Employment of palladium pincer-complexes in phenylselenylation of organohalides

Article abstract of DOI:10.1021/jo051266x

Palladium pincer-complex-catalyzed selenylation of propargyl-, allyl-, benzyl-, and benzoyl halides could be achieved under mild reaction conditions employing trimethylstannylphenylselenide as selenylating agent. This reaction has a high functional group

Full text of DOI:10.1021/jo051266x

Employment of Palladium Pincer-Complexes in
Phenylselenylation of Organohalides
Olov A. Wallner and K a´ lm a´ n J. Szab o´ *
Arrhenius Laboratory, Department of Organic Chemistry, Stockholm University,
SE-106 91 Stockholm, Sweden
kalman@organ.su.se
Received June 20, 2005
Palladium pincer-complex-catalyzed selenylation of propargyl-, allyl-, benzyl-, and benzoyl halides
could be achieved under mild reaction conditions employing trimethylstannylphenylselenide as
selenylating agent. This reaction has a high functional group tolerance as carbmethoxy, tosylamino,
nitro, aryl bromide, and unprotected hydroxy groups are tolerated. Mechanistic studies indicate
that the catalytic cycle is initiated by formation of a selenium-coordinated palladium pincer-complex,
which subsequently reacts with the electrophilic substrate. DFT calculations were performed to
explore the mechanism of the transfer of the organoselenium group from palladium to the substrate.
These modeling studies revealed some interesting differences between the mechanism of the
organoselenium group transfer and (the previously published) organostannane transfer reactions.
1
. Introduction
from the reactivity of commonly used palladium-catalysts
such as Pd(PPh and Pd (dba) . Employment of pincer-
3
)
4
2
3
The structural properties and catalytic reactivity of
complexes offers an attractive synthetic route to prepara-
tion of organometallic compounds due to three important
features: (i) a strong terdentate ligand-metal interaction
provides high stability and durability for the catalyst; (ii)
coordination of the terdentate ligand restricts the number
of the available coordination sites on palladium to a
single one; and (iii) under ambient conditions the oxida-
tion-state of palladium is restricted to +2. Because of
these properties, palladium pincer-complexes have proven
to be robust and effective as catalysts, which are reflected
by their high functional group tolerance and broad
synthetic scope.1
palladium pincer-complexes (e.g., 1a-c) have attracted
considerable attention in inorganic and, more recently,
in organic chemistry.1 We have previously shown that
in the presence of dimetallic organotin reagents pal-
ladium pincer-complexes (1a,b) can be employed as
-6
7,8
efficient catalysts for synthesis of allenyl stannanes,
8
9
allenyl silanes, and allyl stannanes. In these reactions,
palladium pincer-complex catalysts have been employed
to transfer various organometallic groups from a di-
metallic tin reagent (i.e., (Me
propargylic and allylic substrates. It was shown that
the reactivity of pincer-complexes is completely different
3
Sn)
2
or Me
3
SnSiMe7
2
Ph) to
,8
-13
The above-mentioned favorable features of pincer-
complexes inspired us to further extend the catalytic
(
1) Albrecht, M.; Koten, G. v. Angew. Chem., Int. Ed. 2001, 3750.
2) Dupont, J.; Pfeffer, M.; Spencer, J. Eur. J. Inorg. Chem. 2001,
14-19
utility of these species in organoselenium chemistry.
(
1
917.
(
(
(
3) Boom, M. E. v. d.; Milstein, D. Chem. Rev. 2003, 103, 1759.
(10) Solin, N.; Kjellgren, J.; Szab o´ , K. J. Angew. Chem., Int. Ed.
2003, 42, 3656.
(11) Solin, N.; Kjellgren, J.; Szab o´ , K. J. J. Am. Chem. Soc. 2004,
126, 7026.
(12) Solin, N.; Wallner, O. A.; Szab o´ , K. J. Org. Lett. 2005, 7, 689.
(13) Sebelius, S.; Olsson, V. J.; Szab o´ , K. J. J. Am. Chem. Soc. 2005,
127, 10478.
