Immobilized aza-bis(oxazoline) copper catalysts on alkanethiol self-assembled monolayers on gold: Selectivity dependence on surface electronic environments

Article abstract of DOI:10.1016/j.jcat.2010.07.027

Aza-bis(oxazoline) copper complexes have been immobilized onto alkanethiol self-assembled monolayers on gold utilizing five background tail groups with different electronic characteristics. The catalyst was tested in the standard cyclopropanation reaction of ethyl diazoacetate and styrene. The five different tail groups were hydroxyl, bromine, carboxylic acid, ester, and nitrile. Enantioselectivity improved to 95% when the surrounding tail groups were hydroxyl- and bromine-terminated surfaces. The carboxylic acid and ester tail groups reduced the enantioselectivity compared to the homogeneous phase. Additionally, the homogeneous cyclopropanation reaction was performed in methanol, acetonitrile, ethyl acetate, and acetic acid to determine whether similar trends in selectivity could be obtained by varying the homogenous electronic environment. However, the cyclopropanation reaction in these solvents gave greatly reduced selectivity and yield of the cyclopropane products demonstrating the positive aspects of immobilization of self-assembled monolayer supports.

Full text of DOI:10.1016/j.jcat.2010.07.027

Journal of Catalysis 275 (2010) 149–157
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Immobilized aza-bis(oxazoline) copper catalysts on alkanethiol self-assembled
monolayers on gold: Selectivity dependence on surface electronic environments
⇑
Christy C. Paluti, Ellen S. Gawalt
Department of Chemistry and Biochemistry, Duquesne University, Pittsburgh, PA 15282, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Aza-bis(oxazoline) copper complexes have been immobilized onto alkanethiol self-assembled monolay-
ers on gold utilizing five background tail groups with different electronic characteristics. The catalyst was
tested in the standard cyclopropanation reaction of ethyl diazoacetate and styrene. The five different tail
groups were hydroxyl, bromine, carboxylic acid, ester, and nitrile. Enantioselectivity improved to 95%
when the surrounding tail groups were hydroxyl- and bromine-terminated surfaces. The carboxylic acid
and ester tail groups reduced the enantioselectivity compared to the homogeneous phase. Additionally,
the homogeneous cyclopropanation reaction was performed in methanol, acetonitrile, ethyl acetate, and
acetic acid to determine whether similar trends in selectivity could be obtained by varying the homog-
enous electronic environment. However, the cyclopropanation reaction in these solvents gave greatly
reduced selectivity and yield of the cyclopropane products demonstrating the positive aspects of immo-
bilization of self-assembled monolayer supports.
Received 10 June 2010
Revised 27 July 2010
Accepted 28 July 2010
Keywords:
Asymmetric cyclopropanation
Aza-bis(oxazolines) copper
Immobilization
Nitrogen ligand
Chiral shift reagents
Self-assembled monolayers
Heterogeneous catalysis
Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction
ing the catalyst even with the monolayer surface obtained the
greatest selectivity in the cyclopropanation reaction. This en-
Cyclopropane derivatives are important chemical compounds
with interesting biological properties [1] as well as utility in organ-
ic synthesis [2]. Consequently, extensive work has been devoted to
the development of efficient diastereo- and enantio-selective
methods for the synthesis of cyclopropanes [3–8]. One versatile
method of cyclopropane synthesis is the transition metal-catalyzed
cyclopropanation of olefins with diazo compounds. In the litera-
ture, many homogeneous transition metal catalytic systems have
been developed, which show high selectivity in the transition me-
tal-catalyzed cyclopropanation reactions with copper [5–7], rho-
dium [8], and cobalt [4,9]. Nitrogen ligands with C2 symmetry
such as bis(oxazoline) [4–6,8] and aza-bis(oxazoline) [7] form
superior catalysts for the cyclopropanation reaction. Traditional
transition metal catalysts have been immobilized on solid sup-
ports, such as organic polymers [10–15], inorganic solids [10–
15], and colloids [16]. Immobilization of transition metal catalysts
allows for further modification of the ligand systems, which may
produce selectivity enhancements and cost reduction through the
ability to recycle the catalyst multiple times.
hanced selectivity may be the result of hydrogen bonding between
the copper carbene and the surface hydroxyl groups [17]. The en-
hanced selectivity obtained by immobilizing the catalyst even with
the monolayer surface prompted our group to study the influence
of changing the electronic properties of the tail group molecules in
the monolayer surrounding the catalyst.
The generally accepted catalytic pathway for the copper cata-
lyzed cyclopropanation reaction of ethyl diazoacetate and styrene
is shown in Fig. 1 [18,19]. There are two proposed mechanisms that
account for this reaction pathway, the concerted and stepwise [20].
There are opinions and data to support both mechanisms. How-
ever, this work will not focus on this ongoing discussion. However,
one interesting interaction called the ‘‘secondary effect” will be
touched upon in this paper. Initially proposed by Doyle [21], the
nucleophilic oxygen of the polar carbene substituent can stabilize
the developing electropositive center of the reacting alkene as
shown in Fig. 2. This ‘‘secondary” effect is similar to the one that
controls endo selectivity in the Diels–Alder reaction. In the litera-
ture, modification of the polar group of the metal carbene has al-
lowed for effective diastereocontrol [20–25]. Modification of the
polar group of the metal carbene requires synthesis of multiple
diazo compounds.
