Improving reactivity and selectivity of aqueous-based Heck reactions by the local hydrophobicity of phosphine ligands

Article abstract of DOI:10.1016/j.tet.2015.09.010

Modification of a triarylphosphine with a cholate moiety affords a new ligand, 1, which is effective in palladium-catalyzed Heck cross-couplings between acrylates and aryl iodides under mild, aqueous reaction conditions. High yields, up to 99%, were achieved in water at 40 °C. In competition studies, a more hydrophobic substrate (n-Bu acrylate) was preferred over the least hydrophobic substrate (methyl acrylate), supportive of a localized hydrophobic microenvironment near the catalytic center. The enhanced reactivity and selectivity for hydrophobic substrates disappeared when the local hydrophobicity was eliminated using a standard water-soluble phosphine or in organic solvents.

Full text of DOI:10.1016/j.tet.2015.09.010

Tetrahedron 71 (2015) 8263e8270
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Improving reactivity and selectivity of aqueous-based Heck
reactions by the local hydrophobicity of phosphine ligands
*
*
Gina M. Roberts, Shiyong Zhang, Yan Zhao , L. Keith Woo
Department of Chemistry, Iowa State University, Ames, IA 50011-3111, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 June 2015
Received in revised form 2 September 2015
Accepted 3 September 2015
Available online 6 September 2015
Modification of a triarylphosphine with a cholate moiety affords a new ligand, 1, which is effective in
palladium-catalyzed Heck cross-couplings between acrylates and aryl iodides under mild, aqueous re-
action conditions. High yields, up to 99%, were achieved in water at 40 ꢀC. In competition studies, a more
hydrophobic substrate (n-Bu acrylate) was preferred over the least hydrophobic substrate (methyl ac-
rylate), supportive of a localized hydrophobic microenvironment near the catalytic center. The enhanced
reactivity and selectivity for hydrophobic substrates disappeared when the local hydrophobicity was
eliminated using a standard water-soluble phosphine or in organic solvents.
Keywords:
Catalysis
Selectivity
Ó 2015 Elsevier Ltd. All rights reserved.
Microenvironment
Aqueous
Hydrophobic effect
1. Introduction
cholate group was found to exert a strong influence on the activity
and selectivity of the reactions, owing to its ability to create a hy-
Transition metal-catalyzed reactions in water have attracted
significant research attention.1,2 An important motivation comes
from the low cost, abundance, nonflammability, and nontoxicity of
water as a green solvent. If the reactants are organic and phase-
separate from water, while the catalyst stays in the aqueous
phase, the product can be easily separated. Traditionally, transition
metal catalysts are made water-soluble by installing polar groups
such as sulfonate on the metal-coordinating ligands.2e4 However,
this method is limited to reactants with substantial solubility in
water, since highly nonpolar substrates tend to have difficulty
accessing catalysts located in the aqueous phase.
drophobic microenvironment near the catalytic center. This feature
resulted in a preference for hydrophobic substrates over less hy-
drophobic ones by the catalyst. In contrast, the enhanced selectivity
and reactivity was not observed in organic solvents or with a ho-
mogeneous control in which the catalyst was coordinated to water-
soluble phosphines.
2. Results and discussion
2.1. Ligand synthesis
Because surfactant micelles can solubilize a wide variety of
nonpolar compounds, chemists have also performed transition
metal-catalyzed reactions in the micellar phase.5e7 The benefit of
micelles is that they provide a local hydrophobic microenviron-
ment to transition metal catalysts and may help enhance the local
concentration of the substrate near the catalysts. Conversely, the
surfactants may contaminate the products and could also hamper
the product isolation by emulsion formation.
The motivation for designing a cholate-functionalized phos-
phine ligand was derived from the unusual amphiphilicity and
topology of cholic acid. Cholic acid and its associated bile salts
(cholates) are formed in the liver and used as a surfactant for
emulsifying lipids and cholesterol.8 Its rigid ring structure provides
unusual facial amphiphilicity derived from the hydrophilic and
hydrophobic moieties residing on opposite faces rather than in the
conventional head-to-tail arrangement of amphiphiles such as so-
dium dodecyl sulfate (SDS).9 This unique structure and facial
amphiphilicity has been exploited in supramolecular chem-
istry.9e13 According to Small’s primary/secondary aggregation
model, cholates form primary micelles at low concentrations in
water, with 2e10 monomers in the structure stabilized mainly by
hydrophobic interactions.14 As the cholate concentration increases,
these primary micelles can aggregate to larger secondary structures
In this work, we report the synthesis of a cholate-functionalized
phosphine ligand and its application in heterogeneous palladium-
catalyzed Heck cross-coupling reactions. The facially amphiphilic
* Corresponding authors. E-mail addresses: zhaoy@iastate.edu (Y. Zhao), kwoo@
iastate.edu (L.K. Woo).
http://dx.doi.org/10.1016/j.tet.2015.09.010
0040-4020/Ó 2015 Elsevier Ltd. All rights reserved.

8264
G.M. Roberts et al. / Tetrahedron 71 (2015) 8263e8270
through hydrogen bonds, leading to increased polydispersity in
solution.6,15e18
molecule to self-assemble into micellar aggregates, which are sol-
uble in water.18,24 For this reason, we employed a precatalyst
complexation methodology for our catalysis. Sonication of
Pd(OAc)2 and 3 equiv of ligand 1 in MeOH produced a light yellow,
hazy suspension within 30 min at ambient temperature. Removing
MeOH under vacuo provided a solid yellow phosphinocholate Pd
complex that was insoluble in water but formed an evenly dis-
persed suspension upon addition of NEt3. Increased dispersion was
likely due to deprotonation of the carboxylic acid moiety of 1,
producing a charged Pd-L complex.
