вход по аккаунту


Matrix Spectroscopy of 2-Adamantylidene a Dialkylcarbene with Singlet Ground State.

код для вставкиСкачать
Matrix Spectroscopy of 2-Adamantylidene, a
Dialkylcarbene with Singlet Ground State**
Thomas Bally", S t e p h a n Matzinger, Leo T r u t t m a n n ,
M a t t h e w S. Platz,* and Scott Morgan
Identification of the ground state multiplicity (singlet S or
triplet T) of carbene intermediates is of fundamental interest to
synthetic, mechanistic, and theoretical chemistry. The issue is
straightforward to resolve for triplet carbenes, which are persistent at cryogenic temperatures. In this case detection of the triplet
state and assessment whether this is the ground state of the
carbene o r if it is thermally accessible from a singlet ground
state species at the temperature of the experiment is unambiguous
with EPR spectroscopy."] Unfortunately, failure to observe a
triplet EPR spectrum does not constitute proof that a certain
carbene has a singlet ground state. because the triplet may not
be formed at all or be short-lived. even at cryogenic temperatures.
Dialkylcarbenes present a particular challenge in that they may
rapidly decay by 1,2- or 1,3-insertion reactions to form olefins and
cyclopropanes, respectively, even at cryogenic
Therefore. until recently only di-tevt-alkylcarbenes could be observed directly in matrices (that is, not as trapping products), and
these were found to possess triplet ground states.141In 1992 Sheridan et al. reported the spectroscopic identification of the first
directly observed singlet dialkylcarbene. di~yclopropylcarbene.[5~
where migration of the 8-H atoms is prevented by keeping
the corresponding C-H bonds perpendicular to the empty porbital through strong hyperconjugation of the latter with the
Walsh MO's of the cyclopropyl moieties.[61This hyperconjugation also favors the singlet state and may kinetically stabilize the
car bene.
A similar geometrical constraint is realized in 2-adamantylidene (Ad:), which should therefore also be reasonably resistent to
[1,2] H shift at cryogenic temperatures. Photolysis of adamantane-2-spiro-3'diazirine (1) under several low-temperature conditions failed to produce a triplet EPR spectrum, but Ad: can be
intercepted by olefins and by pyridine or thiophene to form the
ylides 2 (Scheme 1) .[71 As the kinetics of this ylide formation are
not influenced at room temperature by oxygen. an excellent trap
of triplet carbenes, it was concluded that Ad: has a singlet ground
state and that the triplet must lie at least 4.8 kcalmol-' higher
in e~iergy.'~.Finally. Ad: isolated in b-cyclodextrin was recently found to rearrange spontaneously to 2,4-dehydroadamantane
( 3 ) at room temperature.["]
Herein we report the spectroscopy of Ad: in an argon matrix.
Our results confirm that Ad: has a singlet ground state. In particular, we have found a characteristic absorption band of Ad:,
which may be of general use in identifying other hydrocarbon
carbenes with singlet ground states.
[*] Prof. T. Bally. Dipl.-Chem. S. Matringer. L. Truttmann
Inctitut de Chimie Physique de I'Universite de Fribourg
Perolles, CH-1700 Fribourg (SwitLerland)
Fax Int. code + (373826.488; e-mail: Thomas. Ballyio
Prof. M S. plat^
Department of Chemistry
The Ohio Slate University
120 West 18th Avenue. Columbus, OH 43230-1173 (USA)
Fax: I n 1 code + (614)292-1685: e-mail: mplatzu magnwacs
Dr. S. Morgan
Department of Chemistry
Dana College. Blair. N E 68008 (USA)
This work M B S supported through grants No. 20.34071-92 of the SWISS
National Science Foundation and No. CHE-8814950 of the US National Science
Scheme 1
The difference spectrum in Figure 1 shows how a broad band
with ,'.,,,rr 620 nm appears at the expense of the sharp diazirine
peaks at 330-370 nm upon 30 minutes irradiation at 365 nm of
I isolated at 10 K in an argon matrix. At the same time the IR
Fig. 1. UV:Vis difference spectra illustrating the formation of Ad: by 365 nm photolysis of 1 (solid lines) and the subsequent bleaching of Ad: by irradiation at
1 > 540 nni (dashed lines). A = absorbance.
