Binuclear versus mononuclear copper complexes as catalysts for asymmetric cyclopropanation of styrene

Article abstract of DOI:10.1016/S0957-4166(99)00009-9

A number of (S)-(-)-2-amino-1,1-diaryl-1-propanol compounds have been synthesized. They were used to form the binuclear copper complexes through a spontaneous assembly of the individual components, the β-aminoalcohol, 2- hydroxy-5-methyl-1,3-benzenedialdehyde, and copper acetate monohydrate, in methanol. These binuclear complexes were examined as asymmetric catalysts for cyclopropanation of styrene by ethyl diazoacetate. Moderate improvement in enantioselectivity has been observed for the binuclear versus mononuclear copper complexes. The e.e. values up to 87% for trans and 93% for cis products and the ratio between trans and cis products up to 9:1 have been obtained.

Full text of DOI:10.1016/S0957-4166(99)00009-9

Pergamon
Tetrahedron: Asymmetry 10 (1999) 411–427
Binuclear versus mononuclear copper complexes as catalysts for
asymmetric cyclopropanation of styrene
Lisheng Cai,∗ Hussein Mahmoud and Ying Han
Department of Chemistry, the University of Illinois at Chicago, Chicago, Illinois 60607, USA
Received 16 October 1998; accepted 22 December 1998
Abstract
A number of (S)-(−)-2-amino-1,1-diaryl-1-propanol compounds have been synthesized. They were used to
form the binuclear copper complexes through a spontaneous assembly of the individual components, the β-
aminoalcohol, 2-hydroxy-5-methyl-1,3-benzenedialdehyde, and copper acetate monohydrate, in methanol. These
binuclear complexes were examined as asymmetric catalysts for cyclopropanation of styrene by ethyl diazoacetate.
Moderate improvement in enantioselectivity has been observed for the binuclear versus mononuclear copper
complexes. The e.e. values up to 87% for trans and 93% for cis products and the ratio between trans and cis
products up to 9:1 have been obtained. © 1999 Elsevier Science Ltd. All rights reserved.
1. Introduction
Most asymmetric catalysts studied to date are mononuclear metal complexes and metal aggregates,
where synergistic effects among the transition metal centers rarely exist. Only recently has the synergistic
cooperation among different metals and chiral templates been incorporated into the design of asymmetric
catalysts.1 The chiral compartmental complexes of a group of binuclear nickel complexes used for
asymmetric O2 oxidation of olefins into epoxides dated back to 1992. Unfortunately, no enantioselectivity
was observed in that reaction.2 This may somehow hinder the rapid development of this class of
complexes as asymmetric catalysts. However, a growing number of binuclear complexes have been
proposed as intermediates in numerous reactions such as the Sharpless epoxidation.3 In our effort
to develop chiral binuclear compartment complexes as asymmetric catalysts, we will select several
mechanistically well understood reactions to demonstrate their advantages in the modulation of chiral
transfers.
Cyclopropanation was selected as our model asymmetric reaction in this study. The reaction bet-
ween diazo compounds and olefins is catalyzed by a variety of transition metal compounds.4 This
∗
Corresponding author. E-mail: cai@uic.edu
0957-4166/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00009-9

412
L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
transformation is stereospecific and generally occurs with ‘electron-rich’ olefins, including substituted
olefins, dienes, and vinyl ethers, but excluding α,β-unsaturated carbonyl compounds or nitriles.5 Relative
reactivities portray a highly electrophilic intermediate and an ‘early’ transition state for cyclopropanation
reactions.6 This may be reflected in the relative difficulty in controlling selectivity. For intermolecular
reactions, the formation of geometrical isomers of regioisomers from reactions with dienes, and of
enantiomers must all be taken into account. Currently, several available catalysts provide exceptional
enantiocontrol in both inter- and intramolecular cyclopropanation reactions for certain types of olefins.7
The most effective ligand systems are those introduced by Arantani,8 Pfaltz,9 Masamune,10 Evans,11
Doyle,12 and Nishiyama.13
These catalysts either induce excellent stereoselectivities in intermolecular cyclopropanation reactions
with olefins of low functionality such as styrene or 2,4-dimethyl-2,4-hexadiene, or give excellent results
in intramolecular reactions. Functionalized olefins such as silyl enol ethers do not give satisfactory
diastereomeric or enantiomeric selectivity.14 The ligands and catalytic systems we have been developing
differ from existing ones. First, two metal centers are involved to preorganize the carbene intermediate
with one metal bound with the carbene and the other with carbonyl oxygen. This property may allow
the modulation of both electronic and steric effects of the catalysts to fit a given class of substrates.
Second, the chiral ‘pocket’ of the catalysts is spacious enough to allow the access of bulkier substrates
other than monosubstituted or 1,1-disubstituted olefins, which gave much better enantioselectivity with
the previously reported catalytic systems described above.
2. Results and discussion
Based on our understanding of the asymmetric cyclopropanation reactions, we are trying to build
a module to optimize our binuclear metal complexes. We selected the Schiff base of salicylaldehyde
as our ligand backbone. A generic structure for the catalyst is shown by 20–24 (Scheme 6). We have
examined the dependence of the enantioselectivity on the steric bulk of the 2-alkoxide group of the
aromatic substitutent.
2.1. Synthesis of 2-alkoxy-5-t-butyl-phenyl bromide
Commercially available 4-tert-butyl-phenol 1 was brominated at the ortho position by bromine at 0°C
in CCl4 and CHCl3 to give 4-tert-butyl-2-bromo-phenol (2) as the only product in almost quantitative
yield (Scheme 1). Formation of phenyl ethers (3–7) was accomplished by alkylation of alkyl bromide
under basic conditions. The reaction worked well for a variety of alkyl groups including primary bromide,
secondary bromide, and triphenyl methyl chloride. However, neither tertiary bromide nor chloride
worked under these conditions or any other conditions have we tried.
Scheme 1.

L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
413
An alternative method for the synthesis of aryl t-alkyl ether 9 is the bromination of the ether 8 instead
of the alkylation of 2 (Scheme 2). Aryl ether 8 was synthesized following an analogous procedure, found
in the literature, in yields varying from 5 to 28%.15 A reduced yield (about 5%) was obtained if tert-butyl
bromide was used. From the rapid disappearance of the bromide, we believe that the elimination reaction
to generate 2-methyl propene predominates under these conditions. Bromination of 8 was anything but
straightforward. Under the conditions used for the bromination of aryl ethers, the only product observed
was 2. No trace of 9 was observed. We reasoned that the byproduct of the reaction, HBr, is responsible for
the cleavage of the tert-butyl group. However, bromination under basic conditions such as with pyridine
or 2,6-lutidine gave no product at all; only the starting material was recovered. It appears that successful
bromination of 8 needs a delicate balance between the acid and base. After multiple trials, we found two
procedures which gave satisfactory results (>80% isolated yields): (1) bromination with solid Na2CO3 at
−22°C in CCl4; (2) bromination with aqueous Na2CO3 solution in CHCl3 at room temperature.
Scheme 2.
2.2. Synthesis of β-aminoalcohols
The β-aminoalcohols 10–14 were synthesized by the reaction of the Grignard reagent with the ester of
alanine hydrochloride (Scheme 3).16 Normally 5–10 equivalents of the Grignard reagent were used. The
characteristic feature of the product in 1H NMR spectrum is the unique quartet given by the CH attached
to the amino group, with chemical shifts within the range of 4.0–4.5 ppm. All of the aminoalcohols we
synthesized appeared as the last band eluted with pure ethyl acetate on a silica gel column. Multiple
attempts to synthesize (S)-(−)-2-amino-1,1-di-(20 -diphenylmethyloxy-50 -tert-butyl-phenyl)-1-propanol
and (S)-(−)-2-amino-1,1-di-(20 -triphenylmethyloxy-50 -tert-butyl-phenyl)-1-propanol failed. This may be
due to the steric bulkiness of the ortho-substitutents.
Scheme 3.

