Synthesis of 1,4-dichloro-1,3-butadienes by rhodium complex catalyzed reaction of terminal alkynes with trichloroacetyl chloride

Article abstract of DOI:10.1021/jo902088p

(Chemical Equation Presented) Chlorinative dimerization of terminal alkynes with trichloroacetyl chloride as chlorine donor proceeds in the presence of rhodium catalysts to give (Z,Z)-1,4-dichloro-1,3-butadienes stereoselectively. Ligand screening has rev

Full text of DOI:10.1021/jo902088p

pubs.acs.org/joc
Synthesis of 1,4-Dichloro-1,3-butadienes by Rhodium Complex Catalyzed
Reaction of Terminal Alkynes with Trichloroacetyl Chloride
Taigo Kashiwabara, Kouichirou Fuse, Takeshi Muramatsu, and Masato Tanaka*
Chemical Resources Laboratory, Tokyo Institute of Technology, Nagatsuta, Midori-ku, Yokohama
226-8503, Japan
m.tanaka@res.titech.ac.jp
Received September 29, 2009
Chlorinative dimerization of terminal alkynes with trichloroacetyl chloride as chlorine donor
proceeds in the presence of rhodium catalysts to give (Z,Z)-1,4-dichloro-1,3-butadienes stereoselec-
tively. Ligand screening has revealed that reactions using sterically bulky and electron-donating
ligands like trimesitylphosphine are high yielding. The reaction is compatible with a range of
functional groups to give the title compounds nearly quantitatively in most cases. A mechanistic
possibility involving coupling of β-chloroalken-1-yl intermediate has been discussed.
Introduction
thiophenes,8 pyrrole,8b siloles and other group 14 conge-
ners,9 phospholes,10 and other pnictogen congeners.11
As for the synthesis of 1,4-dihalo-1,3-butadienes, two
newer general procedures are known, in addition to a
plethora of those for particular structures. One involves
cyclization of two molecules of alkynes with early transition
metal complexes like Zr/Ti complexes forming metalacyclo-
pentadienes followed by quenching with appropriate halo-
gens.12 This procedure, although the yield is not always high,
has been most frequently used to prepare diiodo compounds
in the synthetic application for siloles, phospholes, and other
similar compounds,9-11 due to their high reactivity. The
other is halogenative coupling of terminal alkynes with
appropriate halogen sources like iodine or up to 4-fold excess
of cupric halides in the presence of a catalytic quantity of
platinum or palladium halide.13 However, these methods are
not free from drawbacks. The former requires a stoichio-
metric quantity of Zr or Ti reagent, the metal in which is not
retained in the final products, and the latter with palladium
catalyst is a heterogeneous reaction, which may hamper
utilization in a large scale synthesis.
Since 1,3-dienes are extremely useful intermediates to
synthesize, for instance, biologically active heterocyclic com-
pounds, a wide range of 1,3-diene compounds have been
documented and numerous synthetic methods have been
developed.1 Among the huge family of 1,3-butadienes,
however, the structural variation of 1,4-dihalo-1,3-buta-
dienes is very limited, presumably because synthetic methodo-
logies have not been widely explored although these com-
pounds are also potentially useful in organic synthesis.
Indeed, use of these compounds can be seen in the synthesis
of conjugated polyenes,2 benzene derivatives,3 naphtha-
lenes,4 cyclopentadiene derivatives,5 pyridines,6 furans,7
(1) Luh, T.-Y.; Wong, K.-T. Synthesis 1993, 349.
€
(2) (a) Kiehl, A.; Eberhardt, A.; Mullen, K. Liebigs Ann. 1995, 223. (b)
Yamamoto, T.; Ohya, K.; Kobayashi, K.; Okamoto, K.; Koie, S.; Fukumoto,
H.; Koizumi, T.-a.; Yamaguchi, I. Macromolecules 2009, 42, 3207.
(3) Fukuhara, K.; Takayama, Y.; Sato, F. J. Am. Chem. Soc. 2003, 125,
6884.
(4) Wang, Z.; Wang, C.; Xi, Z. Tetrahedron Lett. 2006, 47, 4157.
(5) (a) Chambers, R. D.; Greenhall, M. P. J. Chem. Soc., Chem. Commun.
1990, 1128. (b) Hu, Q.; Wang, C.; Li, D.; Xi, Z. Org. Biomol. Chem. 2007, 5, 2114.
(6) Wang, C.; Wang, Z.; Liu, L.; Wang, C.; Liu, G.; Xi, Z. J. Org. Chem.
2006, 71, 8565.
(7) Briscoe, M. W.; Chambers, R. D.; Mullins, S. J.; Nakamura, T.;
Vaughan, J. F. S. J. Chem. Soc., Perkin Trans. 1 1994, 3119.
(8) (a) Geering, E. J. J. Org. Chem. 1959, 24, 1128. (b) Briscoe, M. W.;
Chambers, R. D.; Mullins, S. J.; Nakamura, T.; Drakesmith, F. G. J. Chem.
Soc., Chem. Commun. 1990, 1127. (c) Itsuda, H.; Kimura, S. (Seitetsu Kagaku
Co., Ltd.) Jpn. Kokai Tokkyo Koho JP 1990048575 A2, 1990.
(9) (a) Bankwitz, U.; Sohn, H.; Powell, D. R.; West, R. J. Organomet.
Chem. 1995, 499, C7. (b) Yamaguchi, S.; Jin, R.-Z.; Tamao, K.; Sato, F. J.
Org. Chem. 1998, 63, 10060. (c) Ura, Y.; Li, Y.; Tsai, F.-Y.; Nakajima, K.;
Kotara, M.; Takahashi, T. Heterocycles 2000, 52, 1171. (d) Sanji, T.; Mori,
T.; Sakurai, H. J. Organomet. Chem. 2006, 691, 1169. (e) Wang, C.; Luo, Q.;
Sun, H.; Guo, Z.; Xi, Z. J. Am. Chem. Soc. 2007, 129, 3094.
(10) (a) Turcitu, D.; Nief, F.; Ricard, L. Chem.;Eur. J. 2003, 9, 4916. (b)
Nief, F.; de Borms, B. T.; Ricard, L.; Carmichael, D. Eur. J. Inorg. Chem.
2005, 637.
(11) Ashe, A. J. III; Al-Ahmad, S.; Pilotek, S.; Puranik, D. B.; Elschenbroich,
C.; Behrendt, A. Organometallics 1995, 14, 2689.
(12) (a) Negishi, E.-i.; Cederbaum, F. E.; Takahashi, T. Tetrahedron Lett.
1986, 27, 2829. (b) Yamaguchi, S.; Jin, R.-Z.; Tamao, K.; Sato, F. J. Org.
Chem. 1998, 63, 10060.