(14) Nicolau, K. C.; Petasis, N. A. Selenium in Natural Product
Synthesis; CIS: Philadelphia, 1984.
(15) Paulmier, C. Selenium Reagents and Intermediates in Organic
Synthesis; Pergamon Press: Oxford, 1986.
4) Singleton, J. T. Tetrahedron 2003, 59, 1837.
5) Beletskaya, I. P.; Cheprakov, A. V. J. Organomet. Chem. 2004,
6
2
1
1
89, 4055.
(
6) Dupont, J.; Consorti, C. S.; Spencer, J. Chem. Rev. 2005, 105,
527.
(
7) Kjellgren, J.; Sund e´ n, H.; Szab o´ , K. J. J. Am. Chem. Soc. 2004,
26, 474.
(
8) Kjellgren, J.; Sund e´ n, H.; Szab o´ , K. J. J. Am. Chem. Soc. 2005,
27, 1787.
(9) Wallner, O. A.; Szab o´ , K. J. Org. Lett. 2004, 6, 1829.
1
0.1021/jo051266x CCC: $30.25 © 2005 American Chemical Society
Published on Web 10/20/2005
J. Org. Chem. 2005, 70, 9215-9221
9215

Wallner and Szabo´
SCHEME 1. Palladium Pincer-Complex-Catalyzed
Phenylselenylation
TABLE 1. Palladium Pincer-Complex-Catalyzed
Phenylselenylation of Organohalidesa
Because palladium pincer-complexes are known to cata-
lyze the transfer of organostannyl and organosilyl groups
to electrophiles from dimetallic trimethyltin reagents,7,9
we have studied the possibility of employment of tri-
20
methylstannylphenylselenide (2) as a source of phen-
ylselenyl group in related reactions (Scheme 1).2
0-26
Thus,
in this paper we report our results on palladium pincer-
complex-catalyzed transfer of phenylselenyl group from
2
to propargyl-, allyl-, benzyl-, and benzoyl halides. The
mechanism of the phenylselenyl transfer has been in-
vestigated by stoichiometric reactions and DFT modeling
studies. We have also compared the mechanism of the
organoselenium and -stannane8 transfer reactions to
propargylic substrates. The DFT results obtained for the
corresponding TS structures of these processes indicate
some very interesting mechanistic differences between
the two metal transfer processes.
2. Results and Discussion
Considering the fact that SeCSe complex 1a displayed
8
the highest activity in trimethyltin transfer processes,
we employed this readily available27 catalyst in the
selenylation reactions. Indeed, various organic sub-
strates, involving propargyl-, allyl-, benzyl-, and benzoyl
halides (3a-m), readily undergo (Scheme 1) halogen-
selenium exchange in the presence of 2 and catalytic
amounts (2 mol %) of 1a affording organoselenides 4a-m
in good to excellent yield (Table 1). Analysis of the crude
reaction mixtures indicated that the catalytic transfor-
mations proceeded very cleanly providing the selenated
products almost quantitatively, which led to high isolated
yields. The only exception appears to be 4b (entry 2, 66%
yield), which probably underwent partial decomposition
under the isolation process. The reactions were conducted
a
All reactions were conducted in THF using 2 mol % of 1a.
b
Isolated yield.
under mild and neutral conditions without employment
of inert gas atmosphere. In the absence of catalyst 1a,
only traces of the selenylated products were formed.
2
5
In the presented reactions, we did not observe forma-
tion of organostannanes indicating that exclusively the
phenylselenyl group of 2 was transferred to the electro-
phile. The reaction also displays a remarkable high
functional group tolerance as carbmethoxy, tosylamino,
nitro, aryl bromide, and unprotected hydroxy groups are
tolerated. The regioselectivity of the substitution involv-
ing allylic and propargylic substrates is excellent as only
products resulting from the direct substitution of the
halogenated carbon are formed. Thus, employment of
propargyl bromides 3a-d afforded propargyl selenides
(
(
(
(
(
16) Liotta, D. Organoselenium Chemistry; Wiley: New York, 1987.
