Studies on the 1,2-migrations in Pd-catalyzed negishi couplings with JosiPhos ligands

Article abstract of DOI:10.1021/jo801824e

(Chemical Equation Presented) We report an initial investigation with the goal of determining the factors that control the 1,2-migration in the Negishi cross-coupling, which is promoted by palladium catalyst systems generated with JosiPhos ligands. Severa

Full text of DOI:10.1021/jo801824e

Studies on the 1,2-Migrations in Pd-Catalyzed Negishi Couplings
with JosiPhos Ligands
Anders T. Lindhardt, Thomas M. Gøgsig, and Troels Skrydstrup*
Center of Insoluble Protein Structures (inSPIN), Department of Chemistry, Interdisciplinary Nanoscience
Center, UniVersity of Aarhus, Langelandsgade 140, 8000 Aarhus C, Denmark
ts@chem.au.dk
ReceiVed August 14, 2008
We report an initial investigation with the goal of determining the factors that control the 1,2-migration
in the Negishi cross-coupling, which is promoted by palladium catalyst systems generated with JosiPhos
ligands. Several of the factors that were demonstrated to be important for the 1,2-migration include (1)
the nucleophilicity of the organometallic reagent, which possibly influences the transmetalation step in
direct competition with the intermediate ꢀ-hydride elimination of the alkenyl Pd(II) species; (2) the
structural features of the vinyl tosylates and phosphates, in which substrates possessing a bulky C1
substituent displayed highest propensity for undergoing the 1,2-migration under the coupling reaction
conditions; and (3) the structure of the JosiPhos ligand, where both the sterical bulk and choice of
substitutents on the ferrocenyl phosphino group greatly influence the catalytic activity of the palladium
complex and its capacity to facilitate the 1,2-migration.
Introduction
coupling reactions involving vinylic tosylates and phosphates.
Hartwig and co-workers reported a single case of this rear-
rangement in work on the Kumada-Corriu coupling with vinyl
tosylates. The reaction of the alkenyl tosylate 1 with p-tolyl
magnesium bromide catalyzed by an in situ generated Pd:
JosiPhos complex provided the product 2 of 1,2-migration
(Scheme 1a).4,5 Our work on the Pd-catalyzed Mizoroki-Heck
coupling of vinyl tosylates and phosphates revealed that these
migrations are more general,6-8 being promoted with electro-
philes bearing an aryl or a tertiary alkyl substituent at the C1
(Scheme 1b). Additionally, it was found that a bulky electron-
Palladium migrations are becoming increasingly common for
a variety of systems in Pd-catalyzed C-C bond forming
reactions, which include 1,3-, 1,4-, 1,5-, and 1,6-rearrangements
at sp2-carbon centers, thereby expanding the repertoire of
coupling products obtained by these reactions.1-3 Recently, the
1,2-migration has been observed for these transition-metal-based
(1) For examples on the 1,3- to the 1,6-migrations, see: Mota, A. J.; Dedieu,
A. Organometallics 2006, 25, 3130, and references cited therein.
(2) For a recent review on 1,4-migrations, see: Ma, S.; Gu, Z. Angew. Chem.,
Int. Ed. 2005, 44, 7512.
(3) For some recent examples, see: (a) Zhao, J.; Larock, R. C. J. Org. Chem.
2006, 71, 5340. (b) Zhao, J.; Yue, D.; Campo, M. A.; Larock, R. C. J. Am.
Chem. Soc. 2007, 129, 5288. (c) Campo, M. A.; Zhang, H.; Yao, T.; Ibdah, A.;
McCulla, R. D.; Huang, Q.; Zhao, J.; Jenks, W. S.; Larock, R. C. J. Am. Chem.
Soc. 2007, 129, 6298. (d) Blaszykowski, C.; Aktoudianakis, E.; Alberico, D.;
Bressy, C.; Hulcoop, D. G.; Jafarpour, F.; Joushaghani, A.; Laleu, B.; Lautens,
M. J. Org. Chem. 2008, 73, 1888. (e) Mariampillai, B.; Alliot, J.; Li, M.; Lautens,
M. J. Am. Chem. Soc. 2007, 129, 15372. (f) Bour, C.; Suffert, J. Org. Lett.
2005, 7, 653. (g) Mota, A. J.; Dedieu, A.; Bour, C.; Suffert, J. J. Am. Chem.
Soc. 2005, 127, 7171.
(4) Limmert, M. E.; Roy, A. H.; Hartwig, J. F. J. Org. Chem. 2005, 70,
9364.
(5) For some recent reviews on the Kumada-Corriu coupling, see: (a) Corbet,
J.-P.; Mignani, G. Chem. ReV. 2006, 106, 2651. (b) Frieman, B. A.; Taft, B. R.;
Lee, C.-T.; Butler, T.; Lipshutz, B. H. Synthesis 2005, 2989. (c) Schro¨ter, S.;
Stock, C.; Bach, T. Tetrahedron 2005, 61, 2245. (d) Anctil, E. J.-G.; Snieckus,
V. J. Organomet. Chem. 2002, 653, 150. (e) Martin, R.; Buchwald, S. L. J. Am.
Chem. Soc. 2007, 129, 3844. (f) Wolf, C.; Xu, H. J. Org. Chem. 2008, 73, 162.
(g) Ackermann, L.; Althammer, A. Org. Lett. 2006, 8, 3457.
