Comparison of the reactivity of M(allyl)3 (M = Rh, Ir) with donor ligands

Article abstract of DOI:10.1021/om000734k

Reactions of M(allyl)₃ [M = Rh, Ir] with a variety of donor ligands have been investigated. While both complexes are unreactive toward `hard' N- and O-donor ligands, `soft' ligands such as tolylisothiocyanate, tertiary phosphines, CO, and aryl isocyanides are readily added. Whereas ligand additions to Rh(allyl)₃ are often accompanied by reduction of the Rh(III) center, analogous reactions with Ir(allyl)₃ afford stable trivalent products containing σ-allyl ligands. The molecular structures of several derivatives are presented, including Ir(σ-allyl)(π-allyl)₂(PPh₃), Ir(σ-allyl)₂(π-allyl)(PMe₃)₂, and Ir(σ-allyl)₃(C triple bond N-2,6-Me₂C₆H₃)₃.

Full text of DOI:10.1021/om000734k

296
Organometallics 2001, 20, 296-304
Com p a r ison of th e Rea ctivity of M(a llyl)3 (M ) Rh , Ir )
w ith Don or Liga n d s
Kevin D. J ohn,† Kenneth V. Salazar,‡ Brian L. Scott,† R. Thomas Baker,*,† and
Alfred P. Sattelberger*,†
Los Alamos Catalysis Initiative, Chemistry Division, Los Alamos National Laboratory,
Los Alamos, New Mexico 87545, and Materials Science and Technology Division,
Los Alamos National Laboratory, Los Alamos, New Mexico 87545
Received August 22, 2000
Reactions of M(allyl)3 [M ) Rh, Ir] with a variety of donor ligands have been investigated.
While both complexes are unreactive toward “hard” N- and O-donor ligands, “soft” ligands
such as tolylisothiocyanate, tertiary phosphines, CO, and aryl isocyanides are readily added.
Whereas ligand additions to Rh(allyl)3 are often accompanied by reduction of the Rh(III)
center, analogous reactions with Ir(allyl)3 afford stable trivalent products containing σ-allyl
ligands. The molecular structures of several derivatives are presented, including Ir(σ-allyl)-
(π-allyl)2(PPh3), Ir(σ-allyl)2(π-allyl)(PMe3)2, and Ir(σ-allyl)3(CtN-2,6-Me2C6H3)3.
In tr od u ction
remains the key reference on solution reactions of Rh-
(allyl)3.7 They showed that protonolysis of Rh(allyl)3
gave [Rh(allyl)2(µ-Cl)]2, whereas addition of triphen-
ylphosphine afforded monovalent Rh(allyl)(PPh3)2. We
recently reported an improved preparation of Ir(allyl)3
which has made reactivity studies of this complex
practical.8 In this contribution we compare and contrast
the reactivity of tris(allyl)rhodium and -iridium with a
variety of N, O, P, S, and C donor ligands.
Investigations of the reactivity of metal allyl com-
plexes have been central to the development of organo-
metallic chemistry.1 Despite the number of studies
focused on compounds possessing group 9 metal-allyl
fragments, little is known of the reactivity of the
homoleptic tris(allyl) complexes of these metals.2 In the
case of cobalt, the scarcity of studies is likely due to its
extreme air sensitivity and thermal instability of the
parent (decomposes above -40 °C). One reaction of note
is that with carbon monoxide, which gives Co(allyl)(CO)3
and 1,5-hexadiene.3 For iridium, the dearth of synthetic
reports is most likely due to the low literature yield,4
coupled with the high cost of starting materials. The
only reaction reported is the protonolysis of Ir(allyl)3 to
form [Ir(allyl)2(µ-Cl)]2.5 While reactions of tris(allyl)-
rhodium with a variety of metal oxide supports have
been reported,6 the pioneering work by Powell and Shaw
Exp er im en ta l Section
All manipulations were performed under inert atmosphere
or vacuum using standard glovebox, vacuum-line, and Schlenk
techniques. Solvents were dried by elution from columns of
activated alumina and copper oxide BTS catalyst according
to the procedure described by Grubbs et al.9 All reactions were
carried out at atmospheric pressure (600 mTorr at Los Alamos,
elevation 7200 ft). mer-MCl3(tht)3 (M ) Rh, Ir) were prepared
by a modification of the literature procedure.10 Allyllithium
was prepared from tetraallyltin and n-buytllithium by a
modification to the literature procedure.11 All other chemicals
were used as received. NMR spectra were recorded with a
Varian UNITY series 300 MHz spectrometer, and all spectra
were recorded at room temperature in C6D6 unless noted
otherwise. All NMR solvents were stored over 4 Å molecular
sieves. 31P{1H} chemical shifts are reported relative to an
external 85% H3PO4 standard. Infrared spectra were measured
with a Nicolet Magna 760 spectrometer at 4 cm-1 resolution
as mineral oil mulls or solutions contained in an airtight
sodium chloride cell. GC/MS analyses were performed on a HP
6980 GC system with a HP 5973 mass-selective detector.
† Los Alamos Catalysis Initiative, Chemistry Division.
‡ Materials Science and Technology Division.
(1) (a) Guy, R. G.; Shaw, B. L. Adv. Inorg. Chem. Radiochem. 1962,
4, 78. (b) Green, M. L. H.; Nagy, P. L. I. Adv. Organomet. Chem. 1964,
2, 325. (c) Wilke, G.; Bogdanovic, B.; Hardt, P.; Heimbach, P.; Keim,
W.; Kro¨ner, M.; Oberkirch, W.; Tanaka, K.; Steinru¨cke, E.; Walter,
D.; Zimmermann, H. Angew. Chem., Int. Ed. Engl. 1966, 5, 151. (d)
Lobach, M. I.; Babitskii, B. D.; Kormer, V. A. Russ. Chem. Rev. 1967,
36, 1158. (e) Powell, P. In The Chemistry of the Metal-Carbon Bond;
Hartley, F. R., Patai, S., Eds.; J ohn Wiley and Sons: New York, 1982;
Vol. 1, Chapter 8, p 325. (f) Wreford, S. S. In Inorganic Reactions and
Methods; Zuckerman, J . J ., Hagen, A. P., Eds.; VCH: New York, 1991;
Vol. 12a, Chapter 5.8.2.8.2.
(2) (a) Sweany, R. L. (Co); Sharp, P. R. (Rh); Atwood, J . D. (Ir) In
Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G.
A., Abel, E. W., Eds.; Pergamon: Oxford, 1982; Chapter 5, pp 1, 115,
303. (b) Kemmitt, R. D. W.; Russell, D. R. (Co); Hughes, R. P. (Rh);
Leigh, G. J .; Richards, R. L. (Ir) In Comprehensive Organometallic
Chemistry II; Abel, E. W,; Stone, F. G. A.; Wilkinson, G., Eds.;
Pergamon: Oxford, 1995; Chapter 5, pp 1, 277, 541. (c) White, C. The
Organometallic Chemistry of Cobalt, Rhodium, and Iridium; Chapman
and Hall: New York, 1985.
(7) Powell, J .; Shaw, B. L. J . Chem. Soc. (A) 1968, 583.
(8) J ohn, K. D.; Salazar, K. V.; Scott, B. L.; Baker, R. T.; Sattel-
berger, A. P. Chem. Commun. 2000, 581.
(9) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J . Organometallics 1996, 15, 1518.
(10) Allen, E. A.; Wilkinson, W. J . Chem Soc., Dalton Trans. 1972,
613.
(3) Bo¨nnemann, H.; Grard, C.; Kopp, W.; Pump, W.; Tanaka, K.;
Wilke, G. Angew. Chem., Int. Ed. Engl. 1973, 12, 964.
(4) Chini, P.; Martinengo, S. Inorg. Chem. 1967, 6, 837.
(5) Green, M.; Parker, G. J . J . Chem. Soc., Dalton Trans. 1974, 333.
(6) Basset, J . M.; Lefebvre, F.; Santini, C. Coord. Chem. Rev. 1998,
178-180, 1703, and references therein.
(11) (a) Eisch, J . J . In Nontransition Metal Compounds: Organo-
metallic Syntheses; Academic Press: New York, 1981; Vol. 2. (b)
Horikawa, Y.; Takeda, T. J . Organomet. Chem. 1996, 523, 99. Note:
The latter preparation calls for the use of 4 equiv of butyllithium per
1 tetraallyl tin. We have found that the use of 2 equiv of butyllithium
eliminates the presence of butyllithium in the end product.
