H-bomb timimg

1 view
Skip to first unread message

Barnaby Finch

unread,
May 13, 1999, 3:00:00 AM5/13/99
to
Can someone knowledgable in the physics of thermonuclear devices explain
the timeline of a two-stage device?

I know that the fission primary delivers its energy exponentially - the
majority of the energy is released in the last few shakes (a shake being
roughly 10 ns - the time it takes neutrons in oralloy or Pu plasma to
cover their Mean Free Path). I've read how thermodynamic equilibrium is
achieved quite rapidly. How rapidly? How long does it take for this energy
(mostly soft x-rays) to begin abalating the high Z pusher/tamper?

Now, this (de facto) reverse rocket begins to compress the fusion fuel
(usually LiD) and eventually compresses the Pu "sparkplug" enough so that
it undergoes fission, spewing copious quantaties of neutrons into the
highly compressed fuel and initiating fusion. Does the fusion reaction
truly start at this point, or has the compression itself raised the fuel
temperature enough to initiate the reaction, albeit not as vigorously as
when stimulated by Pu neutrons? I know that to achieve good compression
the fusion fuel should not be prematurely heated - but it must be very hot
by the time the compression is complete.

I read somewhere that the fusion reaction is extremely fast - the fuel is
mostly burned within 2 shakes (20 ns). Does the exponential nature of the
energy release from the Pu sparkplug have a bearing on this? It seems, to
a novice like me, that the fusion fuel is "waiting" for the
(comparitively) slow fission parts of the device to do their thing.

Lastly, how long does it take for the fast fission of the tamper/pusher to
take place (assuming that the t/p is U238 or HEU)? I assume, that by the
time the Pu sparkplug has completly fissioned, the primary must be at
least _beginning_ to disassemble the whole device. Or perhaps no
significant kinetic momentum has been imparted until t/p fission is
complete?

I guess what I'm looking for is a microsecond-by-microsecond account of
the state of the various parts of the device. Please forgive the
long-windedness and confusion of this post - there's obviously some basic
stuff I'm not getting.

Barnaby

Paul DeMone

unread,
May 14, 1999, 3:00:00 AM5/14/99
to
Barnaby Finch wrote:
>
> Can someone knowledgable in the physics of thermonuclear devices explain
> the timeline of a two-stage device?

I doubt anyone who is professionally knowledgable in this topic
will risk violating their secrecy oath. Anyone else is just
making an educated guess.

>
> I know that the fission primary delivers its energy exponentially - the
> majority of the energy is released in the last few shakes (a shake being
> roughly 10 ns - the time it takes neutrons in oralloy or Pu plasma to
> cover their Mean Free Path).

One paper from Los Alamos described a computer simulation of a 24 KT
plutonium device with an initial alpha of 1.32e8 s-1 and initial
neutron population of 2.35e17 at time t = 0. The device doesn't
start serious energy release until 100 ns. Max energy release rate
is 0.725 KT/ns at ~130 ns. The device goes subcritical (negative
alpha) at about 140 ns at which time 71% of the yield has been
generated. By 160 ns 90% of the yield is released.

[snip]


> I read somewhere that the fusion reaction is extremely fast - the fuel is
> mostly burned within 2 shakes (20 ns). Does the exponential nature of the
> energy release from the Pu sparkplug have a bearing on this? It seems, to

The rapid energy release in the secondary is the result of the immense
compression of the fusion fuel. Fusion reaction rate increases
quadratically with particle density. The fusion fuel is compressed
to a density probably two or three orders of magnitude higher than
normal.

> a novice like me, that the fusion fuel is "waiting" for the
> (comparitively) slow fission parts of the device to do their thing.

The fusion and fission reactions within the secondary are synergistic
and accelerate each other so it is better not to try to think of them
as independent. The initial "slow" fissioning in the sparkplug heats
the fusion fuel as well as converting Li6 to H3. Once the temperature
and tritium concentration is high enough then fusion rate increases to
the point in which the surplus high energy fusion neutrons rapidly
increase the fission rate in the sparkplug. This in turn heats the
fusion fuel even faster and so on.

>
> Lastly, how long does it take for the fast fission of the tamper/pusher to
> take place (assuming that the t/p is U238 or HEU)? I assume, that by the
> time the Pu sparkplug has completly fissioned, the primary must be at
> least _beginning_ to disassemble the whole device. Or perhaps no
> significant kinetic momentum has been imparted until t/p fission is
> complete?

The weapon casing is likely disassembling itself from the outward pressure
of the inwardly ablating radiation liner long before the primary matter
reaches it.

[snip]

--
Paul W. DeMone The 801 experiment SPARCed an ARMs race of EPIC
Kanata, Ontario proportions to put more PRECISION and POWER into
dem...@mosaid.com architectures with MIPSed results but ALPHA's well
pde...@igs.net that ends well.

Carey Sublette

unread,
May 14, 1999, 3:00:00 AM5/14/99
to

In article <bumples-1305...@c1p152.idl.pe.net>, you say...

> Can someone knowledgable in the physics of thermonuclear devices explain
> the timeline of a two-stage device?
>
> I know that the fission primary delivers its energy exponentially - the
> majority of the energy is released in the last few shakes (a shake being
> roughly 10 ns - the time it takes neutrons in oralloy or Pu plasma to
> cover their Mean Free Path). I've read how thermodynamic equilibrium is
> achieved quite rapidly. How rapidly? How long does it take for this energy
> (mostly soft x-rays) to begin abalating the high Z pusher/tamper?