Phosphine ligand 1 was readily derived from methyl cholate as
shown in Scheme 1. Of the three hydroxyl groups, the most reactive
hydroxyl resides at the C-3 position.19e21 Thus, the 3
a-hydroxyl was
selectively transformed into an azide by initially forming the
b
-
mesylate (2) with methanesulfonic acid under Mitsunobu condi-
tions, and subsequently treating the mesylate with sodium azide.22
The overall stereochemistry of the 3a-hydroxyl was retained in
azide compound 3, thus maintaining the original bifacial topology
of the cholate. Azide reduction to amine 4 was achieved with hy-
drogenation using palladium over carbon as the catalyst, and the
phosphine moiety was added via formation of an amide linkage
between 4 and activated ester 5. The methyl ester functionality of 6
was hydrolyzed with 1.0 M NaOH in methanol (MeOH). Isolation of
the carboxylic acid (1) was achieved by purification on a silica gel
Studies by Jutand and co-workers established that a Pd0 com-
plex was formed between Pd(OAc)2 and PPh3, and the dominate
species was directly dependent on the amount of phosphine added
to the system.25e27 Three equiv of phosphine, in the presence of
wet DMF, led to the overall formation of an anionic palladium (0)
species (7) and phosphine oxide (eq. 1), identified by 31P NMR.
(1)
7
column. Similar to cholic acid, 1 was essentially insoluble in water
and sparingly to moderately soluble in ethanol, MeOH, chloroform
and dichloromethane.16,17 The 31P NMR spectrum of 1 exhibited
a single phosphine signal at ꢁ5.0 ppm with triphenylphosphate
(ꢁ18.0 ppm)23 as an external standard (CDCl3/CD3OD, 1:1).
Under a similar study with ligand 1 and Pd(OAc)2, analogous 31
P
NMR peaks were observed. A solvent mixture of 1:1 CDCl3:CD3OD
sufficiently solubilized the yellow phosphinocholate Pd solid iso-
lated from MeOH (1:3, Pd(OAc)2 to 1), and three 31P NMR signals
were observed (ꢁ5.0, 16.9, and 32.6 ppm). The signals at ꢁ5.0 and
Scheme 1. Synthesis of phosphine ligand 1.
The insolubility of compound 1 in water was not surprising.
Even for cholic acid itself, the dominance of hydrophobic groups
makes the molecule practically insoluble in water. Only when the
carboxylic acid is deprotonated, the ionic carboxylate enables the
32.6 ppm corresponded to free and oxidized ligand 1, respectively.
The peak at 16.9 ppm was assigned to a Pd0 complex, 8, coordinated
with 2 equiv of 1. After 4 h in organic solvents, complex 8 mostly
decomposed to palladium black, resulting in an increase in the

G.M. Roberts et al. / Tetrahedron 71 (2015) 8263e8270
8265
phosphine oxide signal. Minor amounts of uncharacterized phos-
phine products were also detected, exhibiting broad 31P peaks
around 30 ppm. 1H NMR analysis revealed that both aryl and
cholate backbone proton signals were broader than those of free
ligand 1. If only 2 equiv of 1 were used for Pd-complex formation,
the isolated complex decomposed too rapidly for NMR analysis.
Increasing the amount of phosphine to 4 equiv of 1, provided
similar results as those with 3 equiv, with the appearance of
a transient broad peak around 5e7 ppm, which may correspond to
higher coordinate palladium complexes, Pd(L)n (n¼3e4).26,28
Mass spectrometric analysis of the solid isolated from MeOH
(Pd(OAc)2/1¼1:3) supported the formation of a single bis-ligated
palladium complex (8) with a mass cluster centered at 1498 m/z,
and an isotopic mass pattern consistent with a PdL2 complex (see
Fig. S18). A bound acetate ligand was not detected by MS analysis.
In addition to the Pd-1 complex (8), the mass spectrum exhibited
a peak at 710 m/z, corresponding to a phosphine oxide species,
indicating that ligand 1 was oxidized during formation of com-
plex 8.
any significant decomposition.29 It is possible that the extensive
hydrophobic and hydrogen-bonding interactions in the aggregated
form of the complex in water reduced the migration of Pd0, thus
enhancing its stability. Note that, in organic solvents, the complex
easily decomposed into palladium black.
Despite the low solubility of complex 8 in any of the solvents,
the best medium for the heterogeneous Heck catalysis was clearly
water. The results suggest that the local hydrophobicity of the
cholates was important to the catalysis. Most likely, the improved
reactivity of Pd-complex 8 derived from a microenvironment
formed by the cholate ligands that enhanced the local concentra-
tions of reactants around the active Pd center. It is also possible that
the irregular shape and the amphiphilicity of the cholate, known to
make cholate derivatives pack poorly in the solid state,30 enables
similar hydrophobic channels or voids to be formed near the cat-
alysts (vide infra).
Aqueous-based Heck coupling typically gives higher yields with
the use of inorganic bases.2,3,31 However in our cholate phosphine-
Pd system, a substantial decrease in yield was seen when NEt3 was
replaced with either NaOAc or K2CO3 (Table 2). Catalyst de-
composition was extensive within 4 h with either of these inorganic
bases. Conversely, if the alkyl groups of the amine were lengthened
from ethyl to n-octyl, activity was strongly inhibited, but catalyst
robustness was enhanced. For example, with N(octyl)3, only a trace
amount of product was detected for the coupling between 4-
iosoanisole and tert-butyl acrylate, and no catalyst decomposition
was visible after 24 h at 40 ꢀC (entry 3).