10.1021/om000734k CCC: $20.00 © 2001 American Chemical Society
Publication on Web 12/16/2000

Reactivity of M(allyl)3 (M ) Rh, Ir)
Organometallics, Vol. 20, No. 2, 2001 297
Elemental analyses were performed on a Perkin-Elmer 2400
system with a Series II CHNS/O analyzer and an AD-4
autobalance.
(d, 4 H, syn-CH2CHCH2), 2.53 (d, 2 H, anti-CH2CHCH2), 1.53
(d, 4 H, anti-CH2CHCH2). 13C{1H} NMR: δ 100.18 (s, 1 C, 1J RhC
1
) 4.3 Hz, CH2CHCH2), 95.06 (s, 2 C, J RhC ) 4.0 Hz,
CH2CHCH2), 48.23 (s, 2 C, 1J RhC ) 9.2 Hz, CH2CHCH2), 41.22
(s, 4 C, 1J RhC ) 9.1 Hz, CH2CHCH2). Anal. Calcd for C9H15Rh:
C, 47.81; H, 6.69. Found: C, 47.76; H, 6.72.
P r ep a r a tion of Ir (σ-a llyl)(π-a llyl)2(SCN-Tol) (5). To a
room-temperature solution of Ir(allyl)3 (0.100 g, 0.32 mmol)
in toluene (10 mL) was added tolylisothiocyanate (0.047 g, 0.32
mmol). The reaction mixture was stirred for 3 days at room
temperature. The reaction solution was concentrated to ca. 2
mL and the product precipitated by addition of 10 mL of
hexane to provide 5 as colorless crystals in 83.0% yield (0.122
1
g, 0.26 mmol). H NMR: δ 6.77, 6.62 (d, 2 H, C6H4), 5.90 (m,
1 H, CH2-CHdCH2-2), 4.99 (overlapping ddt, 2 H, CH2-CHd
CH2-1), 4.31 (ddt, 1 H, syn-CH2CHCH2-6), 3.93 (d d, 1 H, syn-
CH2CHCH2-9), 3.82 (m, 1 H, CH2CHCH2-5), 3.65 (m, 1 H,
CH2CHCH2-8), 3.34 (d, 1 H, anti-CH2CHCH2-9), 2.96 (ddt, 2
H, CH2-CHdCH2-3), 2.2 (overlapping d, 2 H, CH2CHCH2-4,6),
1.97 (s, 3 H, Me), 1.88 (d, 1 H, syn-CH2CHCH2-7), 1.50 (d, 1
H, anti-CH2CHCH2-4), 0.77 (d, 1 H, anti-CH2CHCH2-7). 13C-
{1H} NMR: δ 132.74 (s, CH2-CHdCH2), 117.61 (s, CH2-CHd
CH2), 81.37 (s, CH2CHCH2), 77.54 (s, CH2CHCH2), 45.85 (s,
CH2CHCH2), 44.66 (s, CH2CHCH2), 39.60 (s, CH2CHCH2),
33.34 (s, CH2CHCH2), 28.34 (s, Me), 24.68 (s, CH2-CHdCH2).
P r ep a r a tion of Ir (σ-a llyl)(π-a llyl)2(P P h 3) (6). To a solu-
tion of Ir(allyl)3 (0.100 g, 0.32 mmol) in 10 mL of toluene at 20
°C was added triphenylphosphine (0.083 g, 0.32 mmol). The
reaction mixture was stirred for 2 h and then filtered. The
filtrate was concentrated to ca. 2 mL and allowed to sit at 20
°C for 24 h to yield 6 as colorless crystals in 93% yield (0.170
g, 0.24 mmol). 1H NMR: δ 7.43 (m, 5 H, Ph), 6.9 (m, 10 H,
Ph), 6.39 (m, 1 H, CH2-CHdCH2-2), 4.61 (overlapping ddt, 2
H, CH2-CHdCH2-1′), 4.39 (m, 1 H, CH2CHCH2-5), 4.10 (m, 1
m er -Ir Cl3(th t)3 (1). To a room-temperature suspension of
IrCl3‚3H2O (2.00 g, 5.67 mmol) in 2-methoxyethanol (100 mL)
was added tetrahydrothiophene (2.50 mL, 28.4 mmol) with
stirring. The solution was refluxed for 12 h, after which time
water (150 mL) was added to the room-temperature solution.
This treatment followed by filtration provided the crude
product as a yellow powder. The product was further purified
by recrystallization from boiling ethanol to yield 1 as yellow
microcrystals in 91.5% yield (2.92 g, 5.19 mmol). 1H NMR (CD2-
Cl2, 25 °C): δ 3.60 (m, 4H, SCH2CH2), 3.23 (m, 2H, SCH2CH2),
2.9 (overlapping m, 6H, SCH2CH2), 2.4-2.1 (overlapping m,
12H, SCH2CH2). 13C{1H} NMR (CD2Cl2, 25 °C): δ 36.92 (2C,
SCH2CH2), 36.51 (4C, SCH2CH2), 30.49 (4C, SCH2CH2), 30.40
(2C, SCH2CH2). Anal. Calcd for C12H24Cl3S3Ir: C, 25.60; H,
4.30. Found: C, 25.70; H, 4.39.
m er -Rh Cl3(th t)3 (2). RhCl3(tht)3 was prepared in the same
manner as above from RhCl3‚3H2O to provide orange micro-
1
crystals of 2 in 93.1% yield. H NMR (CD2Cl2, 25 °C): δ 3.70
(m, 4H, SCH2CH2), 3.28 (m, 2H, SCH2CH2), 2.90 (overlapping
m, 6H, SCH2CH2), 2.3-2.0 (overlapping m, 12H, SCH2CH2).
13C{1H} NMR (CD2Cl2, 25 °C): δ 37.69 (2C, SCH2CH2), 37.31
(4C, SCH2CH2), 30.38 (4C, SCH2CH2), 30.23 (2C, SCH2CH2).
Anal. Calcd for C12H24Cl3S3Rh: C, 30.42; H, 5.11. Found: C,
30.43; H, 5.44.
3
H, CH2CHCH2-8), 3.85 (ddt, J HH ) 6.3 Hz, 1 H, syn-CH2-
CHCH2-6), 3.20 (d, 3J HH ) 6.2 Hz, 1 H, syn-CH2CHCH2-9), 3.07
(m, 1 H, CH2-CHdCH2-3), 2.74 (br s, 1 H, syn-CH2CHCH2-
3
4), 2.26 (d, J HH ) 6.9 Hz, 1 H, syn-CH2CHCH2-7), 1.70 (d,
3
3J HH ) 9.6 Hz, 1 H, anti-CH2CHCH2-6), 1.40 (dd, J HH ) 10.6
P r ep a r a tion of Ir (a llyl)3 (3). To a room-temperature
suspension of IrCl3(tht)3 (0.509 g, 0.90 mmol) in Et2O (40 mL)
was added a solution of allyllithium (0.130 g, 2.70 mmol, in
10 mL of Et2O) dropwise over 20 min with vigorous stirring.
The reaction mixture was stirred for 14 h at room temperature,
during which time the color of the solution changed from pale
yellow to brown. The reaction mixture was then filtered and
the residue rinsed with Et2O (2 × 10 mL). The ether extract
was reduced to dryness under vacuum, and the product was
extracted with hexane (50 mL) and filtered. The hexane extract
was reduced in volume to about 3 mL and transferred to a
sublimator, where the remaining ether was removed and the
brown residue was sublimed onto a coldfinger (-78 °C) at 10-5
Torr with mild heating (<60 °C). Sublimation afforded color-
less microcrystalline Ir(allyl)3 in 44.4% yield (0.126 g, 0.40
mmol). A solution of Ir(allyl)3 can be prepared in over 80% yield
(based on NMR analysis compared to an internal standard)
from the reaction of IrCl3(tht)3 and allyllithium in benzene.
