Ionophilic phosphonium-appended carbopalladacycle catalyst for Suzuki-Miyaura and Heck cross-coupling catalysis

Article abstract of DOI:10.1139/V09-149

An ionic liquid, covalently tethered to an efficient transition-metal catalyst in the presence of an ionic liquid reaction medium, can utilize ionophilic interactions to improve catalyst activity, recyclability, and product isolation while decreasing catalyst leaching. Given the greater stability of phosphonium salts in comparison to imidazolium ionic liquids under basic conditions, phosphonium-tagged oxime carbopalladacycle salts were prepared and employed in both Heck and Suzuki-Miyaura reactions. The desired product was obtained in good yields for up to four catalyst cycles in the case of the Suzuki-Miyaura reaction. While taking advantage of the non-volatile nature of ionic liquids, the product was isolated through simple sublimation from the reaction mixture, eliminating issues associated with catalyst leaching, and the remaining ionic liquid solvent-catalyst mixture was ready for further catalysis.

Full text of DOI:10.1139/V09-149

27
Ionophilic phosphonium-appended
carbopalladacycle catalyst for Suzuki–Miyaura
and Heck cross-coupling catalysis
Jocelyn J. Tindale and Paul J. Ragogna
Abstract: An ionic liquid, covalently tethered to an efficient transition-metal catalyst in the presence of an ionic liquid re-
action medium, can utilize ionophilic interactions to improve catalyst activity, recyclability, and product isolation while de-
creasing catalyst leaching. Given the greater stability of phosphonium salts in comparison to imidazolium ionic liquids
under basic conditions, phosphonium-tagged oxime carbopalladacycle salts were prepared and employed in both Heck and
Suzuki–Miyaura reactions. The desired product was obtained in good yields for up to four catalyst cycles in the case of
the Suzuki–Miyaura reaction. While taking advantage of the non-volatile nature of ionic liquids, the product was isolated
through simple sublimation from the reaction mixture, eliminating issues associated with catalyst leaching, and the remain-
ing ionic liquid solvent–catalyst mixture was ready for further catalysis.
Key words: phosphonium, ionic liquids, ionophilic catalysis, carbopalladacycle, ionic tag.
´
´
´
`
`
´
´
Resume : Un liquide ionique lie d’une fac¸on covalente a un catalyseur efficace a base d’un metal de transition, en pre-
´
´
´
´
sence d’un milieu reactionnel forme d’un liquide ionique, peut utiliser les interactions ionophiles pour ameliorer l’activite
´ ` ˆ
´
´
du catalyseur, sa capacite a etre recycle et la facilite d’isoler le produit tout en diminuant la lixiviation du catalyseur. Se
´
`
basant sur la plus grande stabilite des sels de phosphonium par comparaison aux liquides ioniques a base d’imidazolium
´
´
`
dans des conditions basiques, on a prepare des sels a base carbocycles contenant du palladium sels et portant des phospho-
´
´
´
nium d’oximes comme etiquette ionique et on les a utilises dans des reactions de Heck et de Suzuki–Miyaura. Le produit
´
´
´ ´
´
desire a ete obtenu avec de bons rendements pour au moins quatre cycles catalytiques dans le cas de la reaction de Su-
zuki–Miyaura. En tirant avantage de la nature non volatile des liquides ioniques, on a pu isoler le produit par simple subli-
`
´
´
`
´
mation a partir du milieu reactionnel, eliminant ainsi les problemes associes avec la lixiviation du catalyseur et alors que
ˆ ` ˆ
´
´
´
´
le melange solvant liquide ionique/catalyseur est pret a etre reutilise pour d’autres reactions catalytiques.
´
Mots-cles : phosphonium, liquides ioniques, catalyse ionophile, carbocycles contenant du palladium, marquage ionique.
´
[Traduit par la Redaction]
cations is further amplified given the general framework of
common ionic liquids, such as imidazolium or pyridinium
salts, is susceptible to chemical modification to attain a de-
sired function while concomitantly maintaining the intrinsic
properties of the low melting salts.
Introduction
The use of ionic liquids (ILs) as an alternative solvent in
catalysis has become popular over the last ten years as a
means of rendering the reaction ‘‘green’’.1 This notion pre-
dominantly stems from the ionic nature of the salts, which
give rise to a negligible vapor pressure. However, given re-
cent reports on the toxicity of some ILs, their potential ap-
plication as ‘‘green’’ solvents has faltered.2,3 Nevertheless,
many of their properties beyond non-volatility do render
these materials highly amenable to many chemical reactions
and applications. High thermal stabilities, wide liquid tem-
perature ranges, and unique solubility parameters give rise
to ionic liquids as good alternative solvents. Ionic liquids
generally have the ability to dissolve many organic and or-
ganometallic compounds and are often immiscible with
non-polar organic solvents.4 The breadth of potential appli-
The ease of dissolution of transition metals in ionic
liquids and the immiscibility observed with some organic
solvents are both phenomena associated with ionic liquids
that are often exploited in catalysis. These properties enable
an improvement in the isolation of the organic product from
the catalyst during simple extraction procedures.5,6 Unfortu-
nately, in some cases, the catalyst activity quickly decreases
after several runs.7 This is often a result of catalyst leaching
during the extraction of the product from the reaction mix-
ture or in the purification of the ionic liquid before the ini-
tiation of the next reaction cycle. To ensure retention of the
catalyst within the ionic liquid phase and consequently im-
prove the recyclability of the system and the isolation of
products, modifications to the catalyst are required. A cur-
rent methodology includes preparing a catalyst that mimics
the physico-chemical properties of the ionic medium, thus
increasing the ionophilicity of the catalyst. Shreeve et al.
