ThommyW 0 Posted January 8, 2010 Hello Tom, as you asked me, I am bringing the question to the forum and I added some very important details (!!!) 1) may you explain me what the temperature derivations in the pitzer.dat file mean ? What is the function supposed behind this ? 2) What is the validity ? 3) How to calculate a value for let's say 100 °C ? 4) Why is the value constant ? Is it just a function f = a + b*T ? Here an abstract : ------------------------------------------------------------------------------------------------------------ l- NO3- beta0 = 0.0160 beta1 = 0.0000 beta2 = 0.0000 alpha1 = 99.0 alpha2 = 99.0 dt0/dt = 0.000E+00 d2t0/dt2 = 0.000E+00 dt1/dt = 0.000E+00 d2t1/dt2 = 0.000E+00 dt2/dt = 0.000E+00 d2t2/dt2 = 0.000E+00 -end- end of theta set, begin with cphi set of mu-s H+ Cl- H+ mu = 0.000133 dmmx/dt = 0.104E-04 d2mmx/dt2 = 0.000E+00 --------------------------------------------------------------------------------------------------------------- The standard equation for the dependance of temperature is eg. like that: f = a + b *T + c * T^2 + d/T + e*ln(T) + f/T^2 5) Wouldn't it be preferalble to let the user define the equation to apply ? Will there be any possibility to apply such an equation, to enter the corresponding coefficients a to f and when ? As I have the equation and its coefficients as given before. I may try to calculate the correponding coefficients for the equation for which GWB coefficients are given, but what about its validity range ?????????????? Thanks Thomas Willms Share this post Link to post Share on other sites

Tom Meuzelaar 0 Posted January 8, 2010 5) Wouldn't it be preferalble to let the user define the equation to apply ? Will there be any possibility to apply such an equation, to enter the corresponding coefficients a to f and when ? As I have the equation and its coefficients as given before. I may try to calculate the correponding coefficients for the equation for which GWB coefficients are given, but what about its validity range ?????????????? Hi Thomas: Since questions 1 and 4 are the same as posed in your original email, I will post my email reponse first: >>>>>>>> To see temperature dependence for Pitzer models be sure you are using the dataset thermo_phrqpitz.dat. Also- it would be useful to consult the original reference by Plummer et al., 1988. Temperature dependence is calculated by providing up to 4 coefficients (c1, c2, c3, c4) for beta0, beta1, beta2, cphi, theta, lambda or psi, according to the following equation: val = val25 + c1*(Tk-Tr) + c2*(1/Tk - 1/Tr) + c3*ln(Tk/Tr) + c4(Tk^2-Tr^2) Note that there is scant data, and that only beta0, beta1 and cphi for the Na/Cl pair have data for c2, c3 and c4. >>>>>>>> You can find a slight bit of additional information in the Thermo Datasets Appendix of the version 8 GWB Reference manual. With regards to question 5, there is currently no provision in GWB for a user-defined temperature dependence function, nor are there any development plans for this. Regarding the validity range, I again refer you back to Plummer et al., 1988. Regards, Tom Meuzelaar RockWare, Inc. Share this post Link to post Share on other sites

