Selective carbonylation routes to thiocarbamates. An alternative to phosgene

Article abstract of DOI:10.1021/om990931n

A new route for the synthesis of thiocarbamates using a transition metal complex and CO as a replacement for phosgene has been discovered. The complex (PPh₃)₂PdCl₂ reacts with N-benzylmethylamine, 4-chlorobenzenethiol, and CO to selectively generate a thiocarbamate. Kinetic studies and the isolation of several intermediates in this reaction allow for the proposal of a mechanistic scheme.

Full text of DOI:10.1021/om990931n

Organometallics 2000, 19, 1661-1669
1661
Selective Ca r bon yla tion Rou tes to Th ioca r ba m a tes. An
Alter n a tive to P h osgen e
William D. J ones,* Kelly A. Reynolds, Caroline K. Sperry, and
Rene J . Lachicotte
Department of Chemistry, University of Rochester, Rochester, New York 14627
Stephen A. Godleski and Ronald R. Valente
Eastman Kodak Company, Rochester, New York 14650
Received November 26, 1999
A new route for the synthesis of thiocarbamates using a transition metal complex and CO
as a replacement for phosgene has been discovered. The complex (PPh3)2PdCl2 reacts with
N-benzylmethylamine, 4-chlorobenzenethiol, and CO to selectively generate a thiocarbamate.
Kinetic studies and the isolation of several intermediates in this reaction allow for the
proposal of a mechanistic scheme.
In tr od u ction
Co2(CO)8 or Ru3(CO)12 will catalyze the insertion of
carbon monoxide into thietanes (eq 3).5
The manufacture of acid chlorides as well as carbam-
oyl chlorides and the corresponding amides, urethanes,
and carbamoyl thioesters represent an important class
of specialty chemicals produced in large volume by the
chemical industry. Many of the products are synthesized
via acyl halide intermediates, which in turn are formed
by way of acylating reagents such as thionyl chloride,
phosphorus oxychloride, or phosgene. Thus, novel and
efficient new methods to prepare these intermediates
and/or products would have a great impact in reducing
process-generated waste as well as in the handling of
noxious intermediates.
There are many reports in the literature using carbon
monoxide to couple organic and heteroatom-containing
fragments. For example, the insertion of CO into
σ-bound ligands attached to Ni(II) has been observed
by Hillhouse. Reaction of in situ Ni(0) with thietane
yields a metallathiacyclopentane that reacts with CO.
Insertion occurs to yield a thiobutyrolactone, but no
intermediates were seen to distinguish between initial
Ni-C or Ni-S insertion (eq 1).1 Hillhouse has also been
successful in inserting oxygen into the analogous metal-
lacyclopentane complex using N2O, which also inserts
CO to give δ-valerolactone.2 Adams has reviewed reac-
tions of thietanes including CO insertion reactions.3
Holm has also observed coupling of coordinated alkyl
groups with CO and alcohols or amines in a Ni(II)
system. Reaction of the dialkyl complex with thiol leads
to an alkylthiolate complex, which in turn inserts CO
and eliminates thioester (eq 2).4 Catalytic insertions
have also been observed. Alper has reported that either
Of more relevance to the goal of mixed heterosubsti-
tuted acyl derivatives, Alper has reported the catalytic
synthesis of urethanes in a reductive carbonylation of
aromatic nitro compounds. Reaction of the nitroaromatic
with CO in methanol solvent in the presence of a
montmorillonitebipyridinylpalladium(II) acetate clay
(Pd-clay) and Ru3(CO)12 leads to the direct formation
of the carbamates (eq 4). The Pd-clay is believed to be
responsible for the in situ reduction of the nitroaromatic
to the amine, which then undergoes methanolic carbo-
nylation on the ruthenium center. The reaction works
for a variety of substituted nitroaromatics, giving
urethanes in 70-97% yields.6 Another recent paper by
(1) Matsunaga, P. T.; Hillhouse, G. L. Angew. Chem., Int. Ed. Engl.
1994, 33, 1748.
(2) Matsunaga, P. T.; Mavropoulos, J . C.; Hillhouse, G. L. Polyhedron
1995, 14, 175.
(3) Adams, R. D.; Falloon, S. B. Chem. Rev. 1995, 95, 2587.
(4) Tucci, G. C.; Holm, R. H. J . Am. Chem. Soc. 1995, 117, 6489.
(5) Wang, M. D.; Calet, S.; Alper, H. J . Org. Chem. 1989, 54, 20.
(6) Valli, V. L, K.; Alper, H. J . Am. Chem. Soc. 1993, 115, 3778.
10.1021/om990931n CCC: $19.00 © 2000 American Chemical Society
Publication on Web 03/31/2000

1662 Organometallics, Vol. 19, No. 9, 2000
J ones et al.
Sch em e 1. P r op osed Ca ta lytic Cycle
Alper reports the carbonylation of 1,3-thiazolidines to
give thiazolidinones. The reaction is catalyzed by [RhCl-
(COD)]2 and gives yields of 56-88%.7
Gladfelter has also looked at metal-catalyzed carbo-
nylation reactions of amines to aryl carbamates.8 His
work initially examined the catalytic reactions of nitro-
aromatics and methanol using Ru(dppe)(CO)3 as the
carbonylation catalyst. In situ studies were complicated
by the high pressures required for the reaction.9,10
Detailed kinetic studies were performed on the reaction
of aromatic amines with Ru(dppe)(CO)2(COOMe)2, a
species found to be present under the catalytic condi-
tions. The reactions support the formation of ruthenium
amido derivatives by replacement of the methoxy group
with the amino group. Elimination of aryl isocyanate
then occurs, and the latter reacts rapidly with methanol
to give the observed carbamate products (eq 5).11,12,13
are employed in a palladium-catalyzed insertion into
carbon monoxide to give the desired product (eq 7). The
mechanism suggested for the reaction involves addition
of the sulfenamide S-N bond across a palladium-
thioester bond.