To understand the importance of the local hydrophobicity of the
phosphinocholate ligand, we studied the Heck reactions of differ-
ent acrylates with varying water solubility (Table 3). Since the alkyl
group on the acrylate is three bonds away from the reactive center
(the alkene), it is expected to have negligible electronic and steric
effect on the Heck coupling. Meanwhile, as the alkyl substituent of
an acrylate ester lengthens from methyl to hexyl, its solubility in
water essentially decreases to zero. In agreement with our pro-
posed local hydrophobicity-derived activity, more hydrophobic
acrylates were clearly more reactive than less hydrophobic ones.
2.2. Heterogeneous Heck coupling catalyzed by PdePhosphi-
nocholate complex 8
The instability of the Pd complex in organic solvents (in the
absence of suitable substrates) precluded its isolation. For the same
reason, as described in the experimental section, the mixture ob-
tained from Pd(OAc)2 and ligand 1 was used directly in the Heck
coupling between aryl iodides and acrylates. As revealed in the 31
P
NMR studies mentioned above, the mixture mainly consisted of
complex 8, phosphine oxide, and excess ligand 1.
Complex 8 was catalytically active at ambient temperature
(Table 1, entry 1) and afforded the product in essentially quanti-
tative yield under mild heating (40 ꢀC, entry 3). Most importantly,
the reaction gave the highest yield in water. Reactions in methanol
(entries 5e7), a methanol/water mixture (entries 8e9), or aceto-
nitrile (entry 10) afforded significantly less product, and higher
temperature (60 ꢀC) did not solve the problem.
Table 1
Optimization of Heck coupling reaction conditions using complex 8 in various mediaa
Rxn
Solvent
Amount of tert-butyl acrylate (mmol)
Amount of Et3N (mmol)
t (h)
T (ꢀC)
NMR yield (%)b
1
2
3
H2O
H2O
H2O
H2O
MeOH
MeOH
MeOH
MeOH/H2O
MeOH/H2O
CH3CN
0.12
0.12
0.24
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.18
0.18
0.36
0.18
0.18
0.18
0.18
0.18
0.18
0.18
4
4
4
4
5
2
2
4
4
2
RT
40
40
40
RT
40
60
RT
60
40
41
78
98
95
Trace
14
14
24
29
4c
5d
6d
7d
8
9
10
48
a
Conditions: see experimental for ligand-Pd precoordination method.
Yields determined by NMR using mesitylene as internal standard.
Catalyst was formed, dried and stored for 6 months before use in Heck coupling.
Precatalyst/MeOH solution used directly as reaction medium.
b
c
d
Although many phosphine complexes tend to be air-sensitive, the
For example, 2-ethylhexyl and n-butyl acrylates produced yields of
90 and 92%, respectively, whereas methyl acrylate only afforded
12% yield.
Importantly, the selectivity for hydrophobic acrylates was
completely absent in organic solvent (DMF) in our competition
studies. The catalytic complex was slightly more soluble in DMF
activity of our catalyst was not significantly affected by the presence
of oxygen. As a dry solid, the catalyst residue could be stored under
light and air for more than 6 months with no substantial loss in ac-
tivity (Table 1, entry 4). Inwater, the catalyst complexcould be heated
at 40 ꢀC under atmospheric conditions for more than 3 days without

8266
G.M. Roberts et al. / Tetrahedron 71 (2015) 8263e8270
Table 2
Effect of base on Heck coupling with complex 8 in aqueous mediaa
Rxn
Base
t (h)
Yield (%)b
1
2
3
NaOAc
K2CO3
N(C8H17
4
4
24
4
21
Trace
)
3
a
Conditions: see experimental for ligand-Pd precoordination method. 0.12 mmol iodoanisole, 0.24 mmol tert-butyl acrylate, 0.36 mmol base, 2.4ꢃ10ꢁ3 mmol Pd(OAc)2,
7.5ꢃ10ꢁ3 mmol 1, and 1 mL H2O. Stirred at 40 ꢀC.
b
Average 1H NMR yields for duplicate runs (ꢂ3).
Table 3
Effect of Acrylate solubility on Heck coupling with complex 8 in aqueous mediaa
Rxn
R
Solubilityb (g/L)
t (h)
Yield (%)c
1
2
3
4
Me
Et
60
15
4
4
4
4
12
77
92
90
n-Bu
2-ethylhexyl
1.4
dd
a
Conditions: see experimental for ligand-Pd precoordination method.
Solubility of acrylate in water.
b
c
Average 1H NMR yields for duplicate runs (ꢂ3).
Solubility in water was negligible.
d
than water, but was still predominantly a suspension. When methyl
acrylate and n-butyl acrylate were used in a competition reaction to
couple with 4-iodoanisole, a product ratio of 1:1 was obtained in
DMF (Scheme 2). In an analogous study with water as the solvent,
the product ratio favored the n-butyl cinnamate product over the
methyl analog (2.6:1), confirming the importance of hydrophobic
interactions to the reactivity and selectivity.
However, we were also interested in the distribution of 1 between
the solid and supernatant fractions. In an additional separation
study, a catalytic reaction was removed from heat after 2 h and the
solid and solution phases were separated by centrifugation. Both
fractions were thoroughly dried in vacuo. Upon dissolving each of
the residues in 1:1 CDCl3:CD3OD, 31P NMR analysis indicated that
all the phosphorous containing material remained in the original
Scheme 2. Competitive Heck reaction of methyl and n-butyl acrylate with 4-iodoanisole.
