N-(4-[2.2]paracyclophanyl)-2′-hydroxyacetophenone imine: An effective paracyclophane Schiff-base ligand for use in catalytic asymmetric cyclopropanation reactions

Article abstract of DOI:10.1016/j.molcata.2012.05.009

The synthesis of planar chiral N-(4-[2.2]paracyclophanyl)-2′- hydroxyacetophenone imine (5) and its use as a ligand for the copper(II) catalyzed cyclopropanation of styrene and stilbene derivatives using diazoesters is presented. Catalyst loadings of 0.1

Full text of DOI:10.1016/j.molcata.2012.05.009

Journal of Molecular Catalysis A: Chemical 361–362 (2012) 111–115
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
N-(4-[2.2]paracyclophanyl)-2ꢀ-hydroxyacetophenone imine: An effective
paracyclophane Schiff-base ligand for use in catalytic asymmetric
cyclopropanation reactions
Douglas S. Mastersonb, Caitlyne Shirleyb, Daniel T. Glatzhofera,∗
a Department of Chemistry and Biochemistry, The University of Oklahoma, Norman, OK 73019, United States
b Department of Chemistry and Biochemistry, The University of Southern Mississippi, MS 39406, United States
a r t i c l e i n f o
a b s t r a c t
The synthesis of planar chiral N-(4-[2.2]paracyclophanyl)-2ꢀ-hydroxyacetophenone imine (5) and its
use as a ligand for the copper(II) catalyzed cyclopropanation of styrene and stilbene derivatives using
diazoesters is presented. Catalyst loadings of 0.1 mol% gave conversions of >80% (turnover numbers
approaching 1000) for styrene and its derivatives. When enantiomerically enriched (R)-5 was used to
form the catalyst for cyclopropanation of styrene using ethyldiazoacetate, the cyclopropane products
were obtained in a trans/cis ratio of 1.9–1 and 75.8% and 60.5% ee (corrected), respectively. The use of
t-butyldiazoacetate resulted in an increased trans/cis ratio of 4.6–1 and 88.2% and 77.9% ee, respectively.
Enantioselectivities of up to 95% ee were observed. These are among the highest enantioselectivities
observed for asymmetric reactions using catalysts where chirality is solely derived from the paracyclo-
phanyl moiety. When compared to its non-methylated analog, the simple presence of a methyl group on
the carbon of the imine moiety in 5 resulted in an average increase in enantioselectivity of ca. 60% ee for
a variety of substrates. The origin of this dramatic improvement in selectivity is discussed.
© 2012 Elsevier B.V. All rights reserved.
Article history:
Received 2 January 2012
Received in revised form 8 May 2012
Accepted 9 May 2012
Available online 18 May 2012
Keywords:
Chiral [2.2]paracyclophane derivatives
Chiral Schiff-base
Asymmetric cyclopropanation
Asymmetric catalysis
Copper(II)-catalyzed cyclopropanation
1. Introduction
from the planar chirality of the [2.2]paracyclophane moiety alone;
most ee’s in such cases being low to moderate. The development
Since the seminal work of Pye and co-workers on the use of
chiral bisphosphino[2.2]paracyclophane ligands for the catalytic
asymmetric hydrogenation of pro-chiral alkenes in high enan-
tiomeric excess (ee) [1], there has been continual interest in the
use of non-racemic [2.2]paracyclophane derivatives as sources of
planar chirality in asymmetric transformations. Chiral catalysts
have been constructed from asymmetric [2.2]paracyclophanes and
used in a variety of different reactions [2–5], mainly for asym-
metric 1,2- and 1,4-addition reactions to carbonyls [2,3,6,7] and
asymmetric hydrogenations [1,8,9]. However, the vast majority
of these [2.2]paracyclophane-based ligands incorporate a second,
non-[2.2]paracyclophane-derived chiral center around the active
catalytic site in the ligand construction in order to obtain high ee’s.
Aside from the bisphosphine-type [2.2]paracyclophane catalysts
[1,8,9], and a few other notable exceptions [10,11], chiral [2.2]para-
cyclophane ligands have not generally achieved high ee’s stemming
of highly effective [2.2]paracyclophane ligands where the catalytic
asymmetric induction clearly arises solely from the cyclophane
moiety, is therefore of considerable interest, both synthetically and
from a ligand design perspective.
