Polyolefin-supported recoverable/reusable Cr(III)-salen catalysts

Article abstract of DOI:10.1021/jo102044m

The design of functional soluble polyolefins for use as supports for salen ligands and metal complexes is described. Examples and applications that use both polyisobutylene (PIB)-and polyethylene (PE_Olig)-bound recoverable/recyclable salen ligands/metal complexes are detailed. In the case of using PIB as a support, the polymer-bound complexes can be recovered through the use of latent biphasic or a thermomorphic mixed solvent systems. In the case of PE_Olig-supported complexes, the thermomorphic PE _Olig-bound salen species can be dissolved in "hot" solvents and quantitatively recovered as solids upon cooling to room temperature. Both the PIB-and PE_Olig-bound salen catalysts were shown to catalyze the ring-opening of epoxides with various nucleophiles. Both sorts of polyolefin-bound catalysts can be recycled and reused with no observed loss in activity. However, limitations of catalyst concentration make chiral versions of these complexes uncompetitive in comparison to conventional chiral salen catalysts that can be used in neat substrate at higher concentration to produce high enantioselectivity in the ring-opening products. The preparation of a PIB-bound "half-salen" catalyst was also briefly examined.

Full text of DOI:10.1021/jo102044m

pubs.acs.org/joc
Polyolefin-Supported Recoverable/Reusable Cr(III)-Salen Catalysts
David E. Bergbreiter,* Christopher Hobbs, and Chayanant Hongfa
Department of Chemistry, Texas A&M University, P.O. Box 30012, College Station,
Texas 77842-3012, United States
bergbreiter@chem.tamu.edu
Received October 14, 2010
The design of functional soluble polyolefins for use as supports for salen ligands and metal complexes is
described. Examples and applications that use both polyisobutylene (PIB)- and polyethylene (PEOlig)-
bound recoverable/recyclable salen ligands/metal complexes are detailed. In the case of using PIB as a
support, the polymer-bound complexes can be recovered through the use of latent biphasic or a
thermomorphic mixed solvent systems. In the case of PEOlig-supported complexes, the thermomorphic
PEOlig-bound salen species can be dissolved in “hot” solvents and quantitatively recovered as solids upon
cooling to room temperature. Both the PIB- and PEOlig-bound salen catalysts were shown to catalyze the
ring-opening of epoxides with various nucleophiles. Both sorts of polyolefin-bound catalysts can be
recycled and reused with no observed loss in activity. However, limitations of catalyst concentration make
chiral versions of these complexes uncompetitive in comparison to conventional chiral salen catalysts that
can be used in neat substrate at higher concentration to produce high enantioselectivity in the ring-opening
products. The preparation of a PIB-bound “half-salen” catalyst was also briefly examined.
Introduction
by their description in the literature as members of a class of
“privileged” ligands.3
Salen ligands and the transition metal complexes that they
form are widely used in catalysis. They can be used in a
variety of reactions to convert simple organic starting mate-
rials into useful products. Their use in ring-opening reactions
of epoxides by nucleophiles is especially common,1 and
salen-catalyzed ring-openings are the basis of chemistry that
has led to important new asymmetric syntheses. Similar
epoxide opening chemistry is also the basis of new greener
polymerization chemistry affording polycarbonates from
carbon dioxide and epoxides.2 The broad utility and effec-
tiveness of salen ligands and chiral salen ligands is evidenced
Because of the broad utility of salen transition metal
complexes like 1 and 2 in catalysis, a variety of reactions
and approaches for catalyst and ligand recovery have been
explored.4 Three general strategies to recover, reuse, or
separate the catalyst and product have been used previously.
(3) Yoon, T. P.; Jacobsen, E. N. Science 2003, 299, 1691–1693.
(4) (a) Baleizao, C.; Garcia, H. Chem. Rev. 2006, 106, 3987–4043.
(b) Holbach, M.; Weck, M. J. Org. Chem. 2005, 71, 1825–1836.
(c) Angelino, M. D.; Laibinis, P. E. Macromolecules 1998, 31, 7581–7587.
(d) Pozzi, G.; Shepperson, I. Coord. Chem. Rev. 2003, 242, 115–124. (e) Song,
C. E.; Roh, E. J. Chem. Commun. 2000, 837–838. (f) Bergbreiter, D. E. Chem.
Rev. 2002, 102, 3345–3383. (g) Bergbreiter, D. E.; Tian, J.; Hongfa, C. Chem.
Rev. 2009, 109, 530–582. (h) Cavazzini, M.; Quici, S.; Pozzi, G. Tetrahedron
2002, 58, 3943–3949. (i) Yao, X.; Chen, H.; Lu, W.; Pan, G.; Hu, X.; Zheng,
A. Tetrahedron Lett. 2000, 41, 10267–10271. (j) Dhal, P. K.; De, B. B.;
Sivaram, S. J. Mol. Catal. A: Chem. 2001, 177, 71–87. (k) Madhavan, N.;
Jones, C. W.; Weck, M. Acc. Chem. Res. 2008, 41, 1153–1165.
(1) (a) Gupta, K. C.; Sutar, A. K. Coord. Chem. Rev. 2008, 252, 1420–
1450. (b) McGarrigle, E. M.; Gilheany, D. G. Chem. Rev. 2005, 105, 1563–
1602. (c) Bandini, M.; Cozzi, P. G.; Umani-Ronchi, A. Chem. Commun.
2002, 919–927.
(2) Darensbourg, D. J. Chem. Rev. 2007, 107, 2388–2410.
DOI: 10.1021/jo102044m
r
Published on Web 12/30/2010
J. Org. Chem. 2011, 76, 523–533 523
2010 American Chemical Society

Bergbreiter et al.
JOCArticle
The first strategy uses a support that is insoluble before, during,
and after the reaction.4c These approaches have typically used
cross-linked polymer supports or inorganic solid supports. A
second strategy is to use a biphasic system liquid/liquid system.
Examples of this approach include the use of ionic liquid4e
phase-immobilized catalysts and biphasic fluorous soluble sup-
ports.4h A third strategy has been to use soluble but phase-
separable polymeric supports. This has involved polymers with
pendant salen ligands or polymers that contain salen ligands
that are repeating units in the polymer structure. The most
successful examples of this strategy use salen complexes that are
repeating units of the polymer backbone structure.4k
supports that afford a monophasic mixture of solvents,
substrate, and catalyst during a reaction with a subsequent
polymer-facilitated liquid/liquid catalyst/ligand separation
after the reaction is complete.4f,g,5 The two general schemes
that we have found most effective are shown in parts a and b
of Figure 1a and either involve a thermomorphic phase sepa-
ration where a polymer-bound catalyst and products sepa-
rate into different density liquid phases after a change in
temperature,5a-d or when a latent biphasic mixed solvent
system is perturbed with small amounts of an additive to
become a biphasic mixture.5f,g In either case, the catalyst and
product are separable by a gravity-based separation.
The other general strategy we have developed is based on
the thermomorphic solubility of polymers like polyethy-
lene.5h-k In this case (Figure 1c), a terminally functionalized
polyethylene oligomer or polymer is modified so that it can
bind to a catalyst. Because polyethylene is completely in-
soluble in organic solvents at room temperature but soluble
hot, such catalysts can be used in a heated solution as
homogeneous catalysts but recovered on cooling by a so-
lid/liquid separation. In this case, the product remains in
solution while the catalyst self-separates as a solid.
In the cases where achiral or chiral salen metal complexes
are supported on soluble polymers, the salen ligands and
catalysts are typically separated from products after a reaction
as solids using solvent precipitation.4 This is an effective way to
separate a polymer from other materials but has the disadvan-
tage that it requires relatively large amounts of solvent;a
potential problem on any large-scale application of a supported
catalyst. This paper describes the advantages and limitations of
an alternative approach using polyolefin-bound salen ligands
and salen metal complexes that phase separate from a single
phase reaction either as a separable liquid phase or as solids.
These soluble polymeric species enable the use of common
organic solvents with either a thermomorphic or latent
biphasic separation of a catalyst after a monophasic homo-
geneous reaction with minimal use of solvent. The results
below show that this is an alternative and effective way to
recover and reuse Cr-salen catalysts in epoxide ring-open-
ing reactions. It is also a viable approach to immobilize
catalysts for other salen metal complex mediated reactions.
However, the use of these materials as effective asymmetric
catalysts is frustrated by an inability to achieve sufficiently
high catalyst concentrations.
