The role of ligand transformations on the performance of phosphite- and phosphinite-based palladium catalysts in the Suzuki reaction

Article abstract of DOI:10.1021/om020941f

The orthometalated complex [{Pd(μ-Cl){κ²-P,C-P(OC₆H₂-2,4-^t Bu₂)(OC₆H₃-2,4-^tBu₂ )₂}}₂] reacts with phenylboronic acid hydrate and K₂CO₃ in dimethylacetamide to give [Pd{κ²-P,C-μ²-O-P(O)(C₆H₂-2 ,4-^tBu₂)(C₆H₃-2,4-^tBu ₂)(DMAc)}]. When the reaction is repeated in dimethylformamide 3,3′,5,5′-tetra-tert-butyl-2,2′-biphenol is isolated. Both compounds have been characterized crystallographically. The reaction of palladium dichloride with PⁱPr₂-(OC₆H₄-4-Et) in 2-methoxyethanol followed by recrystallization in the presence of ethanol leads to the formation of trans-[PdCl₂{PⁱPr₂(OEt)}₂], which was also characterized by crystallography. To determine whether related solvolytic processes have a bearing on catalytic activity, the performance of a range of catalysts with hydrolyzed and nonhydrolyzed ligands was assessed in the Suzuki coupling of aryl bromides. In some cases it was evident that hydrolysis plays a significant role on the catalytic activity; however, this depends not only on the ligand, but also on the combination of ligand and palladium precursor.

Full text of DOI:10.1021/om020941f

1364
Organometallics 2003, 22, 1364-1371
Articles
Th e Role of Liga n d Tr a n sfor m a tion s on th e P er for m a n ce
of P h osp h ite- a n d P h osp h in ite-Ba sed P a lla d iu m
Ca ta lysts in th e Su zu k i Rea ction
Robin B. Bedford,* Samantha L. Hazelwood,† and Michael E. Limmert
School of Chemistry, University of Exeter, Exeter EX4 4QD, United Kingdom
J ohn M. Brown, Shailesh Ramdeehul, and Andrew R. Cowley
Department of Chemistry, University of Oxford, South Parks Road,
Oxford OX1 3QY, United Kingdom
Simon J . Coles and Michael B. Hursthouse
Department of Chemistry, EPSRC National Crystallography Service, University of
Southampton, Highfield Road, Southampton SO17 1BJ , United Kingdom
Received November 13, 2002
The orthometalated complex [{Pd(µ-Cl){κ2-P,C-P(OC6H2-2,4-tBu2)(OC6H3-2,4-tBu2)2}}2]
reacts with phenylboronic acid hydrate and K2CO3 in dimethylacetamide to give [Pd{κ2-
P,C-µ2-O-P(O)(C6H2-2,4-tBu2)(C6H3-2,4-tBu2)(DMAc)}]. When the reaction is repeated in
dimethylformamide 3,3′,5,5′-tetra-tert-butyl-2,2′-biphenol is isolated. Both compounds have
been characterized crystallographically. The reaction of palladium dichloride with PiPr2-
(OC6H4-4-Et) in 2-methoxyethanol followed by recrystallization in the presence of ethanol
leads to the formation of trans-[PdCl2{PiPr2(OEt)}2], which was also characterized by
crystallography. To determine whether related solvolytic processes have a bearing on catalytic
activity, the performance of a range of catalysts with “hydrolyzed” and “nonhydrolyzed”
ligands was assessed in the Suzuki coupling of aryl bromides. In some cases it was evident
that hydrolysis plays a significant role on the catalytic activity; however, this depends not
only on the ligand, but also on the combination of ligand and palladium precursor.
Sch em e 1. Th e Su zu k i Bia r yl Cou p lin g Rea ction
In tr od u ction
The coupling of aryl halides with aryl boronic acids,
the Suzuki reaction (Scheme 1), is one of the most
powerful and versatile methods for the synthesis of
biaryls.1 There has recently been considerable interest
in the development of a new, high-activity catalyst that
can be used in low loadings in such reactions, and
palladacyclic complexes have played a significant role
in this regard.
The area was initiated by Beller, Herrmann and co-
workers, who demonstrated that the palladacyclic com-
plex 1 acts as a good catalyst in the coupling of aryl
bromide substrates.2 We demonstrated that the pincer
complexes 2 and the orthopalladated triarylphosphite
and phosphinite complexes 3 show good to excellent
activity in such couplings,3 while Cole-Hamilton and co-
workers demonstrated that the phosphine-based com-
plexes 4 can also be used.4 High activity is not limited
to phosphorus-based palladacycles but is also demon-
strated by the S-donor complexes 5,5 the imine-based
catalyst 6,6 and related oxime-containing species, 7.7
(3) (a) Bedford, R. B.; Draper, S. M.; Scully P. N.; Welch, S. L. New
J . Chem. 2000, 24, 745. (b) Albisson, D. A.; Bedford, R. B.; Lawrence,
S. E.; Scully, P. N. Chem. Commun. 1998, 2095. (c) Bedford, R. B.;
Welch, S. L. Chem. Commun. 2001, 129.
* To whom correspondence should be addressed. Fax: +44 (0) 1392
263434. E-mail: r.bedford@ex.ac.uk.
(4) Gibson, S.; Foster, D. F.; Eastham, G. R.; Tooze, R. P.; Cole-
Hamilton, D. J . Chem. Commun. 2001, 779.