have synthesized a (carbene)palladium(II) catalyst that re-
sembles the skeletal structure of the imidazolium ionic
liquid reaction medium and was used successfully to facili-
Received 21 August 2009. Accepted 5 October 2009. Published
on the NRC Research Press Web site at canjchem.nrc.ca on
17 December 2009.
J.J. Tindale and P.J. Ragogna.1 Department of Chemistry, The
University of Western Ontario, 1151 Richmond St, London,
ON, N6A 5B7, Canada.
1Corresponding author (e-mail: pragogna@uwo.ca).
Can. J. Chem. 88: 27–34 (2010)
doi:10.1139/V09-149
Published by NRC Research Press

28
Can. J. Chem. Vol. 88, 2010
Chart 1. Illustration of the decomposition of an imidazolium cation
in the presence of base resulting in the formation of a carbene, 1,
and alternative imidazolium cations protected at the C-2 position, 2
and 3.
8, were conducted using trihexyltetradecylphosphonium
chloride, [P66614][Cl], as the reaction medium. Increased iso-
lated yields and catalyst recyclability were noted as signifi-
cant improvements to previous reports of ionophilic oxime
carbopalladacycle cross-coupling catalysis.
Experimental section
Materials
The solvents were obtained from Caledon Laboratories.
Dichloromethane, N,N-dimethylformamide, and toluene
were dried using an Innovative Technologies Inc. Solvent
Purification System, which utilizes dual-alumina columns,
and were stored under a nitrogen or argon atmosphere in
˚
Strauss flasks or stored in the glovebox over 4 A molecular
˚
sieves. Methanol was dried by storing over 3 A molecular
sieves. The trihexyltetradecylphosphonium chloride (Strem),
styrene (Aldrich), p-methoxyphenyl boronic acid (Aldrich),
and iodobenzene (Aldrich) were used as received and
stored in a nitrogen-filled MBraun Labmaster 130 glove-
box. The palladium dichloride (Pressure Chemicals), K3PO4
tate C–C bond forming reactions.8,9 A more prevalent strat-
egy to increase a catalyst’s ionophilicity is to covalently in-
tegrate an ionic fragment into the catalyst framework. A
critical review highlighting various successes employing
ionic liquid ligands for the preparation of ionophilic cata-
lysts has recently been published.10,11
¨
(Riedel-deHaen), and LiCl (Caledon) were used as received
and stored in a desiccator. Deuterated NMR solvents
(CDCl3, DMSO) were purchased from Cambridge Isotope
˚
Laboratories and were stored over 4 A molecular sieves.
Carbon–carbon bond forming reactions, such as Heck and
Suzuki–Miyaura, are some of the most important reactions
in organic synthesis. Substantial effort has been put forth to
convert the successful homogeneous methodology into a
heterogeneous system with enhanced catalyst recyclability.
Recently, ionophilic ionic-liquid-tagged catalysts have been
reported as a means of achieving this goal.12,13 The ubiqui-
tous imidazolium cation has been employed as an ion tag;
however, it is susceptible to degradation under basic condi-
tions, a requirement for the Heck and Suzuki–Miyaura reac-
tions. The subsequent formation of a carbene, 1, leads to
poor yields, unwanted side-products, and inactive cata-
lysts.14–16 One approach to avoid this problem is the substi-
tution of a metal-binding ligand such as a pyridine, 2, or
phosphine, 3, at the C-2 position on the imidazolium cation
(Chart 1).17,18
Our methodology, inspired by the work of Corma et al.,19
also evades the susceptibility of imidazolium salts towards
decomposition in the presence of base. Utilizing the iono-
philic ionic-liquid-appended catalyst concept, a phospho-
nium salt was tethered to an oxime carbopalladacycle
moiety, which was previously noted as an exceptional cata-
lyst for Suzuki–Miyaura cross-coupling of aryl halides and
arylboronic acids.20,21 Phosphonium salts do not share the
same vulnerability to basic conditions as observed for imida-
zolium cations and have been successfully used as alterna-
tive solvent media in Suzuki and Heck reactions without
any signs of decomposition products such as Wittig-type re-
agents.22–26 In this context, a phosphonium-tagged oxime
carbopalladacycle catalyst precursor 8 has been synthesized
and comprehensively characterized. It is important to note
here that Singer et al. have prepared similar ionophilic
ligands derived from oximes with pendant imidazolium
salts. The salts were designed as capturing devices for se-
questering metals from aqueous solutions, and their use as
cross-coupling pre-catalysts was not investigated.27 Herein,
cross-coupling reactions catalyzed by the ionophilic catalyst,
Tri-n-butyl phosphine was generously donated by Cytec In-
dustries and was distilled prior to use. Manipulations in-
volving the use of nBu3P and the catalysis experiments
were performed either in the glovebox or by using standard
Schlenk techniques. The precursors 4-hydroxyacetophenone
oxime and 4-(5-bromopentoxy)acetophenone oxime were
synthesized using literature methods.19
Measurements
1
Solution H, 31P{1H} and 13C{1H}, gHMBC, and gHSQC
NMR spectra were recorded on a Varian INOVA 400 MHz,
Varian INVOA 600 MHz, or Mercury 400 MHz spectrome-
ter. All samples for 1H NMR, 13C{1H}, gHMBC, and
gHSQC spectroscopy were referenced to the residual protons
1
in the solvent relative to (CH3)4Si (d (ppm); H: CDCl3 7.26,
DMSO 2.50; 13C{1H}: CDCl3 77.0, DMSO 39.52). Phospho-
rus-31 NMR chemical shifts were reported relative to an ex-
ternal standard (85% H3PO4
d = 0.00 ppm). Mass
spectrometry measurements were recorded in positive and
negative ion modes using an electrospray ionization Micro-
mass LCT spectrometer. Elemental analyses were performed
by Guelph Chemical Laboratories Ltd., Guelph, Ontario,
Canada.