In a previous study conducted by our group, the aza-bis(oxazo-
line) copper complexes were immobilized onto self-assembled
monolayers with three different surface orientations: even, above,
and below monolayer surface [17]. In the previous study orientat-
Another way of controlling the electronic influences in the
cyclopropanation reaction without modification of the polar sub-
stituent of the metal carbene may be the addition of solid supports
with electron withdrawing and donating groups present near the
⇑
Corresponding author. Fax: +1 412 396 5683.
E-mail address: gawalte@duq.edu (E.S. Gawalt).
0021-9517/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2010.07.027

150
C.C. Paluti, E.S. Gawalt / Journal of Catalysis 275 (2010) 149–157
2. Experimental
2.1. Materials
All chemicals and solvents were used as received. Thin-layer
chromatography (TLC) was preformed with Merck 60 F254 silica
gel plates. Visualization of compounds on the silica gel plates
was accomplished by UV light. 1H NMR spectra were obtained at
room temperature on a Bruker Avance 400 MHz spectrometer with
chemical shifts given in ppm relative to the residual solvent peak
(CDCl3, 7.26 ppm).
Starting materials for ligand synthesis were as follows: (S)-tert-
leucinol (98% Aldrich), benzaldehyde (Aldrich), and p-toluenesul-
fonic acid monohydrate (99% Acros) methanol certified ACS
(Fisher).
Starting materials for alkanethiol synthesis were as follows: 10-
undecenoic acid methyl ester (98% TCI), 11-bromo-1-undecene
(96% Alfa Aesar), azo-bis-isobutyronitrile (98% Sigma–Aldrich), thi-
olacetic acid (96% Aldrich), hydrochloric acid (ACS Fisher Scien-
tific), trityl chloride (98% Aldrich), acetonitrile (HPLC grade Fisher
Scientific), acrylonitrite (99% Sigma–Aldrich) trifluroacetic acid
(98% Sigma–Aldrich), and triethylsilane (99% Aldrich).
Starting materials for self-assembled monolayer formation
were as follows: 11-mercapto-1-undecanol (97% Aldrich), 11-mer-
capto-undecanoic acid (95% Aldrich), and 9-mercapto-1-nonanol
(96% Aldrich).
Fig. 1. The catalytic cycle for the copper catalyzed cyclopropanation reaction of
ethyl diazoacetate and styrene.
Starting materials for gold substrate preparation were as fol-
lows: aluminum wire (99.999% Kurt J. Lesker), chromium pieces
(99.998% Kurt J. Lesker), gold wire (99.99% Kurt. J. Lesker), and
microscope cover glass (Fisher Scientific).
Starting materials for the Mitsunobu reaction were as follows:
diethyl azodicarboxylate (40 wt.% solution in toluene Aldrich)
and triphenylphosphine (99% Aldrich).
Fig. 2. The secondary effect of the oxygen of the polar copper carbene stabilizing
the developing electropositive center of the reacting alkene. The orientation of the
phenol or hydrogen depends on the steric influence of the catalyst.
Starting materials for the cyclopropanation reaction were as fol-
lows: styrene (98% Acros), ethyl diazoacetate (Aldrich), dichloro-
methane (99.9% Acros), phenylhydrazine (95% Acros), and
Cu(OTf)2 (98% Aldrich).
catalytic site. These electron withdrawing and donating groups can
interact with the catalytic site through hydrogen bonding and di-
pole–dipole intermolecular interactions. Based on the secondary
effect, the presence of surface carbonyl groups will have a pro-
found influence on the selectivity of the system. The surface car-
bonyl groups near the catalytic site will reduce selectivity of the
system because they can stabilize the electropositive center of
the reacting alkene, allowing the complex to adopt a new orienta-
tion. This new orientation allows for the production of cyclopro-
panes with opposite stereochemistry. By changing the orientation
of the reacting alkene with respect to the catalytic center, the
selectivity of the system will be affected. If addition of surface car-
bonyl groups has little influence on the selectivity of the catalytic
system, then it can be concluded that the secondary effect has min-
imal influence on the selectivity of the system. In Fig. 3, the three
main electronic environments are depicted: (A) hydroxyl surface,
(B) halogen surface where surface bromine groups are present,
and (C) carbonyl surfaces with carboxylic acid and ester substitu-
ents presented at the surface.
2.2. Aza-bis(oxazoline) ligand synthesis
Bis[4,5-dihydro-(4S)-(1,1-dimethylethyl)-1,3-oxazole-2-yl]-
amine (1a) was prepared according to literature procedure [7].
2.3. Preparation of heterogeneous catalyst
2.3.1. Gold substrate preparation
Glass microscope slides were cleaned with ethanol and dried un-
der a stream of nitrogen for 10 min. Aluminum was deposited onto
the glass slides by thermal vapor deposition at a pressure of
3 Â 10À6 mbar. Gold was deposited onto the aluminum-coated
glass slides by thermal vapor deposition at
a pressure of
3 Â 10À6 mbar.