To use the liquid/liquid separation strategies shown in
Figure 1 in salen complex-catalyzed reactions requires an
appropriate polymer-bound salen ligand. As described here,
salen ligands 7 and 8 can be prepared by using a common
intermediate 6 that was formed in a Friedel-Crafts alkylation,
using the acid-catalyzed reaction of commercially available
polyisobutylene oligomers containing terminal alkene groups6
with 2-tert-butylphenol as shown in Scheme 1.7 The product
4-polyisobutyl-2-tert-butylphenol was converted to a salicylal-
dehyde derivative by an acid-catalyzed alkylation and Canniz-
zaro reaction with use of paraformaldehyde. Subsequent imine
formation then afforded both chiral and achiral salen ligands
from suitable 1,2-diamines. These achiral and chiral ligands
were then converted into metal complexes by using the same
chemistry used in the syntheses of structurally similar low
molecular weight salen metal complexes.7,8
Given the availability of the PIB-bound salicylaldehyde 6,
we briefly examined its use in the synthesis of other salen-like
metal complexes. For example, the intermediate 3-tert-butyl-
5-polyisobutylsalicylaldehyde formed in Scheme 1 was used
to prepare a “half-salen” ligand 12 as shown in Scheme 2.
This ligand, like the salen ligands 7 and 8, could be metalated
by using procedures like those used with low molecular
weight analogues to form Cr complex 13.9
Results and Discussion
Electrophilic substitution of phenol by vinyl-terminated
polyisobutylene has the potential to be a general route to
elaborated salen ligands too. This is shown by the chemistry
in Scheme 3 where a 3-(piperindylmethyl)-5-polyisobutyl-
salen ligand was prepared from 4-polyisobutylphenol via
multiple acid-catalyzed formylations by using chemistry
modeled after the chemistry reported by Kozlowski.10
Past work in our group has emphasized two general
strategies for recycling catalysts that use soluble polymers
as phase handles. The first of these strategies uses polymer
(5) (a) Bergbreiter, D. E. J. Polym. Sci., Part A: Polym. Chem. 2001, 39,
2351–2363. (b) Bergbreiter, D. E.; Osburn, P. L.; Liu, Y.-S. J. Am. Chem.
Soc. 1999, 121, 9531–9538. (c) Bergbreiter, D. E.; Osburn, P. L.; Wilson, A.;
Sink, E. M. J. Am. Chem. Soc. 2000, 122, 9058–9064. (d) Bergbreiter, D. E.;
Osburn, P. L.; Frels, J. D. J. Am. Chem. Soc. 2001, 123, 11105–11106.
(e) Bergbreiter, D. E.; Osburn, P. L.; Smith, T. S.; Li, C.; Frels, J. D. J. Am.
Chem. Soc. 2003, 125, 6254–6260. (f) Bergbreiter, D. E.; Hughes, R.;
Besinaiz, J.; Li, C.; Osburn, P. L. J. Am. Chem. Soc. 2003, 125, 8244–8249.
(g) Bergbreiter, D. E.; Blanton, J. R.; Chandran, R.; Hein, M. D.; Huang,
K. J.; Treadwell, D. R.; Walker, S. A. J. Polym. Sci., Part A: Polym. Chem.
1989, 27, 4205–4206. (h) Bergbreiter, D. E.; Chandran, R. J. Am. Chem. Soc.
1987, 109, 174–179. (i) Bergbreiter, D. E.; Chandran, R. J. Am. Chem. Soc.
1985, 107, 4792–4793. (j) Bergbreiter, D. E.; Weatherford, D. A. J. Org.
Chem. 1989, 54, 2726–2730. (k) Bergbreiter, D. E.; Walker, S. A. J. Org.
Chem. 1989, 54, 5138–5141.
(6) BASF: http://daile.com.cn/BASF/Glissopal%20550%201000%
201300%202300.pdf (accessed Dec 2010).
(7) Hongfa, C.; Tian, J.; Andreatta, J.; Darensbourg, D. J.; Bergbreiter,
D. E. Chem. Commun. 2008, 975–977.
(8) Martinez, L. E.; Leighton, J. L.; Carsten, D. H.; Jacobsen, E. N.
J. Am. Chem. Soc. 1995, 117, 5897–5898.
(9) Dossetter, A. G.; Jamison, T. F.; Jacobsen, E. N. Angew. Chem., Int.
Ed. 1999, 38, 2398–2400.
(10) Fennie, M. W.; DiMauro, E. F.; O’Brien, E. M.; Annamalai, V.;
Kozlowski, M. C. Tetrahedron 2005, 61, 6249–6265.
524 J. Org. Chem. Vol. 76, No. 2, 2011

Bergbreiter et al.
JOCArticle
FIGURE 1. Strategies for separation of polyolefin-supported catalysts from products: (a) thermomorphic liquid/liquid separation, (b) latent
biphasic liquid/liquid separation, and (c) thermomorphic solid/liquid separations.
SCHEME 1. Syntheses of PIB-Supported Salen Metal Complexes 9 and 10
SCHEME 2. Synthesis of a PIB-Supported Half-Salen Metal Complex 13
The second strategy for separation of polyolefin-bound
catalysts and products is based on the thermomorphic
properties of polymers like polyethylene that are soluble
hot and insoluble cold. In this case, we prepared salen ligands
and salen metal complexes using terminally functionalized
thermomorphic polyethylene oligomers as supports. These
studies stem from our prior successes with polyethylene
oligomer (PEolig)-bound catalysts5g-k and were prompted
by the success of a DuPont group that used commercially
available terminally functionalized PE oligomers in the
synthesis of PEOlig-bound porphyrin and phthalocyanine
metal complexes.11
To prepare PEOlig-bound salen ligands we used a regiose-
lective Mitsunobu reaction of the less hindered hydroxyl
group of 3-tert-butyl-5-hydroxysalicylaldehyde 20 with the
(11) Older, C. M.; Kristjansdottir, S.; Ritter, J. C.; Tam, W.; Grady,
M. C. Chem. Ind. 2009, 123, 319–328.
J. Org. Chem. Vol. 76, No. 2, 2011 525

Bergbreiter et al.
JOCArticle
SCHEME 3. Synthesis of 3-(Piperindylmethyl)-5-polyisobutyl-Salen Ligands
SCHEME 4. Synthesis of the PEOlig-Bound Salen Metal Complex 24
terminal hydroxyl group of a commercially available hydro-
xyl-terminated polyethylene oligomer12 to prepare the salen
ligand precursor 22. In these cases, the conditions for diimine
formation that we used to form 7 from the PIB-substituted
salicylaldehyde 6 led to formation of the diimine 23a
contaminated with 5-10% of the monoimine result-
ing from incomplete imine formation from the diamine.
Quantitative formation of 23a from a diamine was possi-
ble by using 10 mol % excess of the PEOlig-bound salicy-
laldehyde. However, in this case, we could not easily
separate the unreacted salicylaldehyde 22 from 23a. Any
mixtures of mono- and diimines occasionally seen in
syntheses of a PIB-bound species could be separated by
chromatography because of the high solubility of the PIB
support. Chromatographic separation is not possible with
polyethylene oligomeric ligands as the insolubility of the
PE oligomers at room temperature makes chromato-
graphic separation of these two products impractical. To
obtain PEOlig-bound product free of the salicylaldehyde or
monoimine impurity we chose to use a slight excess of the
diamine and to convert the 9:1 mixture of di- and mono-
imines that formed into a phase separable PEOlig-bound
salen ligand by reaction of the mixture of di- and mono-
imines with excess 3,5-di-tert-butylsalicylaldehyde. This in
effect produces a more polydisperse but equally thermo-
morphic PEOlig-bound salen ligand 23 as a mixture of 23a
and 23b. Metalation of this product mixture in the DMF-
toluene mixture used in metalation of PEOlig-bound por-
phyrins and phthlalocyanines11 then afforded a PEOlig
-
bound salen-Cr complex 24 (Scheme 4), which was isolated
as a dark brown solid.
The properties of polyolefins like polyisobutylene and
polyethylene both have advantages and impose some limita-
tions in these syntheses. First, PIB’s solubility makes it
feasible to separate PIB-containing intermediates from more
polar low molecular weight starting materials or byproducts
by a simple solvent extraction. This makes it possible to use
excess reagents in ligand syntheses and to separate unreacted
reagents from the PIB products by an extraction without a
chromatographic separation. Likewise, PEOlig-bound spe-
cies separate from excess reagents as a solid. Second, both the
(12) Baker Hughes: http://www.bakerhughes.com/assets/media/techni-
caldatasheets/ 4c97b890fa7e1c37f100001b/file/28730_unilin-alcohol-sheet-
2-5-10.pdf.pdf&fs=288204 (accessed Dec 2010).
