вход по аккаунту


Lithium-Ion-Catalyzed Epoxidation by Triplet Dioxygen An ab initio Study.

код для вставкиСкачать
We used aprotic solvents (MeCN, CCI,), with the exception
of water for ArCOONa.['31 The substituent sensitivity was
expressed by Hammett-type correlations; as expected for
carbonyl group properties, the CT scaler'41gave the statistically best fits. Dual substituent parameter treatment,[l51
which separates polar and resonance effects, did not improve
the correlations.
As shown in Table 1, e f diminishes from trifluoroacetophenones to benzoate anions by a factor of 6, following the
diminishing electron demand of the group. The order of
decreasing sensitivity corresponds to increasing resonance
by electron donation from X. The classical qualitative rules
of chemical electrophilicity (and acylating power) are
well reflected by the e' values (COOo < CONH, @ COMe;
COCF,). Bridging oxygen atoms show smaller sensitivity:["] we found a Hammett-type e = 5.6 for -0- of methyl
Table 1 also gives the shift values 6 of the unsubstituted
benzoyl derivatives (measured in MeCN, except for
-COONa). The shielding increases (lower 6) with increasing
donor action of X, as expected for increasing importance of
the polar bond structure C(=X@)-O@.We find a general
parallel between e+ and 6 values, which can be expressed by
a reasonably good correlation with the slope a/@' = 13.1
( R = 0.950). On the other hand, purely inductive effects (as
in COCF, and COCOOEt['61) do not affect the shift values
as much as does resonance.
The chemical shifts of " 0 are essentially determined by
paramagnetic shielding, which is approximated by the
Karplus-Pople equation.[7a1It contains an electronic excitation energy term and a bond-order/charge-density matrix
term. The latter term is related to the n-electron density and
to chemical electrophilicity. Very often, however, the relative
importance of the two terms is not known; for some carbony1 groups the preponderance of the bond order term has
been demonstrated, either by c a l ~ u l a t i o nor~ via
~ ~ comparison with NQR values.['6. 'I In our results the correlation
between 6 and et (which is related to bond order) confirms
the importance of the n-electron density term in the benzoyl
'7Ochemical shifts. This means that the electronic excitation
energy term of the Karplus-Pople equation does not vary in
a systematic way from one type of benzoyl compound to
another; indeed, there exists in this class no correlation between " 0 NMR and UV spectra.['*'
In conclusion it appears that e+ values give a convenient
measure of electrophilicity ; 6 values (simpler to obtain) are
more sensitive to resonance than to polar effects.
Received: January 11. 1990 [Z 3733 I€]
German version: Angew. Chem. 102 (1990) 681
CAS Registry numbers:
PhCOCF,, 434-45-7; PhCHO, 100-52-7, PhCOCO,Et, 1603-79-8; PhCOBr,
618-32-6; PhCOCI, 98-88-4; PhCOSEt. 1484-17-9; PhC0,H . Ma, 532-32-1 ;
PhCOF, 455-32-3; PhCO,Me, 93-58-3; PhCONH,, 55-21-0: PhCO,H.
[6] Reviews: a) I - P . Kintzinger in P. Diehl, E. Fluck, R. Kosfeld (Eds.):
N M R , Basic Principles andProgress, VoI. 17, Springer, Berlin 1981, pp. 1 64; b) W G Klemperer. Angew. Cheni. 90 (1978) 258; Angew. Chem. Inr.
Ed. En$ 17 (1978) 246.
[7] H. A. Christ, P. Diehl, H. R. Schneider. H. Dahn. Helv. Chim. Acru 44
(1961) 865.
[8] H . Dahn, M. N Ung-Truong, Helv. Chim. Aria 70 (1987) 2130.
[9] R. T C. Brownlee, M. Sadek. D. J. Craik, Org. Magn. Reson. 21 (1983)
[lo] T. E.S. Amour, M. I. Burgar, B. Valentine, D . Fiat. J. Am. Chem. SOC.103
(1981) 1128.
