Polyethylene as a nonvolatile solid cosolvent phase for catalyst separation and recovery

Article abstract of DOI:10.1021/ja306719j

The studies described here show that a relatively low molecular weight, narrow polydispersity polyethylene (PE) wax (Polywax) can serve as a nontoxic and nonvolatile alternative to alkane solvents in monophasic catalytic organic reactions where catalysts

Full text of DOI:10.1021/ja306719j

Communication
pubs.acs.org/JACS
Polyethylene as a Nonvolatile Solid Cosolvent Phase for Catalyst
Separation and Recovery
Yanfei Yang, Nilusha Priyadarshani, Tatyana Khamatnurova, Jakkrit Suriboot, and David E. Bergbreiter*
Department of Chemistry, Texas A&M University, College Station, Texas 77842-3012, United States
S
* Supporting Information
hydrocarbon polymers can be used in simpler ways as alkane
ABSTRACT: The studies described here show that a
relatively low molecular weight, narrow polydispersity
polyethylene (PE) wax (Polywax) can serve as a nontoxic
and nonvolatile alternative to alkane solvents in mono-
phasic catalytic organic reactions where catalysts and
products are separated under biphasic conditions. In this
application, a polymer that is a solid at room temperature
substitutes for a conventional alkane solvent at ca. 80 °C.
In addition to the advantages of being a nonvolatile,
nontoxic, reusable solvent, this hydrocarbon polymer
solvent, like heptane, can sequester nonpolar soluble
polymer-bound catalysts after a reaction and separate them
from products. The extent of this separation and its
generality were studied using polyisobutylene (PIB)- and
poly(4-dodecylstyrene)-bound dyes and PE-bound Pd
allylic substitution catalysts, PIB-bound Pd cross-coupling
catalysts, and PE- and PIB-bound metathesis catalysts.
Catalytic reactions were effected using single-phase
reaction mixtures containing Polywax with toluene, THF,
or THF/DMF at ca. 80 °C. These solutions either separate
into two liquid phases on addition of a perturbing agent or
separate as a solid/liquid mixture on cooling. The
hydrocarbon polymer-bound dyes or catalysts either
separate into the hot liquid Polywax phase or coprecipitate
with Polywax and are subsequently isolated as a non-
volatile Polywax solid phase that contains the dye or the
recyclable catalyst.
solvent replacements and can be recovered and separated from
products using a biphasic separation. These approaches are
modeled after other approaches where nonvolatile or nontoxic
solvent phases such as ionic liquids, water, or carbon dioxide
are used to recycle catalysts with biphasic separations and
minimal use of volatile organic solvents.
Our group has also used biphasic thermomorphic and latent
biphasic separations with alkane solvents and alkane phase-
selectively soluble polymer-bound catalysts and reagents.15,16
For example, we have shown that miscible solutions of heptane
and ethanol can be perturbed by a perturbing agent (e.g., a
change in temperature or physical addition of small amounts of
water or salt) to form two phases. After this perturbation,
nonpolar polymer-bound ligands, catalysts, and reagents that
were soluble in the original miscible solvent mixture selectively
partition into the heptane-rich phase, with products remaining
in the polar solvent phase. For example, the nonpolar
polyisobutylene (PIB)-bound azo dye 117 and the nonpolar
poly(4-dodecylstyrene)-bound fluorophore 2 are surrogates of
polymer-bound catalysts and are both phase-selectively soluble
in a heptane-rich phase after a water- or temperature-induced
perturbation of a mixture of heptane and aqueous ethanol or
heptane and DMF, respectively. This is illustrated visually for 1
in Figure 1.
These latent biphasic and thermomorphic separations are
general and are useful ways to separate nonpolar soluble
polymer-bound catalysts from polar solutions of products.15
However, these strategies use heptane. We reasoned that we
eplacement of organic solvents with benign media is a
1−3
Rmajor problem in effecting “greener” reactions.