526 J. Org. Chem. Vol. 76, No. 2, 2011

Bergbreiter et al.
JOCArticle
FIGURE 2. Thermomorphic phase separation of polyolefin-supported salen metal complexes: (a) complex 9 in a miscible mixture of heptane
and ethanol, (b) complex 9 separated into the heptane-rich phase after the addition of 10 vol % water perturbed this miscible solvent mixture,
(c) a monophasic solution of complex 24 in toluene at 80 °C, and (d) the biphasic mixture of toluene and of the PEOlig-bound salen complex
24 formed after cooling to room temperature and centrifugation.
PIB- and PEOlig-bound intermediates can be readily analyzed
by solution-state NMR spectroscopy.5g-k,7,11,13 However,
there are also limitations associated with the use of poly-
olefin supports. For example, neither the PIB nor the PE
supports can be used in a polar solvent like ethanol to form
the salen ligands from a 1,2-diamine and the salicylalde-
hyde precursor because of their insolubility in ethanol.
Instead, mixed solvent systems (e.g., heptane-ethanol)
must be used. Finally, as was true in the synthesis of salen
ligands from PE oligomers, it can be difficult to separate
polymer-bound byproduct from the desired polymer pro-
duct. This is not a problem in Schemes 1 or 2 where
derivatives of the very soluble PIB support can often be
separated from one another by column chromatography
and methods to address this problem for thermomorphic
PE oligomer supports can be devised as described above.
Nonetheless, byproduct separation remains a possible
limitation for syntheses of other polyolefin-bound ligands.
As noted above, polyolefin supports like PIB and PE have
precedent in separation, recovery, and reuse of a variety of
catalysts.4f,g Past work has shown that PIB is effective at
making the Ru alkylidene complexes phase selectively solu-
ble in the heptane phase of biphasic mixtures of heptane and
polar organic solvents to facilitate recovery, separation, and
reuse of ring-closing metathesis catalysts.14 PIB was equally
effective in effecting phase separation of salen complexes like
9 or 10. This is visually apparent by comparison of the
photographs in Figure 2a,b. In this example, a highly colored
single phase mixture of absolute ethanol and heptane con-
taining 9 (Figure 2a) is perturbed into a biphasic solvent
mixture (Figure 2b) by the addition of 10 vol % water.
Similar efficiency in phase separation of these PIB salen
complexes was also achieved by using 9 in a thermomorphic
mixture of heptane and N,N0-dimethylformamide (DMF).
When a mixture of heptane and DMF was used, the PIB-
bound salen complex 9 was soluble in a hot miscible mixture
of these solvents at 80 °C. This solution was monophasic and
deeply colored like that in Figure 2a. However, in the case of
this DMF/heptane mixture, cooling this deeply colored
monophasic solution produced a biphasic liquid/liquid mix-
ture with the salen complex 9 separating into the less dense
heptane-rich phase. The extent of phase separation was
evaluated by UV-visible spectroscopy. On the basis of the
absorbance of 9 at 350 nm, <0.1% of the salen complex
remained in the polar DMF-rich phase after this thermo-
morphic liquid/liquid separation.
PE is also effective at thermomorphic phase separation of
the salen complex 24. In this case, the PEOlig-salen complex
of Cr forms a dark brown solution on heating to 70 °C.
Cooling this solution produces a water white supernatant
and a dark brown PE solid, chemistry that is analogous to
that reported by a DuPont group that used similar ligands to
thermomorphically separate highly colored PEOlig-bound
Co complexes of porphyrins and phthalocyanines from
solution on cooling.11
The penultimate test of the utility of these salen ligands is
the ability of their metal complexes to effect reactions that
their low molecular weight analogues promote. In prelimin-
ary experiments reported previously, we noted that the
achiral catalyst 9 is as competent as its low molecular weight
analogue in the polymerization of cyclohexene oxide and
CO2.7 These studies showed that 9 is kinetically very similar
to a low molecular weight Cr salen catalyst that had a tert-
butyl group in the 5 position (versus the 5-polyisobutyl group
in 9) in CO2-epoxide polymerization. The fact that this PIB-
bound salen species was also readily separated from the
catalyst and that the polycarbonate product was isolated
with low Cr-contamination (so long as an acid workup that
decomposes the Cr-salen complex was avoided) suggested
that polyolefin-bound salen complexes 9, 10, 13, and 24
should be similarly useful as recyclable separable catalysts
in a variety of ring-opening reactions of low molecular
weight epoxides or in other salen complex-mediated cataly-
sis. This is indeed the case though catalyst concentration
imposes some limitations on this chemistry as noted below.
Our initial studies focused on the utility of the achiral
complex 9 as a recyclable catalyst for the ring-opening of
epoxides with various thiolphenols (eqs 1 and 2).15 These
studies employed a heptane/ethanol solvent mixture and
latent biphasic conditions for recovery and recycling of 9 in
a liquid/liquid separation/recovery system. In this chemistry,
we used a miscible mixture of heptane and ethanol which
dissolved both catalyst 9 and the epoxide and thiophenol
reactants. Ring-opening reactions of epoxides shown in eqs 1
(13) Li, J.; Sung, S.; Tian, J.; Bergbreiter, D. E. Tetrahedron 2005, 61,
12081–1209.
(14) (a) Hongfa, C. H.; Tian, J.; Bazzi, H. S.; Bergbreiter, D. E. Org. Lett.
2007, 9, 3259–3261. (b) Hongfa, C. H.; Su, H.-L.; Bazzi, H. S.; Bergbreiter,
D. E. Org. Lett. 2009, 11, 665–667.
(15) Wu, M. H.; Jacobsen, E. N. J. Org. Chem. 1998, 63, 5252–5254.
J. Org. Chem. Vol. 76, No. 2, 2011 527

Bergbreiter et al.
JOCArticle
TABLE 1. Ring-Opening of Epoxides with Thiols Catalyzed by 9a
TABLE 2. ARO of Cyclohexene Oxide with Azidotrimethylsilane
Catalyzed by 10a
yield (%)b
entry
cycle
yield (%)b
ee (%)c
entry
product
cycle 1
cycle 2
cycle 3
cycle 4
cycle 5
96
1
2
3
4
5
6
1
2
3
4
5
6
42
54
70
79
90
95
10
13
22
14
12
15
1
2
3
4
5
6
28
29
32
33
34
35
76
94
77
34
58
73
84
90
76
42
55
70
84
99
99
81
88
84
95
99
99
93
78
98
99
99
99
aReaction conditions: 5 mol % of catalyst, 7.08 mmol of cyclohexene
oxide, and 7.78 mmol of azidotrimethylsilane stirred under N2 for 12 h in
5 mL of heptane and 5 mL of ethanol. Recycling involved removal of the
polar phase and the addition of fresh ethanol and substrates to the
heptane phase. bThe yield is an isolated yield of pure product after
removal of solvent under reduced pressure. cThe % ee was determined by
chiral HPLC analysis (ChiralCel OD column, hexanes/isopropanol
(8:2)).
aThe reactions were conducted with 2 mmol of epoxide, 2 mmol of
thiol, and 0.02 mmol of 9 in a mixture of 3 mL of heptane and 3 mL of
absolute ethanol. The reaction mixture was stirred under N2 at room
temperature for 24 h. Recycling involved removal of the polar phase and
addition of fresh ethanol and substrates to heptane phase. bThe yield is
based on mass of isolated product.
and 2 were carried out with three different epoxides and two
different aromatic thiols. The results in Table 1 show that the
PIB-bound catalyst 9 was as effective as an electronically
analogous low molecular weight catalyst in this chemistry
and that catalyst recycling and separation from product
was feasible. In these recycling experiments, phase separa-
tion and catalyst recovery was achieved by the addition of
10 vol % water to the homogeneous reaction mixture after
the reaction was complete. This added water produced a
biphasic heptane/aqueous ethanol mixture. Gravity filtra-
tion of the less dense catalyst-containing nonpolar phase
afforded a recyclable solution of the catalyst. The products
of the reactions were isolated by removal of the solvent and
concentration of a low molecular weight catalyst was twice as
large as the kobs for either cycles 2 and 3 of the reaction using
the PIB-supported catalyst 9. There was no change in kobs in
cycles 2 and 3 for this ring-opening reaction using the PIB-
supported catalyst 9 supporting the premise that this catalyst
was recoverable and reusable.