ThommyW 0 Posted January 18, 2010 Hello Tom, thanks for your answer and the given equation for the Pitzerdata given in the file phrqpitz Ba++ Br- beta0 = 0.31455 -0.33825E-3 beta1 = 1.56975 6.78E-3 beta2 = 0.0 cphi = -0.0159576 alpha1 = 2.0 alpha2 = 0.0 By the way: how do I know if missing coefficients mean that the coefficients of the equation are zero or that there are no data given? Which means that you can or you cannot apply it for other temperatures. What must I do if there are only data for 25°C to make this clear ? Furthermore: concerning data in thermo-Pitzer.dat : H+ Cl- beta0 = 0.1775 beta1 = 0.2945 beta2 = 0.0000 alpha1 = 2.0 alpha2 = 0.0 db0/dt = -0.308E-03 d2b0/dt2 = 0.000E+00 db1/dt = 0.142E-03 d2b1/dt2 = 0.000E+00 db2/dt = 0.000E+00 d2b2/dt2 = 0.000E+00 What about these data ? Is it the same equation or simply beta = beta(25°C) + B * T because there are only two constant numbers as derivations ? Is this cited in the same article Plummer et al., 1988.? (I don't know this one) In fact this was the file where all my questions refer to. Does this mean there is only one linear dependence of T because the second derivation is zero ? Validity ? It seems that we cannot make parameter files with a dependence on temperature because parameters of modern equations of the general form x = a + b * T + c/T + + d*T^2 + e*ln(T) + f* 1/T^2 which are used for log(K), Pitzer parameters etc. cannot neither be used nor calculated from the given equation by tranformation. Will this change in future ? Thanks for your answer in advance. Sincerely Thomas Willms Share this post Link to post Share on other sites

Tom Meuzelaar 0 Posted January 22, 2010 Hi Thomas: By the way: how do I know if missing coefficients mean that the coefficients of the equation are zero or that there are no data given? Which means that you can or you cannot apply it for other temperatures. What must I do if there are only data for 25°C to make this clear ? You'll want to refer to the PHRQPITZ documentation for the answer to many of these questions. Here is the reference: Plummer, L.N., D.L. Parkhurst, G.W. Fleming and S.A. Dunkle, 1988, PHRQPTZ - A computer program incorporating Pitzer's equations for calculation of geochemical reactions in brines. U.S. Geological Survey Water-Resources Investigations Report 88-4153, 310 p. Furthermore: concerning data in thermo-Pitzer.dat : H+ Cl- beta0 = 0.1775 beta1 = 0.2945 beta2 = 0.0000 alpha1 = 2.0 alpha2 = 0.0 db0/dt = -0.308E-03 d2b0/dt2 = 0.000E+00 db1/dt = 0.142E-03 d2b1/dt2 = 0.000E+00 db2/dt = 0.000E+00 d2b2/dt2 = 0.000E+00 What about these data ? Is it the same equation or simply beta = beta(25°C) + B * T because there are only two constant numbers as derivations ? thermo_pitzer.dat is outdated, and is included with GWB only for compatibility with early releases. We recommend that you use thermo_hmw.dat or thermo_phrqpitz.dat. It seems that we cannot make parameter files with a dependence on temperature because parameters of modern equations of the general form x = a + b * T + c/T + + d*T^2 + e*ln(T) + f* 1/T^2 which are used for log(K), Pitzer parameters etc. cannot neither be used nor calculated from the given equation by tranformation. Will this change in future ? Unfortunately, the equations for temperature dependence are not likely to change in the near future, but if you give me a more detailed explanation of what it is you need (email is fine), I'll add it to the wish list. I hope that helps, Tom Share this post Link to post Share on other sites

ThommyW 0 Posted January 25, 2010 Thanks for the information concerning the outdate format of Pitzer.dat. This means that I will implement the form of temperature dependence as in PHRQPITZ.dat and (only 25°C) hmw.dat respectively for the case that a transformation of the equation is possible. This will be the case if you agree to add one term, as already requested by the GRS (Helge Moog). Thanks ThommyW Share this post Link to post Share on other sites

Tom Meuzelaar 0 Posted January 25, 2010 Hi Thomas: For some reason your response, above, is blank. However, I have received a very detailed email from Helge with a nice explanation of the NEA-compliant temperature dependence, and a simple suggestion for how it might be implemented into GWB. Best regards, Tom Share this post Link to post Share on other sites

ThommyW 0 Posted January 26, 2010 Hello Tom, I edited my last response: the anwer text was hidden. Also, I hope adding one coefficient for 1/T^2 is possible so that I can implement it already right now! Thanks Thomas Share this post Link to post Share on other sites