(5) Zim, D.; Gruber, A. S.; Ebeling, G.; Dupont, J .; Monteiro, A. L.
Org. Letts. 2000, 2, 2881.
(6) Weissman, H.; Milstein, D. Chem. Commun. 1999, 1901.
(7) (a) Alonso, D. A.; Na´jera, C.; Pacheco, M. C. J . Org. Chem. 2002,
67, 5588. (b) Botella, L.; Na´jera, C. Angew. Chem., Int. Ed. 2002, 41,
179.
† Ne´e Welch.
(1) Recent reviews: (a) Miyaura, N.; Suzuki, A. Chem. Rev. 1995,
95, 2457. (b) Stanforth, S. P. Tetrahedron 1998, 54, 263. (c) Suzuki, A.
J . Organomet. Chem. 1999, 576, 147.
(2) Beller, M.; Fischer, H.; Herrmann, W. A.; O¨ fele, K.; Brossmer,
C. Angew. Chem., Int. Ed. 1995, 34, 1848.
10.1021/om020941f CCC: $25.00 © 2003 American Chemical Society
Publication on Web 02/22/2003

Palladium Complex Catalysts in the Suzuki Reaction
Organometallics, Vol. 22, No. 7, 2003 1365
Ch a r t 1
F igu r e 1. Molecular structure of [Pd{κ2-P,C-µ2-O-P(O)-
(C6H2-2,4-tBu2)(C6H3-2,4-tBu2)(DMAc)}], 11.
Ta ble 1. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for [P d {K2-P ,C-µ2-O-
P (O)(C6H2-2,4-tBu 2)(C6H3-2,4-tBu 2)(DMAc)}], 11
Pd(1)-O(1)
Pd(1)-P(1)
P(1)-O(1)
P(1)-O(3)
O(3)-C(7)
2.1349(18)
2.1638(7)
1.5065(18)
1.6216(19)
1.403(3)
Pd(1)-C(2)
Pd(1)-O(4)
P(1)-O(2)
O(2)-C(1)
C(1)-C(2)
1.997(3)
2.1549(19)
1.6198(18)
1.406(3)
1.393(4)
P(1)-Pd(1)-C(2)
O(4)-Pd(1)-C(2)
80.06(8)
95.69(9)
P(1)-Pd(1)-O(1)
O(1)-Pd(1)-O(4)
99.95(5)
85.18(7)
P(1)-Pd(1)-O(4) 173.55(6)
Pd(1)-P(1)-O(1) 123.88(8)
Pd(1)-P(1)-O(3) 104.74(7)
O(1)-Pd(1)-C(2) 179.12(9)
Pd(1)-P(1)-O(2) 108.32(7)
Pd(1)-O(1)-P(1) 118.4(1)
112.45(15) P(1)-O(3)-C(7) 124.57(16)
P(1)-O(2)-C(1)
O(2)-C(1)-C(2)
117.0(2)
Pd(1)-C(2)-C(1) 120.63(19)
PR2(OH) ligands (R ) alkyl, aryl, aryloxide). We now
present evidence that these complexes can indeed
engage in hydrolytic and solvolytic chemistry and ad-
dress whether such processes have a role in the observed
high activity of phosphite- and phosphinite-based cata-
lysts in Suzuki coupling reactions.
Tricyclohexylphosphine adducts of both imine- and
amine-based palladacycles, complexes 8 and 9, show
very good activity when aryl chlorides are used as
substrates.8 Very recently we have found that similar
adducts formed between complexes of the types 3 and
10 with tricyclohexylphosphine show among the highest
activity yet reported in aryl chloride coupling reactions.9
Recently Li demonstrated that palladium-dialkyl-
hydroxyphosphine complexes, Pd-PR2(OH), formed on
reaction of palladium precursors with secondary phos-
phine oxides via a tautomerization of the starting
ligand, show good activity in a range of coupling
reactions of aryl chlorides, including the Suzuki reac-
tion.10 Once formed the Pd-PR2(OH) complexes readily
undergo base-promoted deprotonation of the hydroxyl
group to give anionic species that are proposed to act
as the true catalysts. It occurred to us that palladium
phosphite and phosphinite complexes such as 3 may
well undergo hydrolytic processes under catalytic condi-
tions, which may in turn lead to the in situ generation
of complexes related to those reported by Li with
Resu lts a n d Discu ssion
We were initially alerted to the potential of pallada-
cylic complexes of the type 3 to undergo hydrolytic
reactions during studies on the activation of the catalyst
3a . An NMR scale reaction of the palladacycle (10 mg)
with 1.5 equiv of phenylboronic acid and 2 equiv of
potassium carbonate in dimethylacetamide was heated
1
at 100 °C for 5 h. The H, 13C, and 31P NMR spectra of
the reaction mixture were all broad and proved not to
be useful in the characterization of the products.
However, when the sample was left to stand at room
temperature for several days a small amount of crystals
was obtained and characterized by X-ray crystallogra-
phy. The compound proved to be a new dimeric pal-
ladium complex, [Pd{κ2-P,C-µ2-O-P(O)(C6H2-2,4-tBu2)-
(C6H3-2,4-tBu2)(DMAc)}], 11, the molecular structure of
which is shown in Figure 1, while selected data are
given in Table 1. Complex 11 is, we believe, a unique
example of an orthometalated diarylphosphito complex.