After the discovery that 9 could be used as a recyclable
catalyst for the ring-opening of epoxides with thiols, we
examined the use of complex 10 (Scheme 1) as a catalyst
for the asymmetric ring-opening (ARO) of epoxides.8 We
first tested the use of 10 as a catalyst for reaction of
cyclohexene oxide and azidotrimethylsilane (eq 3). As ex-
pected based on the success of 9 in epoxide-thiol reactions
(vide supra), complex 10 was able to catalyze the opening of
epoxide 25 to afford azido alcohol 36. The PIB-supported
catalyst 10 was successfully recycled six times with use of
latent biphasic conditions to recover the catalyst as shown in
Table 2.
1
characterized by H and 13C NMR spectroscopy and were
>95% pure without the need for any chromatographic
separation of catalyst residues.
However, while 10 proved to be a recyclable catalyst, it
was unfortunately not a very stereoselective catalyst, based
on chiral HPLC analysis of a derivative of 37 formed as
shown in Scheme 5.16 In these cases, analyses of the reac-
tion’s enantioselectivity based on chiral HPLC analysis of
products after reduction and arylation of 36 showed very low
enantioselectivity as noted in Table 2.
The low enantioselectivity for reaction 3 can be explained
by the low concentration of catalyst 10 in the reaction
system. Jacobsen and co-workers have reported the best
enantioselectivities for these Cr-salen catalyzed reactions
are obtained under solvent-free conditions using neat sub-
strate where catalyst concentrations are ca. 0.20 M.8,16,17 In
the case of the PIB-bound catalyst 10, the insolubility of 10 in
substrate required that a solvent be used. Thus, concentra-
tions of 10 in these heptane-ethanol solutions were almost a
factor of 10 lower (0.03 M). Control experiments suggest that
the lower enantioselectivity seen with 10 is mainly due to this
Reuse of the catalyst 9 was affected by simply adding fresh
ethanol and substrates to the recovered heptane-rich phase.
The PIB-supported catalyst could be recycled through at
least 4 cycles with no observed loss in catalytic activity. The
extent of catalyst leaching was also tested using inductively
coupled plasma mass spectroscopy (ICP-MS) analysis of the
product phase. This analysis showed that products from the
reaction between glycidol and 4-methylthiophenol showed
only 0.26% chromium loss per cycle.
We briefly examined the rate of conversion of epoxide to
product using this PIB-bound salen complex. These studies
used similar concentrations of catalyst 1 and 9 and compared
both 1 vs 9 and cycles 2 and 3 for the ring-opening of 31 by the
thiol 26. In these studies, the kobs for reaction using the same
(16) Shaus, S. E.; Larrow, J. F.; Jacobsen, E. N. J. Org. Chem. 1997, 62,
4197–4199.
(17) Hansen, K. B.; Leighton, J. L.; Jacobsen, E. N. J. Am. Chem. Soc.
1996, 118, 10924–10925.
528 J. Org. Chem. Vol. 76, No. 2, 2011

Bergbreiter et al.
JOCArticle
SCHEME 5. Reaction of 36 and 1-Fluoro-2,4-dinitrobenzene To Form 37
SCHEME 6. ARO of 25 with Azidotrimethylsilane Using
Catalyst 2
SCHEME 7. Hetero-Diels-Alder Reaction of Crotonaldehyde
and Ethyl Vinyl Ether Catalyzed by 13
catalyst in the heptane-rich phase and the product in the
polar acetonitrile-rich phase. Even though preliminary trials
of this catalytic system seemed promising (83% yield),
recycling of the polymer-bound catalyst was not feasible.
Diminished yields of the product (ca. 25%) were seen in cycle
1
2. In this case, H NMR analysis showed that the reason
for these diminished yields was decomposition of the cata-
lowered catalyst concentration as opposed to the presence of
the PIB oligomer. Two control experiments with Jacobsen’s
catalyst 2 were performed. First, when the low molecular
weight catalyst 2 was used in place of 10 in reaction 3 at 0.032
M, the enantioselectivity of the reaction was 19%. The 19%
ee seen in Scheme 6 with 0.032 M 2 as a catalyst is very close
to the average 14% ee reported in Table 2. A second
experiment examined the reaction of cyclohexene oxide with
azidotrimethylsilane catalyzed by 5 mol % of 2 in methylene
chloride. When 2 was present at 0.035 M, ring-opened
product was obtained with 33% ee. While these results
suggest a slight solvent effect, these control experiments
suggest that the diminished enantioselectivity seen with 10
is in large part due to the lower catalyst concentration.
Catalyst concentration affects selectivity because of the
importance of the bimolecular reaction mechanism pro-
posed by Jacobsen.17
Given the success of the polyolefin-bound salen complexes
in epoxide ring-opening reactions, we also briefly studied the
use of a PIB-bound “half-salen”9 Cr complex as a catalyst for
other reactions. Low molecular weight analogues of 13 have
previously been used as catalysts for carbonyl-ene reactions,18
Diels-Alder reactions,19 inverse electron demand Diels-Alder
reactions,20 and hetero-Diels-Alder reactions.9 We set out to
investigate the activity of 13 by attempting to catalyze the reac-
tion of crotonaldehyde and ethyl vinyl ether (Scheme 7).
In this study, the PIB-bound catalyst 13 was dissolved in
ethyl vinyl ether and crotonaldehyde was added to the
solution and the reaction mixture was allowed to stir for 2
days. Separation was effected by the removal of solvent and
the addition of heptane and acetonitrile to the reaction flask.
This produced a biphasic mixture with the polymer-bound
˚
lyst due to the 4 A MS present in the reaction mixture.
Because of this, further investigations of this system were not
continued.
We also explored the use of the thermomorphic polyethylene-
supported Cr-salen complex 24 as a recyclable catalyst for the
ring-opening of cyclohexene oxide with azidotrimethylsilane
(eq 4). This chemistry involved the use of thermomorphic solu-
bility of polyethylene and a solid/liquid separation (Figure 1c) to
separate the PEOlig-bound catalyst 24 from product. In the
event, we first dissolved 5 mol % of 24 in 3 mL of toluene at
75 °C under N2. After a homogeneous solution was obtained,
cyclohexene oxide and azidotrimethylsilane were sequentially
added to the reaction flask with a syringe. The reaction mixture
was then stirred 3 h under N2 at 75 °C. Workup and cata-
lyst separation and recovery was then accomplished by simply
cooling the reaction mixture to room temperature. Phase sepa-
ration of solvent containing the product and polymer-bound
catalyst was visually apparent. At this point an additional 5 mL
of toluene was added and the supernatant solution containing
the product was separated from the PEOlig-bound 24. To ensure
quantitative recovery of the product, this isolation procedure
was repeated twice. Recycling of the solid catalyst was achieved
by then dissolving the recovered solid 24 in 3 mL of fresh toluene
at 75 °C and treating this solution with fresh substrate.
As can be seen in Table 3, complex 24 was very successful as a
recyclable separable catalyst in this ring-opening reaction. The
efficiency of catalyst 24 is shown by the recycling data in Table 3.
PEOlig-supported catalyst 24 could be recycled six times with no
loss in catalytic activity. The effectiveness of this process in
minimizing catalyst leaching was also tested by using induc-
tively coupled plasma mass spectroscopy (ICP-MS) analysis.
(18) Grachan, M. L.; Tudge, M. T.; Jacobsen, E. N. Angew. Chem., Int.
Ed. 2008, 47, 1469–1472.
(19) Jarvo, E. R.; Lawrence, B. M.; Jacobsen, E. N. Angew. Chem., Int.
Ed. 2005, 44, 6043–6046.
(20) Gademan, K.; Chavez, D. E.; Jacobsen, E. N. Angew. Chem., Int. Ed.
2002, 41, 3059–3061.
J. Org. Chem. Vol. 76, No. 2, 2011 529

Bergbreiter et al.
JOCArticle
TABLE 3. Ring-Opening of Cyclohexene Oxide with Azidotrimethyl-
mass spectrometer with 2,4,6-trihydroxyacetophenone (THAP) as
matrix.
silane Catalyzed by 24a
yield (%)b
Analysis of Cr(III) Loading on Polymer-Bound Catalysts.
Cr(III) loading of the polymer-bound species was determined
by UV-visible analysis of solutions of the polymer-bound
complexes. First, a UV-visible spectrum was obtained for a
low molecular weight catalyst at varying concentrations. These
data were used in a Beer’s law plot to obtain a value for an
extinction coefficient for the low molecular weight complex. The
UV-visible spectrum of the polymer-bound complex was then
measured and this extinction coefficient was used to determine
Cr(III)-salen loading.