However, the product can only be isolated in ca. 20% yield from
sublimation (the poor yield is presumably due to “cosublima-
tion” of Ir(allyl)3 with the benzene). The solution prepared in
this manner has been used to generate some of the complexes
3
Hz, J HP ) 11.7 Hz, 1 H, anti-CH2CHCH2-9), 1.24 (ddt, 1 H,
3
3
CH2-CHdCH2-3), 1.09 (dd, J HH ) 9.2 Hz, J HP ) 7.8 Hz, 1
3
3
H, anti-CH2CHCH2-4), 0.73 (dd, J HH ) 11.0 Hz, J HP ) 10.2
Hz, 1 H, anti-CH2CHCH2-7). 13C{1H} NMR: δ 151.58 (d, J CP
3
) 2.3 Hz, CH2-CHdCH2), 134.69 (s, Aryl), 129.68 (s, Aryl),
128.17 (s, Aryl), 128.04 (s, Aryl), 102.60 (s, CH2-CHdCH2),
2
85.76 (s, CH2CHCH2), 78.94 (s, CH2CHCH2), 45.73 (d, J CP
)
2
3.0 Hz, CH2CHCH2), 43.34 (d, J CP ) 38.1 Hz, CH2CHCH2),
2
40.72 (s, CH2CHCH2), 32.46 (d, J CP ) 2.5 Hz, CH2CHCH2),
11.22 (d, 2J CP ) 3.7 Hz, CH2-CHdCH2). 31P{1H} NMR: δ 18.59
(s). IR (mineral oil): 1602 cm-1 (vw), ν(CdC). Anal. Calcd for
C
27H30PIr: C, 56.13; H, 5.23. Found: C, 55.93; H, 5.33.
P r ep a r a tion of Ir (σ-a llyl)2(π-a llyl)(P h 2P C6H4P P h 2) (7).
To a solution of Ir(allyl)3 (0.100 g, 0.32 mmol) in 10 mL of
toluene at 20 °C was added 1,2-bis(diphenylphosphino)benzene
(0.142 g, 0.32 mmol). Addition of the phosphine resulted in
an immediate color change to pale orange. The reaction
mixture was heated briefly (10 min) to 85 °C to completely
dissolve the phosphine and was then stirred for 14 h at room
temperature. The orange solution was filtered and the filtrate
concentrated to ca. 2 mL and allowed to sit at -35 °C for 24
h to yield 7 as colorless crystals in 89% yield (0.216 g, 0.28
1
listed below with no effect on the product yields. H NMR: δ
1
5.00 (m, 1 H, CH2CHCH2), 3.30 (m, 2 H, CH2CHCH2), 2.86 (d,
4 H, syn-CH2CHCH2), 2.75 (d, 2 H, syn-CH2CHCH2), 2.66 (d,
2 H, anti-CH2CHCH2), 1.72 (d, 4 H, anti-CH2CHCH2). 13C{1H}
NMR: δ 91.98 (s, 1 C, CH2CHCH2), 82.17 (s, 2 C, CH2CHCH2),
38.07 (s, 2 C, CH2CHCH2), 28.33 (s, 4 C, CH2CHCH2). Anal.
Calcd for C9H15Ir: C, 34.26; H, 4.79. Found: C, 34.57; H, 5.05.
P r ep a r a tion of Rh (a llyl)3 (4). Bright yellow Rh(allyl)3 was
prepared in the same manner as above (in Et2O) from RhCl3-
mmol). H NMR: δ 7.9-6.8 (24 H, Aryl), 6.35 (m, 1 H, CH2-
3
CHdCH2-A), 5.09 (m, 1 H, CH2-CHdCH2-B), 4.96 (d, J HH
)
9.9 Hz, 1 H, syn-CH2-CHdCH2-B), 4.88 (d, 3J HH ) 16.8 Hz, 1
H, anti-CH2-CHdCH2-B), 4.8 (obscured m, 1 H, CH2CHCH2),
3
4.39 (d, J HH ) 14.4 Hz, 1 H, anti-CH2-CHdCH2-A), 4.17 (d,
3J HH ) 9.6 Hz, 1 H, syn-CH2-CHdCH2-A), 2.85 (br s, 1 H,
syn-CH2CHCH2), 2.73 (d, 1 H, anti-CH2CHCH2), 2.69 (d, 1 H,
anti-CH2CHCH2), 2.57 (m, 1 H, syn-CH2CHCH2), 2.0 (overlap-
1
(tht)3 in 85.9% yield. H NMR: δ 5.11 (m, 1 H, CH2CHCH2),
ping ddt, 3 H, CH2-CHdCH2), 1.72 (ddt, 1 H, CH2-CHdCH2).
3
3.68 (m, 2 H, CH2CHCH2), 2.83 (d, 2 H, syn-CH2CHCH2), 2.65
13C{1H} NMR: δ 149.56 (s, CH2-CHdCH2), 148.06 (d, J CP
)

298 Organometallics, Vol. 20, No. 2, 2001
J ohn et al.
6.0 Hz, CH2-CHdCH2), 144.9-127.8 (aromatic),106.46 (d, J CP
) 7.3 Hz, CH2-CHdCH2), 100.83 (s, CH2-CHdCH2), 97.39
CHdCH2-B), 4.87 (ddt, 1 H, syn-CH2-CHdCH2-B), 4.65
(overlapping ddt, 3 H, CH2-CHdCH2), 3.61 (m, 1 H, CH2-
CHCH2), 3.06 (ddt, 2 H, anti-CH2CHCH2), 2.64 (d d, 2 H, syn-
CH2CHCH2), 2.2 (overlapping m, 4 H, CH2-CHdCH2). 13C-
2
(s, CH2CHCH2), 48.89 (s, CH2CHCH2), 48.75 (s, J CP ) 39.4
Hz, CH2CHCH2), 18.95 (s, 2J CP ) 79.5; 5.0 Hz, CH2-CHdCH2),
6.99 (s, CH2-CHdCH2). 31P{1H} NMR: δ 26.97, 23.65. IR
(mineral oil): 1604 cm-1 (vw), ν(CdC). Anal. Calcd for C39H39P2-
Ir: C, 61.48; H, 5.16. Found: C, 61.39; H, 5.37.
3
{1H} NMR: δ 153.06 (m, J CP ) 6.7 Hz, CH2-CHdCH2),
147.91 (m, CH2-CHdCH2), 113.68 (s, CH2CHCH2), 106.03 (s,
CH2-CHdCH2), 104.86 (s, CH2-CHdCH2), 48.93 (m, 2J CP-trans
2
) 47.5 Hz, J CP-cis ) 9.8 Hz CH2CHCH2), 1.87 (br s, CH2-
P r epar ation of Ir (σ-allyl)3(P h 2P (CH2)2P (P h )(CH2)2P P h 2)
(8). To a solution of Ir(allyl)3 (0.100 g, 0.32 mmol) in 10 mL of
toluene at 20 °C was added bis(2-diphenylphosphinoethyl)-
phenylphosphine (0.169, 0.32 mmol). The reaction mixture was
heated briefly (10 min) to 85 °C to completely dissolve the
phosphine and was then stirred for 14 h at room temperature.
The solution was filtered, and the pale yellow filtrate was
concentrated to ca. 2 mL and allowed to sit at -35 °C for 24
h to yield 8 as colorless crystals in 91.8% yield (0.247 g, 0.29
CHdCH2), -0.42 (m, CH2-CHdCH2). 31P{1H} NMR: δ 70.61.
Anal. Calcd for C45H45P2O6Ir: C, 57.74; H, 4.85. Found: C,
57.86; H, 4.82.
P r ep a r a tion of Rh (σ-a llyl)2(π-a llyl)(P Me3)2 (9c). To a
room-temperature solution of Rh(allyl)3 (0.080 g, 0.35 mmol)
in 10 mL of toluene was added trimethylphosphine (72 mL,
0.71 mmol), at which point the color of the reaction mixture
went from golden yellow to a brilliant yellow. The solution was
stirred for 2 h and filtered. The filtrate was then concentrated
to ca. 2 mL and allowed to sit at -35 °C for 24 h to yield 9c as
1
mmol). H NMR: δ 7.6-6.6(25 H, Ph), 6.9 (obscured m, 1 H,
CH2-CHdCH2), 5.92 (m, 2 H, CH2-CHdCH2), 5.40 (ddt, 1
H, anti-CH2-CHdCH2), 5.04 (ddt, 1 H, syn-CH2-CHdCH2),
4.4 (overlapping ddt, 4 H, CH2-CHdCH2), 3.62 (m, 2 H, CH2-
CHdCH2), 2.2 (obscured m, 4 H, CH2-CHdCH2). 13C{1H}
NMR: δ 150.00 (s, 1 C, CH2-CHdCH2), 149.05 (s, 2 C, CH2-
CHdCH2), 132.2-125.6 (aromatic), 106.73 (s, 1 C, CH2-CHd
CH2), 105.82 (s, 2 C, CH2-CHdCH2), 29.9 (m, 4 C, -CH2-),
1
yellow crystals in 98.4% yield (0.132 g, 0.35 mmol). H NMR:
δ 5.7 (overlapping m, 2 H, CH2-CHdCH2), 4.81 (ddt, 1 H, syn-
CH2-CHdCH2-A), 4.7 (overlapping ddt, 2 H, CH2-CHdCH2-
B), 4.60 (ddt, 1 H, anti-CH2-CHdCH2-A), 4.09 (m, 1 H,
CH2CHCH2), 3.05 (ddt, 2 H, anti-CH2CHCH2), 2.19 (d d, 2 H,
syn-CH2CHCH2), 1.32 (m, 2 H, CH2-CHdCH2-A), 0.90 (d, 18
2
2
13.29 (m, J CP ) 78.7; 4.1 Hz, 2 C, CH2-CHdCH2), 12.77 (m,
H, J HP ) 8.4 Hz, PMe3), 0.85 (obscured m, 2 H, CH2-CHd
2J CP ) 79.6 Hz, 1 C, CH2-CHdCH2). 31P{1H} NMR: δ 34.87-
(1P), -2.70 (2P). IR (mineral oil): 1608 cm-1 (vw), ν(CdC).