(13) (a) Ananikov, V. P.; Mitchenko, S. A.; Beletskaya, I. P. Russ. J. Org.
Chem. 2002, 38, 636. (b) Li, J.-H.; Liang, Y.; Xie, Y.-X. J. Org. Chem. 2004,
69, 8125. (c) See also: Ananikov, V. P.; Mitchenko, S. A.; Beletskaya, I. P.
J. Organomet. Chem. 2001, 636, 175.
DOI: 10.1021/jo902088p
r
Published on Web 11/18/2009
J. Org. Chem. 2009, 74, 9433–9439 9433
2009 American Chemical Society

Kashiwabara et al.
JOCArticle
SCHEME 1. Rhodium-Catalyzed Synthesis of 1,4-Dichloro-1,
3-butadienes Starting with Terminal Alkynes with Trichloroacetyl
Chloride
TABLE 1. Rhodium-Catalyzed Reaction of 1-Octyne 1a with
Trichloroacetyl Chloride 2Aa
catalyst
temp (°C), convc yieldd
entry
system
solventb (%) (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Rh(acac)(CO)2 þ 1AsPh3
Rh(acac)(CO)(PPh3)
Rh(acac)(CO)2
Rh(acac)(CO)2 þ 1PMes3
Rh(acac)(CO)2 þ 2PMes3
RhCl(CO)(PPh3)2
130, E
130, E
130, E
130, E
130, E
130, E
130, E
130, E
130, E
130, E
130, E
130, E
130, E
130, E
99 20
98 11
99 30
99 52
99 65
98 19
98 30
nd 28
During the course of our study on the Rh(acac)(CO)-
(AsPh3)-catalyzed addition reaction of chloroacetyl chloride
with terminal alkynes,14 we found accidentally that the
reaction of 1-octyne 1a, run with trichloroacetyl chloride
2A used in place of chloroacetyl chloride (at 110 °C, Rh-
(acac)(CO)(AsPh3), 5 mol %), proceeded in a different
direction, forming (Z,Z)-7,10-dichlorohexadeca-7,9-diene
3a as undesired product in 63% yield (Scheme 1, R =
n-C6H13). Further study has provided us with an easy and
high-yielding access to (Z,Z)-1,4-dichloro-1,3-butadienes in
a homogeneous process (Scheme 1). The details will be
disclosed in this paper.
[RhCl(CO)2]2
[RhCl(CO)2]2 þ 4AsPh3
[RhCl(CO)2]2 þ 4SbPh3
[RhCl(CO)2]2 þ 4PMes3
[RhCl(CO)2]2 þ 4P(o-Tol)3
[RhCl(CO)2]2 þ 4P(1-Naph)3
[RhCl(CO)2]2 þ 4PMes2Ph
[RhCl(CO)2]2
98
6
98 62
98 40
98 39
99 50
100 54
þ 4P[2,4,6-iPr)3C6H2]3
[RhCl(CO)2]2 þ 4BSPe
[RhCl(CO)2]2
15
16
130, E
130, E
100 16
>99
100 36
95
5
þ 4P(p-ClC6H4)3
17
[RhCl(CO)2]2
þ 4P(p-MeOC6H4)3
130, E
18
19
20
21
22
23
24
25
RhCl(CO)(PMe3)2
130, E
130, E
130, E
130, E
130, E
130, E
100, P
100, D
100, T
100, C
130, E
130, C
130, C
8
Results and Discussion
[RhCl(CO)2]2 þ 4PCy3
[RhCl(CO)2]2 þ 2dppe
[RhCl(CO)2]2 þ 2dppp
RhCl(CO)(dppb)
[RhCl(CO)2]2 þ 2dppf
[RhCl(CO)2]2 þ 4PMes3
[RhCl(CO)2]2 þ 4PMes3
[RhCl(CO)2]2 þ 4PMes3
[RhCl(CO)2]2 þ 4PMes3
[RhCl(CO)2]2 þ 4PMes3
[RhCl(CO)2]2 þ 4PMes3
[RhCl(CO)2]2 þ 4PMes3
100 29
98 27
98 23
Trial Experiments for High-Yielding Conditions. Although
the yield of 3a in the accidental discovery of the reaction was not
too low, trial experiments to search for higher-yielding condi-
tions, mainly through ligand and solvent screening, were run in
the presence of various rhodium catalyst systems (2 mol % with
respect to rhodium atom). The procedure using [RhCl(CO)2]2
in conjunction with P(2,4,6-trimethylphenyl)3 (PMes3) in a
chlorobenzene solution has proved satisfactory, as summarized
in Table 1.
43 14
>99 18
100 50
100 69
100 52
100 78
100 70
100 84
100 91
26
27
28f
29f
30f,g
aReaction conditions: 1a (1.0 mmol), 2A (1.0 mmol), rhodium com-
plex (2 mol % Rh), ligand (4 mol %), solvent (1 mL), in a 20 mL Schlenk
tube, 12 h. bE = ethylbenzene, P = propionitrile, D = 1,4-dioxane, T =
toluene, C = chlorobenzene. cDetermined by GC using n-tetradecane as
an internal standard. dDetermined by 1H NMR spectroscopy using
1,1,2,2-tetrachloroethane as an internal standard. eBSP = tris[(3,5-
Initial ligand screening carried out by using (acetylaceto-
nato)rhodium complexes at 130 °C in a 20 mL Schlenk tube
revealed that, as was observed in the addition of chloroacetyl
chloride with alkynes, AsPh3 appeared to be better perform-
ing than PPh3 (entries 1 and 2). However, plain Rh(acac)-
(CO)2 without any additional ligand (entry 3) was better
performing, and Rh(acac)(CO)2PMes3 was even better, in
particular, when 2 equiv of PMes3 was used (entries 4 and 5).
Similar trends were also observed with chlororhodium cata-
lyst systems. Thus, RhCl(CO)(PPh3)2 (entry 6) appears to be
less effective than plain [RhCl(CO)2]2 (entry 7) and the
[RhCl(CO)2]2-4AsPh3 system (entry 8), although better
than [RhCl(CO)2]2-4SbPh3 (entry 9). Among chlororho-
dium catalyst systems, [RhCl(CO)2]2-4PMes3 was the best
(entry 10). Other ligands, like P(o-Tol)3, P(1-Naph)3, and
PMes2Ph, which are more sterically demanding than PPh3
but less bulky than PMes3, displayed intermediate yields
(entries 11-13). P[2,4,6-(iPr)3C6H2]3, an even more bulky
ligand, furnished a somewhat low yield as compared with
PMes3 (entry 14). BSP, a seemingly bulky ligand of a bowl
shape,15 performed much worse than PMes3 and similar to
PPh3 (entry 15); this is presumably because the steric con-
gestion in the close proximity of the rhodium center is not so
serious. Among triarylphosphines (entries 6, 16, and 17),
a distinct electronic effect was also observed; the yield
dimesityl)phenyl]phosphine. Run in a 5 mL Schlenk tube. gReaction
time 3 h, 5 mol % Rh and 10 mol % ligand.