414
L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
2.3. Synthesis of mono-copper complexes
The catalyst 15 was synthesized by means of an one-pot reaction of β-aminoalcohol, 10, salicylalde-
hyde, and copper acetate monohydrate in methanol (Scheme 4). However, no reactions were observed
when analogous conditions were used in attempts to make 16–19. In these cases the addition of solid
NaOH was essential. The catalysts were soluble in a wide range of solvents with different polarity from
pentane to methanol. The crystal structure of an analogous Cu(II) catalyst precursor [Ar=5-tert-butyl-2-
(n-butyloxy)phenyl] was found to be a binuclear copper complex bridged by the phenolate,8 although the
actual catalyst was believed to be a mononuclear copper(I) complex.7a
Scheme 4.
2.4. Optimization of the chiral mono-copper(II) complexes
Aratani and associates pioneered the design of chiral salicylaldimine copper catalysts which gave high
enantioselectivity for intermolecular cyclopropanations.17 The ‘Sumitomo process’, which is employed
for the construction of cilastatin, an in vivo stabilizer of the antibiotic imipenem, is based on this
chemistry. While the mono-copper Schiff base complex 15 was used as the catalyst for cyclopropanation
of styrene with ethyl diazoacetate (Scheme 5), the enantioselectivity significantly depended on the
polarity of the solvents as shown in Table 1. CH3CN is the best solvent for both reaction rate and
enantioselectivity, and is the solvent of choice for subsequent investigation when we vary the substitutents
on the aromatic group.
Scheme 5.
Although variation of the length of the alkoxide group at the 2-position of the aromatic group had
been evaluated by Aratani and associates on the optimization of the enantioselectivity for the synthesis
of cilastatin, this is the first study performed on branched alkoxide groups. We reasoned that these groups
may have better steric shielding effects in the chiral transfer step. The results of the study are summarized

L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
415
Table 1
Dependence of the e.e. values on solvents for catalyst precursor 15a
Table 2
Dependence of the e.e. values on the substituents (R) on mononuclear copper catalystsa
in Table 2. The best enantioselectivity for the trans isomer peaked at the isopropyloxy group, while
that for the cis isomer peaked at the linear chain alkoxide. Further increase of branches of the alkoxide
decreases the enantioselectivity slightly for the trans isomer, but substantially for the cis isomer.
2.5. Synthesis of binuclear copper complexes
The catalysts 20–24 were synthesized by means of a one-pot reaction of β-aminoalcohols 10–14,
2-hydroxy-5-methyl-1,3-benzenedialdehyde, and copper acetate monohydrate in the presence of trie-
thylamine, respectively (Scheme 6). Triethylamine is not necessary but was included to remove AcOH
generated during the reaction. This is in sharp contrast to the synthesis of mononuclear copper complexes,
where a strong base is necessary. Pure complexes 20–24 were obtained by washing with methanol. It is
noteworthy that methanol is one of the reactants which has been demonstrated by an X-ray diffraction
study as shown in Fig. 1. All of these complexes are green or dark gray in color, and are soluble in a
variety of solvents such as benzene, ethyl acetate, CH2Cl2, CHCl3, CH3CN, acetone, DMF, and DMSO.
The relative solubility was found to depend on the substituents.
The molecule of 20 in Fig. 1 has an overall C2 symmetry, and the environments of the two copper
centers are identical. Each copper center adopts a square pyramidal configuration. Among the five
coordination sites, three are from the pentadentate ligand, one is from bridging methoxide, and the other
is from a methanol molecule. The methyl group on the bridging methoxide is disordered and modeled
between two positions. The copper centers, the imine groups, the central phenolate, and the oxygen
atom of the bridging methoxide adopt a planar configuration. The most interesting part of the structure
is the five-membered rings involving chiral centers. Overall the rings adopt a half-chair conformation
with the tertiary carbon at the apex position, with one phenyl at the axial position and the other at the
equatorial position. The methyl group at the chiral carbon occupies an axial position, instead of the
equatorial position. Fig. 2 illustrates the conformational analysis along the C21–C22 single bond. Among

416
L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
Scheme 6.
Figure 1. Molecular structure of complex 20 with atomic numbering (ORTEP, 50% probability ellipsoids, hydrogen atoms
omitted for clarity). Selected bond distances (Å) and angles (°): Cu11–N11 1.924(4), Cu11–O11 1.970(3), Cu11–O21 1.916(3),
Cu11–O31 1.913(3), Cu11–O61 2.493(4), Cu11· · ·Cu11 3.0103(7), N11–Cu11–O11 90.7(2), O11–Cu11–O31 78.29(9),
O21–Cu11–O31 104.3(2), N11–Cu11–O21 86.15(14), O11–Cu11–061 94.5(1), N11–Cu11–O61 95.1(1), O21–Cu11–O61
93.1(1), O31–Cu11–O61 89.7(1), Cu11–O11–Cu11 99.6(2), Cu11–O31–Cu11 103.8(2)
the possible different steric interactions, it appears that the phenyl–methyl interaction predominates. This
makes the methyl at the equatorial position less stable. The crystal structure of an analogous compound
shows the same conformational preference.8
2.6. Optimization of the chiral binuclear copper(II) complexes
We examined the enantioselectivity of catalyst 20 in the asymmetric cyclopropanation reaction in
different solvents as compared with the corresponding mononuclear copper complex 15 (Scheme 5). The
results are summarized in Table 3. CH3CN is still the best solvent for this class of catalysts. Although the
best enantioselectivity did not increase from the mononuclear to binuclear copper catalyst, the binuclear
complex showed much less dependence on solvent, especially for non-polar solvents like toluene (8 vs.

L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
417
Figure 2. Conformational analysis of the complex 20
Table 3
Dependence of the e.e. values on solvents for catalyst precursor 20a
Table 4
Dependence of the e.e. values on the substituents (R) on binuclear copper catalystsa
54% for trans, 5 vs. 50% for cis). The trans to cis ratio uniformly increased for all solvents tested, as
compared with that of 15.
Variation of the alkyl group on the alkoxide substantially changes the e.e. values (Table 4). When the
alkyl group is benzyl, the e.e. values reach 87% and 93%, respectively, for the trans and cis isomers
of the product, ethyl 20-phenyl-1-cyclopropyl carboxylate. No mono-copper Schiff base complexes give
such high e.e. values.
The enantioselectivity does not increase linearly with the bulkiness of the substitutent on the 2-position.
The enantioselectivity reaches a plateau at groups larger than isopropyloxy for the trans isomer. The same

418
L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
Table 5
Solvent dependence of the enantioselectivity based on catalyst 22a
pattern is observed for the cis isomer, but it appears that the selectivity levels off with a much smaller
alkoxide group. This is different from the pattern observed for the mononuclear copper catalysts shown in
Table 2. Efforts to synthesize diphenylmethyl and triphenylmethyl substituted β-aminoalcohols failed as
described above. To our surprise, the effect of the benzyl group is comparable with that of the isopropyl
group.
2.7. Solvent dependence of the enantioselectivity
Binuclear copper complex 22 was used to evaluate the solvent dependence. Non-polar to highly polar
solvents were examined as shown in Table 5. The solvent which gives the highest e.e. values is still
CH3CN. It is interesting to note that the addition of 2,6-lutidine in this solvent dramatically reduces the
rate of the reaction. The reaction in THF or DMF is also slow.
2.8. Temperature dependence of the cyclopropanation
Using catalyst 22 as an example, the reaction yield based on ethyl diazoacetate was 33% at rt,
55% at 50°C, and leveled off (about 60%) at higher temperatures. Temperature has no effect on the
enantioselectivity (e.e. 88% for trans and 93% for cis) of the reaction, however, it does have a dramatic
effect on the yield.
2.9. Mechanistic implications
Synthesis of the binuclear copper complexes utilizing different temperatures, solvents, and reaction
times, does not affect the enantioselectivity significantly, probably the actual catalyst is the same at the
catalytic conditions. After the catalyst precursors reacted with PhNHNH2 at rt or 50°C for a few minutes,
the color of the solution changed from green or dark gray to brown, which remained during the reaction.
This supports the assumption that the copper(II) is reduced to copper(I) by PhNHNH2 in situ. However,
it is not clear at present whether both copper sites in the same molecule are reduced, or only one. The
persistent brown color is consistent with the charge transfer band observed for other mixed valence
binuclear copper complexes.18
2.10. Determination of the e.e. values of the products
The e.e. values for both trans and cis products were determined by GLC analysis of the corresponding
L-menthyl esters.9,10 Others used chiral HPLC to determine the e.e. of the trans product directly,19
but not for the cis product. Evan and associates transformed the cis ester into the amide of (S)-(−)-α-
methylbenzylamine, and determined the e.e. values of the amide.11 In our case, the harsh conditions
for this transformation prevented small scale operations. Instead, the cis-ester was reduced to the
corresponding alcohol by LAH at rt, and the peak separations on Chiralcel OD column increased
from 0.23 min to 0.46 min. The alcohol was then transformed into the (1S)-(−)-camphanic ester by