2-tert-Butyl-4-polyisobutylphenol (5). To a 500-mL round-
bottomed flask equipped with a stir bar and rubber septum
were added 3 (15 g, 100 mmol), 4 (8.9 g, 8.9 mmol) in CH2Cl2
(200 mL), and concd H2SO4 (1.05 g, 10.7 mmol). The mixture
was then stirred under N2 for 3 days at room temperature. At
this point, the solvent was removed under reduced pressure and
250 mL of hexanes was added to the viscous oil. The hexanes
solution was washed with 3 150-mL portions of N,N-dimethyl-
formamide and 3 150-mL portions of 90% ethanol/water. The
organic phase was dried over Na2SO4 and filtered and solvent
was removed under reduced pressure to give 8.27 g of 5as a viscous,
light yellow oil, in 85% yield. Spectroscopic analyses were identical
with those previously reported.6 1HNMR(300MHz, CDCl3) δ7.3
(s, 1H), 7.05 (m, 1H), 6.6 (d, J = 7.75 Hz, 1H), 1.8 (s, 2H), and
0.8-1.6 (m, 140H). 13C NMR (125 MHz, CDCl3) δ 151.8, 142.1,
135.0, 125.4, 124.5, 115.9, multiple poorly resolved peaks between
58-60 and 22-39.
3-tert-Butyl-5-polyisobutylsalicylaldehyde (6). To a 100-mL
round-bottomed flask equipped with a stir bar, rubber septum,
and a water-jacketed reflux condenser were added 5 (3.42 g, 3.13
mmol) and 2,6-lutidine (0.58 g, 5 mmol) in 40 mL of toluene. The
mixture was stirred at room temperature for 30 min. At this
time, SnCl4 (0.15 mL, 1.25 mmol), in 10 mL of toluene, was
added dropwise with a syringe. The reaction mixture was then
stirred at room temperature for 1 h. At this point, paraformal-
dehyde (0.56 g, 18.78 mmol) was added to the flask. The reaction
mixture was then placed on an oil bath at 100 °C and stirred for
12 h. The reaction mixture was then cooled to room temperature
and acidified to pH 2 with 2 M HCl. The organic phase was
isolated and solvent was removed under reduced pressure.
Addition of 250 mL of hexanes to the viscous oil, washing
successively with 3 150-mL portions of N,N-dimethylforma-
mide and 3 150-mL portions of 90% ethanol/water produced a
hexanes phase containing the product 6 that was then dried over
Na2SO4 and filtered then the solvent was removed under reduced
pressure to yield 2.8 g of 6 as a viscous, yellow oil in 76% yield.
Spectroscopic analyses were identical with those previously
reported.7 1H NMR (300 MHz, CDCl3) δ 11.62 (s, 1H), 9.9
(s, 1H), 7.57 (s, 1H), 7.32 (s, 1H), 1.8 (s, 2H), and 0.8-1.6 (m,
140H). 13C NMR (125 MHz, CDCl3) δ 197.6, 159.3, 140. 9, 137.4,
133.2, 128.7, 120.2, multiple poorly resolved peaks between 58 and
60 and 22-39.
entry
cycle
1
2
3
4
5
6
1
2
3
4
5
6
88
90
100
100
100
100
aReaction conditions: 5 mol % of catalyst, 0.95 mmol of cyclohexene
oxide, and 1 mmol of azidotrimethylsilane were stirred under N2 for 3 h
in 3 mL of toluene at 75 °C. Recycling involved cooling the reaction
mixture to room temperature and separating the solid 24 from the
toluene solution of the product by centrifugation. bThese yields are
isolated yields of spectroscopically pure products obtained after removal
of the toluene solvent under reduced pressure.
The analysis showed that for the reaction in Table 3, metal
leaching was found to be ca. 0.3%.
Conclusion
We have shown that recovery and recycling of salen ligands
and their metal complexes can be facilitated with the use of
soluble polyolefin oligomers as phase anchors for such ligands/
catalysts. Using PIB-bound salen-Cr complexes, we have
shown that ring-opening of various epoxides with thiols and
an azide source can be facilitated and the catalyst can be
recycled with minimal metal leaching. However, applications
of this chemistry to asymmetric catalysis are limited by the low
enantioselectivity achieved because of the polyolefin catalyst’s
concentration. Other PIB-bound salen ligands as well as a
“half-salen” catalyst were also prepared. The latter “half-salen”
catalyst effectively catalyzed a hetero-Diels-Alder reaction but
could not be recycled because of the instability of the ligand to
the reaction conditions. The preparation of a PEOlig-bound
salen ligand was also achieved by using a regioselective Mitsu-
nobu reaction and the products were converted into a PEOlig
-
bound salen-Cr(III) complex. This complex was used as a
soluble thermomorphic catalyst in the ring-opening of cyclo-
hexene oxide with azidotrimethylsilane at 75 °C and was
recycled six times with no loss in catalytic activity.
Experimental Section
Vinyl-terminated PIB oligomers (Glissopal 1000 and 1300)
with Mn values of 1000 or 1300 Da and hydroxyl-terminated PE
oligomers (Unilin 550) with a nominal Mn of 550 are commercial
products and were obtained from BASF and Baker-Hughes,
respectively.6,12 All other reagents were purchased from com-
mercial sources and used without further purification unless
1
otherwise stated. H NMR spectra were recorded on an 500
MHz spectrometer operating at 499.95 MHz or on a 300 MHz
spectrometer operating at 299.91 MHz. 13C NMR spectra were
recorded on an 500 MHz spectrometer operating at 125.72 MHz
or on a 300 MHz spectrometer operating at 75.41 MHz. NMR
spectra in the case of PEOlig-bound substrates were obtained
at 70 °C. 13C NMR spectra for PEOlig-bound substrates were
obtained at concentrations of 10 mg/mL. Concentrations higher
than this resulted in poorer S/N ratios. Chemical shifts were
reported in parts per million (δ) relative to residual proton
resonances in deuterated chloroform (CDCl3), deuterated benzene
(C6D6), or deuterated methanol (CD3OH). Coupling constants
(J values) were reported in Hertz (Hz), and spin multiplicities
are indicated by the following symbols: s (singlet), d (doublet),
t (triplet), q (quartet), dd (doublet of doublet), and m (multiplet).
High-resolution MS data were collected from a MALDI/TOF
N,N0-Bis(3-tert-butyl-5-polyisobutylsalicylidene)-1,2-ethylenedi-
imine (7). To a 50-mL round-bottomed flask was added 6 (3.0 g,
2.6 mmol), ethylenediamine (0.08 g, 1.3 mmol), and a catalytic
amount of PTSA in 30 mL of toluene. The reaction mixture was
stirred at reflux overnight with a Dean-Stark trap. The solvent
was removed under reduced pressure and then 150 mL of hexanes
was added to the viscous residue. The hexane solution was
washed with 3 100-mL portions of N,N-dimethylformamide,
then 3 100-mL portions of 90% ethanol/water, dried over Na2SO4,
and filtered, then the hexane was removed under reduced pressure
to give a quantitative yield of 6.1 g of 7 as a viscous yellow oil .
Spectroscopic analyses were identical with those previously
reported.7 1H NMR (300 MHz, CDCl3) δ 8.43 (s, 2H), 7.40 (s,
1H), 7.05 (s, 1H), 3.95 (s, 4H), 1.8 (s, 4H), and 0.8-1.6 (m, 280H).
530 J. Org. Chem. Vol. 76, No. 2, 2011

Bergbreiter et al.
JOCArticle
13C NMR (75 MHz, CDCl3) δ 168.1, 158.2, 139.8, 136.5, 128.4,
127.7, 118.2, multiple poorly resolved peaks between 58 and 60
and 22-39.
32:21c 1H NMR (300 MHz, CDCl3) δ 7.3 (d, J = 8.60 Hz, 2H),
7.1 (d, J = 8.60 Hz, 2H), 3.6 (m, 1H), 3.1 (m, 1H), 2.8 (m, 1H),
2.35 (s, 3H), 1.2-1.6 (m, 14H), and 0.9 (t, J = 6.91 Hz, 3H). 13
C
(R,R)-N,N0-Bis(3-tert-butyl-5-polyisobutylsalicylidene)-1,2-
cyclohexanediimine (8). To a 50-mL round-bottomed flask
equipped with a water-jacketed reflux condenser and a 10-mL
addition funnel was added the L-tartrate salt of cyclohexenedia-
mine (0.265 g, 1 mmol), K2CO3 (1.66 g, 2 mmol), and 3 mL of
water. Then 11 mL of absolute ethanol was added and the
suspension was heated to reflux and became homogeneous. Then
6 (2.965 g, 2 mmol) in 5.5 mL of heptane was added dropwise with
an addition funnel. The yellow reaction mixture was allowed to stir
at reflux for 12 h under N2. At this point, the reaction mixture was
cooled to room temperature and 100 mL of hexanes was added.
