Reference Enthalpy

[Hazem]

If you have any thoughts about the reference enthalpy, please make them known.

 

Over the years we have witnessed the major simulation companies experiment with the reference enthalpy and it took several of them years to get it right.  Pro/II had a major error which led to a discontinuity in the enthalpy at 0°F and they quickly had to withdraw a release and re-release it again.  That was sometime in the mid 90’s.

 

Hysys never had a discontinuity but on occasion, there were peculiar answers that came out of lost work analysis around processes that had reactors.

 

Aspen Plus, always had it right.

 

Speaking of reactors, what I would like is the following:  If a reaction takes place at 1 bar and 298.15°K and goes to completion

 

A === > B

 

Then, the enthalpy of the outlet steam minis the enthalpy of the inlet stream has to be equal to the heat of the reaction.

 

For that to happen, it seems to me that the reference enthalpy at 298.15 and 1 bar, for a pure component has to be the heat of formation.  If we can get that to work, it will save us tremendous trouble later on as we try to calculate heat of reactions.  Furthermore, it seems to me (and correct me if I am wrong) that when calculating heats of reaction at temperatures and pressures other than 1 bar and 298.15K, the enthalpy and heat of formation would be corrected using the normal procedure of subtracting the enthalpy departure.  We do not have to correct the heat of formation separately, it is already in there.

 

I will be experimenting with this today, but please let me know your thoughts.

 

Regards

Hazem

 

[Rafil Elyas]

HYSIM had absolute zero as a reference point.

 

This was changed to heat of formation at 25C for HYSYS.

I don't think the enthalpy problems you encountered were due to the basis selection.

[Rahul Anantharaman]

Hi Hazem....

Heat of formation at 25 C would be the ideal basis for reacting systems. I would be preferrable to give the user a choice of either this, absolute zero or any other case - with Hf at 25 C being the default basis. It is not all that difficult to change the basis. The basis is "stored" in the constant A in the Ideal Gas Heat Capacity Equation
H(ideal) = A + BT + CT^2+DT^3.

One of the things I would probably do is extend this to have a basis such that negative enthalpies are not encountered in the flowsheet. But that is not what anyone else would be interested in.

Rahul

--

Check out my blog @ http://kuchchi.blogspot.com

[Lynn McGuire]

Hi Hazem,

 

The default enthalpy reference state for DESIGN II for Windows is Ideal

Gas at 32 F.  The default entropy reference state for DESIGN II for

Windows is Ideal Gas at 32 F and 14.696 psia.

 

DESIGN II for Window also has two optional enthalpy reference states,

Saturated Liquid at -200 F (API) and Saturated Liquid at 32 F (ASME). 

The corresponding entropy reference states are also at 14.696 psia.

 

Sincerely,

Lynn McGuire

President

WinSim Inc.

www.winsim.com

 

[Hazem]

Thank you Rahul, Thanks Rafil.

 

I just finished the stoichiometric reactor and tested it. It is working just fine with the Heat of Formation as the reference enthalpy.  It is so nice not to worry about heats of reaction.  The reactor does the material balance and a standard enthalpy call calculates the enthalpy of the effluent and it is all done.

 

By the way, can a stoichiometric reactor be solved backwards?

 

Any volunteers for our Kinetic reactor?  Anybody has LSODE is a language other than Fortran or C++?  Maybe LSODE is an overkill, anybody has the fourth order Runge-Kutta?

 

Regards

Hazem

---

[Hazem]

Thank you so much Lynn for your reply.  I almost forgot about entropy.

 

Regards

Hazem

 

---

From: op...@googlegroups.com [mailto:op...@googlegroups.com] On Behalf Of Lynn McGuire
Sent: Friday, April 07, 2006 11:49 AM
To: op...@googlegroups.com
Subject: Re: Reference Enthalpy

 

Hi Hazem,

[Samuel Cartaxo]

Maybe, what we need is here:

Ordinary and Partial Differential Equations (ODEs and PDEs)

Numerical Methods in Pascal page
http://www-rab.larc.nasa.gov/nmp/nmpIndex.htm#Calculus 

Runge-Kutta ODE Algorithm
www.delphiforfun.org/Programs/Math_Topics/rungekutta.htm 