This
interest is evident in the many studies using media such as
water, supercritical carbon dioxide (scCO2), or ionic liquids as
alternative solvents or cosolvents.4−8 Here we describe our
initial studies that illustrate the potential of using inexpensive
nonpolar low-melting polymers as solvents. The studies here
show that relatively low molecular weight polyethylene (PE)
waxes can take the place of volatile alkane solvents, both
reducing use of volatile organic solvents and serving as a vehicle
for catalyst recovery/reuse.
Although polymers such as poly(ethylene glycol) (PEG)
have been used as solvents,9−11 the use of polymers as solvents
is relatively uncommon. There are only limited examples
suggesting other polymers’ use.12−14 Moreover, in the cases
where PEG has been used, the products are entrained in this
polymer, leading to a need to extract the products from PEG
with added organic solvent or with scCO2. As we show here,
Figure 1. Photograph of reversible thermomorphic phase separation of
1 in a heptane/EtOH/H2O mixture or irreversible phase perturbation
after addition of an additional 10 vol% of H2O.
Received: July 10, 2012
Published: August 28, 2012
© 2012 American Chemical Society
14714
dx.doi.org/10.1021/ja306719j | J. Am. Chem. Soc. 2012, 134, 14714−14717

Journal of the American Chemical Society
Communication
might be able to improve this strategy further if we could
replace all or part of the heptane solvent with a nonvolatile,
nontoxic, and fully recyclable solvent. This proved to be the
case using Polywax 400 (3), a commercially available low-
molecular-weight unfunctionalized PE oligomer with a
polydispersity index (PDI) of 1.08.18 We first tested the
efficiency of 3 as a solvent phase by dissolving 9.0 mg of 1 in 4
mL of a THF/DMF mixture (1/1 vol/vol). Next, 1.5 g of the
solid solvent Polywax 3 was added to this homogeneous red
solution. At 25 °C, 3 remained suspended in the solution.
When 3 was removed from this suspension, it was visually
evident that no 1 was sorbed. On heating to 80 °C, 3 dissolved
to form a low-viscosity homogeneous red solution of THF,
DMF, 1, and 3. When this miscible solution was perturbed by
addition of 10−20 vol% water, a liquid/liquid biphasic mixture
formed where the red dye 3 partitioned quantitatively into the
hot liquid Polywax phase. Cooling this mixture formed a red
solid wax that could be physically separated from the organic
solvent phase (Figure 2). UV−visible spectroscopy analysis
showed that <0.1% of the starting dye 1 was in solution.
THF/H2O. Cooling the biphasic solution formed a solid wax
phase that could be physically separated from the THF-rich
phase. Fluorescence analysis showed that <0.01% of 2 remained
in the THF-rich phase.
The miscibility of Polywax in hot organic cosolvents and the
efficient separations seen with nonpolar polymer-bound dyes
after a biphasic separation suggested that Polywax could
substitute for heptane as a hydrocarbon phase for recovery
and reuse of catalysts bound to hydrocarbon polymers. This
was shown to be the case using three different types of catalysts.
The first of these reactions used PEOlig-bound diarylphosphine-
ligated Pd(0) complexes that had previously been used to
prepare PEOlig-bound recyclable analogues of oxygen-sensitive
Pd catalysts such as Pd(P(PPh3))4.20 In this work, a phosphine-
ligated Pd(0) molecular catalyst 8 was prepared using PEOlig
-
PPh2 ligands derived from commercially available functional PE
oligomers that others had used in formation of PEOlig-bound
porphyrin polymerization catalysts.21 The PEOlig-PPh2-bound
Pd(0) catalyst prepared in Scheme 2 was then used as a
Scheme 2. Synthesis of a Polyethylene-Bound Pd(0) Catalyst
Figure 2. Polywax, THF, and DMF at 80 °C form a solution with the
dye 1. Subsequent perturbation of this solution produces a biphasic
mixture consisting of a Polywax phase and a THF/DMF phase. This
nonpolar soluble polymer-bound dye (as well as fluorophore 2)
partitions completely into the hydrocarbon phase which solidifies on
cooling. Separation and drying leads to a 1/Polywax mixture which
redissolves in THF on heating.
recyclable and reusable analogue of tetrakis-
(triphenylphosphine) Pd(0) in a carefully degassed THF/
Polywax system at 85 °C in the allylic substitution reaction
shown in Scheme 3.
