Contributions to the enantioselective Heck reaction using MeO-Biphep ligands. The case against dibenzylidene acetone

Article abstract of DOI:10.1021/om980783l

It is shown that the Pd-catalyzed enantioselective Heck reaction of p-XC₆H₄OTf, X = OMe, H, CO₂Me, with dihydrofuran gives higher enantioselectivities when the chelating diphosphine MeO-Biphep, 1a, is replaced with its disubstituted analogue 3,5-di-tert-butyl MeO-Biphep, 1b. The phenylation of 5-methyl-2,3-dihydrofuran produces a new dihydrofuran containing a quaternary stereogenic center (ee, >98% with 1b, ca. 20% with 1a). Catalytic results for the reaction of phenyl triflate with dhf, together with stoichiometric oxidative addition reactions of aryl halides on Pd complexes of 1, show that the use of Pd(dba)(1), dba = dibenzylidene acetone, slows the oxidative addition relative to the reaction in which the Pd(0) precursor is generated from PdCl₂(1) + NaBH₄. The solid-state structures for two PdI-(aryl)(1a), 3, derivatives, aryl = p-MeOOC-C₆H₄ (3a) and C₆F₅ (3b) are reported.

Full text of DOI:10.1021/om980783l

670
Organometallics 1999, 18, 670-678
Con tr ibu tion s to th e En a n tioselective Heck Rea ction
Usin g MeO-Bip h ep Liga n d s. Th e Ca se Aga in st
Diben zylid en e Aceton e
Matthias Tschoerner and Paul S. Pregosin*
Laboratory of Inorganic Chemistry, ETH Zentrum, 8092 Zu¨rich, Switzerland
Alberto Albinati
Chemical Pharmacy, University of Milan, I-20131 Milan, Italy
Received September 18, 1998
It is shown that the Pd-catalyzed enantioselective Heck reaction of p-XC6H4OTf, X ) OMe,
H, CO2Me, with dihydrofuran gives higher enantioselectivities when the chelating diphos-
phine MeO-Biphep, 1a , is replaced with its disubstituted analogue 3,5-di-tert-butyl MeO-
Biphep, 1b. The phenylation of 5-methyl-2,3-dihydrofuran produces a new dihydrofuran
containing a quaternary stereogenic center (ee, >98% with 1b, ca. 20% with 1a ). Catalytic
results for the reaction of phenyl triflate with dhf, together with stoichiometric oxidative
addition reactions of aryl halides on Pd complexes of 1, show that the use of Pd(dba)(1), dba
) dibenzylidene acetone, slows the oxidative addition relative to the reaction in which the
Pd(0) precursor is generated from PdCl2(1) + NaBH4. The solid-state structures for two PdI-
(aryl)(1a ), 3, derivatives, aryl ) p-MeOOC-C6H4 (3a ) and C6F5 (3b) are reported.
In tr od u ction
The mechanistic aspects of the Heck reaction are
fairly well-known,1-3 and although the groups of
Amatore16-18 and Brown19 have made recent major
contributions, new suggestions continue to appear.20-22
Routinely (see Scheme 1), an aryl halogen (or triflate)
starting material is oxidatively added to a Pd(0) complex
(step a). This latter can be either added as a preformed
complex, e.g., “Pd(L-L)(dba)”, dba ) dibenzylidene
acetone, or generated in situ. The Pd(II) aryl complex
that arises is frequently isolable.1c Olefin complexation
at palladium (step b) is followed by insertion of the
alkene into the Pd-C bond (step c). â-H elimination
The palladium-catalyzed Heck reaction,1 which often
consists of the introduction of an aryl substituent into
a suitable olefin substrate, remains a subject of active
interest.2-5 There are now various protocols,3b,6-8 most
of which involve catalytic amounts of phosphine com-
plexes. The reaction can be relatively slow and/or
require elevated temperatures.9-11 Increasingly, atten-
tion has turned to the enantioselective variant,12 with
Shibasaki,12 Overman,13 Hayashi,14 and Pfaltz,15 among
others, reporting impressive results.
(1) (a) Heck, R. F. Acc. Chem. Res. 1979, 12, 146. (b) Heck, R. F.
Comprehensive Organic Synthesis; Pergamon: Oxford, 1991; Vol. 4.
(c) Garrou, P.; Heck, R. F. J . Am. Chem. Soc. 1976, 98, 4115.
(2) de Meijere, A.; Meyer, F. E. Angew. Chem. 1994, 106, 2473-
2506.
(14) (a) Ozawa, F.; Kubo. A.; Matsumoto, Y.; Hayashi, T.; Nishioka,
I.; Yanagi, K.; Moriguchi, K. Organometallics 1993, 12, 4188. (b)
Ozawa, F.; Kubo, A.; Hayashi, T. J . Am. Chem. Soc. 1991, 113, 1417-
1419.
(15) Pfaltz, A. Acta Chim. Scand. 1996, 50, 189-194. Loiseleur, O.;
Meier, P.; Pfaltz, A. Angew. Chem. 1996, 108, 218.
(16) Amatore, C.; Azzabi, M.; J utand, A. J . Am. Chem. Soc. 1991,
113, 8375-8384.
(17) (a)Amatore, C.; J utand, A.; Khalil, F.; M’Barki, M. A.; Mottier,
L. Orgamometallics 1993, 12, 3168-3178. (b) Amatore, C.; J utand, A.;
Suarez, A. J . Am. Chem. Soc. 1993, 115, 9531-9541.
(18) Amatore, C.; Broeker, G.; J utand, A.; Khalil, F. J . Am. Chem.
Soc. 1997, 119, 5176-5185.
(19) Hii, K. K.; Claridge, T. D. W.; Brown, J . M. Angew. Chem. 1997,
109, 1033-1036. Brown, J . M.; Hii, K. K. Angew. Chem. 1996, 108,
679. Brown, J . M.; Perez-Torrente, J .; Alcock, N.; Clase, H. J .
Organometallics 1995, 14, 207. Deeth, R. J .; Smith, A.; Hii, K. K.;
Brown, J . M. Tetrahedron Lett. 1998, 39, 3229-3232.
(20) Shaw, B. L. New J . Chem. 1998, 77.
(21) Shaw, B. L.; Perera, S. D.; Staley, E. A. J . Chem. Soc., Chem.
Commun. 1998, 1361.
(22) Beller, M.; Fischer, H.; Ku¨hlein, K.; Reisinger, C. P.; Herrmann,
W. J . Organomet. Chem. 1996, 520, 257-259.
(23) (a) Trabesinger, G.; Albinati, A.; Feiken, N.; Kunz, R. W.;
Pregosin, P. S.; Tschoerner, M. J . Am. Chem. Soc. 1997, 119, 6315. (b)
Hayashi, Y.; Rohde, J . J .; Corey, E. J . J . Am. Chem. Soc. 1996, 118,
5502-5503. (c) Rajanbabu, T. V.; Ayers, T. A.; Caasalnuovo, A. L. J .
Am. Chem. Soc. 1994, 116, 4101. (d) Maruoka, K.; Itoh, T.; Shirasaka,
T.; Yamamoto, H. J . Am. Chem. Soc. 1988, 110, 310-312. (e) Trost, B.
M.; Murphy, D. J . Organometallics 1985, 4, 1143.
(3) (a) Cabri, W.; Candiani, I. Acc. Chem. Res. 1995, 28, 2-7. (b)
Cabri, W.; Candiani, I.; De Bernardinis, S.; Francalanci, F.; Penco, S.
J . Org. Chem. 1991, 56, 5796-5800.
(4) Lemaire-Audoire, S.; Savignac, M.; Dupuis, C.; Genet, J . P.
Tetrahedron Lett. 1996, 37, 2003-2006.
(5) Gibson, S. E.; Middleton, R. J . J . Chem. Soc., Chem. Commun.
1995, 1743.
(6) J effery, T.; Galland, J . C. Tetrahedron Lett. 1994, 35, 4103-4106.
(7) Hillers, S.; Sartoti, S.; Reiser, O. J . Am. Chem. Soc. 1996, 118,
2087-2088.
(8) Rigby, J . H.; Huighes, R. C.; Heeg, M. J . J . Am. Chem. Soc. 1995,
117, 7834-7835.
(9) Ohff, M.; Ohff, A.; van der Boom, M. E.; Milstein, D. J . Am.
Chem. Soc. 1997, 119, 11687-11688.
(10) Reetz, M. T.; Lohmer, G.; Schwickardi, R. Angew. Chem. 1998,
110, 492.