Although the catalytic system appeared heterogeneous, it was
possible that the active palladium species during the reaction was
water-soluble. To determine the phase of the active species, a het-
erogeneity test was performed with a reaction that was centrifuged
after 2 h at 40 ꢀC. The solid and supernatant fractions were sepa-
rated, and a second aliquot of reagents was added to each. Water
was also added to the solid fraction. After an additional 2 h at 40 ꢀC,
only the solid fraction proved to remain catalytically active,
resulting in a total yield of 94%. No additional product was detected
by NMR analysis of the supernatant fraction, even after heating at
40 ꢀC for an additional 6 h.
solid fraction. No detectable amount of cholate species was present
in the aqueous solution fraction.
The conventional method in performing transition metal-
catalyzed reactions in water is to install water-solubilizing groups
such as sulfonate on the metal-coordinating ligands.2e4 Our results
so far suggest that the presence of local hydrophobicity near the
catalyst seems to be more important than water-solubility, at least in
this particular system. To further verify the importance of the local
hydrophobicity, we performed similar aqueous Heck coupling using
a water-soluble phosphine ligand, sodium 4-(diphenylphosphino)
benzoate (9), for its similar electronic property of the phosphine.
Consistent with our hypothesis, the reactions catalyzed by the
Pde9 complex (Table 4) gave much poorer results than those
Lack of reactivity with the supernatant fraction indicated that no
active palladium species were dissolved in the aqueous layer.

G.M. Roberts et al. / Tetrahedron 71 (2015) 8263e8270
8267
catalyzed by the corresponding phosphinocholate complex (Table
1). In addition to yields never exceeding 30% with the water-
soluble catalytic system, the selectivity for hydrophobic sub-
strates was completely absent (Table 4, entries 1e5). The yields
showed no improvements when the base or reaction solvent was
varied. These results provide addition support for the importance of
local hydrophobicity for the catalysis. Although fully soluble in
water and containing phenyl groups on the ligand, the Pd complex
with ligand 9 apparently did not possess the necessary local hy-
drophobic environmental as in the cholate-based catalyst 8 for the
efficient Heck coupling between hydrophobic substrates. It should
be mentioned that, with the proper ligands, excellent yields have
been obtained in the literature for Pd-catalyzed Heck reactions in
aqueous/organic solvents, even for less reactive substrates such as
aryl bromides.32e34 This latter approach, however, is not expected
to afford selectivity such as butyl over methyl acrylate demon-
strated by our system.
of the work lies in its implication in catalyst design. Heterogeneous
catalysts are frequently prepared on a hydrophobic support such as
activated carbon or graphene. Such catalysts, however, are ‘globally’
hydrophobic and do not possess the localized feature exemplified by
the phosphinocholate ligand. This work suggests that creation of
a local hydrophobic microenvironment near the catalytic center can
improve both the reactivity and selectivity of the catalyst. Since the
hydrophobicity-derived reactivity and selectivity did not originate
from the Pd metal or the catalytic reaction itself, we expect the
principle could be applied to other systems as well.
4. Experimental section
4.1. General method
All reagents and solvents were ACS-certified grade or higher,
and were used as received from commercial suppliers, unless
Table 4
Results of Heck coupling reaction conditions using water soluble ligand, 9, in various mediaa
Rxn
R
Base
Solvent
Yield (%)b
1
2
3
t-Bu
Me
Et
NEt3
NEt3
NEt3
H2O
H2O
H2O
29
23 (16:7)c
24 (18:6)c
4
5
6
7
8
9
10
11
12
n-Bu
2-ethylhexyl
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
NEt3
NEt3
NaOAc
K2CO3
N(C8H17
NEt3
NEt3
NEt3
NEt3
H2O
H2O
H2O
H2O
19
20
11
14
12
13
16
21
27
)
H2O
3
MeOH/H2O
MeOH
CH3CN/H2O
CH3CN
a
Conditions: see experimental for ligand-Pd precoordination method.
Average 1H NMR yields for duplicate runs (ꢂ3).
trans:cis ratio of cinnamate product.
b
c
The scope of the catalyst’s reactivity is exhibited in Table 5. In
otherwise indicated. The aqueous reactions used deionized water
without further purification or degassing. Routine 1H, 31P and 13C
NMR spectra were recorded on a Varian VXR-400 spectrometer.
Mass spectrometry for compounds 1 and 6 were performed on
a Waters GCT GCeMS, using ESI as the ionization method. Mass
spectrometry for complex 8 was performed on an Agilent 6540 Q-
TOF LC-MS, using APCI as the ionization method. Centrifugation
was performed with a Fisher Scientific Micro 17R Microcentrifuge.
Syntheses of cholate derivatives 2e4 were accomplished as pre-
viously reported.37
general, no (Z)-isomers were detected by 1H NMR. For example, the
olefinic coupling constant for the trans protons was 16.0 Hz in the
products. The catalyst was not active for aryl bromides or aryl
chlorides. Both electron donating and electron withdrawing sub-
stituents on both the aryl and olefinic reagents produced good to
excellent yields at 40 ꢀC.
3. Conclusions
In comparison to the water-soluble Pde9 complex, the
Pdephosphinocholate complex (8) afforded dramatically improved
results in aqueous-based Heck coupling reactions. Although the
catalytic activity was not high for aryl bromides or chlorides, the
local hydrophobicity of the phosphinocholate ligand enabled
quantitative yields for aryl iodides in cases while the water-soluble
Pde9 complex only afforded <30% yields. In the literature, local
hydrophobicity has been found to be important in other systems.35
Cyclodextrin has been used to improve Pd-catalyzed Heck reactions
in aqueous/organic mixtures.36 In our case, despite the heteroge-
neity of the reaction system, the local hydrophobicity with the
phosphinocholate enabled unusual selectivity among acrylates that
was otherwise absent in organic solvents.