The decomposition temperatures were determined using
thermogravimetric analysis (TGA) on a Q600 SDT TA In-
strument for 7 or on a TGA/SDTA 851e Mettler Toledo
instrument for 8. A 0.005–0.010 g sample was heated in a
40 mL alumina crucible at a rate of 10 8C/min over a tem-
perature range of 100–600 8C. Melting and glass transition
points were determined using differential scanning calorim-
etry (DSC) on a Q20 DSC TA instrument for 7 or on a
DSC 822e Mettler Toledo instrument for 8 using aluminum
sample pans. For 7, a 0.005–0.010 g sample was cooled
to –90 8C and then heated to 200 8C at 10 8C/min, cooled
to –90 at 10 8C/min, and finally reheated to 200 8C at
10 8C/min. The glass-transition temperatures were taken
Published by NRC Research Press

Tindale and Ragogna
29
3
3
from the final heat cycle. For 8, a 0.005–0.010 g sample was
cooled to –70 8C where the temperature was sustained for
15 min, followed by heating to 500 8C at 10 8C/min. All ther-
mal analysis experiments were conducted in a N2(g) atmos-
phere.
(d, JP–C = 14.9 Hz), 23.3 (d, JP–C = 17 Hz), 22.6 (d,
2JP–C = 3.9 Hz), 20.4 (d, JP–C = 3.8 Hz), 17.5 (d, JP–C
=
2
1
1
46.1 Hz), 17.3 (d, JP–C = 47.7 Hz), 13.2, 10.9. FTIR (CsI
pellet; cm–1(ranked intensity)): 227(21), 248(19), 280(26),
304(25), 440(24), 639(9), 722(20), 807(12), 917(17),
959(14), 1043(6), 1100(8), 1209(2), 1227(18), 1266(7),
1308(13), 1339(5), 1378(10), 1459(3), 1560(11), 1582(1),
1629(16), 1648(27), 2872(15), 2934(23), 3446(22). MS
(ESI) m/z+/– (%): 562.1 (55) [monomer], 11242 [dimer].
Anal. calcd. (found): C: 46.79 (48.21), H: 6.92 (7.37).
Synthesis
Preparation of 7
To a solution of 4-(5-bromopentoxy)acetophenone oxime
n
(0.970 g, 3.220 mmol) in DMF (17 mL) Bu3P (0.716 g,
Preparation of 4-methoxy-biphenyl via Suzuki–Miyaura
cross-coupling reaction
3.542 mmol) was added, and the mixture was heated at
70 8C for 48 h under a flow of nitrogen. DMF was removed
in vacuo, and the resulting viscous yellow liquid was dis-
solved in a minimal amount of CH2Cl2, which was extracted
first with hexanes (2 Â 15 mL) and then with THF (1 Â
15 mL). The residual solvent was removed in vacuo yielding
7 as a viscous clear yellow liquid (1.083 g, 2.155 mmol,
A 5 mol% mixture of the catalyst 8 (0.025 g, 0.019 mmol)
in trihexyltetradecylphosphonium chloride (0.700 g,
1.348 mmol) was dissolved in CH2Cl2 (1.5 mL) to facilitate
the transfer of the mixture to a small, narrow Schlenk tube.