17) Wirth, T. Top. Curr. Chem. 2000, 208.
18) Beletskaya, I.; Moberg, C. Chem. Rev. 1999, 99, 3435.
19) Ogawa, A. J. Organomet. Chem. 2000, 611, 463.
20) Poleschner, H.; Heydenreich, M.; Schilde, U. Eur. J. Inorg.
Chem. 2000, 1307.
21) Nishiyama, Y.; Tokunaga, K.; Sonoda, N. Org. Lett. 1999, 1,
725.
22) Beletskaya, I. P.; Sigeev, A. S.; Peregudov, A. S.; Petrovskii, P.
V. Mendeleev Commun. 2000, 213.
23) Beletskaya, I. P.; Sigeev, A. S.; Peregudov, A. S.; Petrovskii, P.
V. Mendeleev Commun. 2000, 127.
24) Beletskaya, I. P.; Sigeev, A. S.; Peregudov, A. S.; Petrovskii, P.
V. J. Organomet. Chem. 2000, 605, 96.
25) Beletskaya, I. P.; Sigeev, A. S.; Peregudov, A. S.; Petrovskii, P.
V. Russ. J. Org. Chem. 2001, 37, 1463.
26) Nishiyama, Y.; Kawamatsu, H.; Funato, S.; Tokunaga, K.;
Sonoda, N. J. Org. Chem. 2003, 68, 3599.
27) Yao, Q.; Kinney, E. P.; Zheng, C. Org. Lett. 2004, 6, 2997.
(
1
(
(
4
a-d (entries 1-4) in good to excellent yields. When
(
unsubstituted propargyl bromide 3a was used as elec-
trophile, traces (about 2%) of allenyl selenide product
were also observed in the crude reaction mixture. Pre-
dominant formation of propargyl selenides from propar-
gyl bromides 3a, 3c, and 3d (entries 1, 3, and 4) is in
(
(
(
9216 J. Org. Chem., Vol. 70, No. 23, 2005

Pincer-Complexes in Phenylselenylation of Organohalides
SCHEME 2. Stoichiometric Reaction of 2 with 1a
ppm) was vanished, and at the same time the two signals
of 1a (427.1 and 424.9 ppm) were broadened and shifted
downfield to 432.9 and 425.7 ppm, respectively, and a
new peak appeared at 83.8 ppm. Once again, the exten-
sive line-broadening of the 77Se NMR signals did not
allow clear identification of the product. Our further
efforts to isolate and purify the product of the stoichio-
metric reactions were fruitless, because of its low stabil-
ity.
Although the NMR studies clearly indicated that in the
above stoichiometric process 2 was consumed and that
the NMR spectrum of 1a underwent substantial changes,
conclusive results for the identity of the reaction product
could not be obtained from these experiments. On the
sharp contrast to the corresponding palladium pincer-
complex-catalyzed stannylation and silylation reaction of
propargyl chlorides, which gave the corresponding allenyl
stannane and silane products.7,8
Allyl bromides 3e-g reacted also with high regio-
selectivity affording primary allyl selenides 4e-g as the
sole product in high yields (entries 5-7). Secondary allyl
bromide 3h could also be employed as electrophile (entry
other hand, our previous studies with (Me
SnSiMe
3
Sn)
Ph clearly indicated7 transmetalation of these
2
2 3
or Me -
,8
8); however, this substitution reaction was sluggish even
dimetallic reagents with pincer-complex 1b, and therefore
we assume that the stoichiometric reaction of 1a and 2
results in formation of a phenyl-selenide coordinated
complex 1d (Scheme 2). Formation of 1d is also indicated
by the fact that addition of benzoyl chloride derivative
3m to the reaction mixture of 1b and 2 subjected to the
above 77Se NMR studies results in immediate formation
of 4m and recovery of 1a.