L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
419
reacting with (1S)-(−)-camphanic chloride, which is commercially available. Well-resolved peaks for
both enantiomers were observed under our conditions, as specified in the experimental section. This was
confirmed when the racemic mixture was used as an internal control.
3. Conclusion
As compared with those of the mononuclear counterparts, the binuclear copper(II) catalytic systems
have the following advantages: (1) they showed moderately increased enantioselectivity on the asymme-
tric cyclopropanation of styrene by ethyl diazoacetate; (2) their enantioselectivity is much less dependent
on solvents; and (3) they are easier to assemble from the individual components. The binuclear nature
of the complexes in the precursors is believed to be maintained during the catalytic conditions, since
the pentadentate ligand system provides strong stabilization for such an arrangement. These designed
binuclear complexes represent a new avenue for the assembly of catalysts to control the enantioselectivity
in asymmetric catalysis.
4. Experimental
4.1. 4-tert-Butyl-2-bromophenol (2)20
4-tert-Butylphenol (70 g, 0.467 mol) was dissolved in 220 mL of CHCl3:CCl4 (1:1), and the solution
was cooled to 0°C. Bromine (82 g, 0.513 mol) in 90 mL CHCl3 was added dropwise into the well-stirred
reaction mixture. The addition took about 1 h, and the mixture was then stirred for another hour. Dry
dinitrogen was bubbled through the mixture to remove the HBr. Upon removal of the solvent, the product
1
was distilled at reduced pressure (78–81°C/0.3 torr) to afford 109 g product. Yield 99.5%. H NMR in
CDCl3, δ, 7.43 (d, 4JHH=2.4 Hz, 1H, Ar–H); 7.24 (dd, 3JHH=9.0 Hz, 4JHH=2.5 Hz, 1H, Ar–H); 6.96 (d,
3JHH=8.7 Hz, 1H, Ar–H); 5.34 (s, 1H, OH); 1.29 (s, 9H, CH3). 13C{1H} NMR in CDCl3, δ, 149.8 (C–O,
1C), 145.0 (C–C, 1C), 128.7 (CH, 1C), 126.1 (CH, 1C), 115.5 (CH, 1C), 109.8 (C–Br, 1C), 34.1 (CCH3
1C), 31.3 (CH3, 3C). MS, 230, 228 (25%, M+), 215, 213 (100%, M−CH3 ). High resolution MS for
+
C10H13OBr, found: 228.014429; calcd: 228.014976.
4.2. 4-tert-Butyl-2-bromo-1-octyloxy-benzene (3)
4-tert-Butyl-2-bromophenol (43.5 g, 0.19 mol), octylbromide (40.3 g, 0.209 mol), and K2CO3 (26.2 g,
0.19 mol) were refluxed in dry acetone (100 mL) for 48 h or until the starting phenol almost disappeared.
The solid was filtered off, and extracted with acetone. Upon removal of the solvent, octyl bromide was
distilled off under vacuum. The residue was loaded onto a silica gel column, and the product was eluted
4
with hexane to afford 45 g. Yield 70%. 1H NMR in CDCl3, δ, 7.53 (d, JHH=1.3 Hz, 1H, Ar–H); 7.24
3
4
3
3
(dd, JHH=8.9 Hz, JHH=1.9 Hz, 1H, Ar–H); 6.81 (d, JHH=8.8 Hz, 1H, Ar–H), 3.99 (t, JHH=6.4 Hz,
2H, OCH2); 1.82 (pent, 3JHH=7.4 Hz, 2H, OCCH2); 1.49 (pent, 3JHH=7.7 Hz, 2H, OCCCH2); 1.34–1.20
(m, 8H, CH2); 1.28 (s, 9H, C(CH3)3); 0.886 (t, JHH=6.7 Hz, 3H, CH3). 13C{1H} NMR in CDCl3, δ,
3
153.1 (C–O, 1C), 144.7 (C–C, 1C), 130.3 (CH, 1C), 125.0 (CH, 1C), 112.7 (CH, 1C), 111.8 (C–Br, 1C),
69.2 (OCH2, 1C), 34.0 (C(CH3)3, 1C), 31.7 (CH2, 1C), 31.3 (C(CH3)3, 3C), 29.2 (CH2, 1C), 29.1 (CH2,
1C), 29.0 (CH2, 1C), 25.9 (CH2, 1C), 22.6 (CH2, 1C), 14.0 (CH3, 1C). MS, 342, 340 (5%, M+), 230,

420
L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
228 (12%, M−octene+), 215, 213 (100, M−octene−Me+). High resolution MS for C18H29OBr, found:
340.139887; calcd: 340.140177.
4.3. 4-tert-Butyl-2-bromo-1-isopropyloxy-benzene (4)
This compound was synthesized by the same procedure as above except using isopropyl bromide
1
4
instead of octyl bromide. Yield 95%. H NMR in CDCl3, δ, 7.53 (d, JHH=2.1 Hz, 1H, Ar–H); 7.23
3
4
3
3
(dd, JHH=8.7 Hz, JHH=2.7 Hz, 1H, Ar–H); 6.85 (d, JHH=8.7 Hz, 1H, Ar–H); 4.50 (hepta, JHH=6.0
Hz, 1H, CH(CH3)2); 1.37 (d, 3JHH=5.7 Hz, 6H, CCH(CH3)2); 1.28 (s, 9H, C(CH3)3). 13C{1H} NMR in
CDCl3, δ, 152.2 (C–O, 1C), 145.1 (C–C, 1C), 130.4 (CH, 1C), 125.0 (CH, 1C), 115.5 (CH, 1C), 113.4
(C–Br, 1C), 72.2 (CH, 1C), 34.1 (C(CH3)3, 1C), 31.3 (C(CH3)3, 1C), 22.1 (CH(CH3)2, 2C). MS, 272,
270 (16%, M+), 230, 228 (12%, M−propene+), 215, 213 (100, M−propene−Me+). High resolution MS
for C13H19OBr, found: 270.061441; calcd: 270.061926.
4.4. 4-tert-Butyl-2-bromo-1-benzyloxy-benzene (5)
This compound was synthesized by the same procedure as above except using benzyl bromide instead
of octyl bromide. Yield 96%. 1H NMR in CD3OD, δ, 7.54 (d, 4JHH=2.9 Hz, 1H, Ar–H); 7.48–7.25 (m,
3
6H, Ar–H); 6.98 (d, JHH=8.7 Hz, 1H, Ar–H); 5.12 (s, 2H, OCH2); 1.27 (s, 9H, CH3). 13C{1H} NMR
in CDCl3, δ, 152.7 (C–O, 1C), 145.3 (C–C, 1C), 136.7 (C–C, 1C), 130.5 (CH, 1C), 128.5 (CH, 2C,
Ph), 127.8 (CH, 1C), 126.9 (CH, 2C, Ph), 125.1 (CH, C), 113.4 (CH, 1C, Ph), 112.0 (C–Br, C), 70.8
+
(OCH2, 1C), 34.1 (CCH3, 1C), 31.3 (CH3, 3C). MS, 318, 320 (M+, <5%), 91 (100%, C6H5CH2 ). High
resolution MS for C17H19OBr, found: 318.062333; calcd: 318.061926.
4.5. 4-tert-Butyl-2-bromo-1-diphenylmethyloxy-benzene (6)
This compound was synthesized by the same procedure as above except using diphenylmethyl bromide
instead of octyl bromide. Yield 72%. 1H NMR in CDCl3, δ, 7.54 (d, 4JHH=2.0 Hz, 1H, Ar–H); 7.50 (d,
3
3
3JHH=7.8 Hz, 4H, o-H); 7.34 (t, JHH=7.6 Hz, 4H, m-H); 7.26 (t, JHH=7.4 Hz, 2H, p-H); 7.09 (dd,
3JHH=8.1 Hz, JHH=2.7 Hz, 1H, Ar–H); 6.75 (d, JHH=8.8 Hz, 1H, Ar–H); 6.23 (s, H, OCH); 1.24 (s,
9H, CH3). 13C{1H} NMR in CDCl3, δ, 151.8 (C–O, 1C), 145.1 (C–C, 1C), 141.1 (CH, 1C), 130.4 (CH,
1C), 128.5 (CH, 4C), 127.6 (CH, 4C), 126.6 (CH, 2C), 124.9 (C–Br, 2C), 114.6 (CH, 1C), 112.5 (C–C,
1C), 82.2 (OCH, 1C), 34.0 (CC, 1C), 31.2 (CH3, 3C). EI MS, 167 (100%, (C6H5)2CH+). FAB MS,
4
3
393.0160 (3%, M+), 315.1020 (4%, M−C6H6 ), 167.0841 (100%, Ph2CH+). High resolution FAB MS
+
for C23H22OBr, found: 393.085200; calcd: 393.085402. Elemental analysis for C23H23OBr, found: C,
69.55; H, 5.84; Br, 20.19; calcd: C, 69.88; H, 5.86; Br, 20.21.
4.6. 4-tert-Butyl-2-bromo-1-triphenylmethyloxy-benzene (7)
This compound was synthesized by the same procedure as above except using triphenylmethyl chloride
instead of octyl bromide. Yield 47%. 1H NMR in CDCl3, δ, 7.51 (d, 3JHH=6.6 Hz, 4H, o-H); 7.40 (s, 1H,
Ar–H); 7.26 (brs, 6H, m-H, p-H); 6.77 (d, 3JHH=8.5 Hz, 1H, Ar–H); 6.42 (d, 3JHH=8.7 Hz, 1H, Ar–H);
1.16 (s, 9H, CH3). 13C{1H} NMR in CDCl3, δ, 150.2 (C–O, 1C), 145.3 (C–C, 1C), 143.8 (CH, 1C),
129.7 (CH, 1C), 128.8 (CH, 6C), 127.6 (CH, 6C), 127.1 (CH, 3C), 123.6 (C–Br, 3C), 120.5 (CH, 1C),
115.2 (C–C, 1C), 91.1 (OCH, 1C), 33.9 (CC, 1C), 31.1 (CH3, 3C). MS, 243 (41%, Ph3C+), 228 (4%,