spectrum shows an increase of a group of intense peaks around
2050 cm-', which are indicative of a diazo moiety. Upon photolysis of this sample through a 540 nm cutoff filter for another
30 minutes, the broad visible band and some of the new bands
in the IR spectrum disappear in concert (Fig. 2), while a new
group of peaks arise which are identical to those of authentic 3
in argon; the diazo peaks remain unaffected.
If CO is added to the matrix containing 1, formation of the
620 nm band after 365 nm photolysis is partially suppressed
(Fig. 3), while the IR spectra show new features at 21 10 cinthe region typical of ketene groups, in addition to those of the
diazo compound.
As in the experiments without CO, the 620nm band could be
bleached, and 3 was formed, albeit in smaller overall yield. In
contrast, annealing of the CO-doped matrices led to an increase
of the ketene absorptions (rather than those of 3) at the expense
of those of the primary photoproduct. Annealing the sample at
35 K after complete photolysis of 1 has no effect on the spectra.
Flg. 2. a ) I R spectrum after complete photolysis of 1 at 365 nm. b) difference
spectrum after photolysis of sample of spectrum a ar i > 5 4 0 nm. Peaks pointing
down are assigned to bleached Ad:. those pointing up (except for the band marked
x ) to newly formed 3 ; c) I R spectrum of authentic 3 (bands marked "i" denote a
small impurity).
empty p orbital. Conversely, 3CH, as well as triplet dialkylcarbenes do not absorb above 300 nm.14."1
We have carried out ah initio CASSCF quantum chemical
calculations"21 on the ' A , and 3B, states of Ad: as well as on
tautomers 3 and 4. The results are summarized in Figure 4. According to these calculations, the triplet state of Ad: is 3.1 kcal
mol- more favorable. However, the stability of triplet states of
alkylcarbenes is overestimated (or that of singlet states underestimated) by about 6 kcalmol-' at this level of theory, and large
basis sets and methods for efficient recovery of dynamic correlation effects are needed to remedy this.[14. ''I Therefore we see no
contradiction between this result and the experimental energy
difference of at least 4.8 kcalmol-' in favor of
importantly, both 3 and 4 are distinctly more stable than Ad:.
but the reaction that gives 3 is much more exothermic, which
may explain its exclusive formation from Ad: isolated in B-cyclodextrin.['] The [I ,2] hydrogen shift leading to 4 appears to be
effectively prevented by keeping the empty p orbital on the
carbene perpendicular to the migrating C-H bond.
These results provide strong evidence that the photoinduced
rearrangement of I to diazoadamantane 5 is accompanied by
the formation of Ad:. Both the appearance of ketene 6 in matrices containing C O (in part upon annealing) and the photochemical transformation to 3, a known thermal rearrangement
product of Ad: at room temperature.[91 can most readily be
explained if Ad: is the carrier of the broad 620 nm absorption
and the associated IR bands.
7 -
T I-
Fig. 4. Energy level diagram for Ad:. 3. and 4 from 6-31C* CASSCF calculations
[12]. The first excited states of singlet and triplet Ad: were c'ilculated by the CIS
procedure [16].
Fig. 3. EA and IR difference spectrx illustrating the effect of CO conccnrration o n
the yield of Ad: and ketene 6. Also seen in thia spectrum is the formation of d i a ~ o
compound 5 . Spectra a are obtained after 365 nm photolysis of 1, and b after
annealing at 25 K for 6 minutes. The dashed spectra are before 365 nm photoIysla.