at 60 °C, and therefore an elongated reaction time was
required. Benzyl bromides (3i-k) were also effective as
electrophiles affording benzyl selenides (4i-k) in high
yields (entries 9-11); however, 3i-k reacted somewhat
slower than propargyl and allyl bromides 3a-g. Benzoyl
chlorides (3l,m), on the other hand, reacted rapidly and
afforded carboselenoates (4l,m) in excellent yields in 3 h
(
entries 12 and 13). It was found that elongated reaction
DFT Modeling of the Phenylselenyl Transfer
Reaction. Based on the assumption that the initial step
of the pincer-complex-catalyzed reaction is formation of
1d, we performed DFT modeling studies to explore the
mechanism of the organoselenium group transfer from
the palladium atom to the organic substrate. In these
studies, we investigated the potential energy surface of
the organoselenium transfer to a propargylic substrate
(cf., entries 1-4), to compare the mechanism of this
times (over 8 h) often resulted in a partial disproportion-
ation of 2 to give Ph Se, decreasing the yield of the
2
2
5
corresponding selenylation reactions. We have also at-
tempted to use allyl-, propargyl-, and benzyl-chlorides
instead of the corresponding bromides as electrophiles;
however, under mild conditions these reactions proceed
with a low conversion of the chloride substrates to the
corresponding phenylselenides.
8
process and the previously reported reactions involving
organotin species.
3. Mechanistic Considerations
Computational Methods. All geometries were fully
optimized employing a Becke-type28 three-parameter
density functional model B3PW91 using a double-ú-
Stoichiometric Reactions. Our previous studies
have shown that pincer-complexes such as 1b do not react
2
9-31
with propargyl or allyl halogenides at ambient temper-
(DZ)+P basis constructed from the LANL2DZ
basis
8
ature. On the other hand, dimetallic organostannanes,
by adding one set of d-polarization functions to the heavy
atoms (exponents: C, 0.630; Cl, 0.514; Se, 0.338; Sn,
0.183) and one set of diffuse d-functions on palladium
(exponent: 0.0628). Harmonic frequencies have been
calculated at the level of optimization for all structures
to characterize the calculated stationary points and to
determine the zero-point energies. Fully optimized tran-
sition state structures 5a, 5b, and 8 have been character-
ized by a single imaginary frequency, while the rest of
the optimized structures possess only real frequencies.
All calculations were carried out by employing the
Gaussian 03 program package.32
3 2 3 2
such as (Me Sn) or Me SnSiMe Ph, undergo transmeta-
lation with 1b to give the corresponding organometal (tin
or silicon) coordinated pincer-complex, which could be
1
19
29
7,8
observed by Sn and Si NMR spectroscopy. Thus,
to elucidate the mechanism of the organoselenium trans-
fer reaction, we have conducted a series of stoichiometric
experiments for the reaction of SeCSe complex 1a and
1
77
selenium-stannane 2 monitored by H NMR and Se
NMR at -9 °C (Scheme 2). When the pale-yellow solution
3
of 1a was treated with 1.1 equiv of 2 in CDCl , the color
of the reaction mixture immediately turned to dark
1
orange. The H NMR spectrum of 1a also underwent
Reaction Profile for the Selenium Transfer Pro-
cess. The scope of the DFT studies involved exploration
of the role of the palladium catalyst as well as investiga-
tion of the regioselectivity of the catalytic selenylation
reaction. Because of computational limitations, we em-
ployed slightly simplified model systems to describe the
mechanism of the experimentally studied reaction of 3a
with 2. Accordingly, we have approximated (Scheme 3)
substantial changes upon addition of 2, as the charac-
teristic signals from the methylene groups of 1a (4.73,
.72, 4.32, and 4.27 ppm) were considerably broadened.
4
Unfortunately, the extensive line-broadening of 50 Hz
encumbered the identification of the product of the
stoichiometric reaction. We have also analyzed the reac-
tion mixture of the stoichiometric reaction of 1a and 2
7
7
77
using Se NMR spectroscopy. The Se NMR spectrum
of pure 1a displays two signals at 427.1 and 424.9 ppm,
(28) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(29) Dunning, T. H.; Hay, P. J. Modern Theoretical Chemistry;
Plenum: New York, 1977; Vol. 3.
indicating the presence of two diastereomeric forms,27
while the selenium atom of pure 2 resonates at 0.5 ppm.