Attempts at constructing highly effective asymmetric cyclo-
propanation catalysts using [2.2]paracyclophane moieties as the
sources of chirality have proven difficult. Vogtle et al. have pub-
lished the synthesis and use of novel bipyridine ligands based
on the [2.2]pyridinophane unit (1) and the [2.2]quinolinophane
unit (2) (Fig. 1) [12,13]. However, ligands 1 and 2 required a
lengthy construction of the chiral paracyclophane scaffold and
the asymmetric induction of 1 and 2 with respect to the cop-
per catalyzed cyclopropanation of styrene with ethyldiazoacetate
was low (∼26% ee). Previous studies using N-salicylidene-4-
amino[2.2]paracyclophane (3) as a chiral ligand in the copper(II)
catalyzed cyclopropanation of styrene with diazoesters resulted in
moderate enantioselectivities (40% ee for the trans isomer) [14].
Ligand 3 proved to be highly substrate and diazoester sensitive
with respect to the % ee of the resulting cyclopropane products
(for select results see Table 1B). Adding steric bulk in the form of
tert-butyl groups to the salicylidene moiety improved the enan-
tioselectivity of the cyclopropanation reactions [15]. Using N-(2ꢀ,4ꢀ-
di-tert-butyl)salicylidene-4-amino[2.2]paracyclophane (4) as the
∗
Corresponding author at: Department of Chemistry and Biochemistry, Stephen-
son Life Sciences Research Center, 101 Stephenson Parkway, The University of
Oklahoma, Norman, OK 73019, United States. Tel.: +1 405 325 3834;
fax: +1 405 325 6111.
E-mail address: dtglatzhofer@ou.edu (D.T. Glatzhofer).
1381-1169/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2012.05.009

112
D.S. Masterson et al. / Journal of Molecular Catalysis A: Chemical 361–362 (2012) 111–115
Fig. 1. Chiral cyclophanes used as ligands for catalytic cyclopropanation.
asymmetric ligand resulted in a 67% ee for the cyclopropanation
of styrene (trans isomer) [15]. Although 4 proved to give better
results than 3 in terms of the % ee of the products, 4 was also very
substrate sensitive.
2. Experimental
Cyclopropanation
reactions
were
performed
as
previously described [14]. Enantiomerically enriched (R)-4-
amino[2.2]paracyclophane was prepared as described in the
literature [22]. 2-Hydroxyacetophenone was obtained from
Aldrich Chemical Company, Inc. and used as received. 1H NMR
spectra were recorded on a Varian Unity 400 MHz spectrometer,
13C NMR spectra were recorded on a Varian XL300 spectrometer
and the spectra were referenced to internal solvent resonances.
The enantiomeric purity of (R)-5 and cyclopropane products was
determined using a Chiralcel OJ HPLC column with either a BioRad
UV/Vis detector at 254 nm and SSI pump system or an HPLC APCI
MS system.
It was suspected that for the pyridineophane ligands 1 and 2
the low % ee’s that were observed in the cyclopropane products
were probably due to a relatively open, planar lower face result-
ing in cyclopropanation with little chiral discrimination. In ligands
3 and 4 the low to moderate % ee’s observed were likely due
primarily to rotation of the cyclophane unit about the aromatic-
N bond resulting in conformations with relatively open reaction
sites [14]. Molecular models suggested that replacing the imine
hydrogen in ligand 3 with a methyl group would hinder rotation
of the chiral cyclophane unit significantly and potentially increase
enantioselectivity with respect to the cyclopropane products. Few
highly enantioselective cyclopropanations have been performed
using salen-type ligands [16,17] since the historic asymmetric
cyclopropanation reaction of Nozaki et al. [18,19], which utilized
a copper(II)-based salen-type catalyst system, and the related pio-
neering work of Aratani [20,21]. Herein we report the synthesis of
the title compound (5), a salen ligand, which gave results directly
comparable to those of Nozaki et al. (see Fig. 2), and its effective use
as a ligand in the copper-catalyzed, asymmetric cyclopropanation
of various substrates. The simple addition of a methyl group to the
imine moiety in cyclophane ligand 3 resulted in a dramatic 2- to 7-
fold increase in % ee’s of the products, achieving high % ee’s for most
substrates. Ligand 5 is easily synthesized from the readily available
[2.2]paracyclophane (6), thus alleviating any lengthy construction
of the paracyclophane scaffold.
