Vaulted biaryls in catalysis: A structure-activity relationship guided tour of the immanent domain of the VANOL ligand

Article abstract of DOI:10.1002/chem.201302451

The active site in the BOROX catalyst is a chiral polyborate anion (boroxinate) that is assembled in situ from three equivalents of B(OPh) ₃ and one of the VANOL ligand by a molecule of substrate. The substrates are bound to the boroxinate by Hbonds to oxygen atoms O1-O3. The effects of introducing substituents at each position of the naphthalene core of the VANOL ligand are systematically investigated in an aziridination reaction. Substituents in the 4,4′- and 8,8′-positions have a negative effect on catalyst performance, whereas, substituents in the 7- and 7′-positions have the biggest impact in a positive direction. VANOL destination: The active site in the BOROX catalyst is a chiral polyborate anion (boroxinate; see figure) that is assembled in situ from three equivalents of B(OPh)₃ and one of the VANOL ligand by a molecule of substrate. The effects of introducing substituents at each position of the naphthalene core of the VANOL ligand are systematically investigated in an aziridination reaction. Copyright

Full text of DOI:10.1002/chem.201302451

FULL PAPER
DOI: 10.1002/chem.201302451
Vaulted Biaryls in Catalysis: A Structure–Activity Relationship Guided Tour
of the Immanent Domain of the VANOL Ligand
Yong Guan, Zhensheng Ding, and William D. Wulff*[a]
Abstract: The active site in the
BOROX catalyst is a chiral polyborate
anion (boroxinate) that is assembled in
effects of introducing substituents at
each position of the naphthalene core
of the VANOL ligand are systematical-
ly investigated in an aziridination reac-
tion. Substituents in the 4,4’- and 8,8’-
positions have a negative effect on cat-
alyst performance, whereas, substitu-
ents in the 7- and 7’-positions have the
biggest impact in a positive direction.
situ from three equivalents of BACTHNURGTNEUNG(OPh)3
and one of the VANOL ligand by
a molecule of substrate. The substrates
are bound to the boroxinate by
H bonds to oxygen atoms O1–O3. The
Keywords: asymmetric catalysis
·
biaryls · catalysis · structure–activity
relationships
Introduction
The vaulted biaryl ligands VANOL and VAPOL have been
demonstrated to be effective in a variety of useful catalytic
asymmetric reactions including amidation of imines,[1] ami-
noallylation of aldehydes,[2] aza-Darzens reaction[3] cis-aziri-
dination of imines[4,5] trans-aziridination of imines,[4q,6]
Baeyer–Villiger reactions,[7] benzoyloxylation of aryloxin-
doles,[8] desymmetrization of aziridines,[9] Diels–Alder reac-
tions,[10] heteroatom Diels–Alder reactions,[11] hydroarylation
of alkenes,[12] hydrogenation of alkenes,[13] imidation of
imines,[14] Mannich reactions,[15] multicomponent aziridina-
tion of aldehydes,[16] Petasis reaction,[17] propargylation of
ketones,[18] reduction of imines,[19] chlorination and Michael
reactions of oxindoles,[20] hydroacylation of alkenes,[21] pina-
col rearrangement,[22] reduction of aminals[23] and Michael
addition of alkynes.[24]
Scheme 1. Substrate assembled BOROX catalysts.
Many of the reactions of imines with VANOL and
VAPOL catalysts involve a chiral polyborate catalyst with
a chiral boroxinate anionic core that exists as an ion pair
with the protonated imine and these together constitute the
catalyst–substrate complex (Scheme 1). These boroxinate, or
BOROX catalysts, are assembled in situ by the imine sub-
strate from a molecule of the VANOL or VAPOL ligand
Scheme 2. Imine substrates in BOROX catalyzed reactions.
and BACHTUNGTRENNUNG
(OPh)3.[4j,m,6b,16b] These catalysts have been shown to
occur in cis-[4,5] and trans-aziridination[4q,6] reactions as well
as in aza-Cope rearrangement[2] and are presumably in-
volved in heteroatom Diels–Alder reactions of imines as
well (Scheme 2).[11]
The 3D structure of the (S)-VANOL–BOROX anion
(Figure 1) is based on several X-ray structures determined
for these boroxinate species with the corresponding cationic
moiety removed.[4j,m] In the calculated transition state for
the first step[6d] in the cis-aziridination reaction both sub-
strates are H bonded to various oxygen atoms of the boroxi-
nate anionic core.[6b,c] Both substrates are also H bonded to
the boroxinate anion in the transition state for the first
step[6d] in the trans-aziridination reaction as well.[6b,c] Howev-
er, in the latter case the diazo compound (diazoacetamide)
is bound to the catalyst by two H bonds, one from the hy-
[a] Y. Guan, Dr. Z. Ding, Prof. W. D. Wulff
Department of Chemistry, Michigan State University
East Lansing, MI 48824 (USA)
E-mail: wulff@chemistry.msu.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201302451; including procedures
for the preparation for the 3-phenyl phenols 15 and the 44 new
VANOL derivatives and the asymmetric syntheses of aziridines 12.
Chem. Eur. J. 2013, 19, 15565 – 15571
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
15565

identical asymmetric inductions (Æ1%) over the entire
scope of imines that have been screened in the aziridination
reaction.[25] Nonetheless, substituents of different shapes and
sizes and electronic properties may have interactions with
the bound substrates that are disparate from those of a flat
rigidly projected benzene ring.