Category	Math Description	Source(s)
bsstep	integrate ODEs, Bulirsch-Stoer step	recipes
mmid	integrate ODEs by modified midipoint method	recipes
odeint	integrate ODEs with accuracy monitoring	recipes
rk4	integrate one step of ODEs, fourth-order Runge-Kutta	recipes
rkdumb	integrate ODEs by fourth-order Runge-Kutta	recipes
shoot	solve two-point boundary value problem by shooting	recipes
shootf	solve two-point boundary value problem by shooting to a fitting point	recipes
solvde	two-point boundary value problem, solve by relaxation	recipes
sor	elliptic PDE solved by successive overrelation method	recipes

2006/4/7, Hazem <Hazem_...@hotmail.com>:

--
Prof. Dr. Samuel Jorge Marques Cartaxo
Department of Chemical Engineering
Center of Technology
Federal University of Ceará

"This message is intended exclusively for its
addressee and may contain information that is
confidential, protected by a professional
privilege or which disclosure is prohibited by
law. Unauthorized use of such information is
prohibited and subject to applicable penalties."

[Hazem]

Professor Sam,

 

Please grab the RK4 as quickly as possible, before it disappears.  This is 50% (at least) of a kinetic reactor model.  I’ll review some kinetics tonight to see what else do we need but if somebody can save me the trouble and give us a refresher course in kinetics, it will much appreciated.

 

Regards

Hazem

 

---

[Matt Henley]

Samuel,

If you are looking for numerical methods for lazarus, I invite you to explore the lazarus\fpcsrc\packages\extra\numlib directory in lazarus.  Marco Van de Voort (one of the freepascal authors/maintainers) is a chemical engineer although he does programming now.. While in college he ported the above numerical library to fpc/lazarus.  Some of the docs are still in dutch and he can translate if needed.  Here is a quote:

"I also have some experience with numeric calculating in pascal, and
inherited a pascal lib of relevant numerical routines from my Uni, so be
sure to also run numerical math problems over this list. (it's in
packages/extra/numlib, but unfortunately the docs are still untranslated in
Dutch. I can pick out specific routines and translate/document them if you
need them)"

Here are some of the routines:

{Generic determinant}
procedure detgen(n, rwidth: ArbInt; var a, f: ArbFloat; var k, term: ArbInt);

{determinant symmetrical matrix}
procedure detgsy(n, rwidth: ArbInt; var a, f: ArbFloat; var k, term: ArbInt);

{determinant of a symmetrical positive definitive matrix}
procedure detgpd(n, rwidth: ArbInt; var a, f: ArbFloat; var k, term: ArbInt);

{determinant of a generic bandmatrix}
procedure detgba(n, l, r: ArbInt; var a, f: ArbFloat; var k, term:ArbInt);

{determinant of a symmetrical positive definitive bandmatrix}
procedure detgpb(n, l: ArbInt; var a, f: ArbFloat; var k, term:ArbInt);

{determinant of a tridiagonal matrix}
procedure detgtr(n: ArbInt; var l, d, u, f: ArbFloat; var k, term:ArbInt); 

lots of eigenvector code

{Calc inverse for a matrix with unknown symmetry. General version. }
procedure invgen(n, rwidth: ArbInt; var ai: ArbFloat; var term: ArbInt);

{Calc inverse for a symmetrical matrix}
procedure invgsy(n, rwidth: ArbInt; var ai: ArbFloat; var term: ArbInt);

{Calc inverse for a positive definite symmetrical matrix}
procedure invgpd(n, rwidth: ArbInt; var ai: ArbFloat; var term: ArbInt);

{ Determine natural cubic spline "s" for data set (x,y), output to (a,d2a)
 term=1 success,
     =2 failure calculating "s"
     =3 wrong input (e.g. x,y is not sorted increasing on x)}
procedure ipffsn(n: ArbInt; var x, y, a, d2a: ArbFloat; var term: ArbInt);

{calculate d2s from x,y, which can be used to calculate s}
procedure ipfisn(n: ArbInt; var x, y, d2s: ArbFloat; var term: ArbInt);

{Calculate function value for dataset (x,y), with n.c. spline d2s for
x value t. Return (corrected) y value.
s calculated from x,y, with e.g. ipfisn}
function  ipfspn(n: ArbInt; var x, y, d2s: ArbFloat; t: ArbFloat;
                 var term: ArbInt): ArbFloat;