Anal. Calcd for C43H48P3Ir: C, 60.76; H, 5.69. Found: C, 60.71;
H, 5.97.
CH2-B). 13C{1H} NMR: δ 148.68 (m, 3J CP ) 3.0 Hz, CH2-CHd
3
CH2), 147.16 (m, J CP ) 3.3 Hz, CH2-CHdCH2), 116.65 (s,
CH2CHCH2), 102.57 (s, CH2-CHdCH2), 57.62 (m, 2J CP-trans
)
1
40.7 Hz, CH2CHCH2), 17.69 (m, J CP ) 27.1 Hz, PMe3), 16.15
2
2
(m, J CP ) 15.8 Hz, CH2-CHdCH2), 14.33(m, J CP ) 14.0 Hz,
CH2-CHdCH2). 31P{1H} NMR: δ -7.02 (J RhP ) 149.9 Hz).
P r ep a r a tion of Rh (π-a llyl)(P (OP h )3)2 (11). To a room-
temperature solution of Rh(allyl)3 (0.0313 g, 0.14 mmol) in 5
mL of toluene was added triphenyl phosphite (0.086 g, 0.28
mmol) with stirring. Approximately 30 min after the phosphite
addition, a brilliant yellow precipitate was observed. The
reaction mixture was stirred an additional 2 h, during which
time the precipitate had dissolved. The filtrate was then
concentrated to ca. 2 mL and allowed to sit at -35 °C for 24
h to yield 11 as yellow crystals in 96.7% yield (0.102 g, 0.13
mmol). GC analysis of a portion of this reaction revealed that
the organic byproducts were n-hexane, 2-methylpentane, and
P r ep a r a tion of Ir (σ-a llyl)2(π-a llyl)(P Me3)2 (9a ). To a
room-temperature solution of Ir(allyl)3 (0.100 g, 0.32 mmol)
in 10 mL of toluene was added trimethylphosphine (65 µL,
0.63 mmol). The reaction mixture was stirred for 14 h and then
filtered. The filtrate was concentrated to ca. 2 mL and allowed
to sit at -35 °C for 24 h to yield 9a as colorless crystals in
96.5% yield (0.143 g, 0.31 mmol). Extended vacuum workup
of the reaction mixture resulted in formation of a 50/50 mixture
of 9a and Ir(σ-allyl)(π-allyl)2PMe3 (10). Formation of the latter
complex appears to be reversible, as addition of PMe3 results
in clean conversion of 10 back to 9a . 1H NMR: δ 5.8
(overlapping m, 2 H, CH2-CHdCH2), 4.82 (ddt, 1 H, syn-CH2-
CHdCH2-A), 4.7 (overlapping ddt, 2 H, CH2-CHdCH2-B), 4.64
(ddt, 1 H, anti-CH2-CHdCH2-A), 3.83 (m, 1 H, CH2CHCH2),
2.96 (ddt, 2 H, anti-CH2CHCH2), 2.16 (d d, 2 H, syn-CH2-
1
1-hexene. H NMR: δ 7.55, 6.92 (m. 30 H, Ph), 5.20 (m, 1 H,
CH2CHCH2), 3.03 (ddt, 2 H, syn-CH2CHCH2), 2.51 (ddt, 2 H,
anti-CH2CHCH2). 13C{1H} NMR: δ 148.05 (s, aryl), 124.57 (s,
aryl), 119.29 (s, aryl), 117.01 (s, aryl)109.12 (s, CH2CHCH2),
51.29 (m, CH2CHCH2). 31P{1H} NMR: δ -138.54 (J RhP ) 328.3
Hz). Anal. Calcd for C39H35P2O6Rh: C, 61.27; H, 4.61. Found:
C, 61.20; H, 4.76.
2
CHCH2), 1.62 (m, 2 H, CH2-CHdCH2-A), 1.23 (d, 18 H, J HP
) 8.7 Hz, PMe3), 1.20 (m, 2 H, CH2-CHdCH2-B). 13C{1H}
3
NMR: δ 148.56 (s, J CP ) 3.2 Hz, CH2-CHdCH2), 147.13 (s,
3J CP ) 3.6 Hz, CH2-CHdCH2), 109.52 (s, CH2CHCH2), 104.27
(s, CH2-CHdCH2), 104.06 (s, CH2-CHdCH2), 46.39 (d, 2J CP-trans
2
) 32.4 Hz, CH2CHCH2), 46.35 (d, J CP-trans ) 32.8 Hz, CH2-
P r ep a r a tion of Ir (σ-a llyl)3(CtN-2,6-Me2-C6H3)3 (12). To
a room-temperature solution of Ir(allyl)3 (0.100 g, 0.32 mmol)
in toluene (10 mL) was added xylylisocyanide (0.125 g, 0.95
mmol). Upon addition of the isocyanide, the reaction mixture
immediately turned orange-brown in color and was stirred for
10 min. The reaction mixture was allowed to sit at -35 °C for
24 h to yield 12 as colorless crystals in 84.2% yield (0.143 g,
1
2
CHCH2), 16.68 (s, J CP ) 32.0 Hz, PMe3), 0.43 (s, J CP ) 4.6
Hz, CH2-CHdCH2), -2.66 (s, 2J CP ) 5.4 Hz, CH2-CHdCH2).
31P{1H} NMR: δ -50.75.
Selected NMR data for 10: 1H, 6.42 (m, 1 H, CH2-CHd
CH2-2), 4.80 (ddt, 1 H, anti-CH2-CHdCH2-1), 4.80 (ddt, 1 H,
syn-CH2-CHdCH2-1), 4.1 (overlapping m, 2 H, CH2CHCH2-
5,8), 3.21 (overlapping ddt, 2 H, syn-CH2CHCH2-6,9), 2.74 (m,
1 H, CH2-CHdCH2-3), 2.60 (m, 1 H, syn-CH2CHCH2-4), 1.85
(overlapping m, 2 H, CH2CHCH2-7,6), 1.54 (d, 1 H, anti-CH2-
CHCH2-9), 0.94 (d, 9 H, 2J HP ) 9.3 Hz, PMe3), 0.78 (overlapping
d, 2 H, anti-CH2CHCH2-4,7), 0.60 (t, 1 H, CH2-CHdCH2-3).
31P{1H} NMR: δ -44.53.
1
0.31 mmol). H NMR: δ 6.87 (m, 3 H, CH2-CHdCH2), 6.8-
6.6 (9 H, aromatic), 5.11 (ddt, 3 H, anti-CH2-CHdCH2), 4.64
(d d, 3 H, syn-CH2-CHdCH2), 2.64 (d, 6 H, CH2-CHdCH2),
2.26 (s, 18 H, Me). 13C{1H} NMR: δ 152.22 (s, CH2-CHdCH2),
139.02 (s, CtN), 134.94 (s, aromatic), 128.0 (obscured, aro-
matic), 127.90 (s, aromatic), 104.25 (s, CH2-CHdCH2), 18.96
(s, Me), 14.12 (s, CH2-CHdCH2). IR (mineral oil): 2160 cm-1
(s), 2104 (vs), ν(CtN); 1616 cm-1 (vw), ν(CdC). Anal. Calcd
for C27H30PIr: C, 60.99; H, 5.97; N, 5.93. Found: C, 60.72; H,
6.14; N, 6.11.