Fig. 3. Electronic environments: (A) hydroxyl surface, (B) halogen surface, and (C) carbonyl surface.

C.C. Paluti, E.S. Gawalt / Journal of Catalysis 275 (2010) 149–157
151
2.3.2. Thiol synthesis
2.3.2.1. Bromine tail group thiol
chips were analyzed and then utilized in the immobilization of the
aza-bis(oxazoline) ligand.
2.3.2.1.1. Bromine tail group thiol ester (11-bromoundecyl-ethanethi-
oate). To a three-neck round-bottom flask wrapped with Al foil,
25 ml methanol, azo-bis isobutyronitrile (0.152 mmol, 25 mg), thi-
olacetic acid (8.60 mmol, 0.655 ml), and 11-bromo-1-undecene
(3.92 mmol, 0.915 ml) were added. The reaction mixture was stir-
red under UV light for 12 h.
2.3.2.1.2. Bromine tail group thiol (11-bromoundecane-1-thiol). 11-
Bromoundecyl-ethanethioate (2.92 mmol, 0.9 g) was added to a
round-bottom flask with 20 ml methanol, and then 2 ml of hydro-
chloric acid was added. The reaction mixture was refluxed over-
night. 1H NMR (400 MHz, CDCl3 RT) d 3.38 ppm (t, 2H, J = 7 Hz),
2.49 ppm (q, 2H, J = 7 Hz), 1.83 ppm, (q, 2H J = 7 Hz), 1.58 ppm (q,
2H, J = 7 Hz).
2.3.4. Mitsunobu reaction (ligand immobilization)
The ligand was immobilized to the SAM surface via a surface
Mitsunobu reaction as in the literature [17]. Mixed monolayer
chips were placed in a three-neck round-bottom flask. Aza-
bis(oxazoline) ligand (1a) and triphenylphosphine were added.
THF was added and then the reaction flask was placed in an oil
bath at 50 °C. Under N2, diethyl azodicarboxylate was added and
the reaction was hand-stirred [17] for 2 h. The samples were re-
moved from the reaction mixture and rinsed with dichlorometh-
ane to remove reagents loosely bound to the surface.
2.4. Surface characterization
2.3.2.2. Ester tail group thiol
2.4.1. Diffuse reflectance infrared Fourier transform (DRIFT)
2.3.2.2.1. Ester tail group thiol ester (methyl 10-(acetylthio)-undecan-
oate). To a three-neck round-bottom flask wrapped with Al foil,
25 ml methanol, azo-bis isobutyronitrile (0.152 mmol, 25 mg), thi-
olacetic acid (8.60 mmol, 0.655 ml), and 10-undecenoic acid
(4.61 mmol, 0.915 ml) were added. The reaction mixture was stir-
red under UV light for 12 h.
2.3.2.2.2. Ester tail group thiol (methyl 10-mercaptodecanoate).
Methyl 10-(acetylthio)-undecanoate (3.45 mmol, 0.9 g) was added
to a round-bottom flask with 20 ml methanol, followed by 2 ml of
hydrochloric acid. The reaction mixture was refluxed overnight.
(Yield 72%) 1H NMR (400 MHz, CDCl3 RT) d 3.67 ppm (s, 3H),
2.52 ppm (q, 2H, J = 8, 7 Hz), 2.31 ppm (t, 2H, J = 7 Hz), 1.73 ppm
(s, 1H), 1.62 ppm (m, 2H), 1.50 ppm (m, 2H) 1.32 ppm (m, 6H).
The chips were studied using diffuse reflectance infrared Fou-
rier transform spectroscopy (DRIFT) (Thermo Nicolet-NEXUS 470
FT-IR) to analyze the alkyl chain ordering of the molecules on the
surface. The spectra were recorded under nitrogen to eliminate
the background signals due to CO2 and H2O adsorption bands.
Unmodified gold substrates were used as the background spectra.
We collected 1024 scans for each sample with
resolution.
a
4 cmÀ1
2.4.2. AP MALDI-TOF MS
Atmospheric pressure matrix-assisted laser desorption/ioniza-
tion time-of-flight mass spectrometry (AP MALDI-TOF MS) was
used to analyze the monolayer and to determine whether the
aza-bis(oxazoline) ligands were attached to the monolayer surface
before and after use in the cyclopropanation reaction. A high-res-
olution AP MALDI-TOF MS (Agilent Tech.) with pulsed dynamic
focusing was used. MS analysis of the ions were detected in the
positive mode using a 337 nm N2 laser, pulse width of 20 ns, a
capillary voltage of 3500 V, fragmentor voltage of 260 V, skimmer
voltage of 40 V, and drying temperature of 325 °C. External cali-
bration was done using an ES-TOF tuning mix of 10 masses for
the range of 100–2300 m/z. Three to five areas of each sample
were characterized using the MALDI-TOF. The spectra were col-
lected by uniformly moving the laser in a circular pattern across
the sample for 1 min. The spectra were reproducible within a sin-
2.3.2.3. Nitrile tail group thiol
2.3.2.3.1. Addition of protecting group. 11-Mercapto-1-undecanol
(8.978 mmol, 1.833 g), trityl chloride (6.27 mmol, 1.75 g), and
40 ml THF were added to a three-neck round-bottom flask. The
reaction mixture was stirred under nitrogen for 14 h. Purification
of crude product was performed by filtration of crude product with
2/1 ethyl acetate/hexane. The protected thiol was then used in the
next step of the synthesis.