{Calculate n-degree polynomal b for dataset (x,y) with n elements
 using the least squares method.}
procedure ipfpol(m, n: ArbInt; var x, y, b: ArbFloat; var term: ArbInt);   

{Solve first order, starting value, differential eqs,
Calc y(b) for dy/dx=f(x,y) and y(a)=ae}

Procedure odeiv1(f: rfunc2r; a, ya: ArbFloat; Var b, yb: ArbFloat;
                 ae: ArbFloat; Var term: ArbInt);

{ The same as above, for a set of equations. ya and yb are vectors}
Procedure odeiv2(f: oderk1n; a: ArbFloat; Var ya, b, yb: ArbFloat;
                 n: ArbInt; ae: ArbFloat; Var term: ArbInt);      

{Calculates inproduct of vectors a and b which have N elements. The first
element is passed in a and b}
Function omvinp(Var a, b: ArbFloat; n: ArbInt): ArbFloat;

{Multiplication of two matrices C=AxB }
Procedure omvmmm(Var a: ArbFloat; m, n, rwa: ArbInt;
                 Var b: ArbFloat; k, rwb: ArbInt;
                 Var c: ArbFloat; rwc: ArbInt);

{Multiplication of a matrix(A) with a vector(B), C=A x B}
Procedure omvmmv(Var a: ArbFloat; m, n, rwidth: ArbInt; Var b, c: ArbFloat);

{Calculate 1-Norm of matrix A}
Function omvn1m(Var a: ArbFloat; m, n, rwidth: ArbInt): ArbFloat;

{Calculate 1-Norm of vector A}
Function omvn1v(Var a: ArbFloat; n: ArbInt): ArbFloat;

{Calculate 2-Norm of vector A}
Function omvn2v(Var a: ArbFloat; n: ArbInt): ArbFloat;

{Calculate Frobenius-Norm of mxn matrix A}
Function omvnfm(Var a: ArbFloat; m, n, rwidth: ArbInt): ArbFloat;

{Calculates maximum (infinite) norm of mxn matrix a}
Function omvnmm(Var a: ArbFloat; m, n, rwidth: ArbInt): ArbFloat;

{Calculates maximum (infinite) norm of n-Vector }
Function omvnmv(Var a: ArbFloat; n: ArbInt): ArbFloat;

{Transponate mxn matrix A  (which was declared rwa bytes wide), put
it to C (rwc was declared elements wide)}
Procedure omvtrm(Var a: ArbFloat; m, n, rwa: ArbInt; Var c: ArbFloat;
                        rwc: ArbInt);

{Find the all roots of the binomial eq. x^n=a, with "a" a complex number}

Procedure roobin(n: ArbInt; a: complex; Var z: complex; Var term: ArbInt);

{Find root point of f(x)=0 with f(x) a continuous function on domain [a,b]
 If f(a)*f(b)<=0 then there must be (at least) one rootpoint}

Procedure roof1r(f: rfunc1r; a, b, ae, re: ArbFloat; Var x: ArbFloat;
                 Var term: ArbInt);

{Determine all zeropoints for a given n'th degree polynomal with real
coefficients}

Procedure roopol(Var a: ArbFloat; n: ArbInt; Var z: complex;
                 Var k, term: ArbInt);

{Find roots for a simple 2th degree eq  x^2+px+q=0 with p and q real}

Procedure rooqua(p, q: ArbFloat; Var z1, z2: complex);

{Roofnr is undocumented, but verry big}

Procedure roofnr(f: roofnrfunc; n: ArbInt; Var x, residu: ArbFloat; re: ArbFloat;
                 Var term: ArbInt);

{ term : 1     succesful termination
         2     Couldn't reach the specified precision
               Value X is the best one which could be found.
         3     Wrong input
         4     Too many functionvalues calculated, try to recalc with the
                calculated X
         5     Not enough progress. Possibly there is no solution, or the
               solution is too close to 0. Try to choose a different
               initial startingvalue
         6     Process wants to calculate a function value outside the by
               "deff" defined area.

[Samuel Cartaxo]

Thank you very much for the information.

The library is huge and I never heard about it.

A translation to English of the documentation would be very much welcome.

regards

2006/4/7, Matt Henley <nwm...@gmail.com>:

Federal University of Cear�

"This message is intended exclusively for its
addressee and may contain information that is
confidential, protected by a professional
privilege or which disclosure is prohibited by
law. Unauthorized use of such information is
prohibited and subject to applicable penalties."