The above examples all indicate that carbonylation
of heteroatom-bound ligands can be effected using
transition metals. Mechanisms appear to include attack
on coordinated CO, attack on coordinated acyl groups,
and insertion of CO into metal-heteroatom bonds.
While some catalysis has appeared, there is vast room
for improvement and discovery in the synthesis of acyl
derivatives.
One particularly toxic chemical that is currently used
in the synthesis of carbamoyl thioesters is phosgene (eq
8). In light of the restrictions on handling this reagent,
we felt that carbon monoxide might be a promising
candidate for the replacement of phosgene for the
environmentally benign synthesis of acyl products. We
describe in this paper an alternate route for the syn-
thesis of carbamoyl thioesters using an organometallic
complex and CO.
Catalytic carbonylation of C-H bonds and olefin
insertion has recently been reported by Murai and co-
workers.14 In the report, imidazoles are activated,
carbonylated, and reacted with olefins to give ketone
products (eq 6). This chemistry might be extended to
heteroatom-H bond activation to produce heteroatom
substituted acyl derivatives.
A particularly interesting example of thiocarbamate
Resu lts a n d Discu ssion
synthesis has recently appeared.15 Sulfenamides (RSNR′2)
Ca r bon yla tion Rea ction s. Initial approaches to the
synthesis of thiocarbamates focused on Fisher carbene
complexes formed from a transition metal carbonyl
complex. A potential catalytic cycle is shown in Scheme
1. Attack of a heteroatom nucleophile on coordinated
CO generates an acyl derivative that has Fisher carbene
character. A second nucleophile then attacks, and the
intermediate formed eliminates a difunctionalized acyl
product. For a catalytic cycle, the metal would then have
to be oxidized by two electrons by an external oxidant.
Successful achievement of catalysis would require one
nucleophile to be more reactive with the metal carbonyl
complex and the second nucleophile to be more reactive
with the Fisher carbene complex.
(7) Khumtaveeporn, K.; Alper, H. J . Am. Chem. Soc. 1994, 116, 5662.
(8) Gladfelter, W. L.; Gargulak, J . D. In Benign by Design. Alterna-
tive Synthetic Design for Polution Prevention. Chapter 4. Mechanistic
Study of a Catalytic Process for Carbonylation of Nitroaromatic
Compounds. American Chemical Society: Washington, DC, 1994.
(9) Kunin, A. J .; Noirot, M. D.; Gladfelter, W. L. J . Am. Chem. Soc.
1989, 111, 2739.
(10) Gargulak, J . D.; Berry, A. J .; Noirot, M. D.; Gladfelter, W. L.
J . Am. Chem. Soc. 1992, 114, 8933.
(11) Gargulak, J . D.; Gladfelter, W. L. Inorg. Chem. 1994, 33, 253.
(12) Gargulak, J . D.; Gladfelter, W. L. Organometallics 1994, 13,
698.
(13) Gargulak, J . D.; Gladfelter, W. L. J . Am. Chem. Soc. 1994, 116,
3792.
(14) Chatani, N.; Fukuyama, T.; Kakiuchi, F.; Murai, S. J . Am.
Chem. Soc. 1996, 118, 493.
(15) Kuniyasu, H.; Hiraike, H.; Morita, M.; Tanaka, A.; Sugoh, K.;
Kurosawa, H. J . Org. Chem. 1999, 64, 7305.

Carbonylation Routes to Thiocarbamates
Organometallics, Vol. 19, No. 9, 2000 1663
Ta ble 1. Meta l Com p lexes Scr een ed for Activity
Sch em e 2. Rea ction P a th w a y for Gen er a tion of
4-Ch lor op h en yld ip h en ylp h osp h in e
cationic carbonyls
neutral carbonyls
non-carbonyls
[CpFe(CO)3]+[SbF6]-
[(PPh3)2PtCl(CO)]+[BF4]-
[CpRu(CO)3]+
Mn(CO)5Br
Fe(CO)5
Ru3(CO)12
Cr(CO)6
(PPh3)2PtCl2
(PPh3)2PdX2
(dppe)PdCl2
(tmeda)PdCl2
[Ir(CO)3(PPh3)2]+
2-
PtCl4
(bipy)PtCl2
Several metal carbonyl complexes were screened for
reactivity. Three classes of compounds were studied:
neutral carbonyls, cationic carbonyls, and non-carbonyls
under a CO atmosphere (Table 1). These reactions were
performed using a simple secondary amine, N-benzyl-
methylamine, and an oxidation-resistant aromatic thiol,
4-chlorobenzenethiol, as model reagents. Although many
of these complexes did react with the amine, further
reaction with the thiol was unsuccessful for these
carbonyl-containing complexes.
Use of the coordinatively unsaturated non-carbonyl
complex (PPh3)2PdCl2 (1) proved to be successful. A THF
solution of 1 was treated with 3 equiv of amine followed
by 1 equiv of 4-chlorobenzenethiol. After 24 h under 1
atm of CO at room temperature, the desired thiocar-
bamate product was isolated in 47% yield (eq 9). The
an isomerization toplace the aryl ring on the phosphine.17b
On the basis of the results from these reactions, the
reaction pathway shown in Scheme 2 might be operat-
ing. The oxidative addition of the thiol involves cleavage
of an aryl C-S bond.18 The p-Cl-substituted ring then
exchanges with a phenyl group on the phosphine.
In an attempt to identify the active species in the
thiocarbamate-forming reaction (eq 9), N-benzylmethyl-
amine was added to (PPh3)2PdCl2 suspended in THF-
d8. 31P NMR spectroscopy provides evidence for the
formation of the amine adduct, (PPh3)PdCl2(MeHNCH2-
Ph) (2), along with free PPh3 (eq 10). 2 is stable in the
major byproduct of these reactions was the hydrochlo-
ride salt of the amine, [MeH2NCH2Ph]Cl, which was
isolated as a white precipitate from the reaction solu-
tion. Similar results were obtained with (PPh3)2PdBr2.