2.3.2.3.2. Addition of nitrile group to protected thiol. To a round-bot-
tom flask wrapped with Al foil, 10 mg NaOH, protected thiol
(0.828 mmol, 360 mg), and 15 ml acetonitrile were added and
placed under nitrogen. After drop-wise addition of acrylonitrite
(1.51 mmol, 80 ul), the reaction mixture was stirred overnight un-
der nitrogen. The reaction mixture was quenched with distilled
water and then extracted with CH2Cl2, dried with MgSO4 and then
evaporation of CH2Cl2. The nitrile tail group protected thiol was
then used in the next step of the synthesis.
gle sample and across different samples. The matrix,
a-cyano-4-
hydroxycinnamic acid (CHCA) (Sigma–Aldrich, >99.0% purity),
was used without further purification and dissolved in 9/1 ratio
of the solvents CH2Cl2/EtOH (0.9 ml CH2Cl2 and 0.1 ml ethanol
for a total volume of 1 ml). The ratio of matrix (CHCA) to solvent
was 10 mg/ml. The matrix fragmentation can be observed in each
of the spectra.
2.3.2.3.3. Nitrile tail group thiol (3-((11-mercaptoundecyl)oxy)pro-
pane-nitrile). Nitrile tail group protected thiol (0.260 mmol,
127 mg), trifluroacetic acid (13.15 mmol, 1.5 ml), (1.37 mmol,
0.160 ml) triethylsilane, and 20 ml CH2Cl2 were added to
a
2.5. Cyclopropanation procedures
round-bottom flask. The reaction mixture was then stirred under
nitrogen overnight. Evaporation of CH2Cl2 and then column chro-
matography 50/50 hexane/CH2Cl2 to 100 CH2Cl2 produced a color-
less precipitate. (Yield 54%) 1H NMR (400 MHz, CDCl3 RT) d
3.65 ppm (t, 2H, J = 7, 6 Hz), 3.48 ppm (t, 2H, J = 9, 7 Hz),
2.60 ppm (t, 2H, J = 7 Hz), 2.53 ppm (q, 2H, J = 7 Hz) 1.65 ppm (m,
4H), 1.55 ppm (s, 1H), 1.28 ppm (m, 10H).
2.5.1. Homogeneous asymmetric cyclopropanation of styrene and
ethyl diazoacetate
The cyclopropanation reaction was performed according to the
literature [7] with the following changes.
Under nitrogen atmosphere, Cu(OTf)2(3.6 mg, 0.01 mmol) and
aza-bis(oxazoline) ligand a (6.2 mg, 0.022 mol) were dissolved in
anhydrous CH2Cl2 to produce a light green solution. Phenylhydr-
azine (22 ul of a 5% solution) was added and the green color
disappeared. Styrene (312 mg, 3 mmol, 345 ul) and then ethyl
diazoacetate (EDA) (1 mmol, 1 ml of an 8% solution in CH2Cl2 di-
luted with 7 ml CH2Cl2) were added over 4 h. The reaction mixture
was stirred for 5 h. The solvent was evaporated in vacuo to give
green oil. The reaction products were purified by p-TLC (purifica-
2.3.3. SAM preparation
Gold substrates were immersed in 2 mM ethanolic solution of
alkanethiols (10:90) (hydroxyl:background tail group) and incu-
bated overnight at room temperature (23–25 °C). Upon removal
from solution, the samples were carefully rinsed with CH2Cl2 and
dried under a stream of nitrogen (Scheme 1). The mixed monolayer

152
C.C. Paluti, E.S. Gawalt / Journal of Catalysis 275 (2010) 149–157
Scheme 1. Immobilization of aza-bis(oxazoline) on different surfaces.
tion and separation of the trans and cis isomers was accomplished
by preparation thin-layer chromatography).
CH2Cl2) were added over 4 h. The reaction was hand-stirred for 5 h.
The catalytic chips were removed and the reaction solvent was
evaporated and the reaction products were purified by p-TLC.
2.5.2. Heterogeneous asymmetric cyclopropanation of styrene and
ethyl diazoacetate
2.6. Purification and analysis of cyclopropanation products
The heterogeneous cyclopropanation reaction was performed
according to the literature [17].
The cyclopropane products were purified using p-TLC. The sol-
vent ratio of 96/4 hexane/diethyl ether was used. CH2Cl2 was used
to remove the products from the silica gel. The solvent was evap-
orated under vacuum to yield a colorless oil product. The oil was
then analyzed by 1H NMR and integration of the cis and trans peaks
gave the ratio of cis and trans isomers. The yield of the reaction
was determined by first weighing the products before addition of
The systems 2–4 (a and b) (Scheme 1) were added to a three-
neck round-bottom flask and CH2Cl2 was added under nitrogen.