[ I l l P. Bdlaknshnan, A. L. Baumstark. D. W. Boykin, Org. Magn. Reson. 22
(1984) 753.
[12] K.-T. Liu. T.-R. Wu, Y.-C Lin. J. Ph.y.s. Org Chtm. Z(1989) 363.
[13] Solvent effects upon individual 6 values can be important: H. A. Christ, P.
Diehl, H e h . P h u . Acta 36 (1969) 170. Controls of our series showed,
however, only small solvent effects on @ +
[14] H C. Brown. Y. Okamoto. J. Am. Chem. Soc 80 (1958) 4979; 0. Exner:
Correlurion Analnis of Chemical Datu, Plenum Press, New York 1988,
p. 61.
[IS] S. Ehrenson. R. T. C. Brownlee. R. W. Taft, Prog. Phjs. Org. Chem. 10
(1973) 1.
1161 Also -COCN. 6 = 559: C. P. Cheng. S. C. Lin, G.-S. Shaw, J. Magn.
Reson. 69 (1986) 58.
[17] C P. Cheng, T. L. Brown. J. Am. Cliem. Soc. t 0 f (1979) 2327.
[18] K.Yates, S. L. Klemenko, I. G. Csizmadia, Spectrochint. Acta25A (1969)
Lithium-Ion-Catalyzed Epoxidation
by Triplet Dioxygen: An ab initio Study **
By Heinz Hofmann and Timothy Clark *
Dedicated to Paul von RaguP Schleyer
on the occasion of his 60th birthday
Our recent studies on the metal-ion catalysis of radical,['. cl~sed-sheIl,[~]
and triplet reactions have demonstrated that complexation to a metal cation or radical cation
may facilitate open- or closed-shell reactions, respectively.
Most significantly in the context of the present communication, complexation of triplet dioxygen, 30,,with Li@lowers
the barrier to hydrogen abstraction from C-H donors and
favors the 0 - H product thermodynamically.[41This result,
which may have far-reaching consequences for biological
systems, encouraged us to consider another important oxidation reaction, the epoxidation of olefins. Both the silvercatalyzed epoxidation of ethylene15]and the epoxidation of
aromatics by cytochrome P45OC6]suggest that metals may
have a large effect on the course of this reaction. Our H-abstraction work 14] has revealed the predominantly electrostatic nature of this type of catalysis, so that Lie catalysis provides a useful model for the electrostatic component of
transition-metal catalysis. We have therefore used ab initio
molecular orbital theory"] to investigate the two related reactions (a) and (b) (adduct formation is indicated by, for
example, [Li:O,]@).
302+ C,H,
[l] P. G. Gassmann, A. F. Fentiman, J. Am. Chem. Soc. 91 (1969) 2549; H . C.
Brown. C. G. Rao, M. Ravindranathan, ibid. 99 (1977) 7663; H C.
Brown, P. v. R. Schleyer: The Nonclassical Ion Problem, Plenum Press,
New York 1977, p. 163.
[2] R. T. C. Brownlee, J. Di Stefano, R. D. Topsom. Spectrothim. AFIU31 A
(1975) 1685; R. D Topsom, Prox. Phys. Org. Chem. 16 (1987) 193.
[3] F. H. Allen, 0.Kennard, D. Watson, L. Brammer. A. G. Orpen, R Taylor.
J. Chem. Soc. Perkin Trans. 2 1987, S1.
[4] C. Delseth, T. Nguyen, J:P. Kintzinger. Helv. Chim. Acru 63 (1980) 498:
H. C. Brown, E.N. Peters, J Am. Chem. Soc. 99 (1977) 1712.
[ 5 ] J. B. Stothers: Curbon-f3N M R Specrroscopj, Academic Press. New York
1972. pp 279-310; W. E Reynolds, Prog. P h n Org. Chem. 14 (1982) 165.