No urea is formed. The chelated compounds (dppe)PdCl2
and (TMEDA)PdCl2 were not active. (bipy)PdCl2, how-
ever, did show formation of thiocarbamate, possibly due
2-
to the greater lability of the bipy ligand. K2[PdCl4
]
presence of a large excess of N-benzylmethylamine,
showing no signs of either bis(amine) formation or
deprotonation to form the metal amido. The crystallo-
graphically determined structure of 2 is shown in Figure
1. The amine hydrogen was located and refined, con-
sistent with the Pd(II) structure with a pyramidal
nitrogen. The addition of excess amine and thiol to 2
under CO results in good yields of thiocarbamate.
When a THF solution of 2 is exposed to an atmosphere
of CO in the presence of free PPh3, approximately 1
equiv ((10%) of CO per Pd is consumed and the CO
inserted complex, (PPh3)2PdCl(C(O)NMeCH2Ph) (3) is
formed (eq 11). In solution, two isomers are observed
was also screened for reactivity, but no thiocarbamate
formation was found.
In addition to the desired thiocarbamate product,
several other species are detected by GC-MS of the
reaction mixture upon the completion of the reactions
using the (PPh3)2PdX2 complexes (see Supporting In-
formation). The order of addition of thiol and amine does
not effect the reaction products obtained. A large
quantity of PPh3 is observed. The dominant sulfur-
containing byproduct was identified as S(C6H4Cl)2 (ap-
proximately 62% of byproducts). Minor quantities of Sd
PPh3 (20%), S(C6H5)2 (3%), and (C6H5)S(C6H4Cl) (15%)
were also routinely detected. Similar to this result, the
thermolysis of Ni(SPh)2(PEt3)2 has been reported to
afford PhSPh in quantitative yield.16 Similar sulfur-
containing products have been observed for the ther-
molysis of [H-SNi(C6H4CH3)(PEt3)2].16
One minor species was identified as being 4-chlo-
rophenyldiphenylphosphine, indicating exchange of aryl
groups between the phosphine and the thiol. There is
precedence for the formation of substituted phosphines
from the reaction of Pd(0) with substituted aromatics.17
The aryl group exchange can be thought to occur via
the oxidative addition of the Ar-SH bond followed by
(17) a) Tunney, S. E.; Stille, J . K. J . Org. Chem. 1987, 52, 748. (b)
Herrmann, W. A.; Broâmer, C.; O¨ fele, K.; Beller, M.; Fisher, H. J .
Organomet. Chem. 1995, 491, C1.
(18) For examples of oxidative addition of aryl sulfides to Ni see:
(a) Wenkert, E.; Shepard, M. E.; McPhail, A. T. J . Chem. Soc., Chem.
Commun. 1986, 1390. (b) Osakada, K.; Maeda, M.; Nakamura, Y.;
Yamamoto, T.; Yamamoto, A. J . Chem. Soc., Chem. Commun. 1986,
442. For other transition metal mediated C-S bond cleavage reactions,
see: (c) Curtis, M. D.; Druker, S. H. J . Am. Chem. Soc. 1997, 119,
1027. (d) Luh, T.-Y.; Ni, Z.-J . Synth. 1990, 89.
(16) Osakada, K.; Hayashi, H.; Maeda, M.; Yamamoto, T.; Yama-
moto, A. Chem. Lett. 1986, 4, 597.

1664 Organometallics, Vol. 19, No. 9, 2000
Ta ble 2. Cr ysta l P a r a m eter s for 2, 4, a n d 5
J ones et al.
crystal parameter
2
4
5
chemical formula
C
26H28Cl2NPPd
C42H34Cl2P2PdS
monoclinic
P2(1)/c, 4
18.9364(3)
9.7315(2)
21.5545(2)
106.3830(10)
3810.78(11)
-80
C60H46Cl4P2Pt2S4
monoclinic
C2/c, 8
cryst syst
space group, Z
a, Å
b, Å
c, Å
monoclinic
P2(1)/n, 4
10.84180(10)
15.7204(2)
16.2474(2)
92.1470(10)
2767.22(8)
-80
21.6557(2)
13.6779(2)
24.2998(2)
107.4267(8)
6867.33(8)
-30
â, deg
vol, Å3
temp, °C
abs coeff (mm-1
radiation, λ, Å
range of 2θ
)
1.132
0.795
0.465
Mo, 0.71073
2.0-46.6
16 743
Mo, 0.71073
2.0-56.6
14 157
Mo, 0.71073
2.0-43.9
4566
no. of data collected
no. of unique data
6480
5290
4446
no. of obsd data
5805
3557
3024
agreement between equiv refs
no. of params varied
range of transmn factors
R1(Fo), wR2(Fo2), I > 2σ(I)
R1(Fo), wR2(Fo2), all data
R(Fo), Rw(Fo)
0.0196
312
0.856-0.928
0.0229, 0.0552
0.0281, 0.0572
0.0936
433
0.831-0.928
0.0592, 0.1496
0.1078, 0.1690
0.0322
361
0.66-0.999
0.061, 0.092
3.12
goodness of fit
1.036
1.038
F igu r e 2. ORTEP drawing of (PPh3)2PdCl(SC6H4Cl) (4).
Ellipsoids are shown at 30% probability. Hydrogen atoms
have been omitted for clarity. Selected distances (Å) and
angles (deg): Pd1-P1, 2.345(2); Pd1-P2, 2.330(2); Pd1-
Cl1, 2.331(2); Pd1-S1, 2.300(2), P1-Pd1-P2, 174.09(8);
Cl1-P1-S1, 178.87(8).
F igu r e 1. ORTEP drawing of (PPh3)PdCl2(MeHNCH2Ph)
(2). Ellipsoids are shown at 30% probability. Hydrogen
atoms (except on N) have been omitted for clarity. Selected
distances (Å) and angles (deg): Pd1-P1, 2.2557(4); Pd1-
N1, 2.1217(14); Pd1-Cl1, 2.2851(4); Pd1-Cl2, 2.2968(4),
P1-Pd1-N1, 175.07(4); Cl1-P1-Cl2, 176.95(2).