Scheme 3. Substitution of an Allyl Acetate in THF/Polywax
Using 0.4 mol% of 8 as a Catalyst
A second experiment showed that sequestration of nonpolar
polymer-bound species in Polywax is general for other nonpolar
hydrocarbon-like polymers if the separation process includes a
stage where two liquid phases are present. This was shown in
an experiment with the alkane-soluble poly(4-dodecylstyrene)-
bound dansyl dye 2. In this experiment, a dansyl-labeled
poly(4-dodecylstyrene) 2 was prepared as shown in Scheme 1.
Formation of a Pd(0) catalyst that could be recycled was
accomplished by reaction of excess PEOlig-PPh2 with Pd2(dba)3
Scheme 1. Synthesis of Dansyl-Labeled Poly(4-
dodecylstyrene) 2
in THF at elevated temperature or by reaction of excess PEOlig
-
PPh2 with Pd(PPh3)4. In either case, the (PEOlig-PPh2)4Pd
complex was collected by cooling the mixture and rinsing
several times with degassed THF at room temperature. In the
latter case, several dissolution (heating)/precipitation (cooling)
cycles were carried out to ensure complete exchange of the
PPh3 ligand.
Because of the air sensitivity of the Pd(PEOlig-PPh2)4 catalyst,
the allylic substitution reaction was performed in a septum-
stoppered centrifuge tube under Ar using a mixture of 5 mL of
THF and 1.5 g of Polywax that was degassed before heating.
On heating to 85 °C, a low-viscosity homogeneous orange
solution of THF, Polywax, and the Pd catalyst formed. The
reaction was allowed to proceed overnight, after which a small
portion of water was added. This perturbed the homogeneous
reaction, producing a viscous less dense Polywax phase and a
more dense THF-rich phase. This mixture was cooled and
centrifuged for 20 min to separate the solid Polywax solvent
phase. The THF phase containing the product was transferred
When a ca. 5 mN solution of 2 was prepared in Polywax and
toluene at 70 °C and this homogeneous solution was simply
cooled, the PE precipitated but did not entrap 2. However,
when a solution of 2 was prepared in Polywax and THF (1.5 g
of Polywax, 5 mL of THF, 70 °C), perturbing the solution by
addition of water generated a biphasic mixture of Polywax and
14715
dx.doi.org/10.1021/ja306719j | J. Am. Chem. Soc. 2012, 134, 14714−14717

Journal of the American Chemical Society
Communication
by forced siphon under Ar using a cannula. The waxy mixture
left in the tube was rinsed with one portion (10 mL) of
degassed THF and dried under reduced pressure for 30 min
before more THF and substrates were added. Heating then
regenerated the solution of substrates and catalyst for a
subsequent reaction cycle. When care was taken to avoid air
exposure, the Polywax-containing Pd(0) phase was successfully
recycled four times, with complete conversion of cinnamyl
acetate to the cinnamylamine (average isolated yields of 83%/
cycle). Pd leaching in the THF phase (using ICP-MS) after
cycle 3 or 4 averaged 0.2% of the starting Pd.