10.1021/jo801824e CCC: $40.75
Published on Web 11/26/2008
2009 American Chemical Society
J. Org. Chem. 2009, 74, 135–143 135

Lindhardt et al.
SCHEME 1
rich monodentate ligand, such as P(t-Bu)3, was imperative for
rearrangement.9 Later, we reported a couple of examples of this
migration in the Negishi coupling of aryl zinc reagents to alkenyl
phosphates with another Pd:Josphos complex (Scheme 1c)10,11
employed by Hartwig and co-workers in their example in the
Kumada-Corriu coupling.4
These migrations appear to proceed by a rare pathway
involving first ꢀ-hydride elimination of the intermediate
alkenyl-Pd(II) species with an olefinic hydrogen generating an
alkynyl-coordinated palladium hydride, followed by a hydro-
palladation step to give the isomeric olefinic Pd(II) intermediate.4,7
However, the factors which control this 1,2-migration are not
fully understood. For example, in the Mizoroki-Heck couplings,
we found after ligand screening that t-Bu3P is the best ligand
for promoting this rearrangement,7,8 whereas when the same
vinylic electrophiles were used in the Negishi reaction, the
JosiPhos ligand, PPF-t-Bu, proved more effective for inducing
the migration.10 On the other hand, applying P(t-Bu)3 as the
ligand for the same Negishi reactions favored the normal
coupling product. Furthermore, Hartwig and co-workers reported
the use of another JosiPhos ligand, CyPF-t-Bu, for the
Kumada-Corriu coupling,4 which in our hands only partially
promoted the rearrangement in Negishi couplings.10
(6) Hansen, A. L.; Ebran, J.-P.; Ahlquist, M.; Norrby, P.-O.; Skrydstrup, T.
Angew. Chem., Int. Ed. 2006, 45, 3349.
(7) Ebran, J.-P.; Hansen, A. L; Gøgsig, T. M.; Skrydstrup, T. J. Am. Chem.
Soc. 2007, 129, 6931.
From these preliminary studies, it became evident that subtle
changes in either the ligand or organometallic reagent composi-
tion can greatly influence the outcome of these ꢀ-hydride
elimination processes and subsequent 1,2-migrations. Hence, a
more comprehensive study was undertaken to examine the scope
and limitations of the 1,2-migration in Negishi couplings with
vinyl phosphates and tosylates. Particular emphasis is made on
the nature of the ligand and the nucleophilicity of the organo-
metallic reagents.
(8) For some recent reviews on the Mizoroki-Heck reaction, see: (a) Heck,
R. F. Synlett 2006, 2855. (b) Phan, N. T. S.; Van Der Sluys, M.; Jones, C. W.
AdV. Synth. Catal. 2006, 348, 609. (c) Alonso, F.; Beletskaya, I. P.; Yus, M.
Tetrahedron 2005, 61, 11771. (d) Heck, R. F. Palladium Reagents in Organic
Syntheses; Academic Press: Orlando, FL, 1985. (e) Tsuji, J. Palladium Reagents
and Catalysts: InnoVation in Organic Synthesis; Wiley: Chichester, U.K., 1995.
(f) Brase, S.; de Meijere, A. In Metal-Catalyzed Cross-Coupling Reactions;
DiederichF., Stang, P. J., Eds.; Wiley-VCH: New York, 1998; Chapter 3. (g)
Link, J. T.; Overman, L. E. In Metal-Catalyzed Cross-Coupling Reactions;
DiederichF., Stang, P. J., Eds.; Wiley-VCH: New York, 1998; Chapter 6. (h)
Transition Metal Catalyzed Reactions; Davis, S. G., Murahashi, S.-I., Eds.;
Blackwell Science, Oxford, 1999. (i) Beletskaya, I.; Cheprakov, A. V. Chem.
ReV. 2000, 100, 3009. (j) Handbook of Organopalladium Chemistry for Organic
Synthesis; Negishi, E., Ed.; John Wiley & Sons: New York, 2002.
(9) For the use of this ligand in coupling reactions, see: (a) Hundertmark,
T.; Littke, A. F.; Buchwald, S. L.; Fu, G. C. Org. Lett. 2000, 2, 1729. (b) Littke,
A. F.; Dai, C.; Fu, G. C. J. Am. Chem. Soc. 2000, 122, 4020. (c) Littke, A. F.;
Fu, G. C. J. Org. Chem. 1999, 64, 10. (d) Hills, I. D.; Fu, G. C. J. Am. Chem.
Soc. 2004, 126, 13178. (e) Netherton, M. R.; Fu, G. C. Org. Lett. 2001, 3, 4295.
(f) Littke, A. F.; Schwarz, L.; Fu, G. C. J. Am. Chem. Soc. 2002, 124, 6343. (g)
Dai, C.; Fu, G. C. J. Am. Chem. Soc. 2001, 123, 2719. (h) Littke, A. F.; Fu,
G. C. J. Am. Chem. Soc. 2001, 123, 6989.
Results and Discussion
Negishi Reactions with Aryl Zinc Reagents. In order to
examine the features responsible for the 1,2-migration in the
Negishi couplings, we conducted a more comprehensive study
on the structural importance of the ligand with a series of
commercially available and structurally similar diphosphine
ligands bearing a ferrocene core. The results of this initial study
are depicted in Table 1 for the coupling of o-tolyl zinc chloride
4 with the vinyl phosphate 5.