4.2. Syntheses
4.2.1. Synthesis of compound 5. A mixture of 4-(diphenylphos-
phino)benzoic acid (153 mg, 0.50 mmol), N-hydroxysuccinimide
(87 mg, 0.75 mmol), and 1,3-dicyclohexylcarbodiimide (DCC,
155 mg, 0.75 mmol) was dissolved in 4.0 mL of anhydrous CH3CN/
THF (3:1). The mixture was stirred at room temperature for w12 h.
The urea was separated by filtration and the filtrate evaporated to
dryness leaving the active ester 5 as a light yellow powder
(203.3 mg, 0.50 mmol, 100%), which was used in the next step
without further purification.
Although these results were obtained with an ‘easy’ reaction
(Heck coupling between aryl iodides and acrylates), the significance
4.2.2. Synthesis of compound 6. Compound 5 (202 mg, 0.50 mmol),
diisopropylethylamine (DIPEA, 350 mL, 2.0 mmol), and 4 (253 mg,

8268
G.M. Roberts et al. / Tetrahedron 71 (2015) 8263e8270
to 30:1) to give a white powder (203.5 mg, 0.29 mmol, 57%). 1H
NMR (400 MHz, CDCl3:CD3OD, 1:1):
7.40e7.24 (m, 12H), 5.40 (CH2Cl2), 3.96 (s, 1H), 3.81 (s, 1H), 3.75 (m,
1H), 3.65 (s, 3H, OCH3), 2.45e1.0 (m, 26H), 0.98 (d, J¼8.0 Hz, 3H),
0.94 (s, 3H) 0.70 (s, 3H), NH not observed; 13C NMR (400 MHz,
Table 5
Scope of Heck cross-coupling reactions Catalyzed by 8 in aqueous media
d
7.74 (d, J¼8.0 Hz, 2H),
CDCl3:CD3OD, 1:1)
d 175.0, 167.1, 141.3, 141.2, 135.9, 135.8, 134.3,
133.3, 133.1, 132.8, 132.6, 128.5, 128.1, 128.0, 126.6, 126.5, 72.4, 67.5,
50.8, 49.9, 46.4, 45.8, 41.6, 41.2, 38.9, 35.6, 35.5, 35.0, 34.2, 34.2, 33.1,
30.5, 27.7, 27.0, 26.6, 26.0, 25.1, 24.4, 22.6, 21.9, 16.3, 11.7 (some
aromatic carbon signals were split by the phosphorus); 31P NMR
Rxn
1
Aryl halide
Olefin
NMR yield (%)a Isolated yield (%)b
100
98
96
94
91
(400 MHz, CDCl3:CD3OD, 1:1):
d
ꢁ6.2; High resolution ACPI-MS (m/
z): [MþH]þ calcd for C44H57NO5P, 710.3969; found, 710.3979.
4.2.3. Synthesis of ligand 1. Compound 6 (200 mg, 0.28 mmol) was
dissolved in MeOH (10.0 mL) followed by addition of NaOH (1.0 M,
2.82 mL, 2.8 mmol). The reaction mixture was stirred at room
temperature for 12 h and monitored via TLC. Upon completion of
the reaction, the solvent was removed under reduced pressure. The
yellow solid collected was purified by column chromatography
(CH2Cl2: CH3OH¼20:1) to give a white powder (183 mg, 0.26 mmol,
2
3
89
93%). 1H NMR (400 MHz, CDCl3:CD3OD, 1:1):
d
8.04 (d, J¼12.0 Hz,
1H, NH), 7.74 (d, J¼8.0 Hz, 2H), 7.40e7.12 (m, 12H), 3.96 (s, 1H), 3.81
(s, 1H), 3.76 (m, 1H), 3.35 (CH3OH), 2.40e1.04 (m, 26H), 1.0 (d,
J¼6.0 Hz, 3H), 0.93 (s, 3H) 0.7 (s, 3H),; 13C NMR (400 MHz,
4
5
6
7
8
94
d
d
d
d
82
CDCl3:CD3OD, 1:1):
d 176.4, 166.9, 141.2, 141.1, 135.8, 135.6, 134.1,
133.1, 132.9, 132.6, 132.4, 128.3, 127.9, 127.8, 126.43, 126.36, 72.1,
67.2, 49.8, 46.2, 45.6, 41.5, 41.1, 38.8, 35.3, 34.8, 34.0, 33.9, 32.9, 30.4,
27.6, 26.9, 26.4, 25.9, 24.9, 24.2, 22.4, 21.8, 16.2, 11.6 (some aromatic
carbon signals were split by the phosphorus); 31P NMR (400 MHz,
92
CDCl3:CD3OD, 1:1):
calcd for C43H53NO5P, 694.3656; found, 694.3649.
d
ꢁ5.0; High resolution ESI-MS (m/z): [MꢁH]ꢁ
16 (67)c
94d
62
4.3. General procedure for in situ Heck reaction
To a 20-mL vial, 5ꢃ10ꢁ3 mmol of PdCl2(CH3CN)2 or Pd(OAc)2 and
15ꢃ10ꢁ3 mmol of 1 were added, followed by 0.24 mmol 4-
iodoanisole, 0.48 mmol t-Bu-acrylate, 0.72 mmol NEt3, and
2.0 mL H2O. The reaction mixture was ultrasonicated for 10 min and
then stirred at 40 ꢀC for 18 h. At the completion of the reaction, all
volatiles were removed under vacuum, and 50
standard (10. mM mesitylene in CDCl3) was added to the vial. The
L). All organic
extracts were combined and passed through a plug of Celite and
MgSO4 into an NMR tube. Yields were determined by 1H NMR
spectroscopy by integration of the protons of the cinnamate
product, using mesitylene as the internal standard. In general, both
the aromatic and the methyl protons were used in the calculation
and the results were averaged to give the reported NMR yields.