The purpose of the present work is to systematically map
the contours of the VANOL–BOROX catalyst to observe
how alterations in its domain will effect reactions occurring
at the chiral boroxinate core. The response to the introduc-
tion of various substituents at all five of the open positions
on the naphthalene core on the VANOL ligand (C4–C8) is
monitored as a function of the asymmetric induction in the
cis-aziridination reaction.
It is clear at this point in the development of asymmetric
catalysis, that universal catalysts will not be the norm. Each
reaction class and catalyst type will typically require its own
optimized ligand. This is no more readily evident than in the
chemistry of BINOL ligands.[26] It is commonly observed
that 3,3’-disubstituted BINOLs are superior to BINOL but
for some reactions and with some catalysts the 6,6-disubsti-
tuted BINOLs are the best. It would be interesting to know
how the same substituent introduced into each of the six
open positions in BINOL would effect the asymmetric in-
duction in a given reaction but to the best of our knowledge
this type of study has not been done for BINOL. Given the
established utility of VANOL in the broad range of asym-
metric reactions outlined in the introduction, the synthetic
methods for the preparation of 44 different VANOL deriva-
tives described in this work would be deemed to be of value
not only for boroxinate derived VANOL catalysts but also
for many other types of catalysts based on VANOL.
Figure 1. The 3D structure of the (S)-VANOL–BOROX anion.
drogen of the diazo carbon and one from the amide hydro-
gen.
It can be anticipated that the introduction of substituents
on any of the open positions (C4–C8) of the naphthalene
core of VANOL may have an effect on the binding of either
and/or both of the substrates in the aziridination reaction
leading to either increased or decreased asymmetric induc-
tions (Figure 2). However, the VAPOL ligand is nothing
more than a VANOL ligand with substituents in the C7 and
C8 positions in the form of a fused benzene ring and both
VANOL and VAPOL BOROX catalysts give essentially
Results and Discussion
For each of the five positions on VANOL several different
substituents were investigated including bromide, p-tert-bu-
tylphenyl and p-trifluoromethylphenyl groups (Table 1).
Substituents in the 4- and 8-positions of VANOL had the
biggest negative impact on the asymmetric inductions falling
to 3%ee with ligand 4q and to 8%ee with ligand 8ah. The
large influence of substituents in the 4-position was not an-
ticipated given the remoteness of these groups from the
active site of the boroxinate catalyst. This may have to do
with the influence of groups in the 4-position on the twist
angle between the two naphthalene rings in the VANOL
and the consequences that this might have on the stability
of the boroxinate core since it is believed that its formation
is equilibrium dependent. Nonetheless, the 11B NMR spec-
trum revealed that at least some of the boroxinate species is
formed, but it is certainly possible that there could be signif-
icant background reactions with nonchiral borate species
(see the Supporting Information). The significant negative
effect of the substituents in the 8-position could also be due
to an increase in background reaction as a result of incom-
plete boroxinate formation although, again the 11B NMR
Figure 2. Positional isomers of the VANOL ligand.
15566
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15565 – 15571

W. D. Wulff
FULL PAPER
Table 1. VANOL positional isomer screen in the aziridination reaction.[a]
The fact that the VANOL positional isomer screen identi-
fied the 7-position of VANOL as the “hot” position for ef-
fecting aziridination reactions occurring in the boroxinate
active site, prompted the generation of a library of VANOL
ligands that are substituted at the 7- and 7’-positions. The
fact that the 7-position emerged as optimal was fortunate
since, from the synthetic point of view, it is clear that this is
the best position for diversity to proliferate with the greatest
of ease. These ligands can be readily obtained from p-substi-
tuted phenyl acetic acids through a cycloaddition/electrocyc-
lization cascade (CAEC) process[27] and a few examples are
indicated in Scheme 3. Simply heating the acid chloride 13
with phenyl acetylene led to the corresponding 3-phenyl-1-
naphthol (15) after liberation from the iso-butryate ester 14.
iso-Butyric anhydride was added to the CAEC cascade to
react with 15 and prevent it from being trapped by the aryl
ketene intermediate generated from 13. An oxidative
phenol coupling reaction with air was used to acquire the
racemic ligand in good to high yields. Although a large-scale
resolution method has been developed for VANOL,[27] on
the laboratory scale, the most convenient way to access the
optically pure ligands is to perform a deracemization with
a copper complex of (À)-sparteine (or (+)-sparteine), which
Entry Ligand
R
12d R1 =Ph
Yield
12i R1 =Cy
[b]
ee [%][c] Yield [%][b] ee [%][c]
1
2
3
4
5
6
7
8
VANOL
VAPOL
4q
5q
6q
7q
4d
5d
6d
7d
5p
6p
7p
8h
8ah
9
–
–
84
76
92
93
3
77
78
68
79
77
83
81
80
76
78
84
86
93
72
26[e]
87
81
82
1
p-tBuC6H4 75
p-tBuC6H4 87
p-tBuC6H4 82
p-tBuC6H4 85
Br
Br
Br
Br
p-CF3C6H4 92
p-CF3C6H4 92
p-CF3C6H4 92
Me
Ph
–
84
89
97
45
92
90
89
78
87
84
80
8
74
82
93
45
83
79
85
69
78
78
77
15
90
86
87
83
89
9
10
11
12
13
14
15
16
83
62[d]
90
82
[a] Unless otherwise specified, all reactions were run at 0.5m in imine in
toluene on a 0.5 mmol scale with 1.2 equiv EDA at 258C for 24 h and
went to 100% completion with 5 mol% catalyst. The catalyst was pre-
give the (S)- or (R)-ligands 7 in >99%ee in all cases.[28]
A
pared from 1 equiv ligand, 4 equiv BACHTUNTRGNEUNG(OPh)3 and 1 equiv H2O at 808C in
31-member library of 7,7’-disubstituted VANOL ligands was
generated and their syntheses are all described in the Sup-
porting Information. About a third of these were prepared
from commercially available p-substituted phenyl acetic
acids (Scheme 3) and the rest were prepared by a variety of
toluene for 1 h, followed by removal of volatiles under vacuum
(0.5 mmHg) at 808C for 0.5 h. [b] Yield of isolated cis-aziridine by chro-
matography on silica gel. [c] Determined by HPLC on a Chiralcel OD-H
column. [d] The reaction time was 4 d and conversion was 88%. [e] The
reaction time was 4 d and conversion was 78%.
