Photochemistry of (η6-2,6-X2C5H3 N)Cr(CO)3 (X = H, CH3, (CH3)3Si). First example of a photoinduced ring-slip at an (η6-arene)M(CO)3 center.

Article abstract of DOI:10.1021/om960090+

The photochemistry of (η⁶-2,6-X₂C₅H₃N)Cr(CO) ₃ was investigated both in low-temperature matrices (X = H or (CH₃)₃Si) and in room-temperature solution (X = H, CH₃, or (CH₃)₃ Si), Room-temperature photolysis (λ_exc > 410 nm) in CO-saturated methanol or acetonitrile produced (η¹-2,6-X₂C₅H₃N)Cr(CCO) ₅ which subsequently formed Cr(CO)₆ in a secondary photochemical process (X = H or CH₃), The efficiency of pentacarbonyl formation is lower in CO-saturated cyclohexane and follows the order X = H > X = CH₃, Photolysis in low-temperature matrices resulted in an η⁶ to η¹ pyridine ring-slippage (λ_exc = 460 nm; X = H). Visible irradiation in a CO-doped methane matrix produced (η¹-C₅H₅N)Cr(CO)₅, while in an N₂ matrix fac-(η¹-C⁵H₅N)N_{2 2}Cr(CO)₃ is formed. Irradiation with λ_exc = 308 nm produced both the ring-slippage product and also the CO-loss product (η⁶-C₅H₅)Cr(CO)₂, which in a N₂ matrix is trapped as (η⁶- C₅H₅N)Cr(CO)₂(N₂). Time-resolved infrared spectroscopy in cyclohexane revealed only the CO-loss product (λ_exc = 308 nm; X = H). The apparent difference in room-temperature and low-temperature photochemistry is explained by a rapid regeneration of (η⁶- C₅H₅N)Cr(CO)₃ from the η¹-intermediate. This explanation was supported by laser flash photolysis experiments (λ_exc = 355 nm) in CO-saturated cyclohexane (Sol), where the recovery of the (η⁶- C₅H₅N)Cr(CO)₃ absorption follows a biphasic time profile, whereby the faster process was assigned to the η¹ to η⁶ transformation and the slower to the reaction Of (η⁶- C₅H₅N)Cr(CO)₂(Sol) with CO. Crystals of (η⁶-2,6-(CH₃)₂C₅H ₃N)Cr(CO)₃ and (η⁶-2,6-((CH₃)₃Si)₂C ₅H₃N)Cr(CO)₃ were characterized by X-ray diffraction.

Full text of DOI:10.1021/om960090+

Organometallics 1996, 15, 3679-3687
3679
P h otoch em istr y of (η6-2,6-X2C5H3N)Cr (CO)3 (X ) H, CH3,
(CH3)3Si). F ir st Exa m p le of a P h otoin d u ced Rin g-Slip a t
a n (η6-Ar en e)M(CO)3 Cen ter . Molecu la r Str u ctu r es of
(η6-2,6-(CH3)2C5H3N)Cr (CO)3 a n d
(η6-2,6-((CH3)3Si)2C5H3N)Cr (CO)3
Ciara J . Breheny,1a Sylvia M. Draper,1b Friedrich-W. Grevels,1c
Werner E. Klotzbu¨cher,1c Conor Long,*,1a Mary T. Pryce,1d and
Graham Russell*,1c
School of Chemical Sciences, Dublin City University, Dublin 9, Ireland, and
Max-Planck-Institut fu¨r Strahlenchemie, Postfach 10 13 65,
D-45413 Mu¨lheim an der Ruhr, Germany
Received February 8, 1996X
The photochemistry of (η6-2,6-X2C5H3N)Cr(CO)3 was investigated both in low-temperature
matrices (X ) H or (CH3)3Si) and in room-temperature solution (X ) H, CH3, or (CH3)3 Si).
Room-temperature photolysis (λexc > 410 nm) in CO-saturated methanol or acetonitrile
produced (η1-2,6-X2C5H3N)Cr(CO)5 which subsequently formed Cr(CO)6 in a secondary
photochemical process (X ) H or CH3). The efficiency of pentacarbonyl formation is lower
in CO-saturated cyclohexane and follows the order X ) H > X ) CH3. Photolysis in low-
temperature matrices resulted in an η6 to η1 pyridine ring-slippage (λexc ) 460 nm; X ) H).
Visible irradiation in a CO-doped methane matrix produced (η1-C5H5N)Cr(CO)5, while in an
N2 matrix fac-(η1-C5H5N)(N2)2Cr(CO)3 is formed. Irradiation with λexc ) 308 nm produced
both the ring-slippage product and also the CO-loss product (η6-C5H5)Cr(CO)2, which in a
N2 matrix is trapped as (η6- C5H5N)Cr(CO)2(N2). Time-resolved infrared spectroscopy in
cyclohexane revealed only the CO-loss product (λexc ) 308 nm; X ) H). The apparent
difference in room-temperature and low-temperature photochemistry is explained by a rapid
regeneration of (η6-C5H5N)Cr(CO)3 from the η1-intermediate. This explanation was supported
by laser flash photolysis experiments (λexc ) 355 nm) in CO-saturated cyclohexane (Sol),
where the recovery of the (η6-C5H5N)Cr(CO)3 absorption follows a biphasic time profile,
whereby the faster process was assigned to the η1 to η6 transformation and the slower to
the reaction of (η6-C5H5N)Cr(CO)2(Sol) with CO. Crystals of (η6-2,6-(CH3)2C5H3N)Cr(CO)3
and (η6-2,6-((CH3)3Si)2C5H3N)Cr(CO)3 were characterized by X-ray diffraction.
In tr od u ction
process (M ) Cr).4 To date, there are no reports in the
literature of photoinduced haptotropic reactions for this
class of compound.5 It is accepted that CO loss is the
dominant photoprocess in either low-temperature ma-
trices,6 room-temperature solution,4 or the gas phase.7
However, some indirect evidence for a second photo-
product, possibly one involving a haptotropic or ring-
slip process, was obtained in our studies on the photo-
chemistry of (η6-C6H6)Cr(CO)3.8 Additional transient
signals were observed following flash photolysis which
could not be assigned to the known photoproduct (η6-
C6H6)Cr(CO)2(solvent) or any species formed by a
subsequent reaction of this intermediate. Consequently,
we have investigated the photochemistry of a number
of (η6-arene)M(CO)3 compounds in which the arene
ligand contains a functionality designed to “trap” ring-
slip intermediates, and the first results of these inves-
tigations are reported here.