f
decreased in the order of P(p-MeOC6H4)3 > PPh3 > P(p-
ClC6H4)3, suggesting that a more donating ligand is better
performing, which is consistent with the highest yield
obtained with PMes3. The use of RhCl(CO)(PMe3)2 catalyst
ligated by more electron-donating PMe3 resulted in a very
poor yield (entry 18). However, RhCl(CO)(PCy3)2, ligated
by highly electron-donating but sterically more demanding
ligand, performed better than RhCl(CO)(PMe3)2, which
agrees with the foregoing steric effect of the ligand (entry
19). Chelating phosphines also gave intermediate yields,
more or less similarly to PPh3 (entries 20-23). Despite the
rather clear trends observed in the ligand effect study, the
origin of the ligand-dependent variation in the yield is
ambiguous at this time. This may be associated with a
sterically bulky ligand being able to hamper the formation
of dinuclear rhodium species from mononuclear species,
which are envisioned to be responsible for the catalysis
(vide infra).
Choice of a solvent also plays an important role. Screening
was carried out by using the [RhCl(CO)2]2 þ 4PMes3 catalyst
system at a lower temperature (100 °C) to more clearly
extract discernible effect. The results shown in Table 1
conclude chlorobenzene is the solvent of choice although
(14) Kashiwabara, T.; Fuse, K.; Hua, R.; Tanaka, M. Org. Lett. 2008, 10,
5469.
(15) Niyomura, O.; Iwasawa, T.; Sawada, N.; Tokunaga, M.; Obora, Y.;
Tsuji, Y. Organometallics 2005, 24, 3468. For design and synthesis of BSP
and related heteroatom compounds, see: Ohzu, Y.; Goto, K.; Kawashima,
T. Angew. Chem., Int. Ed. 2003, 42, 5741 and references cited therein.
9434 J. Org. Chem. Vol. 74, No. 24, 2009

Kashiwabara et al.
JOCArticle
1,4-dioxane is also better performing than toluene and
propionitrile (entries 24-27).
TABLE 2. Rhodium-Catalyzed Chlorinative Dimerization of Terminal
Alkynes with Trichloroacetyl Chloride 2Aa
Finally, we turned our attention to the size of the reaction
vessel. Most of the trial experiments were run at 130 °C,
which is higher than the boiling point of trichloroacetyl
chloride (114-116 °C). In a large reaction vessel, the
Cl3CCOCl reagent is probably enriched in the gas phase,
thus reducing its concentration in the solution. We assumed
that the use of a small-sized Schlenk tube could afford a
higher yield. Indeed, when a reaction in ethylbenzene solvent
using the [RhCl(CO)2]2 þ 4PMes3 catalyst system was run in
a 5 mL airtight Schlenk tube, the yield increased to 70%
(entry 28) as compared with 62% obtained in a 20 mL
Schlenk tube reaction (entry 10). This modification, when
applied to a reaction in chlorobenzene, offered 84% yield
(entry 29). By increasing the catalyst quantity from 2 to
5 mol % (with respect to rhodium atom), the yield was
further improved to 91% in a shorter reaction time (3 h) to
provide a satisfactory recipe (entry 30).
yield
(%)
entry
alkyne
R =
product
1
2
3
4
5
6
7
8
9
10
11
12
13
1a
1b
1c
1d
1e
1f
1g
1h
1i
n-hexyl
tert-Bu
3a
3b
3c
3d
88
75
3-chloropropyl
3-cyanopropyl
3-(methoxycarbonyl)propyl 3e
benzyl
trimethylsilyl
1-cyclohexenyl
Ph
98b
90
96
80
70
72
98
94
92
96
97
3f
3g
3h
3i
3j
3k
3l
1j
1k
1l
p-anisyl
p-tolyl
p-fluorophenyl
2-thienyl
1m
3m
Since only chlorine among the constituents in Cl3CCOCl
was found in the products, we occasionally scrutinized the
resulting mixture by GCMS to identify the fate of carbonyl
functionality in Cl3CCOCl, but all attempts failed. Another
approach, somewhat associated with this, is to examine the
possible use of other chlorine compounds. Under the opti-
mized conditions (entry 30, Table 1), the reactions with
C2Cl6, CCl4 and (COCl)2 afforded 3a in 41, 4, and 26%
yields, respectively, suggesting that the presence of the
carbonyl functionality in the starting chlorine sources is
not a prerequisite for the catalysis.
Besides rhodium catalysts, we also attempted to use
ruthenium compounds as catalyst. However, RuCl3, alone
or in combination with PPh3 or AsPh3, [RuCl2(p-cymene)]2,
and RuCl2(CO)2(PPh3)2 were all basically inactive (2 mol %
catalyst, 130 °C in ethylbenzene).
Synthetic Scope and Limitation. The satisfactory recipe
identified in the foregoing section could be applied success-
fully to a variety of terminal alkynes as summarized in Table 2.
Products 3b, 3i, 3k, and 4, documented in the literature,13b
were identified by comparison of the spectroscopic data with
those reported and the others also displayed satisfactory
spectroscopic data. However, to unambiguously confirm the
stereochemistry at least in one case, product 3f was analyzed
by X-ray crystallography, which verified (Z,Z)-configura-
tion as shown in Figure S1 (Supporting Information).
Besides 1-octyne, other aliphatic acetylenes, inclusive of
functionalized ones, such as tert-butylacetylene (1b), 5-chloro-
1-pentyne (1c), 5-hexynenitrile (1d), methyl 5-hexynoate (1e),
3-phenyl-1-propyne (1f), and trimethylsilylacetylene (1g), also
conform to the reaction procedure, without deterioration of the
functional group. Cyclohexenylacetylene (1h), a conjugated
enyne, participates in the reaction as well. Aromatic acetylenes
(1i-l) and 2-thienylacetylene (1m) are also good substrates
affording (near) quantitative yields, irrespective of the nature
of the substituents. However, 4-(tert-butyldimethylsiloxy)-
1-butyne (1n) and 3-methoxy-1-propyne (1o) appear to have
reacted in different directions to end up with a messy mixture of
unidentified products. Methyl propiolate (1p) and ferrocenyla-
cetylene (1q) do not afford the desired products at all. 4-Octyne
(1r), an internal acetylene, is totally inert.
aReaction conditions: A mixture of 1 (1.0 mmol), 2A (1.0 mmol),
[RhCl(CO)2]2 (0.025 mmol), and PMes3 (0.10 mmol) in chlorobenzene
(1.0 mL) was stirred for 3 h at 130 °C in a 5 mL Schlenk tube. bA trace
(<1%) of another isomer appears to have been formed as analyzed by
GCMS.
regio- and stereoisomeric butadiene was observed in these
catalytic reactions.