L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
421
(CH3)3CC6H3(OH)Br+), 215, 213 (8%, (CH3)2CC6H3(OH)Br+), 165 (20%, Ph3C+−C6H6). Elemental
analysis for C29H27OBr, found: C, 73.68; H, 5.80; Br, 17.14; calcd: C, 73.88; H, 5.77; Br, 16.95.
4.7. 4-tert-Butyl-1-tert-butyloxy-benzene (8)
4-tert-Butylphenol (15 g, 0.10 mol), Ni(acac)2 (300 mg, 1.2 mmol), and NaHCO3 (16.8 g, 0.2 mol)
were suspended in 50 mL of tert-butyl chloride. The mixture was refluxed overnight. Ether (200 mL)
was added, and the solution was filtered. Upon removal of the solvent, the crude solid was loaded onto
1
a silica gel column. The product was eluted by hexane to afford 4.6 g product. Yield 23%. H NMR in
CDCl3, δ, 7.24 (d, 3JHH=9.0 Hz, 2H, o-H); 6.89 (d, 3JHH=8.8 Hz, 2H, m-H); 1.32 (s, 9H, CH3); 1.29 (s,
9H, CH3). 13C{1H} NMR in CDCl3, δ, 152.8 (C–O, 1C), 145.8 (C–C, 1C), 125.5 (CH, 2C), 123.4 (CH,
+
2C), 77.9 (C–O, 1C), 34.1 (CC, 1C), 31.4 (CH3, 3C), 28.8 (CH3, 3C). MS, 206 (M+, 1%), 191 (M−CH3 ,
+
+
1%), 150 (M−CH2_C(CH3)2 ), 135 (M−CH2_C(CH3)2–CH3 , 100%). High resolution MS C14H22O,
found: 206.166370; calcd: 206.167066.
4.8. 4-tert-Butyl-2-bromo-1-tert-butyloxy-benzene (9)
Method 1. 4-tert-Butyl-1-tert-butyloxy-benzene (0.60 g, 2.9 mmol) and K2CO3 (5.0 g, 36 mmol) were
suspended in 200 mL of CCl4. The solution was then cooled to −23°C by dry ice/CCl4 cooling bath.
Bromine (0.69 g, 4.3 mmol) was added drop by drop within 24 h. After the reaction, the solid was
removed by filtration, and the solvent was removed to give a viscous liquid which was extracted by ether.
The ether solution was dried by anhydrous Na2SO4. Removal of the solvent gave the crude product,
which was purified on a silica column using hexane as eluent to afford 0.73 g of pure product. Yield
88%.
Method 2. 4-tert-Butyl-1-tert-butyloxy-benzene (1.0 g, 4.9 mmol) in 130 mL of CHCl3 and NaHCO3
(7.3 g, 87 mmol) in 18 mL of water were mixed at rt. Bromine (2.0 g, 12.5 mmol) in 25 mL of CHCl3
was added dropwise overnight. After the reaction was completed as checked by TLC, the organic layer
was separated and dried by anhydrous Na2SO4. The same procedure as above in method 1 was followed
to obtain 1.2 g of the product. Yield 86%. 1H NMR in CDCl3, δ, 7.52 (d, 4JHH=2.0 Hz, 1H, Ar–H); 7.20
(dd, 3JHH=8.8 Hz, 4JHH=2.0 Hz, 1H, Ar–H); 7.02 (d, 3JHH=8.8 Hz, 1H, Ar–H), 1.42 (s, 9H, CH3); 1.29 (s,
9H, CH3). 13C{1H} NMR in CDCl3, δ, 150.6 (C–O, 1C), 147.3 (C–C, 1C), 130.1 (CH, 1C), 124.7 (CH,
1C), 123.2 (CH, 1C), 118.6 (C–Br, 1C), 80.8 (C–O, 1C), 34.2 (CC, 1C), 31.2 (CH3, 3C), 28.9 (CH3, 3C).
+
+
+
MS, 269, 271 (M−CH3 , 3%), 230 (M−CH2_C(CH3)2 , 25%), 213 (M−CH2_C(CH3)2–CH3 , 100%),
+
135 (M−CH2_C(CH3)2–CH3 , 100%). Elemental analysis for C14H21OBr, found: C, 58.92; H, 7.50; Br,
28.22; calcd: C, 58.95; H, 7.42; Br, 28.02.
4.9. (S)-(−)-2-Amino-1,1-diphenyl-1-propanol (10)
PhMgBr (350 mL of 1 M, 350 mmol) was added to a suspension of L-alanine methyl ester hydrochlo-
ride (6.0 g, 43 mmol) in THF at 0°C over a period of 1 h. The mixture was stirred overnight. Saturated
aqueous NH4Cl solution (300 mL) was then added, and the mixture was shaken vigorously until all white
precipitate disappeared. The product was extracted into ether solution (3×200 mL). The combined ether
solution was dried over Na2SO4. Upon removal of the solvent, the crude solid was loaded on a silica gel
column, which was eluted with hexane first, then with hexane/ethyl acetate. The portion of ethyl acetate
was increased gradually from 0.1 to 1. The product was then eluted with ethyl acetate. Removal of the
1
3
solvent affords 4.23 g product. Yield 43%. H NMR in CD3OD, δ, 7.59 (d, JHH=8.6 Hz, 2H, Ar–H);