The bulk of the solvent was removed in vacuo at room tem-
perature, and the remaining residual solvent was removed at
70 8C. The mixture was cooled to room temperature, and
neat K3PO4 (0.275 g, 1.294 mmol), p-methoxyphenyl bor-
onic acid (0.066 g, 0.431 mmol), and iodobenzene (0.080 g,
0.392 mmol) were added followed by the addition of toluene
(0.1 mL) and distilled and N2-purged H2O (0.2 mL). The re-
action mixture was stirred and heated at 70 8C for 5 h, and
then the volatile solvents were removed in vacuo at room
temperature. A cold finger was fitted to the Schlenk tube
and the product was isolated from the ionic liquid mixture
by sublimation at 80 8C for 14 h, yielding 4-methoxy-bi-
67%). Td
(161.96 MHz, CDCl3)
(599.69 MHz, CDCl3) d (ppm): 7.57 (d, 2H, J = 9.0 Hz),
= 343 8C; Tg =
–2 8C. 31P{1H} NMR
d
(ppm): 30 (s). 1H NMR
3
3
3
7.42 (s, br, 1H), 6.86 (d, 2H, J = 8.4 Hz), 4.00 (t, 2H, J =
6.0 Hz), 2.60–2.55 (m, 2H), 2.48–2.43 (m, 6H), 2.25 (s, 3H),
3
1.87 (quintet, 2H, J = 6.6 Hz), 1.74–1.64 (m, 4H), 1.56–
1.51 (m, 12H), 0.97 (s, 9H). 13C{1H} NMR (100.60 MHz,
CDCl3) d (ppm): 159.2, 153.6, 129.1, 126.8, 113.9, 67.2,
3
3
28.2, 27.1 (d, JP–C = 13 Hz), 23.5 (d, JP–C = 12 Hz), 23.4
2
2
1
(d, JP–C = 3 Hz), 21.3 (d, JP–C = 4.6 Hz), 19.1 (d, JP–C
=
1
47.5 Hz), 18.7 (d, JP–C = 47.0 Hz), 13.2, 11.6. FTIR (drop-
cast on KBr; cm–1(ranked intensity)): 510(20), 561(14),
634(12), 725(11), 838(3), 924(4), 1006(7), 1094(16),
1182(9), 1250(1), 1304(10), 1376(15), 1464(8), 1515(6),
1610(5), 1673(21), 2870(17), 2959(2), 3169(13), 3207(19),
3263(18). MS (ESI) m/z+/– (%): 422.3 (100) [M+ – Br],
923.65 [M2Br+], 582.2 (35) [MBr2].
1
phenyl as a white solid (0.057 g, 0.309 mmol, 99%). H
NMR (400.09 MHz, CDCl3) d (ppm): 7.56 (m, 4H), 7.42 (t,
3
3
3
2H, J = 7.6 Hz), 7.30 (t, 1H, J = 7.6 Hz), 6.98 (d, 2H, J =
8.8 Hz), 3.86 (3, 3H).
Preparation of trans-stilbene via Heck cross-coupling
reaction
Preparation of 8
A
2 mol% mixture of the catalyst 8 (0.015 g,
The Li2PdCl4 was first prepared by the addition of PdCl2
(0.212 g, 1.193 mmol) to a solution of LiCl (0.101 g,
2.387 mmol) in MeOH (0.6 mL), which was then heated at
60 8C for 30 min until all the reagents dissolved. Sodium
acetate (0.049 g, 0.597 mmol) and 7 (0.300 g, 0.597 mmol)
in MeOH (0.65 mL) were then added dropwise to the result-
ing red methanolic Li2PdCl4 solution, and the mixture was
stirred at room temperature for 36 h. Methanol was then
added (2 mL), and the heterogeneous mixture was centri-
fuged and the supernatent was decanted. The solids were
washed with MeOH (4 Â 4 mL) followed by acetone (4 Â
4 mL) to selectively remove the red-brown solid. The ace-
tone fractions were combined and cooled to –30 8C. Upon
precipitation of a beige solid, the mixture was centrifuged,
the supernatent was removed, and the solid was dried in va-
cuo giving 8 as a beige powder (0.144 g, 0.112 mmol, 19%).
Td = 290 8C; Tm = 160 8C. 31P{1H} NMR (161.96 MHz,
0.012 mmol) in trihexyltetradecylphosphonium chloride
(0.300 g, 0.578 mmol) was dissolved in CH2Cl2 (1.5 mL) to
more easily facilitate the transfer of the mixture to a small,
narrow Schlenk tube and to assist in the dispersion and sol-
vation of 8 in the ionic liquid solvent. The bulk of the sol-
vent was removed in vacuo at room temperature, and the
remaining residual solvent was removed at 70 8C. The mix-
ture was cooled to room temperature, and neat styrene
(0.094 g, 0.900 mmol), iodobenzene (0.122 g, 0.600 mmol),
and K3PO4 (0.255 g, 1.200 mmol) were added, followed by
the addition of toluene (0.1 mL) and distilled, N2-purged,
H2O (0.2 mL). The reaction mixture was stirred and heated
at 90 8C for 5 h, and then the volatile solvents were re-
moved in vacuo at room temperature. A cold finger was fit-
ted to the Schlenk tube and the product was isolated from
the ionic liquid mixture by sublimation at 80 8C for 14 h,
yielding 4-methoxy-biphenyl as a white solid (0.106 g,
0.588 mmol, 98%). 1H NMR (400.09 MHz, CDCl3)
d (ppm): 7.52 (d, 4H, 3J = 8.4 Hz), 7.36 (t, 4H, 3J =
1
CDCl3) d (ppm): 30 (s). H NMR (400.09 MHz, DMSO) d
3
(ppm): 10.5 (s, br,1H), 7.17 (s, br, 1H), 7.06 (d, 2H, J =
8.4 Hz), 6.54 (d, 2H, 3J = 8.4 Hz), 3.93 (t, 2H, 3J =
3
7.6 Hz), 7.26 (t, 2H, J = 7.6 Hz), 7.12 (s, 2H).
6.4 Hz), 2.22–2.15 (m, 11H), 1.73 (quintet, 2H,
3
3J = 6.0 Hz), 1.53–1.35 (m, 16H), 0.90 (t, 9H, J = 6.8 Hz).