2.1. Synthesis of (R)-5
A
50 mL pear shaped flask was charged with 0.378 g
(1.69 E−3 mol) of enriched (R)-4-amino[2.2]paracyclophane, ca.
20 mg of p-TsOH, 20 mL of toluene, and a stir bar. The flask was fitted
with a Hickman still and a reflux condensor and the resulting solu-
tion was heated to reflux solvent. After all the water was removed
azeotropically from the reaction vessel, fresh toluene was added to
bring the volume to 20 mL. The solution was again heated to reflux
solvent and 0.249 g (1.83 E−3 mol) of 2-hydroxyacetophenone was
added by syringe over a 30 min period. The resulting solution was
heated to reflux solvent overnight. The Hickman still was removed
and activated 4 A molecular sieves were added and the solution
was heated to reflux solvent an additional 24 h. The resulting
yellow mixture was filtered through a 1 in. pad of neutral alu-
mina which was subsequently washed with fresh toluene. The
solvent was removed under reduced pressure to obtain 0.3295 g
(9.650 E−4 mol) of crude (R)-5 (∼57% yield). The crude (R)-5 was dis-
solved in 10 mL of boiling methanol, cooled to room temperature,
and hexanes were added until the solution became cloudy. This
cloudy mixture was placed in a freezer for 48 h and the solid (R)-5
microcrystals were collected and rinsed with cold hexanes. The %
ee of (R)-5 was determined using an analytical Chiralcel OJ column
(1:9 i-PrOH/hexanes at 1.00 mL/min; (S)-5 18 min, (R)-5 23 min).
The % ee was determined to be 86.3% by HPLC. MP (racemic):
178–179.5 ◦C. 1H NMR (400, CDCl3): ı 2.18 (3H,s), 2.93–2.99 (1H,
m), 3.06–3.23 (8H, m with 1 D2O exchangeable proton), 5.76 (1H,
s), 6.44–6.53 (4H, m), 6.60–6.62 (1H, dd 2 Hz, 8 Hz), 6.90 (1H, t,
7 Hz), 7.10 (2H, d 8 Hz), 7.41 (1H, t, 7 Hz), 7.60 (1H, d 8 Hz). 13C NMR
(75 MHz, CDCL3): ı 17.46, 32.33, 34.25, 35.08, 35.41, 118.09, 118.21,
119.67, 127.68, 128.80, 128.99, 130.07, 132.06, 132.27, 132.79,
133.10, 133.30, 134.90, 138.97, 139.58, 140.84, 143.98, 162.21,
Fig. 2. Visual comparison of (R)-5 with Nozaki’s ligand.

D.S. Masterson et al. / Journal of Molecular Catalysis A: Chemical 361–362 (2012) 111–115
113
Scheme 1. Synthesis of (R)-5 from [2.2]paracyclophane (6). CSA, HAP, and p-TsOH stand for 10-camphorsulfonic acid, 2-hydroxyacetophenone, and para-toluenesulfonic
acid, respectively.
169.95. HRMS (FAB): m/z calcd for C24
found 342.1873.
H
24NO (M+H+) 342.1858
with the aromatic hydrogens in the ortho, pseudo-ortho, and
pseudo-gem positions relative to the imine nitrogen on the para-
cyclophane moiety. It is therefore likely that the global minimum
energy conformation of 5 is one in which the methyl group is exo,
causing the 2ꢀ-hydroxyacetophenone imine functionality involved
in complexation to favor an endo conformation, and resulting in
the catalytic metal center residing in a more sterically demanding
environment. Although a similar qualitative analysis can be made
for the corresponding ligand 3, the expected amplification of the
conformational effects by replacing the hydrogen on the imine
carbon with a methyl group suggested that enantioselectivities
for cyclopropanations using ligand 5 to form the copper catalyst
would improve relatively.