(11) Herrmann, W.; Brossmer, C.; Oefele, K.; Reisinger, C.; Prier-
meier, T.; Beller, M.; Fischer, H. Angew. Chem. 1995, 107, 1989-1992.
(12) Sato, Y.; Sodeoka, M.; Shibasaki, M. J . Org. Chem. 1989, 54,
4738. Ohrai, K.; Kondo, K.; Sodeoka, M.; Shibasaki, M. J . Am. Chem.
Soc. 1994, 116, 11737-11748. Oshima, T.; Kagechika, K.; Adachi, M.;
Sodeoka, M.; Shibasaki, M. J . Am. Chem. Soc. 1996, 118, 7108-7116.
(13) Abelman, M. M.; Kado, N.; Overman, L. E.; Sarkar, A. K.
SYNLETT 1997, 1469. Overman, L. E.; Link, J . T. In Metal Catalyzed
Cross-Coupling Reactions; Wiley-VCH: Weinheim, 1998; p 239, and
references therein.
10.1021/om980783l CCC: $18.00 © 1999 American Chemical Society
Publication on Web 01/27/1999

Enantioselective Heck Reaction
Sch em e 1
Organometallics, Vol. 18, No. 4, 1999 671
Ta ble 1. En a n tiom er ic Excesses in th e Heck
Ch em istr ya
ligand
X
yield
ee
1a
H
52%
61%
42%
70%
76%
57%
87
87
71
99
99
91
COOMe
OMe
H
COOMe
OMe
1b
(step d) and olefin decomplexation (step e) afford the
organic product. The palladium hydride generated
reacts with added base (NEt3 or a Hu¨nig base are
commonly used) to regenerate the Pd(0) complex.
We have recently reported23a results for the enantio-
selective Heck addition of phenyl triflate to dihydrofu-
ran (dhf) using the axially chiral MeO-Biphep ligand,
1b, eq 1.23 The relatively small olefin dhf has become
a
Pd(OAc)2, 1, C6H6, dhf, PhOTf, and NEt(i-Pr)2, 40 °C.
Pd(PPh3)4 to be faster when the aryl triflate has
electron-withdrawing groups.
We have also studied the reaction of phenyl triflate
with the trisubstituted olefin, 5-methyl 2,3-dihydrofu-
ran, eq 2:
The reaction is very slow (even at 70 °C) and the
isolated yield only 38%; however, the ee with 1b is
excellent. This 5-methyl 2,3-dihydrofuran reaction is the
most marked example of the meta dialkyl effect yet
observed and represents a relatively rare example of an
enantioselective intermolecular Heck reaction which
generates a new fully substituted stereogenic center.
These few additional catalytic experiments are further
support for the existence of the m-di-tert-butyl effect.
Is d ba Ad va n ta geou s in En a n tioselective Heck
Ch em istr y? During our catalytic studies comparing 1a
with 1b we tested several palladium precursors and
noticed that the reactions starting from Pd(0)-dba
complexes seemed sluggish. To confirm this, we com-
pared the rates of product formation in the catalytic
runs starting from Pd(dba)(1a ,b),23 2, with those from
Pd(0) complexes generated by reducing chloro-deriva-
tives in situ with NaBH4.
something of a test material for this reaction.14,15 On
the basis of the observed enantiomeric excesses, ligand
1b is better than 1a . In view of the continuing develop-
ments in Heck chemistry10,11,20-22 we have expanded our
studies to include new structural and catalytic aspects
involving 1 and report here on (i) an extension of the
m-di-tert-butyl effect on enantioselectivity, (ii) the rate
reduction due to the use of dba, and (iii) the unexpected
instability of several aryl intermediates.
Pd(dba)(1a ) + PhOTf + dhf acetone, 40 °C8
no reaction after 72 h (3)
Resu lts a n d Discu ssion
NaBH4
Th e m -Di-ter t-bu tyl Effect. Our new results for the
palladium-catalyzed Heck reaction of p-XC6H4OTf, X )
OMe, H, CO2Me, with dhf, using 1 as auxiliary, are
summarized in Table 1. The catalyst, Pd(OAc)2 /NEt(i-
Pr)2 /1, in benzene at 40 °C, requires several days, with
yields between 42% and 76%. The ee’s using 1b (91-
99%) are always better than those found for 1a (71-
87%). This is consistent with the m-di-t-Bu effect on
enantioselectivity described earlier23a and with other
reports for B,23b Rh,23c Al,23d and Pd23e based catalysts
containing 3,5-substituted aryl groups. Interestingly, for
both 1a and 1b the p-OMe triflates give much lower ee’s.
Moreover, for 1b, the reaction of the p-OMe derivative
is ca. 5 times faster than for the CO2Me analogue.
J utand and Mosleh24 find the oxidative addition with
PdCl2(1a ) + PhOTf + dhf acetone, 40 °C8
70% conversion after 72 h (4)
Pd(dba)(1b) + PhOTf + dhf acetone, 40 °C8
no reaction after 72 h (5)
NaBH4
PdCl2(1b) + PhOTf + dhf acetone, 40 °C8
44% conversion after 67 h (6)
Equations 3-6 show this chemistry, and Figure 1
represents the development of organic product as a
function of time using several catalyst precursors. The
catalytic reactions were carried out at 313 K, but the
experiments for the relative kinetics, at room temper-
ature, to facilitate monitoring by NMR methods. From
(24) J utand, A.; Mosleh, A. Organometallics 1995, 14, 1810-1817.

672 Organometallics, Vol. 18, No. 4, 1999
Tschoerner et al.
again conclude that for 1a ,b there is no advantage in
using dba to stabilize the zero oxidation state.
Apart from the reduced reactivity of the dba complex,
the relatively rapid oxidative addition of PhI using the
-
BH4 route suggests that this step would most likely
not be rate determining if one were to use the iodide,
and not the triflate, as substrate.
Detailed kinetic studies by Amatore et al. show that,
in the oxidative addition reaction using PhI, “Pd(Binap)”
is more reactive than Pd(dba)(Binap).18 Further, these
authors note that for a variety of chelating diphos-
phines, Pd(dba)(L-L), the dba complex is “involved in
an endergonic equilibrium with the less ligated complex
Pd(L-L) and dba”. When NaBH4 is used to generate
“Pd(1)”, a more active species26 is directly produced, and
consequently one finds a faster reaction.
We have also checked the rates of the stoichiometric
oxidative addition reaction of PhI using Pd(dba)(PPh3)2
and Pd(PPh3)4 and, in agreement with Amatore,17a find
that both of these react relatively quickly (<3 min27) to
form the trans-PdI(Ph)(PPh3)2 product.
F igu r e 1. Development of the 2-phenyldihydrofuran
organic product as a function of time, for the Pd-catalyzed
Heck reaction of dhf with PhOTf. The upper curves arise
from the reactions of “Pd(1a )” (fastest) and “Pd(1b)”,
whereas the bottom curve shows that little product is
formed from Pd(dba)(1a ) after ca. 100 h.
Since the borohydride reduction affords Cl- ions, it
is possible that our observed rate acceleration arises via
an electron-rich Pd(0) anion, e.g., PdClx(1)x-, x ) 1, 2,
as suggested previously for PPh3 derivatives.17b In any
case, the slower reactions of the dba complexes are
probably due to the reduced nucleophilicity of the Pd-
(0), since dba, an R,â-unsaturated ketone, can withdraw
electron density via π-back-bonding.
The above results show that dba is likely to be most
detrimental in enantioselective Heck chemistry, where
bidentate ligands are routinely employed. We do not
suggest that, generally, the PdCl2(L-L)/NaBH4 route
must lead to faster catalysis in all cases, since the
structure of the substrate olefin cannot be neglected.
However, for reactions in which the olefin substrate
employed is a poorer ligand than dba, its presence could
well lead to rate reductions. Our proposed borohydride
route represents a simple alternative.
Str u ctu r a l Stu d ies. As suggested by Scheme 1, the
oxidative addition of PhI is expected to produce PdI-
(aryl)(1a ) as an intermediate. Since this oxidative
addition reaction may not be rate determining (vida
supra), we tried to prepare and isolate both PdI(C6H5)-
(1a ) and PdI(p-OMe-C6H4)(1a ) using the NaBH4 reduc-
tion route, followed by addition of, for example, PhI.