C) VCH I/frlagsgeselischufr mbH, 0-6940 Weinheim, 1990
'13 [ O - O - C H , - C H , ] ~ o x i r a n e + ' 0
+ C,H4j%
3 [ L :~0 - 0 - C H ,-CH,]@
3[oxirane: Li:O]'
Both reactions involve two distinct steps with an intermediate minimum (2 and 5). The first step, the addition of the
Dr. T. Clark, DipLChem. H. Hofmann
lnstitut fur Organische Chemie der Universitat Erlangen-Niirnberg
Henkestrasse 42. D-8520 Erlangen (FRG)
This work was supported by the Deutsche Forschungsgemeinschaft. the
Fonds der Chemischen Industrie. the Volkswagen-Stiftung, and Convex
Computer GmbH.
$13.50 f .25!0
A q e u . . Chem. Int. Ed. Engl. 29 (19901 No. 6
Table 1. Absolute [Hartrees] [a1 and relative [kcal mol-', in parentheses] [b] energies of 1-7.
+ C,H,
NlMAG [c]
H F [a]
MP 2
ZPE [d]
- 221.60923
- 221.58936
- 227.65129
- 234.90058
- 234.88410
- 234.96912
( - 40.0)
- 228.151 11
- 228.1671I
- 235.46566
- 228.14640
- 228.21057
- 228.11102
- 235.50866
- 235.52119
( - 19.3)
3O + oxirane
+ C,H,
+ oxirane
31 7
31 8
Id] Restricted Hartree-Fock (RHF) for closed-shell systems and unrestricted Hartree-Fock (UHF) for radicals. [b] Energies include correction for zero-point vibrations. [c] Number of imaginary frequencies (1, transition state; 0, minimum). [d] Zero-point energy [kcal mol-'1.
dioxygen species to the olefin to give a triplet dimethyleneperoxy diradical, 2, is calculated to be endothermic by
34.4 kcal mol-' in the unperturbed reaction (a) and to have
an activation energy of 40.6 kcal mol-' (via transition state
1). Although it might be expected on the basis of other UHFbased calculated activation energies [*] that the calculated
activation energy should be too high because of spin contamination in the UHF wave function, we can still conclude that
the barrier to addition of triplet oxygen to olefins is considerably higher than those for, say, oxygen
result reflects the expected["] kinetic inactivity of triplet oxygen, even in a reaction that stays on the triplet potential
energy surface.
Complexation of the oxygen with Lie [Eq. (b)] changes
this situation considerably. The activation energy for the
initial addition reaction leading to 5 via transition state 4 is
only 7.8 kcal mol - '. The addition reaction is essentially
thermoneutral with a calculated energy of 0.8 kcal mol-'.
Thus, both the odd-electron bond stabilization of the transition state 4 observed previously for radical addition to
olefins['] and the thermodynamic preference for products
with as many bonds as possible between oxygen and an electropositive element (hydrogen or carbon) are found in the
effect of Lie complexation on this addition reaction. Triplet
oxygen addition to ethylene becomes a fast, reversible process in the presence of Lie.
The second stage of the reaction is the concurrent closing
of the oxirane ring and cleavage of the 0-0 bond. This
process is found to have a high (61.6 kcal mol-', transition
state 3) activation energy for the unperturbed reaction (a).
The entire process shown in Equation (a) is found to be
endothermic by 27.7 kcal mol- '. The calculated activation
energy for the Lie-catalyzed reaction (b) is, however, only
22.3 kcal mol-' (transition state 6) and the reaction sequence (b) as far as the complex 7 is exothermic by
19.3 kcal mol- '. Schematic energy profiles for the two reactions are shown in Figure 1 . There is comparatively little
effect of the Li' on this second stage of the reaction compared with the first step, but the overall effect is significant.
In both transition states for the ring closure/bond cleavage
(3, 6), the 0-0 bond stretching precedes the closure of the
C-C-0 angle. Because these transition states (3,6) bear little
Angen. Chem. hi.Ed. Engl. 29 (1990) No. 6
resemblance to the final products, the intrinsic reaction coordinates were followed from the transition states to the products using the IRC option in Gaussian 88.1"1 We also note
that the orientation of triplet oxygen addition via an s-trans
transition state is opposite to the s-cis orientation found for
singlet oxygen addition to ethylene." 21
[ k c o l mol-'l
'*-* oxirane +
oxirane + 30 LI'
Fig. 1. Schematic energy profiles for the epoxidation of ethylene by 302
(dashed line) and by '[OZ:Li]@(solid line). Energies are relative to the starting
materials and the numbering of the stationary points corresponds to that used
in the text and structures. e = reaction coordinate.