mL of an internal
reaction was then extracted with CDCl3 (3ꢃ200
m
9
61
70
80
10
d
4.4. Coordination of 1 with Pd(OAc)2
11
d
d
d
d
Pd(OAc)2 (5.4 mg, 2.4ꢃ10ꢁ2 mmol) and
1
(50 mg,
7.2ꢃ10ꢁ2 mmol) were combined in MeOH (10 mL) and treated by
sonication at 25 ꢀC for 30 min, resulting in a hazy, yellow solution.
Into 20-mL glass vials, 1.0 mL of this solution was delivered and all
volatiles were removed under vacuum, leaving a dry, yellow resi-
due to be directly used for catalysis.
12
a
Conditions: see experimental for ligand-Pd precoordination method. Average 1
H
NMR yields for duplicate runs (ꢂ3).
b
Isolated yield using scaled-up procedure.
4.5. General procedure for Heck reaction with ligand-Pd
precoordination
c
Parenthetical yield is of doubly coupled product.
d
No doubly coupled product detected.
To the dry, yellow catalyst residue, aryl iodide (0.12 mmol), ac-
rylate (0.24 mmol), 1.0 mL H2O, and NEt3 (51 mL, 0.36 mmol) were
added to the vial in the specified order. The reaction was thoroughly
mixed via sonication for 10 min and subsequently stirred for the
indicated time and temperature. At the completion of the reaction,
0.60 mmol) were dissolved in the mixture of MeOH (2.0 mL) and
THF (2.0 mL). The mixture was stirred at room temperature for 48 h.
The solvent was removed under reduced pressure, and the product
was purified by column chromatography (CH2Cl2:CH3OH from 60:1

G.M. Roberts et al. / Tetrahedron 71 (2015) 8263e8270
8269
all volatiles were removed under vacuum, and 50.
standard (10. mM mesitylene in CDCl3) was added to the vial. The
L). All organic
extracts were combined and passed through a plug of Celite and
MgSO4 into an NMR tube. Yields were determined by 1H NMR
spectroscopy.
m
L of internal
J¼7.2 Hz, 2H), 7.47 (d, J¼8.4 Hz, 2H), 7.36 (t, J¼7.2 Hz, 2H), 7.24 (t,
J¼7.2 Hz,1H), 7.08 (d, J¼16.4 Hz,1H), 7.02 (d, J¼16.4 Hz,1H), 6.91 (d,
J¼8.4 Hz, 2H), 3.85 (s, 3H).
reaction was then extracted with CDCl3 (3ꢃ200
m
4.6.9. (E)-1-Methoxy-4-(4-methylstyryl)benzene. White
108 mg, 0.48 mmol, 80%; 1H NMR (400 MHz, CDCl3)
solid,
(ppm): 7.45
d
(d, J¼8.4 Hz, 2H), 7.39 (d, J¼8.0 Hz, 2H), 7.16 (d, J¼8.4 Hz, 2H), 7.03
(d, J¼16.0 Hz, 1H), 7.00 (d, J¼16.0 Hz, 1H), 6.91 (d, J¼8.4 Hz, 2H),
3.84 (s, 3H), 2.37 (s, 3H).
4.6. General procedure for scaled-up Heck reaction with
Ligand-Pd precoordination
Pd(OAc)2 (5.4 mg, 2.4ꢃ10ꢁ2 mmol) and
1
(50 mg,
4.7. General procedure for heterogeneity tests
7.2ꢃ10ꢁ2 mmol) were combined in MeOH (10 mL) and treated by
sonication at 25 ꢀC for 30 min, resulting in a hazy, yellow solution.
Into 20-mL glass vials, 5.0 mL of this solution was delivered and all
volatiles were removed under vacuum, leaving a dry, yellow resi-
due to be used directly for catalysis. To the dry, yellow catalyst
residue, aryl iodide (0.60 mmol), acrylate (1.2 mmol), 5.0 mL H2O,
To dry, yellow catalyst residue, iodoanisole (0.12 mmol), t-Bu-
acrylate (0.24 mmol), 1.0 mL H2O, and NEt3 (51 mL, 0.36 mmol) were
added to the vial in the specified order. The reaction was thoroughly
mixed via sonication for 10 min and subsequently stirred at 40 ꢀC.
At 2 h, the reaction mixture was centrifuged at 18 ꢀC and 16,200 g
for 30 min. The two fractions were separated and the solid fraction
was reconstituted with more iodoanisole (0.12 mmol), t-Bu-acry-
and NEt3 (250 mL, 1.8 mmol) were added to the vial in the specified
order. The reaction was thoroughly mixed via sonication for 10 min
and subsequently stirred for 12 h at 40 ꢀC. At the completion of the
reaction, all volatiles were removed under vacuum, and the
remaining aqueous solution was extracted with ethyl acrylate
(3ꢃ5 mL). The ethyl acrylate extracts were combined, dried with
MgSO4, and all volatiles were removed under vacuum. The result-
ing crude was purified via column chromatography.
late (0.24 mmol), H2O (1.0 mL), and NEt3 (51 mL, 0.36 mmol). The
reconstituted solid fraction was stirred at 40 ꢀC for an additional
2 h. To the supernatant fraction, more iodoanisole (0.12 mmol), t-
Bu-acrylate (0.24 mmol), and NEt3 (51 mL, 0.36 mmol) were added.