The chemistry of (η6-arene)M(CO)3 (M ) Cr or Mo)
has been the subject of numerous investigations, prin-
cipally because of their use in both directing and
promoting reactions at the arene ligand.2 Also of
interest have been the mechanisms of thermally induced
ring exchange processes, which proceed via a stepwise
ring displacement.3 While arene exchange is also
observed under photolytic conditions, the CO-loss prod-
uct, (η6-arene)M(CO)2, has been implicated in this
X Abstract published in Advance ACS Abstracts, August 1, 1996.
(1) (a) Dublin City University. (b) Present address: Department of
Chemistry, University of Dublin, Trinity College, Dublin 2, Ireland.
(c) The Max-Planck Institute. (d) Present address: Dipartimento di
Chimica Organica e Industriale, Universita` degli Studi di Milano, via
Golgi 19, 20133 Milano, Italy.
(2) Transition metal organometallics for organic synthesis; McQuil-
lin, F. J ., Parker, D. G., Stephenson, G. R., Eds.; Cambridge University
Press: Cambridge, U.K., 1991.
(3) (a) Strohmeier, S.; Starrico, E. Z. Phys. Chem. Neue Folge. 1966,
38, 315. (b) Mahaffy, C. A. L.; Pauson,P. L. J . Chem. Res. (S) 1979,
126. (c) Mahaffy, C. A. L.; Pauson, P. L. J . Chem. Res. (M) 1979, 1752.
(d) Zimmerman, C. L.; Shaner, J . L.; Royh, S. A.; Willeford, B. R. J .
Chem. Res. (S) 1980, 108. (e) Traylor, T. G.; Stewart, K. J .; Goldberg,
M. J . J . Am. Chem. Soc. 1984, 106, 4445. (f) Traylor, T. G.; Stewart,
K. J .; Goldberg, M. J . Organometallics 1986, 5, 2062. (g) Traylor, T.
G.; Stewart, K. J . J . Am. Chem. Soc. 1986, 108, 6977. (h) Traylor, T.
G.; Goldberg, M. J . Organometallics 1987, 6, 2413. (i) Traylor, T. G.;
Goldberg, M. J . Organometallics 1987, 6, 2531.
(4) Gilbert, A.; Kelly, J . M.; Budzwait, M.; Koerner von Gustorf, E.
Z. Naturforsch. 1976, 31B, 1091.
(5) Pryce, M. T. Ph.D. Thesis, Dublin City University, 1994.
(6) Rest, A. J .; Sodeau, J . R.; Taylor, D. J . J . Chem. Soc., Dalton
Trans. 1978, 651.
(7) Zheng, Y.; Wang, W.; Lin, J .; She, Y.; Fu, K.-J . J . Phys. Chem.
1992, 96, 9821.
(8) McGrath, I. M. Ph.D. Thesis, Dublin City University, 1993.
S0276-7333(96)00090-8 CCC: $12.00 © 1996 American Chemical Society

3680 Organometallics, Vol. 15, No. 17, 1996
Breheny et al.
Ta ble 1. Su m m a r y of Cr ysta l Da ta , In ten sity Mea su r em en t, a n d Refin em en t for
(η6-2,6-(CH3)2C5H3N)Cr (CO)3 a n d (η6-2,6-((CH3)3Si)2C5H3N)Cr (CO)3
compd
(η6-(CH3)2C5H3N)Cr(CO)3
(η6-((CH3)3Si)2C5H3N)Cr(CO)3
formula
cryst system
space group
C10H9NO3Cr
orthorhombic
Pna21
C14H21NO3CrSi2
monoclinic
P21/n
a, Å
b, Å
c, Å
13.0308(13)
6.2875(6)
12.7099(10)
7.7088(6)
23.0928(9)
10.4365(8)
97.312(4)
1842.8(2)
4
1.296
0.758
752
â, deg
V, Å3
1041.3(2)
4
1.551
1.082
Z
d(calcd), g cm-3
abs coeff, mm-1
F(000)
496
cryst size, mm
θ range, deg
index ranges: h; k; l
reflcns collcd
indepdt reflcns
refinement method
data/restraints/params
goodness-of-fit on F2
final R indices [I > 2σ(I)]
R indices (all data)
max diff peak, hole, e Å-3
0.5 × 0.5 × 0.25
3.13-25.47
0 to 13; -7 to 7; 0 to 14
1790
940 (Rint ) 0.031)
full matrix on F2
940/1/155
0.4 × 0.4 × 0.3
1.76-24.96
0 to 8; 0 to 23; -11 to 11
3326
3096 (Rint ) 0.0177)
full-matrix on F2
3096/0/203
0.907
1.082
R1 ) 0.024; wR2 ) 0.060
R1 ) 0.055; wR2 ) 0.076
0.302, -0.253
R1 ) 0.032; wR2 ) 0.078
R1 ) 0.074; wR2 ) 0.092
0.308, -0.229
spectroscopy. Crystals of (η6-2,6-((CH3)3Si)2C5H3N)Cr(CO)3 for
structural analysis were grown from diethyl ether/n-hexane
mixtures.
Heteroarenes, such as pyridine, constitute an ideal
ligand type for these studies. These ligands offer the
possibility of coordination via the ring π-electron system
or, alternatively and more commonly, via the nitrogen
atom. Furthermore the electronic spectra of all (η6-
arene)Cr(CO)3 compounds are dominated by metal to
ligand charge transfer transitions.9 Both the HOMO
and LUMO energies in pyridine are lower than the
corresponding orbitals in benzene. Pyridine is therefore
a poorer π-donor but a better π-acceptor ligand than
benzene.10 These differences may have significant
consequences for the photochemistry of η6-coordinated
pyridine compounds. However, to our knowledge, there
are no reports of such investigations. Consequently we
have investigated the photochemistry of (η6-C5H5N)Cr-
(CO)3 and its 2,6-disubstituted derivatives by matrix
isolation and time-resolved spectroscopic techniques in
the hope of detecting haptotropic changes at the het-
eroarene ligand.
Structural determinations of (η6-2,6-(CH3)2C5H3N)-
Cr(CO)3 and (η6-2,6-((CH3)3Si)2C5H3N)Cr(CO)3 were
made to allow comparisons with the published molec-
ular structure of (η6-C5H5N)Cr(CO)3.11 In particular the
effects of the bulky substituents in positions R to the
nitrogen atom were of interest and also whether their
presence could influence the subsequent thermal reac-
tions following photochemical excitation.