(
30) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270.
When 2 was added to 1a, the 77Se NMR signal of 2 (0.5
(31) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
J. Org. Chem, Vol. 70, No. 23, 2005 9217

Wallner and Szabo´
SCHEME 3. Reactions Studied by DFT Modeling; Paths I and II Describe the Pincer-Complex-Catalyzed
Reaction, While Path III Is the Uncatalyzed Reaction
SCHEME 4. PES of the Phenylselenylation
the phenylselenyl groups in palladium pincer-complex 1a
and 1d with methylselenyl groups (1e and 1f), and the
Reaction: a) Paths I and II, b) Path IIIa
bromine atom of 3a was approximated with a chlorine
atom (6). The mechanistic discussions (vide infra) are
restricted to the analysis of the TS geometries and the
relative activation energies of the different reaction
pathways, which are probably not affected by these
approximations. Furthermore, employment of 6 as sub-
strate in the DFT studies simplifies the comparison of
the mechanistic features of the present selenium transfer
8
and the previously published tin transfer reactions.
Using the above-described model systems, we have
investigated three different pathways (Scheme 3) for the
phenylselenylation of propargyl chloride 6. The first
pathway (Path I) involves transfer of the phenylselenyl
a
Energies in kcal mol-1
.
group from pincer-complex 1f to give propargyl selenide
a, while the second pathway (Path II) is the correspond-
4
-1
exothermic (-21.3 kcal mol ) than formation of propar-
gyl product 4a (-14.8 kcal mol ). The activation energy
of the uncatalyzed reaction of 2 with 6 (Path III) is
considerably higher (37.8 kcal mol ) than the corre-
sponding reactions (Path I and II) involving pincer-
complex 1f. These results are in agreement with our
experimental findings that the palladium-catalyzed reac-
tion of 2 with 3a leads predominantly to propargyl
product 4a (entry 1) and that the selenylation reaction
requires palladium catalysis.
ing reaction affording allenyl selenide 7. To clarify the
role of the palladium catalyst in this reaction, we have
investigated the mechanism of the uncatalyzed reaction
-
1
-
1
(
Path III) involving direct transfer of the phenylselenyl
group from 2 to the propargylic substrate to give 4a.
According to the DFT results, the reaction between 1e
-
1
and 2 is exothermic by 2.6 kcal mol affording 1f
Schemes 3 and 4), which is the active selenylating agent
(
in the selenium transfer process. The fact that 1f is such
a stable species also supports our assumptions that
stoichiometric reaction (Scheme 2) of 1a and 2 results in
selenium coordinated complex 1d.
Inspection of the calculated structures of complexes 1e
and 1f, as well as TS structures 5a,b and 8, reveals some
interesting mechanistic details (Figure 1). Complex 1e
The reaction of 1f and 6 affording propargyl selenide
2
has C symmetry with a tightly coordinated pincer ligand.
4
a (Path I) proceeds through TS 5a requiring an activa-
-1
The palladium-carbon bond is relatively short (2.004 Å),
while the palladium-selenium bond lengths (2.432 Å) are
similar to the Pd-Se bond lengths (2.42-2.44 Å) obtained
from X-ray data for a related complex.27 The phenyl-
selenide coordinated structure 1f is characterized by a
similar, strong metal-ligand bonding (Pd-C; 2.039 Å,
Pd-Se; 2.433 and 2.435 Å, respectively); however, the
newly formed Pd-Se bond is relatively long, 2.540 Å
tion energy of 21.5 kcal mol . The subsequent formation
of propargyl selenide 4a is strongly exothermic by 14.8
-
1
kcal mol (Scheme 4a). The second pathway (Path II)
leads to formation of allenyl selenide 7 via TS structure
5
b and requires somewhat higher activation energy (22.8
-
1
kcal mol ) than Path I resulting in 4a. Interestingly,
although the activation barrier of Path II is higher than
that of Path I, formation of allenyl selenide 7 is more
(
Figure 1). In TS 5a, the characteristic five-member ring
topology of the sidearms of the pincer-complex is pre-
served, while the bond length to the phenylselenide group
is elongated (Pd-Se ) 2.574 Å). The distance of the
(
32) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,
V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene,
M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma,
K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.;
Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui,
Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challa-
combe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Gonzalez, C.; Pople, J. A. Gaussian 03; Gaussian, Inc.: Pittsburgh,
PA, 2003.