We emphasize once more['-41 that such results should be
quite general for 30,complexed to metal cations. Lithium,
which is a computationally convenient model, is expected to
show a large effect,[4.l 3 ] but should not be unique. Although
our model studies cannot be considered to relate directly
to enzymatic systems, we should like to point out that
3[0,:Li]b can perform both hydrogen abstraction[4] and
epoxidation reactions analogous to the reactivity of cytochrome P450.r61Our calculations thus suggest indirectly
that the electrostatic component of cytochrome P450 catalysis may be greater than generally thought. Furthermore,
reaction (6) also produces an oxygen atom, so that two
epoxidations should be possible with one oxygen molecule.
Our results and the remarkable gas-phase reactivity of tran-
8 VCH Verlugsgesellschufl mbH, 0-6940 Weinheim. 1990
1783 ..b
181 C. Gonzalez. C. Sosa, H. B. Schlegel, J. P ~ J JChem.
93 (1989) 2435-2440.
[9] C . Sosa. H. B. Schlegel. J. Am. Chem. Soc. 109 (1987) 4193 -4198.
[lo] See. for instance, G. A. Hamilton in 0.Hayaishi (Ed.): Mo/ecu/ur O.xygen Activation. Academic Press, New York 1974, pp. 405-451.
[l 11 Note that possible alternative reactions, such as dioxetane formation, have
not been considered in this study and that they may compete with the
epoxidation considered here.
112) G. Tonachini. H. B. Schlegel, F. Bernardi. M. A. Robb, THEOCHEM 31
1 L;ZO
3 2 C
Structure Determination of NaCD, Powders at 1.5
and 300 K by Neutron and Synchrotron-Radiation
Diffraction **
(1986) 221 -227.
[13] T Clark, J. Am. Chrm. Soc. 111 (1989) 761-763.
[14] H. Schwarz, Ace. Chem. Res 22 (1989) 282-287.
[lS] A K. Ghosh. L. Kevan, J. Phys. Chem. 92 (1988) 4439-4446.
3 5
By Erwin Weiss,* Siegfvied Corbelin,
Jeremy Karl Cockcroft, and Andrew Nicholas Fitch
Dedicated to Projessor P a d von Rag& Schleyer
on the occasion of his 60th birthday
We reported earlierr’]the synthesis of methylsodium containing variable amounts of LiCH, and were able to determine the structures of these preparations from their X-ray
powder data. The preparations were formed from LiCH,
and sodium tert-butoxide by a metal-metal exchange reaction [Eq. (a)] and, depending on the composition of the di-
Scheme 1. Structural parameters for 1-7 calculated at the UHFj6-31G* level
suggest that “naked ion chemistry” may
be a remarkably good model for far more complex systems
such as zeolite^^^.'^] or enzymes.
Received. January 15. 1990 [Z 3740 IE]
German version- Angew. Chem. 102 (1990) 697
CAS Registry numbers
C,H,. 74-85-1, Lie, 17341-24-1
T. Clark. J. Chem. Soc. Chem. Commun. 1986, 1774-1776.
A. von Onciul, T. Clark, J Chem. Soc. Chem. Comnwn. 1989,1082-1084.
T. Clark, J. Am. Chem. Soc. 110 (1988) 1672-1678.
H. Hofmann, T. Clark, unpublished.
[S] See, for instance, K. A. Jsrgensen, Chem. Rev. 89 (1989) 431 -458.
[6] V. Ullrich, Thp. Curr. Chen?.83 (1979) 67-104.
