely an error is; but it's
a useful tool when used properly.
You need very accurate data to start with (which you don't have), and
you have to actually _calculate_ the extrapolation based on the laws
of physics. You need to dig up a formula for the effect of the higher
temperature, a formula for the effect of the longer time, and a
formula for the effect of the different material, and apply those
formulas, including an estimate of the error.
What you have done is not extrapolation, it's just fantasizing.
Jon Fleming
wrote:
>On Fri, 23 May 2003 16:18:10 +0000 (UTC), pven...@hotmail.com (P.
>Venkman) wrote:
>
>snip>
>
>>Actually I have a question about this. You're questioning whether a
>>linear relationship between D/Di and P/Di can be retained during a
>>melt. That pre-supposes a linear relationship exists pre-melt. Take
>>that back to its logical conclusion and that means a linear
>>relationship had to exist at the formation of the earth. Is there any
>>reason to believe this is true?
>
>the reason to believe this is true is that the laws of physics is the
>same today as it always was. And since the only observation of these
>laws that holds true today is the one that say that D originates from
>decay of P,
We have observed creation of D in supernovas.
>then there could be no original oldD at the first
>formation of the earth
Wrong, as has been pointed out many times before.
<snip>
>>I do have one question for you. How do you account for the fact that
>>in new flows that have been measured we do see that D/Di is the same
>>for all samples?
>
>using the Rb/Sr measure?
Yes,and using others. Answer the question.
Jon Fleming
wrote:
>On Fri, 23 May 2003 18:29:57 +0000 (UTC), John Stockwell
><jo...@dix.Mines.EDU> wrote:
>
>snip>
>
>>You forgot "diffusion". Look up something called "Brownian motion".
>
>John, I did look it up. It seems, from what little I've covered so
>far, that diffusion rates depend on temperature of the suspension,
>viscosity of the solvent, and radius of the colloid particle.
For colloid particles, yes. We are not talking about diffusion of
colloid particles. We are talking about diffusion of atoms through a
liquid in which the diffusing atoms and liquid atoms are about the
same size. Viscosity is less important in this situation.
>Therefore, to loosely apply diffusion of particles in water to
>diffusion of isotopes in magma would be like comparing apples to
>oranges.
That's true, although not for the reason you gave.
Also, loosely applying diffusion of different chocolates to diffusion
of isotopes in magma would be like comparing apples to oranges.
Loosely applying diffusion at temperatures you can achieve in your
kitchen to diffusion of isotopes in magma at much higher temperatures
would be like comparing apples to oranges.
Loosely applying diffusion over a period of a few minutes to diffusion
of isotopes in magma over thousands to billions of years would be like
comparing apples to oranges.
You have loosely applied the results of your experiment three times
over, making it totally meaningless.
Jon Fleming
<arensb.no-...@glue.umd.edu> wrote:
>R. Baldwin <res0...@nozirevbackwards.net> wrote:
>> Simple pressure can cause a liquid or colloid to extrude through a
>> medium or a constricted opening, causing mixing.
>
> How, then, do you explain striped toothpaste?
> I'm only being semi-facetious: obviously the toothpaste is not
>thoroughly mixed after being forced by pressure through a constricted
>opening. Either your statement is untrue, or else it doesn't apply to
>toothpaste. If the latter, why?
There's a third alternative ... the stripes are applied at the end of
forcing it through a set of constricted openings, where the
constriction starts opening up. See
<http://www.straightdope.com/classics/a1_173.html>.
Matt Silberstein
<arensb.no-...@glue.umd.edu>:
>R. Baldwin <res0...@nozirevbackwards.net> wrote:
>> Simple pressure can cause a liquid or colloid to extrude through a
>> medium or a constricted opening, causing mixing.
>
> How, then, do you explain striped toothpaste?
> I'm only being semi-facetious: obviously the toothpaste is not
>thoroughly mixed after being forced by pressure through a constricted
>opening. Either your statement is untrue, or else it doesn't apply to
>toothpaste. If the latter, why?
The layers come out swirled. It does start to mix.
--
Matt Silberstein TBC HRL OMM
We are not here to judge other people,
we are just here to be better than they are.
Matt Silberstein
(zoe_althrop):
Try this. Take some light honey, some black molasses, and some
clear corn syrup. Take some clear jars and put layers of each
material in the jar. Put one jar in the fridge. Put another on
the shelf and let it sit for days/weeks. Take a third jar and put
it in a pot. You want water in the pot about as high as the
bottom layer, but certainly not as high as the top. Keep the
water warm and watch the layer boundary. Let it sit in
moderately warm (100 degrees or so) for an hour or so. You could
try this with several different water temps. The really
interesting thing will come when the material starts to show
convection. That is, material on the bottom will be warmer than
the top and rise, material on the top will drop. This will mix
the three materials quite well.
Bigdakine
>Date: 5/24/03 8:04 AM Hawaiian Standard Time
>Message-id: <baoc7s$nu3$2...@grapevine.wam.umd.edu>
>
>R. Baldwin <res0...@nozirevbackwards.net> wrote:
>> Simple pressure can cause a liquid or colloid to extrude through a
>> medium or a constricted opening, causing mixing.
>
> How, then, do you explain striped toothpaste?
> I'm only being semi-facetious: obviously the toothpaste is not
>thoroughly mixed after being forced by pressure through a constricted
>opening. Either your statement is untrue, or else it doesn't apply to
>toothpaste. If the latter, why?
I think it is not enough to force it through an opening. You need to force it
through an opening such that it becomes turbulent. Or the opening is real small
such that diffusion can operate quickly due to the much smaller length scale
imposed by the nozzel.
Neither conditions apply to a tube of toothpaste. The flow of the toothpaste is
laminar, and the nozzle does not represent a small enough length scale to
facilitate rapid mixing. So the amount of mixing is slight.
Bigdakine
>Date: 5/24/03 9:05 AM Hawaiian Standard Time
>Message-id: <3ecfc2a2....@news-server.cfl.rr.com>
hydrothermal system. That alone stirs things up. Diffusion not required.
>Density matters, as appears to be the case in this link:
>
>http://chemed.chem.purdue.edu/analyticalreview/chromatography/ratetheory.htm
>
>"2) flow velocities will vary throughout a packed column and cause
>zone spreading,"
>
>You need to have two or more different species in a solution in order
>for diffusion to occur, right? What are the different species in
>molten rock?
Again mixing in magmas is accomplished by convecting or stirring, not by
diffusion. Thats not to say that diffusion doesn't occur, but it is only
important on smaller length scales.
<snip>
Seppo Pietikainen
> On Fri, 23 May 2003 14:41:23 +0000 (UTC), Seppo Pietikainen
> <s.pietika...@kolumbus.fi> wrote:
>
> snip>
>
>> First of all, *DO NOT EAT* you saran wrap+chocolate mixture!!!
>> (please, tell your next of kin when you're going to do a "scientific
>>experiment").
>
>
> good advice, Seppo. I think I have sufficient chocolate morsels on
> hand that I won't be driven to eat the plastic+chocolate mixture.
>
> snip>
>
>> There *are* some time constraints associated with industrial processes.
>>Hence, every now and then things go awfully wrong, especially when large
>>amount of reagents are mixed... Volcanoes behave "a little"
>>different.(I'd prefer studying volcanoes to stratification problems in
>>chemical industry...[I do neither, but I've been involved with the
>>latter, in a fashion, at least])
>
>
> in what fashion?
In my previous job more than 20 years ago in a research institute
(doing physical oceanography)...
>
> snip>
>
>> Add convection to the list and you might get it somewhat closer.
>>
>> May I suggest a far simpler (and safer for minors and pets...) experiment:
>> 1) Take a kettle (greater than 1 litre capacity [about 1/4 Gallons])
>> 2) Fill it with approximately 1 litre (~ 1 pint) of cold water
>> 3) Add about 30 grams (~1 oz) of ordinary kitchen salt.
>> 4) Heat it to, say, 70 degrees C (~160F), continue for 5 minutes
>>without stirring.
>> 5) Report non-uniformities in the mixture.
>
>
> convection in water that is boiled in an open pot, with access to all
> the oxygen available, is much more rapid than convection in a far more
> dense material in a non-oxygenated chamber, so this experiment would
> not be much of a commentary on the behavior of magma, would it?
>
Availability of oxygen has *nothing* to do with convection mixing.
>
>>Meanwhile, you might want to do a search of "thermal convection" in the
>>web or your local library.
>
>
> if by "thermal" convection you mean heat convection, then, yes, I've
> been doing a bit of reading up on it.
>
> snip>
>
>
>>>Question: What prevents magma from burning up underground?
>>
>> Ummm. Maybe you haven't heard, but *free* oxygen is not terribly
>>abundant underground.
>
>
> yes, I'm hearing about that in this thread. Which would be another
> factor that would rule out an open-pot comparison.
Nope, availability of oxygen has to do with *oxidizing* (=burning) your
chocolate chips.
>
> snip>
>
> ----
> zoe
>
Seppo P.
R. Baldwin
news:20030524200926...@mb-m12.aol.com...
Quite right. IIRC it depends on the viscosity as well, which for
toothpaste is significant. My response to Zoe was simplistic, just to
point out one of the many things she was overlooking. Rheology and
mixing are complicated subjects about which I only know a smattering.
Some rheology-related sites:
http://www.rheology.org/sor/
http://www.sbu.ac.uk/water/hyrhe.html
http://rrc.engr.wisc.edu/
http://www.rheology-esr.org/
http://www.rheology-online.com/TechCentre/TechCentre.htm
http://www.netaccess.on.ca/~dbc/cic_hamilton/mix.html
R. Baldwin
news:3ecfc65a....@news-server.cfl.rr.com...
Within a magma chamber, you should think about convection and
volatiles. Lava is extruded out of the magma chamber, of course,
though I was thinking more of the mantle.
Here are some interesting links for you:
http://www.uwsp.edu/geo/faculty/ritter/glossary/l_n/magma_chamber.html
http://gfd.gly.bris.ac.uk/gfd-people/heidy.mader/studentships/rheology
..html
http://soconnell.web.wesleyan.edu/courses/ees106/lecture_notes/lecture
6_106/sld001.htm
Frank J
I said "correct me if I'm wrong," which means that I just made a
tentative statement. With your answer I will gladly change my
"preconception" to the fact that you are *not* trying to get part of
your audience to infer design. I will now look at some of your other
responses to see if you have actually gotten past square one, which
means that you have offered a better theory of homoginization, and are
on your way to collecting your Nobel Prize.
>
> ----
> zoe
Jon Fleming
Do you have any Web or library references that are a good overview of
this, or could you perhaps discuss it some more?
It appears to me that what mechanism dominates depends on what length
scale we are talking about. For isochron analysis, it appears to me
that the length scale is on the order of the size of the chunk of
magma/rock from which the samples are drawn. That is, if we sample
form a rock that's 1 m diameter, we're also sampling from a 1 m (or
perhaps slightly larger) pool of magma that was the source of that
rock, and the process or processes that dominate at that scale are
what we want to consider. Who cares if the melt was inhomogeneous
over a scale of 1,000 km if we're only sampling that 1 m chunk?
It also appears to me that the appropriate scale for isochron analysis
sampling ranges from maybe 1 cm at the extreme low end to 100 km at
the extreme high end, with most of the population falling in the 10 cm
to 10 m range. Is this reasonable?
What I've managed to dig up suggests that mixing in magmas is
dominated by diffusion over distances something less than 10 m and
dominated by mechanical mixing processes over larger distances. And
that in turn suggests that diffusion is the dominant factor in the
homogenization of the source of most isochron samples ... unless I've
gone wrong somewhere in this chain of reasoning.
At <http://tinyurl.com/cm7u> Dr. Sandiford of the University of
Melbourne says:
"It is an important observation that even at temperatures above the
blocking temperature, the rate of diffusional equilibration between
adjacent solids or non-convective, non-mixing liquids is only of the
order of a meter in 10^7 years. For this reason the establishment of
large separate geochemical reservoirs like the crust or asthenospheric
upper mantle, requires both large scale melt extraction and convective
heat and mass transfer. "
And Dr. White at Cornell says, in a discussion of the requirements for
a valid isochron analysis at
<http://www.geo.cornell.edu/geology/classes/Geo656/656notes03/656%2003Lecture06.pdf>:
"The isotopic composition of the daughter must have been homogeneous
at the time of the event we wish to date. On a small scale,
homogenization takes place through diffusion, which, as we have seen,
is highly temperature dependent. The higher the temperatures obtained
during the 'event', the more homogenized the system will be. On scales
larger than 10 meters or so, homogenization can only be achieved
through convective-driven advective transport. This effectively means
homogenization requires the presence of a liquid. This might be a
magma or a hydrous fluid circulating through rocks undergoing
metamorphism. In any case, both convection and diffusion will be more
efficient at higher temperatures, so homogenization is more likely to
be achieved at high temperatures than at low ones. Finally, the larger
the range in parent/daughter ratios, and hence isotopic composition at
the time we measure them, the less important will be any initial
variations in isotopic composition."
Yet he also says, when discussing the same subject in another document
for the same course, at
<http://www.geo.cornell.edu/geology/classes/Geo656/656notes03/656%2003Lecture04.pdf>
"The agent accomplishing isotopic homogenization of a ‘system’ is
usually diffusion, the rate of which, like other reaction rates,
increases exponentially with temperature. Diffusion rates will vary
depending on the element and the properties of the material through
which the element diffuses. We can nevertheless make the general
observation that the greater the length scale the greater will be the
time (or the higher the temperature required) for isotopic
homogenization to be achieved. For the same temperature and duration
of a thermal event, diffusion will more readily achieve isotopic
homogenization on a small scale than on a large one. Thus, if our
samples or subsystems are 'whole rocks' collected meters or perhaps
kilometers apart, the event dated will generally be a higher
temperature one than an event dated by analysis of individual minerals
from a rock specimen whose scale is only a few centimeters."
Howard Hershey
muz...@aol.com wrote on 5/24/03 7:19 PM:
Zoe, any experiment you wish to extrapolate the results of to the levels of
D/Di in lavas *has* to be consistent with the *direct observations* of
constant D/Di in newly solidified lavas. You cannot use any of your
experimental results to claim that D/Di in newly solidified samples from a
melt will always or mostly produce linear positive slopes which correlate
with the levels of P present in those samples and also correlates with the
position of the rocks in the geological column (meaning that the mechanism
must be different at different geological levels).
Direct observation of constant D/Di uncorrelated with levels of P in newly
formed lavas tells you that your wishful extrapolations from these
experiments is wrong, wrong, wrong. Observation trumps erroneous
extrapolation.
>
> ----
> zoe
>
John Drayton
And melting chocolate in your kitchen does ?!?
How do you tell if an experiment is a good simulation?
If the results agreee with your preconceived ideas?
Gotta wonder why people are wasting time studying and
measuring *real* samples, when they could do chocolate
experiments.
How stupid ... can you eat your rock sample after you're
done?
<Homer Simpson voice>
Mmmmmmm .... chocolate magma
</Homer Simpson voice>
> ----
> zoe
--
John Drayton
zoe_althrop
<jo...@fleming-nospam.com> wrote:
>On Sat, 24 May 2003 zoe_althrop wrote:
snip>
>>viscosity (low versus high) is a critical factor in diffusion, I
>>think.
>
>You are wrong. Viscosity is not a critical factor (in diffusion of
>atoms and molecules). Density is not a critical factor. The major
>critical factors are temperature, concentration gradient, and mobility
>within the material. Although viscosity slightly affects mobility
>within the material, the material form (solid, liquid, or gas) is the
>major effect on mobility.
okay, maybe the word "viscosity" is wrong. I take it back. How about
density/porosity? I mean, there ARE different rates of diffusion,
ranging from high to practically nil, right? Surely, there has to be
SOME factor that is critical to rate of diffusion.
So I'll use your term, "material form." But I would define it more
closely than just solid, liquid, or gas, because liquids come in all
degrees of densities -- if I'm not misusing that word as well.
Mobility within liquid jello is much higher than mobility within
liquid chocolate, from what I have observed? Not to mention liquid
rock -- a molten material far more dense than either jello or
chocolate.
snip>
>
>> See:
>>
>>http://www.uic.edu/classes/phys/phys450/MARKO/N004.html
>>
>>"Water, and solutions which are mostly water, have a viscosity close
>>to hwater = 0.01 Poise or 0.001 Pa·sec."
>
>Note that this link talks about viscosity in relation to "small
>particles" in liquids, _not_ in relation to atoms (which are millions
>of times smaller than the "small particles").
I suppose this means that the behavior of particles in liquids will
change, depending on size? The smaller the particle, the more mobile,
regardless of the density or porosity of the liquid?
>Viscosity is important
>in "diffusion" of particles that are much much larger than the atoms
>or molecules of the fluid; viscosity is _not_ very important in
>diffusion of atoms that are about the same size as the atoms or
>molecules of the fluid.
is what they're teaching students in science classes today incorrect,
then? See:
www.ns.msu.edu/drew/nsc204/PagesLabSS/TAosmos.pdf
"The rate of diffusion is proportional to the number of particles of
solute in solution,,whether they are small molecules, ions or large
polymeric molecules.. The presence of solute particles lessens the
water potential of the solution.. The more concentrated a solution
is,the more negative its water potential"
Sounds to me as if the same principles apply in diffusion, whether the
particles are large or small.
snip>
>Diffusion in high temperature magma may well be much faster than
>diffusion in low-temperature water, I don't know off-hand. I do know
>that temperature is much more important than diffusivity.
maybe you'll take it from Caltech, then?
http://www.gps.caltech.edu/~gurnis/Movies/Science_Captions/stirring.html
"Even after a few billion years of stirring by mantle convection, the
fluid is still quite heterogneous."
>>Indeed, I would expect that the concentration in
>>magma is such that there is no need for exchange of particles from
>>areas of higher concentration to areas of lower concentration.
>
>Indeed, that's an incredibly silly expectation.
>
>Whenever there is a difference in concentration, there is exchange of
>particles from areas of higher concentration to areas of lower
>concentration.
that is just it -- is there really a difference in concentration in
magma? Rock, as it liquefies, compacts and becomes more dense; IF,
that is, it follows the same laws as melting chocolate.
> The only question is how fast it happens.
billions of years, maybe? More than 4.5 billion?
>The only way that there can be "no need for exchange of particles from
>areas of higher concentration to areas of lower concentration" is if
>the concentration is uniform,
exactly. That is what I am saying. That magma is uniform in density.
>and you are trying to argue that the
>concentration is _not_ uniform.
no, uniformity of density is not the same as homogenous. I just got
through experimenting with the chocolate again, this time in a double
boiler, and the chocolate compacted down to half a jar (where it had
started out as a full jar) and the white, caramel, and brown chips
remain in exactly their original formation. But there was no space
between them anymore. That would be uniform density, I would think --
no porosity, no further spaces for the chips to go from higher
concentration to lower concentration. But that is NOT homogeneity.
>>Density matters, as appears to be the case in this link:
>>
>>http://chemed.chem.purdue.edu/analyticalreview/chromatography/ratetheory.htm
>>
>>"2) flow velocities will vary throughout a packed column and cause
>>zone spreading,"
>
>This does not refer to density ... it does not relate to diffusion in
>magma or convection or any kind of magmatic mixing.
what does it refer to then? I thought it described a basic principle
of packing -- or compacting.
>>You need to have two or more different species in a solution in order
>>for diffusion to occur, right?
>
>Wrong. All you need is one species with different concentrations in
>different places in the solution.
well, my D (brown chocolate) and Di (white chocolate) and P (caramel
pieces) were definitely in different concentrations. I did not see
them gravitating to less concentrated areas for the simple reason that
there was nowhere for them to go. They were packed too closely to be
mobile.
>>What are the different species in
>>molten rock?
>
>For example:
>
>P atoms ... if there is more P in one part of the magma than another,
>then P atoms will move from the part with more P to the part with less
>P.
not according to the chocolate experiment. Caramel bits did not move
to areas where there was less caramel bits.
>Or D atoms ... if there is more D in one part of the magma than
>another, then D atoms will move from the part with more D to the part
>with less D.
same thing here. Brown chocolate remained exactly where it had been
placed simply because there was no place for it to go. The
concentration of white and caramel chocolate was such that all spaces
were filled and brown chocolate (P) had nowhere to go but to remain in
place.
>And this will destroy the straight line relationship on the isochron
>diagram ... until the process has gone far enough that the melt is
>almost homogeneous, at which time the points representing the samples
>will "home in on" a horizontal straight line
you're still describing a boiling, mixing, stirring process in order
to get a homogenous mixture. None of this happens with chocolate or
magma.
snip>
>>But in more dense materials, the rate of
>>diffusion should slow down.
>
>Not necessarily. Higher density can slow diffusion somewhat. Higher
>temperature increases the rate of diffusion terrifically.
not from what I observed in my chocolate-melting experiment.
>I really wish you would stop making claims about diffusion that are
>based only on your fantasies. Diffusion has been studied for many
>years, and is pretty well understood. There's no reason for making
>stuff up as you are doing.
I am not making this up. I am telling you what I have observed occurs
in watery solutions versus chocolate melts.
>>And in the case of a magma chamber, all
>>you have is liquid rock, so you won't have magma dissolving into water
>>as you would with salt/sugar/red dye in water.
>
>You have liquid rock with P atoms and D atoms dissolved in it. You
>are claiming that there will be different amounts of P and D in
>different parts of the molten rock (and that will be true for some
>period of time after the rock melts; the question is for how long it
>will be true). Under those conditions, P and D atoms will diffuse.
>The only question is how fast.
billions of years, and the melt will still be heterogenous. See link
above.
snip>
>>if you are invoking billions of years for homogeneity to take place,
>>then your hypothesis of magma being homogenous becomes an
>>unfalsifiable hypothesis.
>
>Nope. Testable predictions can be made, such as recently solidified
>rock should be homogeneous. It almost always is.
please for live examples from the field. If it takes billions of
years, and even then the mix remains heterogenous, how can a recent
solidification be homogenous? Something doesn't fit here.
>Or, samples from a
>rock should often plot as a straight line on an isochron diagram (your
>proposed "memory" would not, and you haven't addressed this). They
>do.
maybe you're talking about Ar-Ar in these cases? I don't know why
they plot on a straight line. But it wouldn't be because of natural
homogeneity, I don't think.
>>>If they waited a few years, the allow would be well mixed without
>>>stirring. They just don't want to wait that long.
>>
>>please for support for the idea that the only reason for mixing is
>>because it would take too long otherwise. And also give support for
>>the idea that the alloy would become well mixed if allowed to sit for
>>a few years.
>
>EVERYTHING becomes well mixed if allowed to sit for long enough.
>Diffusion happens in EVERY inhomogeneous mixture. The only question
>is how long it takes.
not even after billions of years. It is fantasizing to imagine that
cake batter will mix itself if given enough time. But then I must
remember that this is the modus operandus of evolutionary theory --
allow billions of years and surely, a batter of cyanobacteria will
transform into a civilization of human beings.
>Plug the numbers for your molten metal into Fick's law and see what
>happens. If you don't know Fick's law, you don't know enough to
>understand the support for the claim.
I've been reading about Fick's law. It has to do with diffusion, but
usually with gases, not melted solids, as far as I can see (which
probably is not very far, so correct me if I'm wrong.)
>>snip>
>>
>>>Your experiment is irrelevant. The temperature is too low by an order
>>>of magnitude or two, the time is too short by _many_ _many_ orders of
>>>magnitude, the constituents are inappropriate because they cannot
>>>stand up to temperatures that are realistic for underground magma.
>>
>>are you saying that one cannot do an experiment on a lower level and
>>extrapolate the results out to a higher level -- that the laws of
>>physics do change with time?
>
>Certainly you can extrapolate, and certainly the laws of physics
>appear to be the same over time. But you have to _do_ the
>extrapolation, and you have to do it using the laws of physics, and
>you have to do it using physically valid formulas.
I am doing it with real-world observation in this chocolate
experiment. Somebody else can draw up the formulas as to WHY or HOW
the chocolate pieces maintain their original positions, but the data
is that temperature does not cause homogeneity in a non-aqueous
solution.
>You didn't do any extrapolation. You didn't make any use of the laws
>of physics (the effect of temperature on diffusion rate, the effect of
>time on the amount of mixing by diffusion).
I am telling you what I observed from a real-world experiment. If
someone wants to draw up the equations that describe the results that
I have observed, they are welcome to do so.
>Go ahead, do your extrapolation. Show your calculations. Remember the
>exponential dependence of diffusion rate on temperature.
I don't have to do this. You can. I am giving you hard observed data
to work with.
>Doing an experiment under one set of conditions and claiming that the
>result applies directly to wildly different conditions is not
>extrapolation, it's fantasizing.
labeling the two conditions "wildly different" doesn't make it so. I
think there is a lot of similarity.
snip>
>>water in chocolate? Chocolate is a mixture of cocoa paste, cocoa
>>butter, and sugar. What water?
>
>There's water in there, 0.5 to 1.6% of the chocolate, assuming you
>have stored it under very dry conditions. See
><http://www.tis-gdv.de/tis_e/ware/lebensmi/schoko/schoko.htm#feuchte>.
>
>You have an unfortunate tendency to assume you know something about
>many areas in which you are totally ignorant.
I didn't make that up. I got it from a site that neglected to mention
the 0.5 to 1.6%. I stand corrected. But this is still not sufficient
water to make a difference in how the chocolate melts.
>>
>>snip>
>>
>>>The only materials suitable for a useful experiment are molten rocks.
>>>You don't have the facilities to run a realistic experiment. You
>>>can't get temperatures high enough, you can't wait long enough, you
>>>don't have the instrumentation required to make accurate measurements
>>>of what happens over shorter times.
>>
>>well, so much for extrapolation in science; it is a useless tool -- as
>>far as Jon is concerned, anyway.
>
>You haven't even attempted an extrapolation.
>
>Extrapolation is a very dangerous tool, because it's prone to error,
>and the farther you extrapolate the more likely an error is; but it's
>a useful tool when used properly.
you mean, when evolutionary theory extrapolates into the billions of
years, then it is more likely to be in error, too?
snip>
----
zoe
zoe_althrop
<s.pietika...@kolumbus.fi> wrote:
snip>
> In my previous job more than 20 years ago in a research institute
>(doing physical oceanography)...
then maybe you can tell me, is there a difference in the magma that
reaches the surface of the ocean floor and the magma that erupts as
lava on dry ground? Does magma erupting into water produce different
rock types to magma erupting into air?
snip>
----
zoe
zoe_althrop
<mat...@ix.netcom.com> wrote:
snip>
>Try this. Take some light honey, some black molasses, and some
>clear corn syrup. Take some clear jars and put layers of each
>material in the jar. Put one jar in the fridge. Put another on
>the shelf and let it sit for days/weeks. Take a third jar and put
>it in a pot. You want water in the pot about as high as the
>bottom layer, but certainly not as high as the top. Keep the
>water warm and watch the layer boundary. Let it sit in
>moderately warm (100 degrees or so) for an hour or so. You could
>try this with several different water temps. The really
>interesting thing will come when the material starts to show
>convection. That is, material on the bottom will be warmer than
>the top and rise, material on the top will drop. This will mix
>the three materials quite well.
thanks for the suggestion, Matt, but I don't think that honey would
represent rock the way chocolate more nearly does. Both chocolate and
rock are initially in solid forms, and when melted, the observation is
one of how well material of these relative densities and heating will
become mixed. Don't you think that the experiment should start out
with something in solid form first, and then see what happens to its
material as it melts? If you don't think so, why not?