However, our efforts were thwarted by decomposition
to the diiodo derivative, PdI2(1a ), which could be
isolated and characterized. There is no sign of palladium
black. The same diiodide was formed when 1a was
added to PdI(C6H5)(TMEDA).28 This alternative syn-
thetic approach, i.e., displacement of TMEDA from a
complex that already contains the required aryl, sug-
gests that the problem does not arise from decomposi-
tion due to Pd(0). Consequently, in the absence of olefin
substrate, these two aryl products are inherently un-
stable. However, we have been successful in isolating
several compounds of the type PdI(p-X-C6H4,)(1a ), X
Figure 1 it can be seen that the complexes PdCl2(1a or
1b) plus borohydride slowly catalyze the reaction, but
the dba complex is hardly active. Since some organic
substrates will not tolerate reaction at elevated tem-
perature, the correct choice of catalyst precursor can be
critical. With the dba precursor at ambient temperature,
there is, practically speaking, no reaction.
Interestingly, the reaction with 1b is slower than with
1a presumably for steric reasons. The Pd(0) complex of
1a is new (see Experimental Section for this complex
plus benzoquinone and pentenedione analogues), whereas
we have reported the analogue for 1b previously.25
To shed further light on the question of reaction rates
using dba, we have followed the stoichiometric room-
temperature oxidative addition of PhI, in acetone, as
shown in eqs 7 and 8:
NaBH4
PdCl2(1a or 1b) + Ph-I acetone/rt8
PdI(Ph)(1a or 1b) (7)
ca. 100% in < 3 min
Pd(dba)(1a or 1b) + Ph-I f
no reaction after 144 h (8)
The reaction of the p-cyanobromo compound, eqs 9
and 10, was also investigated.
NaBH4
PdCl2(1a ) + p-NCC6H4Br acetone8
PdBr(p-NCC6H4)(1a ) (9)
Pd(dba)(1a or 1b) + p-NCC6H4Br f
no reaction after 120 h (10)
The chemistry of eq 9 is complete in < 1 h, whereas
the reaction does not proceed using the dba compound.
Although the isolated yield of PdBr(p-NCC6H4)(1a ) is
70%, monitoring its progress via NMR suggests that the
yield is essentially quantitative. Given these drastic
differences in rate for both PhI and p-NCC6H4Br, we
(26) A similar approach to generating Pd(0) in solution is given by
Amatore and co-workers in ref 17b.
(27) In ref 17a, Pd(PPh3)4 is reported to react ca. 10 times faster
than Pd(dba)(PPh3)2; however both react on the seconds time scale, so
that our NMR measurements were not quick enough to make this
distinction.
(25) Tschoerner, M.; Trabesinger,G.; Albinati, A.; Pregosin, P. S.
Organometallics 1997, 16, 3447.
(28) de Graaf, W.; van Wegen, J .; Boersma, J .; Spek, A.; van Koten,
G. Recl. Trav. Chim. Pays-Bas 1989, 108, 275-277.

Enantioselective Heck Reaction
Organometallics, Vol. 18, No. 4, 1999 673
F igu r e 2. ORTEP plot of 3a .
Ta ble 2. Selected Bon d Len gth s (Å) a n d Bon d
An gles (d eg) for 3a a n d 3b
3a
3b
Pd-P1
2.262(3)
2.397(3)
2.04(1)
2.646(1)
168.98(8)
175.4(3)
91.8(1)
92.4(3)
92.12(8)
84.2(3)
2.2798(7)
2.3502(7)
2.062(3)
2.6284(3)
174.63(2)
175.73(8)
93.10(3)
91.15(8)
89.21(2)
86.52(7)
Pd-P2
F igu r e 3. ORTEP plot of 3b.
Pd-C1L
Pd-I
P1-Pd-I
P2-Pd-C1L
P1-Pd-P2
P1-Pd-C1L
P2-Pd-I
I-Pd-C1L
crowded bis(diphenylphosphino)propane derivative Pd-
Cl(Ph)(dppp), suggesting that the phenyl ligand does not
always exercise such a marked trans influence. The Pd-
C1L, 2.04(1) Å, and Pd-I1, 2.646(1) Å, separations in
3a are normal.
To confirm the possible correlation between Pd-P
bond length and trans influence of the aryl, an aryl
complex with a different trans influence was needed.
Given the larger number of electron-withdrawing groups
on the C6F5 ring, it seemed likely that this ligand would
possess a smaller trans influence than that of the
p-MeO2C-C6H4 analogue. Crystals of 3b were obtained
from CH2Cl2 /Et2O, and a view of this molecule is given
in Figure 3. The local coordination geometry is again
pseudo-square-planar, with the two P atoms, the iodide,
and C1L completing the immediate coordination sphere.
The Pd-P bond lengths in 3b, Pd-P1 2.2798(7) Å and
Pd-P2 2.3502(7) Å, are different, with the latter
distance now much shorter than the corresponding
separation in 3a . This observation is in agreement with
the idea that increasing the number of electron-
withdrawing groups on the phenyl weakens the σ-donor
component and thus the trans influence.
) CO2Me, CN, NO2, as well as PdI(C6F5)(1a ), all of
which contain electron-withdrawing groups on the aryl
moiety. These aryl complexes could be most effectively
prepared by adding 1a to the known complexes PdI(p-
XC6H4)(TMEDA)28 (although the oxidative addition also
works, see eq 9), and we have determined the structures
for two of these, PdI(p-(CO2Me)C6H4)(1a ), 3a , and PdI-
(C6F5)(1a ), 3b. It was hoped that these X-ray results
might provide a hint with respect to the lack of stability
for both PdI(C6H5)(1a ) and PdI(p-OMe-C6H4)(1a ).
Suitable crystals of 3a were obtained from CH2Cl2
/Et2O, and a view of the molecule is given in Figure 2.
The local coordination geometry is pseudo-square-planar
with the two P atoms, the iodide, and C1L comprising
the immediate coordination sphere. The Pd-P bond
lengths, Pd-P1 2.262(3) Å and Pd-P2 2.397(3) Å, are
quite different, with the latter at the upper end of the
literature range, supporting a large trans influence for
the p-MeO2C-C6H4 ligand (see Table 2). This long Pd-P
bond is interesting in connection with the observed
relative instability of PdI(C6H5)(1a ) in solution. Perhaps
without the electron-withdrawing ester group, the Pd-P
bond trans to the aryl (e.g., in either PdI(C6H5)(1a ) or
PdI(p-OMe-C6H4)(1a )) might be even longer and per-
haps more reactive. Hartwig and co-workers30 report
similar Pd-P distances, 2.283(2) and 2.391(2) Å, for the
structure of the ferrocene diphosphine derivative Pd-
(NPh2)(p-Me2N-C6H4)(dppf), which contains the p-
(dimethylamino) donor. Herrmann et al.31 find 2.2385(5)
and 2.3504(5) Å Pd-P distances in the somewhat less
Figure 4 shows the structure of 3a superimposed on
that for 3b.32 It appears that one of the P-phenyls of 3b
(31) Herrmann, W. A.; Brossmer, C.; Priermeier, T.; Oefele, K. J .
Organomet. Chem. 1994, 481, 97-108.
(32) There are several interesting details for both structures: for
3a the phenyl ring makes an angle of ca. 73° with the P-Pd-P plane,
and there is a modest tetrahedral distortion in that the P1(-0.16 Å)
and I1(-0.14 Å) are slightly below this plane and P2(0.10 Å) and C1L-
(0.13 Å), slightly above. The I1-Pd-P1 and I1-Pd-C1L angles are
relatively small, ca. 169° and 84°, respectively, suggesting some relief
of steric strain by moving the iodide toward the phenyl moiety. The
P-Pd-P bite angle is ca. 92°. For 3b the C6F5 ring makes an angle of
ca. 70° with the five atoms of the coordination plane, and, as in 3a ,
there is a tetrahedral distortion, albeit a smaller one: P1(ca. -0.06
Å) and I1(ca. -0.06 Å) are slightly below, and P2(ca. 0.03 Å) and C1L-
(ca. 0.04 Å), slightly above the coordination plane. The Pd-C1L, 2.062-
(3) Å, and Pd-I1, 2.6284(3) Å, separations are as expected. In contrast
to 3a the coordination angles in 3b do not deviate markedly from those
expected for a slightly distorted square planar geometry. The P-Pd-P
angle is ca. 93°.
(29) Tschoerner, M.; Kunz, R. W.; Pregosin, P. S. To appear in Magn.
Reson. Chem.
(30) Driver, M. S.; Hartwig, J . F. Organometallics 1997, 16, 5706-
5715.

674 Organometallics, Vol. 18, No. 4, 1999
Tschoerner et al.