P r ep a r a tion of Ir (σ-a llyl)2(π-a llyl)[P (OP h )3]2 (9b). To a
room-temperature solution of Ir(allyl)3 (0.060 g, 0.19 mmol)
in 5 mL of toluene was added triphenyl phosphite (0.118 g,
0.38 mmol). The reaction mixture was stirred for 24 h and then
filtered. The filtrate was concentrated to ca. 1 mL, and the
product was precipitated by addition of hexane (5 mL). The
product was isolated by filtration to provide a colorless powder
Rea ction of Rh (a llyl)3 w ith CtN-2,6-Me2C6H3. This
reaction was set up in the same manner as that for 12. Upon
addition of the isocyanide, the golden-yellow color of Rh(allyl)3
quickly turned to dark brown. Attempts to isolate a Rh
complex from this reaction were unsuccessful. GC/MS analysis
1
in 87.5% yield (0.155 g, 0.166 mmol). H NMR: δ 7.2-6.7 (30
H, Ph), 6.31 (m, 1H, CH2-CHdCH2-A), 6.14 (m, 1H, CH2-

Reactivity of M(allyl)3 (M ) Rh, Ir)
Organometallics, Vol. 20, No. 2, 2001 299
Ta ble 1. Cr ysta l Da ta a n d Da ta Collection a n d Refin em en t P a r a m eter s for M(a llyl) Com p lexes
6
7
8
9a
12
empirical formula
fw
space group
a, Å
b, Å
C
54H60Ir2P2
C45H39IrP2‚2(C7H8)
946.17
P1h
C43H48IrP3‚(C7H8)
849.92
P21/n
11.2226(5)
25.4429(13)
16.3141(8)
90
106.450(1)
90
4467.6(4)
1.264
C15H33IrP2
467.55
P21/n
8.3600(4)
27.2735(13)
9.0323(4)
90
115.491(1)
90
1858.94(15)
1.671
C36H42IrN3
708.93
P21/c
10.9408(5)
14.6602(7)
20.4426(9)
90
98.025(1)
90
3246.8(3)
1.450
1155.36
P1h
9.2972(4)
15.7287(7)
16.6995(8)
73.249(1)
82.344(1)
79.303(1)
2289.49(18)
3.352
10.7816(5)
13.5959(6)
13.9441(6)
86.058(1)
71.302(1)
83.341(1)
1921.98(15)
1.441
c, Å
R, deg
â, deg
γ, deg
V, Å3
F
calcd, g cm-3
Z
4
2
4
4
4
µ, mm-1
11.824
3.586
3.121
7.340
4.139
λ(Mo KR), Å
T, K
0.71073
203
0.999
2.62
6.45
0.71073
203
1.003
2.74
7.50
0.71073
203
1.023
2.92
7.33
0.71073
203
1.396
4.97
11.28
5. 57
11.49
3.049 and -3.803
0.71073
203
1.027
3.51
6.89
GooFa
R1 [I > 2σ(I)], %b
wR2 [I > 2σ(I)], %c
R1 (all data), %b
wR2 (all data), %c
lrgst diff peak/hole, e Å-3
3.19
6.70
3.10
7.67
3.92
7.61
5.54
7.49
1.940 and -1.670
0.574 and -0.408
1.136 and -1.018
0.862 and -0.789
GooF ) [∑[w(Fo - Fc2)]2/(n - p)]1/2; n ) number of reflections, p ) total number of parameters refined. R1 ) ∑|Fo| - |Fc|/∑|Fo|.
a
2
b
c wR2 ) [∑[w(Fo - Fc2)2]/∑[w(Fo2)2]1/2
.
2
of a portion of this reaction revealed that CH2dCHCH2-
C(dN-2,6-Me2C6H3)C(dN-2,6-Me2C6H3)CH2CHdCH2 had been
generated (the analogous organic product was observed for the
reaction of Rh(allyl)3 with CNtBu).
2H, CH2-CHdCH2), 5.12 (ddt, 1H, anti-CH2-CHdCH2), 4.82
(ddt, 2H, anti-CH2-CHdCH2), 4.70 (ddt, 1H, syn-CH2-CHd
CH2), 4.63 (ddt, 2H, syn-CH2-CHdCH2), 4.65 (m, 1 H,
CH2CHCH2), 2.49 (m, 2H, CH2-CHdCH2), 1.75 (overlapping
m, 4H, CH2-CHdCH2), 0.80 (s, 9H, PMe3). 13C{1H} NMR: δ
169.12 (s, 2C, CO), 149.35 (s, 1C, CH2-CHdCH2), 148.27 (s,
2C, CH2-CHdCH2), 107.20 (s, 2C, CH2-CHdCH2), 106.30 (s,
1C, CH2-CHdCH2), 16.80 (2, 1C, CH2-CHdCH2, 2J PC ) 51.9
Hz), 15.39 (s, 2C, CH2-CHdCH2, 2J PC ) 5.8 Hz), 14.37 (s, 3C,
Gen er a tion of Ir (σ-a llyl)3(CO)3 (13). To a two-neck round-
bottom flask fitted with a balloon was added Ir(allyl)3 (0.088
g, 0.28 mmol), toluene-d8 (5 mL), and carbon monoxide (ca.
1.5 atm). Aliquots were removed from the reaction flask
periodically for NMR and IR analysis. It was found that all
the Ir(allyl)3 had been consumed after 26 h. NMR analysis
revealed that conversion to Ir(σ-allyl)3(CO)3 was quantitative.
Vacuum workup of the reaction mixture resulted in formation
of a mixture of 13 and presumably Ir(σ-allyl)2(π-allyl)(CO)2
(14a ). Formation of the latter complex is reversible (no other
products are observed by NMR), and addition of CO results
in clean conversion of 14a back to 13. 1H NMR: δ 6.16 (m,
3H, CH2-CHdCH2), 4.84 (ddt, 3H, anti-CH2-CHdCH2), 4.54
(dd, 3H, syn-CH2-CHdCH2), 2.01 (d, 6H, CH2-CHdCH2). 13C-
{1H} NMR: δ 163.77 (s, CO), 147.34 (s, CH2-CHdCH2), 109.67
(s, CH2-CHdCH2), 15.41 (s, CH2-CHdCH2). IR (toluene):
2067(vs), 2121(m) cm-1 (s), ν(CtO); 1622 cm-1 (vw), ν(CdC).
1H NMR for 14a : δ 6.34 (m, 2H, CH2-CHdCH2), 4.90 (ddt,
2H, syn-CH2-CHdCH2), 4.59 (ddt, 2H, anti-CH2-CHdCH2),
4.03 (m, 1 H, CH2CHCH2), 3.75 (ddt, 2H, syn-CH2CHCH2), 1.9
(obscured m, 4H, CH2-CHdCH2), 1.55 (ddt, 2H, anti-CH2-
CHCH2). IR (neat): 2018(vs), 2054(m) cm-1, ν(CtO).
1
PMe3, J PC ) 30.7 Hz). 31P{1H} NMR: δ -61.38. IR (ben-
zene): 2011 (m, br), 2065 (vs) cm-1, ν(CtO).
Sin gle-Cr ysta l X-r a y Diffr a ction Stu d ies. Single-crystal
X-ray diffraction experiments were performed on a Bruker P4/
CCD/PC diffractometer with graphite-monochromated Mo KR
radiation (λ ) 0.71073 Å). Diffraction data were refined using
SHELXTL PC.12 Single crystals of 6, 7, 8, 9a , and 12 were
grown from concentrated toluene solutions at room tempera-
ture. Crystals were coated in mineral oil and mounted on a
glass fiber at -70 °C.
A hemisphere of data was collected using a combination of
φ and ω scans, with 30 s frame exposures and 0.3° frame
widths. Data collection and initial indexing and cell refinement
were performed using SMART13 software. Frame integration
and final cell parameter calculations were carried out using
SAINT14 software. The data were corrected for absorption
using the SADABS15 program. Decay of reflection intensity was
not observed. The crystal and refinement parameters are listed
in Table 1. Average bond distances and angles are listed in
Table 2.The structures were solved using difference Fourier
techniques. The initial solutions revealed Ir and the majority
of all other non-hydrogen positions. The remaining atomic
positions were determined from subsequent Fourier syntheses.
All hydrogen atom positions were fixed in ideal positions. The
C-H distances were fixed at 0.93 (aromatic), 0.96 (methyne),
0.97 (methylene), and 0.98 Å (methyl). The hydrogen atoms
were refined using a riding model, with their isotropic tem-
perature factors set to 1.2 (aromatic, methyne, methylene) or
1.5 (methyl) times the isotropic U of the carbon atom they were
bonded to.