----
zoe
zoe_althrop
<jo...@fleming-nospam.com> wrote:
snip>
>We have observed creation of D in supernovas.
what is the method used for observing this creation of D? I've always
wondered how exactly it is done, seeing how distant these novas are
from hands-on investigation.
>>then there could be no original oldD at the first
>>formation of the earth
>
>Wrong, as has been pointed out many times before.
pointed out via assertions?
>
><snip>
>
>>>I do have one question for you. How do you account for the fact that
>>>in new flows that have been measured we do see that D/Di is the same
>>>for all samples?
>>
>>using the Rb/Sr measure?
>
>Yes,and using others. Answer the question.
not until you answer mine: Why would you use a 48-billion-year half
life to date a new flow that is just a few weeks or years old?
----
zoe
zoe_althrop
<res0...@nozirevBACKWARDS.net> wrote:
snip>
>Here are some interesting links for you:
thanks.
>http://www.uwsp.edu/geo/faculty/ritter/glossary/l_n/magma_chamber.html
why does not the rock around the magma melt, also? If the heat is
constant, from whatever source that is heating the magma, then
eventually, over billions of years, all underground rock should be a
molten mass, shouldn't it? Why isn't that the case?
>http://gfd.gly.bris.ac.uk/gfd-people/heidy.mader/studentships/rheology
>..html
this is a modeling of volcanic flows. Hey, I've learned a more
impressive word for the flow behavior of mixtures: rheology. I shall
use that term from now on in order to communicate more effectively
with your scientific minds. :-)
>http://soconnell.web.wesleyan.edu/courses/ees106/lecture_notes/lecture
>6_106/sld001.htm
very informative. It also answers some of what I was just asking
Seppo about.
----
zoe
zoe_althrop
<rmc...@attbi.com> wrote:
snip>
>With a double boiler you don't need to worry about the oxygen. The
>water in the boiler keeps the chocolate from reaching its ignition
>point. So the chocolate doesn't burn as long as there is water in the
>double boiler.
right, the double boiler method does prevent burning. This time I put
the chocolate into a glass canning jar and immersed it in a pot of
boiling water. The chocolate simply compacted, retaining its original
positions, and after three hours of this, I lifted the jar out one
time too many for examination, and the cover came away in my hand and
the jar fell back into the pot, filling with water. So I covered it
back and left it to boil some more, just to see if the addition of
water would cause convection currents and mix the chocolate this time.
Surprise. I came back to find the water boiling merrily away on the
TOP of the chocolate, and the compacted chocolate continued to
maintain its original formation in the bottom half of the jar.
What did I do wrong?
----
zoe
Bigdakine
>Message-id: <3ed174f1....@news-server.cfl.rr.com>
>
>On Sun, 25 May 2003 06:17:44 +0000 (UTC), "R. Baldwin"
><res0...@nozirevBACKWARDS.net> wrote:
>
>snip>
>
>>Here are some interesting links for you:
>
>thanks.
>
>>http://www.uwsp.edu/geo/faculty/ritter/glossary/l_n/magma_chamber.html
>
>why does not the rock around the magma melt, also?
Depending on what the wall rock is, it can. But much of the heat is removed,
when either the magma escapes and becomes lava, or water carries away the heat
from the wall of the magma chamber (which retards melting). Does "Old Faithful"
ring a bell?
If the heat is
>constant, from whatever source that is heating the magma, then
>eventually, over billions of years, all underground rock should be a
>molten mass, shouldn't it? Why isn't that the case?
Because the Earth is convecting; its internal heat is brought to the surface
and eventually radiated out into space. Hence heat does not build up over
billions of years. Second, the Earth's store of heat sources does diminish with
time.
>
>>http://gfd.gly.bris.ac.uk/gfd-people/heidy.mader/studentships/rheology
>>..html
>
>this is a modeling of volcanic flows. Hey, I've learned a more
>impressive word for the flow behavior of mixtures: rheology. I shall
>use that term from now on in order to communicate more effectively
>with your scientific minds. :-)
Actually, rheology simply meas the mechanical behavior of materials. A fluids
rheology is described in terms of viscosity.
Bigdakine
>Date: 5/24/03 7:54 PM Hawaiian Standard Time
>Message-id: <SUYza.7176$8i4....@nwrddc02.gnilink.net>
Well yes, that is why toothpaste squirting out of a tube is not turbulent.
which for
>toothpaste is significant. My response to Zoe was simplistic, just to
>point out one of the many things she was overlooking. Rheology and
>mixing are complicated subjects about which I only know a smattering.
OK
Seppo Pietikainen
Quite definitely. Magma being released into water create so called
"pillow lava" -formations (and lava tubes), which are quite distinct
from lava flow formations and "bomb" -formations produced when a volcano
erupts above sea level. I suspect the crystallization of lava would also
be quite different due to the different rate of cooling. I think there
are Geologists in the crowd, who would be more than willing to correct
me and expound on the issue.
Seppo P.
Bigdakine
>Date: 5/25/03 3:10 AM Hawaiian Standard Time
>Message-id: <nrc1dv4565il2ece0...@4ax.com>
Not off the top of my head. You could try and see of Frank Spera of UCSB has
any neat stuff on his website. He does alot of magma chamber physcis, or at
least use too.
I did some types of mixing mantle convection mixing numerical simulations some
time ago.
http://archive.ncsa.uiuc.edu/SDG/DigitalGallery/CONVECT.html
>
>It appears to me that what mechanism dominates depends on what length
>scale we are talking about.
Thats quite correct. THe larger the scale the more important stirring is.
Your diffusion mixing time scale is L^2/k where L is the length scale your
interested in, and k is the diffusivity.
In general thermal diffusion is much faster than chemical diffusion at least
for rock materials.
For isochron analysis, it appears to me
>that the length scale is on the order of the size of the chunk of
>magma/rock from which the samples are drawn. That is, if we sample
>form a rock that's 1 m diameter, we're also sampling from a 1 m (or
>perhaps slightly larger) pool of magma that was the source of that
>rock, and the process or processes that dominate at that scale are
>what we want to consider. Who cares if the melt was inhomogeneous
>over a scale of 1,000 km if we're only sampling that 1 m chunk?
Well, you might want to do the analysis for samples obtained from several
places in a large rock body. Because its possible, that on large scales, the
rock may not be well mixed.
Although, all samples should show approximately the same age..
>
>It also appears to me that the appropriate scale for isochron analysis
>sampling ranges from maybe 1 cm at the extreme low end to 100 km at
>the extreme high end, with most of the population falling in the 10 cm
>to 10 m range. Is this reasonable?
>
>What I've managed to dig up suggests that mixing in magmas is
>dominated by diffusion over distances something less than 10 m and
>dominated by mechanical mixing processes over larger distances.
That sounds right. It depends on how much time you allow for the magma to be in
a molten state. Most of the diffusion that takes place will take place when it
is molten. At least thats my guess.
For example allowing the magma to be mostly molten for 10,000 years, this
implies
1,000,000/k = 3.15^11 implies k=3.0E-06cm^2/sec. Sounds in the right ball park
too.
Once the magma has solidified, even if still quite warm, the diffusivity
plummets.
And
>that in turn suggests that diffusion is the dominant factor in the
>homogenization of the source of most isochron samples ... unless I've
>gone wrong somewhere in this chain of reasoning.
Yes, on the scale of the samples collected for isochron analysis.
>
>At <http://tinyurl.com/cm7u> Dr. Sandiford of the University of
>Melbourne says:
>
>"It is an important observation that even at temperatures above the
>blocking temperature, the rate of diffusional equilibration between
>adjacent solids or non-convective, non-mixing liquids is only of the
>order of a meter in 10^7 years.
That implies a diffusivity of ~3E-11, more typical of solid-solid diffusion.
For this reason the establishment of
>large separate geochemical reservoirs like the crust or asthenospheric
>upper mantle, requires both large scale melt extraction and convective
>heat and mass transfer. "
Well that requires more explanation. Maybe I'll give one later when I have more
time.
>
>And Dr. White at Cornell says, in a discussion of the requirements for
>a valid isochron analysis at
>
><http://www.geo.cornell.edu/geology/classes/Geo656/656notes03/656%2003Lec
ture06.pdf>:
>
>"The isotopic composition of the daughter must have been homogeneous
>at the time of the event we wish to date. On a small scale,
>homogenization takes place through diffusion, which, as we have seen,
>is highly temperature dependent. The higher the temperatures obtained
>during the 'event', the more homogenized the system will be. On scales
>larger than 10 meters or so, homogenization can only be achieved
>through convective-driven advective transport. This effectively means
>homogenization requires the presence of a liquid. This might be a
>magma or a hydrous fluid circulating through rocks undergoing
>metamorphism. In any case, both convection and diffusion will be more
>efficient at higher temperatures, so homogenization is more likely to
>be achieved at high temperatures than at low ones. Finally, the larger
>the range in parent/daughter ratios, and hence isotopic composition at
>the time we measure them, the less important will be any initial
>variations in isotopic composition."
Yes. In case that not clear, what he is saying is that if your samples are
strung out along the x-axis on an isochron plot small variations in isotopic
composition due to inhomgeneity will have less effect.
>
>Yet he also says, when discussing the same subject in another document
>for the same course, at
>
><http://www.geo.cornell.edu/geology/classes/Geo656/656notes03/656%2003Lec
ture04.pdf>
>
>"The agent accomplishing isotopic homogenization of a ‘system’ is
>usually diffusion, the rate of which, like other reaction rates,
>increases exponentially with temperature. Diffusion rates will vary
>depending on the element and the properties of the material through
>which the element diffuses. We can nevertheless make the general
>observation that the greater the length scale the greater will be the
>time (or the higher the temperature required) for isotopic
>homogenization to be achieved. For the same temperature and duration
>of a thermal event, diffusion will more readily achieve isotopic
>homogenization on a small scale than on a large one. Thus, if our
>samples or subsystems are 'whole rocks' collected meters or perhaps
>kilometers apart, the event dated will generally be a higher
>temperature one than an event dated by analysis of individual minerals
>from a rock specimen whose scale is only a few centimeters."
On small length scales that is true. However, you will not homogenize a
batholith with diffusion even given the lifetime of the universe.
Harlequin
server.cfl.rr.com:
> On Sat, 24 May 2003 20:55:12 +0000 (UTC), Jon Fleming
> <jo...@fleming-nospam.com> wrote:
>
> snip>
>
>>We have observed creation of D in supernovas.
>
> what is the method used for observing this creation of D? I've always
> wondered how exactly it is done, seeing how distant these novas are
> from hands-on investigation.
Take the light fo the supernova, pass it through a prism (or whatever
the astronomers actually use) to break up the light into its
spectrum. The spectrum will have lines in it. The lines
quite easily show what elements are present. Indeed helium
was first discovered on the Sun before it was found on Earth.
Astronomers had detailed ideas of how these
elements are created in supernovas. These ideas got a big
test with the nearby Supernova 1987a and passed. That supernova
because it was so close was observed in extreme detail.
[snip]
>>>>I do have one question for you. How do you account for the fact that
>>>>in new flows that have been measured we do see that D/Di is the same
>>>>for all samples?
>>>
>>>using the Rb/Sr measure?
>>
>>Yes,and using others. Answer the question.
>
> not until you answer mine: Why would you use a 48-billion-year half
> life to date a new flow that is just a few weeks or years old?
This has been explained to you at least a dozen times. It is called
a test. If Rb-Sr works, then it better produce a date of zero on
freshly solidified lava. And if one to see if D/Di really is the
same in all samples when lava is solified then what better way to
test that then to take a sample of lava that just solidified.
So it now becomes time for you to answer the question. It is one
of the questions that you have been ignorign for a very long time.
That D/Di is the same for all samples in any given lava tested
outright disproves your arguments.
Of course you have also been repeatedly asked about
another way the clocks are tested: by comparing the dates of
one radioactive clock against another radioactive clock and against
the relative geologic timescale. They agree.
I have been trying to convince you to read this one article
since December: http://tinyurl.com/4kjr
Maybe it is time to read it. Jon also posted URLs showing
similiar agreements.
--
Anti-spam: replace "usenet" with "harlequin2"
"...Everybody has opinions: I have them, you have them. And we are all
told from the moment we open our eyes, that everyone is entitled to
his or her opinion. Well, that's horsepuckey, of course. We are not
entitled to our opinions; we are entitled to our _informed_ opinions.
Without research, without background, without understanding, it's
nothing. It's just bibble-babble...."
- Harlan Ellison
Bigdakine
>Date: 5/25/03 5:28 PM Hawaiian Standard Time
>Message-id: <bas1km$307kg$1...@ID-137900.news.dfncis.de>
You're quite right. Lavas that are rapdily quench have a glassy appearance. In
fact sometimes, like obsidian for example, they are essentially natural glass.
On the Big Island of Hawaii, you can find what is sometimes refered as Pele's
hair. These are long thin strands of lava produced by splattering which cool
rapdily forming thin hair-like pieces of glass. Magmas that cool deep
underground are characterized by large crystals, somtimes the size of your fist
or even larger. They are called pegmatites.
John Drayton
> On Sat, 24 May 2003 21:20:39 +0000 (UTC), Matt Silberstein
> <mat...@ix.netcom.com> wrote:
>
> snip>
>
> >Try this. Take some light honey, some black molasses, and some
> >clear corn syrup. Take some clear jars and put layers of each
> >material in the jar. Put one jar in the fridge. Put another on
> >the shelf and let it sit for days/weeks. Take a third jar and put
> >it in a pot. You want water in the pot about as high as the
> >bottom layer, but certainly not as high as the top. Keep the
> >water warm and watch the layer boundary. Let it sit in
> >moderately warm (100 degrees or so) for an hour or so. You could
> >try this with several different water temps. The really
> >interesting thing will come when the material starts to show
> >convection. That is, material on the bottom will be warmer than
> >the top and rise, material on the top will drop. This will mix
> >the three materials quite well.
>
> thanks for the suggestion, Matt, but I don't think that honey would
> represent rock the way chocolate more nearly does.
Based on what measurements?
Do you know the viscosities, densities, and particle
sizes of:
- typical magma in magma chamber
- chocolate at the temperature you heated it to
- honey, molasses and corn syrup at 100 degrees C
If not, on what basis do you say: "I don't think that
honey would represent rock the way chocolate more nearly
does"? Is it just because it matches the results you
want?
> Both chocolate and
> rock are initially in solid forms, and when melted, the observation is
> one of how well material of these relative densities and heating will
> become mixed. Don't you think that the experiment should start out
> with something in solid form first, and then see what happens to its
> material as it melts? If you don't think so, why not?
Why would you care? We already know that not much mixing
will take place when *any* of these substances are solid.
The issue is the behaviour of the substances when they are
liquid: that's when the mixing potentially takes place.
Presumably you would want a substance with the same viscosity,
density, particle size (and probably a bunch of other relevant
stuff I haven't thought of) as liquid rock in the enviroment
where the mixing takes place.
Quite frankly I'd place more credence on observations of
scientists who actually study these phenomena than on the
observations of chocolate being melted in the kitchen.
These "experiments" people suggest are meant to be no more
than illustrations, and my mind boggles that you draw the
conclusions that you do.
Jon Fleming
wrote:
>On Sat, 24 May 2003 20:51:17 +0000 (UTC), Jon Fleming
><jo...@fleming-nospam.com> wrote:
>
>>On Sat, 24 May 2003 zoe_althrop wrote:
>
>snip>
>
>>>viscosity (low versus high) is a critical factor in diffusion, I
>>>think.
>>
>>You are wrong. Viscosity is not a critical factor (in diffusion of
>>atoms and molecules). Density is not a critical factor. The major
>>critical factors are temperature, concentration gradient, and mobility
>>within the material. Although viscosity slightly affects mobility
>>within the material, the material form (solid, liquid, or gas) is the
>>major effect on mobility.
>
>okay, maybe the word "viscosity" is wrong. I take it back. How about
>density/porosity?
That's not particularly good, either. "Density" and "porosity" are
already defined, and their meanings are not quite right for this
situation.
I suggest "mobility", which means exactly one of the major things
which affects diffusion rate.
> I mean, there ARE different rates of diffusion,
>ranging from high to practically nil, right? Surely, there has to be
>SOME factor that is critical to rate of diffusion.
Yep. Actually, there are several ... as I've pointed out many times.
You cannot say in general that just one is critical; you have to
evaluate which ones are important for particular situations or classes
of situations.
Temperature is almost always THE MOST CRITICAL FACTOR. If everything
else happens to be terrifically favorable to diffusion but the
temperature is too low, essentially no diffusion happens. Diffusion
is what's called a "thermally activated process". Each atom/whatever
that moves by diffusion has to be kicked to get it moving, and what
kicks it is the thermal energy that is characterized by the
temperature. As the temperature increases, the atoms/whatever get
kicked harder ... MUCH harder. So, the rate of diffusion goes up a
LOT as the temperature increases.
If the temperature is high enough, then mobility and time are the
other possible critical factors. Which one is more critical depends
on the situation. A lot of diffusion can happen in a material with
low mobility if it happens over a long time, and a lot of diffusion
can happen in a material with high mobility over a very short period
of time..
>So I'll use your term, "material form." But I would define it more
>closely than just solid, liquid, or gas, because liquids come in all
>degrees of densities -- if I'm not misusing that word as well.
>Mobility within liquid jello is much higher than mobility within
>liquid chocolate, from what I have observed?
Maybe, maybe not. Mobility depends on whether or not there are
interactions between what's moving, the nature of those interactions
(if they exist), the availability of "holes" into which what's moving
can move, the amount stuff that has to be pushed out of the way for
the moving thing to move, the difficulty of pushing that stuff out of
the way (there's your viscosity), and some other factors.
>Not to mention liquid
>rock -- a molten material far more dense than either jello or
>chocolate.
Yes, far more dense. That's why I suggested _not_ using density as a
label for mobility. You have already assumed that density (which
means "how much a given size chunk of the stuff weighs") translates
directly to "how mobile stuff is within the material", and that's
wrong. There's a correlation; if all other things are equal, there's
a tendency for diffusion to be slower in denser materials. But it's
not always that way.
Note also the "all other things are equal" phrase. In your comparison
between chocolate and magma, all other things are NOT equal. In fact,
we know that there are two other gigantically different and
significant factors; the temperature and the time. Even if we assume
that mobility of P and D in magma is much less than mobility of white
chocolate in dark chocolate, we cannot immediately conclude that P and
D diffuse less in magma over thousands to billions of years at high
temperatures than you observed in your chocolate experiment at low
temperatures over a few minutes.
>>> See:
>>>
>>>http://www.uic.edu/classes/phys/phys450/MARKO/N004.html
>>>
>>>"Water, and solutions which are mostly water, have a viscosity close
>>>to hwater = 0.01 Poise or 0.001 Pa新ec."
>>
>>Note that this link talks about viscosity in relation to "small
>>particles" in liquids, _not_ in relation to atoms (which are millions
>>of times smaller than the "small particles").
>
>I suppose this means that the behavior of particles in liquids will
>change, depending on size?
Yes, big changes. That's why I assumed that you were talking about
chocolate _molecules_ diffusing in your pot, not chocolate _chunks_
diffusing in your pot. If you are indeed talking about chocolate and
caramel _chunks_ moving around in your pot, then that's another
important flaw in your experiment; chunks of quintillions of molecules
are going to act _very_ different than individual atoms or molecules.
If you are looking for those chunks of chocolate and caramel to move
around in your pot until they are evenly distributed, then your
experiment is much more like diffusion in a solid than it is like
diffusion in a liquid.
>The smaller the particle, the more mobile,
>regardless of the density or porosity of the liquid?
Yes, in diffusion (not necessarily in other types of motion). But
it's not really regardless of the porosity. For particles
significantly larger than an atom or molecule of the liquid, the
liquid looks like a non-porous continuum that has to be physically
shoved out of the way. For atoms, the liquid looks like an extremely
porous collection of loosely connected chunks (atoms).
>>Viscosity is important
>>in "diffusion" of particles that are much much larger than the atoms
>>or molecules of the fluid; viscosity is _not_ very important in
>>diffusion of atoms that are about the same size as the atoms or
>>molecules of the fluid.
>
>is what they're teaching students in science classes today incorrect,
>then?
No, your interpretation is incorrect. For the future, you should
remember that when you see a contradiction it's 99.999% sure that you
are misinterpreting.
>See:
>
>www.ns.msu.edu/drew/nsc204/PagesLabSS/TAosmos.pdf
>
>"The rate of diffusion is proportional to the number of particles of
>solute in solution,,whether they are small molecules, ions or large
>polymeric molecules.. The presence of solute particles lessens the
>water potential of the solution.. The more concentrated a solution
>is,the more negative its water potential"
>
>Sounds to me as if the same principles apply in diffusion, whether the
>particles are large or small.
Yes, the same _fundamental_ principles apply. Note that they didn't
mention viscosity. Viscosity is too "high level", you have to look
into what _causes_ viscosity in order to understand its effects.
Viscosity is a measure of how easily the particles of a liquid slide
past each other. It's affected by the electrochemical interactions
between the liquid particles and the physical shape of the liquid
particles.
Viscosity is important in diffusion of particles that are larger than
the particles (atoms or molecules) of the fluid, because when the
diffusing particle moves it must push a significant amount of the
liquid out of the way, which requires sliding the liquid particles
relative to each other, which is resisted by viscosity.
However, when the diffusing particle is about the same size as the
particles (atoms or molecules) of the fluid, it doesn't have to push
any noticeable amount of fluid out of the way in order to move. It
can move into a space between fluid particles or, in the worst case,
push one or two fluid particles out of the way. Since moving such
small particles doesn't involve moving lots of the fluid particles
relative to each other, viscosity isn't very important.
>>Diffusion in high temperature magma may well be much faster than
>>diffusion in low-temperature water, I don't know off-hand. I do know
>>that temperature is much more important than diffusivity.
>
>maybe you'll take it from Caltech, then?
>
>http://www.gps.caltech.edu/~gurnis/Movies/Science_Captions/stirring.html
>
>"Even after a few billion years of stirring by mantle convection, the
>fluid is still quite heterogneous."
I agree with that statement, but it's irrelevant to this discussion.
That statement refers to large sections of the mantle, hundreds and
thousands of miles across. We're discussing the samples that are
taken for isochron dating. We are talking about chunks of magma that
are a foot (or so) across to maybe 10 yards across at the very
largest. The Caltech statement does not apply to such small chunks.
>>>Indeed, I would expect that the concentration in
>>>magma is such that there is no need for exchange of particles from
>>>areas of higher concentration to areas of lower concentration.
>>
>>Indeed, that's an incredibly silly expectation.
>>
>>Whenever there is a difference in concentration, there is exchange of
>>particles from areas of higher concentration to areas of lower
>>concentration.
>
>that is just it -- is there really a difference in concentration in
>magma?
Well, you are claiming that there is, and you are claiming that it
maintains itself over thousands and millions of years.
Homogeneous equals no difference in concentration.
Heterogenous equals difference in concentration.
If you think there is no difference in concentration on the magma,
then you think the magma is homogeneous, and there's no need to
discuss any farther.
>Rock, as it liquefies, compacts and becomes more dense; IF,
>that is, it follows the same laws as melting chocolate.
Ah, I see; you don't know what concentration means. I'll address that
below. First density ...
Of course, comparing the density of solid chunks of chocolate in a jar
to a fluid mass of chocolate is comparing apples and oranges. What's
meaningful is to compare the density of one chunk of solid chocolate
to one chunk of liquid chocolate.
Almost all materials, including chocolate and rock, become LESS dense
when they liquify. They do NOT become more compact. (Water is one of
the few exceptions; liquid water is denser than ice, that's why ice
floats in water).
When you poured chocolate chunks out of the jar and noted that the
melted chocolate didn't take up a fill jar, you were ignoring the air
that was between the chunks originally. The liquid chocolate _itself_
is less dense that the solid chocolate.
>> The only question is how fast it happens.
>
>billions of years, maybe? More than 4.5 billion?
In some situations, yes. That's one of the reasons isochron dating
works. The P and D atoms are diffusing inside the solid rock as it
sits there waiting for us to sample it ... but the diffusion rate
under those conditions is so low that it would take tens of billions
of years for the rock to homogenize (ignoring for the moment that D is
being continuously created) so, for the most part, we can ignore
diffusion in analyzing solid rocks.
>>The only way that there can be "no need for exchange of particles from
>>areas of higher concentration to areas of lower concentration" is if
>>the concentration is uniform,
>
>exactly. That is what I am saying. That magma is uniform in density.
Yes, but not uniform in concentration. Density is not concentration.
Concentration is the amount of something per unit volume or weight.
For example, the number of D atoms per cubic inch of magma.
If there are more D atoms per cubic inch of magma in one place than
there are in another, there will be diffusion of D atoms between those
two places.
Your _entire_ _argument_ in this thread is based on a claim that there
will be different numbers of P and D atoms per cubic inch in different
parts of the magma, and these differences will remain over millions of
years. That is arguing that there is a difference in concentration
>>and you are trying to argue that the
>>concentration is _not_ uniform.
>
>no, uniformity of density is not the same as homogenous.
Right, but uniformity of concentration _is_ the same as homogeneous.
>I just got
>through experimenting with the chocolate again, this time in a double
>boiler, and the chocolate compacted down to half a jar (where it had
>started out as a full jar) and the white, caramel, and brown chips
>remain in exactly their original formation. But there was no space
>between them anymore. That would be uniform density, I would think --
>no porosity, no further spaces for the chips to go from higher
>concentration to lower concentration. But that is NOT homogeneity.
It is also not uniform density. Uniform density is the same weight
per cubic inch (or whatever) everywhere. Caramel does not weigh the
same as chocolate. Therefore, when you see a liquid in which you can
discern chocolate and caramel parts, it is not uniform density.
Uniform density also does not mean no porosity. When we talk about
diffusion of atoms (which is the appropriate thing to talk about in
relation to isochron dating), there is always some porosity in solids,
lots of porosity in liquids, and tons and tons of porosity in gases.
That's the major reason why (in general and all other things such as
temperature being equal) diffusion is fast in gases, slow in solids,
and intermediate in liquids.
>>>Density matters, as appears to be the case in this link:
>>>
>>>http://chemed.chem.purdue.edu/analyticalreview/chromatography/ratetheory.htm
>>>
>>>"2) flow velocities will vary throughout a packed column and cause
>>>zone spreading,"
>>
>>This does not refer to density ... it does not relate to diffusion in
>>magma or convection or any kind of magmatic mixing.
>
>what does it refer to then? I thought it described a basic principle
>of packing -- or compacting.
That server's down right now, so I can't check. But it probably does
refer to a basic principle of packing or compacting. It definitely
does not refer to diffusion.
>>>You need to have two or more different species in a solution in order
>>>for diffusion to occur, right?