The organic phase was washed with 3 50-mL portions of 90%
ethanol/water, dried over Na2SO4, and filtered, then the solvent
was removed under reduced pressure to afford a quantitative yield
(2.73 g) of 8 as a viscous yellow oil. 1H NMR (300 MHz, CDCl3) δ
8.31 (s, 2H), 7.29 (s, 2H), 6.94 (s, 2H), 3.37-3.30 (m, 2H), 1.8-0.8
(m, 380H). 13C NMR (125 MHz, CDCl3) δ 164.9, 156.9, 137.8,
134.9, 126.9, 125.7, 116.8, multiple poorly resolved peaks were
present between 57 and 60 and 20-40.
NMR (75 MHz, CDCl3) δ 137.0, 13.01, 130.0, 128.0, 69.0, 4.03,
36.0, 29.4, 29.2, 17.5, 23.0, 21.0, and 14.0.
33: 1H NMR (300 MHz, CDCl3) δ 7.4 (d, J = 8.47 Hz, 2H),
6.83 (d, J = 8.47 Hz, 2H), 3.8 (s, 3H), 3.05 (m, 1H), 2.78 (mm,
1H), 1.2-1.6 (m, 14H), and 0.9 (t, J = 7.04 Hz, 3H). 13C NMR
(75 MHz, CDCl3) δ 159.7, 134.0, 125.5, 115.0, 69.2, 55.8, 45.0,
36.5, 32.0, 29.6, 29.3, 27.5, 26.0, 23.0, and 14.3. HRMS calcd for
[C17H28SO2]þ 297.1888, found 297.1782.
34:21d 1H NMR (300 MHz, CDCl3) δ 7.3 (d, J = 8.29 Hz, 2H),
7.1 (d, J = 8.29 Hz, 2H), 3.77 (m, 1H), 3.58 (m, 1H), 3.03 (m,
1H), 2.95 (m, 1H), and 2.3 (s, 3H). 13C NMR (75 MHz, CDCl3) δ
137.4, 135.2, 131.3, 130.2, 69.8, 65.4, 38.9, and 21.3.
35:21e 1H NMR (300 MHz, CDCl3) δ 7.4 (d, J = 8.51 Hz, 2H),
6.9 (d, J = 8.51 Hz, 2H), 3.82 (s, 3H), 3.77 (m, 1H), 3.6 (m, 1H),
3.03 (m, 1H), and 2.9 (m, 1H). 13C NMR (75 MHz, CDCl3) δ
134.2, 132.9, 115.1, 114.8, 69.7, 65.4, 55.6, and 40.2.
ARO of Cyclohexene Oxide with TMS-N3 Catalyzed by 10.
To a 50-mL Schlenk tube equipped with a stir bar and rubber
septum was added 10 (1 g, 0.35 mmol) in a mixture of 5 mL of
heptane and 5 mL of ethanol. To this solution was added
cyclohexene oxide (0.7 mL, 7.08 mmol) and TMS-N3 (1 mL,
7.78 mmol). The dark reaction mixture was allowed to stir for 12
h under N2, at which time stirring was stopped and 0.5 mL of
water was added to the reaction mixture to induce phase
separation. The aqueous layer was removed and dried over
Na2SO4 and solvent was removed from this polar phase under
reduced pressure to yield 36 as a yellow oil. Spectroscopic
analyses were identical with those previously reported.7 1H
NMR (300 MHz, CDCl3) δ 3.28-3.16 (m, 1H), 3.06-2.97 (m,
1H), 1.94-1.79 (m, 3H), 1.66-1.46 (m, 2H), 1.3-0.97 (m, 4H).
13C NMR (75 MHz, CDCl3) δ 74.0, 67.0, 30.0, 29.5, 24.1, 23.5.
FT-IR (neat, cm-1) 3350, 2940, 2860, and 2098. Recycling of the
catalyst was achieved by simply adding fresh ethanol and
substrates to the organic phase.
Reduction and Derivatization of 36. The azido alcohol product
36 (0.63 g, 2.96 mmol) was placed in a vial equipped with a stir
bar. To this vial was added 5 mL of methanol and 0.05 g of 10%
Pd/C. The reaction mixture was equipped with a hydrogen
balloon. The mixture was monitored by IR spectroscopy and
stirred for 12 h, at which time it was diluted with 10 mL of
methanol and passed through a pad of Celite. The solvent was
removed under reduced pressure to yield the 2-aminocyclo-
hexanol as a colorless oil. 1H NMR (300 MHz, CD3-OD)
δ 3.36-3.41 (m, 1H), 2.79-2.85 (m, 1H), 2.01-2.04 (m, 2H),
1.76-1.78 (m, 2H), 1.30-1.40 (m, 4H). Derivatization of this
product was carried out by using a method similar to that
described previously.22 The amino alcohol (0.211 g, 1.83 mmol)
was dissolved in 2 mL of DMSO and placed in a 10-mL round-
bottomed flask equipped with a stir bar and rubber septum. To
the flask was then added K2CO3 (0.28 g, 2.03 mmol) and
1-fluoro-2,4-dinitrobenzene (0.25 mL, 1.83 mmol). The reaction
turned red and was allowed to stir under N2 for 10 h at 80 °C. At
this point, the reaction mixture was cooled to room temperature
and 5 mL of water was added. The yellow solid product that
formed (37) was isolated by filtration, washed with water, and
dried under vacuum and directly used without further purifica-
tion. The yield was 85%. Enantioselectivities were determined
by chiral HPLC [Chiralcel OD (0.40 cm ꢀ5 cm), Chiral Tech-
nologies, Inc., 0.5 mL/min, 80/20 (hexanes/IPA)]. 1H NMR (300
MHz, CDCl3) δ 9.16 (d, J = 2.69 Hz, 1H), 8.68 (bs, 1H), 8.23
(dd, J = 2.72, 6.80 Hz, 1 H), 7.16 (d, J = 9.65 Hz, 1 H), 4.86-
3.67 (m, 1H), 3.77-3.51(m, 1H), 2.4-2.26 (m, 2H), 2.26-2.08
(m, 2H), 1.94-1.35 (m, 5H). 13C NMR (75 MHz, CDCl3)
N,N0-Bis(3-tert-butyl-5-polyisobutylsalicylidene)-1,2-ethyle-
nediimino-Cr(III) Chloride (9). To a 50-mL round-bottomed flask
equipped with a stir bar and rubber septum were added 7 (6.62 g,
2.78 mmol) and CrCl2 (0.374 g, 3.05 mmol) in 30 mL of THF. The
mixture was then stirred under N2 at room temperature for 24 h
then stirred under air for 24 h. The solvent was removed under
reduced pressure and then a 150-mL portion of hexanes was added
to the viscous residue. The hexane solution was washed with a
solution of 3 100-mL portions of a saturated aqueous solution of
NH4Cl and with 3 100-mL portions of a saturated aqueous
solution of NaCl, dried over sodium sulfate, and filtered. The
solvent was removed under reduced pressure to give 4.30 g of 7 as a
dark-brown, viscous residue in 65% yield. IR (neat, cm-1) 1625,
1535, 1467, 1394, 1364, and 1235. UV-visible spectroscopy (THF,
λ
max = 350 nm). Separate experiments with 1 in THF showed it
had a λmax of 350 nm with ε = 4514 M-1 cm-1) and this extinction
coefficient was used to calculate the Cr(III) loading of 9.
(R,R)-N,N0-Bis(3-tert-butyl-5-polyisobutylsalicylidene)-1,2-
cyclohexanediimine-Cr(III) Chloride (10). Complex 10 was pre-
pared by using the same procedure used to prepare 9.
General Procedure for Epoxide Ring-Opening Reactions with
Thiols. To a 25-mL, round-bottomed flask equipped with a
magnetic stirbar and rubber septum were added epoxide (2 mmol),
thiol (2 mmol), 9 (0.05 g, 0.02 mmol), heptane (3 mL), and EtOH
(3 mL). The reaction mixture was placed under N2 and allowed to
stir for 24 h. At this point, approximately 0.3 mL of water was
added to the reaction mixture to induce phase separation to form a
biphasic mixture. The phases were allowed to separate and the
polar phase was removed, dried over Na2SO4, and filtered, then the
solvent was removed under reduced pressure.