While solid Polywax-facilitated separation of 8 from the
product allowed us to recycle 8, Polywax did not protect 8 from
air. Exposure of the solid Polywax phase containing 8 to air
after the fifth cycle oxidized the phosphine ligands, leading to
an incomplete reaction in the sixth reaction cycle. 31P NMR
spectroscopy confirmed that a phosphine oxide formed from
PEOlig-PPh2 in this case, presumably because the polyethylene
wax matrix, like polyethylene, is permeable to oxygen.22
Supported Pd species have been shown to be effective in
cross-coupling chemistry as precatalysts that form some
unspecified Pd(0) colloidal catalysts.23−25 Since we had
previously shown that polyolefin-bound SCS-Pd(II) pincer
species serve in this manner, we examined Polywax as a
medium for recycling the Pd catalyst formed from 9 in a Heck
reaction. In this study, a PIB-bound SCS-Pd complex was
Figure 3. Photographs of (a) Polywax and (b) 10/Polywax (30 mg of
10/1 g of Polywax). (c) Schematic drawing of the greenish 80 °C
solution of 10, Polywax, and THF that forms a biphasic mixture
consisting of a colorless THF phase and a greenish phase of hot liquid
Polywax containing 10 on perturbation of the hot solution by addition
of 20 vol% water. Cooling solidifies the Polywax phase, which on
physical separation from the THF solution re-forms a solid wax phase
like that in photograph (b).
powder redissolved to form a green toluene solution similar to
the initial solution of 10/toluene/Polywax. A Polywax-
entrapped form of 11 was prepared by dissolving 30 mg of
11 and 1 g of Polywax in hot THF. Perturbation of this mixture
with water produced a biphasic solution, and cooling led to an
isolable dispersion of 11/Polywax that was similar in color to
the 10/Polywax solid.
We first tried an RCM reaction using a solvent mixture that
contained 60% Polywax. Although this RCM reaction worked,
cooling this hot toluene mixture led to a gel of Polywax and
toluene that left little or no free toluene solvent, making it
difficult to isolate the product. Thus, recycling was effected by
first cooling the reaction mixture and then washing the gel with
a minimum amount of toluene. As expected, conversion in this
RCM reaction was complete. Centrifugation separated the 10/
Polywax from the toluene solution of the product. Subsequent
reaction cycles involved addition of a fresh substrate to the
recovered 10/Polywax gel followed by heating. Five cycles
afforded an average isolated yield of product of 86%/cycle with
0.19% Ru leaching in cycle 4, as measured by ICP-MS.
The Ru complex 11 in a hot solution of THF/Polywax also
formed a competent catalyst for this same RCM reaction.
However, recycling of this catalyst required perturbing the hot
(80 °C) reaction mixture with water to generate a biphasic
mixture in which the PIB-bound Ru catalyst that formed from
11 partitioned into a hot Polywax phase. While we achieved
complete conversion of diethyl diallylmalonate to a cyclo-
pentene derivative through three cycles, the reaction time
increased in cycle 3, and yields decreased in subsequent cycles.
Prior work has shown that the active methylidene species
formed in RCM reactions of dienes can react with water.29−31
While >70% of a relatively stable Grubbs−Hoveyda complex is
recoverable after a week’s exposure to water at ambient
temperature,31 it is not surprising that water addition could
decompose the active Ru catalysts formed from 11 at 80 °C. In
our case, partitioning of the PIB-bound Ru complex or
unreacted 11 into a Polywax phase may partly protect the Ru
species from decomposition. However, traces of water from
earlier cycles and/or the 80 °C temperature required for the
water-induced biphasic formation make this scheme for
recovery of 11/Polywax unsuitable. Other perturbation agents
presumably could be developed for catalysts that react with
synthesized according to a literature procedure.26 We then
examined coupling between iodoacetophenone and n-butyla-
crylate to evaluate Pd recycling with Polywax as a hydrocarbon
phase. The reaction was conducted in a THF/DMF/Polywax
(3/3/2 by wt) mixture using 2 mol% of SCS-Pd as the
precatalyst at 85 °C in a capped vial. These reactions were
allowed to run overnight because taking aliquots was
experimentally inconvenientremoving an aliquot cooled the
solution and produced a solid suspension. After an overnight
reaction, 15 vol% of water was added, forming a biphasic
mixture consisting of a THF/DMF/H2O phase and a hot
viscous Polywax phase. Cooling this biphasic mixture produced
a solid Polywax phase and a solution of the product. The solid
wax phase that contained the PIB-bound SCS ligand, any
unreacted precatalyst, and the Pd(0) species that we believe is
formed from this SCS-Pd complex was separated by suction
filtration, and this solid could be used to carry out another
reaction cycle. Complete conversion of substrate to product
based on NMR spectroscopy was seen for eight cycles using the
recovered Polywax phase. The average isolated yield is 92% per
cycle. Pd leaching was 0.72 and 0.66% of the starting Pd based
on ICP-MS analysis of the product solution from cycles 3 and
4, respectively.