>>
>>Wrong. All you need is one species with different concentrations in
>>different places in the solution.
>
>well, my D (brown chocolate) and Di (white chocolate) and P (caramel
>pieces) were definitely in different concentrations. I did not see
>them gravitating to less concentrated areas for the simple reason that
>there was nowhere for them to go. They were packed too closely to be
>mobile.
Right. And that is another thing that makes your experiment
unrealistic. When atoms are diffusing in solids, there are always
some holes for them to move into. When atoms are diffusing in
liquids, there are always _lots_ of holes for them to move into.
>>>What are the different species in
>>>molten rock?
>>
>>For example:
>>
>>P atoms ... if there is more P in one part of the magma than another,
>>then P atoms will move from the part with more P to the part with less
>>P.
>
>not according to the chocolate experiment. Caramel bits did not move
>to areas where there was less caramel bits.
The caramel bits themselves wouldn't move. Given enough time at a
high enough temperature, the molecules of the caramel would move into
the chocolate and the molecules of the chocolate would move into the
caramel, and eventually the mixture would be a uniform brown glop of
caramel-chocolate.
>>Or D atoms ... if there is more D in one part of the magma than
>>another, then D atoms will move from the part with more D to the part
>>with less D.
>
>same thing here. Brown chocolate remained exactly where it had been
>placed simply because there was no place for it to go. The
>concentration of white and caramel chocolate was such that all spaces
>were filled and brown chocolate (P) had nowhere to go but to remain in
>place.
Yup. And, as I discussed above, that's another reason why your
experiment is a horribly useless model of what's going on in magma.
In magma, from the point of view of P and D atoms, there are lots of
spaces and motion is fairly easy.
>>And this will destroy the straight line relationship on the isochron
>>diagram ... until the process has gone far enough that the melt is
>>almost homogeneous, at which time the points representing the samples
>>will "home in on" a horizontal straight line
>
>you're still describing a boiling, mixing, stirring process in order
>to get a homogenous mixture.
Nope, I've describing diffusion and nothing else.
>None of this happens with chocolate or
>magma.
Still drawing unwarranted conclusions. You've shown that none of this
happens with chocolate and caramel chucks at the temperature you used
over the time that you waited. Nothing more than that.
>>>But in more dense materials, the rate of
>>>diffusion should slow down.
>>
>>Not necessarily. Higher density can slow diffusion somewhat. Higher
>>temperature increases the rate of diffusion terrifically.
>
>not from what I observed in my chocolate-melting experiment.
You didn't test the effect of higher temperature.
Set up pot of the two chocolates that is barely molten. Wait until
there is measurable diffusion between the two types. Record how long
it took.
Set up another pot of chocolate at a significantly higher temperature.
Wait until the same amount of diffusion has taken place. Record how
long it took.
Compare the two times.
>>I really wish you would stop making claims about diffusion that are
>>based only on your fantasies. Diffusion has been studied for many
>>years, and is pretty well understood. There's no reason for making
>>stuff up as you are doing.
>
>I am not making this up. I am telling you what I have observed occurs
>in watery solutions versus chocolate melts.
I believe you saw what you described. What you _are_ making up is the
connection between what you saw and what happens in magma.
>>>And in the case of a magma chamber, all
>>>you have is liquid rock, so you won't have magma dissolving into water
>>>as you would with salt/sugar/red dye in water.
>>
>>You have liquid rock with P atoms and D atoms dissolved in it. You
>>are claiming that there will be different amounts of P and D in
>>different parts of the molten rock (and that will be true for some
>>period of time after the rock melts; the question is for how long it
>>will be true). Under those conditions, P and D atoms will diffuse.
>>The only question is how fast.
>
>billions of years, and the melt will still be heterogenous. See link
>above.
The mantle will still be heterogeneous. The magma pools from which we
draw isochron dating samples will be and are homogeneous.
Think of your pool of chocolates and caramel as the mantle. Dip just
the very tip of a toothpick into a random part of the pool and pick up
the tiniest amount of liquid that you can. That tiny portion is
analogous to the melt from which we're sampling when we do isochron
dating. Is that tiny bit of stuff that you picked up homogeneous?
Almost certainly (unless you deliberately selected it to be
inhomogeneous).
>snip>
>
>>>if you are invoking billions of years for homogeneity to take place,
>>>then your hypothesis of magma being homogenous becomes an
>>>unfalsifiable hypothesis.
>>
>>Nope. Testable predictions can be made, such as recently solidified
>>rock should be homogeneous. It almost always is.
>
>please for live examples from the field. If it takes billions of
>years, and even then the mix remains heterogenous, how can a recent
>solidification be homogenous? Something doesn't fit here.
Yep, your understanding doesn't fit. The mantle is heterogenous even
after billions of years. The melts from which we draw isochron dating
samples are incredibly smaller than the mantle, and are homogeneous.
The heterogeneity of the mantle is irrelevant.
>>Or, samples from a
>>rock should often plot as a straight line on an isochron diagram (your
>>proposed "memory" would not, and you haven't addressed this). They
>>do.
>
>maybe you're talking about Ar-Ar in these cases?
Nope, I'm talking about _all _ types of isochron dating
> I don't know why
>they plot on a straight line. But it wouldn't be because of natural
>homogeneity, I don't think.
Then stop making any claims until you _DO_ have a theory of why they
plot on a straight line.
A possibly valid theory or hypothesis must be consistent with all the
evidence. Your hypothesis of inhomogeneous melts is not consistent
with the evidence.
>>>>If they waited a few years, the allow would be well mixed without
>>>>stirring. They just don't want to wait that long.
>>>
>>>please for support for the idea that the only reason for mixing is
>>>because it would take too long otherwise. And also give support for
>>>the idea that the alloy would become well mixed if allowed to sit for
>>>a few years.
>>
>>EVERYTHING becomes well mixed if allowed to sit for long enough.
>>Diffusion happens in EVERY inhomogeneous mixture. The only question
>>is how long it takes.
>
>not even after billions of years. It is fantasizing to imagine that
>cake batter will mix itself if given enough time. But then I must
>remember that this is the modus operandus of evolutionary theory --
>allow billions of years and surely, a batter of cyanobacteria will
>transform into a civilization of human beings.
Only in your fantasies.
>>Plug the numbers for your molten metal into Fick's law and see what
>>happens. If you don't know Fick's law, you don't know enough to
>>understand the support for the claim.
>
>I've been reading about Fick's law. It has to do with diffusion, but
>usually with gases, not melted solids, as far as I can see (which
>probably is not very far, so correct me if I'm wrong.)
Fick's law applies to diffusion under all circumstances.
>>>snip>
>>>
>>>>Your experiment is irrelevant. The temperature is too low by an order
>>>>of magnitude or two, the time is too short by _many_ _many_ orders of
>>>>magnitude, the constituents are inappropriate because they cannot
>>>>stand up to temperatures that are realistic for underground magma.
>>>
>>>are you saying that one cannot do an experiment on a lower level and
>>>extrapolate the results out to a higher level -- that the laws of
>>>physics do change with time?
>>
>>Certainly you can extrapolate, and certainly the laws of physics
>>appear to be the same over time. But you have to _do_ the
>>extrapolation, and you have to do it using the laws of physics, and
>>you have to do it using physically valid formulas.
>
>I am doing it with real-world observation in this chocolate
>experiment. Somebody else can draw up the formulas as to WHY or HOW
>the chocolate pieces maintain their original positions, but the data
>is that temperature does not cause homogeneity in a non-aqueous
>solution.
No, the data is that you did not observe homogeneity in chocolate and
caramel at the temperature you chose over a few minutes time.
If you wish to claim that has relevance to magma at much higher
temperature over much longer times, it is _your_ responsibility to
evaluate the effect of the different temperature, _your_
responsibility to evaluate the effect of the longer time, _your_
responsibility to evaluate the effect of the different materials.
>>You didn't do any extrapolation. You didn't make any use of the laws
>>of physics (the effect of temperature on diffusion rate, the effect of
>>time on the amount of mixing by diffusion).
>
>I am telling you what I observed from a real-world experiment.
Yes, and you are claiming that what you observed from a real-world
experiment predicts what will happen in magma under terrifically
different conditions.
Either stop claiming that your experiment has anything to do with
magma or ESTABLISH THE CONNECTION!
> If
>someone wants to draw up the equations that describe the results that
>I have observed, they are welcome to do so.
>
>>Go ahead, do your extrapolation. Show your calculations. Remember the
>>exponential dependence of diffusion rate on temperature.
>
>I don't have to do this. You can. I am giving you hard observed data
>to work with.
You've presented no data. You only took one data point, and you
didn't write down any numbers
>>Doing an experiment under one set of conditions and claiming that the
>>result applies directly to wildly different conditions is not
>>extrapolation, it's fantasizing.
>
>labeling the two conditions "wildly different" doesn't make it so. I
>think there is a lot of similarity.
Nobody cares what you _think_. It's your responsibility to
_establish_ that there is a lot of similarity, and _establish_ exactly
how much similarity there is.
You think that 10 minutes is similar to 500,000,000,000 minutes?
Exactly _why_ do you think that?
You thing that 200 degrees F is similar to 5,000 degrees F? Exactly
why do you think that?
>snip>
>
>>>water in chocolate? Chocolate is a mixture of cocoa paste, cocoa
>>>butter, and sugar. What water?
>>
>>There's water in there, 0.5 to 1.6% of the chocolate, assuming you
>>have stored it under very dry conditions. See
>><http://www.tis-gdv.de/tis_e/ware/lebensmi/schoko/schoko.htm#feuchte>.
>>
>>You have an unfortunate tendency to assume you know something about
>>many areas in which you are totally ignorant.
>
>I didn't make that up. I got it from a site that neglected to mention
>the 0.5 to 1.6%. I stand corrected. But this is still not sufficient
>water to make a difference in how the chocolate melts.
I never said it made a difference in how the chocolate melts. That
water is enough to cause bubbling and an "explosion" when the water
boils and turns to steam, which is what I was talking about.
>>>>The only materials suitable for a useful experiment are molten rocks.
>>>>You don't have the facilities to run a realistic experiment. You
>>>>can't get temperatures high enough, you can't wait long enough, you
>>>>don't have the instrumentation required to make accurate measurements
>>>>of what happens over shorter times.
>>>
>>>well, so much for extrapolation in science; it is a useless tool -- as
>>>far as Jon is concerned, anyway.
>>
>>You haven't even attempted an extrapolation.
>>
>>Extrapolation is a very dangerous tool, because it's prone to error,
>>and the farther you extrapolate the more likely an error is; but it's
>>a useful tool when used properly.
>
>you mean, when evolutionary theory extrapolates into the billions of
>years, then it is more likely to be in error, too?
Evolutionary theory usually doesn't _extrapolate_, more often it
_interpolates_, which is safer and more reliable.
But, when evolutionary theory (or any scientific theory) _does_
extrapolate, yes, it's more likely to be in error. Scientists know
this and acknowledge it. Creationists tend to ignore it; there are
lots of examples of obviously invalid extrapolations presented in the
creationist "literature".
Of course, no "evolutionist", indeed no scientist, would extrapolate
10 minutes worth of observation to thousands or millions or billions
of years; that's _way_ too far an extrapolation o be realistic in
_any_ circumstances.
Jon Fleming
wrote:
>On Sat, 24 May 2003 20:55:12 +0000 (UTC), Jon Fleming
><jo...@fleming-nospam.com> wrote:
>
>snip>
>
>>We have observed creation of D in supernovas.
>
>what is the method used for observing this creation of D? I've always
>wondered how exactly it is done, seeing how distant these novas are
>from hands-on investigation.
Do you ever remember _anything_ from one week to another?
Spectroscopy. We discussed this at length last June, July, and
August. For example, in <http://tinyurl.com/cov8>
>>>then there could be no original oldD at the first
>>>formation of the earth
>>
>>Wrong, as has been pointed out many times before.
>
>pointed out via assertions?
>
>>
>><snip>
>>
>>>>I do have one question for you. How do you account for the fact that
>>>>in new flows that have been measured we do see that D/Di is the same
>>>>for all samples?
>>>
>>>using the Rb/Sr measure?
>>
>>Yes,and using others. Answer the question.
>
>not until you answer mine: Why would you use a 48-billion-year half
>life to date a new flow that is just a few weeks or years old?
To check the validity of our assumptions, something scientists do
regularly. How do you account for the fact that in new flows that
have been measured we do see that D/Di is the same for all samples?
Howard Hershey
muz...@aol.com wrote on 5/24/03 6:37 PM:
> On Fri, 23 May 2003 08:52:48 +0000 (UTC), e...@FAKEADRESS.com (Eric
> Rowley) wrote:
>
> snip>
>
>> But mixing by convection _is_ considered a regular activity in
>> underground magma chambers.
>
> true, but the results of this kind of mixing won't do the job. At
> least not according to this site:
>
> http://www.gps.caltech.edu/~gurnis/Movies/Science_Captions/stirring.html
>
> "Even after a few billion years of stirring by mantle convection, the
> fluid is still quite heterogneous."
>
>>> Heat is not sufficient to produce a
>>> homogenous mixture (see first link above)
>>
>> Have you considered the posibility that heat alone just isn't
>> fast enough for their purposes?
>
> prolonged heat results in burning, dehydration, or pressure buildup.
> It does not contribute to homogeneity, not even by convection currents
> within magma. At least this is what I am learning so far.
>
>> Any liquid that is heated from below and cooled at the top,
>> like molten rock in a magma chamber or soup on your stove,
>> will tend to form convection cells where hot liquid wells
>> up in the center and (slightly) cooler liquid sinks down
>> around the edges.
>
> boiling, you mean?
No. Convection. Boiling requires gas formation (steam bubbles in the case
of boiling water). *Before* water boils, if you look down, you will see
"streams" (due to changes in refractive index because of density differences
in hot and colder water) of water moving up from the bottom. These streams
are the convection cells.
> A magma chamber and soup on a stove carry entirely
> different variables, so I don't think it is fair to compare the
> boiling of soup in an open pot on a stove to the pressurized contents
> of a magma chamber. That is why I have, in the past, used a pressure
> cooker for comparison, because that would have more in common with a
> magma chamber.
Why don't you use the pressure cooker for your chocolate chip experiment?
The advantage is that there will be no boiling, but the temperature
differential will be high enough, from bottom to top, to result in
convection currents.
>
> snip>
>
>>> Question: What prevents magma from burning up underground?
>>
>> Lack of burnable substances!
>> Anything burnable would have been converted to oil, or natural
>> gas way before it reached the melting point of the rest of the rock.
>
> that's interesting. Let me follow up your reply with some research on
> your pointers. Maybe you're right, maybe not.
>
> snip>
>
>> An ounce or so of chocolate in a microwave oven for a minute or
>> two doesn't act the same as a million tons of magma in a magma
>> chamber over millions of years?
>
> the parallel is a relative one. Once the substance (rock or
> chocolate) reaches the point of melting (5000 degrees or 250 degrees),
> the characteristics and behavior of the liquefied rock/chocolate
> should be the same. Boiling is boiling, regardless of what
> temperature it takes to bring a substance to a boil. Same for
> melting.
>
>> Your sample melts may well have been too small to form convection
>> currents.
>
> if by "convection currents" you mean small bubbles percolating up
> through the melted chocolate, then I did see convection currents. The
> chocolate, however, remained pretty much in place, with a probably
> somewhat wider zone.
>
>> Besides convection currents are caused by _uneven_ heating, a
>> microwave oven is designed to heat things evenly from all sides
>> (even the inside).
>
> actually, no, the microwave heats from outside in. If you take the
> item out too soon, the edges will be hotter than the middle.
Actually, microwaves will heat food to whatever depth the microwaves can
reach in the food (usually the top few centimeters). The material in the
center gets heated by conducted heat from the microwave heated surface
layer.
>
>> How much time did they have for difusion to occur before
>> they started to burn or leak?
>
> not much, granted that.
>
>>> The linear relationship between D/Di and P/Di is retained
>>> throughout a melt,
>>
>> Where does the original linear relationship come from in your
>> scenario?
>
> the original linear relationship of D/Di to P/Di would come from an
> original formation in which there was no oldD present. Since D
> originates from P, there is no reason to imagine that there was a time
> when D originated all by itself, without the aid of P.
In the initial outburst of a supernova, but that is besides the point.
> Therefore, the
> moment that P begins to decay to D, the linear relationship would be
> established, and it is this linear relationship that becomes more and
> more fuzzied by remelts. The more remelts, the younger the rock
> appears to be. The less remelts, the older the rock shows up to be.
Until, of course, the present time, when examination of lavas from
historically modern volcanos show constant D/Di uncorrelated with P/Di
rather than positive linear relationships. There are a number of problems
with this.
It cannot explain the general consistency of isochron and argon dating
methods. In the argon method, any time you get a melt you necessarily
generally lose all or nearly all the "retained" argon *in the local area*
due to the nature of the noble gasses in a liquid.
>
>> And how is it retained? Are you suggesting that _no_ mixing
>> takes place in magma chambers or volcanic eruptions?
>
> the linear relationship is retained because very little mixing occurs
> in magma chambers. What serves to fuzzy the relationship is continual
> remelts, leaching, contamination, and other environmental variables
> that interfere with the original linear relationship. Given sufficient
> time, I'm betting that there won't be any rocks left that would
> produce an acceptable isochron.
If your analysis were true, then the fit of the data points to the isochron
line (number of significant figures to which the slope is accurate) of
isochrons that falsely indicate recent dates should be significantly worse
than isochrons that date much older, since only the old isochrons
'accurately' reflect a situation in which D/Di actually started out as
constant and new D has been added by decay of local P. All more modern
isochrons necessarily will have undergone "fuzzification" by random and
local events, producing data that fit a 'line' less well than rocks that
have ungone fewer "fuzzification" events. OTOH, if the clock *resets* at
each melt (D/Di is constant at each melt), then the accuracy of the slope
(how well the data fits the line) will *not* be a function of the age of the
rocks. Do you want to bet on which is observed?
>
>> It seems to me that any mixing at all would destroy the original
>> relationship and only total mixing would create a new linear
>> relationship.
>
> if the mixing is negligible, the original relationship only becomes
> fuzzied (scatter). It is only total mixing that would create a new
> linear relationship, but I am submitting that total mixing is not a
> reality.
No. It is only total mixing *within* the source of the rocks in question
(not the entire mantle, but only the lava within the magma chamber) that
produces samples with a slope of zero upon solidification.
>
>>> with some fuzzying of the borders (scatter),
>>> sufficient to reduce the apparent age of the rock with each
>>> succeeding remelt, but insufficient to bring the slope back to
>>> zero at time of solidification.
>>
>> So where are all the measurements of nonzero isochrons from fresh
>> lava?
>
> if freshly extruded lava is found to be nonzero via Rb/Sr isochron, it
> would mean that the premelt linear relationship is still visible. But
> why would anybody use Rb/Sr to date new lava? It is useless to use
> such long-lived isotopes (48 billion years) to check for recent age.
It is also crucial to do so to check the assumption that new melts have a
constant D/Di uncorrelated with P/Di. That is why people have actually
measured such freshly extruded lavas. It allows one to check the general
validity of the assumptions of isochron dating. It is indeed useless to use
such long-lived isotopes to check for recent age (because the true age will
be wrapped within an error range of several millions of years). It is not
useless to use such measurements as a check of the assumptions of the
method. And guess what?
>
>>> okay, Jon & Company, I have a lot more chocolate morsels on hand.
>>> Are there any other or better experiments that can be done with
>>> them?...quickly, please, before they disappear down my gullet.
>>
>> Yes, take a larger quantity (at least a pound) and heat _gently_
>> in a pot, the best way is in a waterbath (a small pot in a big pot
>> with a little water in between), so that the chocolate is melted but
>> not burned.
>>
>> After a while you should get convection currents, the chocolate
>> should well up in the center of the pot and sink down at the edges
>> causing mixing.
>
> thanks for the suggestion, Eric, but this experiment would not match
> up the variables between melted chocolate and magma. One is in an
> open pot where boiling can take place. The other is in a pressurized
> magma chamber where boiling does not take place.
Then put the bloody chips in a pressure cooker for several hours and let it
solidify. That would be better anyway because the differential between
melting temperature and the temperature in the pressure cooker will be
greater.
>
> ----
> zoe
>
Jon Fleming
(Bigdakine) wrote:
>>Subject: Re: Results of an experiment
>>From: Jon Fleming jo...@fleming-nospam.com
<snip>
>>Do you have any Web or library references that are a good overview of
>>this, or could you perhaps discuss it some more?
>
>Not off the top of my head. You could try and see of Frank Spera of UCSB has
>any neat stuff on his website. He does alot of magma chamber physcis, or at
>least use too.
Ah, <http://magma.geol.ucsb.edu/>. I'll rummage around there some.
Thanks.
>I did some types of mixing mantle convection mixing numerical simulations some
>time ago.
>
>http://archive.ncsa.uiuc.edu/SDG/DigitalGallery/CONVECT.html
Pretty cool. Why do I think that Zoe wouldn't be impressed? {grin}
> For isochron analysis, it appears to me
>>that the length scale is on the order of the size of the chunk of
>>magma/rock from which the samples are drawn. That is, if we sample
>>form a rock that's 1 m diameter, we're also sampling from a 1 m (or
>>perhaps slightly larger) pool of magma that was the source of that
>>rock, and the process or processes that dominate at that scale are
>>what we want to consider. Who cares if the melt was inhomogeneous
>>over a scale of 1,000 km if we're only sampling that 1 m chunk?
>
>Well, you might want to do the analysis for samples obtained from several
>places in a large rock body. Because its possible, that on large scales, the
>rock may not be well mixed.
>
>Although, all samples should show approximately the same age.
Agreed. E.g. I stumbled across the abstract of "Near Kilometer Scale
Strontium Isotopic Homogenization in Meta-Granites, North-Central New
Mexico", by Odom, Platero, & Sachi-Kocher (at
<http://www.magnet.fsu.edu/publications/1999annualreview/pdfs/appen.pdf>),
which is talking about metamorphism but I think it's applicable:
"These data point to an extraordinary scale of strontium isotopic
homogenization in non-convecting rocks. The coincidence (within
uncertainties) of the isochron ages indicate that during the regional
metamorphic episode, the strontium in these rocks were isotopically
homogenized over a scale of 10 to 400 meters (the range of distance
between samples at a single locality). The distinctly different
87Sr/86Sr initial ratio values of the three isochrons indicate that
the scale of isotopic homogenization was less than a few kilometers
(the minimum distance between collecting localities)."
<snip>
>>What I've managed to dig up suggests that mixing in magmas is
>>dominated by diffusion over distances something less than 10 m and
>>dominated by mechanical mixing processes over larger distances.
>
>That sounds right. It depends on how much time you allow for the magma to be in
>a molten state. Most of the diffusion that takes place will take place when it
>is molten. At least thats my guess.
>
>For example allowing the magma to be mostly molten for 10,000 years, this
>implies
>
>1,000,000/k = 3.15^11 implies k=3.0E-06cm^2/sec. Sounds in the right ball park
>too.
>
>Once the magma has solidified, even if still quite warm, the diffusivity
>plummets.
Oh, yes, I'm talking essentially exclusively about liquid.
<snip>
>>At <http://tinyurl.com/cm7u> Dr. Sandiford of the University of
>>Melbourne says:
>>
>>"It is an important observation that even at temperatures above the
>>blocking temperature, the rate of diffusional equilibration between
>>adjacent solids or non-convective, non-mixing liquids is only of the
>>order of a meter in 10^7 years.
>
>That implies a diffusivity of ~3E-11, more typical of solid-solid diffusion.
Yes, that seemed a bit low for liquid diffusion
> For this reason the establishment of
>>large separate geochemical reservoirs like the crust or asthenospheric
>>upper mantle, requires both large scale melt extraction and convective
>>heat and mass transfer. "
>
>Well that requires more explanation. Maybe I'll give one later when I have more
>time.
If you do have the time, I would appreciate the explanation.
>>And Dr. White at Cornell says, in a discussion of the requirements for
>>a valid isochron analysis at
>>
>><http://www.geo.cornell.edu/geology/classes/Geo656/656notes03/656%2003Lec
>ture06.pdf>:
>>
>>"The isotopic composition of the daughter must have been homogeneous
>>at the time of the event we wish to date. On a small scale,
>>homogenization takes place through diffusion, which, as we have seen,
>>is highly temperature dependent. The higher the temperatures obtained
>>during the 'event', the more homogenized the system will be. On scales
>>larger than 10 meters or so, homogenization can only be achieved
>>through convective-driven advective transport. This effectively means
>>homogenization requires the presence of a liquid. This might be a
>>magma or a hydrous fluid circulating through rocks undergoing
>>metamorphism. In any case, both convection and diffusion will be more
>>efficient at higher temperatures, so homogenization is more likely to
>>be achieved at high temperatures than at low ones. Finally, the larger
>>the range in parent/daughter ratios, and hence isotopic composition at
>>the time we measure them, the less important will be any initial
>>variations in isotopic composition."
>
>Yes. In case that not clear, what he is saying is that if your samples are
>strung out along the x-axis on an isochron plot small variations in isotopic
>composition due to inhomgeneity will have less effect.
It's clear to me. Zoe could never comprehend it.
<snip>
Jon Fleming
(Bigdakine) wrote:
>Actually, rheology simply means the mechanical behavior of materials.
Er, flow and deformation of materials. Mechanical behavior subsumes
more than just that.
>A fluids
>rheology is described in terms of viscosity.
Jon Fleming
wrote:
>On Sat, 24 May 2003 21:20:39 +0000 (UTC), Matt Silberstein
Not necessarily. Either way is just as valid, one way may be more
convenient than the other.
Of course, if you're trying to model what happens in magma over
thousands of years, nothing you can do in your kitchen is valid. You
can demonstrate what _might_ happen in various kitchen experiments,
but nothing you can do in your kitchen will establish what _does_
happen.
>If you don't think so, why not?
Because the rate of diffusion at any time depends only on the current
conditions (concentrations of diffusing species, temperature,
mobility, ...). It doesn't matter how those conditions arose.
Howard Hershey
muz...@aol.com wrote on 5/26/03 1:42 AM:
> On Sat, 24 May 2003 20:51:17 +0000 (UTC), Jon Fleming
> <jo...@fleming-nospam.com> wrote:
>
>> On Sat, 24 May 2003 zoe_althrop wrote:
>
> snip>
>
>>> viscosity (low versus high) is a critical factor in diffusion, I
>>> think.
>>
>> You are wrong. Viscosity is not a critical factor (in diffusion of
>> atoms and molecules). Density is not a critical factor. The major
>> critical factors are temperature, concentration gradient, and mobility
>> within the material. Although viscosity slightly affects mobility
>> within the material, the material form (solid, liquid, or gas) is the
>> major effect on mobility.
>
> okay, maybe the word "viscosity" is wrong. I take it back. How about
> density/porosity? I mean, there ARE different rates of diffusion,
> ranging from high to practically nil, right? Surely, there has to be
> SOME factor that is critical to rate of diffusion.
>
> So I'll use your term, "material form." But I would define it more
> closely than just solid, liquid, or gas, because liquids come in all
> degrees of densities -- if I'm not misusing that word as well.