422
L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
3
3
7.50 (d, JHH=8.7 Hz, 2H, Ar–H); 7.33 (t, JHH=7.7 Hz, 2H, Ar–H); 7.28 (t, 3JHH=7.7 Hz, 2H, Ar–H);
7.21 (t, 3JHH=7.1 Hz, 1H, Ar–H); 7.16 (t, 3JHH=7.5 Hz, 1H, Ar–H); 3.98 (q, 3JHH=6.5 Hz, 1H, CH–N);
1.00 (d, 3JHH=6.5 Hz, 3H, CH3). 13C{1H} NMR in CD3OD, δ, 147.6 (C–C, 1C), 147.1 (C–C, 1C), 129.5
(CH, 2C, o-C), 129.1 (CH, 2C, o-C), 127.9 (CH, 1C, p-C), 127.6 (CH, 1C, p-C), 127.3 (CH, 2C, m-C),
127.0 (CH, 2C, m-C), 81.4 (C–O, 1C), 53.6 (C–N, 1C), 17.2 (CH3–C–N). MS, 210 (1%, M−OH+), 105
+
(30%, PhCO+), 77 (48%, Ph+), 44 (100, CH3CH−NH2 ). High resolution MS for C15H18NO, found:
228.138791; calcd: 228.138839.
4.10. (S)-(−)-2-Amino-1,1-di-(20-octyloxy-50-tert-butyl-phenyl)-1-propanol (11)
4-tert-Butyl-2-bromo-1-octyloxy-benzene (33 g, 96.7 mmol) and Mg (2.35 g, 97.9 mmol) were
suspended in 100 mL of THF with slow stirring. I2 (200 mg, 0.79 mmol) was added. A deep brown
solution was formed. The color gradually faded. After the solution had been stirred overnight, or until
no starting material was observed on a TLC plate, the residue Mg was removed by decantation. The
reaction of this new Grignard reagent with L-alanine methyl ester hydrochloride was performed and
1
purification as above afforded 5.5 g of product. Yield 93% based on the ester of the amino acid. H
3
4
NMR in CDCl3, δ, 7.70 (s, 2H, Ar–H); 7.18 (dd, JHH=8.9 Hz, JHH=1.6 Hz, 1H, Ar–H); 7.15 (dd,
3JHH=8.2 Hz, JHH=1.5 Hz, 1H, Ar–H); 6.72 (d, JHH=8.5 Hz, 1H, Ar–H); 6.66 (d, JHH=8.6 Hz, 1H,
Ar–H); 4.28 (q, JHH=6.5 Hz, 1H, CH–N); 3.76–3.67 (m, 4H, OCH2); 1.52 (m, 4H, OCCH2); 1.34
4
3
3
3
3
(s, 9H, CH3); 1.33 (s, 9H, CH3); 1.29–1.20 (m, 10H, CH2); 1.04 (d, JHH=5.0 Hz, 3H, CH3–C–N);
0.886 (t, JHH=6.8 Hz, 6H, CH3). 13C{1H} NMR in CDCl3, δ, 154.0 (C–O, 1C), 153.3 (C–O, 1C),
3
142.1 (C–C, 1C), 141.8 (C–C, 1C), 132.7 (C–C, 1C), 132.0 (C–C, 1C), 125.5 (CH, 1C), 125.1 (CH,
1C), 124.4 (CH, 1C), 124.0 (CH, 1C), 112.1 (CH, 1C), 111.5 (CH, 1C), 80.2 (C–O, 1C), 68.3 (CH2O,
1C), 68.1 (CH2O, 1C), 49.6 (CH–N, 1C), 34.1 (C(CH3)3, 1C), 34.1 (C(CH3)3, 1C), 31.7 (C(CH3)3,
3C), 31.6 (C(CH3)3, 3C), 31.5 (CH2, 1C), 31.4 (CH2, 1C), 29.2 (CH2, 2C), 29.1 (CH2, 2C), 28.9 (CH2,
2C), 25.8 (CH2, 2C), 22.5 (CH2, 2C), 17.8 (CH3–C–N, 1C), 14.0 (CH3–C, 2C). EI MS, 291 (9%), 179
(50%, (CH3)3CC6H3(OH)C(OH)+), 151 (100%, (CH3)3CC6H4(OH)+), 135 (32%), 57 (30%, C4H9 ).
+
FAB MS, 596.4064 (14%, M+), 578.3967 (100%, M−H2O+), 177.0929 (12%, (CH3)3CC6H3(OH)–CO+).
High resolution FAB MS for C39H66NO3, found: 596.504400; calcd: 596.504271. Elemental analysis for
C39H65NO3, found: C, 81.23; H, 11.39; N, 2.98; calcd: C, 78.60; H, 10.99; N, 2.35.
4.11. (S)-(−)-2-Amino-1,1-di-(20-isopropyloxy-50 -tert-butyl-phenyl)-1-propanol (12)
This compound was synthesized by the same procedure as above except using 4-tert-butyl-2-bromo-
1-isopropyloxy-benzene instead of 4-tert-butyl-2-bromo-1-octyloxy-benzene to afford 3.1 g of product.
1
4
Yield 59.3% based on the ester of the amino acid. H NMR in CD3OD, δ, 7.71 (d, JHH=2.3 Hz, 1H,
4
3
4
Ar–H); 7.69 (d, JHH=2.4 Hz, 1H, Ar–H); 7.28 (dd, J=8.7 Hz, JHH=2.7 Hz, 1H, Ar–H); 7.22 (dd,
3J=8.5 Hz, JHH=2.5 Hz, 1H, Ar–H); 6.87 (d, JHH=8.7 Hz, 1H, Ar–H); 6.78 (d, JHH=8.6 Hz, 1H,
4
3
3
Ar–H); 4.55 (hepta, 3JHH=6.1 Hz, 1H, OCH(CH3)2); 4.51 (hepta, JHH=6.0 Hz, 2H, OCH(CH3)2), 4.24
3
(q, 3JHH=6.6 Hz, 1H, CH–N); 1.38 (s, 9H, C(CH3)3); 1.36 (s, 9H, C(CH3)3); 1.09 (d, 3JHH=5.9 Hz, 3H,
3
3
3
CH3); 1.03 (d, JHH=6.0 Hz, 3H, CH3); 1.02 (d, JHH=6.5 Hz, 3H, CH3); 0.929 (d, JHH=5.9 Hz, 3H,
CH3); 0.864 (d, 3JHH=5.9 Hz, 3H, CH3). 1H NMR in CDCl3, δ, 7.71 (s, 1H, Ar–H); 7.69 (s, 1H, Ar–H);
3
4
3
4
7.18 (dd, JHH=8.7 Hz, JHH=2.7 Hz, 1H, Ar–H); 7.14 (dd, JHH=8.2 Hz, JHH=2.0 Hz, 1H, Ar–H);
3
3
3
6.71 (d, JHH=8.7 Hz, 1H, Ar–H); 6.65 (d, JHH=8.6 Hz, 1H, Ar–H); 4.45 (penta, JHH=6.0 Hz, 1H,
OCH); 4.43 (penta, 3JHH=6.0 Hz, 1H, OCH); 4.24 (q, 3JHH=6.6 Hz, 1H, CH–N); 1.35 (s, 9H, C(CH3)3);
1.34 (s, 9H, C(CH3)3); 1.05 (brs, 3H, CH3); 1.04 (brs, 3H, CH3); 1.01 (brs, 3H, CH3); 0.946 (brs, 3H,