Finally, we examined the suitability of Polywax as a cosolvent
in ring-closing metathesis (RCM) catalysis reactions using
PEOlig- and PIB-NHC-supported Ru catalysts 10 and 11.27,28
We first showed that we could entrap both 10 and 11 in
Polywax. In the case of 10, this involved preparing a solution of
30 mg of 10 and 1 g of Polywax in toluene and cooling the
mixture to form the greenish powder shown in Figure 3b. This
14716
dx.doi.org/10.1021/ja306719j | J. Am. Chem. Soc. 2012, 134, 14714−14717

Journal of the American Chemical Society
Communication
(13) Molnar, L. K.; Buchwald, S. L.; Hatton, T. A. Polym. Mater. Sci.
Eng. 1996, 75, 113−114.
water for reactions that use Polywax as a cosolvent with PIB-
bound catalysts.
(14) van den Ancker, T. R.; Cave, G. W. V.; Raston, C. L. Green
Chem. 2006, 8, 50−53.
In summary, a relatively monodisperse low-molecular-weight
polyethylene oligomer is a suitable nonvolatile and recoverable
alternative to heptane. This solid wax serves as a polymeric
solvent or cosolvent that is a direct substitute for heptane in
recovering catalysts bound to both PIB and PE oligomer-bound
ligands as solids at room temperature in biphasic separations at
the end of a reaction. This hydrocarbon polymer solvent
substitute is also a practical cosolvent for selective recovery of
species on other suitably designed nonpolar polymers such as
poly(4-dodecylstyrene). This oligomeric polyethylene solvent
should have low toxicity and low volatility, like polyethylene.
Presumably other polymers with low polydispersity, low
molecular weight, and low viscosity as melts or in solution
can also be used as media to partially substitute or fully
substitute for other conventional volatile organic solvents.
Studies of these possibilities are ongoing.
(15) Bergbreiter, D. E.; Tian, J.; Hongfa, C. Chem. Rev. 2009, 109,
530−582.
(16) Recoverable and Recyclable Catalysts; Benaglia, M., Ed.; John
Wiley & Sons: Chichester, UK, 2009.
(17) Bergbreiter, D. E.; Priyadarshani, N. J. Polym. Sci., Part A: Polym.
Chem. 2011, 49, 1772−1783.
(18) Baker Hughes Inc. Polywax, http://www.bakerhughes.com/
news-and-media/resources/technical-data-sheet/polywax-
polyethylenes (accessed July 25, 2012).
(19) Bergbreiter, D. E.; Chen, Z.; Hu, H.-P. Macromolecules 1984, 17,
2111−2116.
(20) Bergbreiter, D. E.; Weatherford, D. A. J. Org. Chem. 1989, 54,
2726−2730.
(21) Older, C. M.; Dristjansdottir, S.; Ritter, J. C.; Tam, W.; Grady,
M. C. Chem. Ind. 2009, 123, 319−328.
(22) Kanehashi, S.; Kusakabe, A.; Sato, S.; Nagai, K. J. Membr. Sci.
2012, 365, 40−51.