F igu r e 4. Superimposed structures of 3a and 3b.
almost stacks with the C6F5 ring. The distances between
the two rings (e.g., C8-C1L, 3.09 Å, C9-C2L, 3.07 Å,
C13-C6L, 3.6 Å) would be roughly correct for such an
interaction. Generally speaking, arene-arene stacking
is found increasingly33 and is thought to be favored
when one ring is relatively electron poor.34 Specifically,
phenyl-perfluorophenyl stacking is now well-known.35
13C NMR Sp ectr oscop y. Since we observe the diiodo
product, i.e., the aryl is lost, we were interested in
establishing whether the ipso-aryl 13C resonance, C1,
might somehow reflect this effect. As shown in Figure
F igu r e 5. Long-range 13C,1H correlation for 3a showing
the assignment of (a) the aryl ipso-carbon (doublet of
doublets) and (b) the ester carbonyl, via the appropriate
cross-peaks. This correlation method makes use of the
three-bond coupling constants, 3J (13C,1H). Note that the
two doublets for the ipso-carbon arise due to one relatively
large 2J (P,C)trans value, 128 Hz, plus a modest 2J (P,C)cis
value of 6 Hz.
for PhLi, δ ) 174.6-186.8, in a variety of solvents.36
Consequently, we interpret our observed chemical shift
value of 167.2 ppm to suggest increasing anionic char-
acter in the Pd-C bond, with a concomitant larger trans
influence. While this NMR result cannot be taken as a
direct measure of stability, both this chemical shift and
the X-ray results hint at reasons for the instability of
PdI(C6H5)(1a ) and PdI(p-OMe-C6H4)(1a ); that is, both
the σ-bound carbon and the trans P-donor are labilized.
Com m en t. Before concluding, we, return to the
observed 71% ee found for p-OMe-C6H4OTf with 1a ,
relative to the 87% ee for the C6H5 analogue. Although
a reduction in ee when a methoxy group is introduced
is known,14b,37 it demonstrates that this substituent
markedly affects the catalysis. Further, the p-OMe-
C6H4OTf reacts ca. 5 times faster than the p-CO2Me-
C6H4OTf analogue. Admittedly, these catalytic results
and our observations in the preparative chemistry
cannot (yet) be tied together with certainty; neverthe-
less, it seems clear that there are important electronic
effects at work not only in the Heck reaction but also
in the Pd-coordination chemistry.
5, this resonance is best assigned via a long-range 13C,1H
correlation to the proton of C3. This meta proton is
correlated to both the ipso carbon and the ester carbonyl
carbon (CO2Me), δ ) ca. 168, via three-bond interactions
and thus readily reveals the appropriate cross-peaks.
The use of long-range carbon-proton correlations is
more or less unknown in organometallic chemistry, and
we suggest that this methodology has potential, espe-
cially where carbons with long T1 values are to be
measured.
The observed chemical shift for C1 in 3a , δ ) 167.2,
is at fairly high frequency. Relative to the model
compounds PdI(p-CO2MeC6H4)(TMEDA), δ ) 156.6 (C1
trans to N), and PdI(p-CO2MeC6H4)(4), δ ) 159.4 (C1
In summation, we find (i) additional support for the
m-di-tert-butyl effect on enantioselectivity in the Heck
reaction, (ii) that for 1a ,b dba significantly slows the
Heck reaction (and specifically, the oxidative addition),
and (iii) that, for some aryl derivatives, both the σ-bound
carbon and the trans P-donor are labilized.
trans to thioether-S), the observed δ ) 167.2 ppm is
consistent with the expected trans influences for these
three different donor ligands, i.e., P > S > N. Further,
this high-frequency position is not far from those found
Exp er im en ta l Section
Gen er a l Com m en ts. All reactions were done in an atmo-
sphere of dry argon using standard Schlenk techniques.
(33) Barbaro, P. Pregosin, P. S.; Salzmann, R.; Albinati, A.; Kunz,
R. W. Organometallics 1995, 14, 5160.
(34) Bahl, A.; Grahn, W.; Stadler, S.; Feiner, F.; Bourhill, G.;
Bra¨uchle, C.; Reisner, A.; J ones, P. G. Angew. Chem. 1995, 107, 1587.
(35) Coates, G. W.; Dunn, A. R.; Hennling, L. M.; Ziller, J . W.;
Lobkovsky, E. B.; Grubbs, R. H. J . Am. Chem. Soc. 1998, 120, 3641-
3649, and references therein.
(36) Wehman, E.; J astrebzyski, J . T. B. H.; Ernsting, J . M.; Grove,
D. M.; van Koten, G. J . Organomet. Chem. 1988, 353, 133-143.
(37) Casalnuovo, A. L.; Rajanbabu, T. V.; Ayers, T. A.; Warren, T.
H. J . Am. Chem. Soc. 1994, 116, 9869-9882. Schnyder, A.; Togni, A.;
Wiesli, U. Organometallics 1997, 16, 255-260, and references therein.

Enantioselective Heck Reaction
Organometallics, Vol. 18, No. 4, 1999 675
Solvents were freshly distilled from established drying agents
under argon prior to use. The complexes [PdX(TMEDA)(aryl)]
were prepared by the method of de Graf et al.28 NMR spectra
were measured using Bruker DPX300 and DRX400 MHz
spectrometers. Two-dimensional NMR spectra were measured
as described previously.23,33
procedure with (S)-1a gave (+)-2-(p-MeO-C6H4)-2,3-dhf, 74 mg
(42%) with 71% ee.
Ar yla tion of 2,3-d h f Usin g p-MeOOC-C6H4OTf. Pd-
(OAc)2 (6.7 mg, 3 mol %) and (R)-(1b) (61.9 mg, 6 mol %) were
dissolved in 4.5 mL of benzene, dhf (378 µL, 5 mmol) was
added, and the solution was stirred for 40 min. After the
addition of p-MeOOC-C6H4OTf (195 µL, 1 mmol) and EtN(i-
Pr)2 (510 µL, 3 mmol), the solution was degassed and stirred
under Ar at 40 °C. The progress of the reaction was monitored
by NMR: 24 h, 8% conv.; 40 h, 68.8% conv.; 76.5 h, 99% conv.
(78% 2-Ar-2,3-dhf, 21% 2-Ar-2,5-dhf). The reaction mixture was
added to 200 mL of pentane and filtered off. The solution was
washed with 0.1 mol/L HCl and 10% K2CO3 solution, dried
over MgSO4, and concentrated (rot. evap.). The residue was
chromatographed on SiO2 with hexane/ethyl acetate, 95:5.
Yield: 115.2 mg (56.5%) of (-)-2-(p-MeCOO-C6H4)-2,3-dhf and
44.5 mg (21.8%) of 2-(p-MeCOO-C6H4)-2,5-dhf. Total: 78.3%.
Anal. Calcd (found) for C12H12O3 (MW ) 204.2 g/mol): C, 70.58
(70.51); H, 5.92 (5.82). MS (EI+, m/e): 204.1 [M+]. 2-(p-
MeOOC-C6H4)-2,3-dhf: HPLC ChiraGrom, 1 hexane: i-PrOH
Ca ta lytic Heck Rea ction of P d Cl2(1a ) w ith Na BH4, d h f,
a n d P h OTf in Aceton e-d 6. PdCl2(1a ) (7.6 mg, 3 mol %) was
dissolved in 0.6 mL of acetone-d6 in a 5 mm NMR tube under
Ar at -30 °C and reacted with NaBH4 (0.76 mg), dhf (126 µL),
PhOTf (53 µL), and N,N-diisopropylethylamine (170 µL). The
reaction was followed by NMR at room temperature (conver-
sion/time: 15%/1 h, 41%/13 h, 48%/20 h, 90%/162 h). The run
with PdCl2(1b) was performed in the same way (conversion/
time: 15%/12 h, 22%/24 h, 44%/67 h, 61%/188 h, 80%/456 h).
Ca ta lytic Heck Rea ction of P d (d ba )(1a ) w ith d h f a n d
P h OTf in Acet on e-d 6. Pd(dba)(1a ) (9.2 mg, 3 mol %) was
dissolved in 0.6 mL of acetone-d6 at room temperature in a 5
mm NMR tube under Ar. Substrate dhf (126 µL), PhTfl (53
µL), and N,N-diisopropylethylamine (170 µL) were then added.
The tube was then sealed, and the reaction was followed by
NMR. A 4% conversion was reached after 156 h.
99:1, 0.3 mL/min, (-)-isomer tR ) 13.6 min, (+)-isomer tR
)
16.5 min; 98.7% ee, opt. rotation [R]D ) -47.2° c ) 0.6 CHCl3.