propargylic carbon (C
R
) to the selenium atom (2.482 Å)
and the C -Cl bond (2.504 Å) are very long, indicating
R
that the formation of the carbon-selenium bond and the
cleavage of the carbon-chloride bond is a concerted
process. Furthermore, the Se-C
R
-Cl bond angle is
1
51.9°, which is close to the characteristic displacement
N
angle of the S 2-reactions (180°). These geometrical
features indicate that in TS structure 5a the selenium
atom is involved in an S 2-type of displacement of the
leaving group (Cl) without direct involvement of the
palladium atom (Pd-C ) 3.313 Å). This hypothesis is
N
R
9218 J. Org. Chem., Vol. 70, No. 23, 2005

Pincer-Complexes in Phenylselenylation of Organohalides
FIGURE 1. Calculated geometries of 1e and 1f and the TS’s of the selenylation reactions (5a,b and 8). Bond lengths are given
-
1
in angstroms, angles in degrees, energies in kcal mol , and the ZPV-corrected energies in italics.
supported by other geometrical features; the Pd-Se-C
R
modeling results raise the question on the catalytic role
of palladium in the substitution process.
angle is 81.9° and the Pd-Se-CPh angle is 108.7°, clearly
indicating that the nucleophilic attack is initiated by the
lone-pair electrons of the selenium atom. The TS struc-
ture of Path II (5b) leading to allenyl product 7 is
characterized by features similar to those of 5a. The Se-
The above analysis clearly indicates that the nucleo-
philic substitution of the chloride atom in 6 is initiated
by the lone-pair orbital of the selenium atom. Inspection
of the HOMO orbital of 1f and 2 clearly shows that in
C
γ
distance (2.230 Å) and the C
indicate nucleophilic displacement in an S
long Pd-C distance (3.857 Å) indicates that the pal-
R
-Cl distance (1.970 Å)
both species the main component of this MO is the n
π
N
2′ fashion. The
lone-pair of selenium (Figure 2). This explains the similar
geometry of the TS structures of the catalyzed (5a,b) and
uncatalyzed (8) processes. However, the orbital energy
γ
ladium atom is not directly involved in the substitution
reaction. Similarly to 5a, TS structure 8 of the uncata-
lyzed reaction (Path III) also displays the attributes of
of the n lone-pair in palladium complex 1f (-4.67 eV) is
π
much higher than the corresponding orbital energy
(-6.00 eV) in 2, indicating that the lone-pair electrons
of 1f are much easier accessible by electrophiles (such
as 6) than the lone-pairs of 2. Accordingly, formation of
1f from 2 leads to an increase of the lone-pair energy of
selenium, increasing the nucleophilicity of the selenium
atom. The high energy of the lone-pair electrons of
selenium in 1f can probably be explained by a four-
an S
Å) distances as well as a wide Se-C
The long Sn-C distance (3.399 Å) shows that the tin
N
2 process: long Se-C
R
(2.325 Å) and C
R
-Cl (2.636
R
-Cl angle (145.0°).
R
atom is not directly involved in the substitution process.
These geometrical features indicate a similar mechanism
(
S
N
2/ S
N
2′) for the palladium-catalyzed (Paths I and II)
and the uncatalyzed (Path III) reactions. However, these
J. Org. Chem, Vol. 70, No. 23, 2005 9219

Wallner and Szabo´
FIGURE 2. HOMO orbital of complex 1f (ꢀHOMO ) -4.67 eV)
and 2 (ꢀHOMO ) -6.00 eV).
