High throughput catalyst discovery for the direct synthesis of propylene carbonate from propylene glycol and CO 2

Transcription

1 High throughput catalyst discovery for the direct synthesis of propylene carbonate from propylene glycol and C 2 Richard H. Heyn and Elisabeth Tangstad SINTEF Materials and Chemistry, slo, Norway José Antonio Castro-sma, James Comerford, and Michael North Green Chemistry Centre of Excellence, Department of Chemistry, University of York, York, UK

2 Carbon Dioxide Utilization (CDU) IS NT A SLUTIN T THE CLIMATE CHANGE PRBLEM Question of volumes CAN BE A VIABLE, SUSTAINABLE C 1 FEEDSTCK ALTERNATIVE FR THE CHEMICAL INDUSTRY Requires catalysis and energy to overcome inherent thermodynamic and kinetic barriers

3 3 Carbonates Carbonates Sustainable and environmentally benign solvents and reagents whose production is not necessarily sustainable or benign Linear carbonates Dimethylcarbonate Me Me Primary starting material is phosgene Cyclic carbonates C 2 + [cat] R 1 R 2 Industrial process R 1 R 2

4 4 Reaction H H [cat] + C 2 + H 2 Propylene Glycol (PG) Propylene Carbonate (PC) Equilibrium yields ca. 1 % carbonate Ring strain provides chemical energy for reaction Thermodynamic stability of C 2 Improve product yields by removal of water Hydration of acetonitrile (MeCN) C 2 + [cat]

5 5 Carbonates from alcohols or diols industrial utilization of C 2? All one needs is one industrial process utilizing C 2 for carbonates H H + 2 MeH + + C 2 Transesterification chemistry will provide all carbonates

6 6 Literature H H [cat] + C 2 + H 2 Literature gives Zn(Ac) 2 as "best catalyst" Conversion of PG: % (GC) Yield of PC: % (GC) Biproducts: % (propylene glycol acetates) (GC) C, bar, 2.5 mol % catalyst, mol % acetonitrile Huang, S.-Y., et. al., J. Fuel Chem. Technol. 2007, 35, 701 Zhao, X., et. al., Ind. Eng. Chem. Res. 2008, 47, 1365 Data and poor activity of other metal acetates confirmed Relatively large spread in activities

7 7 Establishment of benchmark catalyst Transesterification of carbonates and diols catalyzed by strong acids Addition of p-chlorotoluenesulfonic acid gave better overall yields and conversions Cl S 2 H ptimized system 5 mol % Zn(Ac) mol acid, 280 mol % acetonitrile 60 atm, 145 C, 16 h Conversion PG: 78 % (NMR) Yield PC: 33 % (NMR) Yield byproducts: 40 % (NMR)

8 8 Reactor and Analyses 24-well high throughput reactor for screening of catalyst/il combinations Up to 200 C Up to 100 bar Reaction volume ~11 ml Easy to assemble, disassemble, and clean Analyses of products using a GC HeadSpace instrument Qualititative comparisons between catalysts, not quantitative data (yields, conversions, etc).

9 9 High throughput experiments Confirmed internal reproducibility All reactors with standard GC HeadSpace and NMR analysis of selected wells Experiments contained one standard randomly placed in each row for internal calibration of results Reactions run overnight at 145 C and ca. 50 bar Reaction start: When reactors pressured to 50 bar at 145 C Reaction stop: When oven opened to start cooling Pressure release took ca. 24 h All reactions 5 mol % catalyst Acetonitrile/PG molar ratio 2.8 GAL: Find interesting candidates Selected catalysts checked quantitatively in a Parr reactor