Extension of the procedure to the reaction of tribromoa-
cetyl bromide (2B) with 1-decyne (90 °C, 3 h, in a 5 mL
Schlenk tube), although full optimization had not been
made, revealed the formation of (Z,Z)-9,12-dibromoicosa-
9,11-diene 4 in 43% NMR yield. GC and GCMS analyses
suggested the formation of 1,2-dibromodecene as bypro-
ducts. Quite puzzling is that the 1,2-dibromodecene was a
mixture of two isomers in an approximately 1:1 ratio
(presumably E- and Z-isomers on the basis of GCMS
analysis), although only (Z,Z)-9,12-dibromoicosa-9,
11-diene 4 was formed as the sole dimeric product.16 The
formation of the isomeric 1,2-dibromodecenes might be
associated with partial involvement of external attack of
bromine generating (E)-β-bromoalkenyl intermediate. Such
an elemental process has been verified by Beletskaya and
co-workers in platinum-catalyzed iodinative dimerization of
terminal alkynes (vide infra).13a,c
Mechanistic Consideration. Li and co-workers have deve-
loped a closely related process to synthesize 1,4-dihalobuta-
diene compounds by using palladium halides as catalyst. The
mechanism they proposed comprises halopalladation with
an alkyne leading to XCRdCHPdX species (X = Cl, Br),
insertion of another alkyne molecule to generate (4-halo-1,
3-butadien-1-yl)PdX species, and sp2-C-X reductive elimi-
nation.13b We do not have any decisive evidence to exclude
the same type mechanism for the present rhodium-catalyzed
reaction. The formation of 1,2-dibromodecenes (vide supra)
as byproduct in the present study may be rationalized by
premature reductive elimination at the stage that follows the
similar halometalation process. However, as far as rhodium-
(III) phosphine complexes are concerned, sp2-C-X reduc-
tive elimination involved in the last step leading to the
What is rather amazing through the successful reactions is
the formation of the product as a single isomer; no other
(16) A similar observation in the palladium-catalyzed reaction of phenyl-
acetylene with CuBr2 has been reported by Li and co-workers. See ref 13b.
J. Org. Chem. Vol. 74, No. 24, 2009 9435

Kashiwabara et al.
SCHEME 2. Outline of Possible Mechanism
JOCArticle
dihalo-1,3-diene formation is envisioned to be a tricky
task.17 What is required with this in mind is to work out
another mechanism that does not proceed via successive
insertion of two alkyne molecules. An obvious candidate is
a radical mechanism. To look into this possibility, we ran
two types of experiments in the presence of a radical scaven-
ger or styrene as a probe. A reaction run in the presence of
TEMPO (0.5 mmol, 50 mol %) under the same conditions as
in entry 30 (Table 1) did not result in an appreciable change
in the yield of 3a (89% vs 91% observed in entry 30),
supporting that the formation of 3a does not involve radical
species. However, another reaction run under the same
conditions in the presence of styrene (1.0 mmol) resulted in
a complete conversion of styrene to polystyrene (Mn=880,
Mw=1580) and the yield of 3a decreased to 13% (vs 91%
observed in entry 30), while a control experiment, in which a
chlorobenzene solution of styrene and [RhCl(CO)2]2
þ
4PMes3 catalyst was heated in the absence of 1a and 2A
under the same conditions, did not polymerize styrene to an
appreciable extent (e5% conversion). We presume that
radical species can be generated under the catalytic condi-
tions but the species does not participate directly in the
catalysis forming dichlorobutadienes.
SCHEME 3. Attempted Oxidative Addition of Cl3CCOCl with
RhCl(CO)(PMe3)2
Another aspect worth noting is the high stereoselectivity
forming only Z,Z-isomers in the catalysis, which also
appears to exclude a radical mechanism. Configurational
instability of alkenyl radicals has been substantiated in quite
a few publications18 and utilized in synthetic application
since the work by Stork demonstrated.19 In view of these
precedents, dimerization of free RCCldCH• radical does not
appear to be a realistic route to dichlorobutadiene.
The elimination of possible mechanisms on the basis of the
foregoing consideration leaves a third mechanism via direct
conversion of two β-chloroalkenyl-metal bonds to the
dimer (Scheme 2). Note, however, that the “direct conver-
sion” covers a variety of mechanistic details (vide infra).
Scheme 2 illustrates only reductive elimination of bis-
(β-chloroalkenyl)rhodium species resulting from chlororho-
dation at two Cl-Rh bonds in one molecule with alkyne
(route 3-i followed by 3-ii) or disproportionation of two (β-
chloroalkenyl)rhodium species (route 4).
We presume that the catalysis is initiated by oxidative
addition of trichloroacetyl chloride with a chlororhodium(I)
complex (route 1), which is somehow followed by generation
of RhCl3Ln species such as RhCl3(CO)L2 (route 2), and
insertion of an alkyne takes place (route 3). As for the
oxidative addition, we have been unable, despite our best
efforts, to generate or isolate neat adducts of Cl3CC(O)-
RhCl2(CO)L2. Even Cl3CC(O)RhCl2(CO)(PMe3)2 ligated
by PMe3, a powerful stabilizing ligand for complexes of this
type,14,20 appears extremely labile (Scheme 3). Thus, an
attempted reaction of RhCl(CO)(PMe3)2 with Cl3CCOCl
(1.0 equiv) at 0 °C in CDCl3 furnished, after 15 min, a
mixture that displayed, in 31P NMR spectroscopy, two
doublets, major one centered at -1.27 ppm (1JRh-P=72.8
Hz) arising from known fac,cis-RhCl3(CO)(PMe3)2 (fac,cis-
5)21 and the other minor one at -0.91 ppm (1JRh-P=76.3
Hz), in addition to two more even minor doublets. However,
the doublet at -0.91 ppm, which we presume was arising
from Cl3CC(O)RhCl2(CO)(PMe3)2 disappeared after addi-
tional 1 h at the temperature, and the major doublet became
more intense. When the same reaction was run at room
temperature, only signals due to fac,cis-5 were observed.22
Another reaction run at -30 °C for 1 h provided a 31P NMR
spectrum similar to that obtained at 0 °C after 15 min. A
10 μL portion of the resulting mixture was evaporated
rapidly on a NaCl disk and was subjected to IR spectroscopy,
which showed an absorption band at 1681 cm-1, reasonable
for νCdO due to Cl3CC(O)Rh species. A third support for
Cl3CC(O)RhCl2(CO)(PMe3)2 has come from FAB-MS
measurement; to a solution of RhCl(CO)(PMe3)2 in CH2Cl2
cooled at 0 °C was added Cl3CCOCl (1.0 equiv) in front of a
mass spectrometer to instantaneously develop a red-orange
solution, and the mixture was injected immediately to the
spectrometer to furnish satisfactory HRMS m/z values with
the expected isotope distribution pattern for [M - 2Cl þ H]þ,
(17) (a) Kokubo, K.; Matsumasa, K.; Miura, M.; Nomura, M. J. Org.