1H NMR (300.13 MHz, CDCl3): 2.59 (m, 1H), 3.14 (m, 1H),
3.94 (s, 3H, COOMe), 4.99 (m, 1H), 5.59 (m, 1H, H2), 6.49 (m,
1H), 7.44 (m, 2Har), 8.50 (m, 2Har). 13C{1H} (75.47 MHz,
CDCl3): 167.3 (CdO), 148.6 (C), 145.7 (Colefin), 130.3 (C), 129.8
(C), 125.8 (C), 99.4 (Colefin), 82.5 (C2), 52.5 (OMe), 38.3 (C3).
2-(p-MeOOC-C6H4)-2,5-dhf: HPLC ChiraGrom 2 hexane:
i-PrOH 99:1, 0.3 mL/min, (-)-isomer tR ) 9.25 min, (+)-isomer
tR ) 13.4 min; 93% ee, opt. rotation [R]D ) +174°, CDCl3.1H
NMR (300.13 MHz, CDCl3): 3.93 (s, 3H, OMe), 4.75-4.95 (m,
2H), 5.8-5.9 (m, 2H), 6.05 (m, 1H), 7.38 (m, 2Har), 8.05 (m,
2Har). The same procedure using (S)-1a as ligand gave (+)-2-
(p-MeOOC-C6H4)-2,3-dhf, 74 mg (42%) with 87% ee.
The arylation of 2,3-dhf using PhOTf and 1 was reported
recently.23a
Using Pd(dba)(1b) as catalyst precurser no reaction was
observed after 96 h.
Rea ction of P d Cl2(1b) w ith Na BH4 a n d P h I. PdCl2(1b)
(12.1 mg) was dissolved in 0.6 mL of acetone-d6 at -30 °C and
reacted with NaBH4 (0.76 mg) (formation of H2 and NaCl). PhI
(2 µL) was then added to the dark red-brown suspension. The
color changed to orange while warming the reaction to room
temperature. 31P NMR shows the formation of PdI(C6H5)(1b)
within 3 min. This compound is not stable for prolonged
periods in solution (PdI2(1b) is the observed product).
Rea ction of P d (d ba )(1a ) w ith Br C6H4CN in THF . The
reaction was carried out under Ar in a 5 mm NMR tube with
a sealed D2O-filled capillary for locking. Pd(dba)(1a ) (4 mg,
3.7 µmol) was disolved in 0.6 mL of THF, and 1.2 mg (2 equiv)
of BrC6H4CN was then added. The reaction was followed by
31P NMR. No conversion to product was observed at room
temperature during 120 h. No reaction was observed in runs
at room temperature with Pd(dba)(1a )/PhI or Pd(dba)(1b)/PhI.
Ar yla tion of 2,3-d h f Usin g p-MeO-C6H4OTf. A Schlenk-
ampule was loaded with Pd(OAc)2 (6.7 mg, 3 mol %) and (R)-
1b (61.9 mg, 6 mol %), and these solids were dissolved in 4.5
mL of benzene. After addition of dhf (378 µL, 5 mmol), the
solution was stirred at 40 °C for 40 min, and the solution was
then treated with p-MeO-C6H4-OTf (178 µL, 1 mmol) and
EtN(i-Pr)2 (510 µL, 3 mmol). The solution was then degassed
and stirred under Ar at 40 °C. The progress of the reaction
was monitored by NMR. After 24 h 89% conversion was
observed. After 36 h 100% conversion was observed. One
observes 91% of (-)-2-(p-MeO-C6H4)-2,3-dhf, plus 9% (-)-2-
(p-MeO-C6H4)-2,5-dhf. The reaction mixture was added to 200
mL of pentane and filtered off. The resulting solution was
washed with 0.1 mol/L HCl, 10% K2CO3 solution, dried over
MgSO4, and then concentrated (rot. evap.). The crude product
was purified by column chromatography (silica, hexane:ethyl
acetate 95:5). Yield: 133 mg (75.5%) of (-)-2-(p-MeO-C6H4)-
2,3-dhf: MS (EI+, m/e): 176.1 [M+]. HPLC: ChiraGrom 1
hexane: i-PrOH 99:1, 0.3 mL/min, (+)-isomer tR ) 6.2, (-)-
isomer tR ) 6.9, gave 91.4% ee, opt. rotation [R]D ) -92.7°, c
) 0.51 CHCl3.1H NMR (300.13 MHz, CDCl3): 2.64 (m, 1H),
3.06 (m, 1H), 3.83 (s, 3H, OMe), 4.98 (m, 1H), 5.49 (m, 1H,
H2), 6.45 (m, 1H), 6.91 (m, 2Har), 7.32 (m, 2Har). 13C{1H} (75.47
MHz, CDCl3): 159.5 (Cpara), 145.5 (Colefin), 135.4 (Cipso), 127.4
(Corto), 114.2 (Cmeta), 99.4 (Colefin), 82.5 (C2), 55.7 (OMe), 37.9
(C3). (+)-2-(p-MeO-C6H4)-2,5-dhf: 1H NMR (300.13 MHz,
CDCl3): 3.82 (s, 3H, OMe), 4.7-4.8 (m, 1H, H5), 4.8-4.9 (m,
1H, H5′), 5.75 (m, 1H), 5.88(m, 1H), 6.07 (m, 1H), 6.90 (m,
2Har), 7.25 (m, 2Har). 13C{1H} (75.47 MHz, CDCl3): 159.8
(Cpara), 134.5 (Cipso), 130.4 (Colefin), 128.3 (Corto), 127.1 (Colefin),
114.3 (Cmeta), 87.9 (C2), 75.9 (C5), 55.7 (OMe). The same
Ar yla tion of 5-Me-2,3-d h f. Pd(OAc)2 (6.7 mg, 3 mol %) and
(R)-1b (61.9 mg, 6 mol %) were dissolved in 4.5 mL of benzene,
and then 5-Me-2,3-dhf (456 µL, 5 mmol) was added. After 20
min, PhOTf (157 µL, 1 mmol) and EtN(i-Pr)2 (510 µL, 3 mmol)
were added; the solution was degassed and then stirred under
Ar at 70 °C for 7 days. The reaction mixture was then treated
with 200 mL of hexane and the resulting suspension filtered.
The solution was washed first with 0.1 M HCl and then with
10% K2CO3. Drying over MgSO4 was followed by solvent
distillation in vacuo. The residue was chromatographed on
SiO2 using hexane/ethyl acetate, 97.5:2.5, as eluent to afford
60 mg (38%) of 2-Me-2-Ph-2,3-dhf, rf ) 0.25, and 10 mg (ca.
6%) of 2-Me-2-Ph-2,5-dhf, rf ) 0.04. For 2-Me-2-Ph-2,3-dhf,
only one enantiomer was detected (NMR, Eu(hfc)3, ee esti-
mated to be >98%), opt. rotation [a]D ) +43.1° c ) 0.585
CHCl3. 1H NMR (300.13 MHz, CDCl3, J in Hz): d, 1.66 (s, 3H,
Me), 2.79 (ddd, 1H, H3, J ) 15.0, 2.5, 2.5), 2.87 (ddd, 1H, H3′,
J ) 15.0, 2.5, 2.5), 4.88 (ddd, 1H, H4, J ) 2.5, 2.5, 2.5), 6.42
(ddd, 1H, H5, J ) 2.5, 2.5, 2.5), 7.29 (t, 1H, Hp, J ) 7.2), 7.37
(t, 2H, Hm, J ) 7.9), 7.44 (d, 2H, Ho, J ) 8.0) (from NOESY:
H3 cis to the methyl group). 13C{1H} NMR (75.47 MHz,
CDCl3): d, 29.5 (Me), 44.6 (C3), 87.8 (C2), 98.9 (C4), 124.8 (Co),
127.2 (Cp), 128.7 (Cm), 144.6 (C5), 148.1 (Ci). MS (EI+, m/e):
160.1 [M]+, 145.1 [M - CH3]+, 77.0 [C6H5]+. (R)-1a gives 18%
yield of (+)-2-Me-2-Ph-2,3-dhf with 20% ee (NMR) and the
recovery of 35% PhOTf.
Cr ysta llogr a p h y. Crystals of 3a were mounted on an
CAD4 diffractometer for the unit cell and space group deter-
minations and for the data collection. The unit cell constants,
space group determination, and the data collection for 3b were
carried out on a Siemens SMART diffractometer equipped with
a CCD detector.
