As ds 62, but with:

• deep mantle phases from JPet paper
• hydrous Mg phases: phD, phE, shB
• post perovskite, ppv
• rutile liquid
• hematite liquid corrected for S(m) = 44.9 J approximation
• eskolaite liquid with for S(m) = 50 J approximation
• K-jadeite
• gedrite, adding Fischer et al 1999
• glaucophane, nyboite, tan, nta
• wustite (as well as fper)
• new S for knorringite (Wijbrans et al 2013)
• uvarovite added (Huckenholz & Knittel)
• skiagite (Miller et al in prep)
• changed volume for woL to fit 1bar melting better
• Almandine S->336.7 and Ilmenite S->107.5 (Geiger et al 2012)
• New V, K for eskolaite, Kantor et al 2012; also H from Klemme et al 2000
• Added zrt, tcn endmembers in zircon using Ferry & Watson 2007
• Added Cr-en and Cr-di tschermak endmembers.
• Use Bohlen & Liotta 1986 with W(gr-alm)=6 kJ
• Added picr + cor = sp + esk from Chatterjee et al 84
• refitted geik Cp
• added quandilite (qnd) with data of O'Neill & Scott 2005
• added 2esk + en = 2picr + 2q  (Klemme & O'Neill 1997)
• new spinel & hercynite disorder model (1/2 S_oct) {also qnd, usp}
• Changed Cp and other values for h2oL
• Andradite S->324 (Geiger et al 2017, in press)

Changes since ds61

1. Updated atg, phA
2. Added non-ideality to talc
3. Try Ogorodova et al 2016 Hf(clin) = -8895.00 +/- 20 kJ?, more consistent with S(clin) = 449?
4. Picr treated as perfectly normal spinel (Cr only octahedral)
5. picr cor = sp  esk  Using T = 2185 - 0.01497P (Chatterjee et al 84)
6. Lowered S(py) to 267.0 J/K (was 269.5)
7. Added Mg-eskola pyroxene
8. Refitted Cp(fo) to Richet & Fiquet 1991, therefore modified S(foL)
9. New tpz Cp from Fluortopaz + Mg(OH)2-MgF2
10. MgEs added (Cheng et al in press)
11. Modified caes from cats/2 + 3/2en -3/2fo
12. Vol of ts -> 26.739 from Jenkins 94
13. Added obi and fbi end-members
14. New Landau parameters for quartz from Angell et al 2017
15. Cp(ru) from R&H95 not RHF77, higher values
16. Ilm as sf (1/2S) not Landau
17. Sphene S raised to 129.2 as R&H95

To do:
try Najorka et al 02 formula for staurolite