Due to its low solubility, completely resolved (R)-5 was not read-
ily available. Nonetheless, several milligrams of (R)-5 were resolved
further on a semi-preparative Chiralcel OJ column (1:9 i-PrOH) in
a lengthy process. The sample was determined by Chiral HPLC to
be >99% R. This sample was used to perform a cyclopropanation
(see Table 1) as a check of the validity of correction method for
calculating reaction ee’s (vide infra).
3. Results and discussion
The title compound (R)-5 was constructed as shown in
Enriched (R)-5 was complexed with copper(II) and used in the
copper catalyzed cyclopropanation reaction of various substrates
with both ethyldiazoacetate (EDA) and tert-butyldiazoacetate
(TBDA). The results of the cyclopropanation reactions using
enriched (R)-5 are presented in Table 1A and the % ee’s are cor-
rected for the % ee of ligand (R)-5 [23]. Included for comparison in
Table 1B are selected cyclopropanation results using ligand 3 [14].
When styrene (9) was used as the substrate (Table 1A, entry 1),
and (R)-5 as the ligand, an ee of 75.8% was obtained for the trans iso-
mer and a 60.5% ee was obtained for the cis isomer. This is a dramatic
improvement over ligand 3 as shown in Table 1B, entry 7. When the
non-methylated ligand 3 was used as the ligand in the cyclopropa-
nation of styrene with EDA an ee of 27.4% was obtained for the trans
product and the cis product showed an ee of 12.7%. The cyclopropa-
nation of ␣-methylstyrene (10) with EDA using ligand (R)-5 gave an
ee of 84.3% for the trans product and a 95.0% ee for the cis product.
When methyl substituted (R)-5 was used in the cyclopropanation
Scheme 1. The key step of Scheme
1 is the previously
reported treatment of racemic 4-amino[2.2]paracyclophane (7)
with (+)-10-camphorsulfonic acid (CSA) to obtain enantiomer-
ically enriched (R)-7 [22]. Enriched (R)-7 was condensed with
2-hydroxyacetophenone (8) in the presence of p-TsOH to produce
the title compound (R)-5. After recrystallization, (R)-5 was sub-
jected to chiral HPLC analysis which revealed a 86.3% enantiomeric
purity. The solubility of 5 in solvents appropriate for further chiral
HPLC resolution was very low. However, since the title compound
(R)-5, produces a putative single site catalyst upon complexation
with copper(II), it was decided to use the ligand without further res-
olution and to correct the % ee of the cyclopropane products for the
%ee of the ligand (R)-5 [23]. A small amount of highly resolved (R)-5
was isolated and used for one cyclopropanation reaction to confirm
that use of this correction provided valid results. In principal, enan-
tiomerically pure 5 could be obtained by using completely resolved
7. The synthetic route shown in Scheme 1 has the advantage that
the construction of the [2.2]paracyclophane moiety is unnecessary,
which gives rise to a simple and rapid construction of salen-type
ligand 5.
It was initially suspected that 5 would be more structurally
rigid with respect to the metal center to the cyclophane moiety
than ligands 3 and 4. Molecular modeling suggested that rotation
about the cyclophane–nitrogen bond would be restricted due to
interactions of the methyl group in 5 with the ethano bridges of the
cyclophane moiety (Fig. 3), causing it to rotate in an exo fashion to
the exterior of the cyclophane. Further exo rotation of the methyl
group eventually brings the methyl group into close proximity
Fig. 3. Steric interaction of the imine methyl group with an ethano paracyclophane
bridge in ligand 5.

114
D.S. Masterson et al. / Journal of Molecular Catalysis A: Chemical 361–362 (2012) 111–115
Table 1A
Cyclopropanations using ligand 5.
Entry
Substrate
Diazoestera
% Conv.b
Turnover #c
Trans/Cis
% ee Transd
% ee Cisd
1
2
3
4
5
Styrene
Styrene
␣-Methylstyrene
␣-Phenylstyrene
Trans-stilbene
EDA
TBDA
EDA
EDA
EDA
93
81
86
85
20
929
807
858
847
202
1.9:1
4.6:1
3.3:1
–
75.8
88.2
84.3
74.4e
23.4e
23.8e,f
–
60.5
77.9
95.0
–
–
–
6
Cis-stilbene
EDA
22
220
16.8:1g
–
a
EDA, ethyldiazoacetate; TBDA, tert-butyldiazoacetate.
b
c
d
e
f
Determined by gas chromatography.