Procedure for recycling the catalyst and the ionic liquid
Upon completion of the catalysis, the product was isolated
following the same procedure as noted. Proton NMR spec-
13C{1H}/gHSQC/gHMBC (599.44 MHz, DMSO) d (ppm):
163.4, 157.5, 153.6, 138, 129, 120.8, 109.5, 66.8, 27.8, 26.7
Published by NRC Research Press

30
Can. J. Chem. Vol. 88, 2010
Scheme 1. Synthesis of phosphonium-functionalized oxime carbopalladacycle 8.
troscopy of the resulting IL mixture confirmed the complete
removal of the product. The reagents were then added in the
same amounts, following the initial reaction conditions, and
the catalysis was repeated.
stretching vibration of 1673 cm–1, and the corresponding
C=N vibration in 8 was shifted by 26 cm–1 (observed at
1648 cm–1), which suggested N–Pd coordination (Fig. 2).30
Analysis by UV–vis spectroscopy provided further evidence
of the presence of the carbopalladacycle in 8. An absorption
at lmax 262 nm and an absorption band at lmax 307 nm with
a broad shoulder peak corresponded to ligand centered ab-
sorptions on the phosphonium salt. Ionic liquid 7 displayed
only one absorption band at lmax 260 nm. The absorption
band at lmax 307 nm symbolized charge-transfer transitions,
which were specific to metal–ligand orbital interactions, and
was in accordance with literature results (Fig. 3).19
Results and discussion
Synthesis
The synthetic route towards the preparation of the phos-
phonium salt catalyst 8 is summarized in Scheme 1. Com-
pounds 4 through 6 were prepared according to literature
n
procedures.19 Given the likely reactivity of Bu3P with palla-
dium, the generation of the phosphonium salt 7 was com-
pleted first, followed by formation of the carbopalladacycle
phosphonium salt 8. The quaternization of tri-n-butylphos-
phine with 6 was monitored by 31P{1H} NMR spectroscopy
A typical oxime carbopalladacycle is often a chloro-
bridged dimer, which occurs to fulfill the favorable four-
coordinate geometry about the palladium centre. However,
the imidazolium-tagged oxime carbopalladacycle was found
to be monomeric. The last coordination site was filled by
the imidazolium, preventing the formation of a dimer.19 Evi-
dence for both the monomer and dimer phosphonium ana-
logues, 8, were observed in ESI-MS with the most intense
peak attributed to the monomer. The dimer was the most
likely conformation, since there was no donor capability at
the phosphonium centre in contrast to the imidazolium
example. Attempts to grow crystals of suitable quality for
X-ray diffraction studies resulted in microcrystalline pow-
ders or clusters; thus, solid-state characterization was not
possible. No conclusive statement could be made from these
results to declare the coordination environment about the
palladium.
n
until the full consumption of Bu3P (d = –32 ppm) was ob-
served.28 The complete formation of the phosphonium salt
(d = 35 ppm) was detected within 48 h at 70 8C. The DMF
was removed in vacuo, and the product was purified by ex-
tractions with hexanes and THF, respectively. Comprehen-
sive characterization ascertained the formation of the
phosphonium ionic liquid 7 as a yellow, clear, viscous
liquid. Thermal analysis of 7 revealed a glass transition tem-
perature of –2 8C and a thermal decomposition temperature
of 343 8C. The addition of 7 and NaOAc to methanolic
Li2PdCl4 resulted in immediate precipitation of a red-brown
solid. Upon stirring for 3 days, a beige precipitate also
formed. The solids were isolated and washed with MeOH
followed by acetone and were finally recrystallized in ace-
tone. The beige powder was fully characterized to confirm
the formation of the carbopalladacycle using the following
analytical and spectroscopic methods: 1D and 2D NMR
spectroscopy, FTIR spectroscopy, UV–vis spectroscopy,
electrospray ionization mass spectrometry, differential scan-
ning calorimetry, thermogravimetric analysis, and elemental
Catalysis studies
To examine the catalytic ability of the phosphonium car-
bopalladacycle, two C–C bond forming cross-coupling reac-
tions, Heck and Suzuki–Miyaura, were carried out. The
evaluation of the Heck reaction was accomplished using
styrene and iodobenzene with K3PO4 and, the Suzuki-type
reaction was carried out using p-methoxyphenylboronic acid
and iodobenzene with K3PO4 as the base. The reaction con-
ditions including solvents, bases, reaction times, tempera-
ture, catalyst loading, and isolation techniques were varied,
and the results are summarized in Table 1.