Thermodynamic equilibrium is established on a time scale equal to the
time required for energy to be exchanged between particles in the
system, when a few scale intervals have passed, any initial
irregularities will have vanished. In the case of a radiation dominated
system (like the interior of a TN device) this time scale is the time
between the emission and the absorption of an average photon, 1-2
nanoseconds. The time scale is a little longer to establish equilibrium
throughout the hohlraum since several emission and absorption events are
required to transport the energy from the primary to the most distant
part of the channel.

Ablation begins as soon as the thermal energy arrives. In a modern
boosted weapon the fission process is comparatively sluggish, and the be
tamper and HE becomes ionized and transparent vey rapidly, so there is
an exponentially increasing flow of radiation into the channel for much
of the fission process, followed by an even more rapid increase when
boosting kicks in and the core disassembles to subcritical.

This process is parobably essential for achieving efficient extremely
high compressions in modern designs (much more so than in the 50s
systems).

>
> Now, this (de facto) reverse rocket begins to compress the fusion fuel
> (usually LiD) and eventually compresses the Pu "sparkplug" enough so that
> it undergoes fission, spewing copious quantaties of neutrons into the
> highly compressed fuel and initiating fusion. Does the fusion reaction
> truly start at this point, or has the compression itself raised the fuel
> temperature enough to initiate the reaction, albeit not as vigorously as
> when stimulated by Pu neutrons?

The spark plug neutrons contribute nothing of consequence to the
secondary fusion reaction. It is the heat of the fission that ignites
vigorous fusion.

> I know that to achieve good compression
> the fusion fuel should not be prematurely heated - but it must be very hot
> by the time the compression is complete.

A low grade fusion reaction probably starts in the fusion fuel prior to
spark plug ignition.

In the most modern designs like the W-87 and W-88 it is possible that a
fission spark plug is not used, with DT gas in the center substituting
as the igniter, heated to ignition by compression.

> I read somewhere that the fusion reaction is extremely fast - the fuel is
> mostly burned within 2 shakes (20 ns). Does the exponential nature of the
> energy release from the Pu sparkplug have a bearing on this?

Not really. When the fusion fuel gets hot enough, it undergoes a self-
heating runaway which burns the fuel up in 5-30 nanoseconds. The spark
plug serves the role of simply heating the fuel to this point.

In the early TN designs the spark plug probably heated the entire mass
of fuel to the ignition point, which then burned while confined by the
tamper inertia. The most modern systems may achieve conditions
sufficient for a true thermonuclear detonation, in which a TN shock wave
spreading from the center consumes the fuel mass without requiring
inertial confinement by the tamper.

>It seems, to


> a novice like me, that the fusion fuel is "waiting" for the
> (comparitively) slow fission parts of the device to do their thing.

Yep.

The fission process in a spark plug doesn't take very long, BTW. A huge
population of neutrons is present from the primary fission, so it begins
exponential multiplication from an already high level the instant it
goes critical. It is compressed to densities several times greater than
a high explosive system with correspondingly shorter generation times.

> Lastly, how long does it take for the fast fission of the tamper/pusher to
> take place (assuming that the t/p is U238 or HEU)? I assume, that by the
> time the Pu sparkplug has completly fissioned, the primary must be at
> least _beginning_ to disassemble the whole device. Or perhaps no
> significant kinetic momentum has been imparted until t/p fission is
> complete?

The secondary implosion and TN burn is completed before the expanding
core hits it and begins brekaing it up. In an extremely compact design
they would cut this as close as possible.

> I guess what I'm looking for is a microsecond-by-microsecond account of
> the state of the various parts of the device. Please forgive the
> long-windedness and confusion of this post - there's obviously some basic
> stuff I'm not getting.
>
> Barnaby

These things vary with the sophistication of the device, and with a
number of design choices made by the developers (and to some extent on
the size of the device).

Using the very convenient timeline found in _Fourth Generation Nuclear
Weapons, The Physical Principles Of Thermonuclear Explosives,
Inertial Confinement Fusion, And The Quest For Fourth Generation Nuclear
Weapons_, by Andre Gsponer and Jean-Pierre Hurni, Fifth Edition: March
1999, we have:

Event Time (nanoseconds)
Primary
Chain Reaction 150 - 300
Boosting DT Burn 1 - 4
Thermal X-ray pulse 10 - 50
Fission Core Disassembly 10 - 50
Full disassembly 500 - 2000

Primary/Secondary
X-ray time of flight 1
Neutron time of flight 20
Shock wave arrival time 1000
X-ray thermalization within hohlraum 10

Secondary
Ablative compression 100 - 500
Spark plug chain reaction 10 - 30
Thermonuclear Burn 3 - 20
Fusion fuel disassembly 3 - 20

The order of listing sort of indicates very roughly their temporal
sequence, but most of them overlap each other to varying degrees. The
exact values of any of them is somewhat dependent on the specific design
under discussion.

There are considerable differences between the design of Ivy Mike, and
the recently famous W-88.

Ivy Mike's secondary implosion took 1-2 microseconds, the smaller more
recent designs are just a few hundred nanoseconds.

I could discuss for each of these event the variables that affect their
duration, and how they relate to each other, but I won't right now.

Carey Sublette


NB: This excellent excellent report by Gsponer and Hurni is currently
available, specifics below:

ISBN: 3-933071-02-X. 183 pages, 25 figures, 4 tables, 528 references
(Fifth corrected and expanded version of a report first distributed
at the 1997 INESAP Conference, Shanghai, China, September 8--10, 1997.)

<P>Orders should be sent to IANUS, ia...@hrzpub.tu-darmstadt.de, or by
fax
to No.\ (+49) 6151-16-6309.

<P>Copyright, 1997, 1998, 1999. INESAP, c/o IANUS, Darmstadt University
of
Technology, Germany. All rights reserved. ISBN: 3-933071-02-X. Price:
$20 + postage.

Reply all
Reply to author
Forward
0 new messages