spectrum clearly shows that
some boroxinate has formed.
Alternatively,
the
negative
effect of substituents in the 8-
position could also be due to
a simple disruption of the bind-
ing or of the nature of the bind-
ing of the substrates to the bor-
oxinate core.
Of all of the positional iso-
mers of VANOL that were
screened, the only ligand that
gave an asymmetric induction
that was significantly higher
than VANOL or VAPOL was
the ligand 7q that has a p-tert-
butylphenyl group in the 7- and
Scheme 3. Library synthesis of 7,7’-disubstituted VANOLs.
7’-positions and, curiously, this
was true for both imines 10d
and 10i (Table 1, entry 6). The results from the octahydro-
VANOL ligand 9 were quite intriguing. Although no overall
increase in induction was observed for the catalyst from 9,
the induction for the imines 10d and 10i were reversed. The
induction with the phenyl imine 10d fell from 92 to 82%ee,
the induction for the cyclohexyl imine 10i rose from 81 to
90%ee.
different coupling reactions of the 7,7’-disubstituted diha-
lides (S)-7d and (S)-7e, or their enantiomers.
The data in Table 2 reveal that the majority of all of the
31 different substituents in the 7- and 7’-positions examined
led to improved asymmetric inductions for both aziridines
(20 for 12d and 24 for 12i). It was most remarkable to find
that out of all thirty of the new VANOL ligands, the best
Chem. Eur. J. 2013, 19, 15565 – 15571
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15567

W. D. Wulff et al.
Table 2. Screen of the 7,7’-disubstituted VANOL Library.[a]
strates reveals that the interplay between sterics and elec-
tronics of the aryl group is indeed delicate. The 4-methyl-
phenyl group in ligand 7o leads to a definite jump in asym-
metric induction compared with VANOL and a slight fur-
ther increase is seen with the 4-tert-butylphenyl substituent
in 7q. This increase is more than negated by a 4-trifluoro-
methyl group (7p). Any advantage of a phenyl substituent
in the 7- and 7’-positions is lost with the introduction of
ortho groups on the phenyl ring (7u and 7v).
All reactions in Table 2 were carried out for 24 h in order
to ensure that any differences in rates for the various ligands
could be accommodated. Most of the reactions in fact were
probably complete in substantially less time than 24 h. The
reaction with the optimal ligand 7m with imine 10d was re-
examined with regard to minimum reaction time to find that
the reaction is complete in 4 h with 5 mol% catalyst to give
an 89% yield of 12d in 97.4%ee. This repeat reaction was
performed on a sample of the ligand 7m that was stored in
the refrigerator under nitrogen for two years subsequent to
the first run indicated in Table 2.
The boroxinate catalyst generated from the 7,7-di-tert-bu-
tylVANOL ligand 7m gave higher asymmetric inductions
than the parent VANOL ligand 1 not only for the aziridina-
tion of the imines 10d and 10i (Tables 1 and 2), but all ten
of the imines shown in Table 3. This set of imines was
chosen since all have been previously examined with cata-
lysts generated from VANOL and VAPOL.[13h] The asym-
metric inductions with the di-tert-butylVANOL catalyst
range from 94–99%ee over the ten substrates. It is clear
that there is a dramatic increase in the average asymmetric
induction over these ten substrates that result from the in-
Ligand
R
12d R1 =Ph
12i R1 =Cy
Yield
ee
Yield
ee
[%][b]
[%][c]
[%][b]
[%][c]
VANOL
7b
7c
7d
7e
7 f
7g
7h
7i
7j
H
F
Cl
Br
84
80
91
89[d]
85
96
86
82[e]
90
95
93
94
82[f]
77
94
92
85
82
85
94
77
92
91
88
96
95
96
98
91
86
86
94
92
83
89
89[d]
92
86
96
86[e]
94
94
97
94
98[f]
94
96
84
97
95
97
95
91
92
95
84
95
96
96
95
93
93
92
94
77
76
83
78
80
88
88
83
91
91
91
90
88
72
89
93
83
88
77
90
72
87
94
72
90
77
91
90
90
85
79
87
81
82
78
85
88
85
92
87
87
89
89
87
94
79
90
78
93
92
86
80
82
81
90
63
88
76
90
89
82
84
88
87
I
CF3
OMe
Me
Et
nBu
iPr
Cy
tBu
7k
7l
7m
7n
7o
7p
7q
7r
7s
7t
7u
7v
7w
7x
SiPh2tBu
4-MeC6H4
4-CF3C6H4
4-tBuC6H4
3,5-Me2C6H3
3,5-(tBu)2-4-MeOC6H2
3,5-(CF3)2C6H3
2,6-Me2C6H3
1-naphthyl
2-naphthyl
9-anthracenyl
2-C4H3S
2-C4H3O
3-C4H3S
3-C4H3O
7y
7z
7aa
7ab
7ac
7ad
7ae
7af
À
CH=CH
À ꢀ
C CH
Table 3. Substrate scope: VANOL 1 versus 7,7ꢁ-tBu2VANOL 7m.[a]
À ꢀ
C CSiMe3
À ꢀ
C C-tBu
[a] Unless otherwise specified, all reactions were run at 0.5m in imine in
toluene on a 0.5 mmol scale with 1.2 equiv EDA at 258C for 24 h and
went to 100% completion with 5 mol% catalyst. The catalyst was pre-
pared as indicated in Table 1. [b] Yield of isolated cis-aziridine by chro-
matography on silica gel. [c] Determined by HPLC on a Chiralcel OD-H
column. [d] A repeat of this reaction on 1.0 mmol scale gave 80% yield
and 88%ee. [e] A repeat of this reaction on 1.0 mmol scale gave 87%
yield and 86%ee. [f] A repeat of this reaction revealed that it was com-
plete in 4 h to give 89% yield and 97.4%ee.