[Pd(µ-SPh)(Cl)(PPh3)]2, greatly favored in an equilibri-
um, requiring the use of molten PPh3 to break the dimer
and favor the monomer (eq 13).22
by 1H and 31P NMR spectroscopy for 3, attributed to
syn/anti amide rotamers.19 To verify that these isomers
were due to the asymmetric amine, the same series of
reactions was carried out using the symmetric amine,
dibenzylamine. In this case, only a single isomer of the
Pd complex analogous to 3 was observed. The formation
of Pd carbamoyl complexes under similar conditions has
been previously reported.20
Addition of 4-chlorobenzenethiol to (PPh3)2PdCl2 in
THF results in an instantaneous color change to red-
orange and the formation of the thiol complex, (PPh3)2-
PdCl(SAr) (4) (Ar ) C6H4Cl) (eq 12). The single-crystal
X-ray structure of 4 is shown in Figure 2, showing Pd-
(II) with a single thiolate ligand. Interestingly, 4 is
isostructural with the known complex (PPh3)2PdCl-
(SC6H5).21 In the latter case, however, formation of the
benzenethiol complex is difficult with the dimeric form,
Similarly, the addition of 4-chlorobenzenethiol to the
platinum complex [(PPh3)2PtCl(CO)]+[BF4]- in the pres-
(19) Preliminary X-ray verifies the connectivity of 3. Cell: a ) 28.328
Å, b ) 10.619 Å, c ) 31.432 Å, R ) â ) γ ) 90.00°.
(20) Green, C. R.; Angelici, R. J . Inorg. Chem. 1972, 11, 2095.
(21) Alvarez-Larena, A.; Piniella, J . F. Z. Kristallogr. 1993, 208, 249.
(22) Boschi, T.; Crociani, B.; Toniolo, L.; Belluco, U. Inorg. Chem.
1970, 9, 532.

Carbonylation Routes to Thiocarbamates
Organometallics, Vol. 19, No. 9, 2000 1665
F igu r e 4. Plot showing the effect of excess N-benzyl-
methylamine on rate. (PPh3)2PdCl2 ) 0.028 M, 4-chloro-
benzenethiol ) 0.038 M in THF-d8, PCO ) 1.5 atm, 22 °C.
F igu r e 3. ORTEP drawing of [(PPh3)(µ-SC6H4Cl)(SC6H4-
Cl)Pt]2 (5) Ellipsoids are shown at 30% probability. Hydro-
gen atoms and phosphine phenyl groups have been omitted
for clarity. Selected distances (Å) and angles (deg): Pt1-
P1, 2.269(6); Pt1-S1, 2.373(6); Pt1-S1A, 2.326(6); Pt1-
S2, 2.326(6); P1-Pd1-S1, 177.2(2); S1A-P1-S2, 171.9(2).
appear to substitute for PPh3 when they are present in
high concentrations.
Addition of CO to a solution of 4 resulted in no
reaction. Therefore, the thiol complex itself does not
react with CO in the absence of amine. Apparently,
deprotonation of the amine adduct and loss of Cl from
6 occurs concomitantly with the recoordination of a PPh3
and insertion of CO to give the unobserved intermediate,
7 (eq 16). In accord with the observation of CO insertion
into the Pd-N bond to give 3, 7 is formulated as
containing a Pd carbamoyl moiety.
ence of amine also gives a dimeric thiolate bridged
complex, [(PPh3)(µ-SC6H4Cl)(SC6H4Cl)Pt]2 (5) (eq 14).
A single-crystal X-ray structure of 5 is shown in Figure
3, showing the replacement of both the chloride and CO
ligands by thiolate. We observe no evidence for such
dimer formation in the reaction of 1 with 4-chloroben-
zenethiol. Likewise, the reported reaction of (PhCN)2-
PdCl2 with benzenethiol is proposed to form a mono-
meric (PhCN)2Pd(SPh)Cl complex with the release of
HCl.23 If the Pd and Pt species react similarly, then
these observations suggest the initial substitution of Cl
rather than CO in [(Ph3P)2PtCl(CO)]+, which then
substitutes CO to give 5.
A rearrangement of 7 to a cis form (alternatively, 7
might actually be the cis isomer) is then followed by the
reductive elimination of the thiol with the CO inserted
amide to yield the thiocarbamate product. The Pd is now
(PPh3)2Pd(0), which decomposes to Pd metal and PPh3.
No evidence for the formation of (PPh3)2Pd(CO)2 or
(PPh3)Pd(CO)3 is seen by IR spectroscopy, as these
compounds are not stable at 1 atm CO.24
Kin etic Stu d ies. To more closely investigate the
details of the reaction using the (PPh3)2PdCl2 starting
compound, stoichiometric kinetic studies were under-
taken. The dominant species observed in solutions
containing 1, amine, and thiol was 4. The rate of
consumption of 4 proved to be first order as monitored
by inversed-gated {1H}31P NMR spectroscopy using
methyldiphenylphosphine oxide as an internal standard
(see Supporting Information). The effect of excess N-
benzylmethylamine is shown in Figure 4. As can be
seen, the reaction does not proceed at reasonable rates
with less than 3 equiv of this amine. The amine is acting
as a base in these reactions to form [MeH2NCH2Ph]Cl
salts in the process of generating a reactive intermedi-
ate. To check that the rate increase was due in part to
the amine’s role as a base, additional experiments were
performed using the stronger base, NEt3. The NEt3 was
expected to form [HNEt3]Cl salts from the deprotonation
of the thiol. Upon the addition of 1 or 2 equiv of
triethylamine, the rate increases, but the rate no longer
The 31P signal observed for 4 (s, δ 25.6) is identical to
that seen in the solutions containing both the thiol and
excess amine. This observation was unexpected in light
of the previous observation that the amine adduct, 2,
forms in the presence of excess amine. Upon the
addition of an even larger excess of N-benzylmethyl-
amine to 4 (15 total equiv), the solution becomes a
darker red, free PPh3 is observed, and a peak at δ 31.5
in the 31P NMR spectrum becomes more prominent.
This new species, 6, is in equilibrium with the thiol
complex, with 4 being greatly favored (eq 15). Amines
(23) Ogawa, A.; Ikeda, T.; Kimura, K.; Hirao, T. J . Am. Chem. Soc.