L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
423
CH3); 0.888 (brs, 3H, CH3). 13C{1H} NMR in CDCl3, δ, 152.4 (C–O, 1C), 151.7 (C–O, 1C), 141.5
(C–C, 1C), 141.1 (C–C, 1C), 133.3 (C–C, 1C), 132.5 (C–C, 1C), 125.7 (CH, 1C), 125.3 (CH, 1C), 124.6
(CH, 1C), 124.1 (CH, 1C), 112.4 (CH, 1C), 111.6 (CH, 1C), 80.0 (C–O, 1C), 68.8 (CH–O, 1C), 68.4
(CH–O, 1C), 49.3 (CH–N, 1C), 34.1 (C(CH3)3, 1C), 34.1 (C(CH3)3, 1C), 31.6 (C(CH3)3, 3C), 31.4
(C(CH3)3, 3C), 21.7 (CH3, 1C), 21.5 (CH3, 1C), 21.4 (CH3, 1C), 21.1 (CH3, 1C), 18.0 (CH3–C–N, 1C).
MS, 411 (38%, M−CH3CH(NH2)+), 327 (57%, M−CH3CH(NH2)−2×propene+), 271 (10%), 219 (67%,
(CH3)3CC6H3(OiPr)–CO+), 177 (100%, (CH3)3CC6H3(OH)–CO+), 161 (75%, (CH3)3CC6H4–CO+), 44
+
(85%, CH3CH–NH2 ). Elemental analysis for C29H45NO3, found: C, 77.92; H, 10.30; N, 3.08; calcd: C,
76.44; H, 9.95; N, 3.07.
4.12. (S)-(−)-2-Amino-1,1-di-(20-tert-butyloxy-50 -tert-butyl-phenyl)-1-propanol (13)
This compound was synthesized by the same procedure as above except using 4-tert-butyl-2-bromo-1-
tert-butyloxy-benzene instead of 4-tert-butyl-2-bromo-1-octyloxy-benzene to afford 1.1 g of product.
1
Yield 51% based on the ester of the amino acid. H NMR in CDCl3, δ, 7.67 (s, 2H, Ar–H); 7.11
3
3
(dd, JHH=8.8 Hz, 4JHH=2.5 Hz, 1H, Ar–H); 7.08 (dd, JHH=8.4 Hz, 4JHH=2.0 Hz, 1H, Ar–H); 6.91 (d,
3JHH=8.6 Hz, 1H, Ar–H); 6.86 (d, 3JHH=8.7 Hz, 1H, Ar–H); 4.05 (q, 3JHH=6.5 Hz, 1H, CH–N); 1.31 (s,
9H, C(CH3)3); 1.29 (s, 9H, C(CH3)3); 1.24 (s, 9H, C(CH3)3); 1.22 (s, 9H, C(CH3)3); 1.06 (d, 3JHH=6.1
Hz, 3H, CH3–C–N). 13C{1H} NMR in CDCl3, δ, 151.7 (C–O, 1C), 151.2 (C–O, 1C), 141.7 (C–C, 1C),
141.3 (C–C, 1C), 134.6 (C–C, 1C), 134.4 (C–C, 1C), 125.8 (CH, 2C), 123.7 (CH, 2C), 123.4 (CH,
2C), 116.5 (CH, 1C), 115.9 (CH, 1C), 80.2 (C–O, 1C), 78.4 (C(CH3)3–O, 1C), 78.0 (C(CH3)3–O, 1C),
51.4 (CH–N, 1C), 34.1 (C(CH3)3, 1C), 34.0 (C(CH3)3, 1C), 31.5 (C(CH3)3, 6C), 29.2 (C(CH3)3, 6C),
+
18.0 (CH3–C–N, 1C). EI MS, 439 (M−CH3CH(NH2)+, 5%), 383 (M−CH3CH(NH2)–CH2_C(CH3)2 ,
+
5%), 327 (M−CH3CH(NH2)–2CH2_C(CH3)2 , 100%). FAB MS, 484.2952 (98%, M+), 466.2854 (67%,
M−H2O+), 354.1643 (58%), 327.1135 (M−CH3CH(NH2)–2CH2_C(CH3)2 , 100%), 321.1085 (11%),
+
281.0839 (11%), 177.0851 (30%, (CH3)3CC6H3(OH)–CO+), 161.0540 (13%). High resolution FAB MS
for C31H50NO3, found: 484.379000; calcd: 484.379070. Elemental analysis for C31H49NO3, found: C,
76.79; H, 10.28; N, 2.92; calcd: C, 76.97; H, 10.21; N, 2.90.
4.13. (S)-(−)-2-Amino-1,1-di-(20-benzyloxy-50-tert-butyl-phenyl)-1-propanol (14)
This compound was synthesized by the same procedure as above except using 4-tert-butyl-2-bromo-
1-benzyloxy-benzene instead of 4-tert-butyl-2-bromo-1-octyloxy-benzene to afford 0.27 g of product.
Yield 36% based on the ester of the amino acid. 1H NMR in CDCl3, δ, 7.68 (d, 3JHH=2.2 Hz, 1H, Ar–H);
7.68 (d, 3JHH=2.4 Hz, 1H, Ar–H); 7.21–7.06 (m, 8H, m and p in Ph, Ar–H); 6.82 (d, 3JHH=7.7 Hz, 2H,
o-H in Ph); 6.79 (d, 3JHH=7.3 Hz, 2H, o-H in Ph); 6.67 (d, 3JHH=8.6 Hz, 1H, Ar–H); 6.59 (d, 3JHH=8.6
3
Hz, 1H, Ar–H); 4.93–4.90 (m, 4H, OCH2); 4.48 (q, JHH=6.0 Hz, 1H, CH–N); 1.14 (s, 9H, CCH3),
1.12 (d, 3JHH=6.0 Hz, 3H, CH3–C–N). 13C{1H} NMR in CDCl3, δ, 153.6 (C–O, 1C), 152.9 (C–O, 1C),
142.9 (C–C, 1C), 142.6 (C–C, 1C), 137.2 (C–C, 1C), 136.9 (C–C, 1C), 132.6 (C–C, 1C), 131.9 (C–C,
1C), 128.7 (CH, 2C), 128.7 (CH, 2C), 127.8 (CH, 1C), 127.6 (CH, 1C), 126.9 (CH, 2C), 126.8 (CH,
2C), 126.2 (CH, 1C), 125.8 (CH, 1C), 124.6 (CH, 1C), 124.2 (CH, 1C), 112.7 (CH, 1C), 112.1 (CH,
1C), 80.6 (C–O, 1C), 69.9 (CH2–O, 1C), 69.6 (CH2–O, 1C), 49.3 (CH–N, 1C), 34.3 (C(CH3)3, 2C),
31.6 (C(CH3)3, 6C), 18.1 (CH3–C–N, 1C). EI MS, 267 (8%, (CH3)3CC6H3(OCH2Ph)CO+), 243 (20%),
177 (24%, (CH3)3CC6H3(OH)–CO+), 165 (15%), 91 (100, CH2Ph), 59 (20%), 43 (50%, CH3CH_NH+).
FAB MS, 552.2582 (82%, M+), 534.2596 (100, M−H2O+), 177.0941 (30%, (CH3)3CC6H3(OH)–CO+).

424
L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
High resolution FAB MS for C37H46NO3, found: 552.348000; calcd: 552.347770. Elemental analysis for
C37H45NO3, found: C, 79.51; H, 8.29; N, 2.60; calcd: C, 80.54; H, 8.22; N, 2.54.
4.14. Synthesis of mononuclear copper complex, 15 (Ar=Ph)
10 (26.8 mg, 0.118 mmol), Cu(OAc)2·H2O (23 mg, 0.115 mmol), and salicylaldehyde (14 mg, 0.115
mmol) were suspended in 5 mL of methanol. The mixture was stirred at rt overnight, and the color of
the solution changed to deep green. The solvent was removed, and the solid was washed with pentane
and methanol. Removal of the solvent afforded 39 mg of product 15. Yield 85%. The catalyst was used
without further purification.
15, MS, ESI, 809.0 (40%), 787.1 (100%, 2M+), 515.7 (22%), 412.1 (25%). IR (cm−1), 3057, 2935 (w,
CH), 1633 (s, C6H5), 1602 (w, C6H5), 1537 (w), 1447 (m), 1405 (w), 1316 (w), 1195 (w), 1133 (w), 998
(w), 912 (w), 759 (m), 702 (m), 674 (w), 570 (w), 539 (w), 445 (w). UV–vis (nm), 372 (ε=3198), 365
(ε=3320), 358 (ε=2617), 271 (ε=7121), 246 (ε=10 652), 222 (ε=18 109), 200 (ε=30 993).
4.15. Synthesis of mononuclear copper complexes, 16–19
11 (44.6 mg, 0.078 mmol), Cu(OAc)2·H2O (14.7 mg, 0.074 mmol), and salicylaldehyde (9.1 mg,
0.075 mmol) were suspended in 5 mL of methanol. After about 5 min, solid NaOH (120 mg, 3.0 mmol)
was added, and the mixture was stirred at rt for 18 h, and the color of the solution changed to deep green.
The solvent was removed, and the solid was extracted with pentane. Removal of the solvent afforded
36.6 mg of product 16. Yield 65%. The catalyst was used without further purification. 17, yield 100%;
18, yield 95%; 19, yield 95%.
16, MS, ESI, 1523.6 (100%, 2M+), 1479.1 (76%, 2M−C3H7 ), 1435.1 (34%, 2M−2×C3H7 ), 761.4
+
+
(77%, M+), 717.2 (64%, M−C3H7 ). IR (cm−1), 2953, 2925, 2856 (s, CH), 1630, 1604 (m, C6H3), 1500
+
(m), 1456 (w), 1398 (w), 1264 (m), 1026 (w), 900 (w), 809 (m), 758 (w), 668 (w). UV–vis (nm), 374
(ε=1320), 276 (ε=4640), 246 (ε=4652), 218 (sh, ε=13 180), 208 (ε=19 267).
17, MS, ESI, 1288.7 (45%), 1243.3 (100%, 2M+), 1180.4 (15%, 2M−Cu+), 661.7 (17%, M+CH3CN+),
652.5 (26%, M+CH3OH), 621.3 (42%, M+), 603.4 (30%). IR (cm−1), 2964, 2894, 2852 (s, CH), 1627,
1603 (s, C6H3), 1537 (w), 1498 (m), 1448 (w), 1395 (w), 1321 (w), 1252 (w), 1197 (w), 1147 (w),
1111 (s), 952 (w), 812 (w), 758 (w). UV–vis (nm), 374 (ε=4458), 274 (ε=13 045), 246 (ε=14 787), 226
(ε=30 845), 208 (ε=40 062).
18, MS, ESI, 1344.9 (100%), 1299.3 (96%, 2M+), 1237.4 (33%), 1172.3 (25%), 1132.6 (51%), 859.1
(24%), 680.6 (62%), 649.2 (44%, M+), 588.2 (22%), 524.3 (27%), 519.3 (23%), 466.4 (16%). IR (cm−1),
2965, 2933, 2909, 2872 (m, CH), 1627, 1604 (s, C6H3), 1515 (w), 1494 (s), 1447 (w), 1394 (m), 1366
(w), 1266 (w), 1174 (s), 906 (w), 817 (w), 757 (w), 671 (w). UV–vis (nm), 374 (ε=5552), 274 (ε=16 797),
246 (ε=18 426), 228 (ε=42 539), 208 (ε=60 884).
19, MS, ESI, 1435.2 (100%, 2M+), 782.3 (5%), 717.3 (9%, M+). IR (cm−1), 2951, 2902, 2870 (m,
CH), 1629, 1603 (s, C6H3), 1504 (m), 1444 (w), 1394 (w), 1318 (w), 1267 (w), 1220 (w), 1149 (w),
1020 (w), 900 (w), 809 (w), 757 (w), 696 (w), 670 (w). UV–vis (nm), 374 (ε=2976), 274 (ε=8745), 244
(ε=11 211), 226 (ε=21 586), 212 (ε=26 967).
4.16. Synthesis of binuclear copper complexes, 20–24
The synthesis of catalyst 22 is used as an example. 12 (17 mg, 0.037 mmol), Cu(OAc)2·H2O (7.3 mg,
0.037 mmol), and 2-hydroxy-5-methyl-1,3-benzenedialdehyde (3.0 mg, 0.018 mmol) were suspended in