171 All calculations used the Gaussian 88 (M. J. Frisch, M. Head-Gordon.
H. B. Schlegel, K. RaghaVdChdri, J. S. Binkley, C. Gonzalez. D J. Defrees.
D. J. Fox, R. A. Whiteside, R. Seeger, C . F. Melius. J. Baker, L. R. Kahn.
J J. P. Stewart, E. M. Fluder, S. Topiol. J. A. Pople, Gaussian, Inc.. Pittsburgh PA.) o r Cadpac 4.0 (R. D. Amos, J. E. Rice. CADPAC: The Cumbridge Anulytic Derivutives Package, issue 4.0, Cambridge 1987) programs
on a Convex C 2 or a Cray YMP-432. The 6-31 G* basis set (P. C . Hariharan, J. A. Pople. Theor. Chim. Ac/u. 28 (1973) 213.) was used throughout.
The natures of the stationary points were confirmed by diagonalization of
the force-constant matrices calculated a t the same level. The energies discussed in the text are corrected for zero-point energies and include a
fourth-order Mdler-Plesset correction for electron correlation including
single, double, triple, and quadrupole excitations (MP4sdtq; C. Msller,
M. S. Plesset, Phys. Rev. 46 (1934) 618, J. S. Binkley. J. A. Pople. In/. J.
Quantum Chem. 9 (1975) 229; J. A. Pople, J. S. Binkley. R. Seeger. Int. J
Quanlum Chem. Suppl. 10 (1976) 1; M J. Frisch, R. Krishnan, J. A. Pople,
Chem. Phys. Letl. 75 (1980) 66). Because the transition state 3 occurs very
early in the ring-closure reaction. the intrinsic reaction coordinate was
followed downhill from the transition state to the oxirane product in order
to confirm the nature of the reaction.
?> VCH Verlu~.sgesell.schuftmhH. 0.6940 Weinheim, 1990
+ NaOtBu----ether:hexane N a C H , + LiOtBu
ethyl ether/hexane solvent system, contained NaCH, and
LiCH, in a molar ratio of about 3 :1 to 36: 1. The amount of
LiCH, was lowest in pure ether. In all these reactions, LiCH,
was first prepared and dissolved NaOtBu then added.
X-ray investigation of the fine powder gave a cubic macrocell ( a = 2020 pm, space group F43c) containing 24 tetrameric (NaCH,), aggregates. Although the structure is
therefore closely related to that of methyllithium,[21the arrangement of the (NaCH,), units in the lattice is considerably more complicated. They form a zeolite-like structure
with large cavities, which can incorporate complete
(LiCH,), units up to the structurally determined NaCH,
ratio of 3: 1 without noticeable lattice expansion.
A phase transition, involving formation of a more compact structure, is observed at about 80 “C on drying lithiumpoor methylsodium preparations. We have now succeeded in
directly preparing this methylsodium by dropwise addition
of the LiCH, solution to a previously prepared NaOtBu
[*] Prof. Dr. E. Weiss. DipLChem. S. Corbelin
Institut fur Anorganische und Angewandte Chemie der Universitit
Martin-Luther-King-Platz 6, D-2000 Hamburg 13 (FRG)
Dr. J. K. Cockcroft
Institut Laue-Langevin, 156 X
F-Grenoble 38042 Cedex (France)
Dr. A. N. Fitch
Department of Chemistry, University of Keele
GB-Staffordshire ST5 5BG (UK)
[**I Metal-Alkyl and Metal-Aryl Compounds, Part 43. This work was supported by the Deutsche Forschungsgemeinschaft and the Fonds der Chemischen Industrie. We thank ILL and SERC Daresbury for making possible
[he neutron and synchrotron-radiation diffraction investigations. We also
are grateful to DiplLChem. H. Viebrock for his help in analyzing the data.
Part 42 N P. Lorenzen. E. Weiss, Angen. CIiem. 102 (1990) 322; Angew.
Chem. lnt. Ed. Engl. 29 (1990) 300.
0570-0833~90/0606-0650S 03.50+ .2Sl0
Angevs. Chem. I n / . Ed. EngI 29 (199fJ) No. 6
Без категории
Размер файла
363 Кб
ion, stud, initio, triple, epoxidation, dioxygen, lithium, catalyzed
Пожаловаться на содержимое документа