Str u ctu r a l Stu d y of P d I(p-MeO2C-C6H4)(1a ), 3a . Data
were measured with variable scan speed to ensure constant
statistical precision on the collected intensities. Three standard

676 Organometallics, Vol. 18, No. 4, 1999
Ta ble 3. Exp er im en ta l Da ta for th e X-r a y Diffr a ction Stu d y of Com p ou n d s
Tschoerner et al.
P d I(p-MeO2C-C6H4)(MeO-Bip h ep ) (3a ) a n d P d I(C6F 5)(MeO-Bip h ep ) (3b)
3a
3b
formula
mol wt
diffractometer
cryst syst
space group
a, Å
C
46H39IO4P2Pd
C44H32IF5O2P2Pd
982.94
SMART CCD
monoclinic
P21
10.9401(5)
14.0411(7)
13.7097(7)
104.130(1)
2042.2(2)
2
951.08
CAD4
monoclinic
P21
10.147(4)
19.239(9)
11.419(6)
91.62(4)
2228(2)
2
1.417
12.21
b, Å
c, Å
â, deg
V, Å3
Z
r
(calcd), g cm-3
1.598
13.47
m, cm-1
radiation
θ range, deg
Mo KR (graphite monochrom λ ) 0.710 69 Å)
2.5 < θ < 25.0
1.5 < θ < 29.4
21 699
9964
8655 [|Fo| > 2.0σ(|F|)]
0.8692-1.0000
0.023
no. of data colld
no. of ind data
no. of obs reflns (no)
transmission coeff
4022
3198 [|Fo| > 3.0σ(|F|)]
0.7049-1.2663
a
Rav
R (obs, reflns)
0.052
0.068
2.518
0.026
0.051 (all data)
0.879
b
Rw, R2
w
GOFc
Rav ) ∑|Fo - Fo ,av|/∑i|Fo |, R ) ∑(|Fo - (1/k)Fc|)/∑|Fo|. Rw ) [∑w(Fo - (1/k)Fc)2/∑w|Fo| ]1/2 where w ) [σ2(Fo)]-1; σ(Fo) ) [σ2(Fo2) +
a
2
2
2
b
2
f4(Fo2)]1/2/2Fo and f ) 0.04. R2 ) [∑w(Fo - (1/k)Fc2)2/∑w|Fo | ] where w ) [σ2(Fo2) + (aP)2 + P]-1, a ) 0.023 and P ) {[Fo + max(Fo2)]/
2
2 2
2
w
2
3.0}. c GOF: (a) [∑w(Fo - (1/k)Fc)2/(no - nv)]1/2; (b) [∑w(Fo - (1/k)Fc2)2/(no - nv)]1/2
.
reflections were used to check the stability of the crystals and
of the experimental conditions and were measured every hour.
The collected intensities were corrected for Lorentz and
polarization factors.38 An empirical absorption correction39 was
also applied by using azimuthal (ψ) scans of three “high-ø” (ø
g 85°) reflections. Of the 4022 independent data collected, 3198
were considered as observed and used for the solution and
refinement of the structure. Selected crystallographic and
other relevant data are listed in Table 3 and in Table S1. The
standard deviations on intensities were calculated in term of
statistics alone, while those on Fo were calculated as shown
in Table 3. The structure was solved by direct and Fourier
methods and refined by full-matrix least-squares38 (the func-
were collected (up to sin θ/λ ) 0.692 Å-1) by using ω scans, in
steps of 0.3° with a sample-detector distance fixed at 50 mm.
For each of the resulting 1250 “frames”, counting time was 30
s. Data were corrected for Lorentz and polarization factors and
empirically for absorption using the SADABS program.43
Selected crystallographic and other relevant data are listed
in Table 3 and in Table S1. The standard deviations on
intensities were calculated in terms of statistics alone, while
those on Fo2 were calculated as shown in Table 3. The structure
was solved by direct and Fourier methods and refined by full-
matrix least-squares,44 minimizing the function ∑[w(Fo - (1/
2
k)Fc2)2].
Anisotropic displacement parameters were used for all
atoms. The contribution of the hydrogen atoms in their
tion minimized being ∑[w(Fo - (1/k)Fc)2]). No extinction
2
correction was deemed necessary. The scattering factors used,
corrected for the real and imaginary parts of the anomalous
dispersion, were taken from the literature.40 The final full-
matrix least-squares refinement cycles were carried out using
anisotropic displacement parameters for the “heavy atoms”
(Pd, I, and P), the aryl ligand, and the O-Me moieties of the
Biphep ligand. The remaining atoms were treated isotropically.
The contribution of the hydrogen atoms in calculated positions
(C-H ) 0.95 (Å), B(H) ) 1.3 × B(Cbonded) (Å2)) was taken into
account but not refined. Upon convergence (see Table S1) no
significant features were found in the Fourier difference map.
The handedness of the structures was tested by refining both
enantiomorphs; the coordinates giving the significantly41 lower
Rw factors were used. All calculations were carried out using
the Enraf-Nonius MOLEN package.38
calculated positions (C-H ) 0.95(Å), B(H) ) 1.5 × B(Cbonded
)
(Å2)) was included in the refinement using a riding model. The
handedness of the structure was tested by refining the Flack
parameter.45 No extinction correction was deemed necessary.
Upon convergence (see Table S1) the final Fourier difference
map showed no significant peaks. The scattering factors used,
corrected for the real and imaginary parts of the anomalous
dispersion, were taken from the literature.46 All calculations
were carried out by using the PC version of the SHELX-97
programs.44
P d (d ba )(1a ). Pd2(dba)3‚CHCl3 (40 mg, 38.6 µmol) and (S)-
MeO-Biphep (45.0 mg, 77 µmol) were dissolved in 2 mL of
CH2Cl2 and stirred for 16 h. The orange brown solution was
filtered over Celite and the filtrate evaporated to dryness in
vacuo. The residue was washed with ether (6 × 1.5 mL) and
the complex recrystallized from CH2Cl2/Et2O/hexane to afford
71 mg (100%) as orange microcrystals. mp ) 225 °C (dec). Anal.
Calcd (found) for C55H46O3P2Pd (MW ) 923.3 g/mol): C, 71.55
(70.77); H, 5.02 (5.13). MS (FAB+, m/e): 923 [M + 1]+, 688
[Pd(MeO-Biphep)]+. 31P{1H} NMR: (121.5 MHz, CD2Cl2), δ,
Str u ctu r a l Stu d y of P d I(C6F 5)(1a ), 3b. The space group
was unambiguously determined from the systematic absences,
while the cell constants were refined, at the end of the data
collection, using 8192 reflections (θmax e25°) with the data
reduction software SAINT,42 developed by Siemens. The data
1
25.3 (b), 24.3 (b). H NMR (300.13 MHz, CD2Cl2): δ, 3.39 (s,
(38) MolEN: Molecular Structure Solution Procedure; Enraf-Nonius,
Delft, The Netherlands, 1990.
(39) North, A. C. T.; Philips, D. C.; Mathews, F. S. Acta Crystallogr.
1968, A24, 351.
(40) International Tables for X-ray Crystallography; Kynoch: Bir-
mingham, England, 1974; Vol. IV.
(41) Hamilton, W. C. Acta Crystallogr. 1965, 17, 502.
(42) SAINT: SAX Area Detector Integration; Siemens Analytical
Instrumentation, 1996.
(43) Sheldrick, G. M. SADABS; Universita¨t Go¨ttingen. To be
published.
(44) Sheldrick, G. M SHELX-97. Structure Solution and Refinement
Package; Universita¨t Go¨ttingen, 1997.
(45) Flack, H. D. Acta Crystallogr. 1983, A39, 876.
(46) International Tables for X-ray Crystallography; Wilson, A. J .
C., Ed.; Kluwer Academic Publisher: Dordrecht, The Netherlands,
1992; Vol. C.

Enantioselective Heck Reaction
Organometallics, Vol. 18, No. 4, 1999 677
3H), 3.41 (s, 3H), 4.81 (m, b, 1H), 4.98 (m, b, 1H), 6.25 (m,
(C4), 126.8 (dd, C3, J PC ) 8.7, 4.6 Hz), 123.7-137.2 (16
doublets, 24 CH), 126.5-134.5 (8 doublets, 8 Ci), 138.7 (dd,
C2, J PC ) 4.8, 1.6 Hz), 157.8 (d, C-OMe, J PC ) 9.7 Hz), 158.3
(d, C-OMe, J PC ) 10.6 Hz), 167.2 (dd, C1, J PC ) 128.0, 6.0
Hz) 168.2 (CdO).