1999, 121, 5108.

1666 Organometallics, Vol. 19, No. 9, 2000
J ones et al.
Sch em e 3. P r op osed Mech a n ism
F igu r e 5. Plot showing the effect of added NEt3 on rate.
(PPh3)2PdCl2 ) 0.028 M, 4-chlorobenzenethiol ) 0.038 M,
N-benzylmethylamine ) 0.082 M in THF-d8, PCO ) 1.5 atm,
22 °C.
which the addition of 1 equiv of thiol to a solution of
isolated 3 in THF-d8 yielded significant amounts of the
thiocarbamate product (eq 17).25 The consumption of 3,
F igu r e 6. Plot showing the effect of CO concentration.
(PPh3)2PdCl2 ) 0.028 M, 4-chlorobenzenethiol ) 0.038 M,
N-benzylmethylamine ) 0.082 M in THF-d8.
increases to an appreciable extent upon the addition of
more NEt3 (Figure 5). This result can be interpreted to
mean that the rate increase observed for the addition
of 3 or more equiv of N-benzylmethylamine is due to
its use in the rate-determining step as a ligand, with
the first 2 equiv acting as a base.
as followed by 31P NMR spectroscopy, is relatively fast,
however, taking only approximately 15 min. Therefore,
although it is possible that 3 is capable of acting as an
intermediate species in the reaction, its rate of reaction
with thiol must be inhibited in the presence of amine,
since we are able to observe the formation and con-
sumption of 3 over a much longer period of time in the
presence of both amine and thiol. If 3 is acting as an
intermediate, it may be forming the same mixed CO-
inserted amide and thiol complex, 7, as proposed earlier.
Again, rearrangement, followed by reductive elimina-
tion, would yield the thiocarbamate product and a Pd-
(0) species.
Combining the information gathered from the above
results, the overall mechanistic scheme shown in Scheme
3 can be proposed. According to this mechanism, the
thiocarbamate may be formed from either of two routes,
through 4 or through 3. Rates were obtained by follow-
ing the consumption of the thiol complex, 4. Due to the
lack of reactivity between 4 and CO, it is unlikely that
4 is directly involved in the rate-determining step since
there is a definite dependence on CO pressure. There-
fore, although the consumption of 4 was monitored, it
is more likely that an intermediate complex, 6, is
reacting with the CO to give a carbamoyl species such
as 7.
The effect of CO pressure was also examined. The rate
increases proportionately with increased CO pressure
(Figure 6). This observation implies that CO is incor-
porated in or prior to the rate determining step. Free
PPh3 was added to the reaction mixture in the expecta-
tion that a dramatic rate inhibition might be observed
since free PPh3 is observed in the reaction solution. This
proved to be incorrect, and only a slight inhibition was
observed with 0.5-2.0 equiv of PPh3 (see Experimental
Section). With 5 equiv of PPh3, a new species is formed
and the reaction does not proceed to form thiocarbam-
ates.
An attempt was made to determine the reaction order
in thiol. Rates were not obtained for thiol concentrations
above 0.039 M, as a new species is formed and the
reaction does not form thiocarbamates. Using a smaller
concentration of the thiol, 0.019 M, however, did result
in a slightly decreased rate of reaction, indicating that
the rate is proportional to the thiol concentration.
During the reaction, significant amounts of the CO-
inserted amine complex, 3, are observed by 31P NMR
spectroscopy. 3 is the major species formed as 4 is
consumed and then slowly disappears as thiocarbamate
is formed. An independent reaction was performed in
(24) Inglis, T.; Kilner, M. Nature 1972, 239, 13.
(25) See Experimental Section for details.

Carbonylation Routes to Thiocarbamates
Organometallics, Vol. 19, No. 9, 2000 1667
All 1H, 13C, and 31P NMR spectra were recorded on a Bruker
AMX400 spectrometer. All H chemical shifts are reported in
The slight inhibition of rate with added PPh3 is likely
due to its multiple roles in both the equilibrium between
4 and 6 and the subsequent step to yield 7. With added
phosphine, the equilibrium between 4 and 6 is shifted
back toward 4, decreasing the quantity of 6 available
for further reaction. However, added phosphine should
assist in the step for the formation of 7 from 6,
increasing the rate of this step. These two competing
roles for phosphine would result in little effect on the
overall rate of reaction. Unfortunately, a wide range of
phosphine concentrations could not be explored as side
reactions set in.
The effect of added amine is readily understood, as it
will shift the equilibrium between 4 and 6 toward 6.
Two equivalents of amine are consumed by HCl, so that
only the third equivalent or greater has an effect on the
rate. If the actual rate-determining step is the conver-
sion of 6 to 7, the dependence of the rate on CO pressure
is again easily explained. As 6 is more quickly consumed
to form 7 with increased CO, 4 must also be more
quickly consumed to maintain the equilibrium.
As we do observe the formation of significant amounts
of 3 during these reactions, we invoke it as an inter-
mediate species, capable of reacting with thiol to form
7, although it is possible that 3 is a “dead-end” inter-
mediate and proceeds to 7 by way of reversion to 2, 1,
4, and then 6. At this point, we do not know whether
the thiocarbamate is generated primarily through for-
mation of 3 or 4, since the equilibration reactions are
faster than the rate of thiocarbamate formation. The
recent report of carbonylation of ArSNEt2 was believed
to proceed via a similar species, Pd(PPh3)2(SAr)2.24
Finally, from 7, the reductive elimination of thiocar-
bamate is accomplished leaving a Pd(0) complex. Oxida-
tion of the Pd(0) complex using an external oxidant
could result in a catalytic cycle.
1
ppm (δ) relative to tetramethylsilane and referenced using
chemical shifts of residual solvent resonances (THF-d8, δ 3.58).
31P NMR spectra were referenced to external 30% H3PO4 (δ
0.0). GC-MS was conducted on a 5890 Series II Gas Chro-
matograph fitted with an HP 5970 Series mass selective
device. Analyses were obtained from Desert Analytics. A
Siemens SMART system with a CCD area detector was used
for X-ray structure determination.