Oxygen-free conditions were required, as reactions carried
out in the absence of an inert atmosphere resulted in low
yields (Table 1, entries 1, 2, and 14). It was observed that a
small amount of toluene and water were required to assist
the dissolution of the reagents in the ionic liquid medium,
1
analysis. The most salient feature in the H NMR spectrum
was the change in the pattern of the aromatic protons from
AA’BB’ to ABC, which was indicative of the formation of
the aryl carbopalladacycle environment (Fig. 1).19,29 The
phosphonium carbopalladacycle 8 was isolated as a white
powder in a modest yield (19%). Compound 8 was soluble
in DCM, DMSO, acetone, and trihexyltetradecylphospho-
nium chloride, [P66614][Cl], and was insoluble in CHCl3 and
apolar solvents, such as alkanes and ethers.
The FTIR spectrum for 7 displayed a characteristic C=N
Published by NRC Research Press

Tindale and Ragogna
31
1
Fig. 3. UV–vis spectra of 7 (solid line) and 8 (dotted line). The
Fig. 1. Stacked H NMR spectra for 7 (bottom) and 8 (top), which
emergence of a second band for 8 is indicative of a metal–ligand
interaction.
clearly display the change in the environment about the phenyl
protons, indicative of the incorporation of the palladium metal.
the reaction mixture with hexanes does not always com-
pletely remove all of the biaryl product, given some of the
biaryl species remains in the ionic liquid phase.22 This filtra-
tion strategy is limited by the inability to reuse the catalyst
and solvent because the phosphonium salts predominantly
remain coordinated to the silica. However, it is important to
note, that herein, no evidence of the phosphonium catalyst 8
was present in the hexanes extractions.
Fig. 2. FTIR spectrum of 7 (top) displays a C=N vibration at
1673 cm–1, and the FTIR spectrum of 8 (bottom) shows the shift of
the C=N vibration to 1648 cm–1 revealing N–Pd coordination.
In this context, it was ascertained that the best way to iso-
late the products was through sublimation. One can take ad-
vantage of the non-volatility of the reaction medium and
catalyst and simply remove the volatile product by sublima-
tion. Initial experiments were carried out in a small vial
(15 Â 48 mm) with a septum cap for the delivery of N2(g)
by a needle, given the volume of the reaction mixture was
small. Upon completion of the reaction, the mixture was dis-
solved in CH2Cl2 and transferred to a Schlenk flask. The
volatile solvent was removed under reduced pressure, the
flask was then fitted with a cold finger, and the product was
sublimed at 0.5 mm Hg and 80 8C. To further optimize the
isolation procedure, an unconventionally shaped Schlenk
tube was designed not only to accommodate the small reac-
tion volume, but also to include the capacity to fit a cold
finger to the flask (Fig. 4). Under the optimized reaction
conditions and with the appropriate glassware, high-yielding
Suzuki and Heck cross-coupling reactions were achieved
(Table 1, entries 13 and 19).
as the reaction performed using solely [P66614][Cl] yielded
only trace amounts of the desired product (Table 1, entry
14). To attain a minimal reaction time under mild tempera-
tures, the ideal catalyst loading was found to be 5 mol%
(Table 1, entry 13) for the Suzuki reaction and 2 mol% for
the Heck reaction (Table 1, entry 19).
With the use of viscous, non-volatile ionic liquids as sol-
vents in chemical transformations, modifications to the con-
ventional approaches to synthesis and product isolation are
necessary. Previously reported product isolation techniques
for Suzuki reactions performed in phosphonium ionic liquids
were assessed, including aqueous and organic solvent ex-
tractions, followed by filtration through silica (Table 1, en-
tries 2, 15, and 16) or alternatively, direct filtration of the
reaction mixture through a silica plug (Table 1, entries 1
and 14); all of which proved unsatisfactory for various rea-
sons. Separation of the product by extraction into organic
solvent increased volatile solvent use, which is to be
avoided if possible. Subsequent filtration through silica in-
creased the steps required to obtain a pure product, which
can reduce the final yield. The filtration of the ionic liquid
mixture was formerly deemed necessary, since extracting
Catalyst recyclability studies
While employing the optimized conditions, the recyclabil-
ity of the ionophilic catalyst was demonstrated. For both the
Heck and Suzuki reactions, upon completion of the reac-
tions, the residual toluene and water were removed in vacuo
at room temperature, and subsequently, a cold finger was in-
serted into the flask and the product was sublimed at
1
0.5 mm Hg at 80 8C overnight. H NMR spectroscopy of
the remaining IL mixture confirmed the absence of any
product or starting reagents. If residual product was ob-
served in the IL, then a second sublimation was performed.
Without removal of any residual borates or phosphates, the
starting reagents, not including fresh catalyst, were again
added to the ionic liquid mixture and the catalysis was re-
peated at the optimized conditions, and the results are sum-
marized in Table 2.
Published by NRC Research Press

32
Can. J. Chem. Vol. 88, 2010
Table 1. Reaction conditions studies for the Suzuki- and Heck-type cross-coupling.
