Lewis acid catalyzed hydrogen shuttling between ... - ACS Publications

Apr 5, 1987 - coal was ground for 3 min in a Wig-L-Bug,8 and approx- imately 500 mg of the coal was weighed into a tared bottle and cap. The bottle wa...
0 downloads 0 Views 285KB Size
Energy & Fuels 1989,3, 761-762

761

Communications Lewis Acid Catalyzed Hydrogen Shuttling between Aromatic and Hydroaromatic Hydrocarbons'

Sir: Within the last decade, direct coal liquefaction in a two-stage, low-severity operation has achieved an improved dissolution of coal and efficient heteroatom removal, reducing the undesired cleavage of carbocyclic units and the overconsumption of hydrogen., As usually conducted, in the first stage hydrogen is transferred thermally from hydrogen-donor solvents to the coal. In the second stage, hydrogen gas is catalytically added to the coal slurry, in order to rehydrogenate the solvent, remove heteroatoms, and selectively cleave carbocyclic structures. However, if suitable catalysts could be employed in the first stage for hydrogen shuttling under low-severity conditions,4 both liquefaction and upgrading of coal liquids might be done even more efficiently. Such mild conditions would minimize the overconsumption of hydrogen gas and the repolymerization of phenolic ring clusters that increases the amount of char.5 In our search for hydrogen-shuttling catalysts with model aromatic compounds, we have observed a remarkably facile transfer of hydrogen between aromatic and hydroaromatic hydrocarbons that is catalyzed by Lewis acids.6 For example, admixing equimolar amounts of 1,l-diphenylethene (1) and 9,lO-dihydroanthracene (2) with 10 mol % of AlCl, or MeAlC12in toluene solution at 25 "C leads to almost a complete conversion to 1,l-diphenylethane (3) and anthracene within 15 min (eq l).'

Scheme I1 Q Ph3C-CH-CH

B

\*eA l M e C 1 2

4 Ph

11

H

9

10

7

Ph

MeAlC12

\H

'hP 1

25'C 2

Ph2CH-CH3

3

(1)

+

4

Similarly catalyzed hydrogen transfers have been observed (1)Hydrogen-Transfer Processes. 2. A seminal article dealing with hydrogen shuttling in hydroamaromaticscan be considered as part 1: J.J. Eisch and D.R. Comfort, J. Org. Chem. 1975, 40, 2288. (2) Present address: Chemistry Department, Lebanon Valley College, Annville, PA 17003-0501. (3) Moroni, E. C. Presented at the 193rd National Meeting of the American Chemical Society, Division of Fuel Chemistry, Denver, CO, April 5-10, 1987. (4) Derbyshire, F. Presented at the 196th National Meeting of the American Chemical Society, Division of Fuel Chemistry,Los Anaeles. CA, Sept 25-30,1988. (5) Siskin, M.; Aczel, T. Fuel, 1983,62, 1321. (6) The reactions of Lewis acids with aromatic hydrocarbons in promoting hydrogenation, rearrangement, or ring condensations have been the subject of various publications (For a leading reference, see: Stobart, S. R.; Zaworotko, M. J. Fuel, 1985,64,1623). However, many of these studies did not scrupulously exclude moisture and hence possible catalysis by HC1. (7) mid reaction conditionsand analyses: Admixture of 2.80 mmol each of 1and 2 in 15 mL of toluene with 0.29 mmol of MeAICll at 25 "C led to an orange solution. After 15 min, the reaction mixture was quenched with 5 mL of 1.5 N aqueous HCl. Washing of the organic phase with aqueous NaHCOS, drying over anhydrous MgSO,, and solvent evaporation gave a residue that was analyzed by gas chromatography. Samples were collected from the column and identified by 'H NMR, IR, and MS techniques. In the majority of the reactions (cf. ref a), the yields were essentially quantitative and no side products were observed.

0887-0624/89/2503-0761$01.50/0

between 1,l-diphenylethene and 9,lO-dihydrophenanthrene, 1,l-diphenylethene and 4,5-dihydropyrene, a-methylstyrenes and 9,10-dihydroanthracene, and mesitylethene and 9,lO-dihydroanthracene. In addition, MeA1C12 promoted the disproportionative hydrogen transfer of 1,4-dihydrobenzene into cyclohexene and benzene and of 1,4-dihydronaphthalene into tetralin and na~hthalene.~ That these hydrogen transfers are actually catalyzed by the Lewis acid and not by adventitious source of protons, such as HC1 possibly generated from moisture and the aluminum chloride, is clear from the following experimental tests: (1) the reactions were conducted under rigorously anhydrous conditions in an atmosphere of argon;lo(2) intentional addition of traces of H 2 0 or HC1 did not accelerate but, in fact, retarded the hydrogen transfer somewhat; (3) with MeAlC12 as catalyst, the C-A1 bond would serve as a scavenger for any protons; finally, (4) heating 1 and 2 with anhydrous HCl or p-toluenesulfonic acid (without MeAlClJ gave no hydrogen transfer, but only the well-known proton-catalyzed dimerization of 1.11J2

