[Al-hydroxide in thermo.com.v8.r6]

I am trying to develop a consistent thermodynamic database for Al-hydroxides and noticed the following in thermo.com.v8.r6

 

I am trying to verify that I am using SUPCRT92 correctly by comparing SUPCRT92 output with the thermo.com.v8.r6 database.

This database uses Pokrovskii and Helgeson 1995 (American Journal of Science) for the Al-H2O system.

SUPCRT92 has sprons96.dat which also uses Pokrovskii and Helgeson 1995.

 

The SUPCRT92 calculated log K values for Gibbsite, Boehmite and Diaspore agree at 25 but disagree at other temperatures to varying degrees (See Table below).

Other minerals are exactly consistent, for example, anorthite. Anyone know why or how these values are different?

 

Log K values

Gibbiste

Temp thermo.com SUPCRT

25 C 7.756 7.756

300 C -2.4754 -1.368

 

Boehmite

Temp thermo.com SUPCRT

25 C 7.5642 7.564

300 C -2.7773 -2.705

 

Diaspore

Temp thermo.com SUPCRT

25 C 7.1603 7.160

300 C -2.8906 -2.865

 

Anorthite

Temp thermo.com SUPCRT

25 C 26.578 26.578

300 C -2.5408 -2.541

 

According to the GWB thermo.com.v8.r6 database header, they use supcrt as the "extrapolation algorithm (at least for these phases).