Syn th esis of (η6-(CH3)2C5H3N)Cr (CO)3. Cr(CO)6 (0.6 g)
in 10 mL of dry THF was added to 50 mL of 2,6-(CH3)2C5H3N
and the solution brought to its reflux temperature for 8 h
under a nitrogen atmosphere. The solution was then cooled
to room temperature, and the solvent was removed under
reduced pressure. The resulting yellow solid was purified by
flash chromatography on neutral alumina (7:3 petroleum
ether:diethyl ether (v/v) as the mobile phase). The identity
and purity of this compound was confirmed by the standard
range of spectroscopic and analytical techniques. Crystals for
structural analysis were grown from diethyl ether/n-hexane
mixtures.
Ap p a r a tu s. Standard equipment included a Bruker AC400
NMR, a Perkin-Elmer 2000 FTIR, and a Hewlett Packard
8452A diode array UV/vis spectrophotometer. The matrix
isolation apparatus has been described elsewhere,13 including
the devices for controlling the temperature (10-12 K), the flow
of matrix gas (1.5-2 mmol h-1), and metal complex deposition
rate, the set-up for monochromatic and broad-band irradiation,
and the IR and UV/vis spectrophotometers. The evaporation
temperatures (30-35 °C) for (η6-C5H5N)Cr(CO)3 and (η6-2,6-
((CH3)3Si)2C5H3N)Cr(CO)3 were adjusted (quartz balance) to
achieve a guest/host ratio e 1/1000. The instrumentation for
laser flash photolysis (308 nm, XeCl excimer laser, attenuated
to 90-100 mJ /pulse) in combination with time-resolved IR
detection (CaF2 cell, d ) 1 mm; globar IR source; HgCdTe
detector; in the present configuration the system response is
1-2 µs), including the treatment of the sample solution, was
recently described in detail14 as was the UV/vis flash photolysis
equipment.15 Samples for the UV/vis monitored flash pho-
tolysis experiments were degassed by three cycles of freeze-
pump-thaw to 10-3 Torr, followed by a substantial liquid
pump to remove traces of water introduced in the degassing
procedure. The samples were analyzed in a quartz cuvette
Exp er im en ta l Section
Syn th esis of (η6-C5H5N)Cr (CO)3 a n d (η6-2,6-((CH3)3Si)2-
C5H3N)Cr (CO)3. These complexes were synthesized by the
method of Davies et al.12 and were purified by recrystallization
from diethyl ether/n-hexane mixtures. They were character-
ized by the standard spectroscopic and analytical techniques.
In each case their purity was confirmed by both IR and NMR
(13) (a) Gerhartz, W.; Grevels, F.-W.; Klotzbu¨cher, W. E. Organo-
metallics 1987, 6, 1850. (b) Klotzbu¨cher, W. E. Cryogenics 1983, 23,
554.
(14) Grevels, F.-W.; Klotzbu¨cher, W. E.; Schrickel, J .; Schaffner, K.
J . Am. Chem. Soc. 1994, 116, 6229.
(15) Creaven, B. S.; George, M. W.; Ginzburg, A. G.; Hughes, C.;
Kelly, J . M.; Long, C.; McGrath, I. M.; Pryce, M. T. Organometallics
1993, 12, 3127.
(9) Caroll, D. G.; McGlynn. S. P. Inorg. Chem. 1968, 7, 1285.
(10) J orgenson, W. L.; Salem, L. The Organic Chemists Book of
Orbitals; Academic Press: New York, 1973.
(11) Draper, S. M.; Byrne, J . J .; Breheny, C. J .; Long, C.; Low, J . N.
Acta Crystallogr. 1994, C50, 1669.
(12) Davies, S. G.; Shipton, M. R. J . Chem. Soc., Perkin Trans. 1
1991, 501.

(η6-2,6-X2C5H3N)Cr(CO)3
Organometallics, Vol. 15, No. 17, 1996 3681
Ta ble 2. Sp ectr oscop ic Da ta for th e Com p ou n d s in Th is In vestiga tion
b
compd
λmax (ꢀ)a
ν
CO (cm-1
)
1H NMRc
Cr(CO)6
280
1986d
(η6-C5H5N)Cr(CO)3
(η6-C5H5N)Cr(CO)3
(η6-C5H5N)Cr(CO)3
(η6-C5H5N)Cr(CO)3
316 (8.36 × 103)
1997, 1939, 1926d
2000, 1942, 1926
1996, 1935, 1920
1999, 1938, 1924
6.55 (m, 2H), 5.65 (m, 1H), 5.23 (m, 2H)
e
f
g
(η1-C5H5N)Cr(CO)3
1957, 1841, 1833
1956, 1841, 1833
g
h
(η1-C5H5N)Cr(CO)3
(η6-C5H5N)Cr(CO)2(Sol)d
1950, 1890
1945, 1888
1957, 1910
1967, 1897, 1885
2030, 1909, 1907, 1870
2068, 1938, 1919d
2070, 1938, 1914
(η6-C5H5N)Cr(CO)2
g
(η6-C5H5N)Cr(CO)2(N2)e
fac-(η1-C5H5N)(N2)2Cr(CO)3
e
(η1-C5H5N)Cr(CO)4
h
(η1-C5H5N)Cr(CO)5
410
8.5 (d, 2H), 7.7 (t, 1H), 7.3 (m, 2H)
(η1-C5H5N)Cr(CO)5
f
(η6-2,6-(CH3)2C5H3N)Cr(CO)3
318 (6.07 × 103)
324 (7.69 × 103)
1987, 1926, 1917d
5.65 (t, 1H), 5.12 (d, 2H), 2.41 (s, 6H)
5.48 (d, 2H), 5.29 (t, 1H), 0.3 (s, 18H)
(η6-2,6-((CH3)3 Si)2C5H3N)Cr(CO)3
1978, 1912 (br)d
1986, 1923, 1918
1988, 1926, 1919
(η6-2,6-((CH3)3 Si)2C5H3N)Cr(CO)3
(η6-2,6-((CH3)3 Si)2C5H3N)Cr(CO)3
f
e
(η6-2,6-((CH3)3 Si)2C5H3N)Cr(CO)2
1942, 1890
1947, 1907
f
(η6-2,6-((CH3)3Si)2C5H3N)Cr(CO)2(N2)e
(η1-2,6-((CH3)3Si)2C5H3N)Cr(CO)3
1954, 1854, 1848
f
a
b
c
d
nm in cyclohexane solution ( 2 nm ((1% L mol-1 cm-1). ( 1 cm-1
.
ppm in CDCl3 at 400 MHz. In cyclohexane solution at 298
e
f
g
h
K. In an N2 matrix at 12 K. 10% CO in Ar matrix at 12 K. In a CH4 matrix at 10 K. 10% CO in CH4 matrix at 12 K.
with four optical flat walls. The excitation and monitoring
beams were arranged in a cross-beam configuration.