The copper pre-catalyst (1 mg) was added to the reaction flask.
This was hand-stirred for 10 min and phenylhydrazine (20 ul of a
5% solution) was added. Styrene (312 mg, 3 mmol, 345 ul) and then
EDA (1 mmol, 1 ml of an 8% solution in CH2Cl2 diluted with 7 ml of

C.C. Paluti, E.S. Gawalt / Journal of Catalysis 275 (2010) 149–157
153
NMR solvent and integration of the cis/trans peaks along with the
peak due to side products.
2.6.1. Purification and enantiomeric excess determination
The cis and trans isomers were separated from each other by p-
TLC and the solvent ratio of 96/4 hexane/diethyl ether. CH2Cl2 was
used to remove the respective isomer products from the silica gel.
The solvent was evaporated in vacuo to yield colorless oil for both
isomers. The cis and trans isomers were analyzed by 1H NMR to
show separation of the two isomers. Addition of the chiral shift re-
agent europium tris[3-(heptafluoropropylhydroxymethylene)-(+)-
camphorate] to the NMR samples was used to determine the enan-
tiomeric excess [26–29].
3. Results
3.1. Immobilization
Fig. 5. IR spectrum before (dashed) and after (solid) immobilization of the aza-
bis(oxazoline) ligand. The range of the IR spectrum is 3050–2750 cmÀ1
.
Five different monolayer systems (Scheme 1) were prepared by
formation of mixed monolayers on gold in a 1:9 ratio of alkanethi-
ols of 11-mercapto-undecanol to one of the following alkanethiols:
11-mercapto-undeacanol, 11-bromoundecane-1-thiol, 11-mer-
capto-undecanoic acid, 10-mercaptodecanoate, 3-((11-mercapto-
undecyl) propanenitrile) thus generating the five different
environments (Scheme 1). (9-Mercapto-1-nonanol was used in
the place of 11-mercapto-undecanol for the ester system to en-
hance the packing with the shorter chain ester thiol.) The aza-
bis(oxazoline) ligand was then immobilized on the hydroxyl-ter-
minated alkanethiols using a Mitsunobu reaction (Scheme 1).
These background tail groups can interact with the catalyst result-
ing in differences in the stereo- and enantio-selectivity of the
cyclopropanation products.
The catalytic chips were prepared and characterized by MALDI
and IR. Analysis of the chips before and after use in the cycloprop-
anation reaction was accomplished by MALDI and IR. Fig. 4 shows
the MALDI spectrum of system 2a after use in the cyclopropanation
reaction. The peaks at 630.0579 and 516.9364 indicate the pres-
ence of the catalyst attached to the thiol chain. The peak at m/z
516.9364 is due to the aza-bis(oxazoline) copper complex attached
to the thiol and one proton. The peak m/z 630.0579 is due to the
aza-bis(oxazoline) copper complex and five sodium atoms.
In the IR spectra after attachment of the ligand (Fig. 5 solid line),
the stretch (2955 cmÀ1) due to CH of methyl groups has a greater
intensity compared to the intensity before attachment of the li-
gand. This is due to the presence of the ligand and its methyl-con-
taining t-butyl groups. The width of the peak is due to overlapping
CH stretches from CH groups in the ligand and monolayer (Fig. 5).
3.2. Heterogeneous electronic environment
The catalytic chips were then tested in the cyclopropanation
reaction of ethyl diazoacetate and styrene (Scheme 2). In the cyclo-
propanation reaction, phenylhydrazine was used to reduce the
copper (II) to the copper (I) state, which is the active catalyst for
the reaction. Four potential cyclopropane products were obtained
in this reaction (Scheme 2).
The results of the cyclopropanation reaction are tabulated in
Table 1. The stereoselectivity was determined by 1H NMR, and
enantioselectivity was determined by 1H NMR with the aid of the
chiral shift reagent europium tris[3-(heptafluoropropylhydroxym-
ethylene)-(+)-camphorate].
In the homogeneous phase (Scheme 1, 1a), the trans isomer
enantiomeric excess is 87% (Table 1) and cis isomer enantiomeric
excess is 80% for the standard cyclopropanation reaction (Scheme
2). The homogeneous phase gives good enantioselectivity for the
cyclopropanation reaction of styrene and ethyl diazoacetate, while
the introduction of background tail groups impacts the enantiose-
lectivity depending on the choice of background tail group.
The presence of background surface hydroxyl groups (Scheme 1,
2a) has a positive impact on the enantioselectivity of the system
compared to the homogeneous phase (Table 1, 1a). The enantio-
meric excess for the trans isomer is 95% (Table 1). It should be
noted that 95% is the detection limit of our method so these sys-
tems may have higher enantioselectivity. Additionally, for the hy-
droxyl surface (Scheme 1, 2a), a reversal in the favored enantiomer
is seen for the cis isomer (Table 1, 2a). For the cis isomer in the hy-
droxyl electronic environment system, the (1R,2S) enantiomer is
formed in excess over the (1S,2R) enantiomer which is seen in
the homogeneous phase. This reversal was observed in some of
our previous immobilized systems [17].