> Mobility within liquid jello is much higher than mobility within
> liquid chocolate, from what I have observed? Not to mention liquid
> rock -- a molten material far more dense than either jello or
> chocolate.
Actually, that would depend upon temperature. Syrup and molasses flow much
faster (more like water) as temperature increases above that needed to melt
them. Molasses in February flows less readily than molasses heated to 100°
C. Melted rock flows more readily at higher temperature than at lower
temperature.
Most of the ions and atoms that make up rock are, relatively speaking, small
molecules rather than large polymeric molecules. So I would expect
diffusion of such molecules/ions to be relatively rapid in liquid rock.
>
> snip>
>
>> Diffusion in high temperature magma may well be much faster than
>> diffusion in low-temperature water, I don't know off-hand. I do know
>> that temperature is much more important than diffusivity.
>
> maybe you'll take it from Caltech, then?
>
> http://www.gps.caltech.edu/~gurnis/Movies/Science_Captions/stirring.html
>
> "Even after a few billion years of stirring by mantle convection, the
> fluid is still quite heterogneous."
We are interested in convection within a magma chamber *combined* with
diffusion.
>
>>> Indeed, I would expect that the concentration in
>>> magma is such that there is no need for exchange of particles from
>>> areas of higher concentration to areas of lower concentration.
>>
>> Indeed, that's an incredibly silly expectation.
>>
>> Whenever there is a difference in concentration, there is exchange of
>> particles from areas of higher concentration to areas of lower
>> concentration.
>
> that is just it -- is there really a difference in concentration in
> magma? Rock, as it liquefies, compacts and becomes more dense; IF,
> that is, it follows the same laws as melting chocolate.
No. Rock becomes less dense in the liquid form (water is unusual in that it
does indeed become more dense as a liquid). What you are noticing is the
loss of air pockets between rocks as they melt. That is what causes the
compaction. The 'density' of the original includes both the chips *and* the
air pockets.
>
>> The only question is how fast it happens.
>
> billions of years, maybe? More than 4.5 billion?
>
>> The only way that there can be "no need for exchange of particles from
>> areas of higher concentration to areas of lower concentration" is if
>> the concentration is uniform,
>
> exactly. That is what I am saying. That magma is uniform in density.
>
>> and you are trying to argue that the
>> concentration is _not_ uniform.
>
> no, uniformity of density is not the same as homogenous. I just got
> through experimenting with the chocolate again, this time in a double
> boiler, and the chocolate compacted down to half a jar (where it had
> started out as a full jar) and the white, caramel, and brown chips
> remain in exactly their original formation. But there was no space
> between them anymore. That would be uniform density, I would think --
> no porosity, no further spaces for the chips to go from higher
> concentration to lower concentration. But that is NOT homogeneity.
Use a pressure cooker.
Convection cells will form in a magma chamber, just as they will when the
source of heat is on the bottom of a pressure cooker.
>
> snip>
>
>>> But in more dense materials, the rate of
>>> diffusion should slow down.
>>
>> Not necessarily. Higher density can slow diffusion somewhat. Higher
>> temperature increases the rate of diffusion terrifically.
>
> not from what I observed in my chocolate-melting experiment.
>
>> I really wish you would stop making claims about diffusion that are
>> based only on your fantasies. Diffusion has been studied for many
>> years, and is pretty well understood. There's no reason for making
>> stuff up as you are doing.
>
> I am not making this up. I am telling you what I have observed occurs
> in watery solutions versus chocolate melts.
And at higher temperatures, chocolate will flow more readily (more like
water).
>
>>> And in the case of a magma chamber, all
>>> you have is liquid rock, so you won't have magma dissolving into water
>>> as you would with salt/sugar/red dye in water.
>>
>> You have liquid rock with P atoms and D atoms dissolved in it. You
>> are claiming that there will be different amounts of P and D in
>> different parts of the molten rock (and that will be true for some
>> period of time after the rock melts; the question is for how long it
>> will be true). Under those conditions, P and D atoms will diffuse.
>> The only question is how fast.
>
> billions of years, and the melt will still be heterogenous. See link
> above.
>
> snip>
>
>>> if you are invoking billions of years for homogeneity to take place,
>>> then your hypothesis of magma being homogenous becomes an
>>> unfalsifiable hypothesis.
>>
>> Nope. Testable predictions can be made, such as recently solidified
>> rock should be homogeneous. It almost always is.
>
> please for live examples from the field. If it takes billions of
> years, and even then the mix remains heterogenous, how can a recent
> solidification be homogenous? Something doesn't fit here.
Your model doesn't fit reality.
>
>> Or, samples from a
>> rock should often plot as a straight line on an isochron diagram (your
>> proposed "memory" would not, and you haven't addressed this). They
>> do.
>
> maybe you're talking about Ar-Ar in these cases? I don't know why
> they plot on a straight line. But it wouldn't be because of natural
> homogeneity, I don't think.
And, according to your model, the more melts and re-melts that a rock has is
what determines its slope (the fewer, the more the sample looks like a true
isochron). The problem is that this implies that the data from the rock
samples should have a worse and worse fit to the isochron line as the sample
appears to be younger (has undergone more "fuzzification"). There should be
an increasingly poor fit of the data to the line as the slope decreases
because of the random nature of local "fuzzification" events.
And I am telling you that your observational model conflicts with the
evidence of what is seen in the real lavas you want to extrapolate to. In
science that means that there is something wrong with your model, not with
reality.
>
>> Go ahead, do your extrapolation. Show your calculations. Remember the
>> exponential dependence of diffusion rate on temperature.
>
> I don't have to do this. You can. I am giving you hard observed data
> to work with.
>
>> Doing an experiment under one set of conditions and claiming that the
>> result applies directly to wildly different conditions is not
>> extrapolation, it's fantasizing.
>
> labeling the two conditions "wildly different" doesn't make it so. I
> think there is a lot of similarity.
>
> snip>
>
>>> water in chocolate? Chocolate is a mixture of cocoa paste, cocoa
>>> butter, and sugar. What water?
>>
>> There's water in there, 0.5 to 1.6% of the chocolate, assuming you
>> have stored it under very dry conditions. See
>> <http://www.tis-gdv.de/tis_e/ware/lebensmi/schoko/schoko.htm#feuchte>.
>>
>> You have an unfortunate tendency to assume you know something about
>> many areas in which you are totally ignorant.
>
> I didn't make that up. I got it from a site that neglected to mention
> the 0.5 to 1.6%. I stand corrected. But this is still not sufficient
> water to make a difference in how the chocolate melts.
Chocolate, unlike rock, is composed of molecules that are large polymers
(the fats of cocoa butter, for example) rather than small molecules and
ions. As you noted, there is a large effect of both temperature and the
size of the molecules. So I would say that chocolate is the wrong model to
examine the rates of diffusion of ions in liquid rock. But the real problem
is that direct observation of real lavas always trumps "model-building"
extrapolations.
>
>>>
>>> snip>
>>>
>>>> The only materials suitable for a useful experiment are molten rocks.
>>>> You don't have the facilities to run a realistic experiment. You
>>>> can't get temperatures high enough, you can't wait long enough, you
>>>> don't have the instrumentation required to make accurate measurements
>>>> of what happens over shorter times.
>>>
>>> well, so much for extrapolation in science; it is a useless tool -- as
>>> far as Jon is concerned, anyway.
>>
>> You haven't even attempted an extrapolation.
>>
>> Extrapolation is a very dangerous tool, because it's prone to error,
>> and the farther you extrapolate the more likely an error is; but it's
>> a useful tool when used properly.
>
> you mean, when evolutionary theory extrapolates into the billions of
> years, then it is more likely to be in error, too?
Which is why extrapolations billions of years into the past are always
tentative. All we can say is that the observations are consistent with the
expectations of evolutionary theory and no other known scientific
explanation exists that is consistent with the evidence. That said, one can
always posit magical solutions that can explain anything. Those
explanations are not scientific.
>
> snip>
>
> ----
> zoe
>
Jon Fleming
(John Drayton) wrote:
>> Both chocolate and
>> rock are initially in solid forms, and when melted, the observation is
>> one of how well material of these relative densities and heating will
>> become mixed. Don't you think that the experiment should start out
>> with something in solid form first, and then see what happens to its
>> material as it melts? If you don't think so, why not?
>
>Why would you care? We already know that not much mixing
>will take place when *any* of these substances are solid.
I bet she thinks that the diffusion rate depends on whether you
started with a liquid or a solid.
Jon Fleming
<hers...@indiana.edu> wrote:
>in article 3ed16e41....@news-server.cfl.rr.com, zoe_althrop at
>muz...@aol.com wrote on 5/26/03 1:42 AM:
>
>> On Sat, 24 May 2003 20:51:17 +0000 (UTC), Jon Fleming
>> <jo...@fleming-nospam.com> wrote:
<snip>
>>> Diffusion in high temperature magma may well be much faster than
>>> diffusion in low-temperature water, I don't know off-hand. I do know
>>> that temperature is much more important than diffusivity.
>>
>> maybe you'll take it from Caltech, then?
>>
>> http://www.gps.caltech.edu/~gurnis/Movies/Science_Captions/stirring.html
>>
>> "Even after a few billion years of stirring by mantle convection, the
>> fluid is still quite heterogneous."
>
>We are interested in convection within a magma chamber *combined* with
>diffusion.
<snip>
>> please for live examples from the field. If it takes billions of
>> years, and even then the mix remains heterogenous, how can a recent
>> solidification be homogenous? Something doesn't fit here.
>
>Your model doesn't fit reality.
Indeed it doesn't ... but what she's asking is "how can we expect that
isochron samples came from homogeneous melts when we know that the
entire mantle isn't homogeneous?"
Of course, the answer is that the melts from which the isochron
samples are drawn are much much much smaller than the mantle, and are
locally homogeneous.
<snip>
Eric
HH> in article 3ecfb96c....@news-server.cfl.rr.com,
<snip>
HH> Why don't you use the pressure cooker for your chocolate chip
HH> experiment?
That does NOT sound like a good idea to me.
HH> The advantage is that there will be no boiling,
? A pressure cooker doesn't stop things from boiling!
(Where do you think the pressure comes from?)
It just raises the temperature at which boiling occurs.
And therefore allows one to cook things in water at a
temperature of more than 100 degrees Celsius.
Unless the pressure cooker contains enough water it is
unlikely to prevent burning either.
If you used a pressure cooker as the outer pot in a double
boiler you would have a safe setup that might even work.
HH> but
HH> the temperature differential will be high enough, from bottom
HH> to top, to result in convection currents.
If the pressure cooker is full of hot steam, as it is designed
to be you may not get a good temperature differental.
<snip>
Eric
<My domain is rixtele>
zoe_althrop
<hers...@indiana.edu> wrote:
snip>
zoe wrote:
>> the original linear relationship of D/Di to P/Di would come from an
>> original formation in which there was no oldD present. Since D
>> originates from P, there is no reason to imagine that there was a time
>> when D originated all by itself, without the aid of P.
>
>In the initial outburst of a supernova, but that is besides the point.
why is this besides the point? It is exactly the point that needs to
be explored.
What can you tell me about how supernovas produce D without initial P?
>> Therefore, the
>> moment that P begins to decay to D, the linear relationship would be
>> established, and it is this linear relationship that becomes more and
>> more fuzzied by remelts. The more remelts, the younger the rock
>> appears to be. The less remelts, the older the rock shows up to be.
>
>Until, of course, the present time, when examination of lavas from
>historically modern volcanos show constant D/Di uncorrelated with P/Di
>rather than positive linear relationships.
if a historically modern volcano shows a supposedly constant D/Di, or
zero slope, this would be explained by my hypothesis as the result of
sufficient remelt scatter that would bring the slope down to the zero
range. At least for those rocks that have been through sufficient
remelts to lose the positive slope.
>There are a number of problems with this. It cannot explain the general
>consistency of isochron and argon dating methods. In the argon method,
>any time you get a melt you necessarily generally lose all or nearly all
>the "retained" argon *in the local area* due to the nature of the noble
>gasses in a liquid.
I would expect all dating methods to agree simply because the
accumulated decay product, calculated by ANY method, should reflect
the same length of time the rocks have been present on this earth --
within a general age range, plus or minus X number of years that would
cover contamination, leaching, and the like. The problem is not with
matching ages from different methods, but with the application of
those ages to the history of the rocks.
These ages should not reflect time since solidification. They reflect
should reflect accumulated decay product that gets translated into a
calculated equivalent of years. Applying the calculated number of
years to time since solidification is an unjustified assumption, imo.
snip>
>> the linear relationship is retained because very little mixing occurs
>> in magma chambers. What serves to fuzzy the relationship is continual
>> remelts, leaching, contamination, and other environmental variables
>> that interfere with the original linear relationship. Given sufficient
>> time, I'm betting that there won't be any rocks left that would
>> produce an acceptable isochron.
>
>If your analysis were true, then the fit of the data points to the isochron
>line (number of significant figures to which the slope is accurate) of
>isochrons that falsely indicate recent dates should be significantly worse
>than isochrons that date much older, since only the old isochrons
>'accurately' reflect a situation in which D/Di actually started out as
>constant and new D has been added by decay of local P. All more modern
>isochrons necessarily will have undergone "fuzzification" by random and
>local events, producing data that fit a 'line' less well than rocks that
>have ungone fewer "fuzzification" events. OTOH, if the clock *resets* at
>each melt (D/Di is constant at each melt), then the accuracy of the slope
>(how well the data fits the line) will *not* be a function of the age of the
>rocks. Do you want to bet on which is observed?
that makes sense. What about this link?
http://www.cosis.net/abstracts/EAE03/04785/EAE03-J-04785.pdf
"Rb-Sr whole rock data points obtained from the Tres Piedras Granite
yields a distinct isochron for each outcrop sampled. The age and
apparent initial Sr87/Sr86 ratios of the Tres Piedras Granite outcrops
(approximately 50 square meters of collecting area at each exposure)
are as follows: Tres Piedras Granite Type Locality: 1493 +/- 21 Maand
0.7183 +/- 0.0006; Tres Piedras Granite – Tusas River Canyon: 1501 +/-
44 Maand 0.7145 +/- 0.0013; and Tres Piedras Granite – Tusas Mountain:
1661 +/- 17 Maand 0.7102 +/- 0.0071."
correct me if I'm wrong, but it appears that the younger ages carry
increased scatter:
0.7183 +/- 0.0006
0.7145 +/- 0.0013
0.7102 +/- 0.0071
The piece says that metamorphism could be the cause for the
differences, but why not an alternate supposition that, with each
remelt, the scatter increases?
here's another link.
http://www.lpi.usra.edu/meetings/lpsc97/pdf/1240.PDF
"The absence of scatter of the data points from the line indicates
that the source reservoirs of all these meteorites were formed
contemporaneously and that the Mn-Cr systems of the bulk samples of
these meteorites remained closed since their formation."
This piece talks about meteorites, (which I understand are considered
to be closer to the true age of the earth), therefore it makes sense
that mention is made of an ABSENCE of scatter of the data points.
Another source:
http://www.lpi.usra.edu/meetings/LPSC98/pdf/1663.pdf
"There is substantially more scatter in the St. Séverin data than in
the data for iron meteorites"
Comparison of chondrites from Acapulco shows more scatter than do
whole rock samples of iron meteorites, which would be expected if
meteorites are considered to be older and free from remelts versus
rocks from Earth.
>>
>>> It seems to me that any mixing at all would destroy the original
>>> relationship and only total mixing would create a new linear
>>> relationship.
>>
>> if the mixing is negligible, the original relationship only becomes
>> fuzzied (scatter). It is only total mixing that would create a new
>> linear relationship, but I am submitting that total mixing is not a
>> reality.
>
>No. It is only total mixing *within* the source of the rocks in question
>(not the entire mantle, but only the lava within the magma chamber) that
>produces samples with a slope of zero upon solidification.
why so? Isn't the mantle the source for the magma trapped in
chambers? Why should the part be different from the whole? I think
that it is up to you to demonstrate that convection can totally mix
magma in chambers better than it can mix mantle magma.
snip>
>> if freshly extruded lava is found to be nonzero via Rb/Sr isochron, it
>> would mean that the premelt linear relationship is still visible. But
>> why would anybody use Rb/Sr to date new lava? It is useless to use
>> such long-lived isotopes (48 billion years) to check for recent age.
>
>It is also crucial to do so to check the assumption that new melts have a
>constant D/Di uncorrelated with P/Di. That is why people have actually
>measured such freshly extruded lavas. It allows one to check the general
>validity of the assumptions of isochron dating. It is indeed useless to use
>such long-lived isotopes to check for recent age (because the true age will
>be wrapped within an error range of several millions of years). It is not
>useless to use such measurements as a check of the assumptions of the
>method. And guess what?
guess nothing. I would expect that recent melts that have been
through sufficient remelts, would show no obvious correlation between
D/Di and P/Di (slope) because the fuzzification due to scatter would
have dropped the slope down into the "zero" slope range. The supposed
constancy of D/Di is not an observable phenomenon. This is an
assumption that is used to interpret the isochron.
snip>
>> thanks for the suggestion, Eric, but this experiment would not match
>> up the variables between melted chocolate and magma. One is in an
>> open pot where boiling can take place. The other is in a pressurized
>> magma chamber where boiling does not take place.
>
>Then put the bloody chips in a pressure cooker for several hours and let it
>solidify. That would be better anyway because the differential between
>melting temperature and the temperature in the pressure cooker will be
>greater.
I guess I might consider trying that next -- except the blood might
contaminate the results.
----
zoe
zoe_althrop
<jo...@fleming-nospam.com> wrote:
snip>
>That's not particularly good, either. "Density" and "porosity" are
>already defined, and their meanings are not quite right for this
>situation.
>
>I suggest "mobility", which means exactly one of the major things
>which affects diffusion rate.
okay, so do you think that mobility of isotopes/molecules in melted
rock is higher than mobility of isotopes/molecules in melted
chocolate? Relatively speaking?
snip>
>Temperature is almost always THE MOST CRITICAL FACTOR. If everything
>else happens to be terrifically favorable to diffusion but the
>temperature is too low, essentially no diffusion happens. Diffusion
>is what's called a "thermally activated process". Each atom/whatever
>that moves by diffusion has to be kicked to get it moving, and what
>kicks it is the thermal energy that is characterized by the
>temperature. As the temperature increases, the atoms/whatever get
>kicked harder ... MUCH harder. So, the rate of diffusion goes up a
>LOT as the temperature increases.
>
>If the temperature is high enough, then mobility and time are the
>other possible critical factors. Which one is more critical depends
>on the situation. A lot of diffusion can happen in a material with
>low mobility if it happens over a long time, and a lot of diffusion
>can happen in a material with high mobility over a very short period
>of time..
I think it is a copout to invoke time as the means of accomplishing
changes that are not observable within our own reach. You might as
well invoke miracles at that rate.
snip>
>>Not to mention liquid
>>rock -- a molten material far more dense than either jello or
>>chocolate.
>
>Yes, far more dense. That's why I suggested _not_ using density as a
>label for mobility. You have already assumed that density (which
>means "how much a given size chunk of the stuff weighs") translates
>directly to "how mobile stuff is within the material", and that's
>wrong. There's a correlation; if all other things are equal, there's
>a tendency for diffusion to be slower in denser materials. But it's
>not always that way.
how slow is diffusion in melted rock? Do you have sources that
describe the rate? If diffusion is not noted within a reasonable
period of time, it is not justifiable to say, well, I'm sure if we
gave it more time, say, billions of years, that diffusion will finally
cause homogeneity.
>Note also the "all other things are equal" phrase. In your comparison
>between chocolate and magma, all other things are NOT equal. In fact,
>we know that there are two other gigantically different and
>significant factors; the temperature and the time. Even if we assume
>that mobility of P and D in magma is much less than mobility of white
>chocolate in dark chocolate, we cannot immediately conclude that P and
>D diffuse less in magma over thousands to billions of years at high
>temperatures than you observed in your chocolate experiment at low
>temperatures over a few minutes.
to me, it doesn't matter how high or low the temperature is. What
matters is how the material behaves once it reaches melting point.
Sure, it will take much higher temperatures to melt rock, but why
shouldn't the behavior of melted rock not be similar to the behavior
of melted chocolate, once both reach their melting points?
>>>> See:
>>>>
>>>>http://www.uic.edu/classes/phys/phys450/MARKO/N004.html
>>>>
>>>>"Water, and solutions which are mostly water, have a viscosity close
>>>>to hwater = 0.01 Poise or 0.001 Pa·sec."
>>>
>>>Note that this link talks about viscosity in relation to "small
>>>particles" in liquids, _not_ in relation to atoms (which are millions
>>>of times smaller than the "small particles").
>>
>>I suppose this means that the behavior of particles in liquids will
>>change, depending on size?
>
>Yes, big changes. That's why I assumed that you were talking about
>chocolate _molecules_ diffusing in your pot, not chocolate _chunks_
>diffusing in your pot. If you are indeed talking about chocolate and
>caramel _chunks_ moving around in your pot, then that's another
>important flaw in your experiment; chunks of quintillions of molecules
>are going to act _very_ different than individual atoms or molecules.
>If you are looking for those chunks of chocolate and caramel to move
>around in your pot until they are evenly distributed, then your
>experiment is much more like diffusion in a solid than it is like
>diffusion in a liquid.
I don't understand what you are saying here. The chocolate, once
melted, is no longer a solid. And I don't see why you would separate
atoms or molecules from the whole. If chocolate melts to the point
where its molecules are free to diffuse, then you no longer have
chunks. Yet the molecules that constituted the once-unmelted chunks
remained in essentially their original positions.
>>The smaller the particle, the more mobile,
>>regardless of the density or porosity of the liquid?
>
>Yes, in diffusion (not necessarily in other types of motion). But
>it's not really regardless of the porosity. For particles
>significantly larger than an atom or molecule of the liquid, the
>liquid looks like a non-porous continuum that has to be physically
>shoved out of the way. For atoms, the liquid looks like an extremely
>porous collection of loosely connected chunks (atoms).
you are still talking as if the chocolate chunks did not melt. They
melted, yet I could see where the original chunks were by the
spreading of the molecules into a wider zone, yet they did not diffuse
into each other except at the edges.
>>>Viscosity is important
>>>in "diffusion" of particles that are much much larger than the atoms
>>>or molecules of the fluid; viscosity is _not_ very important in
>>>diffusion of atoms that are about the same size as the atoms or
>>>molecules of the fluid.
snip>
>>www.ns.msu.edu/drew/nsc204/PagesLabSS/TAosmos.pdf
>>
>>"The rate of diffusion is proportional to the number of particles of
>>solute in solution,,whether they are small molecules, ions or large
>>polymeric molecules.. The presence of solute particles lessens the
>>water potential of the solution.. The more concentrated a solution
>>is,the more negative its water potential"
>>
>>Sounds to me as if the same principles apply in diffusion, whether the
>>particles are large or small.
>
>Yes, the same _fundamental_ principles apply. Note that they didn't
>mention viscosity. Viscosity is too "high level", you have to look
>into what _causes_ viscosity in order to understand its effects.
sorry, I didn't mean to use this quote in reference to viscosity. You
had said that "viscosity is _not_ very important in diffusion of atoms
that are about the same size as the atoms or molecules of the fluid."
My quote above was addressing that part of your statement to do with
small molecules or ions diffusing faster than larger particles. In
the lesson I quoted, apparently, large polymeric molecules are
similarly affected by the concentration of the solution as are small
molecules or ions. So why are you making a difference for smaller
particles versus larger?
snip viscosity misunderstanding>
>>>Diffusion in high temperature magma may well be much faster than
>>>diffusion in low-temperature water, I don't know off-hand. I do know
>>>that temperature is much more important than diffusivity.
>>
>>maybe you'll take it from Caltech, then?
>>
>>http://www.gps.caltech.edu/~gurnis/Movies/Science_Captions/stirring.html
>>
>>"Even after a few billion years of stirring by mantle convection, the
>>fluid is still quite heterogneous."
>
>I agree with that statement, but it's irrelevant to this discussion.
>
>That statement refers to large sections of the mantle, hundreds and
>thousands of miles across. We're discussing the samples that are
>taken for isochron dating. We are talking about chunks of magma that
>are a foot (or so) across to maybe 10 yards across at the very
>largest. The Caltech statement does not apply to such small chunks.
the small is part of the whole. If the whole is heterogenous, there
is no reason to assume that the small part will not be heterogenous.
What principles are you using to say it is different?
snip>
>>that is just it -- is there really a difference in concentration in
>>magma?
>
>Well, you are claiming that there is, and you are claiming that it
>maintains itself over thousands and millions of years.
I think I am using the wrong word, then. I don't mean there is
homogeneity when I use the word "concentration." I am referring to
ability for isotopes to move from areas of higher concentration to
lower concentration, isotope-wise. Surely, I don't think you mean
that P isotopes recognize that there are less P isotopes elsewhere in
the mix and they will head over there to even things out. Same for D
and Di isotopes. If there are pockets of lower concentration of ALL
isotopes to which higher concentrations of ALL isotopes will
gravitate, why should a D or Di isotope flow towards a space
preferentially because there are less D or Di isotopes in that area?
Why not a P isotope filling that void?
As I see it, only manual stirring or blending will distribute the
isotopes evenly. Otherwise, any diffusion will be at the mercy of the
next low-concentration pocket and will remain inhomogenous, more
likely retaining original formations than not.
>Homogeneous equals no difference in concentration.
>
>Heterogenous equals difference in concentration.
I guess we have different understandings of "concentration." By
"concentration" you seem to be referring to the isotopes themselves,
whereas for me "concentration" means how closely or loosely those
isotopes are packed. You seem to be saying that no difference in
concentration means all isotopes are of equal distribution, (which
would include P, btw. Why should just D and Di be homogenous and P be
in different concentrations? The ratio of D to Di to P should be the
same for all minerals, if you're going to take this position.)
>If you think there is no difference in concentration on the magma,
>then you think the magma is homogeneous, and there's no need to
>discuss any farther.
if we are using "difference in concentration" to mean different
things, then there is definitely need for further discussion, since at
the moment, we are talking past each other.
>>Rock, as it liquefies, compacts and becomes more dense; IF,
>>that is, it follows the same laws as melting chocolate.
>
>Ah, I see; you don't know what concentration means. I'll address that
>below. First density ...
>
>Of course, comparing the density of solid chunks of chocolate in a jar
>to a fluid mass of chocolate is comparing apples and oranges. What's
>meaningful is to compare the density of one chunk of solid chocolate
>to one chunk of liquid chocolate.
I have NOT been comparing the density of solid chunks of chocolate in
a jar. I am comparing the relative densities of melted chocolate and
melted rock. How do they both behave when melted.
>Almost all materials, including chocolate and rock, become LESS dense
>when they liquify. They do NOT become more compact. (Water is one of
>the few exceptions; liquid water is denser than ice, that's why ice
>floats in water).
>
>When you poured chocolate chunks out of the jar and noted that the
>melted chocolate didn't take up a fill jar, you were ignoring the air
>that was between the chunks originally. The liquid chocolate _itself_
>is less dense that the solid chocolate.