Gen er a tion of Rh (σ-a llyl)2(π-a llyl)(CO)2 (14b). Addition
of CO to Rh(allyl)3 in the same manner as for 13 generated
Rh(σ-allyl)2(π-allyl)(CO)2, 14b. Vacuum workup of the solution
led to significant decomposition and afforded as yet unchar-
acterized Rh products. GC analysis of a portion of this reaction
1
revealed that allylaldehyde had been generated. H NMR: δ
5.79 (m, 2H, CH2-CHdCH2), 4.94 (ddt, 2H, syn-CH2-CHd
CH2), 4.84 (ddt, 2H, anti-CH2-CHdCH2), 4.65 (m, 1 H,
CH2CHCH2), 3.60 (ddt, 2H, syn-CH2CHCH2), 2.64 (d, 4H,
CH2-CHdCH2), 1.53 (ddt, 2H, anti-CH2CHCH2). 13C{1H}
NMR: δ 191.02 (s, CO), 131.48 (m, CH2CHCH2), 118.9
(overlapping m, CH2-CHdCH2), 60.39 (m, CH2CHCH2), 47.1
(overlapping m, CH2-CHdCH2), 30.14 (overlapping m, CH2-
CHdCH2). IR (toluene): 1988 (vs), 2050 (m) cm-1, ν(CtO);
1618 cm-1 (vw), ν(CdC).
Gen er a tion of Ir (σ-a llyl)3(CO)2P Me3 (15). An aliquot was
taken from the reaction used to generate 13 (roughly 1/3 of
the total volume; ca. 0.09 mmol). To this sample was added
PMe3 (46 mL, 0.45 mmol), and the sample was stirred for 1 h.
NMR analysis revealed that the conversion to 15 was quan-
titative. 1H NMR: δ 6.67 (m, 1H, CH2-CHdCH2), 6.31 (m,
(12) SHELXTL Version 5.1; Bruker Analytical X-Ray Systems: 6300
Enterprise Lane, Madison, WI 53719, 1997.
(13) SMART Version 4.210; Bruker Analytical X-Ray Systems: 6300
Enterprise Lane, Madison, WI 53719, 1996.
(14) SAINT Version 4.05; Bruker Analytical X-Ray Systems: 6300
Enterprise Lane, Madison, WI 53719, 1996.
(15) SADABS; George Sheldrick, University of Go¨ttingen, Germany,
1996.

300 Organometallics, Vol. 20, No. 2, 2001
J ohn et al.
tion of the original synthesis.10 Reaction of IrCl3(tht)3
with allyllithium in benzene affords Ir(allyl)3 in >80%
NMR yield (eq 1).
Ta ble 2. Aver a ge Bon d Dista n ces (Å) a n d An gles
(d eg) for M(a llyl) Com p lexesa ,b
nuclei
6
7
8
9a
12
M-C1
2.167[4] 2.190[2]
2.198[4] 2.22[1]
2.160[5]
NA
NA
1.25[1]
136[2]
IrCl3(tht)3 + 3 LiC3H5 benzene8
M-C4(C6) 2.217[5] 2.236[2]
NA
NA
2.23[1]
2.18(1)
25 °C
M-C5
C2-C3
2.145[5] 2.186(2)
Ir(C3H5)3 + 3 LiCl + 3 tht (1)
1.36[1]
1.318[4]
127.2(3)
1.311[4] 1.34[2]
127.0[5] 126[1]
C1-C2-C3 130[1]
C4-C5-C6 119.2[5] 121.1[3]
NA
122.1(7) NA
Attempts to isolate the product from this method have
not provided Ir(allyl)3 in comparable yields (less than
30%); however, use of more volatile solvents such as
Et2O afford sublimed, microcrystalline product in over
47% yield.
M-L
2.311[1] 2.2783[6] 2.324[1] 2.282[1] 1.979[5]
a
Esd’s (standard deviations) are given in parentheses. Values
b
in square brackets are arithmetic means of esd’s. All σ- and
π-allyl bond distances and angles have been averaged, with σ-allyls
being designated by C1-3 and π-allyls by C4-6.
The preparation of Rh(allyl)3 can be achieved via the
same method utilized in the synthesis of the Ir analogue,
although the isolated yield is much higher (ca. 86%).
This method is thus comparable with the synthesis
reported by Powell and Shaw from RhCl3(H2O)x and
allylMgBr.7
Ir (σ-a llyl)(π-a llyl)2(P P h 3) (6). Crystals of 6 were shown
to adopt the space group P1h with two molecules observed in
the asymmetric unit. Refinement revealed that the central
carbon (C2) of the σ-allyl group was disordered over two sites.
The disorder was modeled by refining the carbon atoms with
a shared free variable that assigned the occupation of the sites
on the basis of the observed electron density.
Ir (σ-a llyl)2(π-a llyl)(P h 2P C6H4P P h 2) (7). Crystals of 7 were
shown to adopt the space group P1h. Very disordered toluene
molecules were observed in the lattice of 7, and the program
PLATON/SQUEEZE was used to remove the toluene solvent
electron density.16 The solvent volume modeled for 7 consisted
of two sites centered at 0.5, 0.75, 0.75 and 0.5, 0.25, 0.25 in
the unit cell. The solvent volume modeled was 369 Å3, and
100 electrons per unit cell were removed.
Ir (σ-a llyl)3[P h 2P (CH2)2P (P h )(CH2)2P P h 2] (8). Crystals of
8 were shown to adopt the space group P21/n. Like 7, severely
disordered toluene was observed in the lattice of 8. The solvent
volume modeled for 8 consisted of two sites centered at 0.0,
0.5, 0.0 and 0.0, 1.0, 0.5 in the unit cell. The solvent volume
modeled was 1056 Å3, and 456 electrons per unit cell were
removed. Further refinement of 8 revealed that the terminal
carbon atom (C9) of the σ-allyl group was disordered over two
sites. The disorder was modeled by refining the carbon atoms
with a shared free variable that assigned the occupation of
the sites on the basis of the observed electron density.
Ir (σ-a llyl)2(π-a llyl)(P Me3)2 (9a ). Crystals of 9a were shown
to adopt the space group P21/n. Refinement revealed that the
central carbon (C8) of the σ-allyl group was disordered over
two sites. The disorder was modeled by refining the carbon
atoms with a shared free variable that assigned the occupation
of the sites on the basis of the observed electron density.
Ir (σ-a llyl)3(CtN-2,6-Me2C6H3) (12). The refinements of 12
Rea ction s of M(a llyl)3 w ith N, O, a n d S d on or s.
The tris(allyl) complexes were surprisingly unreactive
toward typical N, O, and S donor ligands. Pyridine,
trialkylamines, anilines (aniline, 3,5-bis(trifluorometh-
yl)aniline), diimines (bipyridine, o-phenanthroline), and
acetonitrile all showed no irreversible binding after days
at 25 °C. Similar results were obtained for ethers (Et2O,
t
thf), alcohols (MeOH, BuOH, PhOH, Ph3SiOH), and
acetone. Even “soft” S donors such as phenylthiol,
thioethers (tht, [PhS(CH2)]2), and thiocamphor were
unreactive. However, addition of tolylisothiocyanate to
Ir(allyl)3 provided Ir(σ-allyl)(π-allyl)2(SCN-Tol) (5) in
moderate yield after stirring for 3 days at 20 °C. No
reaction was observed between Rh(allyl)3 and Tol-NCS.
The composition of 5 was established by 1H and 13C
NMR analysis, which revealed the presence of one
σ-allyl and two inequivalent π-allyl groups in an unsym-
metrical structure. Although thiocamphor may be too
occurred in
observed.
a
straightforward manner. No disorder was
sterically bulky to give a stable adduct, “soft” donors
appeared to be favored by the tris(allyl) complexes, so
we investigated their reactivity toward a variety of
readily accessible phosphorus donors.
Resu lts
M(a llyl)3 Syn th esis. The literature procedure for
preparing Ir(allyl)3 (3) from Ir(acac)3 (acac ) acetylac-
etonate) and (allyl)MgCl affords the desired product in
ca. 20% yield.4 Given the similarly low yield for the
preparation of Ir(acac)3 from commercial precursors, a
new synthetic approach was called for. Wilkinson and
co-workers have reported that the homoleptic methyl
complex [Li(tmeda)]3[IrMe6] can be prepared in over
90% yield from the interaction of IrCl3(tht)3 (1; tht )
tetrahydrothiophene) with methyllithium.17 We have
found that crystalline mer-IrCl3(tht)3 can be prepared
from IrCl3(H2O)x in ca. 90% yield via a slight modifica-
P -Don or s. Treatment of isolated Ir(allyl)3 with PPh3
in toluene gives the 1:1 adduct Ir(σ-allyl)(π-allyl)2(PPh3)
6 in high yield.8 Alternatively, the hexane filtrate from
the synthesis of Ir(allyl)3 can be used directly to prepare
6 with comparable efficiency. Complex 6 did not react
further with excess PPh3, even at 100 °C. The infrared
spectrum of 6 displays a characteristic CdC stretch in
the 1600-1610 cm-1 region due to the σ-allyl group.18
The 1H and 13C NMR spectra of 6 established the
presence of one σ-allyl and two inequivalent π-allyl
groups (Figure 1), even at 100 °C. The PPh3 ligand is
located trans to a methylene group of one of the π-allyl
groups, as evidenced by the large P-C coupling constant
(16) Spek, A. L. Acta Crystallogr. 1990, A46, C-34.