Series
VANOL BOROX[b] 7,7’-tBu2VANOL BOROX
R1
Yield 12
ee 12
Yield 12
ee 12
[%][c]
[%][d]
[%][c]
[%][d]
a
b
c
d
e
f
g
h
i
4-NO2C6H4 86
89
94
82
93
93
90
87
77
82
85
87
96
98
98
95
98
99
97
98
94
94
96
97
4-BrC6H4
2-BrC6H4
C6H5
1-naphthyl
2-MeC6H4
86
90
43[e]
87
78[f]
82
80
91
67[g]
92[h]
71
4-MeOC6H4 67
ligand for the cyclohexyl imine 10i was also the best ligand
for the phenyl imine 10d. Given this singularity for the 7,7’-
di-tert-butylVANOL ligand 7m, it was not surprising to find
that many of the other trends in asymmetric induction ob-
served for the cyclohexyl imine 10i were also observed for
10d. The two groups of substituents that gave the largest in-
creases were bulky alkyl groups and aromatic substituents.
There are definite electronic effects that appear as well that
can be observed for the halogens, in which the inductions
are typically slightly decreased relative to a methoxyl group
that gives rise to a sizable increase. The series of aryl sub-
n-propyl
cyclohexyl
tert-butyl
average
54
79
89
74
77
88
89
85
j
[a] Unless otherwise specified, all reactions were run at 0.5m in imine in
toluene on a 0.5 mmol scale with 1.2 equiv EDA at 258C for 24 h and
went to 100% completion with 5 mol% catalyst. The catalyst was pre-
pared as indicated in Table 1. [b] Data taken from ref. [4h]. [c] Yield of
isolated cis-aziridine by chromatography on silica gel. [d] Determined by
HPLC on a Chiralcel OD-H column. [e] Plus 23% yield trans-6c (cis/
trans=1.9:1). [f] Plus 10% yield trans-6c (cis/trans=8:1). [g] Plus 6%
yield trans-6 f (cis/trans=12:1); [h] cis/trans>100:1.
15568
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15565 – 15571

W. D. Wulff
FULL PAPER
troduction of the tert-butyl in the 7- and 7’-positions of
VANOL (87 vs. 97%ee). Not only are the asymmetric in-
ductions dramatically improved with the di-tert-butylVA-
NOL catalyst, but the general efficiency of the reaction is
also greatly improved with the average yield increasing by
11% over the VANOL catalyst. This increase in efficiency
appears to be the result of a combination of two factors:
a decrease in the amount of enamine side-products and an
increase in the diastereoselectivity in favor of the cis-aziri-
dine. The latter is particularly manifested in the aziridina-
tion of imines derived from ortho-substituted benzaldehydes.
One of the weak points we had previously established for
the aziridination reaction with VANOL and VAPOL cata-
lysts is the lower cis/trans selectivity and the lower asymmet-
ric inductions observed for imines derived from ortho-substi-
tuted benzaldehydes.[13h] For example, the reaction of the 2-
bromophenyl imine 10c with the VANOL catalyst gives
a 1.9:1 ratio of cis to trans and the di-tert-butylVANOL cata-
lyst gives an 8:1 mixture. Improved diastereoselection is also
observed for the ortho-methylphenyl imine 10 f with cis/
trans ratios increasing from 12:1 to >100:1 for the VANOL
and di-tert-butylVANOL catalysts, respectively. Thus, the di-
tert-butylVANOL catalyst is superior to either VANOL or
VAPOL catalysts in providing diastereoselective and enan-
tioselective access to aziridines from benzhydryl imines.
Experimental Section
Preparation of 7-(tert-butyl)-3-phenylnaphthalen-1-ol (15m) through the
cycloaddition/electro-cyclization cascade: A single-neck 500 mL round
bottom flask equipped with a condenser was charged with 4-tert-butyl-
phenylacetic acid 22m (13.44 g, 70.0 mmol) and SOCl2 (18.7 mL,
256 mmol). The top of the condenser was vented to a bubbler and then
into a beaker filled with NaOH (sat. aq.) to trap acidic gases (HCl and
SO2). The mixture was heated to reflux for 1 h in a 908C oil bath, and
then all of the volatiles were carefully removed by swirling it under high
vacuum (1 mm Hg) for 1 h with a second liquid N2 trap to protect the
pump. Under N2 phenylacetylene (10.3 mL, 94 mmol) and (iPrCO)2O
(23.4 mL, 141 mmol) were added to the flask containing the acyl chloride.