While the orthometalation observed in the starting
complex 3a is maintained in the complex 11, one of the
nonorthometalated aryloxide residues in the starting
material has been lost by hydrolysis, possibly with
adventitious water arising from the boronic acid, present
as a hydrate, in the presence of the base. The six-
(8) Bedford, R. B.; Cazin, C. S. J . Chem. Commun. 2001, 1540.
(9) (a) Bedford, R. B.; Hazelwood, S. L.; Cazin, C. S. J . Angew. Chem.,
Int. Ed. 2002, in press. (b) Bedford, R. B.; Cazin, C. S. J .; Hazelwood,
S. L. Chem. Commun. 2002, in press. (c) Bedford, R. B.; Hazelwood,
S. L.; Limmert, M. E. Chem. Commun. 2002, in press.
(10) (a) Li, G. Y. J . Org. Chem. 2002, 67, 3643. (b) Li, G. Y.; Zheng,
G.; Noonan, A. F. J . Org. Chem. 2001, 66, 8677. (c) Li, G. Y. Angew.
Chem., Int. Ed. 2001, 40, 1513.

1366 Organometallics, Vol. 22, No. 7, 2003
Bedford et al.
Sch em e 2a
a
Conditions: (i) PdCl2, 2-MeOC2H4OH, ∆, 18 h. (ii) Recrys-
tallize, CH2Cl2/EtOH, 1-2 weeks. (iii) 0.5 equiv of PdCl2,
EtOH, ∆, 18 h.
F igu r e 2. Molecular structure of [PdCl2{PiPr2(OEt)}2], 14.
membered ring containing the two palladium atoms
adopts a chair conformation. The P-O bond length in
this ring is significantly shorter than those of both the
orthometalated and nonorthometalated aryloxide resi-
dues which are essentially the same length. Most of the
bond lengths of the five-membered metallacycle are the
same as those of complex 3a ,3c except the P-O bond
length which is slightly longer and the O-C bond length
which is slightly shorter. All of the angles in the five-
membered ring are comparable with those of complex
3a except the Pd-C2-C1 angle, which is slightly more
acute. The asymmetric unit contains a molecule of
benzene, presumably derived from the phenylboronic
acid.11
When the reaction was repeated in DMF, again the
the 1H, 13C, and 31P NMR spectra proved to be un-
informative. Leaving the solution to stand resulted in
the formation of a small amount of crystals. This second
compound was characterized by X-ray crystallography
and was shown to be the known biphenol 3,3′,5,5′-tetra-
tert-butyl-2,2′-biphenol, 12.12 While it is evident that the
biphenol 12 is formed by the oxidative coupling of two
2,4-di-tert-butylphenoxy residues,13 it is not clear at this
stage whether the residues are coupled as free phenols
lost during the hydrolytic formation of 11 from 3a or
whether the coupling occurs between two aryloxide
residues still incorporated in phosphite ligands. The
former process could arise from a metal-promoted
pathway, the latter by internal reorganization and
reductive elimination.
Ta ble 2. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for [P d Cl2{P iP r 2(OEt)}2], 14
Pd(1)-P(1)
P(1)-O(1)
P(1)-C(4)
2.3265(10)
1.6156(14)
1.8332(19)
Pd(1)-Cl(1)
P(1)-C(1)
O(1)-C(7)
2.2975(7)
1.830(2)
1.450(2)
Cl(1)-Pd(1)-P(1)
C(4)-P(1)-Pd(1)
O(1)-P(1)-C(1)
C(1)-P(1)-C(4)
88.24(2)
116.50(6)
97.60(8)
C(1)-P(1)-Pd(1)
O(1)-P(1)-Pd(1)
O(1)-P(1)-C(4)
110.23(7)
117.64(5)
105.68(8)
107.18(9)
can be seen the phosphinite ligands adopt a trans-
configuration presumably as a consequence of their high
steric profile, since the cis-configuration in which the
π-donor chlorides would be trans to the π-acidic phos-
phinite ligands would be preferred electronically. Com-
plex 14 is more conveniently prepared in 73% yield by
heating 2 equiv of ligand 13 with palladium dichloride
in ethanol at reflux temperature (Scheme 2).
Interestingly the “transesterification” reaction of the
aryldialkylphosphinite 13 is not seen with palladium
complexes of triarylphosphinites, PAr2(OAr), or tri-
arylphosphites, P(OAr)3. We have previously found that
such ligands readily undergo orthopalladation reac-
tions when heated in 2-methoxyethanol at reflux tem-
perature and that these and their nonorthometalated
palladium(II) counterparts can be recrystallized from
alcohols.3b,c Both of these observations indicate that the
coordinated ligands are stable with respect to trans-
esterification.
Evidently solvolytic processes can occur with both
phosphite and phosphinite complexes of palladium. To
assess whether in situ hydrolysis of such systems could
help account for the high activity observed when com-
plexes of the type 3 are used in Suzuki coupling
reactions with aryl bromide substrates,3b,c we decided
to compare the activity of palladium catalysts with
representative phosphite and phosphinite ligands with
those containing comparable “hydrolyzed” PR2(OH)
ligands. The preformed catalysts synthesized were
complex 16, formed by reaction of the commercially
available compound 17 with dimer 15, which has
previously been found to act as an excellent catalyst
precursor8 (Scheme 3); the complexes 18a ,b, formed by
reaction of the appropriate triarylphosphinite ligands
with palladium [PdCl2(NCMe)2]; and complex 19, formed
by reaction of diphenylphosphine oxide with [PdCl2-
(NCMe)2].