Sin gle-Cr ysta l Str u ctu r a l Deter m in a tion s. X-ray dif-
fraction data were collected on an Enraf-Nonius CAD4 auto-
matic diffractometer at room temperature. The structures
were solved using the Patterson strategy of Shelxs-8616 and
refined by full-matrix least squares on F2 using Shelxl-93.17
Manipulations of the molecular structures were made using
Schakal-92.18 The crystal, intensity, and refinement data are
presented in Table 1.
Resu lts
The spectroscopic details of the compounds in this
investigation are given in Table 2, and the UV/vis
spectrum of (η6-C5H5N)Cr(CO)3 in cyclohexane solution
is given in Figure 1a. The compound shows the char-
acteristic absorptions of (η6-arene)Cr(CO)3 compounds
with λmax at approximately 320 nm. Unfortunately, the
valley in the absorption profile at 290 nm is not as
pronounced in this compound as in the (η6-C6H6)Cr(CO)3
analogue. This valley has been used as a window into
F igu r e 1. (a) UV/vis spectrum of (η6-C5H5N)Cr(CO)3 (2 ×
10-4 M) in cyclohexane and (b) its infrared spectrum in the
the UV region for the detection of transient species
which have little or no absorptions in the visible region
of the spectrum. The infrared spectrum of (η6- C5H5N)-
Cr(CO)3 in the νC-O region is presented in Figure 1b.
Three absorptions of almost equal intensity are observed
for this compound which has Cs symmetry. This makes
the interpretation of the matrix isolation and TRIR
results both simple and unambiguous. The number of
νC-O region also in cyclohexane solution.
are presented in Figure 2 and indicate a reduction in
the intensity of the bands of the parent, with concomi-
tant formation of bands assigned to (η1-C5H5N)Cr(CO)5.
This was confirmed by comparison of the final spectrum
with that of an authentic sample of (η1-C5H5N)Cr(CO)5
prepared independently. The relative quantum ef-
ficiency for the disappearance of (η6-2,6-X2C5H3N)Cr-
(CO)3 depends on the solvent and the substituent X as
measured by UV/vis spectroscopy (λexc > 410 nm). The
reduction in the absorbance of the η6 species was
measured for photochemical conversions of <15% to
minimize internal light filter effects by the product. The
results were normalized to account for minor variations
in the absorbance between the particular solutions. The
νC-O absorption bands will equal the number of carbonyl
ligands in a particular photofragment.
St ea d y-St a t e P h ot olysis of (η6-2,6-X2C5H 3N)Cr -
(CO)3. The photochemistry of (η6-2,6-X2C5H3N)Cr(CO)3
was investigated in either CO-saturated acetonitrile,
1
methanol, or cyclohexane and followed by H NMR or
UV/vis spectroscopy (λexc > 410 nm). The NMR changes
in the aromatic region observed in CD3CN for X ) H
(16) Sheldrick, G. M. SHELXS-86, Programme for Crystal Structure
Determination. Acta Crystallogr. 1990, A46, 467.
(17) Sheldrick, G. M. SHELXL-93, Programme for Crystal Structure
Refinement, University of Go¨ttingen, Go¨ttingen, Germany, 1993.
(18) Keller, E. SCHAKAL-92, A Programme for Graphical repre-
sentation of Molecular Structures, Kristallographisches Institut der
Universitaet, Freiburg, Germany, 1992.

3682 Organometallics, Vol. 15, No. 17, 1996
Breheny et al.
F igu r e 3. (a) Difference spectrum obtained following
irradiation (λexc ) 460 nm) of a CH4 matrix containing (η6-
C5H5N)Cr(CO)3 at 10 K and (b) spectrum obtained in a
separate experiment following irradiation with λexc ) 308
nm (negative peaks indicate depletion of the parent com-
pound).
F igu r e 2. 1H NMR spectral changes in the aromatic region
observed following photolysis (λexc > 410 nm; 275 W Xe-
arc lamp) of (η6-C5H5N)Cr(CO)3 in CO-saturated CD3CN
(at 400 MHz). Spectrum a represents the initial spectrum.
Spectrum b was obtained following 55 min of irradiation
indicating the formation of the (η1-C5H5N)Cr(CO)5 as
verified by comparison with the spectrum of (η1-C5H5N)-
Cr(CO)5 synthesized by an independent route.
new two-band pattern compares with that observed for
(η6-C6H6)Cr(CO)2 (55 cm-1), photogenerated from (η6-
C6H6)Cr(CO)3 in a methane matrix,6 thereby suggesting
the assignment as the CO-loss product (η6-C5H5N)Cr-
(CO)2 (reaction 1). Further support for this assignment
Sch em e 1
(1)
comes from a comparison of the frequency shifts in going
from the respective parent tricarbonyl complexes to (η6-
C6H6)Cr(CO)2 (56 and 43 cm-1)6 and (η6-C5H5N)Cr(CO)2
(54 and 43 cm-1); in the latter case the average wave-
number of the two low-energy bands of the parent
complex (1931 cm-1) was taken.
process is 2.5 times more efficient in CH3OH than C6H12
and 2.3 times more efficient for X ) H than for X )
CH3 in cyclohexane. For X ) (CH3)3Si the efficiency is
essentially zero in cyclohexane. In this latter solvent
the (η1-C5H5N)Cr(CO)5 produced subsequently under-
goes further photochemical loss of the pyridine ligand
producing Cr(CO)6 (confirmed by 13C NMR). The overall
reaction sequence is outlined in Scheme 1.