The mixture was stirred at 1908C for 48 h with a gentle nitrogen flow
across the top of the condenser. Sometimes two condensers are required-
to ensure the efficient return of phenylacetylene. The brown reaction
mixture was cooled to below 1008C (ca. 608C, oil bath temperature) and
a solution of KOH (23.3 g, 415 mmol) in H2O (93 mL) was then added
slowly. This two-phase mixture was stirred at 1008C, overnight. The mix-
ture was cooled to room temperature and ethyl acetate (200 mL) was
added and the mixture stirred for 10 min before the organic layer was
separated. The aqueous layer was extracted twice with ethyl acetate
(100 mLꢂ3) and the combined organic layer was washed with brine
(100 mL), dried over MgSO4, filtered through Celite and concentrated to
dryness. Purification of the crude product by column chromatography on
silica gel (50 mmꢂ250 mm, CH2Cl2/hexanes 1:3 to 1:1 to 1:0) gave 15m
as an off-white solid (9.78 g, 35.4 mmol, 51%). M.p.: 135–1378C; Rf =
0.40 (CH2Cl2). Spectral data for 15m: 1H NMR (CDCl3, 500 MHz): d=
1.43 (s, 9H), 5.25 (s, 1H), 7.06 (d, 1H, J=2.0 Hz), 7.32–7.37 (m, 1H),
7.42–7.47 (m, 2H), 7.59–7.62 (m, 2H), 7.64–7.67 (m, 2H), 7.80 (d, 1H,
J=8.5 Hz), 8.08–8.10 ppm (m, 1H); 13C NMR (CDCl3, 125 MHz): d=
31.31, 35.10, 108.41, 116.25, 118.40, 123.31, 125.80, 127.24, 127.29, 127.81,
128.78, 133.22, 138.19, 141.06, 148.30, 151.66 ppm; IR (thin film): v˜ =3505
(brs), 2961 (s), 1601 (s), 1559 (s), 1458 (s), 1408 (s), 1273 cmÀ1 (s); MS:
m/z (%): 276 [M]+ (54), 261 (96), 233 (15), 202 (24), 189 (15), 165 (9),
130 (13), 116 (100); elemental analysis calcd for C20H20O: C 86.92, H
7.29; found: C 86.92, H 7.04.
Conclusion
A systematic mapping of the sensitivity of the active site in
a VANOL boroxinate catalyst in response to steric and elec-
tronic effects of substitutions in the naphthalene core to the
catalytic transfer of chirality in an aziridination reaction has
been carried out. A screen of all five positional isomers of
VANOL revealed that the chirality transfer was most sensi-
tive in a positive direction to substitution in the 7- and 7’-po-
sitions. A subsequent screen of a set of 31 members of the
7,7’-disubstituted VANOL ligands identified the 7,7’-di-tert-
butylVANOL as optimal. A screen of this ligand for the cat-
alytic asymmetric aziridination of a set of ten different
imines found that this ligand gave higher asymmetric induc-
tions than the unsubstituted VANOL for every imine with
an overall average of 10% higher ee and 11% higher yield.
The 7,7’-di-tert-butylVANOL 7m is clearly the ligand of
choice for the catalytic asymmetric aziridination reaction
with benzhydryl imines, which can be readily prepared from
the commercially available diphenylmethylamine. However,
given the fact that the vaulted biaryl ligands VANOL and
VAPOL ligands have been demonstrated to be effective in
over twenty other catalytic asymmetric reactions, the family
of disubstituted VANOL ligands described here (not just
the 7,7’-derivatives) may well provide the diversity necessary
to identify ligands of optimal performance for other asym-
metric catalysts for a broad range of reactions. A prime
reason for the consideration of such diversity screens is the
ease of synthesis of the various VANOL derivatives.
Preparation of (S)-7,7’-di-tert-butyl VANOL 7m
Air-mediated phenol coupling: 7-(tert-Butyl)-3-phenylnaphthalen-1-ol
15m (201 mg, 0.73 mmol) and mineral oil (1 mL) were added to
a 500 mL flame-dried three neck round bottom flask equipped with
a cooling condenser. Airflow was introduced from one side neck via
a needle located one inch above the mixture. The airflow rate was about
one bubble per second. The mixture was stirred at 1658C for 24 h. After
being cooled to room temperature, CH2Cl2 (1 mL) and hexanes (2 mL)
were added to the flask and the mixture was stirred until all large pieces
were broken up. Purification by column chromatography on silica gel
(30 mmꢂ250 mm, CH2Cl2/hexanes 1:2) gave racemic 7m as an off-white
solid (145 mg, 0.26 mmol, 72%).