[Matt Henley]

He mentioned it to me one day and I promptly forgot.. your table reminded me.

One thing to note..he didnt offer to translate en masse.. just specific sections.. although I notice that quite a bit is at least somewhat documented in english.

[J&S Dudley]

Is there a specification for the ODE solver available? From my experience I would be willing to translate the ROCK2 Runge-Kutta code to Pascal, providing that the Pascal is Level 1 (i.e., the dimension of the array is not part of the array specification, which used to be the problem with Pascal - see Numerical Recipes 1st Edition). ROCK2 performs well for moderately stiff differential equations, and is therefore a good compromise between 'classical' Runge Kutta (poor performance on stiff problems) and Gear methods (poor performance on non-stiff methods). If there is no specification, I'll do a port of the Fortran code. Better still would be looking to support the Fortran directly - I find

 

Y = Y + H * DY

 

easier than

 

for i = 1 to n do

  y(i) := y(i) + h * dy(i);

 

especially as the equations get more involved.

 

Does Pascal now support declaration initialisation? In Fortran 77, this was the DATA statement.

[Hazem]

Thank you Mr. Dudley, I think Sam or Matt or somebody else can answer your question about Pascal better than me.

 

You got me curious though.  I suggested RK4 because my understanding is that it will work for up to 8 to 12 reactions.  If there is something better but still not as involved as LSODE (that’s the most recent version of Gear’s method) I would say let’s go for it.

 

LSODE is available in Fortran and C++ for free.  I converted it to VB.NET but could not get it to work : (

 

Regards

Hazem

---

[Samuel Cartaxo]

Thanks for your addition.

About the method, IMO, if it is more stable than RK, you are welcome to translate it to pascal.

The operation you mentioned is a particular kind of operator modification (could operator  overloading). Free Pascal Compiler does this well in order you can use directly + and * with arrays.

Each pascal unit (source file) may have an optional initialization and finalization sections that work as the names indicate.

Regards

Samuel

2006/4/7, J&S Dudley < sbdp...@blueyonder.co.uk>:

[Evgenii Rudnyi]

> If you have any thoughts about the reference enthalpy, please make them
> known.

You may want to look at the Standard Element Reference used by SGTE

A.T. Dinsdale. SGTE data for pure substances. CALPHAD, 1991, v. 15, N
4, p. 317-425.

In my view, it was pretty a successfull idea.

Evgenii
--
http://Evgenii.Rundyi.Ru/

[Hazem]

Thank you so much Evgenii. The article you cited is available online for
$30.00 and is about 6 MB.

DO you have it? If not, I'll see if Rice University is willing to help.

Regards
Hazem

-----Original Message-----
From: op...@googlegroups.com [mailto:op...@googlegroups.com] On Behalf Of

[Evgenii Rudnyi]

> Thank you so much Evgenii. The article you cited is available online for
> $30.00 and is about 6 MB.

Well, the knowledge is expensive :)

> DO you have it? If not, I'll see if Rice University is willing to help.

Unfortunately, not. Once, I used to have a xerox copy but it was long
time ago.

I guess that the simplest is to find someone from university that has a
subscription. Well, at least it is available online, note that not all
the old journals are online.

So, basically there were two lines of the development. First, is to use
the enthalpy of formation (see JANAF and other databases). This can be
done but according to the definition the reagents and products in the
reaction of the formation should be at the same temperature. This
causes strong headache when one starts using this simple idea in
practice. Again, look at how it has been done at the JANAF tables.

One example. If you take the Gibbs energy of formation, then its
derivative in temperature gives the entropy of formation and not the
absolute entropy.

The solution to this is to fix the standard states at a particular
temperature and pressure. This what has been done by SGTE. They have
chosen standard element phases at 298 K and 1 atm (if I remember
correctly) and then they introduced SRE (Standard Element Reference)

G - Sum_i nu_i H_i(0, 298 K)

The sum is taken over elements in the reaction of formation but this
contribution now is a constant. It does not depend on temperature and
pressure as in the Gibbs energy of formation. Now all the derivatives
gives absolute values, and it is much more convenient in practice.