P d I(C6F 5)(1a ) 3b. Reaction conditions as in 3a : THF, 60
°C, 96 h. Yield: 50 mg (51%) as orange brown crystals from
CH2Cl2 /Et2O/hexane. Mp: 260-265 °C (dec). Anal. Calcd
(found) for C44H32F5O2P2PdI (MW ) 983.0 g/mol): C, 53.76
(53.70); H, 3.28 (3.45). MS (FAB+, m/e): 855 [M - I]+, 815 [M
- C6F5]+, 688.1 [Pd(MeO-Biphep)]+. 31P NMR (121.5 MHz,
THF): δ, 20.9 (m (8 line multiplet)), 13.4 (m (ca. 15 lines)). 1H
NMR (300 MHz, CD2Cl2): δ, 3.39 (s, 3H, OCH3), 3.53 (s, 3H,
OCH3), 6.4-6.5 (m, 3H), 6.84 (m, 1H), 7.1-7.4 (m, 9H), 7.46
(m, 1H), 7.5-7.58 (m, 3H), 7.6-7.8 (m, 5H). 13C-19F HMQC:
δ, 122.1 (d, C1, J PC ) 118 Hz), 136.0/136.7 (C3/C5), 137.9 (C4),
145.0/145.7 (C2/C6). 13C{1H} NMR (100.6 MHz, CD2Cl2): δ,
3H), 6.39 (m, 1H), 6.52 (m, 1H), 6.7-7.8 (m, 33H). 13C{1H}
NMR (75.47 MHz, CD2Cl2): δ, 55.0 (2 OMe), 68.9 (d, b, J PC
)
26.8 Hz), 73.6 (b), 110.5 (CH ortho to OMe), 110.7 (CH ortho
to OMe), 122.0-135 (32 CH), 127-136 (10 Ci), 124.2 (b, CdC,
uncoord.), 143.5 (b, CdC uncoord.), 157.9 (d, C-OMe, J PC
)
7.9 Hz), 158.1 (d, C-OMe, J PC ) 7.6 Hz), 183.01 (b, CdO).
The structure of Pd(dba)(1b) was reported earlier. In contrast
to the observations made for Pd(dba)(1b), there appears to be
no restricted rotation around the P-C6H5 bonds of Pd(dba)-
1
(1a ), based on 31P,1H-correlation and H integration measure-
ments.
Two additional Pd(0) compexes were prepared, although not
discussed.
P d (cyclop en t -4-en e-1,3-d ion e)(1a ). Pd2(dba)3‚dba (57.5
mg, 0.05 mmol) and (S)-MeO-Biphep (58.3 mg, 0.1 mmol)
were dissolved in 2 mL of THF. To the orange-brown solution
formed was added 1.2 equiv of cyclopent-4-ene-1,3-dione (11.4
mg, 0.12 mmol). The color changed to yellow. After stirring
for 16 h the solution was filtered and taken to dryness in a
vacuum. The residue was washed with ether (7 × 1.0 mL).
The product can be recrystallized from CH2Cl2/Et2O. Yield: 78
mg (99%) as yellow microcrystals, mp 265 °C (dec). Anal. Calcd
(found) for C43H36O4P2Pd (MW ) 785.1 g/mol): C, 65.78 (65.78);
H, 4.62 (4.78). MS (FAB+, m/e): 785 [M + 1]+, 688 [Pd-
(MeOBiphep)]+. 31P{1H} NMR (121.5 MHz, CD2Cl2): δ, 28.4
(d, J ) 16.8 Hz), 26.2 (d, J ) 16.8 Hz).1H NMR (400.13 MHz,
CD2Cl2): δ, 2.33 (m, 1H), 2.80 (m, 1H), 3.44 (s, 3H), 3.47 (s,
3H), 4.36 (m, 1H), 4.80 (m, 1H), 6.28-6.36 (m, 2H), 6.51 (ddd,
1H, J ) 10.4, 7.9, 1.0 Hz), 6.59 (ddd, 1H, J ) 10.0, 7.9, 1.0
Hz), 6.83-7.83 (m, 22H). 13C{1H} NMR (100.62 MHz, CD2-
55.0 (OMe), 55.4 (OMe), 112.3 (d, J PC ) 2 Hz), 112.9 (d, J PC
)
2 Hz), 124.4 (dd, CH, J PC ) 11.0, 3.0 Hz), 125.3-137.8 (13
doublets, 19 CH), 126-134 (8 doublets, 8 Ci), 157.9 (d, C-OMe,
J PC ) 10 Hz), 158.3 (d, C-OMe, J PC ) 11 Hz). 19F NMR (282.4
MHz, CD2Cl2): δ, -113.5 (m, 1F), -115.4 (m, 1F), -163.1 (t,
1F, J ) 20), -164.2 (m, 1F), -165.2 (m, 1F). The complexity
arises from 2J (P,P) plus 19F spin-spin coupling.
P d Br (C6F 5)(1a ). Reaction conditions as in 3a : THF, 60 °C,
96 h. Yield: 48.8 mg (52%) as orange crystals from CH2Cl2
/Et2O/hexane. Mp; 275 °C (dec). Anal. Calcd (found) for
C
44H32F5O2P2PdBr (MW ) 936.0 g/mol): C, 56.46 (56.51); H,
3.44 (3.69). MS (FAB+, m/e): 855 [M - Br]+, 769.0 [M - C6F5]+,
688.1 [Pd(MeO-Biphep)]+. 31P NMR (121.5 MHz, CD2Cl2): δ,
27.45 (8 line multiplet, J PP ca. 29 Hz), 15.7 (ca. 15 line
multiplet, J PP ca. 29 Hz). 1H NMR (300 MHz, CD2Cl2): δ, 3.39
(s, 3H, OCH3), 3.53 (s, 3H, OCH3), 6.33-6.5 (m, 3H), 6.85 (td,
1H, J ) 8.1, 2.3 Hz), 7.09-7.34 (m, 9H), 7.41-7.55 (m, 4H),
7.62 (dd, 1H, J ) 11.5, 8.0 Hz), 7.65-7.74 (m, 2H), 7.76-7.86
(m, 2H). 13C-19F HMQC δ, 126.7 (d, C1, J PC ) 133 Hz), 135.9/
136.8 (C3/C5), 137.8 (C4), 145.2/146.0 (C2/C6). 13C{1H} NMR
(75.47 MHz, CD2Cl2): δ, 55.1 (OMe), 55.4 (OMe), 112.5 (d, J PC
) 2.6 Hz), 112.9 (d, J PC ) 2.6 Hz), 124.4 (dd, CH, J PC ) 10.6,
3.0 Hz), 125.1-137 (13 doublets, 19 CH), 126-134 (8 doublets,
8 Ci), 157.9 (d, C-OMe, J PC ) 10.4 Hz), 158.4 (d, C-OMe, J PC
) 11.3 Hz). 19F NMR (282.4 MHz, CD2Cl2): δ, -115.6 (m, 1F),
-116.5 (m, 1F), -163.1 (t, 1F, J ) 20), -164.0 (m, 1F), -165.83
(m, 1F). The complexity arises from 2J (P,P) plus 19F spin-
spin coupling.
Cl2): δ, 55.1 (OMe), 55.2 (OMe), 66.1 (CH2), 71.3 (dd, J PC
)
22.4, 2.7 Hz), 73.4 (dd, J PC ) 22.4, 2.3 Hz), 111.0-134.9 (18
doublets, 26 CH), 127-136 (8 doublets, Ci), 158.0 (d, C-OMe,
J PC ) 7.5 Hz), 158.1 (d, C-OMe, J PC ) 7.7 Hz), 200.8 (dd, Cd
O, J PC ) 4.3, 2.2 Hz), 201.9 (dd, CdO, J PC ) 5.0, 1.6 Hz).