L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
425
5 mL of methanol. After about 5 min, triethylamine (5 equiv.) was added, and the mixture was stirred at
rt for 2 h. The color of the solution changed to deep green. The solvent was removed, and the solid was
washed with methanol. Drying the solid afforded 24 mg of product 22. Yield 100%. The catalyst was
used without further purification. Using this procedure, the yield for 20 was 46%; for 21, 63%; for 23,
97%; for 24, 100%.
20, MS, ESI, 738.9 (100%, M+), 555.0 (33%), 412.1 (51%), 370.1 (68%, M2+), 238.9 (26%), 229.9
(20%). IR (cm−1), 3409 (brs, OH), 1677 (w), 1626 (m, C6H5), 1535 (w), 1433 (w), 1400 (w), 1307
(w), 1229 (w), 973 (w), 887 (w), 836 (w), 766 (w), 686 (w), 652 (w), 533 (w). UV–vis (nm), 413
(ε=3582), 399 (ε=3503), 388 (ε=3574), 383 (ε=3573), 366 (ε=3665), 250 (ε=16 607), 237 (16 620), 198
(ε=19 621), 190 (ε=21 746).
The X-ray quality crystals were grown from methanol solution at rt. C46H62Cu2N2O10; space group
C2, a=18.0489(4) Å, b=13.3410(3) Å, c=10.55680(10) Å, β=117.2490(10)°, V=2259.88(7) Å3, crystal
dimensions 0.3×0.2×0.2 mm, Z=4, dcalcd=1.367 g/cm3, F(000)=980 e, Siemens SMART/CCD diffrac-
tometer, Mo-Kα=radiation (λ=0.71073 Å), T=−60°C. Data were corrected for Lorentz and polarization
effects as well as absorption [empirical, absorption coefficient=0.999 mm−1]. 7528 reflections measured,
4973 [R(int)=0.0245] unique reflections. Non-H atoms were refined with anisotropic displacement para-
meters. Hydrogen atoms were calculated in idealized geometry and included with isotropic contributions.
P
Refined parameters (285) wR2 [unique data]=0.0941, R1[Fo ≥ 2σ(Fo)] = 0.0417, wR2 = [ w(Fo2 −
P
P
P
2
2
Fc2)2]/ [w(Fo2)2]1/2, R1 = (||Fo| − |Fc||)/ |Fo|, w = q/2σ (Fo2) + (ap) + bp, p = (Fo2 +
2Fc2)/3; a = 0.0546, b = 0.0000, ρ
= +0.367/ − 0.385 e A−3. The structure was solved by direct
˚
final
methods and refined by full matrix least–square calculations (SHELX-93, Sheldrick, G. M. SHELXL-93,
Göttingen, 1993).
21, MS, ESI, 1475.4 (100%, M+), 834.8 (29%), 803.4 (21%), 738.4 (51%, M2+), 622.5 (30%). IR
(cm−1), 2967, 2912 (S, CH), 1614 (S, C6H3), 1517 (m), 1404 (m), 1367 (m), 1265 (w), 1172 (m), 1107
(w), 898 (w), 818 (w), 679 (w). UV–vis (nm), 410 (ε=5662), 272 (ε=17 236), 252 (ε=17 774), 216 (sh,
ε=53 593), 210 (ε=61 233).
22, MS, ESI, 1917 (22%), 1521.2 (33%), 1283.5 (62%), 1269.2 (71%), 1255.0 (47%), 1241.2 (40%),
1223.0 (79%), 1209.3 (92%), 1194.6 (100%, M+), 813.3 (47%), 799.2 (52%), 771.3 (25%), 598.4 (32%,
M2+). IR (cm−1), 2968, 2908, 2863 (m, CH), 1635, 1604 (m, C6H3), 1496 (m), 1451 (m), 1395 (m), 1332
(w), 1299 (w), 1266 (m), 1243 (m), 1141 (w), 1112 (m), 1054 (w), 950 (w), 837 (w), 812 (w), 687 (w),
618 (w). UV–vis (nm), 412 (ε=8314), 386 (ε=10 707), 278 (ε=37 069), 250 (ε=39 950), 222 (ε=90 780),
208 (ε=152 173).
23, MS, ESI, 1429.4 (25%), 1338.9 (39%), 1324 (26%), 1309.7 (27%), 1296.9 (9%), 1277.3 (43%),
1262.5 (20%), 1250.5 (100%, M+), 1216.3 (24%), 1174.1 (10%), 838.9 (18%), 797.1 (20%), 715.6
(25%), 626.2 (36%, M2+), 561.0 (18%), 524.1 (13%), 480 (9%). IR (cm−1), 2954, 2857 (m, CH), 1614
(m, C6H3), 1493 (m), 1403 (m), 1367 (w), 1265 (w), 1172 (m), 1113(w), 898 (w), 818 (w), 677 (w), 621
(w). UV–vis (nm), 414 (ε=12 051), 390 (ε=15 408), 276 (ε=56 571), 254 (ε=54 750), 224 (ε=133 270),
212 (ε=179 533).
24, MS, ESI, 1713.2 (19%), 1611.3 (10%), 1445.1 (16%), 1387.3 (100%, M+), 1330.5 (10%), 1293.5
(5%), 1231.7 (12%), 885.2 (15%), 847.3 (7%), 759.2 (13%), 694.3 (15%, M2+). IR (cm−1), 2954, 2862
(m, CH), 1639, 1582 (m, C6H3), 1552 (w), 1456 (m), 1371 (w), 1319 (w), 1266 (w), 1227 (m), 1114
(w), 1020 (m), 808 (w), 734 (w), 697 (w), 613 (w). UV–vis (nm), 528 (ε=272), 412 (ε=8 849), 386
(ε=14 376), 366 (ε=13 550), 266 (ε=49 270), 216 (ε=115 868), 192 (ε=12 893).