To determine the degree to which the hapticity of the
pyridine-chromium interaction has been changed fol-
lowing long-wavelength irradiation, experiments were
conducted in both an N2- and a CO-doped methane
matrix. Irradiation in an N2 matrix (λexc ) 460 nm)
produced three bands (1967, 1897, and 1885 cm-1) along
with two absorptions of comparable intensities in the
Ma tr ix Isola tion P h otoch em istr y of (η6-C5H5N)-
Cr (CO)3. The photochemistry of (η6-C5H5N)Cr(CO)3
was investigated in a methane matrix at 10 K. With
long-wavelength irradiation (λexc ) 460 nm) the bands
of the parent species at 1999, 1938, and 1924 cm-1 were
depleted with concomitant formation of three bands at
1957, 1841, and 1833 cm-1 (Figure 3a), indicative of a
fac-Cr(CO)3 moiety. These bands were assigned to a
heteroarene ring-slip product of the type (ηx-C5H5N)-
Cr(CO)3 (x < 6). Irradiation at shorter wavelength (λexc
) 308 nm) yields the same species but in addition
produces further bands at 1945 and 1888 cm-1 (the
latter band is subject to matrix splitting, Figure 3b)
along with a weak feature at 2136 cm-1 attributable to
free CO. The frequency separation of 57 cm-1 in the
ν
N-N region (2222 and 2189 cm-1) as presented in Figure
4a. This pattern is indicative of a facial arrangement
of three CO groups and two cis-coordinated N2 ligands,19
which logically implies an η1 coordination of the pyridine
(19) The wavenumber difference between the νN-N antisymmetric
and symmetric vibrational modes of 33 cm-1 is similar to the 39 cm-1
difference observed in (η5-C5H5)Nb(CO)2(N2)2: George, M. W.; Haward,
M. T.; Hamley, P. A.; Hughes, C.; J ohnson, F. P. A.; Popov, V. K.;
Poliakoff, M. J . Am. Chem. Soc. 1993, 115, 2286.

(η6-2,6-X2C5H3N)Cr(CO)3
Organometallics, Vol. 15, No. 17, 1996 3683
F igu r e 5. Infrared spectra recorded before (s) and after
(- - -) irradiation (λexc ) 460 nm; 60 min) of (η6-C5H5N)Cr-
(CO)3 in a CH4:CO matrix (10:1) showing the formation of
(η1-C5H5N)Cr(CO)5 (marked *) and Cr(CO)6 (1983 cm-1).
The absorptions above 2080 cm-1 originate from the carbon
monoxide in the matrix.
carbonyl fragment yields (η1-C5H5N)Cr(CO)3 (1956,
1841, and 1831 cm-1), which is identified by comparison
with the data obtained in a methane matrix (cf. Table
2), for (η1-C5H5N)Cr(CO)3 generated from (η6-C5H5N)-
Cr(CO)3. Subsequent long-wavelength photolysis (λexc
> 530 nm) regenerated (η1-C5H5N)Cr(CO)5 at the ex-
pense of both (η1-C5H5N)Cr(CO)3 and (η1-C5H5N)Cr-
(CO)4. Further irradiation with λexc ) 400 nm produced
Cr(CO)6 (1984 cm-1), which dominates the spectrum as
the ultimate product of the overall reaction sequence,
and a band at 1958 cm-1 the assignment of which is
uncertain but could be either the 13C satellite of Cr(CO)6
or possibly one of the bands of Cr(CO)5(CH4).21
F igu r e 4. (a) Difference spectrum obtained following
photolysis of (η6-C5H5N)Cr(CO)3 in an N2 matrix (λexc ) 460
nm) showing the two bands in the νN-N region at 2221 and
2188 cm-1 (cf. inset) assigned to fac-(η1-C5H5N)(N2)2Cr(CO)3
and (b) spectral changes obtained following irradiation with
λexc ) 308 nm with the inset showing the νN-N bands for
the N2 adducts of both ring-slip and CO-loss species.
Ma t r ix P h ot och em ist r y of (η6-2,6-((CH 3)3Si)2-
C5H3N)Cr (CO)3. Photolysis (λexc ) 460 nm) of (η6-2,6-
((CH3)3Si)2C5H3N)Cr(CO)3 in an Ar matrix containing
10% CO failed to cause any significant IR spectral
changes. However photolysis with λexc ) 250 nm
resulted in a depletion of the parent bands and the
appearance of new bands at 1942 and 1890 cm-1
assigned to the CO-loss product (η6-2,6-((CH3)3Si)2-
C5H3N)Cr(CO)2. Subsequent irradiation at 308 nm lead
to a further increase of the bands of the dicarbonyl
species but also produced bands at 1954, 1854, and 1848
cm-1, which, by comparison with the photochemistry of
(η6-C5H5N)Cr(CO)3 in CH4 matrices, are assigned to (η1-
2,6-((CH3)3Si)2C5H3N)Cr(CO)3. White light photolysis
results in a depletion of all product bands and the re-
formation of the parent tricarbonyl compound, thus
illustrating an efficient photoreversal of both the CO-
loss and η1-species.
ligand (reaction 2). Irradiation of a similar matrix with
λexc ) 308 nm produced additional νC-O bands at 1957
and 1910 cm-1 (Figure 4b) while in this case the bands
observed in the νN-N region have significantly different
relative intensities (cf. insets in Figure 4). A weak
feature attributable to free CO (2140 cm-1) is also
observed in this experiment. These observations can
be explained by the simultaneous formation of the ring-
slip and the CO-loss product both of which in turn take
up N2 from the environment producing (η1-C5H5N)Cr-
(CO)3(N2)2 and (η6-C5H5N)Cr(CO)2(N2), respectively. The
νN-N band of the latter species overlaps the low-energy
In an N2 matrix, irradiation of (η6-2,6-((CH3)3Si)2-
C5H3N)Cr(CO)3 with λexc ) 460 nm resulted in the
depletion of the parent bands with concomitant libera-
tion of CO (2140 cm-1) and formation of (η6-2,6-((CH3)3-
Si)2C5H3N)Cr(CO)2(N2) (νC-O 1947 and 1907 cm-1; νN-N
) 2180 cm-1) albeit with low efficiency. This observa-
tion demonstrates that at this wavelength of excitation
the complex is not entirely photochemically inert, as
might be concluded from the experiment in the CO-
doped Ar matrix. On the basis of some additional weak
features in the νC-O region, one might conclude that the
η6 to η1 ring-slippage also takes place, but it was
impossible to determine if this species interacts with
the N2 from the matrix environment because of the
weakness of the bands.
νN-N band of the bis(dinitrogen) species at 2189 cm-1
.
Photolysis of (η6-C5H5N)Cr(CO)3 in a CH4:CO matrix
(10:1) (λexc ) 460 nm) produced a three-band pattern
(2070, 1937, and 1916 cm-1) characteristic of (η1-
C5H5N)Cr(CO)5 (Figure 5),20 thus indicating a photoin-
duced η6 to η1 haptotropic shift, followed by the takeup
of two CO ligands from the matrix. An additional band
at 1983 cm-1 assigned to Cr(CO)6 was also observed.
Subsequent photolysis with λexc ) 250 nm produced
peaks at 2031, 1909, 1907, and 1870 cm-1, which have
been assigned to the (η1-C5H5N)Cr(CO)4 species with a
vacant site cis to the pyridine ligand, produced from (η1-
C5H5N)Cr(CO)5. Photolytic loss of CO from the tetra-
(20) Braterman, P. S. Metal Carbonyl Spectra; Academic Press: New
York, 1975.
(21) Perutz, R. N.; Turner, J . J . J . Am. Chem. Soc. 1975, 75, 4791.

3684 Organometallics, Vol. 15, No. 17, 1996
Breheny et al.
F igu r e 6. TRIR difference spectrum in cyclohexane at
room temperature obtained 1 µs after the laser pulse (λexc
) 308 nm) of (η6-C5H5N)Cr(CO)3 (5 × 10-4 M), showing
depletion of the parent compound and the formation of (η6-
C5H5N)Cr(CO)2(cyclohexane).
F igu r e 7. UV/vis difference spectrum obtained following
5 µs after photolyis (λexc ) 355 nm) of (η6-C5H5N)Cr(CO)3
in CO-saturated cyclohexane at room temperature. The
inset shows the transient signal at 340 nm indicating that
two temporally resolved processes are associated with the
recovery of the parent absorption. The first process,
completed by the time indicated by the arrow, is assigned
to the η1 to η6 haptotropic reversal, while the second process
corresponds to the reaction of the CO-loss species with CO.
Tim e-Resolved In fr a r ed Sp ectr oscop y. Laser
flash photolysis of (η6-C5H5N)Cr(CO)3 was monitored by
TRIR spectroscopy at room temperature in CO-satu-
rated cyclohexane solution ([CO] ) 9.0 × 10-3 M).22
Figure 6 represents the difference spectrum obtained 1
µs after the laser pulse (λexc ) 308 nm). The negative
peaks at 1999, 1940, and 1930 cm-1 correspond to
depletion of the parent absorptions, while the product
bands at 1950 and 1890 cm-1 are close to those obtained
for (η6-C5H5N)Cr(CO)2 in the CH4 matrix experiments
(Table 2) and were assigned to (η6-C5H5N)Cr(CO)2(Sol)
(Sol ) cyclohexane). These bands decay by first-order
kinetics (kobs ) 2.1 × 105 s-1) with partial regeneration
of the starting material (reaction 3).
Ta ble 3. Secon d -Or d er Ra te Con sta n ts (k2) for th e
Rea ction of (η6-2,6-X2C5H3N)Cr (CO)2(Sol) w ith CO
a t 298 K
a
a
X
k2
1.4 TRIR
1.4 UV flash photolysis (CH3)3Si
technique
X
k2
1.5 UV flash photolysis
UV flash photolysis
technique
H
H
CH3
2
a
×10-7 ((10%) M-1 s-1 in cyclohexane solution.
spectrum where the photoproducts absorb significantly
is in the valley at 290 nm of the parent absorption
spectrum and a weak feature at approximately 500 nm.
Analysis of the decay kinetics associated with the
transient absorption at 280 nm (Table 3) yields the rate
constant for the reaction with CO (k2; reaction 3). The
transient signal at this wavelength indicates that the
system is not fully reversible. This absorption is as-
signed to the CO-loss product. Under these conditions,
the parent absorption (monitored at 340 nm) recovers
by two temporally resolved processes. The slower
process, i.e., that occurring after the time indicated by
the arrow in the inset in Figure 7, corresponds to the
re-formation of (η6-2,6-X2C5H3N)Cr(CO)3 species via the
reaction of (η6-2,6-X2C5H3N)Cr(CO)2(Sol) with CO. The
faster process is complete at the time indicated by the
arrow. The terminal impedence was reduced in efforts
to measure this signal accurately, but it appears to be
completed within a time scale too close to the theoretical
rise-time of the equipment (∼40 ns) for it to be reliably
measured and corresponds to the recovery of (η6-2,6-
X2C5H3N)Cr(CO)3 from the (η1-2,6-X2C5H3N)Cr(CO)3
intermediate.
At first sight, loss of CO with formation of the species
(η6-C5H5N)Cr(CO)2(Sol) seems to be the only process
observable by time-resolved IR spectroscopy. However,
on closer inspection it is apparent that some (η1-C5H5N)-
Cr(CO)5 is also detectable, as observed in steady-state
photolyses in CO-saturated solution (Scheme 1). Un-
fortunately, the high-frequency A1 vibration of (η1-
C5H5N)Cr(CO)5 at approximately 2070 cm-1 is inher-
ently weak (cf. Figure 5), which makes this feature
difficult to detect in the time-resolved experiments.
Moreover, the low-frequency A1 and the E stretching
vibrations are almost coincidental with the low-fre-
quency bands of the parent tricarbonyl (Figure 5; see
Table 2). The spectral changes accompanying the
partial recovery of the parent ultimately lead to a
difference spectrum showing positive absorptions at
1948, 1920, and 1880 cm-1. The former pair represent
the E and A1 modes of (η1-C5H5N)Cr(CO)5 mentioned
above, while the latter feature remains unidentified, a
dinuclear species being a plausible candidate.
Cr ysta l a n d Molecu la r Str u ctu r es of (η6-2,6-
(CH3)2C5H3N)Cr (CO)3 an d (η6-2,6-((CH3)3Si)2C5H3N)-
Cr (CO)3. Both molecules consist of a heteroarene ring
to which a chromium tricarbonyl unit is coordinated via
its ring π-system (Figure 8). The chromium atom is
located directly below the arene ring centroid, and the
arene ring is essentially planar. In the case of the 2,6-
(CH3)2C5H3N compound the carbonyl ligands eclipse the
2, 4, and 6 arene carbons; thus carbonyl C10-O2 is
trans to the ring nitrogen. This conformation is also
observed in the unsubstituted (η6-C5H5N)Cr(CO)3 com-
F la sh P h otolysis w ith UV/Vis Detection . The
photochemistry of (η6-2,6-X2C5H3N)Cr(CO)3 (X ) H,
CH3, or (CH3)3Si) was investigated by UV/vis flash
photolysis with λexc ) 355 nm in CO-saturated cyclo-
hexane. The difference spectrum obtained 5 µs after
the laser pulse is given in Figure 7 for X ) H. From
this spectrum it can be seen that the only region of the
(22) Makrancszy, J .; Megyery-Balog, K.; Rosz, L.; Patyt, D. Hung.
J . Indust. Chem. 1976, 4, 269.

(η6-2,6-X2C5H3N)Cr(CO)3
Organometallics, Vol. 15, No. 17, 1996 3685
Ta ble 4. Selected Bon d Len gth s (Å) a n d An gles
(d eg) w ith Esd ’s in P a r en th eses for
(η6-2,6-((CH3)3Si)2C5H3)Cr (CO)3 a n d
(η6-2,6-(CH3)2C5H3N)Cr (CO)3
(η6-2,6-((CH3)3Si)2-
C5H3)Cr(CO)3
(η6-(CH3)2C5H3N)-
Cr(CO)3
Cr(1)-C(9)
Cr(1)-C(10)
Cr(1)-C(11)
Cr(1)-N(1)
Cr(1)-C(2)
Cr(1)-C(5)
Cr(1)-C(6)
Cr(1)-C(3)
Cr(1)-C(4)
1.835(3)
1.846(3)
1.846(3)
2.180(2)
2.183(3)
2.198(3)
2.205(3)
2.204(3)
2.209(3)
1.86(2)
1.847(5)
1.82(2)
2.220(4)
2.21(2)
2.26(2)
2.191(14)
2.22(2)
2.242(5)
C(9)-Cr(1)-C(10)
C(9)-Cr(1)-C(11)
C(10)-Cr(1)-C(11)
C(9)-Cr(1)-N(1)
C(10)-Cr(1)-N(1)
C(11)-Cr(1)-N(1)
C(9)-Cr(1)-C(2)
C(10)-Cr(1)-C(2)
C(11)-Cr(1)-C(2)
N(1)-Cr(1)-C(2)
C(9)-Cr(1)-C(5)
C(10)-Cr(1)-C(5)
C(11)-Cr(1)-C(5)
N(1)-Cr(1)-C(5)
C(2)-Cr(1)-C(5)
C(9)-Cr(1)-C(6)
C(10)-Cr(1)-C(6)
C(11)-Cr(1)-C(6)
N(1)-Cr(1)-C(6)
C(2)-Cr(1)-C(6)
C(5)-Cr(1)-C(6)
C(9)-Cr(1)-C(3)
C(10)-Cr(1)-C(3)
C(11)-Cr(1)-C(3)
N(1)-Cr(1)-C(3)
C(2)-Cr(1)-C(3)
C(5)-Cr(1)-C(3)
C(6)-Cr(1)-C(3)
C(9)-Cr(1)-C(4)
C(10)-Cr(1)-C(4)
C(11)-Cr(1)-C(4)
N(1)-Cr(1)-C(4)
C(2)-Cr(1)-C(4)
C(5)-Cr(1)-C(4)
C(6)-Cr(1)-C(4)
C(3)-Cr(1)-C(4)
86.44(14)
85.44(14)
89.1(2)
118.98(11)
154.56(12)
91.83(12)
93.14(12)
153.91(13)
116.90(13)
36.53(8)
87.6(8)
84.8(2)
89.3(8)
101.8(6)
166.1(2)
101.6(7)
90.8(7)
136.1(8)
134.3(7)
35.2(6)
165.7(7)
103.6(7)
104.0(7)
65.6(5)
74.9(5)
133.6(6)
138.3(8)
88.4(7)
35.3(6)
62.6(2)
37.5(6)
105.9(6)
103.3(7)
163.6(7)
64. 5(6)
35.7(7)
63.2(2)
75.2(6)
136.6(8)
88.6(2)
138.3(8)
77.5(2)
35.2(6)
37.1(8)
66.7(5)
34.1(8)
154.24(13)
90.57(13)
120.12(13)
67.11(10)
78.58(11)
156.13(12)
117.41(13)
93.47(12)
37.16(9)
66.09(10)
37.09(10)
92.64(13)
116.26(14)
154.44(13)
66.85(10)
37.66(10)
65.88(12)
78.15(11)
117.15(13)
89.75(14)
157.25(13)
79.74(10)
67.26(11)
37.17(11)
67.09(11)
36.51(11)
F igu r e 8. ORTEP diagrams of the molecular structures
of (a) (η6-2,6-(CH3)2C5H3N)Cr(CO)3 and (b) (η6-2,6-((CH3)3-
Si)2C5H3N)Cr(CO)3 showing the atom numbering system
used (hydrogen atoms are omitted for clarity).
pound previously published.11 In the (η6-2,6-((CH3)3-
Si)2C5H3N)Cr(CO)3 compound the carbonyl groups are
in a staggered conformation and eclipse the ring bond
centers. This conformation appears to have little effect
on the Cr-ring-centroid distance which is identical in
the trimethylsilyl compound (1.698(3) Å) to that of the
unsubstituted pyridine compound (1.697(4) Å).11 Table
4 contains a selected listing of bond lengths and angles
for both compounds.
The photochemistry of (η6-C5H5N)Cr(CO)3 in a range
of isolating matrices exhibited a wavelength depen-
dency. Irradiation with λexc ) 460 nm results in the
formation of a fac-Cr(CO)3 unit which in a N2 matrix
yields a fac-Cr(CO)3(N2)2 fragment. These results are
consistent with an η6 to η1 haptotropic shift, thus
generating two vacant sites on the metal. In a CO-
doped matrix this species can capture two CO ligands
to produce (η1-C5H5N)Cr(CO)5, while in a N2 matrix fac-
(η1-C5H5N)Cr(CO)3(N2)2 is formed.
The matrix photochemistry of (η6-2,6-((CH3)3Si)2-
C5H3N)Cr(CO)3 proved to be interesting. Even though
experiments were conducted in a CO-doped matrix, no
evidence for the formation of pentacarbonyl or hexa-
carbonyl species was obtained. In this respect the
matrix results are comparable to those obtained for this
compound in room-temperature solution. The products
formed upon short-wavelength photolysis are sensitive
to photoreversal with white light. Hence, the failure
to observe any photochemistry following irradiation at
λexc ) 460 nm in this medium may simply reflect an
Discu ssion
The formation of (η1-C5H5N)Cr(CO)5 and ultimately
Cr(CO)6 in the presence of CO demonstrates that CO
loss is not the only photochemical route accessible to
(η6-C5H5N)Cr(CO)3. The efficiency of formation of the
pentacarbonyl species in solution depends on the nature
of the solvent and also the substituents on the 2,6-
positions. Consequently (η1-2,6,-X2C5H3N)Cr(CO)5 is
formed most efficiently in methanol for X ) H. These
studies indicate that formation of (η1-2,6-X2C5H3N)Cr-
(CO)5 requires the initial formation of an intermediate
species, whose stability is influenced by the Lewis base
character of the solvent and/or the size of the substit-
uents R to the nitrogen atom. The matrix isolation
experiments allow us to propose a model for this
intermediate species.

3686 Organometallics, Vol. 15, No. 17, 1996
Breheny et al.
Sch em e 2a
F igu r e 9. The UV/vis spectrum of (η6-C5H5N)Cr(CO)3 in
a CH4 matrix at 12 K (s) and the same matrix following
irradiation for 60 s with λexc ) 308 nm (- - -). These spectra
are not corrected for matrix scatter.
efficient photoreversal of the initial product. Figure 9
confirms that photolysis of (η6-C5H5N)Cr(CO)3 (λexc
)
308 nm) in a methane matrix, producing both the CO-
loss and ring-slip products, exhibits an unstructured
absorption well into the visible region of the spectrum.
The photochemistry of (η6-2,6-((CH3)3Si)2C5H3N)Cr(CO)3
in an N2 matrix is dominated by a low-efficiency CO loss
even following long-wavelength irradiation. Some evi-
dence for the formation of an η1-species is observed, but
no νN-N bands associated with this species could be
detected. Therefore it is not possible to confirm if this
ring-slip species interacts with the N2 matrix to any
significant degree.
The lack of any evidence for CO or N2 addition to the
(η1-2,6-((CH3)3Si)2C5H3N)Cr(CO)3 species can be ex-
plained in terms of the steric crowding in the coordina-
tion sphere of the chromium atom. The structural
parameters obtained from the X-ray data provide an
estimate of the steric requirements of the 2,6-((CH3)3-
Si)2C5H3N ligand. Using the “Unlock Model” facility in
Schakal-92,18 the chromium tricarbonyl unit was located
2.0 Å from the nitrogen atom. This is approximately
where it would be following an η6 to η1 haptotropic shift.
Analysis of the close contacts for this orientation
indicated that there is sufficient freedom to permit
coordination of the pyridine nitrogen atom to the
chromium tricarbonyl unit but not to higher order
carbonyl units or additional solvent or matrix molecules.
This explains why the presence of the trimethylsilyl
groups in the position R to the coordinating nitrogen
prevents the formation of the metal tetracarbonyl or
metal pentacarbonyl species.
a
CO ligands are represented as s for clarity. Reaction con-
ditions: (i) λexc < 360 nm, Sol ) cyclohexane (reaction 1); (ii)
λexc > 308 nm; (iii) τ1/2 < 40 ns in cyclohexane (reaction 2); (iv)
Sol ) cyclohexane or methanol; (v) CO-saturated cyclohexane
or methonol; (vi) λexc ) 460 nm, N2 matrix; (vii) λexc ) 460 nm,
10% CO in CH4 matrix; (viii) λexc ) 250 nm, 10% CO in CH4
matrix; (ix) λexc > 530 nm, 10% CO in CH4 matrix; (x) hν, 10%
CO in CH4 matrix.
takeup of CO from solution resulting in the relatively
low yield of (η1-C5H5N)Cr(CO)5.
The UV/vis experiments support this explanation.
Monitoring the recovery of the depleted parent absorp-
tion at 340 nm, following pulse photolysis at 355 nm,
demonstrated that this process follows a biphasic time
profile (cf. inset in Figure 6). The faster of these two
processes is completed within 40 ns, which is shorter
than the rise-time of the TRIR equipment. We have
assigned this portion of the recovery to the η1 to η6
haptotropic reversal (reaction 4). Thus, although the
η6 to η1 ring-slippage process accounts for a significant
proportion of the light absorbed by the complex, the
ring-slippage product itself does not show up in the
TRIR experiment. Moreover the yield of (η1-C5H5N)-
Cr(CO)5 as a permanent photoproduct remains rela-
tively low unless the η1-species is trapped by a donor
solvent such as methanol (Scheme 2). This accounts for
the higher relative quantum yield for the conversion of
(η6-C5H5N)Cr(CO)3 into (η1-C5H5N)Cr(CO)5 in CO-
The TRIR experiments indicated that CO loss is a
primary photoinduced process in cyclohexane solution
at room temperature, but there is some indirect evidence
for an η6 to η1 haptotropic shift of the pyridine ligand,
i.e. the observation of (η1-C5H5N)Cr(CO)5, albeit in
minor yield, following parent recovery. In the matrix
experiments, η6 to η1 ring-slippage is clearly predomi-
nant over the loss of CO, which at first sight contrasts
with the time-resolved IR spectroscopic observation in
solution. However this apparent discrepancy is readily
explained in terms of a very fast η1 to η6 haptrotropic
reversal to the parent (reaction 4), which competes with
saturated CH3OH compared to C6H12
photochemical and subsequent thermal reactions of
these compounds are summarized in Scheme 2.
. The overall
Con clu d in g Rem a r k s. It is apparent from this
work that the application of a range of techniques is
essential to fully appreciate the complexity of even a
relatively simple system such as this. For instance the
use of one technique, TRIR, would lead to an overly
simplistic interpretation of the photochemical routes
available. Furthermore these results demonstrate for
the first time the importance of arene ring-slip processes
in the photochemistry of (η6-arene)Cr(CO)3 compounds.
The photochemistry of other (η6-arene)Cr(CO)3 systems

(η6-2,6-X2C5H3N)Cr(CO)3
Organometallics, Vol. 15, No. 17, 1996 3687
containing potential “traps” for ring-slip intermediates
is currently under investigation.
M. Viotte, G. Klihm, and D. Parirnus are thanked for
their skilled technical assistance.
Ack n ow led gm en t. The authors acknowledge the
support of the European Union (HCM Programme,
Contract No. CHRX-CT93-0152), The Irish American
Partnership, and Forbairt (Ireland) (formerly Eolas) for
support for G.R., M.T.P., and C.J .B., respectively.
Professor R. N. Perutz and Professor M. Poliakoff are
thanked for their helpful suggestions and comments. Dr.
Su p p or tin g In for m a tion Ava ila ble: Complete listings
of bond lengths and angles, fractional coordinates, and thermal
parameters and ORTEP diagrams for (η6-2,6-(CH3)2C5H3N)-
Cr(CO)3 and (η6-2,6-((CH3)3)Si)2C5H3N)Cr(CO)3 (12 pages).
Ordering information is given on any current masthead page.
OM960090+