The reconstituted supernatant fraction was stirred at 40 ꢀC for an
additional 6 h. Product workup and analysis were accomplished as
described above. This procedure was performed in triplicate. The
overall average yield of cinnamate product from the solid fractions
was 96%, whereas the supernatant fractions never provided more
than a trace amount of product via NMR.
4.6.1. Methyl (E)-3-(4-methoxyphenyl)acrylate. Clear oil, 76 mg,
0.40 mmol, 65%; 1H NMR (400 MHz, CDCl3)
d (ppm): 7.66 (d,
J¼16.0 Hz, 1H), 7.48 (d, J¼8.4 Hz, 2H), 6.91 (d, J¼8.4 Hz, 2H), 6.32 (d,
J¼16.0 Hz, 1H), 3.85 (s, 3H), 3.80 (s, 3H).
Acknowledgements
4.6.2. Ethyl (E)-3-(4-methoxyphenyl)acrylate. Clear oil, 106 mg,
0.51 mmol, 86%; 1H NMR (400 MHz, CDCl3)
d (ppm): 7.65 (d,
We thank the U.S. Department of Energy (award DESC0002142)
and National Science Foundation (CHE-0809901 and EEC-0813570:
ERC Center for Biorenewable Chemicals, CBiRC, at Iowa State Uni-
versity) for partial support of this work.
J¼15.6 Hz, 1H), 7.48 (d, J¼8.8 Hz, 2H), 6.91 (d, J¼8.8 Hz, 2H), 6.32 (d,
J¼16.0 Hz, 1H), 4.27 (q, J¼7.2 Hz, 2H), 3.85 (s, 3H), 1.34 (t, J¼7.2 Hz,
3H).
4.6.3. Butyl (E)-3-(4-methoxyphenyl)acrylate. Clear oil, 134 mg,
Supplementary data
0.57 mmol, 95%; 1H NMR (400 MHz, CDCl3)
d (ppm): 7.65 (d,
J¼16.0 Hz, 1H), 7.48 (d, J¼8.8 Hz, 2H), 6.91 (d, J¼8.8 Hz, 2H), 6.32 (d,
J¼16.0 Hz, 1H), 4.20 (t, J¼6.8 Hz, 2H), 3.84 (s, 3H), 1.69 (m, 2H), 1.45
(m, 2H), 0.97 (t, J¼7.2 Hz, 3H).
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.tet.2015.09.010.
References and notes
4.6.4. 2-Ethylhexyl (E)-3-(4-methoxyphenyl)acrylate. White solid,
162 mg, 0.56 mmol, 93%; 1H NMR (400 MHz, CDCl3)
d (ppm): 7.65
1. Cornils, B.; Herrmann, W. A. Aqueous-phase Organometallic Catalysis: Concepts
and Applications, 2nd ed.; Wiley-VCH: Weinheim, 2004.
2. Shaughnessy, K. H. Chem. Rev. 2009, 109, 643e710.
3. Li, C.-J. Chem. Rev. 2005, 105, 3095e3166.
(d, J¼16.0 Hz, 1H), 7.49 (d, J¼8.8 Hz, 2H), 6.91 (d, J¼8.8 Hz, 2H), 6.32
(d, J¼16.0 Hz, 1H), 4.12 (m, 2H), 3.85 (s, 3H), 1.65 (m, 1H), 1.48e1.27
(m, 8H), 0.96e0.90 (m, 6H).
€
4. Lindstrom, U. M. Chem. Rev. 2002, 102, 2751e2772.
5. Menger, F. M.; Gan, L. H.; Johnson, E.; Durst, D. H. J. Am. Chem. Soc. 1987, 109,
4.6.5. tert-Butyl (E)-3-(4-methoxyphenyl)acrylate. White solid,
2800e2803.
136 mg, 0.58 mmol, 96%; 1H NMR (400 MHz, CDCl3)
d (ppm): 7.55
6. Dwars, T.; Paetzold, E.; Oehme, G. Angew. Chem., Int. Ed. 2005, 44, 7174e7199.
7. Lipshutz, B. H.; Ghorai, S. Aldrichimica Acta 2008, 41, 59e72.
8. Danielsson, H.; Sjovall, J. Sterols and Bile Acids; Elsevier: Amsterdam, 1985.
9. Zhao, Y. Curr. Opin. Colloid Interface Sci. 2007, 12, 92e97.
10. Brotherhood, P. R.; Davis, A. P. Chem. Soc. Rev. 2010, 39, 3633e3647.
11. Li, Y.; Dias, J. R. Chem. Rev. 1997, 97, 283e304.
12. Virtanen, E.; Kolehmainen, E. Eur. J. Org. Chem. 2004, 2004, 3385e3399.
13. (a) Zhao, Y.; Cho, H.; Widanapathirana, L.; Zhang, S. Acc. Chem. Res. 2013, 46,
2763e2772; (b) Zhao, Y.; Zhong, Z.; Ryu, E.-H. J. Am. Chem. Soc. 2007, 129,
218e225; (c) Cho, H.; Zhong, Z.; Zhao, Y. Tetrahedron 2009, 65, 7311e7316; (d)
Cho, H.; Zhao, Y. J. Am. Chem. Soc. 2010, 132, 9890e9899; (e) Cho, H.; Wida-
napathirana, L.; Zhao, Y. J. Am. Chem. Soc. 2011, 133, 141e147; (f) Zhong, Z.; Li, X.;
Zhao, Y. J. Am. Chem. Soc. 2011, 133, 8862e8865; (g) Gunasekara, R. W.; Zhao, Y.