Best wishes,

Evgenii
--
http://Evgenii.Rudnyi.Ru/

[Hazem]

Thank you so much Evgenii, it is certainly something we will have to worry
about when we get to reactors that need G.

I am now having a problem with enthalpy. I was so excited when the reactor
worked fine with the reference enthalpy being the heat of formation at 25°C.
Basically a standard enthalpy call gave me the enthalpy of the inlet at the
specified T and P, the reactor did the material balance and calculated the
composition of the outlet. I sat the enthalpy of the outlet equal to that
of the inlet by adjusting the outlet temperature.

Well, for hydrogen (or 95% Hydrogen, 5% methane, the enthalpy of a vapor at
-260°F is negative. It is in equilibrium with a liquid which is mostly
methane and now the liquid enthalpy is higher than the enthalpy of a the
vapor it is in equilibrium with.

This can not be right.

Regards
Hazem

-----Original Message-----
From: op...@googlegroups.com [mailto:op...@googlegroups.com] On Behalf Of
Evgenii Rudnyi
Sent: Sunday, April 09, 2006 3:15 AM
To: Open source Process Simulator
Subject: Re: Reference Enthalpy

[Rahul Anantharaman]

On 4/9/06, Evgenii Rudnyi <rud...@imtek.uni-freiburg.de > wrote:

> Thank you so much Evgenii.  The article you cited is available online for
> $30.00 and is about 6 MB.

Well, the knowledge is expensive :)

> DO you have it?  If not, I'll see if Rice University is willing to help.

Unfortunately, not. Once, I used to have a xerox copy but it was long
time ago.

I guess that the simplest is to find someone from university that has a
subscription. Well, at least it is available online, note that not all
the old journals are online.

It is not available online  - only journals from 1998 are - atleast in the database my univ subcribes. I can make a XEROX copy from teh library and then scan it and send it across. I am interested in it too. Unfortunately, this week is easter week in Norway and that means everyone is out skiing. The univ reopens April 18th and I'll get it then.


Rahul 

[Hazem]

Thank you Rahul, here is the link to get it online if you can find someone with subscription.

 

If you travel to Europe often, there is a book by Michelsen that I would like to buy.  I can not find it in the states.  Can you help?

 

 http://www.sciencedirect.com/science/journal/03645916

 

 

Regards

Hazem

 

---

From: op...@googlegroups.com [mailto:op...@googlegroups.com] On Behalf Of Rahul Anantharaman
Sent: Sunday, April 09, 2006 7:24 AM
To: op...@googlegroups.com
Subject: Re: Reference Enthalpy

 

 

On 4/9/06, Evgenii Rudnyi <rud...@imtek.uni-freiburg.de > wrote:

[Dudley, Jeremy]

What is the Michelsen book? is it still in print? (In principle you can order from amazon.co.uk and have mailed to you.)

-----Original Message-----
From: op...@googlegroups.com [mailto:op...@googlegroups.com]On Behalf Of Hazem
Sent: 09 April 2006 15:47
To: op...@googlegroups.com
Subject: RE: Reference Enthalpy

[Dudley, Jeremy]

I missed why we wanted an ODE solver, as I thought we were avoiding dynamic modelling. Is what is wanted really a quadrature code, for solving an integral? And if yes, what equations are involved, that would not be amenable to an explicit solution? The simple polynomial Cp is readily analytically solved.

-----Original Message-----
From: op...@googlegroups.com [mailto:op...@googlegroups.com]On Behalf Of Hazem
Sent: 07 April 2006 23:23
To: op...@googlegroups.com
Subject: RE: Reference Enthalpy

Thank you Mr. Dudley, I think Sam or Matt or somebody else can answer your question about Pascal better than me.

 

You got me curious though.  I suggested RK4 because my understanding is that it will work for up to 8 to 12 reactions.  If there is something better but still not as involved as LSODE (that’s the most recent version of Gear’s method) I would say let’s go for it.