The enantioselectivity (95% Table 1, 3a) of the bromine-termi-
nated surface (Scheme 1, 3a) is enhanced for the trans isomer com-
pared to the homogeneous phase (Table 1, 1a). This
enantioselectivity is comparable to the hydroxyl electronic envi-
ronment (Table 1, 2a). The presence of the surface bromines also
Fig. 4. MALDI spectrum 2a after use in the cyclopropanation reaction. The peak at
m/z 516.9364 is due to the aza-bis(oxazoline) copper complex attached to the thiol
and one proton. The peak m/z 630.0579 is due to the aza-bis(oxazoline) copper
complex and five sodium atoms. The other peaks are due to the matrix.

154
C.C. Paluti, E.S. Gawalt / Journal of Catalysis 275 (2010) 149–157
Scheme 2. Cyclopropanation reaction of ethyl diazoacetate and styrene.
Table 1
Cyclopropanation results of the homogeneous and heterogeneous bis[4,5-dihydro-(4S)-(1,1-dimethyl)-1,3-oxazole-2-yl]-amine copper(I) catalyst in the reaction of styrene and
ethyl diazoacetate in dichloromethane.a
Catalytic system
Cis/trans ratiob
Cis ee% (1S,2R)c
Trans ee% (1S,2S)c
Yieldb,d
1a
2a
3a
4a
5a
6a
20/80 ( 1.5)e
14/86 ( 1.4)
20/80 ( 1.4)
19/81 ( 1.4)
14/86 ( 1.3)
16/84 ( 1.6)
80 ( 1.1)
80f ( 1.0)
62f ( 0.8)
6 ( 0.9)
51 ( 0.8)
80 ( 1.0)
87 ( 1.1)
95 ( 0.5)
95 ( 0.6)
50 ( 0.8)
49 ( 0.7)
78 ( 0.8)
71
70
73
67
64
75
a
Statistical treatment performed by Student’s t-test with a 99.9% confidence level. Ten degrees of freedom and critical value of t equals 4.59. Two systems are statistically
different when test statistic t is greater then the critical value of t.
b
Determined by 1H NMR.
c
Determined by 1H NMR with the aid of chiral shift reagent.
d
Isolated yield.
Sample standard deviation.
The (1R,2S) enantiomer is formed in excess of the (1S,2R).
e
f
causes a reversal in the favored enantiomer for the cis isomer. The
same effect as observed in the hydroxyl system. However, the
enantioselectivity is slightly reduced compared to the hydroxyl
environment.
3.3. Homogeneous electronic environment
To determine whether mimicking the effects of the electronic
systems was possible in the homogeneous system, various solvents
were utilized. The homogeneous cyclopropanation reaction
(Scheme 2) was performed in the solvents presented in Table 2.
When the cyclopropanation reaction (Scheme 2) was performed
in methanol or acetonitrile the selectivity of the system decreased
(Table 2) compared to dichloromethane, which is the standard
solvent for the reaction. However, when the cyclopropanation
reaction was performed in ethyl acetate and acetic acid only
trace amounts of the cyclopropane products were obtained
(Table 2) and this was not enough for proper cis/trans ratio deter-
mination. Thus, changing solvents effectively showed that the
stereo- and enantio-selectivity found in the heterogeneous
systems is unique.
The presence of interfacial carbonyl groups (Scheme 1, 4a and
5a) causes the enantioselectivity of the system to decrease signifi-
cantly compared to the homogeneous phase (Table 1, 1a). This is
seen in both the carboxylic acid and ester cases. The enantiomeric
excess for the trans isomer is 50% (Table 1, 4a) for the carboxylic
acid surface (Scheme 1, 4a) and a comparable enantiomeric excess
for the trans isomer (49% Table 1, 5a) is obtained for the ester sur-
face (Scheme 1, 5a). This is a dramatic decrease in the enantiomeric
excess when compared to the homogeneous phase (87% Table 1,
1a).
The stereoselectivity of the catalytic system with background
nitrile groups (Table 1, 6a) is slightly enhanced compared to the
homogeneous phase (Table 1, 1a). The presence of surface back-
ground nitrile groups gave enantioselectivity comparable to the
homogeneous phase for the cis isomer. However, the addition of
background nitrile groups had a negative impact the enantioselec-
tivity for the trans isomer compared to the homogeneous phase.
4. Discussion
Overall modification of the surface environment around the
aza-bis(oxazoline) copper catalyst influenced the selectivity of
Table 2
Cyclopropanation results of the homogeneous bis[4,5-dihydro-(4S)-(1,1-dimethyl)-1,3-oxazole-2-yl]-amine copper (I) catalyst in the reaction of styrene and ethyl diazoacetate in
various solvents.