-

(8) With a-methylstyrene and mesitylethene there were significant amounts of Lewis acid-catalyzeddimerization and trimerization (cf. 1 6, Scheme I). (9) Such disproportions were also promoted by nickel(0) complexes (Eisch, J. J.; Sexsmith, S. R. Final Report to the US. Department of Energy, Grant DE-FG22-84PC70786, October 1, 1987.) (10)All reactants, solvents, and apparatus were scrupulously dried and deoxygenated prior to reaction. Transfers were made with gastight syringes for liquids or in a drybox for solids.

0 1989 American Chemical Society

Energy & Fuels 1989,3, 762-764

762

Further experimental observations can be adduced to support the proposal that the zwitterionic u complex 5, formed from 1 and the aluminum halide, is the catalytic intermediate in these reactions. This complex would initiate hydrogen transfer by hydride abstraction from 2 to produce aluminate anion 5a and carbenium ion 2a. Proton transfer from 2a to 5a would complete the catalytic cycle (Scheme I). First, such a complex can readily explain the observed Lewis acid catalyzed dimerization of 1 to 6.13 Second, if 5 can abstract hydrogen from 2, it might also react directly with dih~dr0gen.l~Indeed, a 1:l mixture of 1,l-diphenylethene and MeAlC12 in toluene slowly absorbed Hzat 0 "C to yield l,l-diphenylethane.16 Third, if such a zwitterionic complex were to form with other terminal olefins having aryl substituents a to the C=C bond (8), skeletal rearrangements might be induced. In fact, when 3,3,3-triphenylpropene (7) was heated with MeA1C12,the olefin was completely converted into a mixture of 1,1,2-triphenylpropene (10) and 1,1,2-triphenylcyclopropane (11). The formation of both products can be nicely accounted for through the well-precedented carbenium ion induced 1,2-phenyl shift to produce 9.16 Protodealumination to yield 10 or union of positively and negatively polarized carbons to generate 11 would be competing modes of consuming 8 (Scheme 11). Alternative mechanistic pathways for these hydrogen transfers, such as radical-cation formation1' or hydride abstraction by the aluminum chloride,18 appear less probable on the basis of the following observations. First, the course of the reaction between 1 and 2 can be monitored by 'H NMR spectroscopy without the occurrence of signal broadening, which would be expected if paramagnetic intermediates were invo1ved.l' Secondly, treatment of 2 with MeAIClz and hydrolytic workup leads to unchanged 2. Were hydride ion abstracted, 2 should have yielded anthracene. Ongoing work is attempting to convert such homogeneous Lewis acid catalysts to anchored catalysts of predetermined structure, so that hydrogen transfer can be effected by heterogeneous catalysis in a catalytic-bed reactora3

1,4-dihydrobenzene, 628-41-1; cyclohexene, 110-83-8; benzene, 71-43-2.

John J. Eisch,* Stephen R. Sexsmith: Mona Singh Department of Chemistry The State University of New York ut Binghamton Binghamton, New York 13901 Received April 19, 1989 Revised Manuscript Received August 11, 1989

Diffuse-Reflectance FTIR/ESR Study of the Reactivity of Native Free Radicals in Illinois No. 6 Coal

Registry No. 1,530-48-3;2,613-31-0;AlC13,613-31-0;MeAlC12, 917-65-7; 9,10-dihydrophenanthrene,776-35-2; 4,5dihydropyrene, 6628-98-4; a-methylstyrene, 98-83-9; mesitylethene, 769-25-5;