 

LSODE is available in Fortran and C++ for free.  I converted it to VB.NET but could not get it to work : (

 

Regards

Hazem

---

From: op...@googlegroups.com [mailto:op...@googlegroups.com] On Behalf Of J&S Dudley
Sent: Friday, April 07, 2006 4:48 PM
To: op...@googlegroups.com
Subject: RE: Reference Enthalpy

 

Is there a specification for the ODE solver available? From my experience I would be willing to translate the ROCK2 Runge-Kutta code to Pascal, providing that the Pascal is Level 1 (i.e., the dimension of the array is not part of the array specification, which used to be the problem with Pascal - see Numerical Recipes 1st Edition). ROCK2 performs well for moderately stiff differential equations, and is therefore a good compromise between 'classical' Runge Kutta (poor performance on stiff problems) and Gear methods (poor performance on non-stiff methods). If there is no specification, I'll do a port of the Fortran code. Better still would be looking to support the Fortran directly - I find

 

Y = Y + H * DY

 

easier than

 

for i = 1 to n do

  y(i) := y(i) + h * dy(i);

 

especially as the equations get more involved.

 

Does Pascal now support declaration initialisation? In Fortran 77, this was the DATA statement.

-----Original Message-----

From: op...@googlegroups.com [mailto:op...@googlegroups.com]On Behalf Of Hazem

Sent: 07 April 2006 19:49

To: op...@googlegroups.com
Subject: RE: Reference Enthalpy

Professor Sam,

 

Please grab the RK4 as quickly as possible, before it disappears.  This is 50% (at least) of a kinetic reactor model.  I’ll review some kinetics tonight to see what else do we need but if somebody can save me the trouble and give us a refresher course in kinetics, it will much appreciated.

 

Regards

Hazem

 

---

From: op...@googlegroups.com [mailto:op...@googlegroups.com] On Behalf Of Samuel Cartaxo

Sent: Friday, April 07, 2006 1:35 PM

To: op...@googlegroups.com
Subject: Re: Reference Enthalpy

 

Maybe, what we need is here:

> Regards
>
> Hazem
>  
>  ________________________________
>  
>
> From: op...@googlegroups.com [mailto: op...@googlegroups.com] On Behalf Of
> Rahul Anantharaman
>  Sent: Friday, April 07, 2006 8:34 AM
>  To: op...@googlegroups.com
>  Subject: Re: Reference Enthalpy
>  
>
>  
>
>

>  Hi Hazem....
>  
>  Heat of formation at 25 C would be the ideal basis for reacting systems. I
> would be preferrable to give the user a choice of either this, absolute zero
> or any other case - with Hf at 25 C being the default basis. It is not all
> that difficult to change the basis. The basis is "stored" in the constant A
> in the Ideal Gas Heat Capacity Equation
>  H(ideal) = A + BT + CT^2+DT^3.
>  
>  One of the things I would probably do is extend this to have a basis such
> that negative enthalpies are not encountered in the flowsheet. But that is
> not what anyone else would be interested in.
>  
>  Rahul
>  
>  
>  
>
> On 4/7/06, Rafil Elyas <rafil...@east101.com> wrote:
>  
>  
>
> HYSIM had absolute zero as a reference point.
>  
>
>  
>  
>
> This was changed to heat of formation at 25C for HYSYS.
>  
>
> I don't think the enthalpy problems you encountered were due to the basis
> selection.
>  
>  
>

> -----Original Message-----
>  From: op...@googlegroups.com [mailto:op...@googlegroups.com ]On Behalf Of
> Hazem

>  Sent: Friday, April 07, 2006 8:16 PM
>  To: op...@googlegroups.com

>  Subject: Reference Enthalpy
>  
>
> If you have any thoughts about the reference enthalpy, please make them
> known.
>
>  
>

[Rafil Elyas]

I believe the ODE solver's for the kinetic reactor.

-----Original Message-----
From: op...@googlegroups.com [mailto:op...@googlegroups.com]On Behalf Of Dudley, Jeremy
Sent: Monday, April 10, 2006 6:18 PM
To: op...@googlegroups.com
Subject: RE: Reference Enthalpy

[Hazem]

That’s is correct.

 

Just something simple at least to start with for no more than 8 to 10 reactions will probably suffice.  We will assume a batch reactor for now, however, kinetic problems do tend to be stiff.

 

Regards

Hazem

 

---

From: op...@googlegroups.com [mailto:op...@googlegroups.com] On Behalf Of Rafil Elyas
Sent: Monday, April 10, 2006 6:21 AM
To: op...@googlegroups.com
Subject: RE: Reference Enthalpy

[Hazem]

This is Michelsen’s book

 

http://www.tie-techcom/pub/publications.htm

 

 

 

---

<BR

[Lynn McGuire]

Hi Hazem,

 

It is a rather good book.  You might be able to save a lot of time on

your two phase flash (especially retrograde issues).