I didn't pour out the chocolate. I lifted the jar and noted that
where, before it had been full, the chocolate had now settled into a
more tightly packed state.
>>> The only question is how fast it happens.
>>
>>billions of years, maybe? More than 4.5 billion?
>
>In some situations, yes. That's one of the reasons isochron dating
>works. The P and D atoms are diffusing inside the solid rock as it
>sits there waiting for us to sample it ... but the diffusion rate
>under those conditions is so low that it would take tens of billions
>of years for the rock to homogenize (ignoring for the moment that D is
>being continuously created) so, for the most part, we can ignore
>diffusion in analyzing solid rocks.
I meant, does the diffusion happen in a MELT over billions of years,
more than 4.5 billion -- I wasn't talking about solid rock.
>
>>>The only way that there can be "no need for exchange of particles from
>>>areas of higher concentration to areas of lower concentration" is if
>>>the concentration is uniform,
>>
>>exactly. That is what I am saying. That magma is uniform in density.
>
>Yes, but not uniform in concentration. Density is not concentration.
>
>Concentration is the amount of something per unit volume or weight.
>For example, the number of D atoms per cubic inch of magma.
>
>If there are more D atoms per cubic inch of magma in one place than
>there are in another, there will be diffusion of D atoms between those
>two places.
so you're saying that D atoms know when there are less D atoms in
another area of the melt, and even though there are P atoms that can
go to that spot, the D atoms will head there first because they are
supposed to be evenly distributed?
>Your _entire_ _argument_ in this thread is based on a claim that there
>will be different numbers of P and D atoms per cubic inch in different
>parts of the magma, and these differences will remain over millions of
>years. That is arguing that there is a difference in concentration
okay, if that is what you mean by "concentration." Is there a word
for what I mean, then? That the isotopes are so closely packed that
the minerals melt but have nowhere to go. They retain their linear
relationship to each other.
snip>
>You didn't test the effect of higher temperature.
>
>Set up pot of the two chocolates that is barely molten. Wait until
>there is measurable diffusion between the two types. Record how long
>it took.
>
>Set up another pot of chocolate at a significantly higher temperature.
>Wait until the same amount of diffusion has taken place. Record how
>long it took.
>
>Compare the two times.
I suppose that would be a more rigorous method. I don't know what it
would prove, though, because if it turns out that there is very little
diffusion, it would prove nothing because I still would not have
access to the billions of years that you claim for your magma, in
which homogeneity will finally happen.
>>>I really wish you would stop making claims about diffusion that are
>>>based only on your fantasies. Diffusion has been studied for many
>>>years, and is pretty well understood. There's no reason for making
>>>stuff up as you are doing.
>>
>>I am not making this up. I am telling you what I have observed occurs
>>in watery solutions versus chocolate melts.
>
>I believe you saw what you described. What you _are_ making up is the
>connection between what you saw and what happens in magma.
maybe so. It sure makes me question what is assumed to go on in a
magma chamber that no one has access to, though.
>>>>And in the case of a magma chamber, all
>>>>you have is liquid rock, so you won't have magma dissolving into water
>>>>as you would with salt/sugar/red dye in water.
>>>
>>>You have liquid rock with P atoms and D atoms dissolved in it. You
>>>are claiming that there will be different amounts of P and D in
>>>different parts of the molten rock (and that will be true for some
>>>period of time after the rock melts; the question is for how long it
>>>will be true). Under those conditions, P and D atoms will diffuse.
>>>The only question is how fast.
>>
>>billions of years, and the melt will still be heterogenous. See link
>>above.
>
>The mantle will still be heterogeneous. The magma pools from which we
>draw isochron dating samples will be and are homogeneous.
that's an assertion, Jon, admit it. Magma pools are sourced from the
mantle, aren't they? If the mantle is heterogenous, then the magma
pools would reflect their source.
>Think of your pool of chocolates and caramel as the mantle. Dip just
>the very tip of a toothpick into a random part of the pool and pick up
>the tiniest amount of liquid that you can. That tiny portion is
>analogous to the melt from which we're sampling when we do isochron
>dating. Is that tiny bit of stuff that you picked up homogeneous?
>Almost certainly (unless you deliberately selected it to be
>inhomogeneous).
then your six or seven samples are not representative of the entire
cogenetic melt.
>>snip>
>>
>>>>if you are invoking billions of years for homogeneity to take place,
>>>>then your hypothesis of magma being homogenous becomes an
>>>>unfalsifiable hypothesis.
>>>
>>>Nope. Testable predictions can be made, such as recently solidified
>>>rock should be homogeneous. It almost always is.
>>
>>please for live examples from the field. If it takes billions of
>>years, and even then the mix remains heterogenous, how can a recent
>>solidification be homogenous? Something doesn't fit here.
>
>Yep, your understanding doesn't fit. The mantle is heterogenous even
>after billions of years. The melts from which we draw isochron dating
>samples are incredibly smaller than the mantle, and are homogeneous.
>The heterogeneity of the mantle is irrelevant.
is the mantle the source for magma in chambers, or not? Why would the
part not reflect the whole?
snip>
----
zoe
Martin Crisp
(in message <3ed2cbf5....@news-server.cfl.rr.com>):
[rest excised]
> If chocolate melts to the point where its molecules are free to
> diffuse, then you no longer have chunks. Yet the molecules that
> constituted the once-unmelted chunks remained in essentially
> their original positions.
A sort-of frozen-in-place liquid, perhaps?
Have Fun
Martin
--
aa #1792
Almost always SMASHed
Bigdakine
>Message-id: <05a4dv0tbhkasapql...@4ax.com>
>
>On Mon, 26 May 2003 03:35:50 +0000 (UTC), bigd...@aol.comGetaGrip
>(Bigdakine) wrote:
>
>>>Subject: Re: Results of an experiment
>>>From: Jon Fleming jo...@fleming-nospam.com
>
<snip>
>
>>>large separate geochemical reservoirs like the crust or asthenospheric
>>>upper mantle, requires both large scale melt extraction and convective
>>>heat and mass transfer. "
>>
>>Well that requires more explanation. Maybe I'll give one later when I have
>more
>>time.
Since the late 70's when people noticed that Ocena Island Basalts and Mid_ocean
ridge basalts had different rare earth abundances, it was proposed that they
represented two geochemical reservoirs. MORB's having the origins at very
shallow depths in the mantle, where as OIB's are theorized to be fed by mantle
plumes which have their origins deep in the mantle, mnost like at its base. So
intially we had a neat picture, the lower mantle is the OIB reservoir, and the
upper mantle is the MORB reservoir. And, at that time, the observed fact that
earthquakes did not occur below the 670km seismic discontinuity suggested that
the mantle was divided into upper and lower convecting systems which did not
exchange mass, just heat, through their common contact at the 670km. It fit so
well! Valhalla! mantle convection was solved, not only did it explain the most
fundamental isotopic heterogeneity known, but the reservoirs were able to
remain distinct over billions of years since mantle convection was divided into
different layers, each corresponding to a geochemical reservoir, preventing
mass exchange and eventual homogenization. Thats why the above explanation is a
bit confusing. You see convection, left to its own devices, will mix the
mantle, and quickly demolish (well, within the age of the Earth) any long
wavelength heterogeneity. What you need are *boundaries*, which are maintained
by whatever mechanism, to prevent the destruction of geochemical reservoirs by
convection. For example the crust is maintained as a separate geochemical
system, which has survived for billions of years because of the large density
difference it has with the mantle. . Of course if the mantle doesn't convect,
it doesn't differentiate either by realeasing volatiles and light components at
melting centers on the surface. But convection acts to mix the mantle.....
unless you can get it to unmix.
So you see, its not so simple.
By the early 80's, it was realized that subducting slabs, extend
aseismically, into the lower mantle. And thus, the pretty picture that evolved
in the late 70's, was demolished. The number of geochemical reservoirs has
proliferated, I've seen as many as 7 proposed. As a result, we convectioneers
have had an interesting time explaining the longevity of geochemical
reservoirs.
THis is a rich topic, and I could drone on endlessly, but you get the idea.
Richard McBane
>
> On Mon, 26 May 2003 14:36:53 +0000 (UTC), Jon Fleming
> <jo...@fleming-nospam.com> wrote:
>
> snip>
>
> >That's not particularly good, either. "Density" and "porosity" are
> >already defined, and their meanings are not quite right for this
> >situation.
> >
> >I suggest "mobility", which means exactly one of the major things
> >which affects diffusion rate.
>
> okay, so do you think that mobility of isotopes/molecules in melted
> rock is higher than mobility of isotopes/molecules in melted
> chocolate? Relatively speaking?
Yes the mobility of isotopes/molecules in melted rock is different than
in melted chocolate, just as it is different in melted ice.
In a previous post you indicated that you didn't want to use water
because you thought it would behave differently than magma. Chocolate
is very different than magma in that it is composed of complex organic
molecules that formed from carbon chains and rings that are much less
mobile than the molecules in magma. And the properties related to
mobility increase as temperature increases above the melting point. So
far, your chocolate melt is barely above its melting point. The volume
of material is also different. You didn't say how big your melt was but
it sounds relatively small. Convection, which is one element of mixing,
requires differences in density within the melt. Density differences
are caused by different temperatures within the magma chamber. Your
sample may be so small that the temperature difference, and thus density
difference within it are too small to allow mixing.
> snip>
>
> >Temperature is almost always THE MOST CRITICAL FACTOR. If everything
> >else happens to be terrifically favorable to diffusion but the
> >temperature is too low, essentially no diffusion happens. Diffusion
> >is what's called a "thermally activated process". Each atom/whatever
> >that moves by diffusion has to be kicked to get it moving, and what
> >kicks it is the thermal energy that is characterized by the
> >temperature. As the temperature increases, the atoms/whatever get
> >kicked harder ... MUCH harder. So, the rate of diffusion goes up a
> >LOT as the temperature increases.
> >
> >If the temperature is high enough, then mobility and time are the
> >other possible critical factors. Which one is more critical depends
> >on the situation. A lot of diffusion can happen in a material with
> >low mobility if it happens over a long time, and a lot of diffusion
> >can happen in a material with high mobility over a very short period
> >of time..
>
> I think it is a copout to invoke time as the means of accomplishing
> changes that are not observable within our own reach. You might as
> well invoke miracles at that rate.
Your experiment ran for a few hours. In the case of Mt. St. Helens, we
know that that magma chamber sat for hundreds of years before it
erupted.
>
> <snip>
<snip>
>
> to me, it doesn't matter how high or low the temperature is. What
> matters is how the material behaves once it reaches melting point.
> Sure, it will take much higher temperatures to melt rock, but why
> shouldn't the behavior of melted rock not be similar to the behavior
> of melted chocolate, once both reach their melting points?
See comments above.
<snip>
> >>maybe you'll take it from Caltech, then?
> >>
> >>http://www.gps.caltech.edu/~gurnis/Movies/Science_Captions/stirring.html
> >>
> >>"Even after a few billion years of stirring by mantle convection, the
> >>fluid is still quite heterogneous."
> >
> >I agree with that statement, but it's irrelevant to this discussion.
> >
> >That statement refers to large sections of the mantle, hundreds and
> >thousands of miles across. We're discussing the samples that are
> >taken for isochron dating. We are talking about chunks of magma that
> >are a foot (or so) across to maybe 10 yards across at the very
> >largest. The Caltech statement does not apply to such small chunks.
>
> the small is part of the whole. If the whole is heterogenous, there
> is no reason to assume that the small part will not be heterogenous.
> What principles are you using to say it is different?
Its quite possible for the mantle to be heterogeneous over 10's and
100's of thousands of square miles but to be locally homogeneous on a
small scale. One of your previous sources talked about how the fact
that the product in magma chambers is homogenous could not be used to
infer that the mantle is homogeneous. What he was saying is that even
though magma in a magma chamber is homogenous, the mantle is not. In
other words if you sample the magma from Hawaii and from Iceland, you
shouldn't be surprised to find that they are different because the
mantle is not the same composition all over the world.
>
> snip>
<snip>
> >
> >The mantle will still be heterogeneous. The magma pools from which we
> >draw isochron dating samples will be and are homogeneous.
>
> that's an assertion, Jon, admit it. Magma pools are sourced from the
> mantle, aren't they? If the mantle is heterogenous, then the magma
> pools would reflect their source.
> zoe
--
Richard McBane
John Drayton
<snip>
> to me, it doesn't matter how high or low the temperature is. What
> matters is how the material behaves once it reaches melting point.
> Sure, it will take much higher temperatures to melt rock, but why
> shouldn't the behavior of melted rock not be similar to the behavior
> of melted chocolate, once both reach their melting points?
Because they are totally different substances?
Why shouldn't the behaviour of melted rock not be the same
as melted ice once both reach their melting points? They
are both hard substances that melt after a certain
temperature has been reached.
Perhaps I'm missing something here, but what is the point
of the chocoltae experiment? If you *knew* it behaved the
same as magma, then you could perhaps illustrate a point
with it, but you don't know that. There's no reason to
expect that they would behave the same: they are completely
different substances, and the conditions are different also.
And more to the point, can't we tell if mixing is complete
from the actual substances themselves: the rock that has
formed? Won't the isochron show a high scatter if the mixing
is poor? Am I misunderstanding something here?
<snip>
--
John Drayton
Bjoern Feuerbacher
muz...@aol.com (zoe_althrop) wrote in message news:<3ed2bcc6....@news-server.cfl.rr.com>...
> On Mon, 26 May 2003 14:54:06 +0000 (UTC), Howard Hershey
> <hers...@indiana.edu> wrote:
>
> snip>
>
> zoe wrote:
>
> >> the original linear relationship of D/Di to P/Di would come from an
> >> original formation in which there was no oldD present. Since D
> >> originates from P, there is no reason to imagine that there was a time
> >> when D originated all by itself, without the aid of P.
> >
> >In the initial outburst of a supernova, but that is besides the point.
>
> why is this besides the point? It is exactly the point that needs to
> be explored.
>
> What can you tell me about how supernovas produce D without initial P?
Supernovas produce elements by fusion and by neutron captures with
subsequent beta decays, for example.
> >> Therefore, the
> >> moment that P begins to decay to D, the linear relationship would be
> >> established, and it is this linear relationship that becomes more and
> >> more fuzzied by remelts. The more remelts, the younger the rock
> >> appears to be. The less remelts, the older the rock shows up to be.
> >
> >Until, of course, the present time, when examination of lavas from
> >historically modern volcanos show constant D/Di uncorrelated with P/Di
> >rather than positive linear relationships.
>
> if a historically modern volcano shows a supposedly constant D/Di,
Why "supposedly"? It *does* show this (the rock which formed from the
lava from the volcano, not the volcano itself!).
> or zero slope, this would be explained by my hypothesis as the result of
> sufficient remelt scatter that would bring the slope down to the zero
> range.
Could you please explain in detail how this works?
Could you please explain, too, why in such isochrons, no more scatter
is observed than in the ones with non-zero slopes?
> At least for those rocks that have been through sufficient
> remelts to lose the positive slope.
In what way do "sufficient" remelts cause a rock to "lose" the
positive slope? Could you please explain how this works, and
illustrate this with a numerical example? What does "sufficient" mean
here?
> >There are a number of problems with this. It cannot explain the general
> >consistency of isochron and argon dating methods. In the argon method,
> >any time you get a melt you necessarily generally lose all or nearly all
> >the "retained" argon *in the local area* due to the nature of the noble
> >gasses in a liquid.
>
> I would expect all dating methods to agree simply because the
> accumulated decay product, calculated by ANY method, should reflect
> the same length of time the rocks have been present on this earth --
Err, Zoe, for the 100000th time: the isochron method does *not*
calculate "accumulated decay product". This is demonstrated easily by
the simple fact that the slope of newly solidified rocks is zero, no
matter how much "accumulated decay product" is in them.
> within a general age range, plus or minus X number of years that would
> cover contamination, leaching, and the like.
Still ignoring that the isochron method can detect contamination and
leaching...
> The problem is not with
> matching ages from different methods, but with the application of
> those ages to the history of the rocks.
Blah blah blah...
> These ages should not reflect time since solidification.
But they do!
> They reflect
> should reflect accumulated decay product
Zoe, that *STILL* is a completely unsupported assertion!
> that gets translated into a
> calculated equivalent of years. Applying the calculated number of
> years to time since solidification is an unjustified assumption, imo.
Then bring up a numerical example where the slope does *not* give the
time since solidification!!!
[snip]
> >If your analysis were true, then the fit of the data points to the isochron
> >line (number of significant figures to which the slope is accurate) of
> >isochrons that falsely indicate recent dates should be significantly worse
> >than isochrons that date much older, since only the old isochrons
> >'accurately' reflect a situation in which D/Di actually started out as
> >constant and new D has been added by decay of local P. All more modern
> >isochrons necessarily will have undergone "fuzzification" by random and
> >local events, producing data that fit a 'line' less well than rocks that
> >have ungone fewer "fuzzification" events. OTOH, if the clock *resets* at
> >each melt (D/Di is constant at each melt), then the accuracy of the slope
> >(how well the data fits the line) will *not* be a function of the age of the
> >rocks. Do you want to bet on which is observed?
>
> that makes sense. What about this link?
>
> http://www.cosis.net/abstracts/EAE03/04785/EAE03-J-04785.pdf
>
> "Rb-Sr whole rock data points obtained from the Tres Piedras Granite
> yields a distinct isochron for each outcrop sampled. The age and
> apparent initial Sr87/Sr86 ratios of the Tres Piedras Granite outcrops
> (approximately 50 square meters of collecting area at each exposure)
> are as follows: Tres Piedras Granite Type Locality: 1493 +/- 21 Maand
> 0.7183 +/- 0.0006; Tres Piedras Granite ? Tusas River Canyon: 1501 +/-
> 44 Maand 0.7145 +/- 0.0013; and Tres Piedras Granite ? Tusas Mountain:
> 1661 +/- 17 Maand 0.7102 +/- 0.0071."
>
> correct me if I'm wrong, but it appears that the younger ages carry
> increased scatter:
>
> 0.7183 +/- 0.0006
> 0.7145 +/- 0.0013
> 0.7102 +/- 0.0071
Err, Zoe, the number after the "+/-" is not the scatter. It's the
error.
And additionally, this is *one* special example. Do you know how much
isochron datings were done? At how many have you looked?
> The piece says that metamorphism could be the cause for the
> differences, but why not an alternate supposition that, with each
> remelt, the scatter increases?
Because (1) the numbers you gave here are not the scatter and (2) this
isn't consistently observed.
> here's another link.
>
> http://www.lpi.usra.edu/meetings/lpsc97/pdf/1240.PDF
>
> "The absence of scatter of the data points from the line indicates
> that the source reservoirs of all these meteorites were formed
> contemporaneously and that the Mn-Cr systems of the bulk samples of
> these meteorites remained closed since their formation."
>
> This piece talks about meteorites, (which I understand are considered
> to be closer to the true age of the earth), therefore it makes sense
> that mention is made of an ABSENCE of scatter of the data points.
Zoe, it was explained above *why* there is no scatter. Do you think
their explanation makes no sense, didn't you understand it, or what?
> Another source:
>
> http://www.lpi.usra.edu/meetings/LPSC98/pdf/1663.pdf
>
> "There is substantially more scatter in the St. Séverin data than in
> the data for iron meteorites"
>
> Comparison of chondrites from Acapulco shows more scatter than do
> whole rock samples of iron meteorites, which would be expected if
> meteorites are considered to be older and free from remelts versus
> rocks from Earth.
Right. But there are lots of other possible reasons for this: for
example,
rocks on Earth suffer more contamination than meteorites. Did you rule
out this explanation?
Did you examine if rocks which are younger show *consistently* more
scatter? Or did you pick out just a few examples which seemed to say
what you want? (although you misunderstood them - again, the numbers
given above after the +/- are *not* the scatter!).
> >>> It seems to me that any mixing at all would destroy the original
> >>> relationship and only total mixing would create a new linear
> >>> relationship.
> >>
> >> if the mixing is negligible, the original relationship only becomes
> >> fuzzied (scatter). It is only total mixing that would create a new
> >> linear relationship, but I am submitting that total mixing is not a
> >> reality.
> >
> >No. It is only total mixing *within* the source of the rocks in question
> >(not the entire mantle, but only the lava within the magma chamber) that
> >produces samples with a slope of zero upon solidification.
>
> why so? Isn't the mantle the source for the magma trapped in
> chambers?
Yes.
> Why should the part be different from the whole?
Because small parts are more mixed than the whole - mixing is
obviously faster in small parts than in the whole.
> I think
> that it is up to you to demonstrate that convection can totally mix
> magma in chambers better than it can mix mantle magma.
Zoe, newly solidified rocks *are* homogeneous. Hence the magma *is*
mixed in magma chambers. How long do you want to ignore these
observations?
> >> if freshly extruded lava is found to be nonzero via Rb/Sr isochron, it
> >> would mean that the premelt linear relationship is still visible. But
> >> why would anybody use Rb/Sr to date new lava? It is useless to use
> >> such long-lived isotopes (48 billion years) to check for recent age.
> >
> >It is also crucial to do so to check the assumption that new melts have a
> >constant D/Di uncorrelated with P/Di. That is why people have actually
> >measured such freshly extruded lavas. It allows one to check the general
> >validity of the assumptions of isochron dating. It is indeed useless to use
> >such long-lived isotopes to check for recent age (because the true age will
> >be wrapped within an error range of several millions of years). It is not
> >useless to use such measurements as a check of the assumptions of the
> >method. And guess what?
>
> guess nothing.
Well, the answer is: newly solidified rocks are homogeneous - hence
magma *is* thouroughly mixed.
> I would expect that recent melts that have been
> through sufficient remelts,
What means "sufficient"?
> would show no obvious correlation between
> D/Di and P/Di (slope) because the fuzzification due to scatter
Zoe, there is no more scatter in these recent melts than in older
ones.
Would you *please* start facing reality?
> would
> have dropped the slope down into the "zero" slope range.
What does this mean? How does this work? Could you please demonstrate
this with an example?
> The supposed
> constancy of D/Di is not an observable phenomenon.
Huh? Pardon??? It *is* observed - in every recently solidified rock.
> This is an assumption that is used to interpret the isochron.
No, it's an *observation* that rocks which solidify today are
homogeneous. The only "assumption" is that the laws of nature worked
in the same way some billion years ago.
[snip rest]
Bye,
Bjoern
Bjoern Feuerbacher
muz...@aol.com (zoe_althrop) wrote in message news:<3ed2cbf5....@news-server.cfl.rr.com>...
> On Mon, 26 May 2003 14:36:53 +0000 (UTC), Jon Fleming
> <jo...@fleming-nospam.com> wrote:
>
> snip>
>
> >That's not particularly good, either. "Density" and "porosity" are
> >already defined, and their meanings are not quite right for this
> >situation.
> >
> >I suggest "mobility", which means exactly one of the major things
> >which affects diffusion rate.
>
> okay, so do you think that mobility of isotopes/molecules in melted
> rock is higher than mobility of isotopes/molecules in melted
> chocolate? Relatively speaking?
First, replace "isotopes/molecules" in the first instance with
"atoms", and in the second, with "molecules" only. Otherwise it makes
little sense. Second, I think the mobilities could be comparable.
Third, this is of little relevance, because the *temperature* in
melted rock is much higher than in melted chocolate, and, as Jon has
pointed out repeatedly, diffusion depends exponentially on
temperature.
> >Temperature is almost always THE MOST CRITICAL FACTOR. If everything
> >else happens to be terrifically favorable to diffusion but the
> >temperature is too low, essentially no diffusion happens. Diffusion
> >is what's called a "thermally activated process". Each atom/whatever
> >that moves by diffusion has to be kicked to get it moving, and what
> >kicks it is the thermal energy that is characterized by the
> >temperature. As the temperature increases, the atoms/whatever get
> >kicked harder ... MUCH harder. So, the rate of diffusion goes up a
> >LOT as the temperature increases.
> >
> >If the temperature is high enough, then mobility and time are the
> >other possible critical factors. Which one is more critical depends
> >on the situation. A lot of diffusion can happen in a material with
> >low mobility if it happens over a long time, and a lot of diffusion
> >can happen in a material with high mobility over a very short period
> >of time..
>
> I think it is a copout to invoke time as the means of accomplishing
> changes that are not observable within our own reach. You might as
> well invoke miracles at that rate.
Zoe, this objection makes absolutely no sense. With the same argument,
you could say: "It's a copout to invoke time as the means of
accomplishing a decay
of half the amount of Rb we have here. We observe small changes,
right,
but I don't think we are allowed to extrapolate there."
If you doubt that more time can achieve a better homogenization (and
this doubt is absolutely ridiculous! It's *obvious* that more time
gives better homogenization!!!), then you have to doubt, too, that in
48 billion years,
half of the Rb decays.
And I notice that you completely ignored the crucial part, although
Jon stressed again its importance: temperature!!!
> >>Not to mention liquid
> >>rock -- a molten material far more dense than either jello or
> >>chocolate.
> >
> >Yes, far more dense. That's why I suggested _not_ using density as a
> >label for mobility. You have already assumed that density (which
> >means "how much a given size chunk of the stuff weighs") translates
> >directly to "how mobile stuff is within the material", and that's
> >wrong. There's a correlation; if all other things are equal, there's
> >a tendency for diffusion to be slower in denser materials. But it's
> >not always that way.
>
> how slow is diffusion in melted rock? Do you have sources that
> describe the rate? If diffusion is not noted within a reasonable
> period of time, it is not justifiable to say, well, I'm sure if we
> gave it more time, say, billions of years, that diffusion will finally
> cause homogeneity.
Zoe, sources and links were already given which described diffusion in
rocks. Why didn't you look them up?
And for the 100th time: recently solidified rock is *observed* to be
homogeneous, hence obviously magma is homogeneous!!!
> >Note also the "all other things are equal" phrase. In your comparison
> >between chocolate and magma, all other things are NOT equal. In fact,
> >we know that there are two other gigantically different and
> >significant factors; the temperature and the time. Even if we assume
> >that mobility of P and D in magma is much less than mobility of white
> >chocolate in dark chocolate, we cannot immediately conclude that P and
> >D diffuse less in magma over thousands to billions of years at high
> >temperatures than you observed in your chocolate experiment at low
> >temperatures over a few minutes.
>
> to me, it doesn't matter how high or low the temperature is.
Because you are *absolutely* ignorant of diffusion.
TEMPERATURE IS THE MOST IMPORTANT FACTOR IN DIFFUSION!
Did you get it this time?
> What
> matters is how the material behaves once it reaches melting point.
Why does this matter???
> Sure, it will take much higher temperatures to melt rock, but why
> shouldn't the behavior of melted rock not be similar to the behavior
> of melted chocolate, once both reach their melting points?
BECAUSE THE TEMPERATURE IS VASTLY DIFFERENT!?!?!?!
[snip]
> >>I suppose this means that the behavior of particles in liquids will
> >>change, depending on size?
> >
> >Yes, big changes. That's why I assumed that you were talking about
> >chocolate _molecules_ diffusing in your pot, not chocolate _chunks_
> >diffusing in your pot. If you are indeed talking about chocolate and
> >caramel _chunks_ moving around in your pot, then that's another
> >important flaw in your experiment; chunks of quintillions of molecules
> >are going to act _very_ different than individual atoms or molecules.