Evidence that phosphinite complexes are also able to
undergo solvolytic processes was obtained when we
attempted the synthesis of the orthopalladated complex
3d from palladium dichloride and the ligand 13 in
2-methoxyethanol at reflux temperature. Slow recrys-
tallization of the product mixture from dichloromethane/
ethanol did not yield the expected product 3d , but rather
gave the new complex trans-[PdCl2{PiPr2(OEt)}2], 14
(Scheme 2).
The structure of 14 was confirmed unequivocally by
X-ray crystal analysis and the molecule is shown in
Figure 2, while selected data are given in Table 2. As
(11) For recent examples of this process see: Goosse, L. J . Chem.
Commun. 2001, 669 and references therein.
(12) See Supporting Information for the crystal structure of 12.
(13) For recent examples of the production of 12 by oxidative
coupling see: (a) Gupta, R.; Mukherjee, R. Tetrahedron Lett. 2000,
41, 7763. (b) Nishino, H.; Satoh, H.; Yamashita, M.; Kurosawa, K. J .
Chem. Soc., Perkin Trans. 2 1999, 1919. (c) Lockwood, M. A.; Blubaugh,
T. J .; Collier, A. M.; Lovell, S.; Mayer, J . M. Angew. Chem., Int. Ed.
1999, 38, 225 and references therein.
The 31P NMR spectrum of 16 shows a singlet at δ 95.2
ppm, whereas that of the starting material 17 shows a

Palladium Complex Catalysts in the Suzuki Reaction
Organometallics, Vol. 22, No. 7, 2003 1367
Sch em e 3. Syn th esis of P r efor m ed Ca ta lystsa
Sch em e 4. Syn th esis a n d Or th om eta la tion of a
Bu lk y Tr ia r ylp h osp h itea
a
Conditions: (i) CH2Cl2, rt, 1 h. (ii) 0.5 equiv of [PdCl2-
(NCMe)2], CH2Cl2, rt, 1 h.
a
Conditions: (i) NEt3, toluene, ∆, 18 h. (ii) PdCl2, toluene,
∆, 18 h.
1
doublet with a large (592 Hz) J (PH) coupling at δ 95
1
ppm. The H NMR of complex 16 shows a broad singlet
two Pd-Cl stretches at 312 and 345 cm-1, indicating a
cis geometry about the palladium.
at 5.04 ppm for the O-H proton whereas that of the
ligand 17 shows a doublet at 8.07 ppm for the P-H,
with a phosphorus coupling of 592 Hz. This indicates
that it is the P-OH tautomer that coordinates to the
palladium center. The resonances associated with the
orthometalated N,N-dimethylbenzylamine are closely
similar to those found for other phosphine adducts,14,8
supporting the formulation of 16 as a simple phosphine
adduct. The presence of the trifluoroacetate ligand was
confirmed by IR spectroscopy, which showed a peak at
1545 cm-1 corresponding to the CdO stretch.
The phosphite ligand 20 was prepared by heating the
dichlorophosphite 21 with the diol 22 in toluene in the
presence of triethylamine (Scheme 4). Ligand 20 was
then used to generate the palladacyclic catalyst 3e by
heating it with palladium dichloride in toluene.
The application of complexes 3e, 16, 18a ,b, and 19
and related complexes formed in situ as catalysts in the
Suzuki coupling reaction was then investigated. In all
cases the initial coupling studied was that of phenyl-
boronic acid with 4-bromoanisole as this is an electroni-
cally challenging bromide, thus results from this reac-
tion are a useful indicator of catalyst performance. The
results of this study are summarized in Table 3.
As can be seen, the preformed catalyst 16 showed
slightly lower activity than catalysts formed in situ from
complex 15 and either 1 or 2 equiv of the ligand 17
(entries 1-4). This demonstrates that complexes with
P-OH tautomers of comparatively π-acidic ligands can
give good activity in coupling reactions. Next we com-
pared the performance of the catalysts formed in situ
from complex 15 and 2 equiv per palladium of the
The 31P NMR spectra of the complexes 18a ,b show
singlets at δ 101.1 and 102.8 ppm, respectively, ca. 52-
55 ppm upfield of their analogous orthopalladated
complexes 3b,c and consistent with nonorthometalated
phosphinite ligands on a palladium(II) center.3c The H
1
NMR spectra indicated that all the ortho-protons are
indeed present. The IR spectra of complex 18a shows
two Pd-Cl stretches at 305 and 335 cm-1, indicative of
a cis disposition about the palladium center, whereas a
single Pd-Cl stretch at 370 cm-1 is seen for 18b
indicating a trans geometry. The IR data are very
similar to those reported previously for analogous
triarylphosphite complexes.15 The cis geometry of 16a
is preferred electronically as this renders the π-acidic
phosphinite ligands trans to the π-basic chlorides; the
bulk of the phosphinite ligands in 18b overrides this
electronic preference.
The 31P spectrum of complex 19 shows a singlet at δ
79.6 ppm compared with a doublet at δ 45.0 for
diphenylphosphine oxide, indicating that the ligand has
coordinated as the hydroxyphosphine tautomer. The H
NMR spectrum shows a broad singlet at δ 5.30 corre-
phosphinite ligands PR2(OC6H3-2,4-tBu2) (R ) Ph, Pr)
i
with the analogous systems containing the ligands
PR2(OH) (entries 5-8). As can be seen, the hydroxy-
phosphine systems show good activity, but only about
half that of the analogous phosphinites. This demon-
strates that in this case hydrolysis cannot account for
all the activity seen with the phosphinite ligands.