SCHEME 5. Differences in the Regioselectivity of
the Selenylation and the Stannylation Reactions
Catalyzed by Palladium Pincer-Complexes
FIGURE 3. TS structures of selenylation (5a and 5b) and
8
stannylation (9a and 9b) of propargyl chloride determined
at the B3PW91/LANL2DZ+P level of theory.
SCHEME 6. The Catalytic Cycle of the
Phenylselenylation Reaction
π σ
electron interaction between the n /n orbitals of sele-
nium and the high-lying nonbonding d-orbitals of palla-
dium. These types of interactions do not occur in 8
because of the lack of d-valence orbitals in tin.
In summary, the catalytic effect of palladium can be
explained by the presence of high energy lone-pair
electrons in 1f, which leads to an increase of the nucleo-
philicity of the selenium atom in 1f compared to 2. This
higher nucleophilicity is reflected by the fact that pal-
ladium-catalyzed nucleophilic selenylation requires lower
-
1
activation barrier (21-22 kcal mol ) than the uncata-
-
1
lyzed process (37.8 kcal mol ).
Comparison of the Mechanism of the Organose-
lenium and -tin Transfer Reactions. As shown above,
the phenylselenylation of propargyl chloride (6) affords
the propargylic product 4a via displacement of the
chloride by the lone-pair electrons of the selenium atom.
This regioselectivity is in sharp contrast to the reaction
through TS structure 9a. Because of the different mech-
anism in the selenylation process, formation of the
propargyl product (4a) via 5a is favored over the forma-
tion of the allenyl derivative (7) by 1.3 kcal mol . A
possible explanation is that the TS structure of the direct
stannylation process (9a) is sterically more congested
than the corresponding TS of the selenylation process
(5a), in which the palladium atom does not participate
in the displacement of the leaving group.
-
1
3 2
of (Me Sn) with 6, which exclusively affords the allenyl
8
product (Scheme 5), indicating mechanistic differences
between the palladium-catalyzed selenylation and stan-
nylation process of propargylic substrates.
The differences in the TS structures of the selenylation
and stannylation reactions are highlighted in Figure 3.
The Catalytic Cycle of the Phenylselenyl Trans-
fer Reaction. On the basis of the above presented
results, we propose a catalytic cycle for the palladium
pincer-complex-catalyzed phenylselenyl transfer from 2
to the organic substrates (Scheme 6). Thus, the catalytic
cycle of the substitution reaction is initiated by formation
of intermediate 1d by transmetalation of 2 with 1a. In
the transmetalation process, the phenylselenyl group is
transferred to palladium. Formation of 1d is then fol-
8
As mentioned above, the selenylation reaction is based
on nucleophilic displacement of the leaving group by the
high-lying lone-pair electrons of the selenium atom. On
the other hand, the stannylation reaction (9a,b) is
initiated by the nucleophilic attack of the palladium-
8
tin σ-orbital. This is indicated by the relatively short
Pd-C
stannylation reaction (Pd-C
2.638 Å in 9b), which is shorter by 0.35-1.23 Å than
the corresponding distances in 5a and 5b. Furthermore,
the Sn-C -Cl angle in 9a (132°) is more acute by almost
0° than the Se-C
R
and Pd-C
γ
distances in the TS structures of the
8
R
) 2.961 in 9a and Pd-C
γ
N
lowed by an S 2-type of displacement of the bromide
)
leaving group by the high-lying lone-pair electrons of
selenium regenerating catalyst 1a. It is interesting to
point out that the mechanistic features of the above-
described S 2-type of selenation process show an inter-
N
esting analogy to the fluoride ion-catalyzed selenyl
R
2
R
-Cl angle (150.9°) in 5a. According
8
to the DFT calculations, formation of allenyl stannane
via TS structure 9b requires lower activation energy (by
transfer reaction developed by Beletskaya and co-work-
-
1
22,23
1
3 kcal mol ) than formation of the propargyl product
ers.