Gen er a l Sa m p le P r ep a r a tion . Kinetic reactions were
carried out using the following general procedure. A sample
of (PPh3)2PdCl2 (9.6 mg, 0.014 mmol) was placed in a resealable
NMR tube along with methyldiphenylphosphine oxide (used
as an internal standard). THF-d8 was vacuum transferred into
the NMR tube. In the glovebox, N-benzylmethylamine (5.3 µL,
0.041 mmol) was added via microliter syringe. Solid 4-chlo-
robenzenethiol (2.8 mg, 0.019 mmol) was added as well. The
tube was vigorously shaken to give a dark red solution with
some white solids. The solution was freeze-pump-thaw
degassed, and CO was added to the NMR tube while it was
still frozen. Reactions were monitored by NMR spectroscopy.
P r ep a r a tion a n d Isola tion of Th ioca r ba m a te. (PPh3)2-
PdCl2 (138 mg, 0.196 mmol) and methyldiphenylphosphine
oxide were suspended in THF. N-Benzylmethylamine (71.3 µL,
0.588 mmol) and 4-chlorobenzenethiol (39.7 mg, 0.274 mmol)
were added, resulting in a dark red solution. The solution was
freeze-pump-thaw degassed, and CO was added while the
solution was frozen. The solution was stirred overnight.
Solvents were removed in vacuo. The products were dissolved
in minimal THF and filtered through silica gel. The filtrate
was coated onto a silica plate, and a solution of 10% ethyl
acetate in hexanes was used as eluent to give a clean band of
thiocarbamate. The product band was isolated, CH2Cl2 was
added, and the solution was filtered. Solvents were removed
to give a pale yellow oil (yield 27 mg, 47%).
P r ep a r a tion a n d Resu lts of Rea ction s P er for m ed w ith
Excess P P h 3. A sample of (PPh3)2PdCl2 (9.6 mg, 0.014 mmol)
and PPh3 (1.8 mg, 0.0069 mmol or 7.2 mg, 0.027 mmol) was
placed in a resealable NMR tube along with methyldiphen-
ylphosphine oxide (used as an internal standard). THF-d8 was
vacuum transferred into the NMR tube. In the glovebox,
N-benzylmethylamine (5.3 µL, 0.041 mmol) was added via
microliter syringe. Solid 4-chlorobenzenethiol (2.8 mg, 0.019
mmol) was added as well. The solution was freeze-pump-
thaw degassed, and CO was added to the NMR tube while it
was still frozen. The rate of disappearance of 4 for 0.5 equiv
of PPh3 was found to be 2.8 × 10-4 s-1; for 2 equiv, 9.7 × 10-5
s-1. These rates are slower than the rate determined for no
Con clu sion s
The organometallic complex (PPh3)2PdCl2 is capable
of reacting with amines and thiols in the presence of
CO to generate thiocarbamate products. This observa-
tion is significant in that it allows for the replacement
of the environmentally toxic reagent phosgene for the
synthesis of carbamoyl thioesters. Ureas and/or dithio-
carbonates are not formed. Further mechanistic studies
using the Ni analogue are in progress.
added PPh3, 3.4 × 10-4 s-1
.
P r ep a r a tion of (P P h 3)P d Cl2(MeNHCH2P h ) (2). N-Ben-
zylmethylamine (0.550 mL, 4.25 mmol) was added dropwise
to a stirred slurry of (PPh3)2PdCl2 (249 mg, 0.354 mmol) in
THF. The solution became clear yellow-orange after a few
minutes. The solution was then filtered. X-ray quality crystals
of 2 were obtained from a concentrated methylene chloride
Exp er im en ta l Deta ils
Gen er a l Con sid er a tion s. Most manipulations were per-
formed under an N2 atmosphere, either on a high-vacuum line
using modified Schlenk techniques or in a Vacuum Atmo-
spheres Corp. glovebox. Tetrahydrofuran (THF), benzene, and
toluene were distilled from sodium/benzophenone ketyl solu-
tions. THF-d8 was purchased from Cambridge Isotope Lab. and
distilled under vacuum from sodium/benzophenone ketyl solu-
tions. (PPh3)2PdCl2 was purchased from Strem and used as
received. Methyldiphenylphosphine oxide was purchased from
Alfa-ÆSAR. CO was purchased from Air Products. 4-Chlo-
robenzenethiol, N-benzylmethylamine, and dibenzylamine
were purchased from Aldrich Chemical Co. The liquids were
freeze-pump-thawed prior to use. [(PPh3)2PtCl(CO)]+[BF4]-
was synthesized as reported in the literature.26
1
solution by vapor diffusion with hexanes. H NMR (THF-d8):
δ 7.62-7.54 (m, 6H, PPh3), 7.42-7.19 (m, 12H, PPh3, NCH2Ph),
4.31 (t, 1H, NCH2Ph), 4.17 (br, 1H, NH), 3.58 (m, 1H, NCH2-
Ph), 2.65 (q, 3H, NMe). 31P{1H} NMR (THF-d8): δ 28.39. 13C-
{1H} NMR (C6D6): δ 135.3 (PPh3), 135.2 (PPh3), 130.7
(NCH2Ph), 130.5 (NCH2Ph), 130.3 (NCH2Ph), 128.7 (NCH2Ph),
128.1 (PPh3), 56.9 (NCH2Ph), 37.9 (NMe). Solids suitable for
elemental analysis were not obtained due to the sensitivity of
the product to solvent loss.
P r ep a r a tion of (P P h 3)2P d Cl(C(O)NMeCH2P h ) (3). N-
Benzylmethylamine (19.9 µL, 0.154 mmol) was added dropwise
to a resealable NMR tube containing a slurry of (PPh3)2PdCl2
(36.0 mg, 0.0513 mmol) in THF-d8. The solution was freeze-
pump-thaw degassed, and CO was added over the frozen
solution. The sample was thawed and then mixed. After 15
(26) Clark, H. C.; Dixon, K. R.; J acobs, W. J . J . Am. Chem. Soc. 1968,
90, 2259.