Sir: Since the independent discovery of free radicals in coal by Uebersfeld' and Ingram2 in 1954, radicals in coal and coal components have been receiving much attention in an attempt to understand their role in the chemical processes that take place during coal conversion. Many of these studies employed electron spin resonance (ESR) techniques and are well r e ~ i e w e d . ~ -The ~ fundamental question of the reactivity of native free radicals in coal remains unanswered. This question has been addressed Our approach recently in a clever study by Stock et uses the radicals in coal to initiate the free-radical polymerization of a styrene analogue. 4-Vinylpyridine was chosen to determine whether the coal radicals initiate polymerization. 4-Vinylpyridine swells coals readily and does not undergo cationic polymerization. After swelling pyridine-extracted Illinois No. 6 coal with 4-vinylpyridine and drying to constant weight, the coal contained 20% 4-vinylpyridine. Diffuse-reflectance FTIR (DRIFT) spectroscopy clearly showed the presence of poly(4vinylpyridine). ESR spectra demonstrated a decrease in radical concentration and a change in radical structure. Taken together, these observations suggest that the polymerization of 4-vinylpyridine is induced by radicals present in the coal. Illinois No. 6 coal (77.8 %C, 5.7 %H, 9.1 % O (difference), 1.4 %N, 16.2% ash) all daf, from the Argonne Coal Sample Bank' was Soxhlet extracted with pyridine and dried in a vacuum oven at 100 "C to constant weight. The coal was ground for 3 min in a Wig-L-Bug,8and approximately 500 mg of the coal was weighed into a tared bottle and cap. The bottle was then placed into a vacuum desiccator and vapor swelled with 4-vinylpyridine for 48 h, with the coal increasing in weight by 28%. The swelled sample was then dried to constant weight in a vacuum oven

(11)Much stronger Bronsted acids, such as CF3S03H or hectorite sheet silicate, are able to catalyze such hydrogen transfers as well (Eisch, J. J.; Sin h, M. Unpublished studies. Adams, J. M.; Davies, S. E.; Graham, S. J. Chem. Soc., Chem. Commun. 1979 627),but such acids have far greater pK, values ((-12) than HCl or ArS03H (-7). (12)Bergmann, E.; Taubadal, H.; Weiss, H. Chem. Ber. 1931,64B, 1493. (13)Wolovsky, R.; Maoz, N. J . Org. Chem. 1973,523,4040. (14)A mixture of HCl and AlC13 has been reported to catalyze the hydrogenation of ropene at -80 OC under 100 atm pressure of dihydrogen (Koch, Gilfert, W. Brennst.-Chem. 1949,30,413). (15)Interaction of 2.0mmol of 1 with 0.20 m o l of MeAlCl? in 15 mL of toluene at -78 OC under 6 atm of dihydrogen and then warming to 0 OC over 2 h led, upon hydrolytic workup, to 1,l-diphenylethane (3)and the dimer 6. (16)1,3,3-Trimethylcyclopropenereacts with triisobutylaluminum to yield acyclic products evidently arising from carbenium ion-like rearrangements: Richey, H. G., Jr.; Kubala, B.; Smith, M. A. Tetrahedron Lett. 1981,22,3471. (17)Bock, H.; Kaim, W. Chem. Ber. 1978,111,3552. (18)Deno, N.C.; Peterson, H. J.; Saines, G. S. Chem. Reu. 1960,60, 7.

(1)Ingram, D.J. F.; Tapley, J. G.; Jackson, R.; Bond, R. I.; Murnaghan, A. R. Nature (London) 1954,174,797. (2)Uebersfeld, J.; Etienne, A.; Combrisson,J. Nature (London) 1954, 174, 614. (3) (a) Grandy, D. W.; Petrakis, L. Fuel 1979,58,239. (b) Petrakis, L.; Grandy, D. W. Fuel 1980,59,227.(c) Petrakis, L.; Grandy, D. W. Fuel 1981,60, 115. (d) Petrakis, L.; Grandy, D. W. Fuel 1981,60, 120. (e) Petrakis, L.; Grandy, D. W.; Ruberto, R. G. Fuel 1981,60, 1013. (0 Petrakis, L.; Grandy, D. W. Fuel 1981,60,1017.(g) Petrakis, L.; Grandy, D. W.; Jones, G. L. Fuel 1983,62,669.(h) Petrakis, L.;Grandy, D. W.; Jones, G. L. Fuel 1983,62,671.(i) Petrakis, L.; Jones, G. L. Fuel 1983, 62,681. (j)Petrakis, L.; King, A. B. Feu1 1983,62,681, (4)Gorbaty, M. L., Larsen, J. W., Wender, I., Eds. Coal Science; Academic Press: New York, 1982. (6) Petrakis, L., Fraisard, J. P., Eds. Magnetic Resonance. Introduction. Advanced Tooies and ADDlications to Fossil Energy; -- Reidel: Dordrecht, 1984. (6) Stock, L. M.; Muntean, J. V.; Botto, R. E. Energy Fuels 1988,2, 108. (7) Samples were Soxhlet extracted with pyridine. Further characterization data for this coal are available from Dr. Karl Vorres at Argonne National Laboratory. (8)Fuller, M. P.; Griffiths, P. R. Anal. Chem. 1978,50, 1906.

Acknowledgment. We are indebted to the Department of Energy for the support of this research under Grants DE-FG22-84PC70786 and DE-FG22-88PC88930.

8.

8;

0887-0624/89/2503-0762$01.50/0

__

0 1989 American Chemical Society