(17) (a) Hay-Motherwell, R. S.; Wilkinson, G.; Hussain, B.; Hurst-
house, M. B. J . Chem. Soc., Chem. Commun. 1989, 1436. (b) Hay-
Motherwell, R. S.; Wilkinson, G.; Hussain-Bates, B.; Hursthouse, M.
B. Polyhedron 1990, 9, 2071.
(18) Nakamoto, K. Infrared and Raman Spectra of Inorganic and
Coordination Compounds; Wiley: New York, 1997.

Reactivity of M(allyl)3 (M ) Rh, Ir)
Organometallics, Vol. 20, No. 2, 2001 301
F igu r e 1. 1H NMR of 6.
(2J PC ) 38 Hz). This was further confirmed by an X-ray
crystallographic study of 6 (Figure 2), which shows the
F igu r e 2. Thermal ellipsoid representation of 6 shown at
the 50% probability level. Hydrogen atoms have been
omitted for clarity. Only 1 of the 2 molecules in the
asymmetric unit is shown. Only one orientation of the
disordered allyl group is shown (see the Experimental
Section).
π-allyl ligand trans to P is symmetrically bound with
typical Ir-C bond distances19 (Ir-CH ) 2.145[5], Ir-
CH2 ) 2.217[5] Å). The other π-allyl ligand is unsym-
metrically bound as a result of the trans influence of
the σ-allyl {C(1)-Ir-C(7) ) 161.6[2]°, Ir-C(7) ) 2.268-
[4] vs Ir-C(9) ) 2.196[4] Å}. The upfield shift of the
methylene protons induced by the trans P (or S) and
σ-allyl groups allowed us to assign all 15 inequivalent
allyl proton resonances (see Experimental Section) for
both 5 and 6.
C6H4) 7 in good yield. Complex 7 has inequivalent σ-allyl
groups as determined by NMR analysis, even at 100 °C,
with the two phosphorus donors trans to one methylene
of the π-allyl (2J PC ) 39.4 Hz) and to one of the σ-allyl
groups (2J PC ) 79.5 Hz), respectively. The cis geometry
of the σ-allyl groups was confirmed by X-ray structural
analysis (Figure 3). Like 6, the structure of 7 reveals
that there is a significant trans influence by one of the
σ-allyl groups on the single π-allyl group [C(1)-Ir-C(7)
) 162.1(1)°, Ir-C(7) ) 2.270(2) vs Ir-C(9) ) 2.202(3)
Å].
Addition of the chelating bis(phosphine) 1,2-(PPh2)2-
C6H4 to Ir(allyl)3 affords Ir(σ-allyl)2(π-allyl)(1,2-(PPh2)2-
(19) Manger, M.; Wolf, J .; Teichert, M.; Stalke, D.; Werner, H.
Organometallics 1998, 17, 3210.

302 Organometallics, Vol. 20, No. 2, 2001
J ohn et al.
F igu r e 3. Thermal ellipsoid representation of 7 shown at
the 50% probability level. Hydrogen atoms have been
omitted for clarity.
F igu r e 5. Thermal ellipsoid representation of 9a shown
at the 50% probability level. Hydrogen atoms have been
omitted for clarity. Only one orientation of the disordered
σ-allyl group is shown (see the Experimental Section).
temperature below their decomposition points. Con-
versely, the single 31P resonance for each of 9a and 9b
F igu r e 4. Thermal ellipsoid representation of 8 shown at
the 50% probability level. Hydrogen atoms have been
omitted for clarity. Only one orientation of the disordered
allyl group is shown (see the Experimental Section).
did not undergo sufficient broadening at -80 °C for us
to estimate the activation barrier for this fluxional
process. The trans configuration in 9a ,b was further
confirmed through single-crystal X-ray diffraction analy-
sis (Figure 5).20 Attempts to isolate a mono-PMe3
complex have not been successful. Addition of 1 equiv
of PMe3 to Ir(allyl)3 results in the formation of a mixture
containing the parent tris(allyl)iridium, 9a , and pre-
sumably Ir(σ-allyl)(π-allyl)2(PMe3) 10. A single 31P NMR
resonance for 10 appears at -44.5 ppm, and the 1H
NMR spectrum is similar to that observed for the PPh3
analogue 6. Furthermore, addition of a 10-fold excess
of PMe3 provides only 9a even after heating (85 °C, 24
h). Attempts to prepare the Rh analogues of complexes
6-8 were unsuccessful. In the case of triphenylphos-
phine, we were able to isolate the previously reported
Rh(I) species, Rh(π-allyl)(PPh3)2, in rather low yield (less
than 10%). This compound has been prepared previously
from a refluxing methanol solution of Rh(allyl)3 and
PPh3 in 85% yield.7 A similar reaction with P(OPh)3
Addition of the tridentate phosphine PhP(CH2CH2-
PPh2)2 to Ir(allyl)3 gives tris(σ-allyl) complex 8. This
triphos complex has two sets of σ-allyl groups in a 1:2
ratio, consistent with Cs symmetry. X-ray structural
analysis of 8 (Figure 4) reveals that there is a significant
difference in the bond distances associated with the
terminal and central Ir-P bonds (2.339[1] and
2.294(1) Å, respectively). This difference is not observed
in the corresponding σ-allyl groups, and the esd’s on the
Ir-C bonds preclude further discussion. Less bulky
phosphorus ligands such as PMe3 and P(OPh)3 react
with Ir(allyl)3 to afford IrL2(σ-allyl)2(π-allyl) (9a ,b) in
high yield. NMR analysis of 9a ,b indicated that, in
contrast with 7, the inequivalent σ-allyl ligands are
trans to one another. In particular, we observe overlap-
ping methyne proton resonances for the σ-allyl ligands
at δ 5.8 and significant phosphorus coupling to both
methylene carbons of the π-allyl group (av J CP ) 32.6
Hz). Proton resonances due to the inequivalent σ-allyl
ligands in 9a ,b all broaden at higher temperatures, but
we were unable to reach a well-defined coalescence
(20) While the structural analysis of 9a confirmed the proposed
structure, the refinement was only fair, as was that of the phosphite
complex 9b. Full details for 9a ,b are provided in the Supporting
Information.

Reactivity of M(allyl)3 (M ) Rh, Ir)
Organometallics, Vol. 20, No. 2, 2001 303
gave a higher yield of analogous 11, in addition to the
C6 organic products, hexane, hexene, and 2-methylpen-
tane. In contrast, Rh(allyl)3 and the powerful donor
PMe3 provided the trivalent Rh complex of Rh(σ-allyl)2-
(π-allyl)(PMe3)2 (9c) in good yield. The formulation of
9c was based on the similarity of its NMR spectrum to
that of 9a . There was no evidence to suggest the
formation of any Rh(I) species. Specifically, 31P NMR
analysis reveals that the rhodium-phosphorus coupling
is approximately 150 Hz (cf. 200 Hz for Rh(π-allyl)-
(PPh3)2), indicative of a Rh(III) species.21 The PMe3
complexes thus constitute the only isostructural pair of
Ir/Rh-tris(allyl) adducts observed to date.
C-Don or s. As with phosphorus donors, we observe
significant reactivity differences between tris(allyl)-
rhodium and -iridium with C-donor ligands. Addition
of xylylisocyanide to Ir(allyl)3 provides Ir(σ-allyl)3(CN-
2,6-Me2C6H3)3 (12) in high yield. The geometry of the
F igu r e 6. Thermal ellipsoid representation of 12 shown
at the 50% probability level. Hydrogen atoms have been
omitted for clarity.
(CH2dCHCH2C(dNR)C(dNR)CH2CHdCH2; R ) cyclo-
t
hexyl, Bu, xylyl) that presumably forms via insertion
of an isonitrile into the Rh-Callyl bond followed by
reductive elimination.