28:21a 1H NMR (300 MHz, CDCl3) δ 7.4 (d, J = 7.83 Hz, 2H),
7.1 (d, J = 7.83 Hz, 2H), 3.3 (m, 1H), 2.7 (m, 1H), 2.37 (s, 3H),
2.1 (m, 2H), 1.7 (m, 2H), and 1.3 (m, 4H). 13C NMR (75 MHz,
CDCl3) δ 138.4, 134.8, 13.00, 128.0, 72.0, 56.8, 34.0, 32.7, 26.4,
24.5, and 21.4.
29:21b 1H NMR (300 MHz, CDCl3) δ 7.42 (d, J = 8.82 Hz,
2H), 6.85 (d, J = 8.82 Hz, 2H), 3.8 (s, 3H), 3.23 (m, 1H), 2.6 (m,
1H), 2.1 (m, 2H), 1.75 (m, 2H), and 1.25 (m, 4H). 13C NMR (75
MHz, CDCl3) δ 160.0, 137.3, 122.0, 114.7, 71.6, 57, 55.6, 33.9,
32.5, 26.4, and 24.5.
(21) (a) Yang, M.-H.; Yan, G.-B.; Zheng, Y.-F. Tetrahedron Lett. 2008,
49, 6471–6474. (b) Reddy, M. S.; Srinivas, B.; Sridhar, R.; Narender, M.;
Rao, K. R. J. Mol. Catal. 2006, 255, 180–183. (c) Solladie, G.; Demailly, G.;
Greck, C. Tetrahedron Lett. 1985, 26, 435–438. (d) Culvenor, C. C. J.;
Davies, W.; Savige, W. E. J. Chem. Soc. 1949, 2198–2206. (e) Owen, L. N.;
Muhammad, S. J. Chem. Soc. C 1971, 1442–1447.
ꢀ
(22) Buckingham, J.; Guthrie, R. D. J. Chem. Soc. C 1970, 106–108.
J. Org. Chem. Vol. 76, No. 2, 2011 531

Bergbreiter et al.
JOCArticle
δ 148.7, 135.9, 130.3, 130.0, 124.4, 115.1, 75.9, 74.5, 58.7,
52.8, 34.2, 33.7, 31.4, 30.8, 30.5, 24.6, 24.4, 23.9, 23.4. Mp
184 -187 °C.
4.90 mmol) in glacial acetic acid (3.19 mL). The mixture was stirred
at room temperature for 3 h. At this point, 16 (5 g, 4.46 mmol) in a
mixture of heptane and ethanol (1:1, 13 mL) was added to the
reaction flask and placed on an oil bath regulated at 90 °C and
allowed to stir for 5 days. At this point, the reaction was cooled to
room temperature and hexanes (50 mL) was added. The mixture
was washed with 1 25-mL portion of saturated NaHCO3(aq) and 3
25-mL portions of 90% ethanol/water. The organic phase was
dried over Na2SO4 and filtered and solvent was removed under
reduced pressure to give a viscous yellow oil that was then subjected
to column chromatography (silica, hexanes:ethylacetate (9:1)) to
give 2.6 g of 17, 50%. 1H NMR (500 MHz, CDCl3) δ 10.40 (s, 1H),
7.65 (d, J = 2.5 Hz, 2H), 7.22 (d, J = 2.4 Hz, 2 Hz), 3.70 (s, 2H),
2.55 (m, 4H), 1.8 (s, 2H), and 0.8-1.6 (m, 140H). 13C NMR (125
MHz, CDCl3) δ 191.7, 159.7, 140.9, 133.8, 125.4, 122.2, multiple
poorly resolved peaks between 58 and 60, 38-39, and 30-33.
N,N0-Bis((3-piperidylmethyl)-5-polyisobutylsalicylidene)-1,2-
ethylenediimine (18). This compound was prepared by using the
same procedure used to prepare 7. 1H NMR (500 MHz, CDCl3) δ
8.39(s, 2H), 7.31 (s, 2H), 7.19 (s, 2H), 3.70 (s, 2H), 2.60 (m, 8H), 1.8
(s, 4H), and 0.8-1.6 (m, 300H). 13C NMR (125 MHz, CDCl3) δ
166.3, 157.1, 139.9, 132.0, 127.2, 124.1, 117.8, multiple poorly
resolved peaks between 58 and 60.0, 38-39, and 30-33.
ARO of Cyclohexene Oxide with TMS-N3 Catalyzed by 2. To
a 50-mL Schlenk tube equipped with a stir bar and rubber
septum was added Jacobsen’s catalyst 2 (0.116 g, 0.184 mmol),
cyclohexene oxide (0.37 mL, 3.68 mmol), and azidotrimethylsi-
lane (0.537 mL, 4.05 mmol) in a mixture of 5 mL of heptane and
5 mL of absolute ethanol. The reaction mixture was stirred
under N2 for 12 h at room temperature. At this time, the solvent
was removed and the viscous oil was worked up by dissolving it
in 20 mL of hexanes and passing the resulting solution through a
plug of Celite. The solvent was removed under reduced pressure
to give 0.514 g of 36 in 66% yield. Spectroscopic analyses
showed that this product was identical with that formed in
reactions catalyzed by 10.
3-tert-Butyl-5-polyisobutylsalicylidene-(1R,2S)-(1-amino-2-inda-
nol)imine (12). To a two-necked, 50-mL round-bottomed flask
equipped with a stir bar, water-jacketed reflux condenser, and a
Dean-Stark trap were added 6 (2.94 g, 2.028 mmol) in 21 mL of
toluene, 11 (0.333 g, 2.230 mmol), and a catalytic amount of PTSA.
The reaction mixture was stirred with azeotropic removal of water
for12handthencooledtoroomtemperature. Thesolventwasthen
removed under reduced pressure and to the viscous oil was added
150 mL of hexanes. The solution was washed with 3 100-mL
portions of DMF and 3 100-mL portions of 90% ethanol/water.
The organic phase was dried over Na2SO4 and filtered and the
solvent was removed under reduced pressure to yield 3.19 g of 12 as
(R,R)-N,N0-Bis((3-piperidylmethyl)-5-polyisobutyl)salicyli-
dene-1,2-cyclohexanediimine (19). This compound was prepared by
using the same procedure used to prepare 8. 1H NMR (500 MHz,
CDCl3) δ 8.31 (s, 2H), 7.36 (s, 2H), 7.03 (s, 2H), 3.60 (d, J = 12.7
Hz, 2H), 3.47 (d, J = 12.71 Hz, 2H), 3.3 (m, 2H), 2.60 (m, 8H), 1.8
(s, 4H), and 0.8-1.6 (m, 300H). 13C NMR (125 MHz, CDCl3) δ
163.9, 156.1, 138.7, 130.8, 127.4, 126.5, 116.7, multiple poorly
resolved peaks between 58 and 60, 38-39, and 30-33.
1
a viscous yellow oil in quantitative yield. H NMR (300 MHz,
CDCl3) δ8.62 (s, 1H), 7.40 (s, 1H), 7.35-7.17 (m, 4H), 7.12 (s, 1H),
4.80 (d, J = 5.32 Hz, 1H), 4.73 (bs, 1H), 3.25 (m, 2H), 1.80-0.80
(m, 160H). 13C NMR (75 MHz, CDCl3) δ 168.3, 158.0, 141.0,
140.8, 139.5, 136.5, 128.6, 128.5, 127.2, 127.0, 125.5, 125.0, 117.7,
75.8, 75.3, multiple poorly resolved peaks were present between 57
and 60 and 20-40.
PIB-Supported Half-Salen-Cr Complex (13). To a 50-mL
round-bottomed flask equipped with a stir bar and rubber
septum was added 12 (2.74 g, 1.73 mmol) and CrCl2 (0.233 g,
1.90 mmol) in 21 mL of THF. The dark reaction mixture was
allowed to stir under N2 for 24 h at room temperature, at which
point the reaction system was exposed to air and stirred for 24 h.
The solvent was removed under reduced pressure and 150 mL of
hexanes was added to the dark, viscous oil. The hexanes solution
was then washed with 3 50-mL portions of a saturated aqueous
solution of NH4Cl and 3 50-mL portions of a saturated aqueous
solution of NaCl. The organic phase was dried over Na2SO4 and
the solvent was removed under reduced pressure to yield 2.82 g
of 13 as a dark viscous oil in quantitative yield. UV-visible
spectroscopy (THF, λmax = 348 nm).