Deracemization:[28,29] To a 50 mL round bottom flask was added (À)-spar-
teine (782 mg, 3.34 mmol), CuCl (160 mg, 1.62 mmol) and MeOH
(26 mL) under an atmosphere of air. The reaction mixture was sonicat-
ed[29] in a water bath for 60 min with exposure to air.[29] The flask was
then sealed with a septum and purged with argon, which was introduced
by a needle under the surface for 60 min. At the same time, to a 250 mL
flame-dried round bottom flask was added racemic 7m (525 mg,
0.95 mmol) and CH2Cl2 (103 mL). The resulting solution was purged with
argon for 60 min under the surface. The green CuII-sparteine solution was
then transferred via cannula to the solution of racemic 7m under argon
and then the combined mixture was sonicated[29] for 15 min and then al-
lowed to stir at room temperature, overnight, with an argon balloon at-
tached to the flask which was covered with aluminum foil. The reaction
was quenched by slow addition of sat. aq. NaHCO3 (12 mL), H2O
(40 mL) and most of the organic solvent was removed under reduced
pressure. The residue was then extracted with CH2Cl2 (30 mLꢂ3). The
combined organic layer was dried over MgSO4, filtered through Celite
and concentrated to dryness. The product was purified by column chro-
matography on silica gel (30 mmꢂ250 m, CH2Cl2/hexanes 1:2) to afford
Chem. Eur. J. 2013, 19, 15565 – 15571
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15569

W. D. Wulff et al.
(S)-7m as an off-white foamy solid (404 mg, 0.73 mmol, 77%). The opti-
cal purity was determined to be >99%ee by HPLC analysis (Pirkle d-
Phenylglycine column, 99:1 hexane/iPrOH at 254 nm, flow-rate:
1.0 mLminÀ1). Retention times: tR =8.67 min for (R)-7m (minor) and
tR =10.19 min for (S)-7m (major). M.p.: 154–1568C; Rf =0.26 (1:2
Acknowledgements
This work was supported by the National Institute of General Medical
Sciences (GM 094478).
1
CH2Cl2/hexanes). Spectral data for 7m: H NMR (CDCl3, 500 MHz): d=
1.48 (s, 18H), 5.81 (s, 2H), 6.61 (dd, 4H, J=8.0, 1.0 Hz), 6.95 (t, 4H, J=
8.0 Hz), 7.03–7.07 (m, 2H), 7.28 (s, 2H), 7.66 (dd, 2H, J=8.5, 2.0 Hz),
7.73 (d, 2H, J=8.5 Hz), 8.29–8.30 ppm (m, 2H); 13C NMR (CDCl3,
125 MHz): d=31.33, 35.19, 112.72, 117.71, 121.63, 122.65, 126.39, 127.43,
127.45, 128.90, 132.83, 140.01, 140.40, 148.58, 150.24 ppm (1 sp2 C not lo-
cated); IR (thin film): v˜ =3519 (brs), 3058 (w), 2961 (s), 1597 (s), 1497
(s), 1385 (s), 1265 cmÀ1 (s); MS: m/z (%): 550 [M]+ (47), 535 (13), 275
(29), 260 (89), 232 (29); elemental analysis calcd for C40H38O2: C 87.23,
[1] G. B. Rowland, H. Zhang, E. B. Rowland, S. Chennamadhavuni, Y.
Wang, J. C. Antilla, J. Am. Chem. Soc. 2005, 127, 15696–15707.
[2] a) H. Ren, W. D. Wulff, J. Am. Chem. Soc. 2011, 133, 5656–5659;
b) H. Ren, W. D. Wulff, Org. Lett. 2013, 15, 242–245.
[3] S. E. Larson, G. Li, G. B. Rowland, D. Junge, R. Huang, H. Wood-
cock, J. C. Antilla, Org. Lett. 2011, 13, 2188.
[4] a) J. C. Antilla, W. D. Wulff, J. Am. Chem. Soc. 1999, 121, 5099–
5100; b) J. C. Antilla, W. D. Wulff, Angew. Chem. 2000, 112, 4692–
4695; Angew. Chem. Int. Ed. 2000, 39, 4518–4521; c) C. Loncaric,
W. D. Wulff, Org. Lett. 2001, 3, 3675–3678; d) A. Patwardhan, V. R.
Pulgam, Y. Zhang, W. D. Wulff, Angew. Chem. 2005, 117, 6325–
6328; Angew. Chem. Int. Ed. 2005, 44, 6169–6172; e) Y. Deng, Y. R.
Lee, C. A. Newman, W. D. Wulff, Eur. J. Org. Chem. 2007, 2068–
2071; f) Z. Lu, Y. Zhang, W. D. Wulff, J. Am. Chem. Soc. 2007, 129,
7185–7194; g) P. Wipf, M. A. Lyon, ARKIVOC 2007, 91–98; h) Y.
Zhang, A. Desai, Z. Lu, G. Hu, Z. Ding, W. D. Wulff, Chem. Eur. J.
2008, 14, 3785–3803; i) Y. Zhang, Z. Lu, A. Desai, W. D. Wulff, Org.
Lett. 2008, 10, 5429–5432; j) G. Hu, R. H. Huang, W. D. Wulff, J.
Am. Chem. Soc. 2009, 131, 15615–15617; k) M. Mukherjee, A. K.
Gupta, Z. Lu, Y. Zhang, W. D. Wulff, J. Org. Chem. 2010, 75, 5643;
l) H. Ren, W. D. Wulff, Org. Lett. 2010, 12, 4908–4911; m) G. Hu,
A. K. Gupta, R. H. Huang, M. Mukherjee, W. D. Wulff, J. Am.
Chem. Soc. 2010, 132, 14669–14675; n) L. Huang, W. D. Wulff, J.