The chosen catalytic system was extrapolated from our earlier
work, consisting of a palladium(0) source, Pd(dba)2, in com-
bination with an equimolar amount of a hindered JosiPhos
ligand12 in THF at 70 °C with a reaction time between 18 and
20 h. A characteristic trait for all the diphosphine ligands tested
is the presence of the bulky di-tert-butyl phosphine moiety,
(10) Hansen, A. L.; Ebran, J.-P.; Gøgsig, T. M.; Skrydstrup, T. J. Org. Chem.
2007, 72, 6464.
(11) For reviews and general literature on the Negishi reaction, see: (a)
Organozinc Reagents, A Practical Approach; Knochel, P., Jones, P., Eds.; Oxford:
New York, 1999. (b) Berk, S. C.; Yeh, M. C. P.; Jeong, N.; Knochel, P.
Organometallics 1990, 9, 3053. (c) Knochel, P.; Singer, R. D. Chem. ReV. 1993,
93, 2117. (d) Reiser, O. Angew. Chem., Int. Ed. 2006, 45, 2838. (e) Yin, N.;
Wang, G.; Qian, M.; Negishi, E. Angew. Chem., Int. Ed. 2006, 45, 2916. (f)
Tan, Z.; Negishi, E. Angew. Chem., Int. Ed. 2006, 45, 762. (g) Organ, M. G.;
Avola, S.; Dubovyk, I.; Hadei, N.; Kantchev, E. A. B.; O’Brian, C. J.; Valente,
C. Chem.sEur. J. 2006, 12, 4749. (h) Frisch, A. C.; Beller, M. Angew. Chem.,
Int. Ed. 2005, 44, 674. (i) Arp, F. O.; Fu, G. C. J. Am. Chem. Soc. 2005, 127,
10482. (j) Milne, J. E.; Buchwald, S. L. J. Am. Chem. Soc. 2004, 126, 13028.
136 J. Org. Chem. Vol. 74, No. 1, 2009

1,2-Migrations in Pd-Catalyzed Negishi Couplings
TABLE 1. Substitution Pattern Effects of the JosiPhos Ligands on
the 1,2-Migration
reagent, 4-methoxy phenyl zinc chloride (11), coupled success-
fully in good to excellent yields though producing varying ratios
of the normal and rearranged product depending on the starting
vinyl tosylate/phosphate (entries 1-3). For example, for the
vinyl tosylate 3 bearing a C1-t-butyl group, the product of
migration was predominant, whereas with a less bulky alkyl
C1 substituent or an o-tolyl group, migration was substantially
inhibited. Exchanging the p-methoxy group of the aryl zinc
reagent with a less electron-donating p-methyl substituent, as
shown in entries 4 and 5, led to a slight decrease in the isolated
yields, but more importantly, the selectivity for the 1,2-migration
was increased. Switching to o-methylphenyl zinc chloride 4 and
thereby increasing the steric bulk of the aryl zinc reagent further
enhanced the migration (entries 6-8). For example, with tosylate
3 as the coupling partner, only the 1,2-disusbstituted alkene from
rearrangement was obtained. Nevertheless, this sterical bulk also
led to a diminished reactivity with the cyclohexyl vinyl tosylate
9, where only a 39% conversion after 20 h was observed.
An aryl zinc reagent 13 with an electron-withdrawing cyano
group at the C4-position of the aromatic ring was examined
with the three electrophiles 3, 8, and 9 (entries 10-12).
Characteristic for these three reactions was an increased
selectivity for migration compared to the other para-substi-
tuted organozinc reagents tested. Again, the tosylate 9 displayed
modest reactivity with this zinc reagent but nonetheless favored
the 1,2-substituted olefin.
Finally, the effect of the leaving group on the selectivity of
the coupling products was also examined via the reaction of
the aryl zinc reagent 4 with a halide equivalent to the alkenyl
phosphate 8 (entry 13). In this example, the ratio of the product
of migration 26 to the normal product 27 was measured to be
5:1, which was essentially the same result observed for the
similar reaction between vinyl phosphate 8 and 4 (entry 7). In
addition, the reaction yields of the two cross-coupling were
comparable.
a Conversion and ratio determined by 1H NMR analysis of crude
reaction mixture.
which was previously observed to be crucial for generating an
active catalyst for this cross-coupling reaction.10 Whereas all
palladium catalysts composed of these ligands promoted the
reaction with varying degrees of conversion, the Pd:PPF-t-Bu
and Pd:NPF-t-Bu complexes (entries 4 and 5, respectively)
provided the products 6 and 7 with highest selectivities toward
the 1,2-migration. As earlier reported, here too the catalyst
generated from CyPF-t-Bu displayed selectivity for the migrated
product with an aryl zinc reagent less than that of PPF-t-Bu.
Both electron-donating and -withdrawing substituents on the
phenyl ring of the JosiPhos lowered the selectivity (entries 2
and 6), suggesting that subtle electronic and sterical effects on
this moiety of the ligand may be effecting the outcome of the
migration in the Negishi coupling. Finally, interchanging the
two phosphines on the ligand was detrimental for the selectivity,
essentially leading to a 1:1 mixture of the coupling products 6
and 7 (entries 7 and 8).
Work was then undertaken to investigate the electronic and
sterical influence of the organometallic reagent on the migration.