4-(Polyisobutyl)phenol (15). This compound was prepared by
using the procedure described above leading to 5. Spectroscopic
analyses were identical with those previously reported.13 1H
NMR (500 MHz, CDCl3) δ 7.23 (d, J = 8.79 Hz, 2H), 6.75 (d,
J = 8.79 Hz, 2H), 1.8 (s, 2H), and 0.8-1.6 (m, 140H). 13C NMR
(125 MHz, CDCl3) δ 153.1, 142.9, 127.5, 114.7, multiple poorly
resolved peaks between 58 and 60, 38-39, and 30-33.
5-Polyisobutylsalicylaldehyde (16). This compound was pre-
pared analogously to 6. Spectroscopic analyses were identical
with those previously reported.14a 1H NMR (500 MHz, CDCl3)
δ 10.88 (s, 1H), 9.9 (s, 1H), 7.57 (dd, J = 2.44, 8.54 Hz, 1H), 7.48
(m, 1H), 6.94 (d, J = 8.54 Hz, 1H), 1.8 (s, 2H), and 0.8-1.6 (m,
140H). 13C NMR (125 MHz, CDCl3) δ 197.0, 159.6, 142.4,
135.7, 130.7, 120.2, 117.2, multiple poorly resolved peaks be-
tween 58 and 60, 38-39, and 30-33.
3-tert-Butyl-5-hydroxysalicylaldehyde (20). This compound
was prepared according to a literature procedure.23
PEOlig-Bound Salicylaldehyde Derivative (22). To a 150-mL,
round-bottomed flask equipped with a stir bar, pressure-equal-
ized addition funnel, and a rubber septum was added 20 (1.88 g),
21 (1.33 g, 6.84 mmol), and triphenylphosphine (3.47 g, 17.09
mmol) in toluene (34 mL). The mixture was placed in an oil bath
regulated at 80 °C and the contents were allowed to dissolve. At
this point, a solution of DEAD (2.69 mL, 17.09 mmol) in toluene
(10 mL) was added dropwise with the addition funnel. The dark
reaction mixture was allowed to stir at 80 °C for 24 h. Then the
reaction mixture was cooled to room temperature whereupon
the PEOlig-bound product precipitated. This product was col-
lected by vacuum filtration and washed with toluene (50 mL)
and THF (50 mL). This product was then placed in a 25-mL
round-bottomed flask equipped with a stir bar and rubber
septum and 10 mL of toluene containing 0.5 mL of water was
added to this flask along with 0.45 g of PTSA. The flask was
placed on an oil bath at 80 °C for 2 h. This procedure converted
any imine derivative of the aldehyde group of 22 back into an
aldehyde group. The PEOlig-bound product 22 that precipitated
on cooling of the reaction mixture at this point was collected by
vacuum filtration and washed with toluene (50 mL) and THF
(50 mL) to yield 1.1 g of 22 as an light orange solid (80% yield).
1H NMR (500 MHz, benzene-d6, 70 °C) δ 11.87 (s, 1H), 9.38 (s,
1H), 7.26 (d, J = 3.06 Hz, 1H), 6.43 (d, J = 1.9 Hz, 1H), 3.71 (t,
J = 6.26 Hz, 2H), 1.71 (m, 1H), 1.49-1.30 (bs, 230H), 0.91 (t,
J = 6.58 Hz, 5H). 13C NMR (125 MHz, benzene-d6, 70 °C) δ
196.0, 155.9, 151.7, 139.9, 123.9, 120.1, 113.1, 68.6, 34.8, 31.9,
29.7, 29.3, 29.2, 28.9, 26.1, 22.5, 13.9.
PEOlig-Bound Salen Derivative (23). To a 25-mL, round-
bottomed flask equipped with a stir bar and a rubber septum
was added 22 (0.80 g) in toluene (6 mL). The mixture was placed
under N2 and the reaction flask was placed on an oil bath regulated
at 80 °C. To the reaction mixture was added ethylenediamine
3-(Piperindylmethyl)-5-polyisobutylsalicyladehyde(17). To a
50-mL round-bottomed flask equipped with a stir bar, rubber
septum, and a water-jacketed reflux condenser was added piper-
idine (0.44 mL, 4.46 mmol) and paraformaldehyde (0.15 g,
(23) Anyanwu, U. K.; Venkataraman, D. Green Chem. 2005, 7, 424–425.
532 J. Org. Chem. Vol. 76, No. 2, 2011

Bergbreiter et al.
JOCArticle
(0.011 mL, 0.163 mmol). Upon addition of ethylenediamine, the
reaction became bright yellow. The reaction was allowed to stir for
1 h, at which point 3,5-di-tert-butylsalicylaldehyde (0.25 g, 1.30 mmol)
was added to the reaction mixture, which was allowed to stir for an
additional hour. At this point, the reaction was cooled to room
temperature and the product was collected by vacuum filtration,
then washed with toluene (20 mL) and THF (20 mL) to give 0.7 g
of 23 as a yellow solid (as a mixture of the bis-PEOlig and mono-
a stir bar and rubber septum were added complex 24 and toluene
(3 mL). The reaction vessel was then placed under N2 and placed
on an oil bath regulated at 75 °C. Once dissolution of the solid
was achieved, cyclohexene oxide (0.097 mL, 0.95 mmol) was
added by syringe followed by azidotrimethylsilane (0.126 mL,
0.95 mmol). The reaction mixture was allowed to stir at 75 °C for
3 h. At this point, the reaction vessel was removed from the oil
bath and allowed to cool to room temperature, inducing pre-
cipitation of the catalyst. The reaction vessel was then exposed
to air and 5 mL of toluene was added. The system was then
subjected to centrifugation for 10 min at 1100 rpm. The solvent
was decanted and the process was repeated. The toluene layers
were combined and the solvent was removed under reduced
pressure to yield the product as a light yellow oil. Spectra were
identical with those found in the literature.7 To facilitate recy-
cling of the catalyst, toluene (3 mL) was added to the recovered
catalyst and the entire process was repeated.
1
PEOlig-bound species), 90%. H NMR (500 MHz, benzene-d6,
70 °C) δ7.86 (s, 2H), 7.54 (s, 1H), 6.51 (m, 2 H), 3.83 (t, J=6.42Hz,
4 H), 3.35 (s, 4 H), 1.49-1.30 (bs, 420H), 0.91 (t, J = 6.58 Hz,
5H)). 13C NMR (125 MHz, benzene-d6, 70 °C) δ 167.5, 166.9,
155.1, 151.3, 143.6, 139.9, 138.9, 119.0, 118.2, 113.1, 68.6, 59.1,
34.9, 31.8, 31.2, 30.8, 29.6, 29.4, 29.2, 26.1, 22.5, 13.6.
PEOlig-Bound Salen-Cr Complex (24). To a 10-mL, round-
bottomed flask equipped with a stir bar, a rubber septum,
and a water-jacketed reflux condenser were added 23 (0.7 g),
CrCl2 (0.086 g, 0.7 mmol), toluene (2 mL), and DMF (2 mL)
and the flask was placed under N2. The mixture was placed
on an oil bath regulated at 100 °C and allowed to stir for 12 h.
At this point, the flask was exposed to air and the reaction
allowed to stir for 12 h. The reaction was then allowed to cool
to room temperature and the product was collected by
vacuum filtration and washed with THF (100 mL) to give
0.63 g of 24 as a brownish/yellow solid, 90%. UV-visible
spectroscopy (toluene, 70 °C, λmax = 366 nm). Separate
UV-visible analysis of 1 in toluene showed it had a λmax of
366 nm with ε = 7566 M-1 cm-1, and this extinction coeffi-
cient was used to calculate the Cr(III) loading of 24.
Acknowledgment. Support from the Robert A. Welch
Foundation (A-0639) and National Science Foundation
(CHE-0952134) is gratefully acknowledged. We thank
BASF and Baker-Hughes for the PIB and PE oligomers,
respectively. Dr. Jianhua Tian and Mr. Yun-Chin Yang
are gratefully acknowledged for their help with the metal
analysis.
Supporting Information Available: NMR spectra and struc-
tures for selected compounds, phase-selectivity, and ICP-MS
digestion procedures for metal analysis. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
Ring-Opening of Cyclohexene Oxide with TMS-N3 Catalyzed
by 24. To a 50-mL round-bottomed Schlenk tube equipped with
J. Org. Chem. Vol. 76, No. 2, 2011 533