 

Contact Dr. Stenby at  E...@kt.dtu.dk .   He helped me to get a

copy (via wire transfer if I remember correctly) about a year ago.

 

Sincerely,

Lynn McGuire

President

WinSim Inc.

www.winsim.com

 

----- Original Message -----

From: Hazem

To: op...@googlegroups.com

[Hazem]

Thank you so much Lynn.  I emailed Michelsen himself and did not get a response, so I’ll try Dr. Stenby.

 

Regards

Hazem

 

---

[Rafael de Pelegrini Soares]

Hi All,

As Mr. Jeremy pointed before I missed why an ODE solver??
Batch reactors doesn't make sense for a static simulator!

With the best regards,
Rafael.

Hazem escreveu:

> That’s is correct.
>
>
>
> Just something simple at least to start with for no more than 8 to 10
> reactions will probably suffice. We will assume a batch reactor for
> now, however, kinetic problems do tend to be stiff.
>
>
>
> Regards
>
> Hazem
>
>
>

> ------------------------------------------------------------------------
>
> *From:* op...@googlegroups.com [mailto:op...@googlegroups.com] *On
> Behalf Of *Rafil Elyas
> *Sent:* Monday, April 10, 2006 6:21 AM
> *To:* op...@googlegroups.com
> *Subject:* RE: Reference Enthalpy

>
>
>
> I believe the ODE solver's for the kinetic reactor.
>
> -----Original Message-----

> *From:* op...@googlegroups.com [mailto:op...@googlegroups.com]*On
> Behalf Of *Dudley, Jeremy
> *Sent:* Monday, April 10, 2006 6:18 PM
> *To:* op...@googlegroups.com
> *Subject:* RE: Reference Enthalpy

>
> I missed why we wanted an ODE solver, as I thought we were
> avoiding dynamic modelling. Is what is wanted really a quadrature
> code, for solving an integral? And if yes, what equations are
> involved, that would not be amenable to an explicit solution? The
> simple polynomial Cp is readily analytically solved.
>
>
>
>
>
> >
>

--
:: MSc(Eng) Rafael de Pelegrini Soares
:: raf...@enq.ufrgs.br

[Hazem]

I know what you mean and you are right, but it can and has been done.

The inlet stream represents the composition at time=0 and the outlet stream
composition is that at time=t (specified by the user).

The reason I suggested a Batch reactor to start with is because it is an
initial value problem, the simplest of all ODE's. The more realistic plug
flow reactor is a more difficult boundary value problem, not an initial
value problem. Add the heat equation to it and you have a system of coupled
and stiff differential equations. Add the radial temperature profile and
you have a system of coupled and stiff partial differential equations. If
somebody would like to volunteer for this, be my guest, otherwise, let's
stick to a batch reactor for now.

Although my company (www.engineeringsoftware.biz) has plans for a Pyrolysis
program, to even start on something like this requires all other projects to
be finished and my table to be clean and my mind to be clear with no
distractions. This is not the case right now :)

Regards
Hazem

-----Original Message-----
From: op...@googlegroups.com [mailto:op...@googlegroups.com] On Behalf Of
Rafael de Pelegrini Soares
Sent: Monday, April 10, 2006 12:03 PM
To: op...@googlegroups.com
Subject: Re: Reference Enthalpy

[Dudley, Jeremy]

The plug flow reactor is a BVP if what is wanted is to specify an outlet condition. If a reactor length is specified then it is no different to a batch reactor. If the outlet conditions are 'carefully' chosen then many design problems are ODEs, integrated until the required outlet criteria is reached (e.g., how long should the reactor be until the concentration of component A has fallen below/exceed some value?)

SUpporting other specifications may not be tood difficult, if we can have a suitable specification NOW. I agree, it seems like a step too far, given what else has been done; but in my day job we still beleive in documenting what we would like to do, becasue (a) it puts in on the record, and may influence design decisions before they impeded that future desire, and (b) having done the 'simple case it may be 'obvious' that the future can easily be delivered now.

BUT - this needs someone familiar with whatever specification is likely to be required.