To start off with, different aryl zinc chlorides were tested in
their coupling with two alkenyl tosylates, 3 and 9, and an alkenyl
O,O-diethyl phosphate represented by 8 in the presence of a
catalyst formed with PPF-t-Bu as the diphosphine ligand. We
have earlier demonstrated that the use of a tosylate or O,O-
diethyl phosphate has essentially no consequence on either the
yield or selectivity for such couplings, whereas the correspond-
ing O,O-diphenyl phosphate led to the formation of a dimeric
biproduct.10 As illustrated in Table 2, the electron-rich aryl zinc
Increasing the sterical bulk of the JosiPhos ligand with NPF-
t-Bu had a substantial effect on the outcome of some of these
reactions, by producing either a more active or a more stabile
catalyst. In some instances, a dramatic outcome on the migration
was also observed. As seen from Table 3, entry 2, the use of
this palladium complex with this ligand increased the ratio of
the 1,2-alkene versus 1,1-alkene to 4:3 from 1:5 with PPF-t-Bu
(Table 2, entry 2). Similar enhanced migration selectivity was
seen in a few other cases (entries 1, 7, and 11). Even the phenyl
vinyl phosphate displayed an analogous reactivity with the
o-tolyl zinc chloride upon switching between catalysts generated
from these two ligands (Table 2, entry 9 vs Table 3, entry 9).
With a complex composed of PPF-t-Bu, only the normal
coupling product was isolated, where in contrast the naphthalene
analogue was capable of promoting migration to some degree.
Significant yield increases were also obtained for certain cases
upon the use of a Pd:NPF-t-Bu complex. For example, the
1-cyclohexyl vinyl tosylate 9 reacted poorly in the presence of
a catalyst bearing the PPF-t-Bu ligand (Table 1, entries 8 and
12). But with NPF-t-Bu, synthetically more useful yields were
secured.
Finally, the effect of the solvent was examined in order to
compare the results from the Negishi couplings to the
Kumada-Corriu couplings obtained by Hartwig and co-workers
(Scheme 2), which were performed in toluene.4 Whereas the
degree of migration was not significantly affected by increased
amounts of added toluene in these Negishi couplings, the yields
(12) (a) Pregosin, P. S.; Salzmann, R.; Togni, A. Organometallics 1995, 14,
842. (b) Togni, A.; Breutel, C.; Schnyder, A.; Spindler, F.; Landert, H.; Tijani,
A. J. Am. Chem. Soc. 1994, 116, 4062. (c) Abbenhuis, H. C. L.; Burckhardt, U.;
Gramlich, V.; Togni, A.; Albinati, A.; Muel, B. Organometallics 1994, 13, 4481.
J. Org. Chem. Vol. 74, No. 1, 2009 137

Lindhardt et al.
TABLE 2. Negishi Couplings with PPF-t-Bu as the Ligand
a
Conversion and ratio determined by H NMR spectrum of crude reaction mixture. b Isolated yields. c Not determined.
1
were improved for some of the substrates. For example, the
product 24 was isolated in almost quantitative yield in the
coupling of 3 with o-tolyl zinc chloride using standard PPF-t-
Bu coupling conditions but in a 7:3 THF/toluene mixture
(Scheme 2). Unfortunately, the cyclohexyl vinyl tosylate 9
showed no reactivity when toluene was used as cosolvent, and
hence further cross-coupling reactions were not carried out in
this solvent mixture.
of experiments were then performed with an alkyl zinc reagent
and two aryl Grignard reagents.
Coupling Reactions with an Alkyl Zinc Reagent and
Aryl Grignard Reagents. Table 4 illustrates the outcome of a
series of Negishi couplings performed with n-butyl zinc chloride
as a representative example of an alkyl zinc reagent. As with
the reactions employing the aryl zinc reagents, Pd(dba)2 was
used as the palladium(0) source in combination with equimolar
ratios of a diphosphine ligand with respect to Pd. All reactions
were initially performed in THF at 70 °C. The use of a palladium
catalyst formed with PPF-t-Bu in reactions carried out with
alkenyl tosylate 3 and phosphate 8 provided good coupling
yields but importantly did not promote any degree of migration
(entries 1 and 2). Exploiting instead a catalyst prepared with
NPF-t-Bu in the coupling of t-Bu-vinyl tosylate 3 resulted in a
10:9 ratio in favor of the 1,1-disubstituted alkene (entry 3).
However, again in the case of vinyl phosphate 8, no sign of the
rearranged product could be detected (entry 4).
From these initial studies, it becomes evident that aryl zinc
chlorides with increased nucleophilicity have a detrimental effect
on the ꢀ-hydride elimination step and subsequently the 1,2-
migration, whereas decreasing the nucleophilicity of the orga-
nometallic species with electron-reducing groups significantly
improved the product distribution in favor of the rearranged
disubstituted alkene. Furthermore, increased sterical bulk of the
aryl zinc reagent as noted with the o-tolyl zinc chloride again
provided a preference for the 1,2-alkene. To further examine
the role of the nucleophilicity of the coupling reagent, a series
138 J. Org. Chem. Vol. 74, No. 1, 2009

1,2-Migrations in Pd-Catalyzed Negishi Couplings
TABLE 3. Negishi Couplings with NPF-t-Bu as the Ligand
a
Conversion and ratio determined by H NMR spectrum of crude reaction mixture. b Isolated yields. c Not determined.
1
SCHEME 2
The temperature effect on the 1,2-migration was examined
in the Negishi coupling between the alkenyl tosylate 3 and
n-butyl zinc chloride applying NPF-t-Bu as the ligand, the results
of which are shown in Table 5. Quite clearly, lowering the
reaction temperature resulted in a reduction of the ratio between
the 1,2-migrated and the normal coupling products.