Chem. 1996, 61, 6941. (b) Kampmeier, J. A.; Harris, S. H.; Rodehorst, R. M.
J. Am. Chem. Soc. 1981, 103, 1478.
(18) (a) Fessenden, R. W.; Schuler, R. H. J. Chem. Phys. 1963, 39, 2147.
(b) Sargent, G. D.; Browne, M. W. J. Am. Chem. Soc. 1967, 89, 2788. (c)
Jenkins, P. R.; Symons, M. C. R.; Booth, S. E.; Swain, C. J. Tetrahedron Lett.
1992, 33, 3543. (d) Galli, C.; Guarnieri, A.; Koch, H.; Mencarelli, P.;
Rappoport, Z. J. Org. Chem. 1997, 62, 4072.
(19) Stork, G.; Baine, N. H. J. Am. Chem. Soc. 1982, 104, 2321.
(20) (a) Hua, R.; Shimada, S.; Tanaka, M. J. Am. Chem. Soc. 1998, 120,
12365. (b) Hua, R.; Onozawa, S.-y.; Tanaka, M. Chem.;Eur. J. 2005, 11,
3621. (c) Kashiwabara, T.; Kataoka, K.; Hua, R.; Shimada, S.; Tanaka, M.
Org. Lett. 2005, 7, 2241.
(21) Browing, J.; Goggin, P. L.; Goodfellow, R. J.; Norton, M. G.;
Rattray, A. J. M.; Taylor, B. F.; Mink, J. J. Chem. Soc., Dalton Trans.
1977, 2061.
(22) Another reaction at room temperature was run in toluene (5 mL)
using RhCl(CO)(PMe3)2 (38.0 mg, 0.119 mmol) and Cl3CCOCl (50 μL, 0.45
mmol) for 1 h to afford a near-quantitative yield of mer,trans-RhCl3(CO)-
(PMe3)2 (mer,trans-5, yellow powder), which was confirmed to be identical
with an authentic sample prepared via separate route (ref 21). The formation
of mer,trans-isomer in this experiment is associated with the polarity of the
solvent (vide infra).
9436 J. Org. Chem. Vol. 74, No. 24, 2009

Kashiwabara et al.
JOCArticle
i.e., HRMS (FAB, matrix=2-nitrobenzyl alcohol) calcd for
C9H1935Cl237ClO2P2Rh ([M - 2Cl þ H]þ) m/z 430.8952,
found 430.8962; calcd for C9H1935Cl37Cl2O2P2Rh ([M - 2Cl þ
H]þ) m/z 432.8924, found 432.8929.
Mechanistic detail of the transformation from Cl3CC-
(O)RhCl2(CO)(PMe3)2 to fac,cis-5 (Scheme 3), or to
RhCl3Ln in a broader sense with Scheme 2 in mind (route
2), is uncertain at this time. It may involve β-chlorine
abstraction with concomitant formation of dichloroketene,
similar to the chemistry of R-iodoacetylrhodium complex as
Cole-Hamilton reported,23 although we have been unable to
detect any species derived from dichloroketene.
To discuss the next insertion step (route 3), it is worth
noting beforehand that coordinatively saturated complex 5,
either cis or trans, is also labile in two directions (Scheme 3).
One is its configurational lability. According to Goggin and
co-workers,21 the stereochemistry of the major product
formed in chlorination of trans-RhCl(CO)(PMe3)2 with
chlorine depends on the polarity of the solvent, benzene
preferring mer,trans-RhCl3(CO)(PMe3)2 (mer,trans-5) and
dichloromethane fac,cis-5.21,24 In our experiment, an
authentic sample of fac,cis-5 (synthesized in dichloro-
FIGURE 1. Molecular structure of 6. Thermal ellipsoids are drawn
at the 30% probability level. Hydrogen atoms were omitted for
clarity.
is able to participate in the catalysis, if we admit the
chlororhodium species being ill-defined.27 In the other ex-
periment, a solution of fac,cis-5 and 1-octyne (approxi-
madely 3 equiv) in chlorobenzene was heated at 130 °C for
3 h. Although PMe3 was not a better perfoming ligand in the
present catalysis (entry 18, Table 1), 3a was formed in 27%
yield based on Rh.28,29
As for the dimerization of the rhodium-bound β-chloro-
alkenyl ligand leading to 1,4-dichloro-1,3-butadiene forma-
tion, one can think mainly of the following two possibilities.
First, the transformation is realized by reductive elimination
of two β-chloroalkenyl ligands bonded to the same rhodium
center. The pioneering work by Beletskaya and co-workers
on metal-catalyzed halogenative dimerization of alkynes
suggested such transformation of two β-iodo-(E)-alkenyl
ligands bound to a single platinum center to (E,E)-1,4-
diiodo-1,3-butadienes, although the β-iodo-(E)-alkenyl
ligand in this case was formed via external attack of iodine
to the coordinated triple bond.13a,c These authors were
also able to cleanly generate [β-iodo-(E)-alkenyl]platinum
1
methane) dissolved in benzene displayed H and 31P NMR
signals identical with those of mer,trans-5, and mer,trans-5
(synthesized in benzene) dissolved in CDCl3 displayed sig-
nals assignable to fac,cis-5, suggesting that cis-trans isomer-
ization had taken place rapidly. The other lability we
encountered is facile formation of a dinuclear rhodium
complex. Upon recrystallization in a dichloromethane/hex-
ane mixture, fac,cis-5 and mer,trans-5 were converted near
quantitatively to known dimeric rhodium complex [Rh-
(μ-Cl)Cl2(PMe3)2]2 (6),25 the structure of which is interesting
in that one rhodium has two mutually cis phosphines and the
other trans, as verified unequivocally by X-ray diffraction as
illustrated in Figure 1 although it displays a disorder at PMe3
ligands.26
The cis-trans interconversion is envisaged to involve
dissociation of a ligand, which provides an unoccupied
coordination site, a prerequisite for alkyne insertion. Like-
wise, the formation of 5 proceeds via dissociation of CO
leaving a vacant site. In the dinuclear rhodium complex
formation, the vacant site is satisfied by bridging chlorines,
but in the present catalysis, the site is available for an alkyne
molecule toward insertion. Although we do not have con-
vincing evidence for insertion of alkyne into a Cl-Rh bond,
our previous observations14,20 and following two experi-
ments combined together indicate that route 3-i (Scheme 2)
is reasonable. Thus, we ran a catalytic reaction under the
standard conditions (130 °C, 12 h, in ethylbenzene) using a
yellow powder obtained by mixing RhCl3 and 2 equiv of
P(o-Tol)3 in ethanol and subsequent evaporation to see if the
material could initiate the catalysis. The yield of 3a was 28%,
which was lower than that obtained using the [RhCl(CO)2]2
þ 4P(o-Tol)3 catalyst system (40%; entry 11, Table 1), but
not too low to consider that even the chlororhodium species
(27) In view of the yellow color of the material we obtained, we presume
that it contained Rh(III) species, but the material might not be pure, which
can be the origin of the somewhat low yield of 3a as compared with the
[RhCl(CO)2]2 þ 4P(o-Tol)3 catalyst system. Formation of Rh(II) species,
which should be blue-green, can be generated even by a slight modification of
the procedure and appears dependent on the ligand used. For instance,
treatment of RhCl3 with PPh(o-Tol)2 was assumed to generate a dimeric
Rh(II) species [RhCl2{PPh(o-Tol)2}2]2 or, less probably, a chlorine-bridged
Rh(I)-Rh(III) mixed valence dinuclear species, like [{PPh(o-Tol)2}2Rh(μ-
Cl)2RhCl2{PPh(o-Tol)2}2]. On the other hand, another similar treatment
with PPh2(o-Tol) may have afforded a chlorine-bridged Rh(I) dimer
[RhCl{PPh2(o-Tol)}2]2. For details, see: (a) Sacco, A.; Ugo, R.; Moles, A.