1668 Organometallics, Vol. 19, No. 9, 2000
J ones et al.
min, a colorless solution was produced that contained white
salts. The salts were then removed by filtration to give a clean
solution of 3. Solids were obtained from a concentrated THF
solution vapor diffused with hexanes. Two isomers are ob-
served. 1H NMR (THF-d8): δ 7.77 (m, H, PPh3), 7.72 (m, H,
PPh3), 7.39 (m, H, PPh3), 7.32 (m, H, PPh3), 7.04 (m, 2H,
NCH2Ph), 6.98 (m, 2H, NCH2Ph), 6.92 (m, 2H, NCH2Ph), 6.63
(d, 2H, NCH2Ph), 6.38 (d, 2H, NCH2Ph), 4.64(s, 2H, NCH2-
Ph), 3.25 (s, 2H, NCH2Ph), 2.83 (s, 3H, NMe), 1.84 (s, 3H,
NMe). 31P{1H} NMR (THF-d8): δ 19.87, 19.59. 13C{1H} NMR
(THF-d8): δ 183.7 (m, CO), 138.5 (PPh3), 138.4 (PPh3), 135.9
(m, PPh3), 133.5 (t, NCH2Ph), 133.1 (t, NCH2Ph), 130.7 (d,
NCH2Ph), 130.0 (m, NCH2Ph), 129.2 (m, NCH2Ph), 129.0 (m,
NCH2Ph), 128.7 (m, PPh3), 52.2 (s, NCH2Ph), 35.3 (NCH2Ph),
32.9 (NMe), 21.4 (NMe). For C45H40ClNOP2Pd, calcd (found):
C 66.4 (65.7), H 4.92 (4.90).
was approximately 12 h. Frames for 2 were integrated to a
maximum 2θ angle of 56.6° with the Siemens SAINT program
to yield a total of 16 743 reflections, of which 6480 were
independent (Rint ) 1.96%, Rsig ) 2.51%)27 and 5805 were above
2σ(I). Frames for 4 were integrated to a maximum 2θ angle of
46.6° with the Siemens SAINT program to yield a total of
14 157 reflections, of which 5290 were independent (Rint
)
9.36%, Rsig ) 12.49%)27 and 3557 were above 2σ(I). Laue
symmetry revealed monoclinic crystal systems for both crys-
tals, and the final unit cell parameters (at -80 °C) were
determined from the least-squares refinement of three-
dimensional centroids of 6801 reflections for 2 and 4570
reflections for 4.28 Data were corrected for absorption with the
SADABS29 program.
The space groups were assigned as P21/n and P21/c for 2
and 4, respectively. The structures were solved by using direct
methods and refined employing full-matrix least-squares on
F2 (Siemens, SHELXTL,30 version 5.04). For Z values of 4,
there is one molecule and a CH2Cl2 in the asymmetric unit of
2, and one molecule in the asymmetric unit of 4. All of the
non-H atoms for both structures were refined anisotropically,
and hydrogen atoms were included in idealized positions,
except the amine hydrogen (H1) in 2, which was located and
its position and isotropic thermal parameter refined. For 2,
the final refinement revealed a goodness of fit (GOF)31 of 1.036
and final residuals32 of R1 ) 2.29% (I > 2σ(I)), wR2 ) 5.52%
(I > 2σ(I)). For 4, the final refinement revealed a goodness of
fit (GOF)31 of 1.038 and final residuals32 of R1 ) 5.92% (I >
2σ(I)), wR2 ) 14.96% (I > 2σ(I)).
Yellow crystals of 5 were grown from a CHCl3 solution. A
fragment of approximate dimensions 0.06 × 0.08 × 0.14 mm3
was mounted with epoxy on a glass fiber and immediately
placed on the X-ray diffractometer in a cold nitrogen stream
at -60 °C. The X-ray intensity data were collected on an Enraf-
Nonius CAD4 diffractometer with graphite-monochromated
Mo KR radiation. Data were collected to a maximun 2θ angle
of 43.9°. A total of 4566 reflections were collected, of which
4446 were independent (Rint ) 0.032)33 and 3024 were above
3σ(I). Laue symmetry revealed a monoclinic crystal system,
and the final unit cell parameters (at -60 °C) were determined
from the least-squares refinement of reflections. Data were
corrected for absorption with the DIFABS34 program. The
space group was assigned as C2/c (No. 15), and the structure
was solved by heavy-atom Patterson methods35 and expanded
using Fourier techniques (teXsan).36 The non-hydrogen atoms
were refined anisotropically, and the hydrogen atoms were
included in idealized positions. For a Z value of 8, there is half
a molecule in the asymmetric unit and one molecule of CHCl3.
The structure refined to a goodness of fit (GOF)37 of 3.12 and
final residuals38 of 6.1% (I > 3σ(I), Rw ) 9.2% (I > 3σ(I)).
Rea ction of 3 w ith 4-Ch lor oben zen eth iol. (PPh3)2PdCl2
(36.0 mg, 0.051 mmol) was placed in a resealable NMR tube.
THF-d8 was vacuum transferred into the tube, and N-benzyl-
methylamine (19.9 µL, 0.154 mmol) was added via syringe.
The solution was degassed and CO was added. Within 15 min,
the solution became colorless with a white precipitate. The
solids were removed by filtration. 4-Chlorobenzenethiol (8.9
mg, 0.062 mmol) was then added to the solution, producing a
red color. 31P NMR spectroscopy showed that 3 had been
entirely consumed within 15 min. GC-MS analysis of the
solution after 6 h showed formation of thiocarbamate.