With regard to the question of the ground state multiplicity of
Ad: we note that singlet dicyclopropylcarbene shows a similarly
broad band with L,,, = 490 nm.['I and that 'CH, has a weak
transition with a long vibrational progression whose vertical
component lies at about 1 .S eV (820 nm).[tolwhich corresponds
to excitation of an electron from the sp hybrid lone pair to the
With the CASSCF-optimized geometries, CIS calculations"'] yielded a vertical excitation energy for 'Ad: of 1.78 eV
(695 nm), whereas the first excited state of 3Ad: was predicted
to be 6.88 eV (180 nm). Although the former energy is about
0.2 eV too low, the calculations clearly support the assignment
of the 620 nm band to 'Ad: (rather than 3Ad:).["1 Apparently.
alkyl substitution does not induce a major shift in this transition. and hence other singlet dialkylcarbenes should show similar bands which may facilitate their identification.
E.xperinieri tal
I w'as prepared according to literature procedures 1181. A sample of 3 was kindly
provided by Prof. Udo Brinker (State University of New York at Binghamton) who
also shared his recent data with us prior to their publication. The apparatus and
procedure for matrix isolation experimrnls has been previously dcacrihed [19].
0.1 Torr (approximately the equilibrium vapor pressure at room temperature) of I
were mixed with 100 Torr ofargon in a 2.2 L bulb from which 80 Torr (approximately 10 mbi Ar) were deposited on a CsI window held at 20 K for 2 h. Compound 1 was
photolyzed with a SOOW medium-presssure mercury lamp fitted with a 365 nm
interference filter; converyion of Ad: to 3 was achieved with a 1 k W argon resonance
lamp and a 540 nm cutoff filtcr. Calculations were carried out with the 6-31C;* basis
set uvng the GAUSSIANYq [17]and GAMESS (US) program\ [20] on SGI and IBM
Received April 19 1994
J u l y 22 1994 [z6861 IE]
German version Aitge" Chwn 1994 106 2048
A M Tiozzolo E Was\ermdn in Curhrnr, Pol I I (Eda R A Moss M
Jones). Wiley. New York. 1975. p. 185.
a ) S. Wierlacher. W. Wander. M. T. H. Liu. J A m Chef??.Soc. 1993, 115. 8943;
h) E. J Dix. M. S. Herman, J. L. Goodman ihid. 1993. 115. 10424; c) Private
coinmuiiication from Prof. R . Sheridan. University of Nevada. Reno.
J. W Storer. K. N . Houk, .
A m . Clirm. Sor.. 1993. 115. 10426.
a) Di-r[,.,.r-butylcarbene:J. E. G a n o , R. H. Wettach, M. S. Platz. V. P. Senthiln;ilhan. J A m . Ciirm. Sot.. 1982. f04, 2326: b) Diadaniantylcarheiie: D. R .
Myers. V. P. Senthilnathan. M. S. Platz, M. Jones. Jr.. f h d . 1986. IUH. 4232.
J. R. Ammann. R. Subramanian. R. S. Sheridan. J Am. Cheni. SOC.1992. 114,
P €3. Shevlin. M. L. McKee. J Am. C%eni.Soc. 1989. 111, 519. and references
a ) R A . Moss. M. J. Chang. Etruhrilrori Leu. 1981. 22. 3749: b) S. Morgan.
J. E. Jackson. M. Platz. J . A t i i . Che17i.Soc. 1991. 113. 2728.
d that the equilibrium concentration of 'Ad: must be less
that T.AS,, = 0.62 kcalmol-' is theonlyentropycomponeiil th'it changes. the louer limit for the S,T enthalpy gap in Ad: is
4 17 kcal mol '.
U . H. Bi-inker. R . Buchkremer. M. Kolodziejczyk. R. Kupfer. M. Rosenberg,
M D Poliks. M Orlando. M. Gross. Angrii.. Cheni. 1993. 105. 1427: Anyeit,.
C'hrn7 1111.E d Eny/. 1993. 32. 1344.
W. H. Green, N. C. Handy. P. J. Knowles. S. Carter. J Chein. PIIJS.1991. 94,
118. a n d references therein.