Finally, several couplings were performed with two aryl
Grignard bromides to examine the effect of increasing the
nucleophilicity of the aryl metal reagent (Table 6). These
reactions performed best in toluene as earlier reported,4 where
in THF poorer coupling yields were observed, although with
no change in the distribution of the two coupling products. Here
too, the normal coupling products were favored in the four cases,
which is the opposite preference for the coupling reactions with
the corresponding aryl zinc reagents. Even in the same reaction
earlier studied by Hartwig,4 though with PPF-t-Bu as the ligand,
migration was not the favored pathway (entry 4).
Mechanistic Considerations. In this study on the 1,2-
migration in the Negishi couplings, several main conclusions
can be made on the factors controlling this unsual rearrangment.
J. Org. Chem. Vol. 74, No. 1, 2009 139

Lindhardt et al.
TABLE 4. Negishi Couplings with Alkyl Zinc Reagents
Entry
Electrophile
Ligand
Conversion (%)a
Yield (%)b
Ratio A:Ba
Cmpd No.
1
2
3
4
3 R ) t-Bu
8 R ) o-tolyl
3 R ) t-Bu
8 R ) o-tolyl
PPF-t-Bu
PPF-t-Bu
NPF-t-Bu
NPF-t-Bu
100
100
100
100
85
96
58
86
0:1
0:1
9:10
0:1
36
37
38/36
37
a
Conversion and ratio determined by H NMR spectrum of crude product. b Isolated yields.
1
TABLE 5. Temperature Effect on the 1,2-Migration
Entry
Temperature (°C)
Ratio 38:36a
Yield (%)b
1
2
3
40
55
70
23:77
36:64
47:53
62
29
58
a
Ratio determined by H NMR spectrum of crude product. b Isolated yield.
1
TABLE 6. Kumada-Corriu Couplings with 1,2-Migration
b
a Isolated yield. All reactions proceeded with full conversion. Ratio determined by H NMR spectrum of crude product.
1
First, the nature of the organometallic species has a substantial
influence on the efficacy of the 1,2-migration for this cross-
coupling reaction catalyzed by a palladium-JosiPhos complex.
Essentially, the less nucleophilic the coupling reagent is, the
greater its tendency for promoting the 1,2-migration.13
Second, the 1,2-migration is profoundly dependent on the
structure of the starting alkenyl tosylate/phosphate. As observed
in our studies on the Mizoroki-Heck reaction, vinyl tosylates
and phosphates bearing a quaternary alkyl C1 substituent led
to products of rearrangement.6,7 Whereas increased sterical bulk
at this position seems to favor the ꢀ-hydride elimination step,
there are some observations that need further exploration. For
example, the cyclohexyl vinyl tosylate 9, which incorporates a
secondary alkyl substituent at C1 of the electrophile, reacts
cleanly with the aryl zinc chlorides promoted by a catalyst with
either PPF-t-Bu or its naphthalene equivalent. Yet, no coupling
could be encouraged with 1-isopropyl vinyl tosylate or the
structurally similar alkenyl tosylate 1 bearing a methyl group
at C2. In both cases, the starting tosylates were fully recovered.
The absence of reactivity for these cases is difficult to understand
considering that both vinyl tosylates 1 and 9 underwent
Kumada-Corriu couplings with an aryl Grignard reagent using
the JosiPhos ligand, PPF-t-Bu (Table 6).4
(13) Efforts were made to couple either p-tolyl or p-methoxyphenyl trim-
ethylstannane with 1-cyclohexyl vinyl tosylate, with the presence of lithium
chloride, with the expectation that cross-coupling with less reactive organotin
compounds would lead to higher preferences for the 1,2-disubstituted alkene
compared to that observed for the corresponding zinc reagents. Nevertheless,
the reaction conditions employed for the described Negishi couplings were not
effective for promoting cross-couplings with the aryltin compounds. Further work
is ongoing to identify suitable conditions which may promote migrations in the
Kosugi-Migita-Stille couplings.
The 1,2-migrated product is not the favored product in the
coupling of phenyl vinyl phosphate 10 with an aryl zinc reagent
using either PPF-t-Bu or NPF-t-Bu as the ligand. The stability
of a η3-benzyl-metal complex, which is formed after the
140 J. Org. Chem. Vol. 74, No. 1, 2009

1,2-Migrations in Pd-Catalyzed Negishi Couplings
oxidative addition step, may retard the ꢀ-hydride elimination
step.14 However, increasing the steric bulk of the C1-aryl group
with an ortho-substituent as with the o-tolyl vinyl phosphate 8
disfavors formation of the η3-benzyl-metal intermediate and
thereby promotes the elimination step. This trend was also
observedinourstudiesonthe1,2-migrationintheMizoroki-Heck
coupling.7
Third, a large ligand effect was noted for these Pd-catalyzed
cross-couplings with the alkenyl tosylates and phosphates tested.
The JosiPhos ligands, generally considered to display bidentate
metal binding properties,15 can promote these 1,2-migrations
and with high efficiency depending on the starting electrophile.
In our work on the same rearrangements for the Mizoroki-Heck
reactions, the most effective catalyst systems were formed with
a sterically encumbered and electron-rich monodentate phos-
phine, such as tri-tert-butylphosphine. In our explanation,6 we
invoked the intermediacy of a tricoordinated Pd(II) species after
the oxidative addition step due to the sterically demanding
trialkylphosphine ligand.9b,16 Hence, an empty coordination site
on the palladium nucleus is secured, which is necessary for the
concominant ꢀ-hydride elimination step.