> >If you are looking for those chunks of chocolate and caramel to move
> >around in your pot until they are evenly distributed, then your
> >experiment is much more like diffusion in a solid than it is like
> >diffusion in a liquid.
>
> I don't understand what you are saying here. The chocolate, once
> melted, is no longer a solid. And I don't see why you would separate
> atoms or molecules from the whole. If chocolate melts to the point
> where its molecules are free to diffuse, then you no longer have
> chunks.
Right. Jon simply pointed out that *if* there are unmelted chunks
left, their behaviour is different from the molecules.
> Yet the molecules that constituted the once-unmelted chunks
> remained in essentially their original positions.
Right - at this temperature and during this time.
What means "essentially"? Did it stay in *exactly* the same position,
or did it move a bit? If it did move, did you measure the change in
position? If it moves a bit in a short time, what does stop it from
moving more in a longer time?
[snip]
> They
> melted, yet I could see where the original chunks were by the
> spreading of the molecules into a wider zone, yet they did not diffuse
> into each other except at the edges.
And what would happen, in your opinion, if you would carry out this
experiment for a longer time? Let's say you heat the chocolate for a
week or so - what happens?
If magmas would behave in that way (not mix, only diffuse into each
other "at the edges"), we wouldn't get homogeneous rocks from volcanic
eruptions. Nevertheless, exactly this is observed.
[snip]
> >>www.ns.msu.edu/drew/nsc204/PagesLabSS/TAosmos.pdf
> >>
> >>"The rate of diffusion is proportional to the number of particles of
> >>solute in solution,,whether they are small molecules, ions or large
> >>polymeric molecules.. The presence of solute particles lessens the
> >>water potential of the solution.. The more concentrated a solution
> >>is,the more negative its water potential"
> >>
> >>Sounds to me as if the same principles apply in diffusion, whether the
> >>particles are large or small.
> >
> >Yes, the same _fundamental_ principles apply. Note that they didn't
> >mention viscosity. Viscosity is too "high level", you have to look
> >into what _causes_ viscosity in order to understand its effects.
>
> sorry, I didn't mean to use this quote in reference to viscosity. You
> had said that "viscosity is _not_ very important in diffusion of atoms
> that are about the same size as the atoms or molecules of the fluid."
> My quote above was addressing that part of your statement to do with
> small molecules or ions diffusing faster than larger particles. In
> the lesson I quoted, apparently, large polymeric molecules are
> similarly affected by the concentration of the solution as are small
> molecules or ions.
Right. But what has this to do with the question if they diffuse
slower or not???
"A has the same effect on B than on C" does not mean "B and C behave
in completely the same way"!
> So why are you making a difference for smaller
> particles versus larger?
Because they *do* behave differently!
> snip viscosity misunderstanding>
>
> >>>Diffusion in high temperature magma may well be much faster than
> >>>diffusion in low-temperature water, I don't know off-hand. I do know
> >>>that temperature is much more important than diffusivity.
> >>
> >>maybe you'll take it from Caltech, then?
> >>
> >>http://www.gps.caltech.edu/~gurnis/Movies/Science_Captions/stirring.html
> >>
> >>"Even after a few billion years of stirring by mantle convection, the
> >>fluid is still quite heterogneous."
> >
> >I agree with that statement, but it's irrelevant to this discussion.
> >
> >That statement refers to large sections of the mantle, hundreds and
> >thousands of miles across. We're discussing the samples that are
> >taken for isochron dating. We are talking about chunks of magma that
> >are a foot (or so) across to maybe 10 yards across at the very
> >largest. The Caltech statement does not apply to such small chunks.
>
> the small is part of the whole.
Yes. So what?
> If the whole is heterogenous, there
> is no reason to assume that the small part will not be heterogenous.
This makes no sense at all. If something is heterogenous on a large
scale,
nothing stops it from being homogenous on a small scale. And we have
two different arguments which show that it *is* homogeneous on a small
scale:
1) the law of diffusion
2) the observation that recently solidified rocks are homogeneous.
> What principles are you using to say it is different?
See above.
> >>that is just it -- is there really a difference in concentration in
> >>magma?
> >
> >Well, you are claiming that there is, and you are claiming that it
> >maintains itself over thousands and millions of years.
>
> I think I am using the wrong word, then. I don't mean there is
> homogeneity when I use the word "concentration."
And no one claimed that you do mean homogenity when you use the word
concentration, so what on Earth are you talking about???
> I am referring to
> ability for isotopes to move from areas of higher concentration to
> lower concentration, isotope-wise.
In other words, mobility.
> Surely, I don't think you mean
> that P isotopes recognize that there are less P isotopes elsewhere in
> the mix and they will head over there to even things out.
No. But obviously, if there are more atoms of P in region A than in
region B, and all these atoms move around, in the mean more atoms will
move out of A and into B than in the opposite direction (that's again
statistics!).
> Same for D and Di isotopes.
Right, the same.
> If there are pockets of lower concentration of ALL
> isotopes to which higher concentrations of ALL isotopes will
> gravitate,
"gravitate" is really the wrong word here!
> why should a D or Di isotope flow towards a space
> preferentially because there are less D or Di isotopes in that area?\
Statistics.
> Why not a P isotope filling that void?
Could happen, too. But in the end, after everything has settled, you
will have a homogeneous contribution (hint: that's a consequence of
the 2nd law of thermodynamics, in principle).
> As I see it, only manual stirring or blending will distribute the
> isotopes evenly.
You see it wrong.
And you *still* haven't explained why recently solidified rocks *are*
homogeneous!
> Otherwise, any diffusion will be at the mercy of the
> next low-concentration pocket and will remain inhomogenous, more
> likely retaining original formations than not.
I don't understand what you say here, sorry.
> >Homogeneous equals no difference in concentration.
> >
> >Heterogenous equals difference in concentration.
>
> I guess we have different understandings of "concentration."
Big surprise!
> By
> "concentration" you seem to be referring to the isotopes themselves,
"Concentration" means "how many of the stuff is there, compared to the
other stuff that's there", in principle.
> whereas for me "concentration" means how closely or loosely those
> isotopes are packed.
That's "density".
> You seem to be saying that no difference in
> concentration means all isotopes are of equal distribution,
Yes - this follows from the *definition* of "concentration". If
everythere there is exactly the same amount (relative to the other
stuff that's there), obviously there is an equal distribution.
> (which
> would include P, btw. Why should just D and Di be homogenous and P be
> in different concentrations?
In the magma, P *is* homogeneously distributed. But because of
chemistry, in the rock, it isn't.
> The ratio of D to Di to P should be the
> same for all minerals, if you're going to take this position.)
Right from the viewpoint of the magma, but nevertheless wrong, because
you ignore the laws of chemistry which say that D can bind to P only
in certain
ratios.
> >>> The only question is how fast it happens.
> >>
> >>billions of years, maybe? More than 4.5 billion?
> >
> >In some situations, yes. That's one of the reasons isochron dating
> >works. The P and D atoms are diffusing inside the solid rock as it
> >sits there waiting for us to sample it ... but the diffusion rate
> >under those conditions is so low that it would take tens of billions
> >of years for the rock to homogenize (ignoring for the moment that D is
> >being continuously created) so, for the most part, we can ignore
> >diffusion in analyzing solid rocks.
>
> I meant, does the diffusion happen in a MELT over billions of years,
> more than 4.5 billion -- I wasn't talking about solid rock.
Well, try reading the links which were given. They talked about
diffusion in melts.
> >Concentration is the amount of something per unit volume or weight.
> >For example, the number of D atoms per cubic inch of magma.
> >
> >If there are more D atoms per cubic inch of magma in one place than
> >there are in another, there will be diffusion of D atoms between those
> >two places.
>
> so you're saying that D atoms know when there are less D atoms in
> another area of the melt,
No, they don't "know" this. This simply happens because of statistics!
Zoe, imagine you have a box with two gases, separated by a wall. Then
you take out the wall and wait a while (how long depends on the
temperature mainly). Afterwards, the two gases will have mixed -
although none of the gas molecules "knew" that there were none of them
in the other part of the box!
> and even though there are P atoms that can
> go to that spot, the D atoms will head there first because they are
> supposed to be evenly distributed?
Sometimes the P atoms will go to that spot, sometimes the D atoms. But
in the mean, all of these movements will lead to a homogeneous
mixture.
Imagine the experiment with the two gases above again, but this time,
take three gases and two walls. After taking out the walls and waiting
for a while, all three gases will have mixed.
[snip]
> >Your _entire_ _argument_ in this thread is based on a claim that there
> >will be different numbers of P and D atoms per cubic inch in different
> >parts of the magma, and these differences will remain over millions of
> >years. That is arguing that there is a difference in concentration
>
> okay, if that is what you mean by "concentration."
That's what *everyone* mean by "concentration".
> Is there a word
> for what I mean, then? That the isotopes are so closely packed that
> the minerals melt but have nowhere to go.
"density". But this makes no sense - essentially, if the molecules or
atoms are packed so closely that they can't move, you have a solid,
not a melt. A common characteristic, I would even say a *defining*
characteristic of *all* liquids is that the atoms or molecules can
move (relatively) freely in it and are not bound to the same place.
> They retain their linear relationship to each other.
Contradicted by the observation that recently solidified rocks are
homogeneous.
> >You didn't test the effect of higher temperature.
> >
> >Set up pot of the two chocolates that is barely molten. Wait until
> >there is measurable diffusion between the two types. Record how long
> >it took.
> >
> >Set up another pot of chocolate at a significantly higher temperature.
> >Wait until the same amount of diffusion has taken place. Record how
> >long it took.
> >
> >Compare the two times.
>
> I suppose that would be a more rigorous method.
Yes.
> I don't know what it
> would prove, though,
It would show how diffusion depends on temperature, and that
temperature is a crucial factor.
> because if it turns out that there is very little
> diffusion, it would prove nothing because I still would not have
> access to the billions of years that you claim for your magma, in
> which homogeneity will finally happen.
Zoe, why don't you extrapolate the diffusion you observe properly?
Look up how diffusion depends on temperature and time, and then use
the formulas to extrapolate your measured values.
> >>>I really wish you would stop making claims about diffusion that are
> >>>based only on your fantasies. Diffusion has been studied for many
> >>>years, and is pretty well understood. There's no reason for making
> >>>stuff up as you are doing.
> >>
> >>I am not making this up. I am telling you what I have observed occurs
> >>in watery solutions versus chocolate melts.
> >
> >I believe you saw what you described. What you _are_ making up is the
> >connection between what you saw and what happens in magma.
>
> maybe so. It sure makes me question what is assumed to go on in a
> magma chamber that no one has access to, though.
Zoe, we have access to the end product - the lava. This lava *is*
homogeneous. This has been tested several, I would even say lots of
times.
And we know the laws of diffusion, hence we can *calculate* what goes
on in magma chambers. Do you think the laws of diffusion we know are
wrong, or that there is a reason why they don't apply in magma
chambers?
[snip]
> >>billions of years, and the melt will still be heterogenous. See link
> >>above.
> >
> >The mantle will still be heterogeneous. The magma pools from which we
> >draw isochron dating samples will be and are homogeneous.
>
> that's an assertion, Jon, admit it.
No, that's an *observation*. Freshly solidified rocks are homogeneous,
hence the magma is homogeneous.
> Magma pools are sourced from the
> mantle, aren't they?
Yes.
> If the mantle is heterogenous, then the magma
> pools would reflect their source.
The mantle is heterogenous *ON LARGE SCALES*. Magma pools have
relatively small sizes, compared to the whole mantle, hence there is
no problem with them being homogeneous.
> >Think of your pool of chocolates and caramel as the mantle. Dip just
> >the very tip of a toothpick into a random part of the pool and pick up
> >the tiniest amount of liquid that you can. That tiny portion is
> >analogous to the melt from which we're sampling when we do isochron
> >dating. Is that tiny bit of stuff that you picked up homogeneous?
> >Almost certainly (unless you deliberately selected it to be
> >inhomogeneous).
>
> then your six or seven samples are not representative of the entire
> cogenetic melt.
Right. Why should it? The goal of isochron dating is to find samples
which are representative of the time since solidification, not of the
whole melt!
[snip]
> >Yep, your understanding doesn't fit. The mantle is heterogenous even
> >after billions of years. The melts from which we draw isochron dating
> >samples are incredibly smaller than the mantle, and are homogeneous.
> >The heterogeneity of the mantle is irrelevant.
>
> is the mantle the source for magma in chambers, or not? Why would the
> part not reflect the whole?
What's your problem with understanding that something which is
heterogenous on large scales can be homogenous on small scales?
Bye,
Bjoern
Matt Silberstein
(zoe_althrop):
>On Sat, 24 May 2003 21:20:39 +0000 (UTC), Matt Silberstein
><mat...@ix.netcom.com> wrote:
>
>snip>
>
>>Try this. Take some light honey, some black molasses, and some
>>clear corn syrup. Take some clear jars and put layers of each
>>material in the jar. Put one jar in the fridge. Put another on
>>the shelf and let it sit for days/weeks. Take a third jar and put
>>it in a pot. You want water in the pot about as high as the
>>bottom layer, but certainly not as high as the top. Keep the
>>water warm and watch the layer boundary. Let it sit in
>>moderately warm (100 degrees or so) for an hour or so. You could
>>try this with several different water temps. The really
>>interesting thing will come when the material starts to show
>>convection. That is, material on the bottom will be warmer than
>>the top and rise, material on the top will drop. This will mix
>>the three materials quite well.
>
>thanks for the suggestion, Matt, but I don't think that honey would
>represent rock the way chocolate more nearly does.
When you do science you try to control for the variables. The
materials here are liquid, yes, the temperatures are low and the
time frame small. You can easily freeze the items to make them
solid or even crystalize them. (At least the honey will
crystalize, I don't know about the corn syrup and probably not
the molasses.)
> Both chocolate and
>rock are initially in solid forms, and when melted, the observation is
>one of how well material of these relative densities and heating will
>become mixed. Don't you think that the experiment should start out
>with something in solid form first, and then see what happens to its
>material as it melts? If you don't think so, why not?
Are you willing to deal with the kinds of temperatures involved?
Like well above 1,000? And will you wait for a few years while
you observe? Anyway, the mantel is liquid, not solid.
--
Matt Silberstein TBC HRL OMM
We are not here to judge other people,
we are just here to be better than they are.
Jon Fleming
How do you account for the fact that in new flows that
have been measured we do see that D/Di is the same for all samples?
On Tue, 27 May 2003 01:51:41 +0000 (UTC), muz...@aol.com (zoe_althrop)
wrote:
>
>http://www.cosis.net/abstracts/EAE03/04785/EAE03-J-04785.pdf
>
>"Rb-Sr whole rock data points obtained from the Tres Piedras Granite
>yields a distinct isochron for each outcrop sampled. The age and
>apparent initial Sr87/Sr86 ratios of the Tres Piedras Granite outcrops
>(approximately 50 square meters of collecting area at each exposure)
>are as follows: Tres Piedras Granite Type Locality: 1493 +/- 21 Maand
>0.7183 +/- 0.0006; Tres Piedras Granite – Tusas River Canyon: 1501 +/-
>44 Maand 0.7145 +/- 0.0013; and Tres Piedras Granite – Tusas Mountain:
>1661 +/- 17 Maand 0.7102 +/- 0.0071."
>
>correct me if I'm wrong, but it appears that the younger ages carry
>increased scatter:
>
>0.7183 +/- 0.0006
>0.7145 +/- 0.0013
>0.7102 +/- 0.0071
You are wrong. Those are the Y-intercepts. The greatest scatter is
in the intermediate age
Age Scatter
1493 +/- 21
1501 +/- 44
1661 +/- 17
>The piece says that metamorphism could be the cause for the
>differences, but why not an alternate supposition that, with each
>remelt, the scatter increases?
Well, first of all, what's wrong with it is that the data that you
presented indicates otherwise. And the differences in the ages are
way too small
But you have presented FAR too little data to make a determination.
However, you can of course go ahead and make that supposition. Then
collect a significant number of published ages ranging from a few
million years ago to around 4 billion years ago. About 25 to 100
should do. Make sure that, as much as possible, you include dates
that cover all the possible dates in the range of a few million years
ago to around 4 billion years ago; to do this you'll probably have to
examine 250-500 papers.
Then graph or otherwise analyze the scatter as a function of age.
Let's assume that you _do _ see a tendency for younger ages to have
greater scatter Then you analyze the statistical significance. Next
you consider other possible sources of the trend (such as the fact
that younger rocks involve measuring smaller differences in isotopic
makeup) and prepare discussion s of why you think that the effect is
due to you supposed mechanism rather than other possible mechanisms.
Then get back to us with your results.
...
(change nospam to group to email)
Jon Fleming
How do you account for the fact that in new flows that
have been measured we do see that D/Di is the same for all samples?
On Tue, 27 May 2003 02:58:47 +0000 (UTC), muz...@aol.com (zoe_althrop)
wrote:
>On Mon, 26 May 2003 14:36:53 +0000 (UTC), Jon Fleming
><jo...@fleming-nospam.com> wrote:
>
>snip>
>
>>That's not particularly good, either. "Density" and "porosity" are
>>already defined, and their meanings are not quite right for this
>>situation.
>>
>>I suggest "mobility", which means exactly one of the major things
>>which affects diffusion rate.
>
>okay, so do you think that mobility of isotopes/molecules in melted
>rock is higher than mobility of isotopes/molecules in melted
>chocolate? Relatively speaking?
I don't know. It's difficult to make a comparison. Since you can't
have molten chocolate and molten rock at the same temperature, you
can't compare them at the same temperature. We know that temperature
has a gigantic effect.
>snip>
>
>>Temperature is almost always THE MOST CRITICAL FACTOR. If everything
>>else happens to be terrifically favorable to diffusion but the
>>temperature is too low, essentially no diffusion happens. Diffusion
>>is what's called a "thermally activated process". Each atom/whatever
>>that moves by diffusion has to be kicked to get it moving, and what
>>kicks it is the thermal energy that is characterized by the
>>temperature. As the temperature increases, the atoms/whatever get
>>kicked harder ... MUCH harder. So, the rate of diffusion goes up a
>>LOT as the temperature increases.
>>
>>If the temperature is high enough, then mobility and time are the
>>other possible critical factors. Which one is more critical depends
>>on the situation. A lot of diffusion can happen in a material with
>>low mobility if it happens over a long time, and a lot of diffusion
>>can happen in a material with high mobility over a very short period
>>of time..
>
>I think it is a copout to invoke time as the means of accomplishing
>changes that are not observable within our own reach.
Nope, not a copout, just an acknowledgement of a fact. Molten rocks
tend to be molten for millions of years. In considering what happens
to them, we have to include processes that act over that time period.
>You might as
>well invoke miracles at that rate.
Totally different. Miracles involve events that are incompatible with
our understanding of the laws of physics. Events that take place over
millions of year in accordance with our understanding of the laws of
physics are not at all comparable to miracles.
>snip>
>
>>>Not to mention liquid
>>>rock -- a molten material far more dense than either jello or
>>>chocolate.
>>
>>Yes, far more dense. That's why I suggested _not_ using density as a
>>label for mobility. You have already assumed that density (which
>>means "how much a given size chunk of the stuff weighs") translates
>>directly to "how mobile stuff is within the material", and that's
>>wrong. There's a correlation; if all other things are equal, there's
>>a tendency for diffusion to be slower in denser materials. But it's
>>not always that way.
>
>how slow is diffusion in melted rock? Do you have sources that
>describe the rate?
_NOW_ you're asking about the rate of diffusion in molten rock? After
all the assertions you've made about what diffusion cannot do in
molten rock, you are now admitting that you don't have any idea how
fast it is?
It varies widely, depending on all sorts of things. Diffusivity in
magma ranges from 10^-18 to 10^-6 square meters per second (3*10"-18
to 3*10^-6 square feet per second). For Rb in a dry rhyolite melt at
1,000 degrees C it's about 10^-12 square meters per second (3*10^-12
square feet per second) For Sr in a dry rhyolite melt it's about
4*10^12 square meters per second
The diffusion distance L, which is approximately the size of the
volume that can be homogenized by diffusion, is the square root of the
product of diffusivity times the time elapsed in seconds. E.g, for Rb
diffusing for a million years (3.15*10^13 seconds):
L = sqrt(10^-12*3.15*10^13)
L = sqrt(31.5)
L = 5.6 meters = 18 feet
So, in a million years, the Rb approximately an 18 foot diameter
"bubble" of a dry rhyolite magma at 1,000 degrees C will be
homogenized
>If diffusion is not noted within a reasonable
>period of time, it is not justifiable to say, well, I'm sure if we
>gave it more time, say, billions of years, that diffusion will finally
>cause homogeneity.
The question doesn't make sense. We do measure diffusion in magma in
the lab, so diffusion _is_ noted in a reasonable period of time.
>>Note also the "all other things are equal" phrase. In your comparison
>>between chocolate and magma, all other things are NOT equal. In fact,
>>we know that there are two other gigantically different and
>>significant factors; the temperature and the time. Even if we assume
>>that mobility of P and D in magma is much less than mobility of white
>>chocolate in dark chocolate, we cannot immediately conclude that P and
>>D diffuse less in magma over thousands to billions of years at high
>>temperatures than you observed in your chocolate experiment at low
>>temperatures over a few minutes.
>
>to me, it doesn't matter how high or low the temperature is.
Then you don't have the slightest idea of what diffusion is or how it
works, and you are just fantasizing and blowing smoke. Temperature is
terrifically important. For example, the diffusivity of Rb in a dry
rhyolite magma at 450 degrees C is about 10^-16 square meters per
second. 10,000 times slower than at 1,000 degrees C!!!!
> What
>matters is how the material behaves once it reaches melting point.
>Sure, it will take much higher temperatures to melt rock, but why
>shouldn't the behavior of melted rock not be similar to the behavior
>of melted chocolate, once both reach their melting points?
Wrong question. You're the one that's claiming that they are the
same, so it's your responsibility to answer "why _would_ the behavior
of melted rock be similar to the behavior of melted chocolate, once
both reach their melting points?".
Of course, any reference you consult on diffusion is going to point
out that temperature matters (unless they're talking about everything
at the same temperature). The answer to your question is that
different materials, different atoms, different temperatures, and
different times all make significant difference.
<snip>
>>>The smaller the particle, the more mobile,
>>>regardless of the density or porosity of the liquid?
>>
>>Yes, in diffusion (not necessarily in other types of motion). But
>>it's not really regardless of the porosity. For particles
>>significantly larger than an atom or molecule of the liquid, the
>>liquid looks like a non-porous continuum that has to be physically
>>shoved out of the way. For atoms, the liquid looks like an extremely
>>porous collection of loosely connected chunks (atoms).
>
>you are still talking as if the chocolate chunks did not melt. They
>melted, yet I could see where the original chunks were by the
>spreading of the molecules into a wider zone, yet they did not diffuse
>into each other except at the edges.
EXCEPT AT THE EDGES????
In other words, they _did_ diffuse into each other at the edges.
That's diffusion.
Diffusion is motion of particles that are pushed by thermal energy.
Your molten chips are not particles; if something pushes on one part
of a chip, the whole chip doesn't move, only part of the chip moves..
Your molten chips aren't going to diffuse as chips because they are
not the kind of thing that diffuses.
The molecules of chocolate and the molecules of caramel _are_ the sort
of thing that diffuse, and they are going to diffuse.
For the reasons that I provided already.
No, they don't recognize that there are less P atoms elsewhere in the
mix, they don't have the capability ... but they _DO_ head over there
to even things out. That's what diffusion is. It's another case in
which non-random results arise from a random fundamental process.
The P atoms are knocked about randomly by thermal energy. The higher
the temperature, the more they are knocked about. Which direction
they go each time they're kicked is random, and how far they go each
time they're kicked is random (within limits).
In areas where there are more P atoms, they are more likely to be
kicked _out_ of that area, just because there are more P atoms. In
areas where there are fewer P atoms, they are less likely to be kicked
out of that area, just because there are fewer P atoms. Some will be
kicked out of areas with few P atoms ... but more will be kicked out
of the areas with more P atoms and _in_ to the areas with few P atoms.
So the net result is that random kicks in random directions move P
atoms from areas with lots of P atoms to areas with fewer P atoms, and
the concentration of P atoms gets more homogeneous over time.
This is exactly what Howard and others have been trying to get across
\t you with their experiments with dyed water and ice cubes and the
like. Make an ice cube of colored water. Put it in a glass of
un-colored water. Wait a few days. The color of the water will be
uniform. How did the color atoms "know" to spread out that way? They
didn't; random thermal processes did it.
See
<http://www.timedomaincvd.com/CVD_Fundamentals/xprt/intro_diffusion.html>.
>Same for D
>and Di isotopes. If there are pockets of lower concentration of ALL
>isotopes to which higher concentrations of ALL isotopes will
>gravitate, why should a D or Di isotope flow towards a space
>preferentially because there are less D or Di isotopes in that area?
>Why not a P isotope filling that void?
>
>As I see it, only manual stirring or blending will distribute the
>isotopes evenly.
Well, you see wrong. Diffusion distributes isotopes evenly.
>Otherwise, any diffusion will be at the mercy of the
>next low-concentration pocket and will remain inhomogenous, more
>likely retaining original formations than not.
You have no idea how diffusion works.
>>Homogeneous equals no difference in concentration.
>>
>>Heterogenous equals difference in concentration.
>
>I guess we have different understandings of "concentration." By
>"concentration" you seem to be referring to the isotopes themselves,
>whereas for me "concentration" means how closely or loosely those
>isotopes are packed.
Ah, you are making up new meanings for words again. Don't do that.
From Merriam Webster:
"Concentration: Chemistry. The amount of a specified substance in a
unit amount of another substance."
>You seem to be saying that no difference in
>concentration means all isotopes are of equal distribution, (which
>would include P, btw. Why should just D and Di be homogenous and P be
>in different concentrations?
It wouldn't, in the melt.
>The ratio of D to Di to P should be the
>same for all minerals, if you're going to take this position.)
In the melt it's the same. When crystals solidify, different crystals
take up different amounts of P and different amounts of D or Di,
depending on their crystal structure. This establishes the
differences that we see later.
>>If you think there is no difference in concentration on the magma,
>>then you think the magma is homogeneous, and there's no need to
>>discuss any farther.
>
>if we are using "difference in concentration" to mean different
>things, then there is definitely need for further discussion, since at
>the moment, we are talking past each other.
Try using the dictionary definition of words instead of making up your
own definitions.
>>>Rock, as it liquefies, compacts and becomes more dense; IF,
>>>that is, it follows the same laws as melting chocolate.
>>
>>Ah, I see; you don't know what concentration means. I'll address that
>>below. First density ...
>>
>>Of course, comparing the density of solid chunks of chocolate in a jar
>>to a fluid mass of chocolate is comparing apples and oranges. What's
>>meaningful is to compare the density of one chunk of solid chocolate
>>to one chunk of liquid chocolate.
>
>I have NOT been comparing the density of solid chunks of chocolate in
>a jar. I am comparing the relative densities of melted chocolate and
>melted rock. How do they both behave when melted.