J. Chem. Soc., A 1966, 1670. (b) Bennett, M. A.; Longstaff, P. A. J. Am.
Chem. Soc. 1969, 91, 6266. Even if the yellow powder we generated contains a
dimeric Rh(II) species, one can anticipate that the dimeric Rh complex
undergoes “oxidative addition” with Cl3CCOCl to afford Cl-Rh(III) and
Cl3CC(O)-Rh(III) species, which are envisioned to enter the proposed
catalytic cycle. Similar arguments are also possible for a chlorine-bridged
Rh(I)-Rh(III) mixed valence dinuclear species and also for a chlorine-
bridged Rh(I) dimer.
(28) Somewhat puzzling is the formation of another product, which
appeared, in GCMS analysis, to be an isomer of 3a, in some 5%. Although
its structure has not been fully characterized, this is the only case where an
regio- or stereoisomer of 3a, if any, was formed.
(29) In a similar experiment, to [Rh(μ-Cl)Cl2(PMe3)2]2 (6) (8.2 mg, 0.046
mmol with respect to rhodium atom) and toluene-d8 (0.5 mL) placed in an
NMR tube was added 1-octyne (7 μL, 0.046 mmol) and the mixture was
heated at 110 °C for 15 h. Analysis by 1H NMR spectroscopy, after addition
of 1,1,2,2-tetrachloroethane (5.4 mg) as internal standard, revealed (Z,Z)-
7,10-dichlorohexadeca-7,9-diene (3a) being formed in 4% yield.
(23) Weston, W. S.; Cole-Hamilton, D. J. Inorg. Chim. Acta 1998, 280, 99.
(24) A similar observation was also reported by Intille for somewhat
different rhodium complexes. See: Intille, G. M. Inorg. Chem. 1972, 11, 695.
(25) This compound appeared to be the same as a rhodium complex
reported without detailed information of the stereochemistry. See ref 24.
(26) For similar complexes ligated by other phophine ligands, see: (a)
Cotton, F. A.; Kang, S.-J.; Mandal, S. K. Inorg. Chim. Acta 1992, 206, 29. (b)
Cotton, F. A.; Eglin, J. L.; Kang, S. J. Inorg. Chem. 1993, 32, 2332.
J. Org. Chem. Vol. 74, No. 24, 2009 9437

Kashiwabara et al.
JOCArticle
and bis[β-iodo-(E)-alkenyl]platinum complexes as discrete
intermediates, which clearly indicates high stereospecificity
of the reductive elimination. The other possibility is dispro-
portionation of (β-chloroalkenyl)RhCl2 species, which pro-
duces (β-chloroalkenyl)2RhCl and RhCl3 species. There have
been a plethora of publications reporting disproportionation
of this type.30 Whitesides and co-workers reported stereospe-
cific thermolysis of (E)- and (Z)-propen-1-yl copper and silver
complexes forming 2,4-hexadiene with retention of configura-
tion at the double bonds. They have proposed disproportio-
nation and other possibilities like bimolecular reductive
elimination.31 The latter possibility cannot be excluded in
our reaction and remains to be studied further.
IR (neat, cm-1) 2247, 1604. GCMS (70 eV) m/z (% relative
intensity) 256 ([M]þ, 65), 221 (39), 185 (40), 180 (100), 166 (17),
144 (35). HRMS (EI) calcd for C12H14Cl2N2 m/z 256.0534,
found 256.0530.
Dimethyl (Z,Z)-5,8-dichloro-5,7-dodecadienedioate (3e). Iso-
lated by preparative TLC (silica gel, hexane/acetone = 95/5),
pale yellow oil. 1H NMR (300 MHz, CDCl3) δ 6.38 (s, 2H), 3.64
(s, 6H), 2.43 (t, J=7.16 Hz, 4H), 2.30 (t, J=7.40 Hz, 4H), 1.89
(m, 4H). 13C{1H} NMR (75 MHz, CDCl3) δ 173.4, 136.6, 121.1,
51.5, 38.9, 32.5, 22.5; IR (neat, cm-1) 1734, 1603; GCMS (70 eV)
m/z (relative intensity) 322 ([M]þ, 57), 286 (28), 255 (60), 219
(100), 186 (29), 174 (31), 139 (26), 117 (26); HRMS (EI) calcd for
C14H20Cl2O4 m/z 322.0739, found 322.0732.
(Z,Z)-1,6-Diphenyl-2,5-dichloro-2,4-hexadiene (3f). Isolated
by column chromatography (silica gel, hexane/methyl tert-butyl
ether=10/1): colorless solid; mp 125.9-126.9 °C; 1H NMR (300
MHz, CDCl3) δ 7.35-7.23 (m, 10H), 6.52 (s, 2H), 3.71 (s, 4H);
13C{1H} NMR (75 MHz, CDCl3) δ 137.3, 137.2, 129.4, 128.9,
127.4, 122.2, 46.5; IR (KBr, cm-1) 1603; GCMS (70 eV) m/z
(relative intensity) 302 ([M]þ, 100), 267 (20), 231 (26), 211 (22),
125 (22), 91 (97); HRMS (EI) calcd for C18H16Cl2 m/z 302.0629,
found 302.0630.
In conclusion, the reaction of terminal alkynes with tri-
chloroacetyl chloride has proved to provide an operationally
easy and high yielding access to (Z,Z)-1,4-dichloro-1,
3-butadienes stereo- and regioselectively. Further mechan-
istic study on this and related catalysis is in progress.