P r ep a r a tion of (P P h 3)2P d Cl(SC6H4Cl) (4). 4-Chloroben-
zenethiol (39.7 mg, 0.274 mmol) dissolved in THF was added
dropwise to a stirred slurry of (PPh3)2PdCl2 (160.5 mg, 0.2287
mmol). An instantaneous color change to red-orange was
observed. The solvent was removed in vacuo after 1 h. The
red product was washed with hexanes, then extracted into
THF and filtered. Upon slow evaporation of the THF, crystal-
1
line material was obtained. H NMR (THF-d8): δ 7.65 (m, H,
PPh3), 7.36 (m, H, PPh3), 7.29 (m, H, PPh3), 6.76 (d, 2H, SC6H4-
Cl), 6.46 (d, 2H, SC6H4Cl). 31P{1H} NMR (THF-d8): δ 25.6. 13C-
{1H} NMR (CD2Cl2): δ 137.5 (PPh3), 135.1 (PPh3), 134.1
(SC6H4Cl), 130.8 (SC6H4Cl), 128.8 (SC6H4Cl), 128.2 (PPh3). For
C
42H35Cl2P2Pd: calcd (found): C 62.2 (61.88), H 4.32 (4.04).
P r ep a r a tion of [(P P h 3)(µ-SC6H4Cl)(SC6H4Cl)P t]2 (5).
Piperidine (0.023 mL, 0.24 mmol) was added dropwise to a
THF suspension of [(PPh3)2PtCl(CO)]+[BF4]- (42 mg, 0.048
mmol) with vigorous stirring. The resulting bright yellow
solution was then stirred at room temperature for 45 min until
the color turned to a pale yellow. Solvents were then removed
in vacuo to give a pale yellow oil. The oil was then dissolved
in benzene and filtered. The benzene was removed in vacuo,
and THF was added. 4-Chlorobenzenethiol (7.8 mg, 0.054
mmol) in THF was added to give a bright yellow solution and
a precipitate. This solution was stirred for 35 min before
solvents were removed in vacuo. 1H NMR (CDCl3): δ 7.54 (m,
12H, PPh3), 7.39 (m, 6H, PPh3), 7.19 (m, 12H, PPh3), 6.67 (m,
8H, SC6H4Cl), 6.58 (m, 8H, SC6H4Cl).
2
(27) Rint ) ∑|Fo - Fo2(mean)|/∑[Fo2]; Rsigma ) ∑[σ(Fo2)]/[Fo2].
(28) It has been noted that the integration program SAINT produces
cell constant errors that are unreasonably small, since systematic error
is not included. More reasonable errors might be estimated at 10×
the listed value.
X-r a y Str u ctu r a l Deter m in a tion of 2, 4, a n d 5. Crystals
of 2 were grown from a solution of CH2Cl2/hexanes, and
crystals of 4 were grown by slow evaporation from THF. A
yellow prism of 2 with approximate dimensions 0.18 × 0.18 ×
0.16 mm3 was mounted under Paratone-8277 on a glass fiber
and immediately placed in a cold nitrogen stream at -80 °C
on the X-ray diffractometer. A thin purple plate of 4 with
approximate dimensions 0.18 × 0.12 × 0.08 mm3 was mounted
in a similar manner, and data were also collected at -80 °C.
The X-ray intensity data for both crystals were collected on a
standard Siemens SMART CCD area detector system equipped
with a normal focus molybdenum-target X-ray tube operated
at 2.0 kW (50 kV, 40 mA). For both crystals, a total of 1321
frames of data (1.3 hemispheres) were collected using a narrow
frame method with scan widths of 0.3° in ω and exposure times
of 30 s/frame using a detector-to-crystal distance of 5.09 cm
(maximum 2θ angle of 56.6°). The total data collection time
(29) The SADABS program is based on the method of Blessing;
see: Blessing, R. H. Acta Crystallogr., Sect A 1995, 51, 33.
(30) SHELXTL: Structure Analysis Program, version 5.04; Siemens
Industrial Automation Inc.: Madison, WI, 1995.
(31) GOF ) [∑[w(Fo - Fc2)2]/(n - p)]1/2, where n and p denote the
2
number of data and parameters
2
2
2
1/2
2
2
(32) R1 ) (∑||(Fo| - Fc||)/∑|Fo|; wR2 ) [∑[w(Fo - Fc) ]/∑[w(Fo) ]]
where w ) 1/[σ2(Fo2) + (aP)2 + bP] and P ) [(max;0,Fo2) + 2Fc2]/3.
2
2
(33) Rint ) ∑|Fo - Fo2(mean)|/∑[Fo
]
(34) DIFABS: Walker, N.; Stuart, A. Acta Crystallogr. 1983, A39,
158-166. An empirical absorption correction program.
(35) PATTY: Beurskens, P. T.; Admiraal, G.; Beurskens, G.; Bos-
man, W. P.; Garcia-Granda, S.; Gould, R. O.; Smits, J . M. M.; Smykalla,
C. The DIRDIF program system, Technical Report to the Crystal-
lography Laboratory; University of Nijmegen: The Netherlands, 1992.
(36) DIRDIF92: Beurskens, P. T.; Admiraal, G.; Beurskens, G.;
Bosman, W. P.; Garcia-Granda, S.; Gould, R. O.; Smits, J . M. M.;
Smykalla, C. The DIRDIF program system, Technical Report to the
Crystallography Laboratory; University of Nijmegen: The Nether-
lands, 1992. teXsan, Crystal Structure Analysis Package; Molecular
Structure Corporation: 1985 and 1992.

Carbonylation Routes to Thiocarbamates
Organometallics, Vol. 19, No. 9, 2000 1669
Su p p or tin g In for m a tion a va ila ble includes kinetic plots
for the production of thiocarbamate, a GCMS trace of the
reaction mixture, and tables of data collection parameters,
atomic coordinates, distances and angles, and anisotropic
thermal parameters for 2, 4, and 5. This material is available
free of charge via the Internet at http://pubs.acs.org.
Ack n ow led gm en t is made to the National Science
Foundation (CHE-9616601) for their support of this
work.
(37) GOF ) (∑[w(Fo - Fc)2]/(No - Nv)]1/2, where No ) no. of
observations and Nv ) no. of variables.
(38) R ) (∑||(Fo| - Fc||)/∑|Fo|, Rw ) [∑[w(Fo - Fc)2]/∑[w(Fo2)]]1/2 where
w ) 1/[σ2(Fo)].
OM990931N