Solvent
Cis/trans ratioa
Cis ee% (1S,2R)b
Trans ee% (1S,2S)b
Yielda,c
Dichloromethane
Methanol
Acetonitrile
Ethyl acetate
Acetic acid
20/80
40/60
37/63
–
–
80
48
71
–
87
53d
64
–
71
45
68
e
Trace amount
Trace amount
–
–
a
b
c
Determined by 1H NMR.
Determined by 1H NMR with the aid of chiral shift reagent.
Isolated yield.
d
e
The (1R,2S) enantiomer is formed in excess over the (1S,2R) enantiomer.
Not determined due to trace amount of product formation.

C.C. Paluti, E.S. Gawalt / Journal of Catalysis 275 (2010) 149–157
155
Fig. 6. (A)–(C), and (E) are the potential intermolecular interactions between the catalytic site and the hydroxyl surface. (D) is the potential interaction between the catalytic
site and the bromine surface. (F) is the potential interaction between the catalytic site and the carbonyl surfaces.
Fig. 7. Proposed reaction pathways.
the systems in the cyclopropanation reaction of styrene and ethyl
diazoacetate. The selectivity obtained in the heterogeneous phase
(Table 1) could not be obtained in the homogenous phases (Table
2) where the solvents contained the same functional groups; hy-
droxyl, nitrile, carboxylic acid, and ester. This demonstrates that
these functional groups are only effective in the heterogeneous

156
C.C. Paluti, E.S. Gawalt / Journal of Catalysis 275 (2010) 149–157
phase to give their respective enhancements or reductions in
enantioselectivity.
tion of the experimentally favored enantiomers (Fig. 8). The pres-
ence of surface carbonyl groups allows for surface stabilization of
the electropositive center of the reacting alkene thus generating
the opposite enantiomers of the trans and cis isomers (Fig. 9).
The ability to stabilize the electropositive center of the reacting al-
kene has a negative effect on the enantioselectivity of the system
because this allows for the production of both enantiomers of
the trans and cis isomer.
If both points of stabilization (Figs. 8 and 9) are available, then a
mixture of the two enantiomers will be formed thus reducing the
enantiocontrol of the catalytic system. For the substrates where
surface carbonyl groups are present, the enantioselectivity of the
system decreases (Table 1). These results support the influence of
the secondary effect in the selectivity of our heterogeneous cata-
lytic systems.
The control of enantioselectivity by modification of the elec-
tronic environment around the catalytic site is believed to be
caused by intermolecular interactions between the background
surface molecules and the copper carbene. Fig. 6 demonstrates
the potential intermolecular interactions that may occur with each
respective surface. In the hydroxyl and bromine surfaces, the elec-
tric forces cause the catalytic site to orient closer to the monolayer
surface therefore enhancing surface steric influence on the cata-
lytic site and selectivity. The alignment of the catalytic site and
the background surface molecules according to attraction of their
partially positive and partially negative regions enhances the sur-
face steric influence thus combining both an electronic (dipole–di-
pole or hydrogen bonding) effect along with steric effect. The
enantiocontrol for both the hydroxyl and bromine systems are
comparable for the trans isomer. These electric forces would favor
the formation of trans isomer B as seen in Fig. 7.
5. Conclusion
A hydrogen bonding interaction between the copper carbene
and the hydroxyl tail groups accounts for the reversal in the favored
enantiomer for the cis isomer (Fig. 7). In Fig. 7, it can be seen that
because of this hydrogen bonding interaction with the surface hy-
droxyl groups the (1R,2S) is favored over the (1S,2R) enantiomer
for the cis isomers. This environment allows for hydrogen bonding
between the surface hydroxyl groups and the ester of the copper
carbene stabilizing the same interaction seen in our past catalytic
study [17]. Increasing the concentration of the background surface
hydroxyl groups has a positive impact on the enantioselectivity for
the trans isomer. Due to the strong intermolecular interaction of
hydrogen bonding, this orientation is favored over that of the ester
of the copper carbene positioned away from the monolayer surface.
In the bromine system, an enhancement in enantioselectivity
was also observed. This may be due to dipole–dipole interactions
between the copper carbene and the electronegative bromine tail
groups. This dipole–dipole interaction causes the catalyst to orient
closer to the monolayer surface.
In an immobilized catalytic system, modification of the elec-
tronic character of the molecules around the catalyst influences
the selectivity of the catalyst in the cyclopropanation reaction of
ethyl diazoacetate and styrene. The presence of surface hydroxyl
and bromine groups enhances the enantioselectivity of the trans
isomer. Also presence of these surface groups causes the reversal
of the favored enantiomer for the cis isomer. Carbonyl groups re-
duce the enantioselectivity. To demonstrate the fact that these
functional groups can only be present in the heterogeneous phases
to give their respective enhancements or reductions in selectivity,
homogenous systems were evaluated. The solvent of the reaction
was changed to mimic the heterogeneous systems. Comparison
of the homogeneous and heterogeneous electronic phases did not
show a similar trend in the selectivity. The homogeneous systems
showed greatly reduced selectivity and yield of cyclopropane prod-
ucts for all solvents. Our group has developed a simple method of
influencing enantioselectivity of the aza-bis(oxazoline) copper cat-
alyst by modification of the electronic environment around the
catalyst.