They behave as I said; the density (weight per unit volume) goes down.
>>Almost all materials, including chocolate and rock, become LESS dense
>>when they liquify. They do NOT become more compact. (Water is one of
>>the few exceptions; liquid water is denser than ice, that's why ice
>>floats in water).
>>
>>When you poured chocolate chunks out of the jar and noted that the
>>melted chocolate didn't take up a fill jar, you were ignoring the air
>>that was between the chunks originally. The liquid chocolate _itself_
>>is less dense that the solid chocolate.
>
>I didn't pour out the chocolate. I lifted the jar and noted that
>where, before it had been full, the chocolate had now settled into a
>more tightly packed state.
>
>>>> The only question is how fast it happens.
>>>
>>>billions of years, maybe? More than 4.5 billion?
>>
>>In some situations, yes. That's one of the reasons isochron dating
>>works. The P and D atoms are diffusing inside the solid rock as it
>>sits there waiting for us to sample it ... but the diffusion rate
>>under those conditions is so low that it would take tens of billions
>>of years for the rock to homogenize (ignoring for the moment that D is
>>being continuously created) so, for the most part, we can ignore
>>diffusion in analyzing solid rocks.
>
>I meant, does the diffusion happen in a MELT over billions of years,
>more than 4.5 billion -- I wasn't talking about solid rock.
Depends on how big a melt you are talking about,
In the melts from which we sample for isochron dating, which vary
widely in size, the melt is homogenized by diffusion over times
ranging from a few thousand to a few hundred million years.
Mechanical mixing (which does happen, despite your unsupported
assertions to the contrary) reduces the homogenization time.
>>>>The only way that there can be "no need for exchange of particles from
>>>>areas of higher concentration to areas of lower concentration" is if
>>>>the concentration is uniform,
>>>
>>>exactly. That is what I am saying. That magma is uniform in density.
>>
>>Yes, but not uniform in concentration. Density is not concentration.
>>
>>Concentration is the amount of something per unit volume or weight.
>>For example, the number of D atoms per cubic inch of magma.
>>
>>If there are more D atoms per cubic inch of magma in one place than
>>there are in another, there will be diffusion of D atoms between those
>>two places.
>
>so you're saying that D atoms know when there are less D atoms in
>another area of the melt, and even though there are P atoms that can
>go to that spot, the D atoms will head there first because they are
>supposed to be evenly distributed?
Nope, I'm saying that the random thermal motions of the D atoms lead
to a decidedly non-random even distribution of the D atoms throughout
the melt.
>>Your _entire_ _argument_ in this thread is based on a claim that there
>>will be different numbers of P and D atoms per cubic inch in different
>>parts of the magma, and these differences will remain over millions of
>>years. That is arguing that there is a difference in concentration
>
>okay, if that is what you mean by "concentration." Is there a word
>for what I mean, then? That the isotopes are so closely packed that
>the minerals melt but have nowhere to go. They retain their linear
>relationship to each other.
There's no word for it because there is no such thing, at least in
liquids. In liquids it is impossible to have the atoms so closely
packed that the atoms of other minerals have no place to go.
It's possible in solids, but it can only be achieved in very small
solids in laboratories with lots of expensive equipment, and with the
expenditure of lots of effort. No naturally-occurring solid is packed
that tightly (although many solids are packed tightly enough to make
diffusion very slow).
In your chocolate melt, the atoms of chocolate and caramel are
diffusing. The packing is not so tight that they can't.
<snip>
>>I believe you saw what you described. What you _are_ making up is the
>>connection between what you saw and what happens in magma.
>
>maybe so. It sure makes me question what is assumed to go on in a
>magma chamber that no one has access to, though.
We assume that the laws of physics that apply in the lab also apply in
the magma chamber. We run experiments in the lab using the materials
that are found in the magma chamber, under the temperatures and
pressures found in the magma chamber, in order to minimize the
possible errors.
<snip>
>>Think of your pool of chocolates and caramel as the mantle. Dip just
>>the very tip of a toothpick into a random part of the pool and pick up
>>the tiniest amount of liquid that you can. That tiny portion is
>>analogous to the melt from which we're sampling when we do isochron
>>dating. Is that tiny bit of stuff that you picked up homogeneous?
>>Almost certainly (unless you deliberately selected it to be
>>inhomogeneous).
>
>then your six or seven samples are not representative of the entire
>cogenetic melt.
Absolutely! So freakin' what!?!?! There's no requirement that the
samples be representative of the entire melt. And we don't care
whether or not the _melt_ is co-genetic, all we care about is whether
the _samples_ are co-genetic. Co-genetic samples solidified at the
same time from a homogeneous PORTION of a melt, and we don't care
about the rest of the melt.
For example, if we pick up a rock that's a foot in diameter and do a
mineral isochron analysis of several samples, then that rock came from
a melt that was a foot in diameter. If that teeny small portion of
the entire melt was homogeneous, the samples from that rock are
co-genetic.
Or say we pick up seven rocks from one lava flow, and we select those
rocks from an area 400 feet in diameter, and we do a whole-rock
isochron analysis of those seven rocks. Here it's a little less
clear, but our samples came from a melt that was about 100 to 1,000
feet in diameter. If that portion of the entire melt was homogeneous,
our samples are co-genetic.
Most of our isochron samples come from melt portions that are small
enough to be homogenized by diffusion. Some come from larger melts,
and those require mechanical mixing to have taken place in order to be
homogeneous. In spite of your unsupported claims, mechanical mixing
is common in magma.
If the PART OF THE MELT FROM WHICH THE SAMPLES ARE DRAWN was uniform,
the assumption of co-geneticity of the samples is met, and the
isochron method gives the time since solidification (provided the
samples remained closed).
>>>>>if you are invoking billions of years for homogeneity to take place,
>>>>>then your hypothesis of magma being homogenous becomes an
>>>>>unfalsifiable hypothesis.
>>>>
>>>>Nope. Testable predictions can be made, such as recently solidified
>>>>rock should be homogeneous. It almost always is.
>>>
>>>please for live examples from the field. If it takes billions of
>>>years, and even then the mix remains heterogenous, how can a recent
>>>solidification be homogenous? Something doesn't fit here.
>>
>>Yep, your understanding doesn't fit. The mantle is heterogenous even
>>after billions of years. The melts from which we draw isochron dating
>>samples are incredibly smaller than the mantle, and are homogeneous.
>>The heterogeneity of the mantle is irrelevant.
>
>is the mantle the source for magma in chambers, or not? Why would the
>part not reflect the whole?
Huh? The mantle is heterogeneous .. that does not mean that every
possible rock solidified from the mantle is heterogeneous. Parts
often do not have the same properties as the whole.
Look at your chocolate-caramel melt. It's not homogeneous, is it?
Now barely touch the very tip of a toothpick to one of the chocolate
areas and extract the smallest amount that you can. Is that tiny bit
of chocolate on the end of the toothpick homogeneous? Does it contain
caramel? Is it the same as the whole?
Jon Fleming
feue...@thphys.uni-heidelberg.de (Bjoern Feuerbacher) wrote:
>> http://www.cosis.net/abstracts/EAE03/04785/EAE03-J-04785.pdf
>>
>> "Rb-Sr whole rock data points obtained from the Tres Piedras Granite
>> yields a distinct isochron for each outcrop sampled. The age and
>> apparent initial Sr87/Sr86 ratios of the Tres Piedras Granite outcrops
>> (approximately 50 square meters of collecting area at each exposure)
>> are as follows: Tres Piedras Granite Type Locality: 1493 +/- 21 Maand
>> 0.7183 +/- 0.0006; Tres Piedras Granite ? Tusas River Canyon: 1501 +/-
>> 44 Maand 0.7145 +/- 0.0013; and Tres Piedras Granite ? Tusas Mountain:
>> 1661 +/- 17 Maand 0.7102 +/- 0.0071."
>>
>> correct me if I'm wrong, but it appears that the younger ages carry
>> increased scatter:
>>
>> 0.7183 +/- 0.0006
>> 0.7145 +/- 0.0013
>> 0.7102 +/- 0.0071
>
>Err, Zoe, the number after the "+/-" is not the scatter. It's the
>error.
And the number before the "+/-" in her table is not the age ...
Jon Fleming
(Bigdakine) wrote:
Interesting. Thanks.
Jon Fleming
feue...@thphys.uni-heidelberg.de (Bjoern Feuerbacher) wrote:
>
>> The ratio of D to Di to P should be the
>> same for all minerals, if you're going to take this position.)
>
>Right from the viewpoint of the magma, but nevertheless wrong, because
>you ignore the laws of chemistry which say that D can bind to P only
>in certain
>ratios.
Er, she is indeed ignoring the laws of chemistry, but you are ignoring
the fact that D and P are essentially never part of the chemical
formula of a mineral. It's not quite as simple as D binding to P in
certain ratios.
D and P are incorporated by substitution or physical entrapment. Of
course, substitution leads directly to the effect of differing amounts
of P and D in different minerals, and entrapment can too (but not
quite so strongly).
Jon Fleming
wrote:
>
>http://www.cosis.net/abstracts/EAE03/04785/EAE03-J-04785.pdf
>
>"Rb-Sr whole rock data points obtained from the Tres Piedras Granite
>yields a distinct isochron for each outcrop sampled. The age and
>apparent initial Sr87/Sr86 ratios of the Tres Piedras Granite outcrops
>(approximately 50 square meters of collecting area at each exposure)
>are as follows: Tres Piedras Granite Type Locality: 1493 +/- 21 Maand
>0.7183 +/- 0.0006; Tres Piedras Granite – Tusas River Canyon: 1501 +/-
>44 Maand 0.7145 +/- 0.0013; and Tres Piedras Granite – Tusas Mountain:
>1661 +/- 17 Maand 0.7102 +/- 0.0071."
>
>correct me if I'm wrong, but it appears that the younger ages carry
>increased scatter:
>
>0.7183 +/- 0.0006
>0.7145 +/- 0.0013
>0.7102 +/- 0.0071
Actually I'm half right and half wrong, Bjoern's half right and half
wrong, and you're 100% wrong.
The numbers before the +/- in your table are not ages, the numbers
after the +/- are not scatter.
Jon Fleming
wrote:
>if a historically modern volcano shows a supposedly constant D/Di, or
>zero slope, this would be explained by my hypothesis as the result of
>sufficient remelt scatter that would bring the slope down to the zero
>range. At least for those rocks that have been through sufficient
>remelts to lose the positive slope.
This is new. You've been claiming that young rocks would show greater
scatter (and your attempts to support that are pretty pitiful), but
this is the first claim of a systematic reduction in slope in your
"model".
So please fill in the details. What, physically, reduces the slope to
zero if melts do not homogenize?
John Stockwell
>
>On Fri, 23 May 2003 20:56:53 +0000 (UTC), pven...@hotmail.com (P.
>Venkman) wrote:
>
>>muz...@aol.com (zoe_althrop) wrote in message news:<3ecd8ced....@news-server.cfl.rr.com>...
>><SNIP>
>>
>>> okay, Jon & Company, I have a lot more chocolate morsels on hand. Are
>>> there any other or better experiments that can be done with
>>> them?...quickly, please, before they disappear down my gullet.
>>>
>>
>>Add 3 drops of red ink to the water on one side of the glass, then add
>>three drops of blue ink to the water on the opposite side of the
>>glass. Cover the glass with some plastic wrap (just to avoid
>>evaporation problems) and leave it for a week.
>>
>>Now pour the contents of the glass into an ice cube tray, stick the
>>tray in the freezer, and make ice cubes. After they're frozen, pull
>>them out. Are the ice cubes all the same color, or do you have some
>>mostly red and some mostly blue cubes?
>>
>>What do you conclude from this experiment?
>
>without even conducting this experiment, I can conclude that it does
>not in any way simulate magma mixing.
Yes it does. We are not talking about the gross mixing of large amounts
of materials of different miscibilities. We are talking about the
diffusion of trace materials, which the ink in water experiment
amply describes.
>----
>zoe
>
>
--
John Stockwell | jo...@dix.Mines.EDU
Center for Wave Phenomena (The Home of Seismic Un*x)
Colorado School of Mines
Golden, CO 80401 | http://www.cwp.mines.edu/cwpcodes
voice: (303) 273-3049
Our book:
Norman Bleistein, Jack K. Cohen, John W. Stockwell Jr., [2001],
Mathematics of multidimensional seismic imaging, migration, and inversion,
(Interdisciplinary Applied Mathematics, V. 13.), Springer-Verlag, New York.
howard hershey
Probably right. But how about the following? Start out with chocolate
chips in honey or water in the pressure cooker and freeze the entire
batch (making it like solid rock with materials of different melting
points and densities). Then putting the frozen block in the cooker on
the stove. The water in the honey should be sufficient to provide the
steam for the pressure. In a sense, the steam will be like entrapped
volitiles in rocks (like Ar). These would dissipate into the atmosphere
in volcanos because that would be like opening up the pressure cooker
before solidification, but in our case would re-condense during cooling.
The embedded chips should melt at the temperature of a pressure cooker
(which is further above the melting point of chocolate than a double
boiler is), but the low level of oxygen availability should prevent the
chips from burning or charring significantly (which is why I suggested a
pressure cooker rather than a microwave, which was a very dangerous and
foolish experiment). But the chips should melt as the temperature
climbs. Then, after a couple of hours, she should refreeze the mix and
examine it. According to Zoe, we should still see the fuzzied chips
embedded in the frozen ice block. Somehow, I don't think she will.
Because of the immiscibility of lipids in water, she may see *layers* of
chocolate fats over ice form during the cooling process rather than a
uniform mixture (separation out of materials -- but not as dramatically
-- and crystalization due to different chemical properties does occur in
lavas too).
howard hershey
zoe_althrop wrote:
> On Mon, 26 May 2003 14:54:06 +0000 (UTC), Howard Hershey
> <hers...@indiana.edu> wrote:
>
> snip>
>
> zoe wrote:
>
>
>>>the original linear relationship of D/Di to P/Di would come from an
>>>original formation in which there was no oldD present. Since D
>>>originates from P, there is no reason to imagine that there was a time
>>>when D originated all by itself, without the aid of P.
>>
>>In the initial outburst of a supernova, but that is besides the point.
>
>
> why is this besides the point? It is exactly the point that needs to
> be explored.
It is irrelevant because all one ever measures is totalD/Di in samples.
OldD is not and never was defined as the amount of D present at the
formation of the earth or present before the earth formed (after all, P
did not poof into existence either). OldD/Di is defined as the ratio of
D/Di present in all samples at the time of last solidification of the
samples used to create the isochron plot. OldD/Di, defined correctly,
is empirically estimated from the ratio of D/Di at x = 0 (aka, the
y-intercept) in an extrapolation (usually, since one rarely actually has
a data point with no P) of the line of best fit plotted through the
measured sample data points (with totalD/Di of each sample being the y
value and currentP/Di being the x value). The reason for the
y-intercept being a good estimate is that a sample with 0 P will not
change its level of totalD/Di after solidification since there is no
source of newD.
>
> What can you tell me about how supernovas produce D without initial P?
It is formed by the same sort of mechanisms that produce P in supernovae.
>
>>>Therefore, the
>>>moment that P begins to decay to D, the linear relationship would be
>>>established, and it is this linear relationship that becomes more and
>>>more fuzzied by remelts. The more remelts, the younger the rock
>>>appears to be. The less remelts, the older the rock shows up to be.
>>
>>Until, of course, the present time, when examination of lavas from
>>historically modern volcanos show constant D/Di uncorrelated with P/Di
>>rather than positive linear relationships.
>
>
> if a historically modern volcano shows a supposedly constant D/Di, or
> zero slope, this would be explained by my hypothesis as the result of
> sufficient remelt scatter that would bring the slope down to the zero
> range. At least for those rocks that have been through sufficient
> remelts to lose the positive slope.
Why do essentially *all* modern volcanos show this constant D/Di level?
If your hypothesis were true, shouldn't we be seeing a significant
number of volcanos (those whose magma contain rocks with fewer remelts)
which present clear positive slopes correlated with P levels? And how
does that explain the consistency with layers and the fossils embedded
between them? Are you claiming that the entire mantle has solidified
and remelted equally so that all volcanoes that produce lava indicating,
say, a age of 100 mybp just happen to have certain fossils above them
and other fossils below them?
>
>
>>There are a number of problems with this. It cannot explain the general
>>consistency of isochron and argon dating methods. In the argon method,
>>any time you get a melt you necessarily generally lose all or nearly all
>>the "retained" argon *in the local area* due to the nature of the noble
>>gasses in a liquid.
>
>
> I would expect all dating methods to agree simply because the
> accumulated decay product, calculated by ANY method, should reflect
> the same length of time the rocks have been present on this earth --
> within a general age range, plus or minus X number of years that would
> cover contamination, leaching, and the like. The problem is not with
> matching ages from different methods, but with the application of
> those ages to the history of the rocks.
That does not answer the criticism. If the reason for getting a certain
date is because melted rocks merely reform before significant diffusion
of the previous isotopes has occurred (ala your 'chocolate' model),
giving a falsely old date, that model does not hold in the case of a
method whereby age is indicated by entrapped volitiles (like Ar). If
the rock melts at all, it is no more able to hold Ar nearby (ala your
'chocolate' model) than I am able to grab ahold of the He released from
a balloon with my bare hands. Thus, we should be seeing huge and
consistent discrepancies between ages determined by the Ar methods
(which can only measure Ar accumulated since the last solidification,
because Ar cannot be significantly retained in liquid rock) and Rb/Sr
methods (which you claim gives false ages older than the time of last
solidification). Specifically, the Ar methods should give consistently
massively younger ages to a given rock than the Rb/Sr method. OTOH, if
both in fact measure age since solidification, the methods should
largely agree.
>
> These ages should not reflect time since solidification. They reflect
> should reflect accumulated decay product that gets translated into a
> calculated equivalent of years. Applying the calculated number of
> years to time since solidification is an unjustified assumption, imo.
>
I agree. If your 'chocolate' model is correct, only the Ar method
should reflect time since solidification. It should produce the very
young ages you require and should differ significantly from the ages
produced by Rb/Sr since the mechanism of "fuzzification" is quite
different in the two methods.
That would be the scatter around the y-intercept value. The scatter
around the slope is better estimated by the error in the ages (since
slope is what is used to determine age, not y-intercept). That is:
1493 +/- 21 Ma
1501 +/- 44 Ma
1661 +/- 17 Ma
Personally, I don't see a lot of difference in the amount of scatter
around the slope here. You need a wider spread in slope measurements
and a few examples of recent slope calculations (those which are, to all
extents and purposes a slope of zero) would be nice.
>
> The piece says that metamorphism could be the cause for the
> differences, but why not an alternate supposition that, with each
> remelt, the scatter increases?
>
> here's another link.
>
> http://www.lpi.usra.edu/meetings/lpsc97/pdf/1240.PDF
>
> "The absence of scatter of the data points from the line indicates
> that the source reservoirs of all these meteorites were formed
> contemporaneously and that the Mn-Cr systems of the bulk samples of
> these meteorites remained closed since their formation."
>
> This piece talks about meteorites, (which I understand are considered
> to be closer to the true age of the earth), therefore it makes sense
> that mention is made of an ABSENCE of scatter of the data points.
There is *always* some scatter, if for no other reason there will be
scatter due to the limits of measurement.
>
> Another source:
>
> http://www.lpi.usra.edu/meetings/LPSC98/pdf/1663.pdf
>
> "There is substantially more scatter in the St. Séverin data than in
> the data for iron meteorites"
>
> Comparison of chondrites from Acapulco shows more scatter than do
> whole rock samples of iron meteorites, which would be expected if
> meteorites are considered to be older and free from remelts versus
> rocks from Earth.
>
The above is mere hand-waving until you do an actual comparison plot of
scatter vrs. determined age. I would imagine that the primary cause of
scatter is the limitation of measurement accuracy with actual sample to
sample variation due to "fuzzification" events (which is due to
violations of the assumption of constant D/Di at last solidification --
which is the violation you are proposing -- and/or the assumption of
closed system), but not clearly identifiable as such, being the other
contributor. The problem you have is that there is very little
deviation in any of the useful isochrons. You could look at the
literature for the last 5 years, say, to make the measurement accuracy
as close to constant as possible in an improving technology.
>
>>>>It seems to me that any mixing at all would destroy the original
>>>>relationship and only total mixing would create a new linear
>>>>relationship.
>>>
>>>if the mixing is negligible, the original relationship only becomes
>>>fuzzied (scatter). It is only total mixing that would create a new
>>>linear relationship, but I am submitting that total mixing is not a
>>>reality.
>>
>>No. It is only total mixing *within* the source of the rocks in question
>>(not the entire mantle, but only the lava within the magma chamber) that
>>produces samples with a slope of zero upon solidification.
>
>
> why so?
Because it is only the rocks from a particular eruption from a
particular magma chamber that solidified at a particular time that you
are interested in dating, not the entire mantle.
> Isn't the mantle the source for the magma trapped in
> chambers?
Yes.
> Why should the part be different from the whole?
It probably is not *very* different in D/Di ratio from the whole. But
whether it is or isn't is irrelevant. What matters is that the lava in
a particular chamber (or from crystals taken from rocks not more than 10
m away) be essentially uniform in the ratio of D/Di (uniformity in P/Di
ratios is not important or relevant) as it comes out in a liquid.
> I think
> that it is up to you to demonstrate that convection can totally mix
> magma in chambers better than it can mix mantle magma.
The fact that new lavas produce samples with constant D/Di ratios in a
given eruption and place is not enough? Your 'model' (unless you are
saying that everything is different now) requires that there
*frequently* be new lavas (those that have not undergone frequent
remelts) that produce positive D/Di ratios corellated with the levels of
P in the samples. Can you tell me how your model manages that neat trick?
>
> snip>
>
>>>if freshly extruded lava is found to be nonzero via Rb/Sr isochron, it
>>>would mean that the premelt linear relationship is still visible. But
>>>why would anybody use Rb/Sr to date new lava? It is useless to use
>>>such long-lived isotopes (48 billion years) to check for recent age.
>>
>>It is also crucial to do so to check the assumption that new melts have a
>>constant D/Di uncorrelated with P/Di. That is why people have actually
>>measured such freshly extruded lavas. It allows one to check the general
>>validity of the assumptions of isochron dating. It is indeed useless to use
>>such long-lived isotopes to check for recent age (because the true age will
>>be wrapped within an error range of several millions of years). It is not
>>useless to use such measurements as a check of the assumptions of the
>>method. And guess what?
>
>
> guess nothing. I would expect that recent melts that have been
> through sufficient remelts, would show no obvious correlation between
> D/Di and P/Di (slope) because the fuzzification due to scatter would
> have dropped the slope down into the "zero" slope range. The supposed
> constancy of D/Di is not an observable phenomenon. This is an
> assumption that is used to interpret the isochron.
IOW, you have to assume that essentially *all* rocks from *all* magma
chambers in *all* modern volcanos have undergone sufficient remelts so
that *all* recent lavas now show the *same* zero slope. World-wide.
And *all* rocks from *all* magma chambers right around the K-T boundary
just *happened* to have undergone the correct number of 'remelts' so
that they misleadingly, but consistently, give an age of 60 million
years or so. World-wide. Regardless of the nature of the material in
the melt. Regardless of the apparent consistency with Ar methods which
would not be expected to behave in the same way. And do you still
believe in the tooth fairy?
P. Venkman
> On Fri, 23 May 2003 16:18:10 +0000 (UTC), pven...@hotmail.com (P.
> Venkman) wrote:
>
> snip>
>
> >Actually I have a question about this. You're questioning whether a
> >linear relationship between D/Di and P/Di can be retained during a
> >melt. That pre-supposes a linear relationship exists pre-melt. Take
> >that back to its logical conclusion and that means a linear
> >relationship had to exist at the formation of the earth. Is there any
> >reason to believe this is true?
>
> the reason to believe this is true is that the laws of physics is the
> same today as it always was. And since the only observation of these
> laws that holds true today is the one that say that D originates from
> decay of P, then there could be no original oldD at the first
> formation of the earth -- unless you want to posit some miraculous
> appearance of oldD that did not originate from P. Occam would not
> approve of that, now would he?
Supernovas, as I believe has been pointed out elsewhere in the thread,
are a source of oldD.
So now that we've cleared that up, explain again why this linear
relationship would exist at the formation of the earth?
>
> ><SNIP some references>
> >
> >> Apparently, some of the actions needed to produce a homogenous mixture
> >> are:
> >>
> >> stirring
> >> shaking
> >> vibrating
> >> wet mixing
> >>
> >> None of the above actions are considered regular underground activity
> >> in the earth. Heat is not sufficient to produce a homogenous mixture
> >> (see first link above) and, evidently, pressure will not do it,
> >> either, judging from my latest experiment in my kitchen-lab.
> >
> >Well, it depends. I know that uneven heating can lead to convection
> >currents which can cause stirring. For instance, salt gets pretty
> >evenly mixed in boiling water.
>
> again, you're equating the viscosity of boiling water with the
> viscosity of magma.
No, I'm simply pointing out that there are other mechanisms that cause
mixing other than the ones listed above.
>
> >The other element missing from the references you gave is time. Think
> >about stirring for a moment. Obviously just swishing a spoon through
> >cake batter once isn't enough to mix the batter evenly, you have to
> >keep mixing for some length of time. The same thing is true with
> >convection, it takes time. 24 hours might not be enough when creating
> >metal alloys, but maybe 24 million years would be enough.
>
> I don't think that my cake batter would be any better mixed if I left
> the ingredients to sit for several months or years. Ask any woman in
> her kitchen if she thought her cake batter would become nicely mixed
> if she let it sit for a few months or years, and she would question
> your sanity.
>
> I don't think you should invoke time as the solution. To do so only
> makes your theory unfalsifiable, since millions of years is not
> available to test your theory.
We don't have millions of years, of course, but some extrapolation is
possible. For instance consider a paint shaker. Leaving a can of
paint in the shaker for one second won't thouroughly mix the paint,
but maybe 15 minutes would. If we lived in some alternate universe
where 15 minutes was too long of a time we could mix paint for 1
second, 2 seconds, 3 seconds, etc. and see if homogenization increased
as mixing time increased.
>
> snip>
>
> >I don't know that I'm qualified to speak about this, but maybe putting
> >your 'crystals' in a double-boiler over relatively low heat for a few
> >weeks? I realize that may not be practical. Maybe try it for 2
> >hours, then try another batch for 6 hours, and another for 12 and see
> >if the homogenity increases as time increases?
>
> I would need to eliminate the presence of oxygen, too. Any
> suggestions?
I'm not sure this is necessary. You could cover the surface of the
chocolate with some heat-resistant plastic wrap if you were very
careful, perhaps.
> >I do have one question for you. How do you account for the fact that
> >in new flows that have been measured we do see that D/Di is the same
> >for all samples?
>
> using the Rb/Sr measure?
Among others.
And to address the question I've seen you ask elsewhere in the thread,
this is done to validate the theory. One of the predictions we can
make if isochron dating works is getting a zero slope at the time of
solidification; part of science is testing these predictions.