10 10 The 106 different catalysts Zn(Ac) 2 + 1* Zn(Ac) x 1 Zn(Ac) 2 + 2* Zn(Ac) Zn(Ac) ZnCl ZnBr ZnI ZnCl ZnBr ZnI Zn(TFA) 2 Zn(TFA) Zn(acac-F 6 ) 2 Zn(acac-F 6 ) Zn(BF 4 ) 2 Zn(BF 4 ) Zn(S 4 ) 2 Zn(S 4 ) Zn(acac) Zn(Ac) 2 + Li(Tf) Zn(Ac) 2 + Mg(Tf) 2 Zn(Ac) 2 +Ca(Tf) 2 Zn(Ac) 2 + Al(Tf) 3 Zn(Ac) 2 + Sm(Tf) 3 Zn(Ac) 2 + La(Tf) 3 Zn(tosylate) 2 + KH + bipy Zn(Ac) 2 + Y(Tf) 3 Zn(tosylate) bipy Zn(Ac) 2 + Yb(Tf) 3 Zn(tosylate) 2 + K 2 C 3 Zn(Ac) 2 + Sc(Tf) 3 Zn(Tf) 2 + bipy Zn(Ac) 2 + KI Zn(Tf) 2 + phen Sm(Tf) 3 Zn(Ac) 2 + KI + 1 Zn(Tf) 2 + P(C 6 F 5 ) 3 La(Tf) 3 Zn(Ac) 2 + NEt 4 Br Zn(Tf) 2 + dppe Y(Tf) 3 Zn(Ac) 2 + NEt 4 Br + 1 Zn(Tf) 2 + 1,2-(NH 2 ) 2 (C 6 H 4 ) Yb(Tf) 3 Zn(Ac) 2 + KH Zn(Tf) 2 + P(tol) 3 Sc(Tf) 3 Zn(Ac) 2 + K 2 C 3 Zn(Tf) 2 + P(Ar*) 3 Li(Tf) Zn(Ac) 2 + bipy + 1 Zn(tosylate) 2 + phen Mg(Tf) 2 Zn(Ac) 2 + bipy + 2 Zn(tosylate) 2 + P(C 6 F 5 ) 3 Ca(Tf) 2 Zn(Ac) 2 + bipy Zn(tosylate) 2 + dppe Al(Tf) 3 Zn(tosylate) 2 Zn(tosylate) 2 + 1,2-(NH 2 ) 2 (C 6 H 4 ) Fe(Tf) 3 Zn(tosylate) Zn(tosylate) 2 + P(tol) 3 Fe(Tf) Zn(tosylate) Zn(tosylate) 2 + P(Ar*) 3 Co(Ac) Zn(tosylate) Zn(Tf) 2 + Yb(Tf) 3 CoC 3 Zn(tosylate) Zn(Tf) 2 + La(Tf) 3 CoC Zn(triflate) 2 Zn(Tf) 2 + Y(Tf) 3 Zn(triflate) Zn(Tf) 2 + Sm(Tf) 3 Zn(triflate) Zn(Tf) 2 + Sc(Tf) 3 Zn(triflate) Zn(tosylate) 2 + La(Tf) 3 Zn(triflate) Zn(tosylate) 2 + Y(Tf) 3 Zn(tosylate) 2 + bipy Zn(tosylate) 2 + Sm(Tf) 3 Zn(tosylate) 2 + KH Zn(tosylate) 2 + Sc(Tf) 3 1 = chlorobenzenesulfonic acid 2 = p-tolylsulfonic acid 3 = 4-nitrobenzenesulfonic acid hydrate 4 = dibenzenesulfonimide = Tested in batch scale reactor * = 10 mol % Ni(Ac) Ni(Ac) (dppe)nicl 2 NiBr 2 NiBr NiBr Ni(Tf) 2 Ni(Tf) Cu(Ac) Cu(Ac) Cu(acac) 2 Cu(acac) Cu(acac) Cu(Tf) 2 Cu(Tf) Pd(Ac) 2 Pd(Ac) 2 + 1