WK have also carried out I N D O 3 calculations (using geometries optimized
with ah initio methods) on 3 and 4 which indicate that neither compound
should shcw visible absorptions and hence cannot be the carriers of the 620 nm
tive space multirererence S C F including the sp hybrid orbitals and
the p atomic orbital at the carbene center in the active space The closed-shell
singlet states of carbenes require a [2,2] multireference wavefunction for a
correct description. For the 'B, states of Ad:. this amounts to a single-determiiimt R O H F calculation. For 4 ("adamantene") that has bii-adicaloidcharacter
due to its twisted double bond [13], the active space included the K and x*
MO's. whereas 3 was treated at the single-determinant R O H F level. Unconstrained seometry optimizations were carried out for all species and states
considered. The 'B, state of Ad: converged in this procedure to a structure of
C,, symmetry. and the closed-shell ' A , state to a distorted C, geometry that,
however. lies only 0.03 kcalmol- below the C,,-constrained minimum. As
expected. 3 has C, symmetry and 4 is unsymmetrical.
J. Michl. G. J. RadLiszewski, J. W. Downing. K. B. Wiberg, F. H. Walkerm R.
D. Miller. P. Kovacic, M. Jawdosiuk. V. BonaZ-Koutecky. Pure Appl. C/irm.
1983. 5.7. 315.
a) M. M . Gallo. H. F. Schaefer 111, J PI7ys. Chhpn7. 1992, 96, 1515; b) an
excellent discussion of the factors influencing the S/T gap in methylcarbene has
recently been provided by S. Khodabandeh, E. Carter, ihid. 1993, 97, 4360.
We have calculated the SIT gap in methyl- and dimethylcarbene at the same
level of theory used for our calculations on Ad:. both at their equilibrium
geoinetries (where the angle 3 at the carbene carbon center is 105 and 1 1 0 for
the singlets. respectively. and 129- for the triplets of both compounds) and also
by iixing the angle 1at the value it adopts in optimized 'Ad: (109.5") and 'Ad:
(11h.5 ) , respectively. Thus, the calculated S T gap in methylcarbene is
10.2 kcalmol-I, that is. some 6 kcalmol-' too high [14], and it reduces to
8.4 kcalmol-' at the Ad: valence angles. For dimethylcarbene the S / T gap is
6.1 and 3 6 kcalmol-'. respectively. Subtracting the approximately 6 kcal
mol I error (from methylcarbene) yields a preference for the triplet ground
state in dimethylcarbene (at the Ad: valence angles) of 2-3 kcalmol-I. which
should he similar in Ad:, in agreement with experiment.
CI bvith all singly excited configurations as implemented in GAUSS1ANrz 117):
J. B. Foresman. M. Head-Gordon. J. A. Pople. M . J. Frisch. .
Phys. C/zem.
1992. 96. 135.
SG1-G97 Rev. E.2and IBM-RS6000-G92 Rev. C). M. J. Frisch, G. W. Trucks,
M Head - Gordon, P. M. W. Gill. M. W. Wong. J. B. Foresman, B. G. Johnson,
H. B. Schlegel. M. A. Robbs. E. S. Replogle. R. Gomperts, J. L. Andres. K.
Raghavachari, J. S. Binkley. C. Gonzales. R. L. Martin. D. J. Fox. D. J. Defrees. J. Baker. J. J. P. Stewart, J. A. Pople, Gaussian Inc., Pittsburgh, PA, 1992.
B ) A . C. G . Adding. J. Strating. H. Wynberg. J. L. M. A. Schlatmann, Cliem.
('ommu1 1966. 657; b) S. D. Isaev. A. G. Yurchenko. F. N . Stepanov, G. C.