On the other hand, the bidentate nature of the JosiPhos ligands
should inhibit the hydride elimination step by the opposite
principle in that no vacant site on the metal center will be
available due to coordination of both phosphino groups of the
ligand to palladium center. This is observed when DPPF is
exploited as a ligand for the Negishi couplings where no
migration is observed.10,17 Furthermore, a recent and extensive
study performed by Hartwig and co-workers on the amination
of aryl and heteroaryl halides revealed that palladium complexes
generated from JosiPhos ligands and in particular with CyPF-
t-Bu are highly effective catalysts for aminations of heteroaryl
and aryl halides.15a The efficiency of these catalysts for these
C-N bond forming reactions was in part attributed to the rigid
backbone of the ligand reinforcing a tight binding to the metal
center. Yet, the observations of 1,2-migration for the Negishi
couplings in this study and that of Hartwig and co-workers in
the Kumada-Corriu reaction suggest that the JosiPhos ligands
may also exhibit monodentate character if the ꢀ-hydride
elimination mechanism is operating for these migrations.
There is an alternative mechanism for the ꢀ-hydride elimina-
tion with a chelating diphosphine ligand, involving first dis-
sociation of the tosylate from the metal center to generate a
cationic vinyl complex as that observed for the Mizoroki-Heck
reaction with electron-rich olefins.18 However, under the reaction
conditions used in these Negishi reactions, up to 30 equiv of
LiCl are generated in solution relative to the Pd-catalyst which
would trap any cationic intermediate. Support for this assump-
tion was provided by comparing the outcome of coupling
experiments between the aryl zinc reagent 4 with the alkenyl
phosphate 8 and the corresponding vinyl chloride (Table 2).
Both the yields and the ratios of 1,2-disubstituted olefin versus
its 1,1-disubstituted isomer were similar, thereby suggesting that
a cationic mechanism is not operating for this migration.19
For the Negishi couplings with aryl zinc reagents, catalysts
generated from either PPF-t-Bu or NPF-t-Bu were more efficient
for promoting the 1,2-migration than CyPF-t-Bu. We explain
this observation by the difference in binding capacity between
the two phosphino groups on PPF-t-Bu and NPF-t-Bu than
for the latter ligand. It is highly plausible that the bulky alkyl-
di-tert-butylphosphino unit of these three ligands ensures the
formation of a reactive catalyst for the first step of the catalytic
cycle involving the demanding oxidative addition of the C-O
bond in the vinyl tosylate and phosphate. We have earlier seen
that replacement of the di-tert-butylphosphino moiety with a
dicyclohexylphosphino unit leads to a catalyst which cannot
promote these Negishi couplings.10 After this step, the weaker
donating diphenyl- and dinaphthylphosphino groups could
temporarily dissociate, providing a trigonally coordinated pal-
ladium complex with a vacant site, which represents a situation
identical to that proposed for the Mizoroki-Heck coupling with
a catalyst generated from tri-tert-butylphosphine (Scheme
3).6,7,20 Subsequent ꢀ-hydride elimination will then be possible
by the presence of the vacant site on the metal center. Therefore,
the rate of the transmetalation will have a marked influence on
the degree of ꢀ-hydride elimination and subsequent hydropal-
ladation. This model then explains the higher tendency for 1,2-
migration with organometallic reagents of low nucleophilicity/
reactivity because of a slow transmetalation step.
In general, better catalytic activity and either higher or similar
ratios of 1,2- to 1,1-disubstituted olefins were noted employing
a palladium complex formed from NPF-t-Bu than with PPF-t-
Bu. Here, the greater sterical bulk of the dinaphthylphosphino
group compared to that of the diphenylphosphino moiety allows
NPF-t-Bu to operate as a more sterically hindered monodentate
ligand. Electronic effects do not appear to be important, as the
introduction of electron-withdrawing groups on the phenyl ring
of PPF-t-Bu should lead to increased isomerization compared
to PPF-t-Bu. Yet, the opposite was observed with the PPF-t-
Bu analogue possessing a p-substituted trifluoromethyl group
(Table 1, entry 2).
(14) (a) Roberts, J. S.; Klabunde, K. J. J. Am. Chem. Soc. 1977, 99, 2509.
(b) Gatti, G.; Lopez, J. A.; Mealli, C.; Musco, A. J. Organomet. Chem. 1994,
483, 77. (c) Hayashi, T.; Matsumoto, Y.; Ito, Y. Tetrahedron: Asymmetry 1991,
2, 601. (d) Rix, F. C.; Brookhart, M.; White, P. S. J. Am. Chem. Soc. 1996, 118,
2436. (e) LaPointe, A. M.; Rix, F. C.; Brookhart, M. J. Am. Chem. Soc. 1997,
119, 906. (f) Nozaki, K.; Komaki, H.; Kawashima, Y.; Hiyama, T.; Matsubara,
T. J. Am. Chem. Soc. 2001, 123, 534. (g) Nettekoven, U.; Hartwig, J. F. J. Am.
Chem. Soc. 2002, 124, 1166. (h) Kuwano, R.; Kondo, Y.; Matsuyama, Y. J. Am.
Chem. Soc. 2003, 125, 12104.