Experimental Section
(Z,Z)-1,4-Dichloro-1,4-bis(trimethylsilyl)-1,3-butadiene (3g).
Isolated by preparative TLC (silica gel, hexane): colorless oil;
1H NMR (300 MHz, CDCl3) δ 7.06 (s, 2H), 0.23 (s, 18H);
13C{1H} NMR (75 MHz, CDCl3) δ 143.6, 130.5, -2.4; 29Si{1H}
NMR (60 MHz, CDCl3) δ 0.17; IR (neat, cm-1) 1604. GCMS
(70 eV) m/z (relative intensity) 266 ([M]þ, 26), 143 (33), 123 (42),
93 (63), 73 (100); HRMS (EI) calcd for C10H20Cl2Si2 m/z
266.0481, found 266.0480.
(Z,Z)-1,4-Dichloro-1,4-bis(1-cyclohexenyl)-1,3-butadiene (3h).
Isolated by preparative TLC (silica gel, hexane): colorless solid;
mp 95.8-97.2 °C; 1H NMR (300 MHz, CDCl3) δ 6.84 (s, 2H),
6.48 (t, J=1.8 Hz, 2H), 2.34-2.20 (m, 8H), 1.74-0.58 (m, 8H);
13C{1H} NMR (75 MHz, CDCl3) δ 137.1, 133.6, 130.6, 119.1,
31.6, 26.3, 26.1, 22.7; IR (KBr, cm-1) 1633; GCMS (70 eV) m/z
(relative intensity) 282 ([M]þ, 55), 247 (100), 211 (89), 179 (91),
131 (15), 81 (49); HRMS (EI) calcd for C16H20Cl2 m/z 282.0942,
found 282.0940.
Typical Procedure for the Reaction of Alkyne with Trichloro-
acetyl Chloride: The Reaction with 1-Octyne Affording (Z,Z)-
7,10-Dichlorohexadeca-7,9-diene (3a). To a mixture of [RhCl-
(CO)2]2 (9.7 mg, 0.025 mmol), PMes3 (39 mg, 0.10 mmol), and
chlorobenzene (1.0 mL) placed in a 5 mL Schlenk tube were
added trichloroacetyl chloride (180 mg, 1.0 mmol) and 1-octyne
(110 mg, 1.0 mmol). The mixture was heated at 130 °C for 3 h.
After being cooled to room temperature, the resulting mixture
was analyzed by 1H NMR spectroscopy after addition of
1,1,2,2-tetrachloroethane (24.3 mg) as internal standard and
CDCl3 (1.0 mL) to reveal the formation of (Z,Z)-7,10-dichlor-
ohexadeca-7,9-diene 3a in 91% NMR yield. Evaporation left a
brown residue, which was subjected to preparative TCL (silica
gel, hexane), leading to isolation of 3a (127.6 mg, 88%) as a
colorless oil: 1H NMR (300 MHz, CDCl3) δ 6.40 (s, 1H), 2.39 (t,
J=7.54 Hz, 4H), 1.54 (quart, J=7.26 Hz, 4H), 1.29-1.09 (m,
12H), 0.86 (t, J=6.80 Hz, 6H); 13C{1H} NMR (75 MHz, CDCl3)
(Z,Z)-1,4-Dichloro-1,4-di(4-methoxyphenyl)-1,3-butadiene (3j).
The reaction mixture was evaporated, and the residue was
extracted with hot hexane. Evaporation of the extract gave an
analytically pure sample: pale yellow solid; mp 193.2-194.1 °C;
1H NMR (300 MHz, CDCl3) δ 7.67 (d, J=7.2 Hz, 4H), 7.21 (s,
2H), 6.92 (d, J=7.2 Hz, 4H), 3.85 (s, 6H); 13C{1H} NMR (75
MHz, CDCl3) δ 160.3, 135.0, 130.2, 127.8, 120.4, 113.8, 55.4; IR
(neat, cm-1) 1605. GCMS (70 eV) m/z (relative intensity) 334
([M]þ, 77), 299 (44), 264 (100); HRMS (EI) calcd for
C18H16Cl2O2 m/z 334.0527, found 334.0528. Anal. Calcd for
C18H16Cl2O2: C, 64.49; H, 4.81. Found: C, 64.18; H, 5.23.
(Z,Z)-1,4-Dichloro-1,4-di(4-fluorophenyl)-1,3-butadiene (3l).
Isolated by exactly the same procedure as for 3j: colorless
δ 133.7, 120.3, 39.9, 31.5, 28.3, 27.5, 22.5, 14.1; IR (neat, cm-1
)
1605; GCMS (70 eV) m/z (relative intensity) 290 ([M]þ, 100), 255
(4), 206 (8), 185 (12), 149 (14), 135 (18), 109 (40), 95 (21), 81 (21),
67 (13); HRMS (EI) calcd for C16H28Cl2 m/z 290.1568, found
290.1566.
(Z,Z)-1,4,7,10-Tetrachloro-4,6-decadiene (3c). Isolated by
preparative TLC (silica gel, hexane): colorless oil; bp 210-215
°C/0.1 mmHg (Kugelrohr); 1H NMR (300 MHz, CDCl3) δ 6.48
(s, 2H), 3.55 (t, J=6.31 Hz, 4H), 2.59 (t, J=7.08 Hz, 4H), 2.07
(quint-like, 4H); 13C{1H} NMR (75 MHz, CDCl3) δ 136.1,
121.4, 43.5, 36.9, 30.0; IR (neat, cm-1) 1604; GCMS (70 eV)
m/z (relative intensity) 276 ([M]þ, 100), 239 (27), 211 (34), 175
(17), 167 (29), 139 (78), 113 (15), 108 (23). Anal. Calcd for
C10H14Cl4: C, 43.51; H, 5.11. Found: C, 43.41; H, 4.91.
(Z,Z)-5,8-Dichloro-5,7-dodecadienedinitrile (3d). Isolated by
preparative TLC (silica gel, hexane/acetone=90/10): colorless
oil; 1H NMR (300 MHz, CDCl3) δ 6.49 (s, 2H), 2.58 (t, J=7.12
Hz, 4H), 2.36 (t, J=7.08 Hz, 4H), 1.98 (quart-like, 4H); 13C{1H}
NMR (75 MHz, CDCl3) δ 135.5, 121.9, 118.9, 38.2, 23.0, 15.9.