Addition of carbon monoxide to M(allyl)3 generates
Ir(σ-allyl)3(CO)3 (13) and Rh(σ-allyl)2(π-allyl)(CO)2 (14b),
respectively. Vacuum workup of 13 results in the
formation of a mixture of 13 and Ir(σ-allyl)2(π-allyl)(CO)2
1
(14a ), as evidenced by the similarity of the H NMR to
14b. The formation of 14a appears to be reversible, as
CO exposure regenerates pure 13 in nearly quantitative
yield. Removal of solvent under vacuum from product
14b results in the insertion of CO into the Rh-allyl
bond followed by reductive elimination to yield allyl-
aldehyde and as yet uncharacterized organometallic
products.
Studies of the reactivity of the aforementioned car-
bonyl complexes are currently underway. Preliminary
studies indicate that both 13 and 14b are reactive
toward phosphines. For instance, 13 reacts with PMe3
to provide the monophosphine adduct Ir(σ-allyl)3-
(CO)2PMe3 (15) in quantitative yield. This reactivity is
tris-isocyanide complex is fac, as evidenced by the lone
set of σ-allyl resonances in the H NMR. Furthermore,
1
IR analysis reveals two CN stretches at 2160 and 2104
cm-1 (presumably of a1 and e symmetry, respectively)
consistent with pseudooctahedral C3v symmetry.18 This
is borne out by the structural analysis of 12 (Figure 6),
which shows the fac-geometry with essentially linear
M-CNR fragments ( (Ir-CtN) ) 177.3[4]°; (CtN-
C
xylyl) ) 173.7[5]°). Also of note is the significant
difference in the Ir-Callyl bond distances between com-
pound 12 (av 2.160[5] Å) and triphos derivative 8 (av
2.198[4] Å). This difference directly reflects the superior
σ-donor strength and trans influence of the triphos
ligand relative to isocyanides. Addition of isocyanides
to Rh(allyl)3 results in the initial formation of a tris-σ-
1
allyl complex (as observed by H NMR), which quickly
(less than 5 min) decomposes at room temperature via
reductive elimination to provide as yet uncharacterized
metal-containing products. GC/MS analysis of the or-
ganic products revealed the formation of an R-diimine
(21) Verkade, J . G.; Mosbo, J . A. In Phosphorus-31 NMR Spectoscopy
in Stereochemical Analysis; Organic Compounds and Metal Complexes;
Verkade, J . G., Quin, L. D., Eds.; VCH: New York, 1987; Chapter 13,
and references therein.
consistent with the fact that 13 reversibly loses a CO
to form 14a . Attempts to isolate a mixed carbonyl/
phosphine complex from 14b have so far been unsuc-

304 Organometallics, Vol. 20, No. 2, 2001
J ohn et al.
cessful. It appears that addition of PMe3 to 14b results
in a rapid equilibrium between 14b and a product
bulkier tricyclohexylphosphine complexes, allyl ex-
change was observed for the Pd complex, but not for the
Pt analogue, even at 90 °C.28 The steric bulk of 6 is
apparent from both its molecular structure in the solid
state and its reluctance to add a second phosphine.
The formation of isomeric bis(ligand) tris(allyl)iridium
adducts can also be attributed to steric constraints.
While the 1H NMR spectrum of the Rh and Ir dicarbonyl
adducts 14a ,b are indicative of Cs symmetry, the bis-
(PMe3) complexes 9a ,c both exhibit unsymmetrical
structures, due presumably to hindered rotation about
the M-C bonds of the σ-allyl groups. With the larger
arylated bis(phosphine), the trans structure is presum-
ably no longer sterically viable for 7 and the σ-allyl
groups adopt the cis orientation.
The redox stability of the heavier group 9 tris(allyl)
complexes parallels that of their group 10 homologues.
Pd(σ-allyl)(π-allyl)(PPh3) eliminates 1,5-hexadiene above
0 °C to give the monovalent, allyl-bridged dimer,
[Pd(PPh3)(µ-allyl)]2,27 and excess phosphine gives
Pd(PPh3)4. In contrast, ligand-induced hydrocarbon
elimination from Pt(allyl)2 has not been reported. In our
work, only the powerful donor phosphine PMe3 is
capable of stabilizing trivalent tris(allyl)rhodium com-
plexes; the dicarbonyl adduct can be generated in
solution, but removal of the solvent in vacuo induces
elimination of allylaldehyde.
1
analogous to 15, as evidenced by H and 31P NMR.
Discu ssion
Although allyllithium is not typically the reagent of
choice for preparing transition metal allyl complexes,1e
we have been unable to obtain high yields of tris(allyl)-
iridium (3) using magnesium or tin reagents. The
M(allyl)3 product is particularly sensitive to heat as a
solid or in concentrated solution, although we have not
characterized its decomposition reaction or the resulting
insoluble brown residue in detail. As a result, the crude
hexane extracts of 3 are often the preferred starting
material for investigating its chemistry. Similar prob-
lems are not encountered for the Rh analogue, which
can be purified by sublimation in high yield, although
solutions cannot be heated much above 60 °C without
noticeable decomposition.
The ligand addition chemistry of M(allyl)3 is remark-
ably selective for “soft” donors. Such is not the case for
the 16 e- group 10 bis(allyl) complexes, for which the
NMR spectra are highly solvent dependent due to facile
dynamic π-σ allyl conversion in donor solvents.22
Pd(allyl)2 reacts readily with alcohols, and diethylamine,
for example, gives the amido-bridged dimer.23 While
Rh(allyl)3 reacts readily with the O-H functions of solid
metal oxides, no reaction is observed with Ph3SiOH in
the absence of an ancillary donor ligand such as a
phosphine.24
Con clu sion
Reactions of tris(allyl)iridium with “soft” donor ligands
afford a variety of trivalent adducts all containing σ-allyl
groups. The increased reactivity of the Ir-C bond in
these products (relative to Ir(allyl)3) will be especially
useful for our future efforts to prepare bis(allyl)iridium-
(III) moieties bonded to metal oxide surfaces.24 We
foresee that the increased thermal and redox stability
of these supported organoiridium centers may lead to
useful catalytic properties for hydrocarbon functional-
ization reactions.
The structural rigidity of Ir(σ-allyl)(π-allyl)2(PPh3) (6)
in solution is likely a result of strong Ir-C bonds and
the saturated electron count, i.e., a second π-σ allyl
conversion would be required in the transition state for
allyl exchange. For the 16 e- group 10 homologues,
M(σ-allyl)(π-allyl)(PMe3), no additional coordination site
is required and allyl proton exchange barriers are
reported to be 9.6 ( 1.5 and 21.1 ( 1.5 kcal/mol for Pd
and Pt, respectively.25 Allyl exchange is also observed
for both PPh3 analogues,26,27 whereas in the case of the
Ack n ow led gm en t. We thank the Department of
Energy’s Laboratory Directed Research and Develop-
ment (LDRD) program for financial support and Dr.
David C. Smith for preliminary work on the Ir(allyl)3
synthesis.
(22) J olly, P. W.; Wilke, G. The Organic Chemistry of Nickel, Vol1;
Academic Press: New York, 1974.
(23) Hegedus, L. S. In The Chemistry of the Metal-Carbon Bond;
The Nature and Cleavage of Metal-Carbon Bonds; Hartley, F. R.,
Patai, S., Eds.; J ohn Wiley and Sons: New York, 1985; Vol. 2, Chapter
6D, and references therein.
(24) J ohn, K. D.; Baker, R. T.; Sattelberger, A. P. Unpublished
results.
(25) Henc, B.; J olly, P. W.; Salz, R.; Stobbe, S.; Wilke, G.; Benn, R.;
Mynott, R.; Seevogel, K.; Goddard, R.; Kruger, C. J . Organomet. Chem.
1980, 191, 449.
Su p p or tin g In for m a tion Ava ila ble: Tables of all atomic
coordinates, anisotropic thermal parameters, bond lengths and
bond angles, and hydrogen atom coordinates for 6, 7, 8, 9a ,
and 12. This material is available free of charge via the
Internet at http//pubs.acs.org.
(26) Bertani, R.; Carturan. G.; Scrivanti, A.; Wojcicki, A. Organo-
metallics 1985, 4, 2139.
(27) J olly, P. W.; Kru¨ger, C.; Schick, K.-P.; Wilke, G. Z. Naturforsch.,
Teil B 1980, 35b, 926.
OM000734K
(28) Ku¨hn, A.; Werner, H. Chem. Ber. 1980, 113, 2308.