(15) For some applications of JosiPhos ligands in Pd-catalyzed C-heteroatom
bond forming reactions, see: (a) Shen, Q.; Ogata, T.; Hartwig, J. F. J. Am. Chem.
Soc. 2008, 130, 6586. (b) Fernandez-Rodriguez, M. A.; Shen, Q.; Hartwig, J. F.
J. Am. Chem. Soc. 2006, 128, 2180. (c) Roy, A. H.; Hartwig, J. F. Organome-
tallics 2004, 23, 194. (d) Roy, A. H.; Hartwig, J. F. J. Am. Chem. Soc. 2003,
125, 8704.
The higher degree of migration observed at increased reaction
temperatures (Table 5) also provides support for this mechanism.
(18) For a few papers on the topic of the Mizoroki-Heck reaction with
electron-rich alkenes, see: (a) Ozawa, F.; Kubo, A.; Hayashi, T. J. Am. Chem.
Soc. 1991, 113, 1417. (b) Cabri, W.; Candiani, I. Acc. Chem. Res. 1995, 28, 2.
(c) Ludwig, M.; Stromberg, S.; Svensson, M.; Akermark, B. Organometallics
1999, 18, 970. (d) Larhed, M.; Hallberg, A. In Handbook of Organopalladium
Chemistry for Organic Synthesis; Negishi, E., Ed.; John Wiley & Sons: New
York, 2002. (e) Mo, J.; Xu, L.; Xiao, J. J. Am. Chem. Soc. 2005, 127, 751. (f)
Cabri, W.; Candiani, I.; Bedeschi, A.; Santi, R. J. Org. Chem. 1992, 57, 3558.
(g) Amatore, C.; Godin, B.; Jutand, A.; Lemaˆıtre, F. Organometallics 2007, 26,
1757.
(19) There is additionally another possible mechanism involving base-
promoted elimination of the alkenyl tosylate or phosphate that could also be
invoked followed by addition of the organometallic reagent. However, in this
scenario, it would be expected that greater levels of the 1,2-disubstituted alkene
would be formed employing the more nucleophilic and most likely also more
basic organometallic reagents. Yet, the opposite trend is observed in many of
these coupling reactions.
(16) Stambuli, J. P.; Incarvito, C. D.; Bu¨hl, M.; Hartwig, J. F. J. Am. Chem.
Soc. 2004, 126, 1184.
(17) It should be noted that DPPF has also been reported to be a ligand with
a capacity to bind Pd(II) either with one phosphorus atom or with two phosphorus
atoms as demonstrated for example by Cabri et al. Cabri, W.; Candiani, I.;
DeBernardinis, S.; Francalanci, F.; Penco, S.; Santo, R. J. Org. Chem. 1991,
56, 5796. Nevertheless, DPPF does not promote the 1,2-migration in these
Negishi reactions, suggesting that certain JosiPhos ligands may possess a greater
tendency for monodentate character than DPPF.
(20) Lindhardt, A. T.; Skrydstrup, T. Chem. Eur. J. 2008, 14, 8756.
J. Org. Chem. Vol. 74, No. 1, 2009 141

Lindhardt et al.
SCHEME 3
In general, it would be expected that the rearrangement,
representing a unimolecular reaction, is favored at lower
temperatures, whereas as the bimolecular transmetalation is
favored at higher reaction temperatures. If this were the case,
then the unrearranged process would be favored at higher
reaction temperatures. Yet the opposite is observed, thereby
indicating that phosphine dissociation may be important for the
migrations observed.
Nevertheless, whereas these observations lend support to this
mechanistic proposal, rigorous experimental support is still
required to rule out alternative pathways involving ꢀ-hydride
elimination. For example, if transmetalation is fast, providing
a vinyl alkyl (or aryl) Pd(II) species, perhaps ꢀ-hydride
elimination competes with the reductive elmination step rather
than taking place before the transmetalation step. Alternatively,
ꢀ-hydride elimination occurs via a penta-coordinated Pd(II)
intermediate promoted with JosiPhos as the ligand but not with
DPPF. Work is now underway to distinguish between these
different pathways.21
Experimental Section
(E)-1-(3,3-Dimethylbut-1-enyl)-2-methylbenzene and 1-(3,-
3-Dimethylbut-1-en-2-yl)-2-methylbenzene (2410 and 25;10
Table 2, entry 6). General Procedure for the Negishi Cou-
pling Using PPF-t-Bu as Ligand. 3,3-Dimethyl but-1-en-2-yl
tosylate (127.2 mg, 0.50 mmol), PPF-t-Bu (6.8 mg, 0.013 mmol),
and Pd(dba)2 (7.2 mg, 0.013 mmol) were dissolved in 1.0 mL of
THF. A 0.5 M solution of 2-methylbenzene zinc chloride in THF
(1.5 mL, 0.75 mmol) was added, and the mixture was reacted for
20 h at 70 °C. NMR analysis of the crude reaction mixture provided
a >19:1 ratio of migrated/nonmigrated products. The crude product
was purified by flash chromatography on silica gel using pentane
as eluent. This afforded 61.0 mg of the major isomer (70% yield)
as a colorless oil.