However, when a similar comparison is made between
the triarylphosphite 20 and its hydrolyzed analogue 23
(entries 9 and 10), it can be seen that there is essentially
no difference in performance. Therefore it seems, at first
sight at least, as if hydrolysis may be playing a role in
this case. As well as being dependent on the ligand, the
extent to which hydrolysis may play a role also seems
to be dependent on the palladium source. Comparing
1
sponding to the hydroxyl proton. The IR spectrum shows
(14) Bedford, R. B.; Cazin, C. S. J . Unpublished data.
(15) Ahmad, N.; Ainscough, E. W.; J ames, T. A.; Robinson, S. D. J .
Chem. Soc., Dalton Trans. 1973, 1148.

1368 Organometallics, Vol. 22, No. 7, 2003
Ta ble 3. Th e Su zu k i Cou p lin g of 4-Br om oa n isole w ith P h en ylbor on ic Acid a
Bedford et al.
a
Reaction conditions: phenylboronic acid (15 mmol), 4-bromoanisole (10 mmol), K2CO3 (20 mmol), toluene (30 mL), 110 °C, 18 h.
Based on conversion of 4-bromoanisole to 4-methoxybiphenyl, determined by GC (hexadecane standard). c Ar ) C6H3-2,4-tBu2.
b
the performance of the catalysts formed in situ from
palladium bis(dibenzylideneacetone) and either the
It is not possible to perform analogous studies with the
PiPr2(OAr) and PiPr2(OH) ligands at this stage as we
were unable to cleanly synthesize the appropriate
palladium dichloride adducts of the latter ligands.
To establish whether the observed similarity in
performance of some of the catalysts and their hydro-
lyzed counterparts is indeed significant, rather than just
a coincidence for this particular coupling, we investi-
gated their performance in the coupling of three further
bromides, namely 4-bromotoluene, 2-bromotoluene, and
2-bromo-m-xylene (Table 4).
i
phosphinite ligands PR2(OC6H3-2,4-tBu2) (R ) Ph, Pr)
or their hydrolyzed counterparts, it can be seen that
hydrolysis cannot account fully for the activity observed
with the phosphinites. It is also interesting to note that
here palladium bis(dibenzylideneacetone) appears to be
a better palladium source than complex 15, in contrast
to the findings obtained in the Suzuki coupling of aryl
chlorides catalyzed by analogous tricyclohexylphosphine
complexes.8,16
By contrast when the preformed aryldiphenylphos-
phinite complexes 18a ,b are compared with the hy-
droxydiphenylphosphine complex 19 (entries 15-17)
very similar activity results, again indicating that, at
least with triarylphosphinites, hydrolysis is important.
Comparing the activities of the preformed phosphinite
complex 18a with the analogous hydroxyphosphine
complex 19 in the coupling of all three aryl bromide
substrates (entries 1-6) it can be seen that, as in the
case with 4-bromoansiole, they are essentially identical.
The performances of the catalysts formed in situ from
complex 15 and either ligand 20 or ligand 23 are again
(16) Andreu, M. G.; Zapf, A.; Beller, M. Chem. Commun. 2000, 2475.

Palladium Complex Catalysts in the Suzuki Reaction
Organometallics, Vol. 22, No. 7, 2003 1369
Ta ble 4. Su zu k i Cou p lin g of Ar yl Br om id es w ith P h en ylbor on ic Acid a
a
Reaction conditions: phenylboronic acid (15 mmol), 4-bromoanisole (10 mmol), K2CO3 (20 mmol), toluene (30 mL), 110 °C, 18 h.
Based on conversion of aryl bromide to Suzuki coupled biphenyl, determined by GC (hexadecane standard). c Ar ) C6H3-2,4-tBu2.
b

1370 Organometallics, Vol. 22, No. 7, 2003
Bedford et al.
3
3
3J PH ) 15.2 Hz, CH(CH3)2), 1.19 (dd, 6H, J HH ) 7.1 Hz, J PH
essentially identical except in the coupling of 2-bromo-
toluene where the hydrolyzed system shows slightly
lower activity. Therefore it may be concluded that for
both preformed catalyst systems of the type [PdCl2-
{PPh2(OAr)}2] or catalysts formed in situ from complex
15 and the triarylphosphite 20, hydrolysis occurs during
the activation of the pre-catalysts and that the true
active catalysts contain hydrolyzed forms of the ligands.
It is difficult to extrapolate these data to all systems
with triarylphosphinite and triarylphosphite ligands
since both catalytic activity and the influence of hy-
drolysis are not only dependent on the ligand type but
also on the nature of the palladium precursor. In all
cases the activity observed is substantially lower than
when palladacycles with orthometalated triarylphos-
phite or phosphinites ligands are used under the same
conditions. For instance, the palladacycle formed from
ligand 20, complex 3e, shows a maximum TON of
430 000 in the coupling of phenylboronic acid with
4-bromoanisole (Table 4, entry 13), while complex 3c
shows TONs of up to 2.6 million in the same reaction.3c
Therefore it is not possible at this stage to determine
what extent the hydrolysis of the ligands has on the
performance of these very high activity catalysts. Re-
gardless, the data obtained here certainly point to the
fact that hydrolysis does play a role. From this study it
is apparent that potential hydrolytic processes should
be taken into account during both the rational design
of new precatalysts and studies into their in situ
activation.