In the homogeneous phase, the secondary effect stabilizes the
electropositive center of the reacting alkene leading to the forma-
References
*Duquesne University
[1] C. Suckling, J. Angew. Chem. Int. Ed. Engl. 27 (1988) 537.
[2] H.N.C. Wong, M.Y. Hon, C.W. Tse, Y.C. Yip, J. Tanko, T. Hudlicky, Chem. Rev. 89
(1989) 165.
[3] (a) T. Ye, M.A. McKervey, Chem. Rev. 94 (1994) 1091;
(b) V.K. Singh, A.D. Gupta, G. Sekar, Synthesis (1997) 137;
(c) M.P. Doyle, M.N. Protopopova, Tetrahedron 54 (1998) 7919.
[4] H. Lebel, J.F. Marcoux, C. Molinaro, A.B. Charette, Chem. Rev. 103 (2003) 977
(and references therein).
[5] M.P. Doyle, D.C. Forbes, Chem. Rev. 98 (1998) 911.
[6] A. Pfaltz, Cyclopropanation and C–H insertion with Cu, in: E.N. Jacobsen, A.
Pfaltz, H. Yamamoto (Eds.), Comprehensive Asymmetric Catalysis, vol. 2,
Springer-Verlag, Berlin, 1999, pp. 513–538.
Fig. 8. Homogeneous phase secondary effect.
[7] M. Glos, O. Reiser, Org. Lett. 2 (2000) 2045.
[8] K.M. Lydon, M.A. McKervey, Cyclopropanation and C–H insertion with Rh, in:
E.N. Jacobsen, A. Pfaltz, H. Yamamto (Eds.), Comprehensive Asymmetric
Catalysis, vol. 2, Springer-Verlag, Berlin, 1999, pp. 539–580.
[9] (a) For recent examples see: T. Ikeno, I. Iwakura, S. Yabushita, T. Yamada, Org.
Lett. 4 (2002) 517;
(b) T. Ikeno, I. Iwakura, T. Yamada, J. Am. Chem. Soc. 124 (2002) 15152;
(c) T. Ikeno, I. Iwakura, T. Yamada, Org. Lett. 6 (2004) 949.
[10] D. Rechavi, M. Lemaire, Chem. Rev. 102 (2002) 3467.
[11] Q.-H. Fan, Y.–M. Li, A.S.C. Chan, Chem. Rev. 102 (2002) 3385.
[12] A. Mandoli, S. Orlandi, D. Pini, P. Salvadori, Chem. Commun. 19 (2003) 2466.
[13] H. Wang, W. Sun, C. Xia, J. Mol. Catal. A: Chem. 206 (2003) 199.
[14] M. Diaz-Requejo, T. Beldevrain, C. Nicasio, P. Perez, Organometallics 19 (2000)
285.
[15] K. Hara, R. Akiyama, S. Takakusagi, K. Dosaki, T. Yoshino, H. Kagi, M. Sawamura,
Angew. Chem. Int. Ed. 47 (2008) 5627.
[16] T. Belser, M. Stohr, A. Pfaltz, J. Am. Chem. Soc. 127 (2005) 8720.
[17] C.C. Paluti, E.S. Gawalt, J. Catal. 267 (2009) 105.
Fig. 9. Heterogeneous phase secondary effect where surface carbonyl groups are
present.
[18] T. Pintauer, J. Organometallic Chem. 691 (2006) 3948.

C.C. Paluti, E.S. Gawalt / Journal of Catalysis 275 (2010) 149–157
157
[19] C. Ricardo, T. Pintauer, J. Organometallic Chem. 692 (2007) 5165.
[24] H.M.L. Davies, T.J. Clarks, L.A. Church, Tetrahedron Lett. 30 (1989) 5057.
[25] P.E. O’Bannon, W.P. Dailey, Tetrahedron 46 (1990) 7341.
[26] P. Reisberg, I. Brenner, I. Jerome, J. Pharmaceut. Sci. 65 (1976) 592.
[27] Q. Wang, M. Mayer, C. Brennan, F. Yang, M. Hossain, D. Grubisha, D. Bennett,
Tetrahedron 56 (2000) 4881.
[28] H. Goering, J. Eikenberry, G. Koermer, J. Am. Chem. Soc. 93 (1971) 5913.
[29] Q. Wang, H. Forsterling, M. Hossain, Organometallics 21 (2002) 2596.
[20] Doyle P. Micheal, McKervey M. Anthony, Tao Ye, Modern Catalytic Methods for
Organic Synthesis with Diazo Compounds from Cyclopropanation to Ylides,
Wiley-Interscience Publication, 1998.
[21] M.P. Doyle, J.H. Griffin, V. Bagheri, R.L. Dorow, Organometalics 3 (1984) 53.
[22] M.P. Doyle, R.L. Dorow, W.E. Buhro, J.H. Griffin, W.H. Tamblyn, M.L. Trudell,
Organometalics 3 (1984) 44.
[23] M.P. Doyle, K.L. Loh, K.M. DeVries, M.S. Chinn, Tetrahedron Lett. 28 (1987) 833.