Calculated using the following equation: (mole olefin × % conversion)/mole catalyst.
Determined using a Chiralcel OJ HPLC analytical column and the % ee was corrected for the % ee of the ligand using the following equation: (observed ee/ligand ee) × 100.
The terms cis and trans are irrelevant.
Run using >99% (R)-5 to check validity of d.
Refers to exo/endo ratio.
g
Table 1B
Results using ligand 3 [14].
Entry
Substrate
Diazoester
% Conv.
Turnover#
Trans/cis
% ee Trans
% ee Cis
7
8
9
10
11
Styrene
Styrene
␣-Methylstyrene
␣-Phenyl styrene
Trans-stilbene
EDA
TBDA
EDA
EDA
EDA
96
96
91
93
39
960
960
910
930
390
2.4:1
5.9:1
1.8:1
–
27.4
40.5
18.2
9.5a
12.7
12.7
13.5
–
–
11.8a
–
a
The terms cis and trans are irrelevant.
reaction of ␣-phenylstyrene (Table 1A, entry 4), an ee of 74.4% was
obtained. When ␣-phenylstyrene was used as the substrate in the
catalytic cyclopropanation reaction using non-methylated ligand
3, the resulting cyclopropane product was obtained in less than
10% ee. It is clear from the results presented in Table 1A that the
ability of 5 to induce enantioselectivity in the cyclopropane prod-
ucts is a substantial improvement over ligands 1–4. It is possible
that the improved enantioselectivity in the cyclopropanation reac-
tion using ligand 5 could be due in part to the electronic effects of
the electron donating methyl substituent. However, the improved
enantiomeric induction of 5 is largely attributable to restricted
rotation about the cyclophane–nitrogen bond as discussed earlier.
Restricted rotation prevents the bulky chiral cyclophane unit from
orienting itself away from the catalytic copper center, allowing it
to efficiently transfer its chiral information to the resulting cyclo-
propane products.It is also worth noting that the conversions of
substrate to product were quite good when (R)-5 was used as the
ligand. Table 1A illustrates that conversions were typically >80% for
the styrene derivatives, with turnover numbers approaching 1000
(catalyst loading 0.1 mol%). The advantage of low catalyst load-
ings is that unwanted diazo coupling products are not detected
in these reactions. To the best of our knowledge this is the low-
est catalyst loading used for a salen-type ligand. Substituting the
chiral phenethylamine unit in Nozaki’s original ligand with the
more bulky chiral 4-amino[2.2]cyclophane has a dramatic effect on
the % ee of the resulting cyclopropane products (Fig. 2). The cyclo-
propanation of styrene with EDA in the presence of Nozaki’s ligand
resulted in ee’s less than 10% for both the cis and trans cyclopropane
products [18], whereas salen ligands 3 and 5 resulted in much
improved enantioselectivities (Tables 1A, entry 1 and 1B, entry 7).
This shows that the simple chiral [2.2]paracyclophanyl moiety can
be much more effective in chiral induction compared to the chi-
ral phenethylamine unit of Nozaki’s ligand and is indicative of the
potential for the sterically demanding [2.2]paracyclophanyl unit to
construct highly effective chiral ligands for asymmetric catalysis.
2-hydroxyacetophenone and that 5 can be used as a ligand for
copper-catalyzed asymmetric cyclopropanation of styrene deriva-
tives. The simple substitution of a methyl group for the imine
hydrogen of 3 makes the cyclophane ligands much more effective
for asymmetric induction. Cyclopropanations of styrenes run using
5 as a ligand showed a dramatic average increase of 60% ee (2-
to 8-fold increases in % ee) compared to those run with ligand 3.
Ligand 5 was also less substrate sensitive with respect to enantios-
electivity while maintaining a good catalytic turnover. We have
grown crystals of ligands 3 and 5 and are currently carrying out X-
ray crystal structure analysis to obtain exact atom coordinates for
computational studies of the conformation differences between 3
and 5. These studies, together with spectroscopic studies of 3 and
5 in solution to elucidate the exact nature of the factors behind
the dramatic increase in asymmetric induction with ligand 5, are
ongoing.