) 11.0 Hz, CH(CH3)2), 1.30 (s, 9H, tBu), 1.41 (s, 9H, tBu), 2.04
2
3
(apparent d of heptets, 2H, J PH ) 2.3 Hz, J HH ) 7.2 Hz,
3
4
CH(CH3)2), 7.11 (dd, 1H, J HH ) 8.5 Hz, J HH ) 2.6 Hz, H5),
4
3
7.29 (d, 1H, J HH ) 2.6 Hz, H3), 7.50 (dd, 1H, J HH ) 8.5 Hz,
4J HP ) 6.2 Hz, H6). 31P NMR (121.5 MHz, CDCl3): δ 138.4 (s).
Anal. Calcd for C20H35OP: C, 74.49; H, 10.94. Found: C, 74.0;
H, 10.6.
P r ep a r a tion of 2,4-Di-ter t-bu tylp h en yl Dip h en ylp h os-
p h in ite, P P h 2(OC6H3-2,4-tBu 2). A mixture of predried (tolu-
ene azeotrope) 2,4-di-tert-butylphenol (6.89 g, 33.4 mmol),
chlorodiphenylphosphine (6.0 mL, 33.4 mmol), and triethyl-
amine (7.0 mL, 50.0 mmol) in toluene (50 mL) was heated at
reflux temperature for 17 h. The mixture was filtered through
Celite to remove [Et3NH]Cl, the precipitate was washed with
toluene (2 × 10 mL), and the volatiles were removed from the
combined organic fractions under reduced pressure to yield
the title compound as a white solid that was not purified
1
further. Yield: 12.78 g (98%). H NMR (300 MHz, CDCl3): δ
t
t
3
1.32 (s, 9H, Bu), 1.39 (s, 9H, Bu), 7.05 (dd, 1H, J HH ) 8.5
4
3
4
Hz, J HP ) 3.0 Hz, H6), 7.12 (dd, 1H, J HH ) 8.5 Hz, J HH
)
2.5 Hz, H5), 7.36 (d, 1H, 4J HH ) 2.5 Hz, H3), 7.40 (m, 6H, Ph),
7.63 (m, 4H, Ph). 31P NMR (121.5 MHz, CDCl3): δ 108.5 (s).
Anal. Calcd for C26H31OP: C, 80.0; H, 8.0. Found: C, 80.1; H,
7.95.
P r ep a r a tion of P (OC6H3-2,4-tBu 2){(OC6H2-2-tBu -4-Me-
6-)2-CH2}, 20. A mixture of 2,2′-methylenebis(6-tert-butyl-4-
methyl)phenol (1.00 g, 2.94 mmol), dichloro-2,4-di-tert-butyl-
phenol phosphite (1.00 g, 3.23 mmol), and triethylamine (1.0
mL, 7.2 mmol) in toluene (100 mL) was heated at reflux
temperature for 18 h. The suspension was allowed to cool to
room temperature and then filtered through Celite. The clear
solution is evaporated to dryness in vacuo, yielding the crude
product as a gum that is washed repeatedly with n-pentane
to give a colorless solid. Yield: 1.39 g (82%). 1H NMR (300
Exp er im en ta l Section
t
t
MHz, CDCl3): δ 1.32 (s, 18H, Bu), 1.53 (s, 9H, Bu), 1.34 (s,
9H, tBu), 2.32 (s, 6H, CH3), 3.48 (d, 1H, 2J HH ) 12.8 Hz, CH2),
Gen er a l Meth od s. All reactions were carried out under
nitrogen following standard Schlenk techniques. Solvents were
dried and freshly distilled prior to use. All other chemicals
were used as received. Complex 15, compound 23, and
PPh2(OPh) were prepared according to literature methods.8,17,18
GC analyses were performed on a Varian 3800 GC fitted with
a 25 m CP Sil 5CB column and data were recorded on a Star
workstation.
2
5
4.50 (dd, 1H, J HH ) 12.8 Hz, J HP ) 2.8 Hz, CH2), 7.05 (d,
2H, 4J HH ) 2.2 Hz, H3′), 7.15 (d, 2H, 4J HH ) 2.2 Hz, H5′), 7.16
(dd, 1H, J HH ) 8.5 Hz, J HH ) 2.5 Hz, H5), 7.42 (d, 1H, J HH
3
4
4
3
4
) 2.5 Hz, H3), 7.59 (dd, 1H, J HH ) 8.5 Hz, J HP ) 2.7 Hz,
H6). 31P{1H} NMR (CDCl3, 121.5 MHz): δ 133 (s). Anal. Calcd
for C37H51O3P: C, 77.32; H, 8.94. Found: C, 76.8; H, 9.4.
P r ep a r a t ion of [{P d (µ-Cl){K2-P ,C-P (OC6H2-2,4-t Bu 2)-
{(OC6H2-2-tBu -4-Me-6-)2-CH2}}}2], 3e. A mixture of PdCl2
(0.150 g, 0.85 mmol) and the ligand 20 (0.500 g, 0.91 mmol)
in toluene (15 mL) was heated at reflux temperature for 18 h.
The solution was allowed to cool to room temperature and then
the solvent was removed in vacuo. The crude product was
dissolved in dichloromethane (25 mL), filtered through Celite,
and concentrated to ca. 5 mL. Addition of methanol (15 mL)
gave a bright yellow precipitate of the product, which was
collected by filtration and dried in vacuo. Yield: 0.325 g (53%).