Kolyada. S. Novikov. J. Org. Cheni. (En,./. Eansl.) 1973, 9, 724; c) H. Bayley,
J. R Knowles, Bioclwniistrr. 1978, 17, 2420; hid. 1980. 19. 3883
[19] T. Bally in Radicd Ionic Systems (Eds.: A. Lund, M . Shiotani), Kluwer,
Doordrccht. 1991. p.3.
M W. Schmidt. K . K . Baldridge. J. A. Boatz. J. H. Jensen, S. Koseki. M. S.
Gordon. K . A. Nguyen. T. L. Windus. S. T. Elbert. QC'PE Bull. 1990, 10, 52.
VCH Ver/uysge.w/lschufrmhH, 0-69451 Weinheim, 1994
3,3'-Bis(dicyanomethyiene)-4,4,4',4'-tetra~ethyi2,2'-bithiolanyiidene, a Compound Containing
the Fundamental Chromophore of Thioindigo:
( E / Z ) and Valence Isomers, Thermoand Photochromism""
Andreas Pawlik, Walter Grahn," Axel Reisner,
Peter G. Jones, and Ludger Ernst
Dedicated to Professor CliriJtiun Reicliurdt
on the occasion of his 60th hirthduy
The reversible interconversion of the E / Z isomers of thioindig0 1 a (X = oxygen, Scheme 1 ) has been thoroughly investigated in solution and in the adsorbed state.['. Possible applications of the photochromism of 1 a, particularly for optical data
storage and the conversion of solar energy, have stimulated considerable work.['] The valence isomer pair ( Z ) - 1 / 2should show
a stronger photochromism than the ( E / Z )isomers 1. However,
cis-thioindigo (Z)-la cannot be transformed into the corresponding heteroanalogous cyclohexadiene 2 a. In contrast, the
indigoid dyes l b (X = sulfur)[31 and I c (X = dicyanom e t h y l e ~ ~ e )cannot
be isolated and occur exclusively in the
form of the 1,2-dithiine 2 b and cyclohexadiene 2c, respectively.
( E ) - la-c
- la-c
Scheme 1. Thioindigodyesl(1a:X = O ; l b [ 3 ] : X = S ; l c [ 4 ] : X = C(CN)J,their
valence isomers 2 [ 2 a :X = 0;2 b [3]: X = S; 2c [4]: X = C(CN),], and corresponding systems 3 with the fundamental chromophore of indigo [3a 161: X = 0 . n = 1.
3b: X = S, n = I ; 3c [8]: C(CN),, 17 = l ; 3d [9]: X = 0. n = 21.
This is certainly due in part to the gain in n stabilization energy
through the benzothiophene unit. This kind of stabilization is
impossible for the dicyanomethylene analogue 3cC5' of 2.2'bithiolanylidene 3 a,161 a coinpound having the fundamental
chromophore of thioindigo. Quantum mechanical calculations
(PM3"I) underline this: the heats of formation for ( E ) - 3 c and
( Z ) - 3 c , 202.1 and 202.2 kcalmol-' respectively, are approxi[*] Priv.-Doz. Dr. W. Grahn, DipLChem. A. Pawlik. DipLChem. A. Reisner
lnstitut fur Organische Chemie der Technischen Universitlt
Hagenring 30, D-38106 Braunschweig (FRG)
Telefax: Int. code + (531)391-5388
Prof. Dr. P. G. Jones
Institut fur Anorganische und Analytische Chemie
der Technischen UniversitHt Braunschweig
Prof. Dr. L. Ernst
NMR-Laboratorium der Chemischen Institute
Technische Universitit Braunschweig
This work was supported by the Fonds der Chemischen Industrie. A. P. thanks
the state of Lower Saxony for a fellowship. We thank Prof. W. Luttke. Universitit Gottingen, for helpful discussions.
0570-0833:Y4,'1919-1966 5 10.00+ .25;0
Angov. CIiem. Itit. Ed. Enxl. 1994. 33. No. I Y
Без категории
Размер файла
351 Кб
grounds, spectroscopy, adamantylidene, matrix, single, state, dialkylcarbene
Пожаловаться на содержимое документа