Entry
1
2
3
4
5
6
7
8
Reaction
Suzukia
T (8C)
55
70
70
70
70
70
70
70
70
80
95
70
130
90
90
90
90
Time (h)
Catalyst 8 (mol%)
Yield (%)
1
16
16
16
2
4
6
8
4
7
4
5
24
16
16
4.5
6.5
5
1
1
1
1
1
1
1
1
5
1
1
5
2
2
2
2
2
2
Tracec,d
29c,e
76f
18g
42g
76g
82g
88g
9
83h
10
12
13
14
15
16
17
18
19
78h
66h
98j
Heckb
Tracec,d,i
75e
Tracee
73h
74h
98j
90
aIodobenzene (1 equiv.), p-methoxyphenylboronic acid (1.1 equiv.), K3PO4 (3.3 equiv.), 0.7 g [P66614][Cl],
0.2 mL H2O, and 0.1 mL toluene.
bIodobenzene (1 equiv.), styrene (1.5 equiv.), K3PO4 (2 equiv.), 0.3 g [P66614][Cl], 0.2 mL H2O, and 0.1 mL
toluene.
cUnder atmospheric conditions.
dProduct isolation by silica column filter.
eProduct isolation by extraction with hexanes followed by silica column filter.
fProduct isolation by extraction with hexanes, removal of solvent under reduced pressure followed by
sublimation.
gProduct isolation by aqueous extraction and sublimation of the remaining reaction mixture.
hProduct isolation by sublimation from reaction mixture.
iNo toluene or water added; NaOAc as base.
jProduct isolation by sublimation in reaction flask.
Fig. 4. Sublimation of trans-stilbene from the ionophilic catalyst
8 – [P66614][Cl] reaction mixture using the specially designed reac-
tion tube.
is likely due to the formation of Pd black and is probably
associated with the decrease in conversion. The phospho-
nium catalyst 8 exhibited good catalytic activity in Heck re-
action as well; however, the recyclability of the catalyst was
poor. The yields decreased by 20% on the second cycle and
continued to diminish rapidly over the subsequent cycles.
Again, the colour of the reaction mixture changed from yel-
low to a red-black on the first cycle, indicative of the initia-
tion of catalyst decomposition. However, it is important to
note that while 8 exhibits moderate recyclability, the imida-
zolium carbopalladacycle ionophilic catalyst by Corma et al.
displayed quite low reactivity. The Heck and Suzuki cou-
pling reactions exhibited yields ranging from 3%–25% on
the first run, and therefore, the catalyst was not recyclable.19
They proposed that the imidazolium tag was unstable under
the basic conditions required, and the formation of carbene
by deprotonation at C-2 on the imidazolium prevailed. The
resulting carbene would then coordinate to the palladium
and ultimately result in the decomposition of the catalyst,
the formation of Pd black, and poor catalytic activity. There-
fore, the substitution of a phosphonium salt for an imidazo-
lium cation proves to be a more effective route towards the
generation of an ionophilic tag for oxime carbopalladacycle
cross-coupling reaction catalysts, given basic conditions are
required for these types of reactions.
The ionophilic catalyst 8 – [P66614][Cl] mixture was re-
cycled three times before any decrease in activity was ob-
served for the Suzuki reaction. After the third reuse of the
catalyst, the IL mixture changed from yellow to black. This
Published by NRC Research Press

Tindale and Ragogna
Table 2. Recyclability of 8 in Suzuki and Heck cross-coupling reactions.
33
Reaction
Cycle
Isolated yield (%)
Suzuki
1
2
3
4
5
92
95
99
93
66
Heck
1
2
3
4
5
87
67
37
18
—
(6) Welton, T. Coord. Chem. Rev. 2004, 248 (21-24), 2459.
doi:10.1016/j.ccr.2004.04.015.
(7) Buijsman, R. C.; van Vuuren, E.; Sterrenburg, J. G. Org.
Lett. 2001, 3 (23), 3785. doi:10.1021/ol016769d. PMID:
11700138.
Conclusions
A phosphonium-tagged oxime carbopalladacycle catalyst
8 has been synthesized and comprehensively characterized.
The catalyst displayed ionophilic properties in the C–C
bond forming cross-coupling Heck and Suzuki reactions
conducted in an ionic liquid medium. Based on the ionic na-
ture of 8, the catalyst selectively partitioned into the ionic
liquid reaction phase, and furthermore, the ionic catalyst –
solvent media allowed the isolation of the reaction products
simply by sublimation without the need to use volatile or-
ganic solvents for extraction/separation techniques. This
straightforward product-isolation methodology enabled and
simplified the recycling of the catalyst and the reaction sol-
vent. The substitution of a phosphonium appendage on an
oxime carbopalladacycle as opposed to an imidazolium
functionality resulted in increased tolerance towards the ba-
sic conditions required in Heck and Suzuki reactions and
therefore increased catalytic activity.
(8) Wang, R.; Twamley, B.; Shreeve, J. M. J. Org. Chem. 2006,
71 (1), 426. doi:10.1021/jo052098b. PMID:16388677.
(9) Wang, B.; Zeng, Z.; Twamley, B.; Piekarski, M.; Shreeve, J.
M. Eur. J. Org. Chem. 2007, 2007 (4), 655. doi:10.1002/
ejoc.200600828.
ˇ
ˇ
´
(10) Sebesta, R.; Kmentova, I.; Toma, S. Green Chem. 2008, 10
(5), 484. doi:10.1039/b801456f.
(11) Lin, I. J. B.; Vasam, C. S. J. Organomet. Chem. 2005, 690
(15), 3498. doi:10.1016/j.jorganchem.2005.03.007.