1H NMR (CDCl3, 300 MHz): δ 0.91, 1.10, 1.14, 1.20 (s, br, 36H,
tBu,), 2.14 (s, 6 H, CH3,), 3.65 (m, br, 1H, CH2), 4.47 (m, br,
1H, CH2), 6.91, 6.95, 7.18, 7.35 (m, 6H, br). 31P{1H} NMR
(CDCl3, 121 MHz): δ 119.0 (s, br, major isomer), 117.0 (s,
minor isomer). Anal. Calcd for C31H39O3P: C, 62.10, H, 7.04.
Found: C, 61.9; H, 7.1.
Gen er a l Meth od for th e P r ep a r a tion of Ar yl Diiso-
p r op ylp h osp h in ite Liga n d s. A mixture of the appropriate
predried (toluene azeotrope) phenol (31.4 mmol), chlorodiiso-
propylphosphine (5.0 mL, 31.4 mmol), and triethylamine (5.0
mL, 35.9 mol) in toluene (80 mL) was heated at reflux
temperature for 17 h. Petroleum ether 60-80 (50 mL) was
added to the cooled reaction mixture, which was then filtered
through Celite to remove [Et3NH]Cl. The precipitate was
washed with petrol (3 × 10 mL) and the solvents removed from
the combined organic fractions under reduced pressure to yield
the diisopropylphosphinite ligands as pale yellow oils which
were not purified further.
1
P iP r 2(OC6H4-4-Et), 13. Yellow oil. Yield: 7.26 g (97%). H
3
3
NMR (300 MHz, CDCl3): δ 1.32 (dd, 6H, J HH ) 7.2 Hz, J PH
3
) 15.8 Hz, CH(CH3)2), 1.40 (t, 3H, J HH ) 7.0 Hz CH2CH3),
P r ep a r a tion of tr a n s-[P d Cl2{P iP r 2(OEt)}2], 14. A mix-
ture of PdCl2 (0.500 g, 2.82 mmol) and ligand 13 (1.344 g, 5.64
mmol) in ethanol (50 mL) was heated at reflux temperature
overnight and the resulting orange solution was cooled and a
yellow-orange solid precipitated. The crude solid was recrys-
tallized from CH2Cl2/pentane to give the title complex as a
yellow solid. Yield: 1.347 g (73%). 1H NMR (300 MHz,
3
3
1.42 (dd, 6H, J HH ) 7.2 Hz, J PH ) 11.0 Hz, CH(CH3)2), 2.18
2
3
(apparent d of heptets, 2H, J PH ) 2.34 Hz, J HH ) 7.2 Hz,
3
CH(CH3)2), 2.75 (q, 2H, J HH ) 7.0 Hz CH2CH3), 7.18 (d, 2H,
3J HH ) 7.9 Hz), 7.22 (d, 2H, 3J HH ) 8.0 Hz, aromatic). 31P NMR
(121.5 MHz, CDCl3): δ 149.4 (s). Anal. Calcd for C14H23OP:
C, 70.56; H, 9.73. Found: C, 70.15; H, 9.2.
P iP r 2(OC6H3-2,4-tBu 2). Pale yellow oil. Yield: 9.72 g (96%).
3
1H NMR (300 MHz, CDCl3): δ 1.13 (dd, 6H, J HH ) 7.0 Hz,
3
CDCl3): δ 1.30 (t, 6H, br, J HH ≈ 7 Hz, CH2CH3), 1.45 (dd,
3
3
6H, J HH ) 7.4 Hz, J PH ) 10.7 Hz, CH(CH3)2), 1.51 (dd, 6H,
3J HH ) 7.4 Hz, 3J PH ) 10.5 Hz, CH(CH3)2), 2.50 (apparent d of
(17) Kumaraswamy, S.; Kumar, K. S.; Raja, S.; Swamy, K. C. K.
Tetrahedron 2001, 57, 8181.
(18) Sander, M. Chem. Ber. 1960, 93, 1220.
3
2
heptets, 4H, J HH ) 7.1 Hz, J PH ≈ 2 Hz, CH(CH3)2), 4.20 (q,
4H, J HH ≈ 7 Hz, CH2CH3). 31P NMR (121.5 MHz, CDCl3): δ
3

Palladium Complex Catalysts in the Suzuki Reaction
Organometallics, Vol. 22, No. 7, 2003 1371
140.9. IR (KBr) ν(Pd-Cl): 315, 347 cm-1. Anal. Calcd for
18H, tBu), 1.39 (s, 18H, tBu), 7.30 (dd, 2H, 3J HH ) 7.5 Hz, 4J HH
) 2.5 Hz, H5 OAr), 7.36 (d, 2H, 2.5 Hz, H3 OAr), 7.41 (m,
C
16H38Cl2O2P2Pd: C, 51.43; H, 7.09. Found: C, 51.0; H, 7.4.
3
3
1H NMR (300 MHz, CDCl3): δ 1.30 (t, 6H, br, J HH ≈ 7 Hz,
12H, Ph), 7.78 (m, 8H, Ph), 7.90 (d, 2H, J HH ) 7.5 Hz, H6
3
3
CH2CH3), 1.45 (dd, 6H, J HH ) 7.4 Hz, J PH ) 10.7 Hz,
OAr). 31P NMR (121.5 MHz, CDCl3): δ 102.8 (s). IR (KBr)
ν(Pd-Cl): 370 cm-1. Anal. Calcd for C52H62Cl2O2P2Pd: C,
65.17; H, 6.52. Found: C, 64.7; H, 6.4.