426
L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
4.17. Asymmetric cyclopropanation
In a glovebox, the catalyst 22 (25 mg, 0.020 mmol) was dissolved in 3.0 mL of CH3CN. PhNHNH2
(2.4 equiv.) in CH3CN was added in one portion. Ethyl diazoacetate (236 mg, 2.1 mmol) in 3.0 mL of
CH3CN was placed in an addition funnel. The green color of the catalyst remained after the mixture was
taken out of the glovebox. A few drops of N2CHCOOC2H5 solution were added, and the reaction mixture
was heated to 50°C until the color of the solution changed to brown (about 2 minutes). Styrene (1 mL,
8.74 mmol) was added in one portion. After 5 minutes, the solution of N2CHCOOC2H5 was added and
the reaction stirred overnight. After the reaction, the solvent was removed and the product was purified
using the TLC plate.
4.18. Determination of the e.e. values for the trans product
The e.e. value of the trans isomer was determined directly by a Chiralcel OD column with the eluent
hexane:isopropanol=95:5. The retention times were 10.4 and 13.7 minutes, respectively.
4.19. Determination of the e.e. values for the cis product
4.19.1. Reduction of the cis ester to the cis alcohol
In a typical example, the cis ester (58 mg, 0.31 mmol) was dissolved in 5.0 mL of THF, and LiAlH4
(11.6 mg, 0.31 mmol) was added in one portion. After 0.5 h, 1 mL of water was added to quench
the excess LiAlH4. The product was extracted by ether, and the ether solution was dried by Na2SO4.
Purification of the product by silica gel column or TLC plate (ethyl acetate:hexane=30:70) afforded 36
mg of product. Yield 78%.
4.19.2. Synthesis of (1S)-(−)-camphanic ester
The cis alcohol (20 mg, 0.14 mmol) and (1S)-(−)-camphanic chloride (45 mg, 0.21 mmol) were
dissolved in 4 mL of CH2Cl2:(C2H5)3N (1:1). DMAP (5 mg, 0.041 mmol) was added in one portion.
After 3 h, or when no starting alcohol was detectable on TLC, the solvent was removed, and the product
was purified by TLC plate to afford 40 mg of the ester. Yield 87%. The e.e. values were determined as
above, with the retention times, 24.2 and 25.4 minutes, respectively.
References
1. (a) Shibasaki, M.; Sasai, H.; Arai, T. Angew. Chem., Int. Ed. Engl. 1997, 36, 1236–1256. (b) Kless, A.; Lefeber, C.;
Spannenberg, A.; Kempe, R.; Baumann, W.; Holz, J.; Börner, A. Tetrahedron 1996, 52, 14599–14606. (c) Sawamura, M.;
Sudoh, M.; Ito, Y. J. Am. Chem. Soc. 1996, 118, 3309–3310. (d) Süss-Fink, G. Angew. Chem., Int. Ed. Engl. 1994, 33, 1,
67–69. (e) Konsler, R. G.; Karl, J.; Jacobsen, E. N. J. Am. Chem. Soc. 1998, 120 (41), 10780–10781.
2. Oda, T.; Irie, R.; Katsuki, T.; Okawa, H. Synlett 1992, 641–643.
3. (a) Williams, I. D.; Pedersen, S. F.; Sharpless, K. B.; Lippard, S. J. J. Am. Chem. Soc. 1984, 106, 6430–6431. (b) Noyori,
R. Asymmetric Catalysis in Organic Synthesis; John Wiley & Sons: New York, 1994; p. 264. (c) Girard, C.; Kagan, H. B.
Angew. Chem., Int. Ed. Engl. 1998, 37, 2922–2959.
4. (a) Tamblyn, W. H.; Hoffman, S. R.; Doyle, M. P. J. Organomet. Chem. 1981, 216, C64–C68. (b) Hanks, T. W.; Jennings,
P. W. J. Am. Chem. Soc. 1987, 109, 5023–5025. (c) Tomilov, Y. V.; Kostitsyn, A. B.; Shulishov, E. V.; Nefedov, O. M.
Synthesis 1990, 246–248.
5. (a) Adams, J.; Spero, D. M. Tetrahedron 1991, 47, 1765–1808. (b) Maas, G. Top. Curr. Chem. 1987, 137, 75–253. (c)
Doyle, M. P. Chem. Rev. 1986, 86, 919–939. (d) Doyle, M. P. Acc. Chem. Res. 1986, 19, 348–356.

L. Cai et al. / Tetrahedron: Asymmetry 10 (1999) 411–427
427
6. (a) Anciaux, A. J.; Hubert, A. J.; Noels, A. F.; Petiniot, N.; Teyssié, P. H. J. Org. Chem. 1980, 45, 695–702. (b) Doyle, M.
P.; Griffin, J. H.; Bagheri, V.; Dorow, R. L. Organometallics 1984, 3, 53–61.
7. (a) Doyle, M. P. Asymmetric Cyclopropanation in Catalytic Asymmetric Synthesis; Ojima, I., Ed. VCH: New York, 1993;
pp 63–99. (b) Singh, V. K.; Datta Gupta, A.; Sekar, G. Synthesis 1997, 137–149. (c) Doyle, M. P.; McKervey, M. A. J.
Chem. Soc., Chem. Commun. 1997, 983–989. (d) Noyori, R. Asymmetric Catalysis in Organic Synthesis; John Wiley &
Sons: New York, 1994; pp. 199–207.
8. Aratani, T. Pure Appl. Chem. 1985, 57, 1839–1844.
9. Fritschi, H.; Leutenegger, U.; Pfaltz, A. Helv. Chim. Acta 1988, 71, 1553–1565.
10. (a) Lowenthal, R. E.; Abiko, A.; Masamune, S. Tetrahedron Lett. 1990, 31, 6005–6008. (b) Lowenthal, R. E.; Masamune,
S. Tetrahedron Lett. 1991, 32, 7373–7376. (c) Masamune, S.; Lowenthal, R. E. U.S. Patent 5298623, 1994. [Chem. Abstr.
1994, 121, 205742 g.]
11. (a) Evans, D. A.; Woerpel, K. A.; Hinman, M. M.; Faul, M. M. J. Am. Chem. Soc. 1991, 113, 726–728. (b) Evans, D. A.;
Woerpel, K. A.; Scott, M. J. Angew. Chem., Int. Ed. Engl. 1992, 31, 430–432.
12. (a) Doyle, M. P.; Protopopova, M. N.; Brandes, B. D.; Davies, H. M. L.; Huby, N. J. S.; Whitesell, J. K. Synlett 1993,
151–153. (b) Doyle, M. P.; Winchester, W. R.; Protopopova, M. N.; Müller, P.; Bernadinelli, G.; Ene, D.; Motallebi, S.
Helv. Chim. Acta 1993, 76, 2227–2235. (c) Müller, P.; Baud, C.; Ené, D.; Motallebi, Doyle, M. P.; Brandes, B. D.; Dyatkin,
A. B.; See, M. M. Helv. Chim. Acta 1995, 78, 459–470.
13. (a) Nishiyama, H.; Itoh, Y.; Matsumoto, H.; Park, S. -B.; Itoh, K. J. Am. Chem. Soc. 1994, 116, 2223–2224. (b) Nishiyama,
H.; Itoh, Y.; Sugawara, Y.; Matsumoto, H.; Aoki, K.; Itoh, K. Bull. Chem. Soc. Jpn. 1995, 68, 1247–1262. (c) Park, S.-
B.; Murata, K.; Matsumoto, H.; Nishiyama, H. Tetrahedron: Asymmetry 1995, 6, 2487–2494. (d) Park, S.-B.; Sakata, N.;
Nishiyama, H. Chem. Eur. J. 1996, 2, 303–306.
14. Schumacher, R.; Dammast, F.; Reibig, H.-U. Chem. Eur. J. 1997, 3, 614–619.
15. Camps, F.; Coll, J.; Moretó, J. M. Synthesis 1982, 186–188.
16. For examples of related derivatives: (a) Itsuno, S.; Nakano, M.; Miyazaki, K.; Masuda, H.; Ito, K. J. Chem. Soc., Perkin
Trans. 1 1985, 2039–2044. (b) Wey, S.; O’Connor, K. J.; Burrows, C. J. Tetrahedron Lett. 1993, 34, 1905–1908.
17. Doyle, M. P. Asymmetric Cyclopropanation in Catalytic Asymmetric Synthesis; Ojima I, Ed. VCH: New York, 1993; pp.
63–99.
18. (a) Mahroof-Tahir, M.; Karlin, K. D. J. Am. Chem. Soc. 1992, 114, 7599–7601. (b) Harding, C.; McKee, V.; Nelson, J. J.
Am. Chem. Soc. 1991, 113, 9684–9685. (c) Long, R. C.; Hendrickson, D. N. J. Am. Chem. Soc. 1983, 105, 1513–1521. (d)
Hendrickson, D. N.; Long, R. C.; Hwang, Y. T.; Chang, H. R. In Biological and Inorganic Copper Chemistry; Karlin, K.
D.; Zubieta, J., Eds. Adenine Press: Guilderland, New York, 1985. (e) Gagné, R. R.; Koval, C. A.; Smith, T. J.; Comolino,
M. C. J. Am. Chem. Soc. 1979, 101, 4571–4580. (f) Gagné, R. R.; Henling, L. M.; Kistenmacher, T. J. Inorg. Chem. 1980,
19, 1226–1230. (g) Gagné, R. R.; Koval, C. A.; Smith, T. J. J. Am. Chem. Soc. 1977, 99, 8367–8368.
19. Kanemasa, S.; Hamura, S.; Harada, E.; Yamamoto, H. Tetrahedron Lett. 1994, 43, 7985–7988.
20. Jones, B. J. Chem. Soc. 1941, 358–364.