11 HT results 11

12 12 The 106 different catalysts Zn(Ac) 2 + 1* Zn(Ac) x 1 Zn(Ac) 2 + 2* Zn(Ac) Zn(Ac) ZnCl ZnBr ZnI ZnCl ZnBr ZnI Zn(TFA) 2 Zn(TFA) Zn(acac-F 6 ) 2 Zn(acac-F 6 ) Zn(BF 4 ) 2 Zn(BF 4 ) Zn(S 4 ) 2 Zn(S 4 ) Zn(acac) Zn(Ac) 2 + Li(Tf) Zn(Ac) 2 + Mg(Tf) 2 Zn(Ac) 2 +Ca(Tf) 2 Zn(Ac) 2 + Al(Tf) 3 Zn(Ac) 2 + Sm(Tf) 3 Zn(Ac) 2 + La(Tf) 3 Zn(tosylate) 2 + KH + bipy Zn(Ac) 2 + Y(Tf) 3 Zn(tosylate) bipy Zn(Ac) 2 + Yb(Tf) 3 Zn(tosylate) 2 + K 2 C 3 Zn(Ac) 2 + Sc(Tf) 3 Zn(Tf) 2 + bipy Zn(Ac) 2 + KI Zn(Tf) 2 + phen Sm(Tf) 3 Zn(Ac) 2 + KI + 1 Zn(Tf) 2 + P(C 6 F 5 ) 3 La(Tf) 3 Zn(Ac) 2 + NEt 4 Br Zn(Tf) 2 + dppe Y(Tf) 3 Zn(Ac) 2 + NEt 4 Br + 1 Zn(Tf) 2 + 1,2-(NH 2 ) 2 (C 6 H 4 ) Yb(Tf) 3 Zn(Ac) 2 + KH Zn(Tf) 2 + P(tol) 3 Sc(Tf) 3 Zn(Ac) 2 + K 2 C 3 Zn(Tf) 2 + P(Ar*) 3 Li(Tf) Zn(Ac) 2 + bipy + 1 Zn(tosylate) 2 + phen Mg(Tf) 2 Zn(Ac) 2 + bipy + 2 Zn(tosylate) 2 + P(C 6 F 5 ) 3 Ca(Tf) 2 Zn(Ac) 2 + bipy Zn(tosylate) 2 + dppe Al(Tf) 3 Zn(tosylate) 2 Zn(tosylate) 2 + 1,2-(NH 2 ) 2 (C 6 H 4 ) Fe(Tf) 3 Zn(tosylate) Zn(tosylate) 2 + P(tol) 3 Fe(Tf) Zn(tosylate) Zn(tosylate) 2 + P(Ar*) 3 Co(Ac) Zn(tosylate) Zn(Tf) 2 + Yb(Tf) 3 CoC 3 Zn(tosylate) Zn(Tf) 2 + La(Tf) 3 CoC Zn(triflate) 2 Zn(Tf) 2 + Y(Tf) 3 Zn(triflate) Zn(Tf) 2 + Sm(Tf) 3 Zn(triflate) Zn(Tf) 2 + Sc(Tf) 3 Zn(triflate) Zn(tosylate) 2 + La(Tf) 3 Zn(triflate) Zn(tosylate) 2 + Y(Tf) 3 Zn(tosylate) 2 + bipy Zn(tosylate) 2 + Sm(Tf) at York 3 Zn(tosylate) 2 and SINTEF + KH Zn(tosylate) 2 + Sc(Tf) 3 1 = chlorobenzenesulfonic acid 2 = p-tolylsulfonic acid 3 = 4-nitrobenzenesulfonic acid hydrate 4 = dibenzenesulfonimide = Tested in batch scale reactor * = 10 mol % High throughput results mirror Parr reactor results Ni(Ac) Ni(Ac) (dppe)nicl 2 NiBr 2 NiBr NiBr Ni(Tf) 2 Ni(Tf) Cu(Ac) Cu(Ac) Cu(acac) 2 Cu(acac) Cu(acac) Cu(Tf) 2 Cu(Tf) Pd(Ac) 2 Pd(Ac) 2 + 1

13 13 Effect of acid Addition of 10 mol % p-chlorobenzenesulfonic acid did not improve significantly catalyst activity No observable effect of acid pk a on catalyst activity Cl 7.2 S 2 H N S 2 H 6.60 pk a 's in acetonitrile S 2 H H 2 S S 2 H N H J. rg. Chem. 2011, 76, 391 J. Comp. Chem. 2009, 30, 799

14 14 And the winners are. Zn salts with very poorly coordinating anions Zn(S 2 CF 3 ) 2 (Zn-triflate) Zn(p-S 2 C 6 H 4 CH 3 ) 2 (Zn-tosylate) Variations of these salts with ligands (e.g. bipy) or strong Lewis acid triflate salts [Sm(Tf) 3 ] HYPTHESIS: Effect of acid is to Protonate off Ac-ligands to provide a "naked" Zn(II) cation, but In situ deprotonation is incomplete Batch scale reactions Zn(S 2 CF 3 ) 2 Zn(p-S 2 C 6 H 4 CH 3 ) 2 Zn(S 2 CF 3 ) 2 + bipy

15 15 NMR analysis and impurities H H H Conversion, yield and selectivity from a normalized integration (sum = 100) CH 3 C N H 2 H H H + + NH 3 NH 2 H 2 H H H + + H 2 H + NH 3

16 16 Reaction profile of three new Zn catalysts 145 C, 50 bar C 2, 5 mol % catalyst

17 17 Reaction profile of three new Zn catalysts 145 C, 50 bar C 2, 5 mol % catalyst

18 18 Crystalline material from Zn(tosylate) 2 reaction 1 H NMR data shows no evidence of PG or PC nly acetonitrile and acid protons (exchange with D of MeH-d 4 )

19 19 Comparison of new catalysts with standard 145 C, 50 bar C 2, 5 mol % catalyst

20 Screening of conditions 20

21 21 Summary New catalyst systems for the direct reaction of PG and C 2 to PC Assisted by acetonitrile hydrolysis Discovered via high-throughput screening 100 % improvement over best known homogeneous system Faster Unexplained curiosities with product development Rapid increase in PC and byproduct yields Byproduct yields level out while PC yields continue to increase PC yield profile suggests two different mechanisms Work is continuing to elucidate mechanism and isolate and characterize potential catalytic intermediates, resting states, or dead catalysts

22 22 Acknowledgement European Union Seventh Framework program (FP7/ ) project "CyclicC2R", under grant agreement number Experimental assistance Kari Anne Andreassen Anne Andersen Aud M. Bougza Dr. Silje F. Håkonsen Dr. Anna Lind Ruth Elisabeth Stensrød