1
needles; mp 143.2-144.5 °C; H NMR (300 MHz, CDCl3) δ
3
4
7.70 (AB spin system, JH-H=7.70 Hz, JH-F=4.9 Hz, 4H),
7.22 (s, 2H), 7.09 (dd, 3JH-H=7.70 Hz, 3JF-H=8.61 Hz, 4H);
13C{1H} NMR (75 MHz, CDCl3) δ 162.3 (d, J = 252.2 Hz),
135.0, 133.3 (d, J=3.54 Hz), 128.4 (d, J=8.45 Hz), 121.6, 115.5
(d, J = 20.6 Hz); IR (neat, cm-1) 1597; GCMS (70 eV) m/z
(relative intensity) 310 ([M]þ, 100), 275 (51), 240 (94), 220 (18).
Anal. Calcd for C16H10Cl2F2: C, 61.76; H, 3.24. Found: C,
61.96; H, 3.54.
(30) (a) Tsou, T. T.; Kochi, J. K. J. Am. Chem. Soc. 1979, 101, 7547. (b)
Carvajal, J.; Muller, G.; Sales, J.; Solans, X.; Miravitlles, C. Organometallics
1984, 3, 996. (c) Ozawa, F.; Fujimori, .; Yamamoto, T.; Yamamoto, A.
Organometallics 1986, 5, 2144. (d) Ozawa, F.; Hidaka, T.; Yamamoto, T.;
Yamamoto, A. J. Organomet. Chem. 1987, 330, 253. (e) van Asselt, R.;
Elsevier, C. J. Organometallics 1994, 13, 1972. (f) Suzaki, Y.; Yagyu, T.;
Osakada, K. J. Organomet. Chem. 2007, 692, 326. (g) Wakioka, M.; Nagao,
M.; Ozawa, F. Organometallics 2008, 27, 602.
(Z,Z)-1,4-Dichloro-1,4-di(2-thienyl)-1,3-butadiene (3m). Iso-
lated by preparative TLC (silica gel, hexane), yellow needles, mp
1
127.3-127.8 °C. H NMR (300 MHz, CDCl3) δ 7.39 (dd, J=
1.15, 3.77 Hz, 2H), 7.30 (dd, J=1.15, 5.02 Hz, 2H), 7.18 (s, 2H),
7.03 (dd, J = 3.77, 5.02 Hz, 2H). 13C{1H} NMR (75 MHz,
CDCl3) δ 142.5, 129.2, 128.2, 127.2, 126.3, 120.2. IR (neat,
cm-1) 1575. GCMS (70 eV) m/z (% relative intensity) 286 ([M]þ,
(31) Whitesides, G. M.; Casey, C. P.; Krieger, J. K. J. Am. Chem. Soc.
1971, 93, 1379.
9438 J. Org. Chem. Vol. 74, No. 24, 2009

Kashiwabara et al.
JOCArticle
39), 251 (35), 216 (100), 171 (15). HRMS (FAB, matrix =
2-nitrobenzyl alcohol) calcd for C12H9Cl232S2 (value for [M þ
H]þ) m/z 286.9523, found 286.9523.
diffraction data collection on a diffractometer with Mo KR
radiation (= 0.7107 A). Atom C8 was refined isotropically.
˚
Crystal data for 5: C12H30Cl6P4Rh2, M=716.79, orthorhombic,
˚
Pnma (#62), a=11.217(5) A, b=13.036(6) A, c=17.985(9) A,
mer,trans-RhCl3(CO)(PMe3)2 (mer,trans-5). 1H NMR (400
MHz, C6D6) δ 1.31 (t, JP-H =3.91 Hz); 31P{1H} NMR (162
MHz, C6D6) δ -2.28 (d, JRh-P=72.3 Hz); IR (KBr, cm-1) 2085
(νCO). An authentic sample synthesized by us according to the
Goggin’s procedure21 displayed 1H and 31P{1H} NMR signals
at δ 1.31 (t, JP-H=4.00 Hz) and at δ -2.30 (d, JRh-P=72.3 Hz),
respectively.
˚
˚
V=2360(2) A , Z=4, Dcalc=1.810 g/cm3. GOF=0.883, R (I >
3
˚
3.00σ(I))=0.0416, Rw (I > 3.00σ(I), all reflections)=0.0602.
Selected bond distances and angles are summarized in Tables
S2-1 and S2-2 in the Supporting Information.
Reaction of fac,cis-RhCl3(CO)(PMe3)2 (fac,cis-5) with
1-Octyne. A mixture of the rhodium complex (3.8 mg, 9.8 ꢀ
10-3 mmol) and 1-octyne (3.1 mg, 0.028 mmol) in chloroben-
zene (0.5 mL) was heated at 130 °C for 3 h in an NMR tube,
cooled to room temperature, and analyzed, after addition of
n-tetradecane (3.0 μL; internal standard), by 1H NMR spectro-
scopy to reveal 3aA and 3aA0 being formed in 27 and 5% yield,
respectively.
Conversion of mer,trans-RhCl3(CO)(PMe3)2 (mer,trans-5) to
[Rh(μ-Cl)Cl2(PMe3)2]2 (6). mer,trans-RhCl3(CO)(PMe3)2 (mer,
trans-5) obtained by the reaction of trans-RhCl(CO)(PMe3)2
with trichloroacetyl chloride22 was recrystallized from dichloro-
methane/hexane (95/5) to afford brown crystals. The solvent
was removed with a syringe, and the crystals were rinsed with
hexane and dried to furnish [Rh(μ-Cl)Cl2(PMe3)2]2 (6; 37.8
mg, 88%), brown crystals, mp 248.3 °C (under nitrogen, dec)
(lit.24 mp 250-254 °C dec).
[Rh(μ-Cl)Cl2(PMe3)2]2 (6). This compound has been docu-
mented without spectral data,24 which are as follows: 1H NMR
400 MHz, CDCl3) δ 1.87 (virtual t, JP-H=3.92 Hz, 18H), 1.64
(d, JP-H=11.7 Hz, 18H); 13C{1H} NMR (75 MHz, CDCl3) δ
15.7 (d, JC-P=20.6 Hz), 12.6 (virtual t, JC-P=15.8 Hz); 31P{1H}
NMR (162 MHz, CDCl3) δ 25.1 (d, JRh-P=113.1 Hz), -4.5 (d,
JRh-P=80.1 Hz).
Acknowledgment. This work was supported by a Grant-
in-Aid for Scientific Research on Priority Areas (No.
18065008) from the Ministry of Education, Culture, Sports,
Science and Technology, Japan and by a research fellowship
to T.K. from the Japan Society for the Promotion of Science
and Technology.
Supporting Information Available: Experimental details,
NMR spectra of 3a-m and crystallographic data for 3f and 6
(CIF). This material is available free of charge via the Internet at
http://pubs.acs.org.
X-ray Analysis of [Rh(μ-Cl)Cl2(PMe3)2]2 (6). Single crystals
were obtained from a CH2Cl2/hexane (95/5) solution. A
brown crystal (= 0.20 ꢀ 0.10 ꢀ 0.10 mm) was used for X-ray
J. Org. Chem. Vol. 74, No. 24, 2009 9439