Am. Chem. Soc. 2011, 133, 8892–8895; o) A. A. Desai, R. Morꢃn-
Ramallal, W. D. Wulff, Org. Synth. 2011, 88, 224–237; p) Z. Wang,
F. Li, L. Zhao, Q. He, F. Chen, C. Zheng, Tetrahedron 2011, 67,
9199–9203; q) L. Huang, Y. Zhang, R. J. Staples, R. H. Huang,
W. D. Wulff, Chem. Eur. J. 2012, 18, 5302–5313.
H
6.95; found:
C
86.90,
H
7.16; [a]2D0 =À215.2 (c 1.0, CH2Cl2) on
>99%ee (S)-7m (HPLC).
Asymmetric catalytic aziridination with a BOROX catalyst prepared
from (S)-7,7’-di-tert-butyl VANOL 7m: A 25 mL pear-shaped single
necked Schlenk flask, which had its 14/20 joint replaced by a threaded
high-vacuum Teflon valve was flame dried (with a stir bar in it) and
cooled to room temperature under N2 and charged with (S)-tBu2VANOL
7m (13.8 mg, 0.025 mmol) and triphenyl borate (29 mg, 0.10 mmol). The
mixture was dissolved in dry toluene (1 mL). After the addition of H2O
(0.45 mL, 0.025 mmol), the Teflon valve was closed and the flask was
heated at 808C for 1 h. Toluene was carefully removed by exposing to
high vacuum (0.1 mmHg) by slightly cracking the Teflon valve. After re-
moval of the solvent, the Teflon valve was completely opened and the
flask was heated to 808C under high vacuum for 30 min. Imine 10d
(136 mg, 0.50 mmol) and dry toluene (1 mL) were added to the Schlenk
flask containing the catalyst. The reaction mixture was stirred at room
temperature for 5 min and then ethyl diazoacetate (62 mL, 0.6 mmol) was
added via syringe. The Teflon valve was closed and the reaction mixture
was stirred at room temperature for 24 h. The mixture was then diluted
with hexanes (5 mL) and transferred to a 25 mL round bottom flask.
Rotary evaporation of the solvent followed by exposure to high vacuum
(0.5 mmHg) for 30 min gave the crude mixture as an off-white solid. The
conversion was determined from the 1H NMR spectrum of the crude re-
action mixture by integration of the aziridine ring methine protons rela-
tive to either the imine methine proton or the H on the imine carbon.
[5] For reviews, see: a) Y. Zhang, Z. Lu, W. D. Wulff, Synlett 2009,
2715–2739; b) A. A. Desai, H. Ren, M. Mukherjee, W. D. Wulff,
Org. Process Res. Dev. 2011, 15, 1108–1115.
[6] a) A. Desai, W. D. Wulff, J. Am. Chem. Soc. 2010, 132, 13100–
13103; b) M. Vetticatt, A. Desai, W. D. Wulff, J. Am. Chem. Soc.
2010, 132, 13104–14107; c) M. J. Vetticatt, A. A. Desai, W. D. Wulff,
J. Org. Chem. 2013, 78, 5142–5152. The rate-limiting step in this re-
1
The cis/trans ratio was determined to be >100:1 from the H NMR spec-
trum of the crude reaction mixture by integration of the ring methine
protons for each aziridine. The cis (J=6–8 Hz) and the trans (J=1–3 Hz)
coupling constants were used to differentiate the two isomers. The yields
of the acyclic enamine products were determined to be <1% from the
1H NMR spectrum of the crude reaction mixture by integration of the
À
action is the second step which involves ring closure via a C N
bond formation; see ref. [6c].
[7] C. Bolm, J.-C. Frison, Y. Zhang, W. D. Wulff, Synlett 2004, 1619–
1621.
[8] Z. Zhang, W. Zheng, J. C. Antilla, Angew. Chem. 2011, 123, 1167–
1170; Angew. Chem. Int. Ed. 2011, 50, 1135–1138.
À
N H proton of the enamine relative to the aziridine ring methine pro-
tons with the aid of the isolated yield of the cis-aziridine. The crude prod-
uct was purified by column chromatography on silica gel (35 mmꢂ
400 mm, EtOAc/hexanes 1:19) to afford 12d as a white solid (146 mg,
0.41 mmol, 82%). The optical purity was determined to be 98%ee by
HPLC (Chiralcel OD-H column, 222 nm, 90:10 hexane/iPrOH, flow rate:
0.7 mLminÀ1). Retention time: tR =4.42 min for (2S,3S)-12d (minor) and
tR =8.17 min for (2R,3R)-12d (major). The reaction was repeated and
stopped after a reaction time of 4 h to give an 89% yield of 12d with
97.4%ee. M.p.: 126–1278C; Rf =0.13 (1:9 EtOAc/hexanes). Spectral data
[9] a) E. B. Rowland, G. B. Rowland, E. Rivera-Otero, J. C. Antilla, J.
Am. Chem. Soc. 2007, 129, 12084–12085; b) G. Della Sala, A. Lat-
tanzi, Org. Lett. 2009, 11, 3330–3333; c) S. E. Larson, J. C. Baso, G.
Li, J. C. Antilla, Org. Lett. 2009, 11, 5186–5189; d) M. Senatore, A.
Lattanzi, S. Santoro, C. Santi, G. Della Sala, Org. Biomol. Chem.
2011, 9, 6205–6207; e) G. Della Sala, Tetrahedron 2013, 69, 50–56.