Regards,

Jeremy

Regards
Hazem

Hi All,

Hazem escreveu:

-------------------------------------------------------------------------------------------------------------------
Visit our websites www.wrcplc.co.uk and www.wrcnsf.com, as well as www.waterportfolio.com for collaborative research projects.
-------------------------------------------------------------------------------------------------------------------
The Information in this e-mail is confidential and is intended solely for the addressee. Access to this e-mail by any other party is unauthorised. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on information contained on this e-mail is prohibited and maybe unlawful. When addressed to WRc clients, any opinions or advice contained in this e-mail are subject to the terms and conditions expressed in the governing WRc client Agreement.

[Rahul Anantharaman]

Hazem... let me know if you do not recieve any response from Dr Stenby. I can try to get you the book.

Rahul

[Rafil Elyas]

In SS, I generally attach kinetic reactions to a plug flow reactor, reactive distillation & CSTR. Residence time can be computed from flow rates or are spec'd.

-----Original Message-----
From: op...@googlegroups.com [mailto:op...@googlegroups.com]On Behalf Of
Rafael de Pelegrini Soares
Sent: Tuesday, April 11, 2006 1:03 AM
To: op...@googlegroups.com
Subject: Re: Reference Enthalpy

[Evgenii Rudnyi]

> It is not available online - only journals from 1998 are - atleast in the
> database my univ subcribes.

Well, ScienceDirect shows the listing differently depending on the
subscription options and it may not show everything that is available.

Evgenii

[Samuel Cartaxo]

I have a full access subscription, and can confirm it is NOT on-line.
Sorry to say that.

Samuel

2006/4/10, Evgenii Rudnyi <rud...@imtek.uni-freiburg.de>:

[Evgenii Rudnyi]

Hi Hazem,

> I am now having a problem with enthalpy. I was so excited when the reactor
> worked fine with the reference enthalpy being the heat of formation at 25°C.
> Basically a standard enthalpy call gave me the enthalpy of the inlet at the
> specified T and P, the reactor did the material balance and calculated the
> composition of the outlet. I sat the enthalpy of the outlet equal to that
> of the inlet by adjusting the outlet temperature.
>
> Well, for hydrogen (or 95% Hydrogen, 5% methane, the enthalpy of a vapor at
> -260°F is negative. It is in equilibrium with a liquid which is mostly
> methane and now the liquid enthalpy is higher than the enthalpy of a the
> vapor it is in equilibrium with.

I am not sure if I completely understand your problem. One typical
problem is that quite often people mix enthalpy and enthalpy of
formation between each other.

Well, the problem with enthalpy is that we cannot determine its
absolute value. Hence, we need some reference point. Yet, during an
analysis we can introduce enthalpy of a pure substance that is a
function of temperature and presure. In the case of a solution,
enthalpy also depends on concentration. Let us say for simplicity H(T,
p). The derivative of the enthalpy over temperature gives the absolute
heat capacity.

The enthalpy of formation is the enthalpy of a reaction of formation,
Del_f_H, the substance from elements in the stable phases (see IUPAC
Green book, http://www.iupac.org/publications/books/author/mills.html,
as well as other IUPAC documents). Say, if we take molecular hydrogen,
this is

H2(g) = H2(g)

and the enthalpy of formation is zero. The derivative of the enthalpy
of formation over temperature in this case will give zero as well.
Funny, isn't it?

If you express your problem in more detail, I will try to answer your
question.

Evgenii

[Hazem]

Thanks Jeremy. It is interesting to know that a plug flow reactor can still
be an initial value problem.

Regards
Hazem

-----Original Message-----
From: op...@googlegroups.com [mailto:op...@googlegroups.com] On Behalf Of

Dudley, Jeremy
Sent: Monday, April 10, 2006 2:10 PM
To: op...@googlegroups.com
Subject: RE: Reference Enthalpy

[Rahul Anantharaman]

Turns out that I do have acess to the paper in question from my office but not through VPN from home. The paper is a bit too large to post here. Any suggestions?

Rahul

[Hazem]

How big is big? 5 MB? 10?  I can receive emails up to 10MB at Hazem_...@Hotmail.com then we’ll make it available through the project directory or per request (to avoid getting into trouble).

[Samuel Cartaxo]

Dear Rahul,

Please, if possible I´d like to see this paper too.

This account samuel...@gmail.com has enough room to receive it.

Thanks

Samuel

2006/4/11, Hazem <Hazem_...@hotmail.com>: