Solid-state photochemical diag-lat isomerization of (η5-C5H4R)Re(CO)(L)X2 on the surface of silica gel

Article abstract of DOI:10.1021/om960726x

Room-temperature visible illumination (200 W light source) of diag-(η⁵-C₅H₄R)Re(CO)-(L)X₂ (R = H, alkyl; L = CO, P(OR)₃; X = Br, I) adsorbed on silica gel results in isomerization of the complexes into the corresponding lateral isomers in good yield (typically >80%). For L = CO, the isomerization reactions occurred in the direction reverse to that found in solution in which lat-to-diag isomerization was observed. It was found that the ratio of diagonal to lateral isomers established on the surface of silica gel was influenced by both X and L, viz. (η⁵-C₅H₄Me)Re(CO)[P(OR)₃]Br ₂ > (η⁵-C₅H₄Me)Re(CO)₂Br₂ > (η⁵-C₅H₄Me)Re(CO)₂I₂. Chemical removal of surface hydroxyl groups with Me₂SiCl₂ revealed that only a monolayer of the rhenium complex chemically adsorbed on the surface of the silica gel underwent isomerization. A mechanism to rationalize the role of the silica surface hydroxyl groups is described.

Full text of DOI:10.1021/om960726x

Organometallics 1997, 16, 591-596
591
Solid -Sta te P h otoch em ica l d ia g-la t Isom er iza tion of
(η5-C5H4R)Re(CO)(L)X2 on th e Su r fa ce of Silica Gel
Lin Cheng and Neil J . Coville*
Centre for Applied Chemistry and Chemical Technology, Department of Chemistry,
University of The Witwatersrand, Private Bag 3, WITS 2050, J ohannesburg, South Africa
Received August 22, 1996X
Room-temperature visible illumination (200 W light source) of diag-(η5-C5H4R)Re(CO)-
(L)X2 (R ) H, alkyl; L ) CO, P(OR)3; X ) Br, I) adsorbed on silica gel results in isomerization
of the complexes into the corresponding lateral isomers in good yield (typically >80%). For
L ) CO, the isomerization reactions occurred in the direction reverse to that found in solution
in which lat-to-diag isomerization was observed. It was found that the ratio of diagonal to
lateral isomers established on the surface of silica gel was influenced by both X and L, viz.
(η5-C5H4Me)Re(CO)[P(OR)3]Br2 > (η5-C5H4Me)Re(CO)2Br2 > (η5-C5H4Me)Re(CO)2I2. Chemical
removal of surface hydroxyl groups with Me2SiCl2 revealed that only a monolayer of the
rhenium complex chemically adsorbed on the surface of the silica gel underwent isomer-
ization. A mechanism to rationalize the role of the silica surface hydroxyl groups is described.
In tr od u ction
to be little developed in the synthesis of organometallic
compounds.
Compared to the development of synthetic chemistry
in the solution state, synthetic chemistry employing
solid-state reactivity has been less developed.1 Indeed,
a survey of the organometallic chemistry literature
reveals that little exploration of the use of the solid
phase in the synthesis of new molecules has been
reported. From what has been reported, the reactions
can be divided into three types.2
Finally, the surfaces of solids can be used as reagents
for carrying out reactions. Thus, reagents can be added
to a surface and the surface-reagent interaction used
to induce reactivity such that the reagent is converted
into a product which can be extracted from the surface
in a later step. In these reactions the product is formed
in a solid-state reaction and is to be differentiated from
the use of the heterogeneous complexes to bring about
reactions in the solution phase.9,10 Studies of surface-
organometallic complexes have been directed at the
production of well-controlled heterogeneous materials,
e.g. Mo(CO)n on a range of surfaces,11 for use in gas-
and liquid-phase catalytic reactions.12 The use of the
surface as a reagent for the synthesis of new organo-
metallic complexes has been less extensively
exploited.7a,13
Firstly, solid-state reactions can occur within a solid.
For example, a range of isomerization reactions have
been carried out both thermally and photochemically
in the solid-state. Reactions include nitrito to nitro
isomerization3 and cis to trans isomerization of chro-
mium,4 ruthenium,5 platinum,6 and rhenium7 com-
plexes.
Second, reactions can occur between solid particles.
Here the intimacy and method of interaction (heating,
shaking, grinding, etc.) become critical and particle size
typically plays a dominant role in determining the
reaction rate. This methodology has been exploited in
organic chemistry1 and materials science8 but appears
In this publication we wish to report on a novel
example of the third type of solid-state reaction. We7a
and others14 have reported that lat-(η5-C5R4R′)Re-
(CO)2X2 isomerizes to diag-(η5-C5R4R′)Re(CO)2X2 (R )
H, Me; R′ ) H, alkyl group, X ) halogen) in the solution
phase. Herein we report that surface-supported (η5-
X Abstract published in Advance ACS Abstracts, J anuary 1, 1997.
(1) (a) Toda, F. Acc. Chem. Res. 1995, 28, 480. (b) West, A. Solid
State Chemistry and its Applications; Wiley: Chichester, U.K., 1994.
(2) Reactions in matrices, e.g. matrix isomerization reactions, have
been excluded from this discussion since the matrix molecules/atoms
can be regarded as frozen “solvent” molecules.
(3) (a) Boldyreva, E. V. Mol. Cryst. Liq. Cryst. 1994, 242, 17. (b)
Grenthe, I.; Nordin, E. Inorg. Chem. 1979, 18, 1869. (c) Dulepov, V.
E.; Boldyreva, E. V. React. Kinet. Catal. Lett. 1994, 53, 289. (d)
Masciocchi, N.; Kolyshev, A.; Dulepov, V.; Boldyreva, E.; Sironi, A.
Inorg. Chem. 1994, 33, 2579.
(4) Sedova, G. N.; Demchenko, L N. Zh. Neorg. Khim. 1981, 26, 435.
(5) (a) Katsuki, K.; Ooyama, Y.; Okamoto, M.; Yamamoto, Y. Inorg.
Chim. Acta 1994, 217, 181. (b) Krassowski, W. D.; Lemay, H. E., J r.;
Nelson, J . H. Inorg. Chem. 1988, 27, 4307.
(6) (a) Macdonald, F. M.; Sadler, P. J . Polyhedron 1991, 10, 1443.
(b) Louw, W. J . Inorg. Chem. 1977, 16, 2147. (c) Favez, R.; Roulet, R.;
Pinkerton, A. A.; Schwarzenbach, D. Inorg. Chem. 1980, 19, 1356.
(7) (a) Cheng, L.; Coville, N. J . Organometallics 1996, 15, 867. (b)
Palmer, B. J .; Becalska, A.; Hader, R.; Hill, R. H. Polyhedron 1991,
10, 877.
(9) Gates, B. C. J . Mol. Catal. 1994, 86, 95.
(10) (a) Mayo, P.; Nakamura, A.; Tsang, P. W. K.; Wong, K. S. J .
Am. Chem. Soc. 1982, 104, 6824. (b) Mayo, P.; Ramnath, N. Can. J .
Chem. 1986, 64, 1293.
(11) (a) Reddy, K. P.; Brown, T. L. J . Am. Chem. Soc. 1995, 117,
2845. (b) Ang, H. G.; Chan, K. S.; Chuah, G. K.; J aenicke, S.; Neo, S.
K. J . Chem. Soc., Dalton Trans. 1995, 3753. (c) Nakamura, R.; Pioch,
D.; Bowman, G.; Burwell, R. L. J . Catal. 1985, 93, 388, 399. (d)
Brenner, A.; Hucul, D. A.; Hardwick, S. J . Inorg. Chem. 1979, 18, 1487.
(e) Gonzales, P. N.; Villa Garcia, M. A.; Brenner, A. J . Catal. 1989,
118, 360.
(12) Chem. Rev. 1995, 95. Issue 3 of this volume contains numerous
articles on solid-state reactions.
(13) (a) Roberto, D.; Cariati, E.; Ugo, R.; Psaro, R. Inorg. Chem. 1996,
35, 2311. (b) Roberto, D.; Cariati, E.; Psaro, R.; Ugo, R. J . Organomet.
Chem. 1995, 488, 109. (c) Theolier, A.; Smith, A. K.; Leconte, M.;
Basset, J . M.; Zanderighi, G. M.; Psaro, R.; Ugo, R. J . Organomet.
Chem. 1980, 191, 415.
(14) (a) King, R. B.; Reimann, R. H. Inorg. Chem. 1976, 12, 183. (b)
Einstein, F. W. B.; Klahn-Oliva, A. H.; Sutton, D.; Tyers, K. G.
Organometallics 1986, 5, 53. (c) Hill, R. H.; Palmer, B. J . Organo-
metallics 1989, 8, 1651.
(8) (a) Moore, J . J .; Feng, H. J . Prog. Mater. Sci. 1995, 39, 243. (b)
Thadhani, N. N. Prog. Mater. Sci. 1993, 37, 117.
S0276-7333(96)00726-1 CCC: $14.00 © 1997 American Chemical Society

592 Organometallics, Vol. 16, No. 4, 1997
Cheng and Coville
C5H4Me)Re(CO)[P(OPh)3]Br2 (92 mg, 58% yield) from a brown
band, respectively.
C5H4R)Re(CO)(L)X2 (L ) CO, phosphite) undergoes a
solid-state photochemical isomerization reaction in the
direction reverse to that observed in the solution phase,
a reaction which indicates the influence of the surface
as a ligand in the overall reaction. This reaction
provides for a facile synthesis of lat-(η5-C5H4R)Re(CO)-
(L)X2.
Syn th eses of d ia g- a n d la t-(η5-C5H4Me)Re(CO)2I2. To
a solution of (η5-C5H4Me)Re(CO)3 (730 mg, 2.09 mmol) in 50
mL of dimethyl sulfoxide, a solution of I2 (530 mg, 2.09 mmol)
in 20 mL of dimethyl sulfoxide was added dropwise. After the
reaction mixture was stirred for an additional 1.0 h at room
temperature, the reaction was quenched by adding 100 mL of
water at 25 °C. The resulting brown precipitate, which
contained unreacted (η5-C5H4Me)Re(CO)3 and both isomers of
(η5-C5H4Me)Re(CO)2I2, was filtered and dried at 25 °C (0.1
mmHg). This solid was dissolved in dichloromethane and
chromatographed on a silica gel column prepared in hexane.
Successful elution with hexane, hexane/dichloromethane (1:
1), and dichloromethane gave unreacted (η5-C5H4Me)Re(CO)3
(350 mg, 48% recovery), diag-(η5-C5H4Me)Re(CO)2I2 (245 mg,
20% conversion, 39% yield) from a red band, and lat-(η5-C5H4-
Me)Re(CO)2I2 (106 mg, 9% conversion, 17% yield) from a brown
band, respectively.
Exp er im en ta l Section
The diagonal and lateral (η5-C5H4R)Re(CO)(L)Br2 (R ) Me,
Et, tBu, Si(CH3)3, L ) CO; R ) H, L ) CO, P(OMe)3, P(OPh)3)
complexes were prepared by the literature methods.7a,14a
Kieselgel PF254 (Merck) silica gel for preparative thin-layer
chromatography was used as the support for most of the
reactions. Dichlorodimethylsilane (98%) was obtained from
Merck. Solvents were dried by conventional methods, distilled
under nitrogen, and used immediately. Irradiation was car-
ried out with two 200 W incandescent lamps. The BET surface
area measurements of the inorganic solids were performed
using the N2 adsorption method with classical gas-phase
surface area equipment constructed in our department.15
Melting points were recorded on a Kofler hot stage melting
point apparatus. Infrared spectra were measured on a Midac
FTIR spectrometer, usually in KBr cells (solutions). NMR
spectra were measured on a Bruker AC200 spectrometer
operating at 200 MHz. Microanalyses were carried out at the
CSIR, Pretoria, South Africa.
Syn th eses of d ia g- a n d la t-(η5-C5H4iP r )Re(CO)2Br 2. A
solution of Br2 (136 mg, 0.85 mmol) in 2 mL of trifluoroacetic
i
acid was added dropwise at 25 °C to a solution of (η5-C5H4 -
Pr)Re(CO)3 (304 mg, 0.81 mmol) in 3 mL of trifluoroacetic acid.
After the reaction mixture was stirred for an additional 1.0 h
at room temperature, the reaction was quenched by pouring
the mixture into 100 mL of water at 25 °C. The resulting
i
brown precipitate, which contained unreacted (η5-C5H4 Pr)Re-
i
(CO)3 and both isomers of (η5-C5H4 Pr)Re(CO)2Br2, was filtered
and dried at 25 °C (0.1 mmHg). This solid was dissolved in
dichloromethane and chromatographed on a silica gel column
prepared in hexane. Successive elution with hexane, hexane/
dichloromethane (1:1), and dichloromethane gave unreacted
Ir r a d ia tion Exp er im en ts. Experiments were performed
as follows. Pyrex cylinders (22 mm × 90 mm) were used as
reactors and were loaded with diag- or lat-(η5-C5H4R)Re-
(CO)(L)X2 (0.038 mmol) dissolved in CH2Cl2 (3 mL). Silica gel
(300 mg) was added to the reactor, the reagents were well-
mixed, and the CH2Cl2 was removed under vacuum. The
cylinders were sealed under nitrogen (atmospheric pressure)
and rotated with a motor (8 rpm) with the cylinder axis in a
horizontal position. Two 200 W incandescent lamps (λ > 370
nm) were located 30 cm from the cylinder, and the whole
experimental apparatus was set up in a ventilated fume hood;
this ensured minimum local heating from the lamp at the
sample.16 A thermometer placed next to the sample cylinder
indicated that the temperature of the reaction vessel never
exceeded 25 °C. A control experiment was performed in the
dark by completely wrapping a cylinder, containing the silica-
supported diag-(η5-C5H4Me)Re(CO)2Br2, in aluminum foil.
After completion of the irradiation, the products were extracted
from the surface of the silica gel with CH2Cl2, separated on a
silica gel column (CH2Cl2/hexane), weighed, and identified by
IR and NMR spectroscopy.
The photochemical isomerization reaction of the pure amor-
phous solid diag- or lat-(η5-C5H4Me)Re(CO)2Br2 was carried
out in the same manner as described above. The irradiation
of diag-(η5-C5H4Me)Re(CO)2Br2 on other inorganic solids was
also performed in the same way as described above, except
that two or more Pyrex cylinders were sometimes used in order
to irradiate the large quantities of material required in the
study.
Syn th eses of dia g- an d la t-(η5-C5H4Me)Re(CO)[P (OP h )3]-
Br 2. A solution of diag-(η5-C5H4Me)Re(CO)2Br2 (100 mg, 0.208
mmol) and P(OPh)3 (212 mg, 0.683 mmol) in toluene (25 mL)
was refluxed under N2 for 20 h. IR spectra indicated that all
the starting material had reacted. After solvent removal under
reduced pressure, the red residue was dissolved in dichloro-
methane and chromatographed on a silica gel column prepared
in hexane. Successive elution with dichloromethane/hexane
(1:1) and dichloromethane gave diag-(η5-C5H4Me)Re(CO)-
[P(OPh)3]Br2 (48 mg, 30% yield) from a red band and lat-(η5-
i
i
(η5-C5H4 Pr)Re(CO)3 (171 mg, 56% recovery), diag-(η5-C5H4 -
Pr)Re(CO)2Br2 (66 mg, 16% conversion, 37% yield) from a red
i
band, and lat-(η5-C5H4 Pr)Re(CO)2Br2 (60 mg, 12% conversion,
33% yield) from a brown band, respectively.
Syn th eses of d ia g- a n d la t-(η5-C5H4Me)Re(CO)2Br I. (i)
A solution of diag-(η5-C5H4Me)Re(CO)2Br2 (57 mg, 0.118 mmol)
and diag-(η5-C5H4Me)Re(CO)2I2 (68 mg, 0.118 mmol) in ben-
zene (25 mL) was refluxed under nitrogen for 20 h. After
solvent removal with a rotary evaporator, the red solid was
dissolved in dichloromethane and chromatographed on a silica
gel column (1.5 cm × 60 cm) prepared in hexane. Careful
elution with hexane/dichloromethane (4:1) and hexane/di-
chloromethane (1:1) gave diag-(η5-C5H4Me)Re(CO)2I2 (35.8 mg,
53% recovery), diag-(η5-C5H4Me)Re(CO)2BrI (48.5 mg, 39%
conversion, 67% yield), diag-(η5-C5H4Me)Re(CO)2Br2 (18.0 mg,
32% recovery), lat-(η5-C5H4Me)Re(CO)2I2 (2.4 mg, 1.8% conver-
sion, 3.0% yield), lat-(η5-C5H4Me)Re(CO)2BrI (7.1 mg, 5.7%
conversion, 9.8% yield), and lat-(η5-C5H4Me)Re(CO)2Br2 (8.9
mg, 7.8% conversion, 13% yield).
(ii) Silica gel (900 mg) was added to a solution of NaI (17.1
mg, 0.114 mmol) in water (2 mL). The slurry was well stirred
and dried in an oven at 120 °C. This pretreated silica gel was
then added to a solution of diag-(η5-C5H4Me)Re(CO)2Br2 (55
mg, 0.114 mmol) in dichloromethane (3 mL), the reagents were
well mixed, and the CH2Cl2 was removed under vacuum. The
reaction mixture was placed into three Pyrex cylinders (22 mm
× 90 mm) and sealed. After irradiation with two 200 W
incandescent lamps at room temperature for 48 h, the reaction
products were extracted from silica gel with dichloromethane
and chromatographed on a silica gel column prepared in
hexane. Elution with hexane/dichloromethane (1:1) gave first
a mixture of diag-(η5-C5H4Me)Re(CO)2I2, diag-(η5-C5H4Me)Re-
(CO)2BrI, and diag-(η5-C5H4Me)Re(CO)2Br2 (14.7 mg, 24%
yield) and then lat-(η5-C5H4Me)Re(CO)2I2 (7.5 mg, 11% yield),
lat-(η5-C5H4Me)Re(CO)2BrI (12.0 mg, 20% yield), and lat-(η5-
C5H4Me)Re(CO)2Br2 (12.0 mg, 22% yield), respectively.
(15) Duvenhage, D. J . Ph.D. Thesis, University of the Witwaters-
rand, J ohannesburg, South Africa, 1993.
(16) Abdel-Malik, M. M.; De Mayo, P. Can. J . Chem. 1984, 62, 1275.
All new complexes were characterized by elemental analysis
and by IR and NMR spectroscopy (Table 1).

Isomerization of (η5-C5H4R)Re(CO)(L)X2
Organometallics, Vol. 16, No. 4, 1997 593
Ta ble 1. Sp ectr oscop ic a n d An a lytica l Da ta for (η5-C5H4R)Re(CO)(L)X2 Com p lexes
1H NMRb (ppm)
anal (%)
C
compd
mp (°C)
νcoa (cm-1
)
Cp ring
4.53 (t, 2H), 4.65 (t, 2H)c
R
L
H
diag-(η5-C5H4Me)Re(CO)[P(OPh)3]Br2 164-166 1988
1.64 (s, 3H)c
6.88-7.56
calcd 39.33 2.90
(m, 15H)c found 39.39 2.68
calcd 23.59 2.18
found 23.35 1.88
calcd 16.71 1.23
found 16.54 1.04
calcd 18.19 1.34
found 18.08 1.15
diag-(η5-C5H4iPr)Re(CO)2Br2
diag-(η5-C5H4Me)Re(CO)2I2
diag-(η5-C5H4Me)Re(CO)2BrI
lat-(η5-C5H4Me)Re(CO)[P(OPh)3]Br2
lat-(η5-C5H4iPr)Re(CO)2Br2
lat-(η5-C5H4Me)Re(CO)2I2
112-114 1997, 2062 5.38 (t, 2H) 5.71 (t, 2H)
142 -144 1982, 2048 5.29 (t, 2H), 5.71 (t, 2H)
125-127 1989, 2054 5.52 (t, 2H), 5.70 (t, 2H)
1.29 (d, 6H),
2.83 (m, 1H)
2.44 (s, 3H)
2.36 (s, 3H)
1.73 (s, 3H)c
183-185 1959
4.62 (d, 1H), 4.72 (d, 1H),
4.77 (d, 1H), 5.07 (d, 1H)c
6.75-7.55
calcd 39.33 2.90
(m, 15H)c found 39.38 2.69
calcd 23.59 2.18
127-129 1978, 2054 5.88 (d, 2H), 5.92 (d, 2H)
127-129 1972, 2040 5.78 (m, 4H)
1.28 (d, 6H),
2.88 (m, 1H)
2.58 (s, 3H)
found 23.36 1.92
calcd 16.71 1.23
found 16.54 1.04
calcd 18.19 1.34
lat-(η5-C5H4Me)Re(CO)2BrI
135-137 1974, 2043 5.76 (m, 2H), 5.81 (m, 1H),
2.39 (s, 3H)
5.88 (m, 1H)
found 17.95 1.14
a
b
Recorded in CH2Cl2. Recorded in CDCl3, relative to TMS: s, singlet; d, doublet; t, triplet; m, multiplet. c Recorded in C6D6.
F igu r e 2. Solid-state photochemical isomerization reac-
tion of diag-(η5-C5H4Me)Re(CO)2Br2 as a function of silica
gel surface area.
F igu r e 1. Solid-state photochemical isomerization reac-
tion of diag-(η5-C5H4Me)Re(CO)2Br2 after different times.
and suggests that only (η5-C5H4Me)Re(CO)2Br2 in physi-
cal contact with the SiO2 will undergo the photoisomer-
ization reaction. A plot of SiO2 surface area against the
percentage of lateral isomer formed for diag-(η5-
Resu lts a n d Discu ssion
diag-(η5-C5H4Me)Re(CO)2Br2 was loaded onto silica
as described in the Experimental Section and irradiated
with two 200 W lamps. A series of experiments were
performed in which the reaction was terminated after
different reaction times (1-72 h; Figure 1). The prod-
ucts were extracted from the silica (>85%), separated
on a chromatographic column, weighed, and analyzed
by IR and NMR spectroscopy. The data indicated that
after 24 h a photostationary state of the isomers had
formed with a [lat]/[diag] ratio of ∼7. Thus, an isomer-
ization reaction takes place on irradiation of silica in
the solid-state which favors the lateral isomer. This
ratio is different from that achieved in the thermal solid-
state isomerization reaction.7a Further, Hill and co-
workers have reported that photoisomerization of (η5-
C5Me5)Re(CO)2Br2 on Si(111) at 77 K gave a reaction
in which the diagonal isomer was formed from the
lateral isomer.7b We have further observed that ir-
radiation of either pure amorphous solid diag- or lat-
(η5-C5H4Me)Re(CO)2Br2 in the absence of silica resulted
in no photoisomerization activity.
i
C5H4R)Re(CO)2Br2 (R ) Me, Pr) gave similar results,
suggesting that the substituents on the cyclopentadienyl
ring ligand had little effect on the solid-state photo-
chemical isomerization reaction. This result was fur-
ther confirmed by the similar isomerization yields
obtained for (η5-C5H4R)Re(CO)2Br2 (R ) H, Me, Et, iPr,
tBu, Si(CH3)3) at a fixed complex loading (Table 2). All
the above reactions were terminated after 48 h; by this
time a photostationary state had been established, as
revealed by studying the isomerization reactions of both
the lateral and diagonal isomers. The photostationary
ratio [lat]/[diag] increased as follows: (η5-C5H4Me)Re-
(CO)(L)Br2 > (η5-C5H4Me)Re(CO)2Br2 > (η5-C5H4Me)-
Re(CO)2I2 (Table 2). As can be seen from Table 2, most
of the product was extractable from silica (generally 90-
95%) and suggests that the method provides a facile
procedure for isomerization of the rhenium complexes.
In contrast, the solution isomerization reaction for (η5-
i
t
C5H4R)Re(CO)2Br2 (R ) H, Me, Et, Pr, Bu, Si(CH3)3)
has been reported7a and indicated that the diagonal
isomer was the dominant isomer formed in the equilib-
rium reaction. We have now also established that diag-
(η5-C5H4Me)Re(CO)[P(OPh)3]Br2 isomerizes completely
into the lateral isomer in CHCl3 after several days at
A series of experiments were performed using a
varying diag-(η5-C5H4Me)Re(CO)2Br2/SiO2 ratio achieved
by changing the amount of SiO2. BET surface areas for
the loaded and unloaded SiO2 samples were determined,
and a plot of the SiO2 surface area against the percent-
age of lateral isomer formed (after 48 h reaction) is
shown in Figure 2. Clearly a photostationary state is
established when the SiO2 surface area is >29 m2 for
0.038 mmol of diag-(η5-C5H4Me)Re(CO)2Br2. This cor-
(17) The “footprint”18 of diag-(η5-C5H4Me)Re(CO)2Br2 was calculated
from its X-ray structure19 as 22.5 Å2. Thus, 0.038 mmol of diag-(η5-
C5H4Me)Re(CO)2Br2 will cover an area of 5.2 m2.
(18) Matsuishi, T.; Shimada, T.; Morihara, K. Bull. Chem. Soc. J pn.
1994, 67, 748.
(19) Boeyens, J . C. A.; Cheng, L.; Coville, N. J .; Levendis, D. C.;
McIntosh, K. S. Manuscript in preparation.
17
responds to less than a monolayer coverage of the SiO2

594 Organometallics, Vol. 16, No. 4, 1997
Cheng and Coville
Ta ble 2. Solid -Sta te P h otoch em ica l d ia g-la t Isom er iza tion of (η5-C5H4R) Re(CO)(L)X2 on th e Su r fa ce of
Silica Gela
entry no
complex
diag (%)b
lat (%)b
total recovery (%)c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
diag-(η5-C5H5)Re(CO)2Br2
23.4
15.7
12.2
10.1
10.3
18.6
71.1
0.0
76.6
84.3
87.8
89.9
89.7
81.4
28.9
100.0
100.0
100.0
88.3
89.0
84.1
86.3
86.7
88.5
100.0
86.5
94.0
87.2
76.7d
87.9
89.3
93.2
79.4
84.2
82.4
91.6
89.1
87.2
72.0
98.5
89.3
90.3
diag-(η5-C5H4Me)Re(CO)2Br2
diag-(η5-C5H4Et)Re(CO)2Br2
diag-(η5-C5H4iPr)Re(CO)2Br2
diag-(η5-C5H4tBu)Re(CO)2Br2
diag-[η5-C5H4Si(CH3)3]Re(CO)2Br2
diag-(η5-C5H4Me)Re(CO)2I2
diag-(η5-C5H5)Re(CO)[P(OMe)3]Br2
diag-(η5-C5H5)Re(CO)[P(OPh)3]Br2
diag-(η5-C5H4Me)Re(CO)[P(OPh)3]Br2
lat-(η5-C5H5)Re(CO)2Br2
0.0
0.0
11.7
11.0
15.9
13.7
13.3
11.5
0.0
lat-(η5-C5H4Me)Re(CO)2Br2
lat-(η5-C5H4Et)Re(CO)2Br2
lat-(η5-C5H4iPr)Re(CO)2Br2
lat-(η5-C5H4tBu)Re(CO)2Br2
lat-[η5-C5H4Si(CH3)3]Re(CO)2Br2
lat-(η5-C5H4Me)Re(CO)[P(OPh)3]Br2
a
b
Conditions: (η5-C5H4R)Re(CO)(L)X2, 0.038 mmol; silica gel, 300 mg; reaction time, 48 h; room temperature. Isolated yields. c Based
d
on starting materials used. Silica gel 200 mg.
Ta ble 3. Solid -Sta te P h otoch em ica l d ia g-la t
Isom er iza tion of (η5-C5H4Me) Re(CO)2Br 2 on
Va r iou s In or ga n ic Solid sa
surface
area
total
lat recovery
entry
no.
amt
diag
(%)
inorganic solid
(m2/g) (mg)
(%)
(%)
1
2
3
4
5
6
7
8
9
ZnO
ZnO
CaSO4
CaSO4
Al2O3 (neutral)
MgO
29.7
29.7 2400
28.1 300
28.1 2400
300
300
99.4
99.0
94.3
92.0
15.8 84.2
83.6 16.4
66.9 33.1
10.0 90.0
56.3 43.7
0.6
1.0
5.7
8.0
90.2
29.0
95.6
68.3
10.4
33.3
69.4
6.0
95.1
100.0
80.9
94.0
62.5
108
108
271
300
300
750
300
TiO2
F igu r e 3. Solid-state photochemical isomerization reac-
tion of diag-(η5-C5H4Me)Re(CO)2Br2 on silica gel containing
different amounts of surface hydroxyl groups.
TiO2
zeoliteb
Nil
10
11
12
100.0
9.5 90.5
15.7 84.3
0.0
amorphous SiO2 258
Silica gel 240
300
300
Rea ction Mech a n ism . The solid-state photochemi-
cal isomerization reactions of (η5-C5H4Me)Re(CO)2Br2
were carried out on a series of silica gels containing
different amounts of surface hydroxyl groups (obtained
by pretreatment of the surface with different amounts
of dichlorodimethylsilane). The results showed that the
yield of isomerized product increased with the amount
of surface hydroxyl groups (Figure 3). Clearly, the
surface hydroxyl groups play a crucial role in the solid-
state photochemical isomerization reaction.
The solid-state photochemical isomerization reactions
of an equimolar mixture of diag-(η5-C5H5)Re(CO)-
[P(OMe)3]Br2 and diag-(η5-C5H4Me)Re(CO)[P(OPh)3]Br2
were carried out, and no intermolecular exchange
between P(OMe)3 and P(OPh)3 ligands was found (Table
4). The intermolecular exchange between CO and
P(OPh)3 ligands was also not observed for the solid-state
isomerization reaction of diag-(η5-C5H4Me)Re(CO)2Br2
in the presence of excess P(OPh)3 ligand (Table 4). This
suggests that the solid-state photochemical diag-lat
isomerization reactions of (η5-C5H4R)Re(CO)(L)X2 do not
proceed via intermolecular exchange of CO, L, or cyclo-
pentadienyl ligands.
a
Conditions: diag-(η5-C5H4Me)Re(CO)2Br2, 0.038 mmol; reac-
b
tion time, 48 h; room temperature. HZSM-5.
room temperature. The isomerization reaction was
much slower (>2 weeks) when benzene was used as
solvent. In contrast, lat-(η5-C5H4R)Re(CO)2I2 (R ) H,
Me) converted (>90%) into the diagonal isomer after
several hours at room temperature in CHCl3; the diag
to lat isomerization did not occur under these conditions.
The diag-lat isomerization reaction of (η5-C5H4Me)-
Re(CO)2Br2 was also performed in the dark (no reaction)
and by irradiation with sunlight and UV light. The
latter reactions also yielded photoisomerized products
with conversions little different from that obtained with
the 200 W lamp. This suggests that the solid-state
isomerization reactions on the surface of silica gel can
be achieved by a range of wavelengths.
Other inorganic solids were also tested as potential
supports for the photoisomerization reaction of (η5-C5H4-
Me)Re(CO)2Br2, and the data are shown in Table 3. In
general only the silica supports gave good conversions
to the lateral isomer. ZnO, MgO, Al2O3, and CaSO4
were found to be poor supports; they either caused
complex decomposition or resulted in low conversions.
Increasing the product by increasing the amount (sur-
face area) of the supports which gave poor results was
also found to be unsuccessful. It should be pointed out
that the quality of the surface coating, which depends
on the preparation method used, could also perturb the
efficiency of the photochemical isomerization reaction.
The solid-state photochemical isomerization reaction
of an equimolar mixture of diag-(η5-C5H4Me)Re(CO)2Br2
and diag-(η5-C5H4Me)Re(CO)2I2 was carried out, and a
mixture of dibromide and diiodide complexes, as well
as 48% (η5-C5H4Me)Re(CO)2BrI (Table 1), was formed
in the reaction (Table 4). Facile exchange of Br and I
(ions, free radicals) is possible on photolysis. This was
also indicated by the formation of molecular iodine on

Isomerization of (η5-C5H4R)Re(CO)(L)X2
Organometallics, Vol. 16, No. 4, 1997 595
Ta ble 4. Solid -Sta te P h otoch em ica l Isom er iza tion of Mixed Rh en iu m Com p lexesa
entry no
1
reacn mixture
diag (%)
lat (%)
total recovery (%)
82.2
diag-(η5-C5H4Me)Re(CO)2Br2
diag-(η5-C5H4Me)Re(CO)2I2
38.6b
(A) 19.6c
(B) 48.0c
(C) 32.4c
(A) 2.0c
(B) 5.9c
(C) 7.4c
(A)1.3c
2h
3i
diag-(η5-C5H4Me)Re(CO)2Br2
diag-(η5-C5H4Me)Re(CO)2I2
(H) 29.7j
(I) 40.2j
(J ) 14.9j
(H) 34.1j
(I) 0.0j
96.6
72.6
diag-(η5-C5H4Me)Re(CO)2Br2
diag-(η5-C5H4Me)Re(CO)2I2
(B) 1.4c
(C) 6.0c
(E) 35.7d
(J ) 57.3j
(D) 64.3d
4
5
6
7
8
diag-(η5-C5H4Me)Re(CO)2Br2
20-fold NaI
99.0
84.7
78.1
90.5e
75.4
diag-(η5-C5H4Me)Re(CO)2Br2
20-fold NaCl
14.2e
70.2f
15.0e
0.0
85.8e
29.8f
85.0e
diag-(η5-C5H4Me)Re(CO)2I2
20-fold NaCl
diag-(η5-C5H4Me)Re(CO)2Br2
5-fold P(OPh)3
diag-(η5-C5H5)Re(CO)[P(OMe)3]Br2
diag-(η5-C5H4Me)Re(CO)[P(OPh)3]Br2
(F) 45.3g
(G) 54.7g
a
Two equimolar complexes were mixed together and coadsorbed on silica in the impregnation method. Reaction conditions were the
same as those recorded in Table 2. The mixture consisted of diag-(η5-C5H4Me)Re(CO)2Br2, diag-(η5-C5H4Me)Re(CO)2BrI, and diag-(η5-
b
C5H4Me)Re(CO)2I2. c Legend: (A) lat-(η5-C5H4Me)Re(CO)2I2; (B) lat-(η5-C5H4Me)Re(CO)2BrI; (C) lat-(η5-C5H4Me)Re(CO)2Br2. Legend:
d
(D) diag-(η5-C5H4Me)Re(CO)2I2; (E) lat-(η5-C5H4Me)Re(CO)2I2. e Only (η5-C5H4Me)Re(CO)2Br2 was found in the product. f Only (η5-
C5H4Me)Re(CO)2I2 was found in the product. Legend: (F) lat-(η5-C5H5)Re(CO)[P(OMe)3]Br2; (G) lat-(η5-C5H4Me)Re(CO)[P(OPh)3]Br2.
g
The complexes were refluxed in the dark in benzene for 20 h without silica gel. iSilica gel was treated with 30% dichlorodimethylsilane.
h
Legend: (H) diag-(η5-C5H4Me)Re(CO)2I2; (I) diag-(η5-C5H4Me)Re(CO)2BrI; (J ) diag-(η5-C5H4Me)Re(CO)2Br2.
j
Ta ble 5. Solid -Sta te P h otoch em ica l Isom er iza tion in th e P r esen ce of a F r ee-Ra d ica l Sca ven ger a
entry no.
complex
diag (%)
lat (%)
total recovery (%)
1
2
3
diag-(η5-C5H4Me)Re(CO)2Br2 + equimolar NaI
diag-(η5-C5H4Me)Re(CO)2Br2 + 3-fold galvinoxyl
diag-(η5-C5H4Me)Re(CO)2Br2 + equimolar NaI +
3-fold galvinoxyl
31.8b
78.3
(A) 23.8,c (B) 38.1c (C) 38.1c
21.7
76.7
83.1
90.2
81.2b
(A) 1.1,c (B) 6.1,c (C) 11.6c
a
b
Reaction conditions were the same as those recorded in Table 2. The mixture of diag-(η5-C5H4Me) Re(CO)2Br2, diag-(η5-
C5H4Me)Re(CO)2BrI, and diag-(η5-C5H4Me)Re(CO)2I2. c Legend: (A) lat-(η5-C5H4Me)Re(CO)2I2; (B) lat-(η5-C5H4Me)Re(CO)2BrI; (C) lat-
(η5-C5H4Me) Re(CO)2Br2.
irradiation of silica-supported diag-(η5-C5H4Me)Re-
(CO)2I2.20 Irradiation of a mixture of diag-(η5-C5H4Me)-
Re(CO)2Br2 and excess NaI on SiO2 gave exclusively a
photostationary mixture of diag- and lat-(η5-C5H4Me)-
Re(CO)2I2 (Table 4). No halogen exchange between
NaCl and diag-(η5-C5H4Me)Re(CO)2X2 (X ) Br, I) took
place under the irradiation conditions. This may be
related to the ease of formation of the halogen radical
X.21
Further experiments were performed to correlate
solid-state photochemical isomerization reactions with
a halogen dissociation-association reaction. The solid-
state photochemical isomerization reaction of an equi-
molar mixture of diag-(η5-C5H4Me)Re(CO)2Br2 and diag-
(η5-C5H4Me)Re(CO)2I2 was carried out on a completely
dehydroxylated surface of silica gel. Neither isomer-
ization activity nor any Br and I exchange was detected
(Table 4). The solid-state photochemical isomerization
reaction of an equimolar mixture of diag-(η5-C5H4Me)-
Re(CO)2Br2 and NaI in the presence of the free-radical
scavenger galvinoxyl was also carried out, and it was
observed that galvinoxyl greatly suppressed both Br and
I ligand exchange and the isomerization reaction (Table
5). Because the photochemical reactions were carried
out in the solid-state and not in solution, the role of
galvinoxyl is not completely clear. It is possible that
competitive light absorption by galvinoxyl or surface
occupation by galvinoxyl molecules reduced the ef-
ficiency of the photochemical reactions. However, the
results clearly suggest that a close relationship exists
between the halogen dissociation-association reaction
and the solid-state photochemical isomerization reac-
tion.
The Br and I ligand exchange was also observed when
an equimolar mixture of diag-(η5-C5H4Me)Re(CO)2Br2
and diag-(η5-C5H4Me)Re(CO)2I2 was refluxed in benzene
in the dark (Table 4), which indicated that the halogen
dissociation-association of (η5-C5H4R)Re(CO)(L)X2 can
occur under both photochemical and thermal conditions
and in both solid and solution phases.
Two mechanisms for the solid-state photochemical
isomerization reactions of (η5-C5H4R)Re(CO)(L)X2 on the
surface of silica gel can be proposed (Schemes 1 and 2).
Both mechanisms include interaction of the halogen
ligand X of (η5-C5H4R)Re(CO)(L)X2 with the surface of
silica gel. These mechanisms are different from the
original proposal made by Hill7b for similar compounds,
in which CO dissociation was shown to be the key step
for photochemical isomerization at low temperature.
However, a more recent report by Hill and co-workers
has implicated metal-halogen bond cleavage in a diag-
lat isomerization reaction of related rhenium com-
plexes.22
These two mechanisms are shown in diagram form
in Schemes 1 and 2. In Scheme 1, upon irradiation, the
(20) Molecular iodine was separated from the isomerization products
of diag-(η5-C5H4Me)Re(CO)2I2 on a chromatography column and con-
firmed by reaction with starch in water (blue color).
(21) The bond energies of NaI, NaBr, and NaCl are 72.7, 86.7, and
97.5 kcal/mol, respectively.
(22) Xia, W.; Hill, R. H.; Klahn, A. H.; Leiva, C.; Buono-Core, G. E.
Polyhedron 1996, 10, 3093.

596 Organometallics, Vol. 16, No. 4, 1997
Cheng and Coville
Sch em e 2. P r op osed Mech a n ism 2 for th e
Solid -Sta te P h otoch em ica l d ia g-la t Isom er iza tion
of d ia g-(η5-C5H4R)Re(CO)(L)X2 on th e Su r fa ce of
Silica Gel
Sch em e 1. P r op osed Mech a n ism 1 for th e
Solid -Sta te P h otoch em ica l d ia g-la t Isom er iza tion
of d ia g-(η5-C5H4R)Re(CO)(L)X2 on th e Su r fa ce of
Silica Gel
reactions, the ring substituent has little effect on the
photochemical solid-state isomerization reactions of (η5-
C5H4R)Re(CO)(L)X2. The photostationary ratio [lat]/
[diag] was mainly determined by the ligands L and X.
The solution and matrix isomerization reactions of (η5-
C5R4R′)Re(CO)2X2 (R ) H, Me; R′ ) Me) gave results
completely different from those observed for the solid-
state thermal and photochemical isomerization reac-
tions discussed here. It was found that only monolayer
adsorbed molecules on the silica surface underwent
isomerization and studies showed that the hydroxyl
groups on silica gel play a crucial role in the isomer-
ization reactions. The isomerization reactions could be
rationalized by a mechanism in which a halogen dis-
sociative-associative process involving ions or radicals
occurs under irradiation with visible light.
ligand X dissociates to form a halogen ion and a diagonal
three-legged cation intermediate which rearranges to
the lateral configuration. The later recombines with X-
(or X′-) ion to form the lateral isomer. In Scheme 2,
irradiation results in Re-X bond homolysis to form the
halogen radical X• and a diagonal three-legged radical
intermediate, which converts to a lateral radical inter-
mediate and recombines with X• (or X′•) to form a lateral
isomer. While the details of the two mechanisms
proposed presently lack direct experimental support,
they provide a basis for further mechanistic studies on
this unusual solid-state photochemical isomerization
reaction on silica gel.
Con clu sion s
The solid-state diagonal to lateral isomerization of
monosubstituted cyclopentadienyl four-legged piano-
stool rhenium complexes (η5-C5H4R)Re(CO)(L)X2 (R )
H, alkyl; L ) CO, P(OR)3; X ) Br, I) proceeds in good
yield when the complexes are adsorbed on the surface
of silica gel and irradiated with visible light. Unlike
the thermal solid-state or molten-state isomerization
Ack n ow led gm en t. We wish to thank the FRD and
the University of The Witwatersrand for financial
support.
OM960726X