So it's been tried, with Rb/Sr, and we do indeed see that D/Di is the
same in all samples for a melt. How do you account for this?
>
> ----
> zoe
P. Venkman
> On Fri, 23 May 2003 20:56:53 +0000 (UTC), pven...@hotmail.com (P.
> Venkman) wrote:
>
> >muz...@aol.com (zoe_althrop) wrote in message news:<3ecd8ced....@news-server.cfl.rr.com>...
> ><SNIP>
> >
> >> okay, Jon & Company, I have a lot more chocolate morsels on hand. Are
> >> there any other or better experiments that can be done with
> >> them?...quickly, please, before they disappear down my gullet.
> >>
> >
> >OK, here's one for you. Get a large glass and fill it with water.
> >Add 3 drops of red ink to the water on one side of the glass, then add
> >three drops of blue ink to the water on the opposite side of the
> >glass. Cover the glass with some plastic wrap (just to avoid
> >evaporation problems) and leave it for a week.
> >
> >Now pour the contents of the glass into an ice cube tray, stick the
> >tray in the freezer, and make ice cubes. After they're frozen, pull
> >them out. Are the ice cubes all the same color, or do you have some
> >mostly red and some mostly blue cubes?
> >
> >What do you conclude from this experiment?
>
> without even conducting this experiment, I can conclude that it does
> not in any way simulate magma mixing.
> zoe
Why? Why would you say your experiment with the chocolate does an
adequate job of simulating magma mixing, and this one doesn't?
howard hershey
zoe_althrop wrote:
> This is an experiment I've been trying to get around to for the last
> week or so, and I finally found time tonight.
>
> Consider this a mop-up operation on the isochron threads (drum
> roll...adjust my scientific beret here).
>
> Preamble (or postramble): The issue finally came down to whether or
> not a linear relationship between D/Di and P/Di is retained during a
> melt, causing a premelt slope to remain in evidence at time of
> solidification. Right, Jon? Was that where we left off? I believe
> so.
This, believe it or not, is progress. Zoe has finally recognized that
in order for isochrons to be presenting falsely large dates at least one
of the three assumptions of the method must be consistently and causally
false: 1) cogenetic source with a reset of the clock at solidification
-- which requires that D/Di be constant at the time of solidification in
all samples used to generate the isochron line. 2) closed system. And
3) constant rate of radioactive decay. She has chosen to claim that it
is the cogenetic source assumption, and specifically the assumption that
the clock (which depends upon the relationship between D and P) is reset
at each solidification, that is false. Her claim is that the clock is
only partially reset upon each melting, with many samples (but not
modern ones) producing, in the newly solidified rock, not samples with
uniform D/Di but samples with D/Di levels that vary positively with the
amount of P in that sample.
Her claim is that liquified rock (in the liquid state) in a magma
chamber (which is churning around, releasing volitiles retains the
relationship between D and P that existed locally in the rocks it came
from over extended periods of time.
The alternative claim is that the level of D/Di gets homogenized in
magmas and the change in D/Di ratios is entirely due to decay of P
*since solidification*.
There are several simple tests of the implications of her hypothesis
that I can think of. But I am no geologist (never having taken any
course in the science), so others, far more knowledgeable than I am, may
well be able to add others or correct errors I have made here.
1) Lavas should show ages uncorrelated by position in the geological
column but correlated to how many times it has melted and remelted.
2) There should be a large discrepancy in ages between methods where the
daughter nucleides are ions in solution (and thus susceptible to
entrapment and slow diffusion) and methods where the daughter nucleides
are noble gasses. There should also be large discrepancies in dates
using different methods, because the change in slope should be a
consequence of the size of the ions and their chemical natures (features
that affect the rate of diffusion or "fuzzification") rather than due to
the rate of radioactive decay since solidification.
3) There should be a significantly greater number of modern
(historically known and recent) lavas which exhibit positive (as opposed
to negative) slopes on plotting an isochron. [If the slope of the
isochron of a modern lava is zero, as is the case if the clock resets,
then one should expect a roughly equal number of plotted isochrons that
have, insignificantly different from zero but positive and
insignificantly different from zero but negative slopes. If the slope
is a consequence of number of melts and remelts with a bias toward
retaining a positive slope, I would expect an excess of positive slopes
even in modern lavas, since there must be some variability in the number
of melts and remelts.]
4) There should be more variation and less fit to the isochron line as
the 'age' of the sample decreases, since random events (like a
convection current or a steam bubble or hydrodynamic shear during
expulsion can easily disrupt the correlation between P and D levels in a
*liquid* leading to samples with erratic P and D levels.
5) The age of isochrons generated by samples taken from closely spaced
crystals (with differing levels of P and D/Di) should date consistently
younger (because diffusion will have more effect on equalizing D/Di
levels in closely spaced crystals meaning that the slope of such samples
will be closer to zero at solidification) than in samples taken from
more distantly spaced crystals (where the level of P will be more
closely correlated to the local level of D because diffusion will not
have equalized D/Di over this larger area).
Call me a skeptic, but I think Zoe's hypothesis will fail all of these
tests of the consequences of her idea of "retained memory". And all
these are tests on the material in question, not in chocolate.
>
> In order for the linear relationship between D/Di and P/Di to be
> erased, melted rock would have to be reduced to a homogenous state.
> The question is: what does it take for a liquid to become homogenous?
> Pulling up a few sources that deal with homogenization, this is what I
> see:
>
[snip dangerous experiment]
>
> CONCLUSION:
>
> The linear relationship between D/Di and P/Di is retained throughout a
> melt, with some fuzzying of the borders (scatter), sufficient to
> reduce the apparent age of the rock with each succeeding remelt, but
> insufficient to bring the slope back to zero at time of
> solidification.
>
A model is useful only to the extent that it is a valid model and its
results are consistent with observation of the thing being modeled.
[snip]
zoe_althrop
except that coming from my perspective, where the Y-intercept
represents accumulated decay product, NOT oldD/Di, the values at the
intercept would represent the common point of decay for all samples at
a particular point in time.
----
zoe
zoe_althrop
<jo...@fleming-nospam.com> wrote:
>You forgot to answer an important question:
>
> How do you account for the fact that in new flows that
>have been measured we do see that D/Di is the same for all samples?
let's see. What have I learned from you guys so far, and what have I
done with the information? Hmmm...where's my beret. Hang on.
Okay. So far it has been agreed that zero slope covers a range of a
couple million to tens of millions of years. That leads to the
question: If zero slope covers a range of time, how do you know that
your zero slope (or D/Di appearing to be the same) represents
homogeneity? There may very well be a retained premelt linear
relationship between D/Di and P/Di, which linearity just does not show
up because it is hiding within zero slope. Your new flow could be
hiding a few million years worth of D/Di to P/Di, which, if it could
have been visible, would have led you to conclude that the new melt
must really be a few million years old.
So, proffering new flows with zero slopes (or same D/Di) is not
necessarily evidence that the melt is homogenous.
But, it might be asked, why would a recent flow register below "zero"
slope today when, in the past, it has registered a positive slope?
Answer: If a new flow plots a zero slope it would be because
sufficient remelts have occurred to reduce its positive slope to the
unobservable level within "zero" slope.
New question: What would be the mechanism for reducing an original
positive slope to zero?
Let's see. (adjusting beret with furrowed brow)
Answer: Remelts and diffusion could be the mechanism. Remelted rock
would allow the process of diffusion to begin, sufficient to remove
from the lineup X quantity of newD to P, but not sufficient to cause
homogenization. With solidification, the slope would be decreased due
to the fuzzification of the original slope, and with each new
solidification, the slope begins to increase again. But with each
succeeding remelt, there is again a loss of D and P to diffusion. I
am betting that there are crossovers, newD accumulating in one
direction, total D diminishing due to diffusion in the opposite
direction. P being in larger quantity than D, yet diffusing at the
same rate, would continue to be in linear relationship with D, but
appearing to give a younger age over time, due to loss of isotopes to
diffusion.
well now, that was fun. Now comes the hard part -- to test it.
>On Tue, 27 May 2003 01:51:41 +0000 (UTC), muz...@aol.com (zoe_althrop)
>wrote:
>
>>
>>http://www.cosis.net/abstracts/EAE03/04785/EAE03-J-04785.pdf
>>
>>"Rb-Sr whole rock data points obtained from the Tres Piedras Granite
>>yields a distinct isochron for each outcrop sampled. The age and
>>apparent initial Sr87/Sr86 ratios of the Tres Piedras Granite outcrops
>>(approximately 50 square meters of collecting area at each exposure)
>>are as follows: Tres Piedras Granite Type Locality: 1493 +/- 21 Maand
>>0.7183 +/- 0.0006; Tres Piedras Granite – Tusas River Canyon: 1501 +/-
>>44 Maand 0.7145 +/- 0.0013; and Tres Piedras Granite – Tusas Mountain:
>>1661 +/- 17 Maand 0.7102 +/- 0.0071."
>>
>>correct me if I'm wrong, but it appears that the younger ages carry
>>increased scatter:
>>
>>0.7183 +/- 0.0006
>>0.7145 +/- 0.0013
>>0.7102 +/- 0.0071
>
>You are wrong. Those are the Y-intercepts. The greatest scatter is
>in the intermediate age
>
> Age Scatter
> 1493 +/- 21
> 1501 +/- 44
> 1661 +/- 17
let me rephrase: The younger ages carry increased scatter in the
Y-intercepts. This would be a good chance to observe what the
identification of the intercept will do to how the data is
interpreted.
If the Y-intercept is identified as oldD/Di, then you have the problem
of explaining why a cogenetic source shows differences in oldD/Di. If
the Y-intercept is identified as exhausted P, then it makes sense that
there would be different intercepts, depending on how much P the
smallest samples started out with, that has now decayed to near zero.
yes, yes, I know the ratios can make the smallest sample register
higher than a larger sample, but I expect the jump to remain well
within acceptable scatter.
snip>
>But you have presented FAR too little data to make a determination.
now, about THAT, you are correct.
>However, you can of course go ahead and make that supposition. Then
>collect a significant number of published ages ranging from a few
>million years ago to around 4 billion years ago. About 25 to 100
>should do. Make sure that, as much as possible, you include dates
>that cover all the possible dates in the range of a few million years
>ago to around 4 billion years ago; to do this you'll probably have to
>examine 250-500 papers.
it's doable. Do I have the time, though? That's the question.
>Then graph or otherwise analyze the scatter as a function of age.
>Let's assume that you _do _ see a tendency for younger ages to have
>greater scatter Then you analyze the statistical significance. Next
>you consider other possible sources of the trend (such as the fact
>that younger rocks involve measuring smaller differences in isotopic
>makeup) and prepare discussion s of why you think that the effect is
>due to you supposed mechanism rather than other possible mechanisms.
>
>Then get back to us with your results.
(salute) yes, sir! Maybe, sir!
----
zoe
zoe_althrop
<hers...@indiana.edu> wrote:
>
>
>zoe_althrop wrote:
>> On Mon, 26 May 2003 14:54:06 +0000 (UTC), Howard Hershey
>> <hers...@indiana.edu> wrote:
>>
>> snip>
>>
>> zoe wrote:
>>
>>
>>>>the original linear relationship of D/Di to P/Di would come from an
>>>>original formation in which there was no oldD present. Since D
>>>>originates from P, there is no reason to imagine that there was a time
>>>>when D originated all by itself, without the aid of P.
>>>
>>>In the initial outburst of a supernova, but that is besides the point.
>>
>>
>> why is this besides the point? It is exactly the point that needs to
>> be explored.
>
>It is irrelevant because all one ever measures is totalD/Di in samples.
> OldD is not and never was defined as the amount of D present at the
>formation of the earth or present before the earth formed (after all, P
>did not poof into existence either).
how did P come into existence? "I don't know" would actually be a
better answer than strong assertions that lead to incorrect
conclusions.
> OldD/Di is defined as the ratio of
>D/Di present in all samples at the time of last solidification of the
>samples used to create the isochron plot. OldD/Di, defined correctly,
>is empirically estimated from the ratio of D/Di at x = 0 (aka, the
>y-intercept) in an extrapolation (usually, since one rarely actually has
>a data point with no P) of the line of best fit plotted through the
>measured sample data points (with totalD/Di of each sample being the y
>value and currentP/Di being the x value). The reason for the
>y-intercept being a good estimate is that a sample with 0 P will not
>change its level of totalD/Di after solidification since there is no
>source of newD.
neither will a sample whose P has exhausted change its level of
totalD/Di.
>
>>
>> What can you tell me about how supernovas produce D without initial P?
>
>It is formed by the same sort of mechanisms that produce P in supernovae.
what are these mechanisms?
>>>>Therefore, the
>>>>moment that P begins to decay to D, the linear relationship would be
>>>>established, and it is this linear relationship that becomes more and
>>>>more fuzzied by remelts. The more remelts, the younger the rock
>>>>appears to be. The less remelts, the older the rock shows up to be.
>>>
>>>Until, of course, the present time, when examination of lavas from
>>>historically modern volcanos show constant D/Di uncorrelated with P/Di
>>>rather than positive linear relationships.
>>
>>
>> if a historically modern volcano shows a supposedly constant D/Di, or
>> zero slope, this would be explained by my hypothesis as the result of
>> sufficient remelt scatter that would bring the slope down to the zero
>> range. At least for those rocks that have been through sufficient
>> remelts to lose the positive slope.
>
>Why do essentially *all* modern volcanos show this constant D/Di level?
> If your hypothesis were true, shouldn't we be seeing a significant
>number of volcanos (those whose magma contain rocks with fewer remelts)
>which present clear positive slopes correlated with P levels?
you do, but those modern volcanos are not recognized as modern because
they have slopes. A volcano that might have happened a few thousand
years ago will be considered ancient because it shows a slope, and you
have no way to know if it was recent or not.
>And how
>does that explain the consistency with layers and the fossils embedded
>between them? Are you claiming that the entire mantle has solidified
>and remelted equally so that all volcanoes that produce lava indicating,
>say, a age of 100 mybp just happen to have certain fossils above them
>and other fossils below them?
no, the relative dating applies both to the fossil layers and the
igneous layers. But all you can do is say that, according to the laws
of superposition, an igneous layer clearly formed before or after a
fossil layer. When it comes to absolutely dating the fossils,
however, it cannot be done because the absolute dates are not tied to
time since solidification, but are merely a reflection of accumulated
decay product.
It is like trying to tie the age of the oils in a painting on a canvas
to the age of the canvas itself. The oils themselves may have sat
around for years, and you might be able to date them to be, say, 20
years old, but the canvas upon which the oils were placed might have
been manufactured as recently as a year ago. It would be correct to
say that the application of the oil to the canvas happened at a later
time than the acquisition of the canvas itself, yet it would be
incorrect to say that since the oils were placed on the canvas, that
this means the oils are younger than the canvas. The fact is that the
oils can be dated as 20 years old, but their application to the canvas
is "younger" than the manufacture of the canvas itself. Ability to
date the oils as 20 years old does not mean that the canvas is older
than 20 years.
This makes complete sense to me, but guess what, I can guarantee there
will be retorts like "irrelevant" "meaningless" "garbage."
K/Ar dating is known to retain excess argon that is why they prefer
Ar/Ar. But that is no help since Ar/Ar harks back to K/Ar through the
J factor. I would submit that the retention of excess argon is the
equivalent of a retained linear relationship of D/Di to P/Di. And
both methods, K/Ar and Rb/Sr will give similar ages within a RANGE.
>> These ages should not reflect time since solidification. They reflect
>> should reflect accumulated decay product that gets translated into a
>> calculated equivalent of years. Applying the calculated number of
>> years to time since solidification is an unjustified assumption, imo.
>>
>I agree. If your 'chocolate' model is correct, only the Ar method
>should reflect time since solidification.
not even then. Excess argon is a factor in the real world. The only
time all argon is removed would be in the lab where the experiment is
conducted to perfection and it is observed that all excess argon can
indeed be eliminated. But the lab is not the real world. Indeed, see
Jon's argument against my kitchen experiment. It would apply to the
lab as well.
>It should produce the very
>young ages you require and should differ significantly from the ages
>produced by Rb/Sr since the mechanism of "fuzzification" is quite
>different in the two methods.
I don't think the mechanics are different. In one case, the linear
relationship is retained between D/Di and P/Di. In the other case,
the linear relationship is retained between excess argon and K. And
because the ages are within broad ranges of millions of years, the
discrepancies will fit within those ranges.
snip>
----
zoe
zoe_althrop
(John Drayton) wrote:
snip>
>This latest thread has been an absolute smorgasboard from Zoe.
>This is my favourite:
bon appetit, John.
>
> I don't think that honey would represent
> rock the way chocolate more nearly does
Honey differs from chocolate and rock in that it does not start out in
solid form. Saran-wrapped (representing minerals) chocolate pieces
(representing isotopes) can be stacked together in a solid manner
(rocklike), as can minerals be packed together in a solid manner.
Honey cannot do that. I want to compare what happens to the
arrangement of atoms or chips when melting does take place. How long
do they maintain their positions under certain conditions?
----
zoe
zoe_althrop
Venkman) wrote:
snip>
>Supernovas, as I believe has been pointed out elsewhere in the thread,
>are a source of oldD.
you could not possibly be suggesting that I should accept these
assertions without question, could you? Surely, that is not the type
of student you would encourage, now would you?
Again: In what way are supernovas a source of original D, and how is
this original D recognized to be disconnected from P's decay?
>So now that we've cleared that up,
it has NOT been cleared up. Sorry.
>explain again why this linear
>relationship would exist at the formation of the earth?
no explanation until it is cleared up. If it is not cleared up, that
leaves the way open for the default assumption that D comes only from
decay of P. And if that is the case, then the linear relationship
would exist from the moment P starts to decay to D.
snip>
>> I don't think you should invoke time as the solution. To do so only
>> makes your theory unfalsifiable, since millions of years is not
>> available to test your theory.
>
>We don't have millions of years, of course, but some extrapolation is
>possible. For instance consider a paint shaker. Leaving a can of
>paint in the shaker for one second won't thouroughly mix the paint,
>but maybe 15 minutes would. If we lived in some alternate universe
>where 15 minutes was too long of a time we could mix paint for 1
>second, 2 seconds, 3 seconds, etc. and see if homogenization increased
>as mixing time increased.
but there is no shaking activity like in a paint shaker that occurs
under the earth's crust.
snip>
----
zoe
zoe_althrop
<jo...@fleming-nospam.com> wrote:
snip>
>Of course, if you're trying to model what happens in magma over
>thousands of years, nothing you can do in your kitchen
or the laboratory?
>is valid. You
>can demonstrate what _might_ happen in various kitchen experiments,
>but nothing you can do in your kitchen will establish what _does_
>happen.
same for the laboratory?
snip>
----
zoe
zoe_althrop
<jo...@fleming-nospam.com> wrote:
snip>
>I bet she thinks that the diffusion rate depends on whether you
>started with a liquid or a solid.
no, I am not comparing the diffusion rate between natural liquids and
melted liquids. I am interested in what happens to the structural
arrangements of atoms/molecules in a solid when it melts. Will these
arrangements be retained under conditions that do not actively mix?
If you start with a liquid that never was a solid, you have no
premelt version for comparison of the structure and arrangement of
that liquid's atoms/molecules.
----
zoe
zoe_althrop
<mat...@ix.netcom.com> wrote:
snip>
>Are you willing to deal with the kinds of temperatures involved?
>Like well above 1,000? And will you wait for a few years while
>you observe? Anyway, the mantel is liquid, not solid.
of course the mantle is liquid. I am talking about remelts when I say
"solid". A remelt implies the rock was once solid, right?
----
zoe
Harlequin
server.cfl.rr.com:
After all this time, Zoe still cannot comprehend the concept
of experimental error correctly. Zoe your objections
continue to get lamer and lamer. You are a case study
in utter desperation.
> On Tue, 27 May 2003 13:01:38 +0000 (UTC), Jon Fleming
> <jo...@fleming-nospam.com> wrote:
>
>>You forgot to answer an important question:
>>
>> How do you account for the fact that in new flows that
>>have been measured we do see that D/Di is the same for all samples?
>
> let's see. What have I learned from you guys so far, and what have I
> done with the information? Hmmm...where's my beret. Hang on.
>
> Okay. So far it has been agreed that zero slope covers a range of a
> couple million to tens of millions of years.
Remember that this "zero slope" you are refering to is nothing more
than shorthand for zero plus or minus [some small number]. And keep
in mind that that small number is small enough not to have a major
major impact on the date. (Major impact being defined something like
making a 4000 year old rock into a 4,000,000,000 year old rock.)
> That leads to the
> question: If zero slope covers a range of time, how do you know that
> your zero slope (or D/Di appearing to be the same) represents
> homogeneity? There may very well be a retained premelt linear
> relationship between D/Di and P/Di, which linearity just does not show
> up because it is hiding within zero slope. Your new flow could be
> hiding a few million years worth of D/Di to P/Di, which, if it could
> have been visible, would have led you to conclude that the new melt
> must really be a few million years old.
This is wrong on just so many levels. You continue to assume that
scientists are nothing more than idiotic protectors of dogma or
at least you make comments which could only be true if scientists
were nothing more than idiotic protectors of dogma. The behavior
of isotopes is something that is well known and well understood both
from a theoretical point of view and from empirical data.
Measurements of isotopes did not start when isochrons were dreamed
up and they _many_ applications of isotopic data.
Has it ever occured to you that scientist are full capable of melting
two substances together and taking measurements before and after?
Of course if you are ever convinced of this you will find yet
another excuse.
"There may very well be a retained premelt linear
relationship between D/Di and P/Di, which linearity just does not show
up because it is hiding within zero slope."
Yet another Chez Watt worthy statement. How do you "hide" a
significantly non-zero slope in slope with no significant deviation
from zero? Do you have a physically possible explanation for
the formation of linearity? Do you have any sort of experimental
data?
> So, proffering new flows with zero slopes (or same D/Di) is not
> necessarily evidence that the melt is homogenous.
This is silly. "Zero slope" again means zero plus or minus some small
number. If the real slope is greater than exactly zero but smaller to
or equal to tha small number than the melt is insignificantly different
than homogenous. In plain not-tech English, it is trivially different
that completely homogenous. This is not enough to make a 4,000 year
old rock look like a 4 billion year old rock.
[snip the rest since IMHO it is rather pointless]
Of course we again must come to the questions that you continue to
utterly ignore.
Rb-Sr Isochrons are hardly the only type of isochrons of its type.
Why do the other isochrons methods of the same time agree?
Then there are other self-testing methods like Pb-Pb isochron
(which inspite of the name is a completely different method),
concordia-discordia, and Ar-Ar. Why do these methods which are
all independent of each other agree with each other?
Then why do they agree with the relative geologic timescale which
was developed (at least in course detail) in the 19th century
before radiometric dating was even conceived?
Why does Ar-Ar dating when presented with samples from a historic
eruption nail the date of that eruption dead-on?
If radiometric dating does not give the real age of the rock then
how come rocks of x age generally show the same polarity of
Earth's magnetic field. (If you don't know sometimes the polarity
of the magnetic field reverses itself and a compass needle that would
have pointed north before the change would point south after the
change.)
One could go on a very long time giving such examples.
None of this would have happened if radiometric dating did not
work.
--
Anti-spam: replace "usenet" with "harlequin2"
"...Everybody has opinions: I have them, you have them. And we are all
told from the moment we open our eyes, that everyone is entitled to
his or her opinion. Well, that's horsepuckey, of course. We are not
entitled to our opinions; we are entitled to our _informed_ opinions.
Without research, without background, without understanding, it's
nothing. It's just bibble-babble...."
- Harlan Ellison
Martin Crisp
(in message <3ed6cbf7....@news-server.cfl.rr.com>):
> On Tue, 27 May 2003 12:12:12 +0000 (UTC), bitbu...@hotmail.com
> (John Drayton) wrote:
>
> snip>
>
>> This latest thread has been an absolute smorgasboard from Zoe.
>> This is my favourite:
>
> bon appetit, John.
>
>>
>> I don't think that honey would represent
>> rock the way chocolate more nearly does
>
> Honey differs from chocolate and rock in that it does not start out in
> solid form. Saran-wrapped (representing minerals) chocolate pieces
Chocolate differs in that it starts in a variety of _phases_ and
when those are combined correctly ends up as a solid.
Or at least so the Cadbury factory 1.5 suburbs over claims. "A
glass and a half of full cream milk in every 200g block" -> about
300ml (combined) of solids suspended in liquid at the 'start' of
the process.
> (representing isotopes) can be stacked together in a solid manner
> (rocklike), as can minerals be packed together in a solid manner.
> Honey cannot do that. I want to compare what happens to the
<goes to cupboard, grabs 500g jar of 'home brand' "Pure Australian
Honey" - smells like leatherwood, but jar doesn't specify.>
About 25% of the jar has been used.
<turns jar upside down>
<waits>
Current progress indicates that the contents of the jar are a
solid. [lots of sugar crystals... we could be here for a while...]
And it's not even particularly cold here. I could see if there's
any honey at Mawson base, and ask what it does @ -30 or so Celcius,
if you like. [Would take a little while to organise, since I'm not
directly associated with the Antarctic Division]
> arrangement of atoms or chips when melting does take place. How long
> do they maintain their positions under certain conditions?
"When a crystal melts the long-range order breaks down, but some
short-range order persists in the liquid at temperatures right up
to the boiling point. For _a few_ substances a large amount of one-
or two-dimensional order exists in the liquid. These are the
fascinating materials known as liquid crystals. The breakdown of a
lattice requires energy..."
p 11, _First Year Chemistry_, J. M. Coxon, J.E. Fergusson, & L.F.
Phillips (of Dept Chemistry, Uni of Canterbury, Auckland, NZ);
Edward Arnold (Publishers) Ltd. 1980.
[_underline_ added]
Put another way: comparing rock to chocolate is invalid because of
the variability, across various substances, exhibited in the
behaviour you claim to be interested in. Why don't you watch what
happens to a drop of food colouring placed into the liquid wax at
the top of a candle?
BTW: I've given the honey a good 3/4 of an hour, and there's no
sign of movement (at any scale I can measure at least...).
Have Fun
Martin
--
aa #1792
Almost always SMASHed
Martin Crisp
(in message <ce43f6e.03052...@posting.google.com>):
> Martin Crisp <Spam....@tesseract.com.au> wrote in message
> news:<0001HW.BAF914BE...@news.ozemail.com.au>...
>> On Tue, 27 May 2003 12:58:47 +1000, zoe_althrop wrote
>> (in message <3ed2cbf5....@news-server.cfl.rr.com>):
>>
>> [rest excised]
>>
>>> If chocolate melts to the point where its molecules are free to
>>> diffuse, then you no longer have chunks. Yet the molecules that
>>> constituted the once-unmelted chunks remained in essentially
>>> their original positions.
>>
>> A sort-of frozen-in-place liquid, perhaps?
>>
>> Have Fun
>> Martin
>
> This latest thread has been an absolute smorgasboard from Zoe.
> This is my favourite:
>
> I don't think that honey would represent
> rock the way chocolate more nearly does
I must have missed that.
Zoe is great value, and for different reasons (as so many have said
during the isochron threads) so are her knowledgeable respondents
(me not included!!)