(E)-3,3-Dimethyl-1-(2-methylphenyl)butene: 1H NMR (400 MHz,
CDCl3) δ (ppm) 7.45 (d, 1H, J ) 6.8 Hz), 7.21-7.15 (m, 3H),
6.54 (d, 1H, J ) 16.0 Hz), 6.15 (d, 1H, J ) 16.0 Hz), 2.38 (s, 3H),
1.18 (s, 9H); 13C NMR (100 MHz, CDCl3) δ (ppm) 143.5, 137.4,
135.2, 130.2, 126.8, 126.1, 125.6, 122.5, 33.7, 29.8, 19.9; GC-MS
C13H18 [M] calcd 174, found 174.
1
3,3-Dimethyl-2-(2-methylphenyl)butene: H NMR (400 MHz,
CDCl3) δ (ppm) 7.21-7.14 (m, 2H), 7.11 (td, 1H, J ) 7.6, 2.0
Hz), 7.05 (dd, 1H, J ) 7.2, 1.2 Hz), 5.31 (d, 1H, J ) 2.0 Hz), 4.77
(d, 1H, J ) 2.0 Hz), 2.26 (s, 3H), 1.13 (s, 9H); 13C NMR (100
MHz, CDCl3) δ (ppm) 158.0, 142.7, 136.0, 130.1, 129.6, 126.5,
124.5, 112.4, 37.1, 30.0, 20.7; GC-MS C13H18 [M] calcd 174, found
174.
Conclusion
In this work, we have reported an initial investigation with
the goal of determining the factors that control this unusual
rearrangement in the Negishi cross-coupling reaction catalyzed
by palladium complexes generated from JosiPhos ligands.
Several of these factors, which are believed to be important for
the 1,2-migration, include (1) the nucleophilicity of the orga-
nometallic reagent, which influences the transmetalation step
in direct competition with the intermediate ꢀ-hydride elimination
of the alkenyl Pd(II) species; (2) the structural features of the
vinyl tosylates and phosphates, in which substrates possessing
a bulky C1 substituent displayed highest propensity for under-
going the 1,2-migration under the coupling reaction conditions
examined; and (3) the structure of the JosiPhos ligand, where
both the sterical bulk and choice of substitutents on the
ferrocenyl phosphino group greatly influences the catalytic
activity of the palladium complex and its capacity to facilitate
the 1,2-migration.
(E)-4-(2-Methylstyryl)benzonitrile and 4-(1-o-Tolylvinyl)-
benzonitrile (31 and 32;10 Table 3, entry 11). General
Procedure for the Negishi Coupling Using NPF-t-Bu as
Ligand. o,o-Diethyl 1-o-tolylvinyl phosphate (135.1 mg, 0.50
mmol), NPF-t-Bu (8.0 mg, 0.013 mmol), and Pd(dba)2 (7.2 mg,
0.013 mmol) were dissolved in 1.0 mL of THF. A 0.5 M solution
of 4-cyanophenyl zinc chloride in THF (1.5 mL, 0.75 mmol) was
added, and the mixture was reacted for 18 h at 70 °C. NMR analysis
of the crude reaction mixture provided a >19:1 ratio of migrated/
nonmigrated products. The crude product was purified by flash
chromatography on silica gel using CH2Cl2/pentane (2:3) as eluent.
This afforded 74.5 mg of the major isomer (68% yield) as a
colorless oil.
(E)-4-(2-Methylstyryl)benzonitrile: 1H NMR (400 MHz, CDCl3)
δ (ppm) 7.65-7.56 (m, 5H), 7.46 (d, 1H, J ) 16.0 Hz), 7.26-7.22
(m, 3H), 6.99 (d, 1H, J ) 16.0 Hz), 2.45 (s, 3H); 13C NMR (100
MHz, CDCl3) δ (ppm) 142.3, 136.4, 135.5, 132.6, 130.8, 130.3,
128.6, 128.2, 127.1, 126.5, 125.7, 119.2, 110.7, 20.0; GC-MS
C16H13N [M] calcd 219, found 219.
Further work is now in progress to provide more details about
the factors for this remarkable migration and to investigate
reaction conditions which will promote this rearrangement in
other transition-metal-catalyzed reactions.
1
4-(1-o-Tolylvinyl)benzonitrile: H NMR (400 MHz, CDCl3) δ
(ppm) 7.58 (d, 2H, J ) 8.8 Hz), 7.37 (d, 2H, J ) 8.8 Hz), 7.31-7.19
(m, 4H), 5.88 (d, 1H, J ) 0.8 Hz), 5.38 (d, 1H, J ) 0.8 Hz), 2.03
(21) We thank the reviewers for helpful comments and suggestions to this
work.
142 J. Org. Chem. Vol. 74, No. 1, 2009

1,2-Migrations in Pd-Catalyzed Negishi Couplings
(s, 3H); 13C NMR (100 MHz, CDCl3) δ (ppm) 148.4, 145.3, 140.4,
136.1, 132.5, 130.6, 130.2, 128.4, 127.3, 126.3, 119.1, 118.2, 111.3,
20.3; GC-MS C16H13N [M] calcd 219, found 219.
graduate schools, and the University of Aarhus. Solvias is
thanked for generous gifts of the JosiPhos ligands.
Supporting Information Available: Experimental details for
1
all reactions. Copies of H NMR and 13C NMR spectra for all
Acknowledgment. We are deeply appreciative of generous
financial support from the Danish National Research Foundation,
the Danish Natural Science Research Council, the Carlsberg
Foundation, the Lundbeck Foundation, the OChem and iNANO
new coupling products. This material is available free of charge
via the Internet at http://pubs.acs.org.
JO801824E
J. Org. Chem. Vol. 74, No. 1, 2009 143