ASSOCIATED CONTENT
* Supporting Information
(23) Bergbreiter, D. E.; Osburn, P. L.; Frels, J. D. Adv. Synth. Catal.
2005, 347, 172−184.
■
S
(24) Yu, K.; Sommer, W.; Richardson, J. M.; Weck, M.; Jones, C. W.
Adv. Synth. Catal. 2005, 347, 161−171.
Experimental procedures for the synthesis of copolymer 2;
description of the fluorescence experiments; and details of the
catalytic reactions, recycling, and ICP-MS analyses. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(25) Crabtree, R. H. Chem. Rev. 2012, 112, 1536−1554.
(26) Bergbreiter, D. E.; Li, J. Chem. Commun. 2004, 42−43.
(27) Hobbs, C.; Yang, Y.-C.; Ling, J.; Nicola, S.; Su, H.-L.; Bazzi, H.
S.; Bergbreiter, D. E. Org. Lett. 2011, 13, 3904−3907.
(28) Hongfa, C.; Su, H.-L.; Bergbreiter, D. E.; Hassan, B. S. Org. Lett.
2009, 11, 665−667.
AUTHOR INFORMATION
Corresponding Author
bergbreiter@tamu.edu
■
(29) Kirkland, T. A.; Lynn, D. M.; Grubbs, R. H. J. Org. Chem. 1998,
63, 9904−9909.
(30) Burtscher, D.; Grela, K. Angew. Chem., Int. Ed. 2009, 48, 442−
454.
Notes
(31) Hong, S. H.; Grubbs, R. H. J. Am. Chem. Soc. 2006, 128, 3508−
3509.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The generous gift of samples of Polywax and end-functionalized
polyethylene oligomers by Dr. Paul K. Hanna of Baker-Hughes,
assistance in ICP-MS analyses by Dr. Yun-Chin Yang, and
support by the National Science Foundation (CHE-0952134)
and the Robert A. Welch Foundation (Grant A-0639) are
gratefully acknowledged.
REFERENCES
■
(1) Anastas, P. T. ACS Symp. Ser. 2002, 819, 1−9.
(2) Henderson, R. K.; Jimenez-Gonzalez, C.; Constable, D. J. C.;
Alston, S. R.; Inglis, G. G. A.; Fisher, G.; Sherwood, J.; Binks, S. P.;
Curzons, A. D. Green Chem. 2011, 13, 854−862.
(3) Clark, J. H.; Tavener, S. J. Org. Process Res. Dev. 2007, 11, 149−
155.
(4) Pollet, P.; Hart, R. J.; Eckert, C. A.; Liotta, C. L. Acc. Chem. Res.
2010, 43, 1237−1245.
(5) Jessop, P. G.; Subramaniam, B. Chem. Rev. 2007, 107, 2666−
2694.
(6) Jessop, P. G. Green Chem. 2011, 13, 1391−1398.
(7) Hallett, J. P.; Welton, T. Chem. Rev. 2011, 111, 322−353.
(8) Hagiwara, H. Synlett 2012, 23, 837−850.
(9) Heldebrant, D. J.; Jessop, P. G. J. Am. Chem. Soc. 2003, 125,
5600−5601.
(10) Chen, J.; Spear, S. K.; Huddleston, J. G.; Rogers, R. D. Green
Chem. 2005, 7, 64−82.
(11) Karnakar, K.; Narayana, M. S.; Ramesh, K.; Satish, G.;
Nanubolu, J. B.; Nageswar, Y. V. D. Tetrahedron Lett. 2012, 53,
2897−2903.
(12) Heldebrant, D. J.; Witt, H. N.; Walsh, S. M.; Ellis, T.; Rauscher,
J.; Jessop, P. G. Green Chem. 2006, 8, 807−815.
14717
dx.doi.org/10.1021/ja306719j | J. Am. Chem. Soc. 2012, 134, 14714−14717