(12) Zhao, D.; Fei, Z.; Geldbach, T. J.; Scopelliti, R.; Dyson, P. J.
J. Am. Chem. Soc. 2004, 126 (48), 15876. doi:10.1021/
ja0463482. PMID:15571412.
(13) Alonso, F.; Beletskaya, I. P.; Yus, M. Tetrahedron 2008, 64
(14), 3047. doi:10.1016/j.tet.2007.12.036.
(14) Aggarwal, V. K.; Emme, I.; Mereu, A. Chem. Commun.
(Camb.) 2002, (15): 1612. doi:10.1039/b203079a. PMID:
12170807.
(15) Magna, L.; Chauvin, Y.; Niccolai, G. P.; Basset, J.-M. Orga-
nometallics 2003, 22 (22), 4418. doi:10.1021/om021057s.
(16) Mathews, C. J.; Smith, P. J.; Welton, T. J. Mol. Catal.
Chem. 2004, 214 (1), 27. doi:10.1016/j.molcata.2003.11.030.
(17) Wang, R.; Xiao, J. C.; Twamley, B.; Shreeve, J. M. Org.
Biomol. Chem. 2007, 5 (4), 671. doi:10.1039/b616529j.
PMID:17285176.
Acknowledgements
We thank the Natural Sciences and Engineering Research
Council of Canada (NSERC), the Canada Foundation for In-
novation (CFI), the Ontario Ministry of Research and Inno-
vation, and The University of Western Ontario for funding.
We also thank Professor K. M. Baines for the use of TGA
and DSC instrumentation, D. Hairsine for the acquisition of
mass spectrometry data, and Y. Rambour for glassware fab-
rication.
(18) Wan, Q.-X.; Liu, Y.; Cai, Y.-Q. Catal. Lett. 2009, 127 (3-4),
386. doi:10.1007/s10562-008-9700-5.
´
(19) Corma, A.; Garcıa, H.; Leyva, A. Tetrahedron 2004, 60 (38),
8553. doi:10.1016/j.tet.2004.06.121.
(20) Alacid, E.; Alonso, D. A.; Botella, L.; Najera, C.; Pacheco,
References
´
(1) Rogers, R. D.; Seddon, K. R. Ionic Liquids as Green Sol-
vents: Progress and Prospects; American Chemical Society:
WA, 2003.
M. C. Chem. Rec. 2006, 6 (3), 117. doi:10.1002/tcr.20077.
PMID:16795006.
´
(21) Alonso, D. A.; Botella, L.; Najera, C.; Pacheco, M. C. Synth-
¨
(2) Ranke, J.; Stolte, S.; Stormann, R.; Arning, J.; Jastorff, B.
esis 2004, (10), 1713.
Chem. Rev. 2007, 107 (6), 2183. doi:10.1021/cr050942s.
PMID:17564479.
(3) Welton, T. Green Chem. 2008, 10 (5), 483. doi:10.1039/
b805586f.
(22) McNulty, J.; Capretta, A.; Wilson, J.; Dyck, J.; Adjabeng,
G.; Robertson, A. J. Chem. Commun. (Camb.) 2002, (17):
1986. doi:10.1039/b204699g. PMID:12271707.
(23) Kaufmann, D. E.; Nouroozian, M.; Henze, H. Synlett 1996,
1996 (11), 1091. doi:10.1055/s-1996-5658.
(4) Wasserscheid, P.; Welton, T. Ionic Liquids in Synthesis, 2nd
ed.; WILEY-VCH: Weinheim, 2008; Vol. 1.
(24) Gerritsma, D. A.; Robertson, A.; McNulty, J.; Capretta, A.
Tetrahedron Lett. 2004, 45 (41), 7629. doi:10.1016/j.tetlet.
2004.08.103.
ˆ
(5) Parvulescu, V. I.; Hardacre, C. Chem. Rev. 2007, 107 (6),
2615. doi:10.1021/cr050948h. PMID:17518502.
Published by NRC Research Press

34
(25) Fraser, K. J.; MacFarlane, D. R. Aust. J. Chem. 2009, 62 (4),
Can. J. Chem. Vol. 88, 2010
(28) Kosolapoff, G. M.; Maier, L. Organic Phosphorus Com-
pounds; John Wiley & Sons, Inc.: Toronto, ON, 1972;
Vol. 2.
309. doi:10.1071/CH08558.
(26) Wong, H.-t.; Pink, C. J.; Ferreira, F. C.; Livingston, A. G.
Green Chem. 2006, 8 (4), 373. doi:10.1039/b516778g.
(27) Naik, P. U.; McManus, G. J.; Zaworotko, M. J.; Singer, R.
D. Dalton Trans. 2008, (36): 4834. doi:10.1039/b811232k.
PMID:18766212.
´
(29) Botella, L.; Najera, C. J. Organomet. Chem. 2002, 663 (1-2),
46. doi:10.1016/S0022-328X(02)01727-8.
(30) Nielson, A. J. J. Chem. Soc., Dalton Trans. 1981, (1): 205.
doi:10.1039/dt9810000205.
Published by NRC Research Press