P d (ben zoqu in on e)(1a ). Synthesis was performed as for
[Pd(cyclopente-4-ene-1,3-dione){(S)-MeO-Biphep}], but 2 equiv
of benzoquinone (21.6 mg, 0.2 mmol) was employed. Yield: 52
mg (65%) as red microcrystals, mp 210 °C (dec). Anal. Calcd
(found) for C44H36O4P2Pd (MW ) 797.1 g/mol): C, 66.3 (65.9);
H, 4.55 (5.09). MS (FAB+, m/e): 797 [M + 1]+, 688 [Pd(MeO-
Biphep)]+. 31P{1H} NMR (121.5 MHz, CD2Cl2): δ, 30.9. 1H
NMR (300.13 MHz, CD2Cl2): δ, 3.44 (s, 6H), 5.37 (m, 2H), 5.69
(m, 2H), 6.32 (m, b, 1H), 6.35 (m, b, 1H), 6.6 (m, 2H), 6.9 (m,
2H), 7.1-7.8 (m, 20H). 13C{1H} NMR (75.47 MHz, CD2Cl2):
δ, 55.1 (2 OMe), 104.21 (d, CH, J PC ) 3.5 Hz), 104.26 (d, CH,
J PC ) 3.5 Hz), 105.26 (d, CH, J PC ) 3.5 Hz), 105.31 (d, CH,
J PC ) 3.5 Hz), 111.3-135.2 (12 doublets, 3 broad signals, 26
P d Br (4-cya n op h en yl)(1a ). This was prepared as de-
scribed for 3a . Reaction conditions: THF, 40 °C, 14 h. Yield:
60 mg (69%) as yellow crystals from CH2Cl2/Et2O. Mp: 173
°C (dec). Anal. Calcd (found) for C45H36NO2P2PdBr (MW )
871.1 g/mol): C, 62.05 (61.78); H, 4.17 (4.35); N, 1.61 (1.62).
MS (FAB+, m/e) 790 [M - Br]+, 769.0 [M - C6H4CN]+, 688.1
[Pd(MeO-Biphep)]+. 31P NMR (121.5 MHz, THF): δ, 12.4 (d,
J ) 41.4 Hz), 26.9 (d, J ) 41.4 Hz). 1H NMR (400.13 MHz,
CD2Cl2): δ, 3.38 (s, 3H, OCH3), 3.49 (s, 3H, OCH3), 6.36-6.47
(m, 3H), 6.88 (td, 1H, J ) 8.2, 2.2 Hz), 6.96 (m, 2H, H3), 7.05-
7.37 (m, 12H), 7.43 (m, 1H), 7.47-7.52 (m, 3H), 7.57 (m, 2H,
H2), 7.64-7.72 (m, 2H), 7.74-7.85 (m, 4H). 13C{1H} NMR
(100.6 MHz, CD2Cl2): δ, 55.1 (OMe), 55.3 (OMe), 105.9 (d, C4,
J PC ) 1.6 Hz), 112.16 (d, J PC ) 2.5 Hz), 112.3 (d, J PC ) 1.6
Hz), 120.5 (d, CtN, J PC ) 0.7 Hz), 123.6-136.6 (16 doublets,
CH), 127.5-135.5 (8 doublets, Ci), 158.06 (d, C-OMe, J PC
)
5.2 Hz), 158.14 (d, C-OMe, J PC ) 5.2 Hz), 185.2 (d, CdO, J PC
) 3 Hz), 185.24 (d, CdO, J PC ) 3 Hz).
P d I(MeOOC-C6H4)(1a ), 3a . (S)-MeO-Biphep (58.3 mg,
0.1 mmol) and [PdI(TMEDA)(MeOOC-C6H4)] (48.5 mg, 0.1
mmol) were dissolved in 5 mL of THF and stirred at 40 °C for
14 h. The solution was filtered and taken to dryness. The
residue was washed with Et2O (5 × 2 mL) and the crude yellow
product recrystallyzed from CH2Cl2 and Et2O (1 mL CH2Cl2
condensed into the Schlenk tube and layered with Et2O). Over
a period of 24 h crystals were formed, which were washed with
Et2O and dried in a vacuum. Yield: 72.7 mg (76.4%) of yellow
crystals, mp 160 °C (dec). Anal. Calcd (found) for C46H39O4P2-
PdI (MW ) 951.1 g/mol) C, 58.09 (57.81); H, 4.13 (4.31). MS
(FAB+, m/e): 823.0 [Pd(C6H4COOMe)(MeO-Biphep)]+, 814.9
[PdI(MeO-Biphep)]+. 31P{1H} NMR (121.5 MHz, CD2Cl2): δ,
10.1 (J ) 42.7 Hz), 21.5 (J ) 42.7 Hz). 1H NMR (300 MHz,
CD2Cl2): δ, 3.34 (s, 3H, OCH3), 3.48 (s, 3H, OCH3), 3.79 (s,
3H, COOCH3), 6.36-6.41 (m, 2H), 6.44-6.51 (m, 1H), 6.8-
6.9 (m, 1H), 7.0-7.1 (m, 2H), 7.11-7.8 (m, 24H). 13C{1H} NMR
(100.6 MHz, CD2Cl2): δ, 51.7 (COOMe), 55.0 (OMe), 55.3
(OMe), 111.2 (d, J PC ) 1.6 Hz), 112.2 (d, J PC ) 1.6 Hz), 124.6
24 CH), 126.7-133.5 (8 doublets, 8 Ci), 129.2 (dd, C3, J PC
)
8.9, 1.8 Hz), 138.4 (dd, C2, J PC ) 5.0, 1.8 Hz), 157.8 (d, C-OMe,
J PC ) 10 Hz), 158.4 (d, C-OMe, J PC ) 11 Hz), 172.0 (dd, C1,
J PC ) 132.7, 4.1 Hz).
Bor oh ydr ide Rou te to P dBr (4-cyan oph en yl)(1a). NaBH4
and LiBHEt3 have been recently used to prepare zerovalent
complexes.47,48 NaBH4 (3.8 mg, 0.1 mmol) was added to a
(47) Wicht, D. K.; Zhuravel, M. A.; Gregush, R. V.; Glueck, D. S.;
Guzei, I. A.; Liable-Sands, L. M.; Rheingold, A. L. Organometallics
1998, 17, 1412-1419.

678 Organometallics, Vol. 18, No. 4, 1999
Tschoerner et al.
suspension of [PdCl2{(S)-MeO-Biphep}] (38 mg, 50 µmol) in
acetone at -50 °C. There was an immediate color change from
yellow to red-orange. The mixture was slowly warmed to room
temperature and 4-bromo-benzonitrile (10 mg, 55 µmol) added.
The suspension became light yellow within 1 h. Filtration,
distillation in vacuo of the solvent, and recrystallization of the
residue from CH2Cl2/Et2O gave 30.4 mg (70%) of the pure title
compound. Analytical data are almost identical to those given
above.
P d Br (p -n it r op h en yl)(1a ). This was prepared as in 3a
except that the mixture was refluxed for 2 h in THF. The
product (28.5 mg, 32%) was obtained from CH2Cl2/hexane, mp
165 °C (dec). Anal. Calcd (found) for C44H36NO4P2PdBr (MW
) 891.0 g/mol): C, 59.31 (59.05); H, 4.07 (4.10); N, 1.57 (1.66).
MS (FAB+, m/e): 810.2 [M - Br]+, 769.0 [M - C6H4NO2]+,
688.1 [Pd(MeO-Biphep)]+. 31P NMR (121.5 MHz, THF): δ,
12.5 (d, J ca. 41 Hz), 26.6 (d, ca. 41 Hz). 1H NMR (300.13 MHz,
CD2Cl2): δ, 3.40 (s, 3H), 3.49 (s, 3H), 6.35-6.5 (m, 3H), 6.89
(m, 1H), 7.0-8.0 (m, 26H). 13C{1H} NMR (75.47 MHz, CD2-
Cl2): δ, 55.15 (OMe), 55.30 (OMe), 112.27 (d, J PC ) 2.3 Hz),
112.36 (d, J PC ) 1.8 Hz), 120.32 (dd, C3, J PC ) 9.1, 1.8 Hz),
123.7-136.4 (16 doublets, 24 CH), 120-136.7 (8 doublets, 8
Ci), 137.9 (dd, C2, J PC ) 4.6, 1.8 Hz), 144.5 (C4), 157.8 (d,
C-OMe, J PC ) 10.5 Hz), 158.4 (d, C-OMe, J PC ) 11.3 Hz),
176.7 (dd, C1, J PC ) 133.0, 4.0 Hz).
Ack n ow led gm en t. P.S.P. thanks the Swiss Na-
tional Science Foundation and the ETH Zurich for
financial support. A.A. thanks MURST for partial
support. We also thank F. Hoffmann-La Roche AG for
the gift of MeO-Biphep ligands, as well as J ohnson
Matthey for the loan of PdCl2.
Su p p or tin g In for m a tion Ava ila ble: Text giving experi-
mental details and a full listing of crystallographic data for3a
and 3b, including tables of positional and isotropic equivalent
displacement parameters, calculated positions of the hydrogen
atoms, anisotropic displacement parameters, bond distances
and angles, and ORTEP figures showing the full numbering
schemes. This material is available free of charge via the
Internet at http://pubs.acs.org.
(48) Hodgson, M.; Parker, D.; Taylor, R. J .; Feruson, G. Organome-
tallics 1988, 7, 1761-1766.
OM980783L