An important feature of the presented catalytic
9220 J. Org. Chem., Vol. 70, No. 23, 2005

Pincer-Complexes in Phenylselenylation of Organohalides
cycle (Scheme 6) is that the catalytic process occurs in a
single free coordination site of palladium, as well as the
oxidation state of palladium is restricted to +2.
pincer-complex-catalyzed stannylation process, in which
the palladium-tin σ-bond initiates the nucleophilic dis-
placement of the leaving group. The different mecha-
nisms in the selenylation and stannylation reactions lead
to different regioselectivity for the substitution of pro-
pargylic substrates.
Several publications have appeared in the literature
for coupling of selenium-tin reagents with organohalides
2
1-26
for preparation of organoselenides.
Beletskaya, Nish-
have shown that
iyama, and their co-workers2
1,24-26
Pd(PPh ) efficiently catalyzes the coupling of tributyl-
3 4
5
. Experimental Section
stannylphenylselenide with organohalides. The presented
palladium pincer-complex-catalyzed phenylselenylation
of organohalides represents an interesting synthetic
alternative to the palladium(0) catalyzed and other
transformations. The catalytic process with 1a proceeds
via a redox-free reaction mechanism extending the
functional group tolerance (e.g., unprotected hydroxy
group and aryl bromide) of the palladium-catalyzed
selenylation reactions. In addition, the catalytic reactions
can be performed in the absence of phosphines under so-
called ligand-free conditions without employment of inert-
gas atmosphere or use of manifold technique.
General Procedure for the Palladium Pincer-Complex-
Catalyzed Phenylselenylation. Trimethylstannylphenyl-
2
0
selenide 2 (88 mg, 0.275 mmol) and the corresponding
electrophile (0.25 mmol) in THF (1.0 mL) were added to
catalyst 1a (2.8 mg, 0.005 mmol) at room temperature. This
reaction mixture was then stirred for the allotted temperatures
and times given in Table 1. Thereafter, the solvent was
removed and the product was purified by silica gel chroma-
tography.
Phenyl-(2-propynyl)-selenide (4a). This product was
prepared by the above general procedure. For purification,
silica gel chromatography was employed using pentane as
1
eluent. The obtained H NMR data are in agreement with the
3
3 1
3
published literature values. H NMR: δ 2.25 (t, J ) 2.7 Hz,
3
2
4. Conclusions
1H), 3.49 (d, J ) 2.7 Hz, J (Se, H) ) 13.0 Hz, 2H), 7.30 (m,
H), 7.61 (m, 2H). C NMR: δ 12.7, 72.1, 81.2, 128.1, 129.5,
1
3
3
1
Palladium pincer-complexes can be employed as cata-
29.8, 133.6.
lysts in the coupling of a phenylselenyl group with
organohalides extending the synthetic scope of the phen-
ylselenylation reactions. These reactions proceed under
mild and neutral conditions, and therefore many func-
Acknowledgment. This work was supported by the
Swedish Natural Science Research Council (VR). The
computer time granted by the Swedish National Alloca-
tion Committee (SNAC) is greatly appreciated. We
acknowledge the financial support by the Knut and Alice
Wallenberg Foundation (KAW).
2
tionalities, including COOMe, OH, NHTs, NO , and even
aryl Br, are tolerated. The selenylation of functionalized
allylic and propargylic substrates proceeds with excellent
regioselectivity as only the linear products are formed.
Mechanistic studies for selenylation of propargyl chloride
Supporting Information Available: Detailed experi-
(
6) indicate that the lowest energy path provides pro-
mental section, computational details of the DFT studies, and
1
3
C NMR spectra of compounds 4a-m. This material is
pargyl selenide product via a phenylselenyl-coordinated
palladium pincer-complex. In this reaction, the lone-pair
electrons of the selenium atom initiate the nucleophilic
available free of charge via the Internet at http://pubs.acs.org.
JO051266X
N
displacement of the leaving group in an S 2-type of
process. This mechanism is significantly different from
the previously reported mechanism of the palladium
(
33) Donkevoort, J. G.; Gordon, A. R.; Johnstone, C.; Kerr, W. J.;
8
Lange, U. Tetrahedron 1996, 52, 7391.
J. Org. Chem, Vol. 70, No. 23, 2005 9221