J. Am. Chem. Soc. 2015, 137, 843e849.
14. Small, D. M. In Molecular Association in Biological and Related Systems; Goddard,
E., Ed.; Advances in Chemistry; American Chemical Society: Washington, DC,
1968; Vol. 84, pp 31e52.
15. Sugioka, H.; Moroi, Y. Biochim. Biophys. Acta, Lipids Lipid Metab. 1998, 1394,
99e110.
(d, J¼16.0 Hz, 1H), 7.46 (d, J¼8.8 Hz, 2H), 6.89 (d, J¼8.8 Hz, 2H), 6.24
€
(d, J¼16.0 Hz, 1H), 3.85 (s, 3H), 1.54 (s, 9H).
4.6.6. tert-Butyl (E)-3-(p-tolyl)acrylate. White solid, 123 mg,
0.56 mmol, 94%; 1H NMR (400 MHz, CDCl3)
d (ppm): 7.57 (d,
J¼16.0 Hz, 1H), 7.41 (d, J¼8.0 Hz, 2H), 7.18 (d, J¼8.0 Hz, 2H), 6.33 (d,
J¼16.0 Hz, 1H), 2.38 (s, 3H), 1.54 (s, 9H).
4.6.7. tert-Butyl
(E)-3-(naphthalen-1-yl)acrylate. White
solid,
139 mg, 0.55 mmol, 91%; 1H NMR (400 MHz, CDCl3)
d
(ppm): 8.45
(d, J¼15.6 Hz, 1H), 8.21 (d, J¼8.4 Hz, 1H), 7.88 (d, J¼8.4 Hz, 2H), 7.75
(d, J¼7.2, 1H), 7.62e7.42 (m, 3H), 6.47 (d, J¼15.6 Hz, 1H), 1.60 (s, 9H).
4.6.8. (E)-1-Methoxy-4-styrylbenzene. White solid, 103 mg,
16. Ju, C.; Bohne, C. J. Phys. Chem. 1996, 100, 3847e3854.
17. O’Connor, C. J.; Wallace, R. G. Adv. Colloid Interface Sci. 1985, 22, 1e111.
0.49 mmol, 82%; 1H NMR (400 MHz, CDCl3)
d
(ppm): 7.50 (d,

8270
G.M. Roberts et al. / Tetrahedron 71 (2015) 8263e8270
18. Small, D. M. In The Bile Acids; Chemistry, Physiology, and Metabolism; Nair, P. P.,
Kritchevsky, D., Setchell, K. D. R., Eds.; Plenum: New York, 1971; pp 249e356.
19. Blickenstaff, R. T.; Orwig, B. J. Org. Chem. 1969, 34, 1377e1381.
30. For two reviews, see: (a) Miyata, M.; Sada, K. In Comprehensive Supramolecular
€
Chemistry; Atwood, J. L., Davis, J. E. D., MacNicol, D. D., Vogtle, F., Eds.; Elsevier:
Oxford, 1996; Vol. 6, Chapter 6; (b) Miyata, M., Sada K., Yoswathananont N., In:
L. Atwood N., Steed J. W. (Eds.), Encyclopedia of Supramolecular Chemistry,
Marcel Dekker, New York, p 441e451.
20. Wolf, G. C.; Foster, E. L.; Blickenstaff, R. T. J. Org. Chem. 1973, 38, 1276e1279.
21. Fieser, L. F.; Rajagopalan, S. J. Am. Chem. Soc. 1950, 72, 5530e5536.
22. Davis, A. P.; Dresen, S.; Lawless, L. J. Tetrahedron Lett. 1997, 38, 4305e4308.
31. Beletskaya, I. P.; Cheprakov, A. V. Chem. Rev. 2000, 100, 3009e3066.
32. DeVasher, R. B.; Moore, L. R.; Shaughnessy, K. H. J. Org. Chem. 2004, 69,
7919e7927.
ꢀ ꢀ
23. Rentsch, D.; Hany, R.; Barthelemy, S.; Steinauer, R. Tetrahedron Lett. 2003, 44,
6987e6990.
24. Oakenfull, D. G.; Fisher, L. R. J. Phys. Chem. 1977, 81, 1838e1841.
25. Amatore, C.; Carre, E.; Jutand, A.; M’Barki, M. A.; Meyer, G. Organometallics
1995, 14, 5605e5614.
26. Amatore, C.; Jutand, A.; M’Barki, M. A. Organometallics 1992, 11, 3009e3013.
27. Amatore, C.; Carre, E.; Jutand, A.; M’Barki, M. A. Organometallics 1995, 14,
1818e1826.
33. Moore, L. R.; Shaughnessy, K. H. Org. Lett. 2004, 6, 225e228.
34. Jagtap, S.; Deshpande, R. Tetrahedron Lett. 2013, 54, 2733e2736.
35. Bistri, O.; Reinaud, O. Org. Biomol. Chem. 2015, 13, 2849e2865.
36. Senra, J. D.; Malta, L. F. B.; Souza, A. L. F.; Aguiar, L. C. S.; Antunes, O. A. C. Adv.
Synth. Catal. 2008, 350, 2551e2558.
37. Zhao, Y.; Zhong, Z. J. Am. Chem. Soc. 2005, 127, 17894e17901.
ꢀ
ꢀ
28. Csakai, Z.; Skoda-Foldes, R.; Kollar, L. Inorg. Chim. Acta 1999, 286, 93e97.
29. In the presence of NEt3, heat and water, the catalyst visibly formed palladium
black within 24 h and fully decomposed within 48 h.