Temperature extrapolation in Act2 vs. Rxn

[Tom Meuzelaar]

[admin notice: the below is from the former GWB users group email distribution list. This message was originally posted 8/17/2006]

 

Posted by: Joel Brugger

 

Dear all,

 

We have observed different outcomes for temperature extrapolation (between 2 temperatures for which LogKs are available on the grid) in rxn and act2.

 

We're trying to model organic pH buffers. We have modified our database with a new "redox" basis species:

 

 

DESPEN(aq)

* formula= C12H28N2O6S2

charge= 0.0 ion size= 4.0 A mole wt.= 360.492 g

6 species in reaction

-3.000 H2O -17.500 O2(aq) 16.0000 H+

12.0000 HCO3- 2.000 NH3(aq) 2.000 SO4--

2.7420 2.7420 2.7420 2.7420

2.7420 500.0000 500.0000 500.0000

* Added by Pascal Grundler, 16/8/2006

* source: Kandegedara, A. & Rorabacher, D.B. Anal. Chem. 1999, 71, 3140

* RANDOM VALUE; FOR USE IN DECOUPLED REDOX SYSTEMS

 

Note that the LogK’s here are random values – we’re only interested in reaction among DESPEN species. Possibly, this if what causes the strange behaviour described below?

 

The two deprotonated forms are entered as derived species:

 

DESPEN-

* formula= C12H27N2O6S2

charge= -1.0 ion size= 4.0 A mole wt.= 359.484 g

2 species in reaction

1.000 DESPEN(aq) -1.0000 H+

5.956 5.62 5.176 4.796

4.421 -500.0000 -500.0000 -500.0000

* Added by Pascal Grundler, 16/8/2006

* Data from Kandegedara, A. & Rorabacher, D.B. Anal. Chem. 1999, 71, 3140

* values at 0 , 60 , 100 and 150 C obtained by extrapolation from the

* variable temperature (15 to 45 C) data reported by Kandegedara & Rorabacher

 

 

 

DESPEN--

* formula= C12H26N2O6S2

charge= -2.0 ion size= 4.0 A mole wt.= 358.476 g

2 species in reaction

1.000 DESPEN(aq) -2.0000 H+

15.378 14.68 13.765 12.978

12.203 -500.0000 -500.0000 -500.0000

* Added by Pascal Grundler, 16/8/2006

* Data from Kandegedara, A. & Rorabacher, D.B. Anal. Chem. 1999, 71, 3140

* values at 0 , 60 , 100 and 150 C obtained by extrapolation from the

* variable temperature (15 to 45 C) data reported by Kandegedara & Rorabacher

 

 

Now if we calculate the pKa’s in rxn at 99 and 100C we get the correct values:

 

 

DESPEN-- + 2 H+ = DESPEN(aq)

 

Log K's:

0 C: 15.3780 150 C: 12.2030

25 C: 14.6800 200 C: ???

60 C: 13.7650 250 C: ???

100 C: 12.9780 300 C: ???

 

Polynomial fit:

log K = 15.38 - .02695 T - 7.705e-5 T^2 + 1.654e-6 T^3 - 5.888e-9 T^4

 

Log K at 99 C = 12.9944

(Value interpolated using polynomial fit)

 

Assumptions implicit in equilibrium equation:

temperature = 99 C

 

Equilibrium equation:

12.99 = - log a[DESPEN--] - 2 log a[H+] + log a[DESPEN(aq)]

 

 

-------------------------------------------------------------------------------

 

DESPEN- + H+ = DESPEN(aq)

 

Log K's:

0 C: 5.9560 150 C: 4.4210

25 C: 5.6200 200 C: ???

60 C: 5.1760 250 C: ???

100 C: 4.7960 300 C: ???

 

Polynomial fit:

log K = 5.956 - .01285 T - 4.337e-5 T^2 + 8.658e-7 T^3 - 3.068e-9 T^4

 

Log K at 99 C = 4.8039

(Value interpolated using polynomial fit)

 

Assumptions implicit in equilibrium equation:

temperature = 99 C

 

Equilibrium equation:

4.804 = - log a[DESPEN-] - log a[H+] + log a[DESPEN(aq)]

 

-------------------------------------------------------------------------------

 

DESPEN-- + H+ = DESPEN-

 

Log K's:

0 C: 9.4220 150 C: 7.7820

25 C: 9.0600 200 C: ???

60 C: 8.5890 250 C: ???

100 C: 8.1820 300 C: ???

 

Polynomial fit:

log K = 9.422 - .01409 T - 3.368e-5 T^2 + 7.879e-7 T^3 - 2.82e-9 T^4

 

Log K at 99 C = 8.1905

(Value interpolated using polynomial fit)

 

Assumptions implicit in equilibrium equation:

temperature = 99 C

 

Equilibrium equation:

8.191 = - log a[DESPEN--] - log a[H+] + log a[DESPEN-]

 

 

However, calculations in act2 give the right answer at 100C, but not at 99C.

 

So, why is the interpolation different in both programs? Is this a bug or a feature? What would react do?

 

We’d appreciate some insider knowledge before we dive too deeply...

 

Thanks!

 

Joel & Pascal

 

 

Posted by: Craig Bethke

 

Hi Joel,

 

Looks like you're using (positive) 500 as the flag for a missing log K in some of your entries, and -500 in others. The correct flag is 500.

 

Hope this helps,

Craig