3
2
CH(CH3)2), 1.51 (dd, 6H, J HH ) 7.4 Hz, J PH ) 10.5 Hz,
CH(CH3)2), 2.50 (apparent d of heptets, 4H, 3J HH ≈ 2 Hz, 2J PH
) 7.1 Hz, CH(CH3)2), 4.20 (q, 4H, J HH ≈ 7 Hz, CH2CH3). 31P
cis-[P d Cl2{P P h 2(OH)}2], 19. Yellow solid. Yield: 0.78 g
3
1
NMR (121.5 MHz, CDCl3): δ 140.9.
(69.8%). H NMR (300 MHz, CDCl3): δ 5.30 (s, br, 2H, POH),
P r epar ation of [P d(K2-N,C-NMe2CH2C6H4)(TFA){P (OH)-
(cyclo-OC6H4-2-C6H4)}], 16. A solution of [{Pd(µ-TFA)(κ2-N,C-
C6H4CH2NMe2)}2] (1.00 g, 1.41 mmol) and 6H-dibenz[c,e][1,2]-
oxaphosphorin-6-oxide (0.61 g, 2.83 mmol) in CH2Cl2 (10 mL)
was stirred at room temperature for 1 h during which time a
precipitate formed. The resulting white solid was collected by
filtration and recrystallized from CH2Cl2/MeOH to yield the
title complex as a white solid. Yield: 0.51 g (63%). 1H NMR
(300 MHz, CDCl3): δ 2.21 (s, 3H, CH3), 2.35 (s, 3H, CH3), 3.28
(d, 1H, 2J HH ) 12 Hz, CH2N), 3.32 (d, 1H, 2J HH ) 12 Hz, CH2N),
5.04 (1H, s, br, OH), 7.19 (m, br, 4H, orthometalated ring),
7.86 (complex overlapping multiplets, 8H, aromatic). 31P NMR
(121.5 MHz, CDCl3): δ 95.2 (s). IR (KBr) ν(CdO): 1545 cm-1
Anal. Calcd for C23H21F3NO4PPd: C, 48.48; H, 3.71; N, 2.46.
Found: C, 48.75; H, 3.3; N, 2.6.
7.23 (m, 8H, H2, Ph), 7.38 (m, 4H, J HH ) 6.0 Hz, H4), 7.56
3
(m, 8H, Ph). 31P NMR (121.5 MHz, CDCl3): δ 79.6 (s). IR (KBr)
ν(Pd-Cl): 312, 345 cm-1. Anal. Calcd for C24H22Cl2O2P2Pd: C,
49.55; H, 3.81. Found: C, 49.85; H, 3.64.
Ca ta lysis. In a three-necked flask under an atmosphere of
nitrogen were placed the appropriate aryl bromide (10.0
mmol), phenyl boronic acid (1.83 g, 15.0 mmol), K2CO3 (2.76
g, 20.0 mmol), and toluene (30 mL total, including catalys
solution). The correct amount of catalyst was added as a
toluene solution made up by multiple volumetric dilutions of
stock solutions and the mixture was heated at reflux temper-
ature for 18 h. The mixture was cooled in an ice bath, quenched
with aqueous HCl (2 M, 100 mL), extracted with dichloro-
methane (3 × 100 mL), dried over MgSO4, and evaporated to
dryness. Hexadecane (0.068 M in CH2Cl2, 3.0 mL) and di-
chloromethane (5-7 mL, to ensure complete dissolution) were
added. The conversion to coupled product was then determined
by GC analysis.
Gen er a l Meth od for th e P r ep a r a tion of th e Com p lexes
[P d Cl2(L)2]. A solution of [PdCl2(NCMe)2] (0.50 g, 1.93 mmol)
and appropriate ligand (3.85 mmol) in CH2Cl2 (20 mL) was
stirred for 1 h at room temperature. Hexane (15 mL) was
added to precipitate the product, which was then recrystallized
from CH2Cl2/hexane and dried in vacuo.
Ack n ow led gm en t. We thank the EPSRC for fund-
ing (studentship for S.L.H. and PDRAs for M.E.L. and
S.R.) and J ohnson Matthey Chemicals for funding and
the loan of palladium salts.
cis-[P d Cl2{P P h 2(OP h )}2], 18a . Pale orange solid. Yield:
1
0.62 g (44%). H NMR (300 MHz, CDCl3): δ 7.07 (m, 2H, H4
of OPh), 7.20 (m, 4H, OPh), 7.28 (m, 4H, OPh), 7.46 (m, 12H,
PPh), 7.67 (m, 8H, PPh). 31P NMR (121.5 MHz, CDCl3): δ 101.1
Su p p or tin g In for m a tion Ava ila ble: Complete crystal
structure data for compounds 11, 12, and 14. This material is
available free of charge via the Internet at http://pubs.acs.org.
(s). IR (KBr) ν(Pd-Cl): 305, 335 cm-1. Anal. Calcd for C36H30
-
Cl2O2P2Pd: C, 58.92; H, 4.12. Found: C, 59.6; H, 4.6.
tr a n s-[P d Cl2{P P h 2(OC6H3-2,4-tBu 2)}2], 18b. Yellow solid.
Yield: 0.44 g (88%). 1H NMR (300 MHz, CDCl3): δ 1.36 (s,
OM020941F