Acknowledgments
The authors would like to thank Maureen Smith and Chaz Bur-
rows for technical assistance. Support from the Department of
Education, the National Science Foundation (MCB 0844478, CHE
0639208, CHE 0840390), the University of Oklahoma, and the Uni-
versity of Southern Mississippi is greatly appreciated.
References
[1] P.J. Pye, K. Rossen, R.A. Reamer, N.N. Tsou, R.P. Volante, P.J. Reider, J. Am. Chem.
Soc. 119 (1997) 6207–6208.
[2] S. Braese, in: M. Christmann, S. Braese (Eds.), Asymmetric Synthesis, 2nd ed.,
Wiley-VCH Publishers, Weinheim, 2007, pp. 67–71.
[3] S. Braese, in: D. Enders, K.-E. Jaeger (Eds.), Asymmetric Synthesis with Chemical
and Biological Methods, Wiley-VCH Publishers, Weinheim, 2007, pp. 196–214.
[4] X.-W. Wu, W. Sun, X.-L. Hou, Youji Huaxue 23 (2003) 906–913.
[5] S.E. Gibson, J.D. Knight, Org. Biomol. Chem. 1 (2003) 1256–1269.
[6] O. Plietzsch, C.I. Schilling, M. Nieger, T. Muller, S. Braese, Tetrahedron: Asym-
metry 21 (2010) 1474–1479.
[7] N.V. Vorontsova, G.S. Bystrova, D.Y. Antonov, A.V. Vologzhanina, I.A. Godovikov,
M.M. Il’in, Tetrahedron: Asymmetry 21 (2010) 731–738.
[8] T.-Z. Zhang, D. Chen, X.-L. Hou, L.-X. Dai, Gaodeng Xuexiao Huaxue Xuebao 29
(2008) 2015–2019.
[9] M.N. Cheemala, M. Gayral, J.M. Brown, K. Rossen, P. Knochel, Synthesis 24
(2007) 3877–3885.
4. Conclusion
It has been demonstrated that
structed from commercially available [2.2]paracyclophane and
5 can be readily con-

D.S. Masterson et al. / Journal of Molecular Catalysis A: Chemical 361–362 (2012) 111–115
115
[10] D.Y. Antonov, V.I. Rozenberg, T.I. Danilova, Z.A. Starikova, H. Hopf, Eur. J. Org.
[16] T. Ikeno, M. Sato, T. Yamada, Chem. Lett. 28 (1999) 1345–1346.
[17] T. Yamada, T. Ikeno, H. Sekino, M. Sato, Chem. Lett. 28 (1999) 719–720.
[18] H. Nozaki, S. Moriuti, R. Noyori, Tetrahedron Lett. 7 (1966) 5239–5244.
Chem. 6 (2008) 1038–1048.
[11] I. Sato, A. Ohno, Y. Aoyama, T. Kasahara, K. Soai, Org. Biomol. Chem. 1 (2003)
244–246.
[12] U. Worsdorfer, F. Vogtle, M. Nieger, M. Waletzke, S. Grimme, F. Glorius, A. Pfaltz,
Synthesis 31 (1999) 597–602.
[13] U. Worsdorfer, F. Vogtle, F. Glorius, A. Pfaltz, J. Prakt. Chem. 341 (1999) 445–448.
[14] D.S. Masterson, T.L. Hobbs, D.T. Glatzhofer, J. Mol. Catal. A: Chem. 145 (1999)
75–81.
[19] H. Nozaki, H. Takaya, S. Moriuti, R. Noyori, Tetrahedron Lett.
3655–3679.
9
(1968)
[20] T. Aratani, Y. Yoneyoshi, T. Nagase, Tetrahedron Lett. 23 (1982) 685–688.
[21] T. Aratani, Pure Appl. Chem. 57 (1985) 1839–1844.
[22] A. Cipiciani, F. Finguelli, V. Mancini, O. Piermatti, F. Pizzo, R. Ruzziconi, J. Org.
Chem. 62 (1997) 3744–3747.
[15] D.S. Masterson, D.T. Glatzhofer, J. Mol. Catal. A: Chem. 146 (2000) 65–68.
[23] A. Cawley, J.P. Duxbury, T.P. Kee, Tetrahedron: Asymmetry 9 (1998) 1947–1949.