[10] a) J. Bao, W. D. Wulff, A. L. Rheingold, J. Am. Chem. Soc. 1993,
115, 3814–3815; b) J. Bao, W. D. Wulff, Tetrahedron Lett. 1995, 36,
3321–3324; c) D. P. Heller, D. R. Goldberg, W. D. Wulff, J. Am.
Chem. Soc. 1997, 119, 10551–10552; d) D. P. Heller, D. R. Goldberg,
H. Wu, W. D. Wulff, Can. J. Chem. 2006, 84, 1487–1503.
[11] a) C. A. Newman, J. C. Antilla, P. Chen, A. V. Predeus, L. Fielding,
W. D. Wulff, J. Am. Chem. Soc. 2007, 129, 7216–7217; b) C. A.
Newman, Ph.D. Thesis, Michigan State University, 2007.
[12] H. Harada, R. K. Thalji, R. G. Bergman, J. A. Ellman, J. Org. Chem.
2008, 73, 6772–6779.
1
for 12d: H NMR (CDCl3, 500 MHz): d=0.96 (t, 3H, J=7.0 Hz), 2.65 (d,
1H, J=7.0 Hz), 3.19 (d, 1H, J=7.0 Hz), 3.93 (s, 1H), 3.90–3.94 (m, 2H),
7.14–7.20 (m, 2H), 7.20–7.26 (m, 5H), 7.30–7.34 (m, 2H), 7.37–7.40 (m,
2H), 7.46–7.49 (m, 2H), 7.57–7.60 ppm (m, 2H); 13C NMR (CDCl3,
125 MHz): d=13.93, 46.38, 48.03, 60.55, 77.71, 127.19, 127.21, 127.31,
127.40, 127.54, 127.75, 127.78, 128.48, 135.03, 142.38, 142.52, 167.72 ppm
(1 sp2 C not located); IR (thin film): v˜ =3031 (w), 2982 (w), 1738 (s),
1456 (m), 1204 cmÀ1 (s); MS: m/z (%): 357 [M]+ (0.05), 190 (46), 167
(100), 117 (61); [a]2D0 = +36.2 (c 1.0, CH2Cl2) on 98%ee (2R,3R)-12d.
[13] P. W. N. M. Van Leeuwen, D. Rivillo, M. Raynal, Z. Freixa, J. Am.
Chem. Soc. 2011, 133, 18562.
[14] Y. Liang, E. B. Rowland, G. B. Rowland, J. A. Perman, J. C. Antilla,
Chem. Commun. 2007, 4477–4479.
[15] S. Xue, S. Yu, Y. Deng, W. D. Wulff, Angew. Chem. 2001, 113, 2331–
2334; Angew. Chem. Int. Ed. 2001, 40, 2271–2774.
15570
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15565 – 15571

W. D. Wulff
FULL PAPER
[16] a) A. K. Gupta, M. Mukherjee, W. D. Wulff, Org. Lett. 2011, 13,
5866–5869; b) A. K. Gupta, M. Mukherjee, G. Hu, W. D. Wulff, J.
Org. Chem. 2012, 77, 7932–7944; c) L. Huang, R. J. Staples, W. D.
Wulff, Chem. Sci. 2013, 4, 622–628.
[17] S. Lou, S. E. Schaus, J. Am. Chem. Soc. 2008, 130, 6922–6923.
[18] D. S. Barnett, S. E. Schaus, Org. Lett. 2011, 13, 4020.
[19] G. Li, Y. Liang, J. C. Antilla, J. Am. Chem. Soc. 2007, 129, 5830–
5831.
[24] G. Blay, L. Cardona, J. R. Pedro, A. Sanz-Marco, Chem. Eur. J.
2012, 18, 12966–12969.
[25] See pages S26 to S28 in the Supporting Information in footnote 4k.
[26] a) J. M. Brunel, Chem. Rev. 2005, 105, 857–897; b) Y. Chen, S.
Yekta, A. K. Yudin, Chem. Rev. 2003, 103, 3155–3211; c) T. Akiya-
ma, Chem. Rev. 2007, 107, 5744; d) M. Terada, Synthesis 2010, 1929.
[27] Z. Ding, W. E. G. Osminski, H. Ren, W. D. Wulff, Org. Process Res.
Dev. 2011, 15, 1089–1107.
[20] W. Zheng, Z. Zhang, M. J. Kaplan, J. C. Antilla, J. Am. Chem. Soc.
2011, 133, 3339–3341.
[28] G. Hu, D. Holmes, B. F. Gendhar, W. D. Wulff, J. Am. Chem. Soc.
2009, 131, 14355–14364.
[21] T. J. Hoffman, E. M. Carreira, Angew. Chem. 2011, 123, 10858–
10862; Angew. Chem. Int. Ed. 2011, 50, 10670–10674.
[22] S. A. Snyder, S. B. Thomas, A. C. Mayer, S. P. Breazzano, Angew.
Chem. 2012, 124, 4156–4160; Angew. Chem. Int. Ed. 2012, 51, 4080–
4084.
[29] Sonification is called for in the original procedure reported in
ref. [28] and is employed in the present work, but we have recently
found that the deracemization of VAPOL gives the same result
whether or not sonication is employed: Y. Guan, W. D. Wulff, un-
published results.
[23] M.-W. Chen, Q.-A. Chen, Y. Duan, Z.-S. Ye, Y.-G. Zhou, Chem.
Commun. 2012, 48, 1698–1700.
Received: June 25, 2013
Published online: October 2, 2013
Chem. Eur. J. 2013, 19, 15565 – 15571
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15571