Critical role of external axial ligands in chirality amplification of trans-cyclohexane-1,2-diamine in salen complexes

Article abstract of DOI:10.1021/ja904635n

A series of Mn^IV(salen)(L)₂ complexes bearing different external axial ligands (L ) Cl, NO₃, N₃, and OCH₂CF₃) from chiral salen ligands with trans-cyclohexane-1,2-diamine as a chiral scaffo

Full text of DOI:10.1021/ja904635n

Published on Web 08/10/2009
Critical Role of External Axial Ligands in Chirality
Amplification of trans-Cyclohexane-1,2-diamine in Salen
Complexes
Takuya Kurahashi,† Masahiko Hada,‡ and Hiroshi Fujii*,†
Institute for Molecular Science and Okazaki Institute for IntegratiVe Bioscience, National
Institutes of Natural Sciences, Myodaiji, Okazaki, Aichi 444-8787, Japan, and Department of
Chemistry, Graduate School of Science, Tokyo Metropolitan UniVersity, 1-1 Minami-Osawa,
Hachioji-shi, Tokyo 192-0397, Japan
Received June 8, 2009; E-mail: hiro@ims.ac.jp
Abstract: A series of MnIV(salen)(L)2 complexes bearing different external axial ligands (L ) Cl, NO3, N3,
and OCH2CF3) from chiral salen ligands with trans-cyclohexane-1,2-diamine as a chiral scaffold are
synthesized, to gain insight into conformational properties of metal salen complexes. X-ray crystal structures
show that MnIV(salen)(OCH2CF3)2 and MnIV(salen)(N3)2 adopt a stepped conformation with one of two
salicylidene rings pointing upward and the other pointing downward due to the bias from the trans-
cyclohexane-1,2-diamine moiety, which is in clear contrast to a relatively planar solid-state conformation
for MnIV(salen)(Cl)2. The CH2Cl2 solution of MnIV(salen)(L)2 shows circular dichroism of increasing intensity
in the order L ) Cl < NO3 , N3 < OCH2CF3, which indicates MnIV(salen)(L)2 adopts a solution conformation
of an increasing chiral distortion in this order. Quantum-chemical calculations with a symmetry adapted
cluster-configuration interaction method indicate that a stepped conformation exhibits more intense circular
dichroism than a planar conformation. The present study clarifies an unexpected new finding that the external
axial ligands (L) play a critical role in amplifying the chirality in trans-cyclohexane-1,2-diamine in
MnIV(salen)(L)2 to facilitate the formation of a chirally distorted conformation, possibly a stepped conformation.
linkage, leading to unique reactivity.7-9 It has been also shown
that salicylidene rings are appropriately modified for creating a
Introduction
Salen ligands, which are diimine-diphenolate ligands bearing
a linkage between two imine donors, have emerged as one of
the most versatile synthetic ligands. Salen ligands, which are
easily obtained by the condensation of a salicylaldehyde with a
diamine, securely bind various metal ions in a tetradentate
fashion. The malleability inherent to the salen ligand system
has enabled access to metal centers with a wide variety of
stereochemical and electronic properties and, thus, has led to
their extensive use. Salen metal complexes are still expanding
their utilities in catalysis,1 in asymmetric synthesis,2 in modeling
enzymes,3 in polymer synthesis,4 in biological application,5 and
as synthons for supramolecular materials.6
The most promising feature of salen complexes is easily
tunable stereochemical properties, including coordination ge-
ometries, conformations, and even supramolecular structures.
It has been shown that coordination geometries as well as
conformations could be modulated by changing a diamine
chiral environment around the metal center, resulting in exquisite
selectivity in asymmetric catalysis.10 There is much current
interest in constructing salen metal complexes with a helical
motif, in which intramolecular hydrogen bonding, π-π staking,
and metal binding sites are rationally incorporated to the
salicylidene rings to stabilize higher-order helical structures.11-13
A recent intriguing example that makes use of fine-tuned
(3) (a) Wu, A. J.; Penner-Hahn, J. E.; Pecoraro, V. L. Chem. ReV. 2004,
104, 903–938. (b) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L.,
Jr. Chem. ReV. 2004, 104, 939–986.
(4) Darensbourg, D. J. Chem. ReV. 2007, 107, 2388–2410.
(5) (a) Muller, J. G.; Kayser, L. A.; Paikoff, S. J.; Duarte, V.; Tang, N.;
Perez, R. J.; Rokita, S. E.; Burrows, C. J. Coord. Chem. ReV. 1999,
185-186, 761–774. (b) Thompson, K. H.; McNeill, J. H.; Orvig, C.
Chem. ReV. 1999, 99, 2561–2571. (c) Riley, D. P. Chem. ReV. 1999,
99, 2573–2587.
(6) (a) Miyasaka, H.; Saitoh, A.; Abe, S. Coord. Chem. ReV. 2007, 251,
2622–2664. (b) Wezenberg, S. J.; Kleij, A. W. Angew. Chem., Int.
Ed. 2008, 47, 2354–2364. (c) Kleij, A. W. Chem.sEur. J. 2008, 14,
10520–10529.
† National Institutes of Natural Sciences.
(7) Wang, Y.; DuBois, J. L.; Hedman, B.; Hodgson, K. O.; Stack, T. D. P.
Science 1998, 279, 537–540.
‡ Tokyo Metropolitan University.
(1) (a) Atwood, D. A.; Harvey, M. J. Chem. ReV. 2001, 101, 37–52. (b)
(8) Hormnirun, P.; Marshall, E. L.; Gibson, V. C.; Pugh, R. I.; White,
A. J. P. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 15343–15348.
(9) Kim, H.-J.; Kim, W.; Lough, A. J.; Kim, B. M.; Chin, J. J. Am. Chem.
Soc. 2005, 127, 16776–16777.
Cozzi, P. G. Chem. Soc. ReV. 2004, 33, 410–421.
(2) (a) Canali, L.; Sherrington, D. C. Chem. Soc. ReV. 1999, 28, 85–93.
(b) Jacobsen, E. N. Acc. Chem. Res. 2000, 33, 421–431. (c) Bandini,
M.; Cozzi, P. G.; Umani-Ronchi, A. Chem. Commun. 2002, 919–927.
(d) Irie, R.; Katsuki, T. Chem. Rec. 2004, 4, 96–109. (e) Katsuki, T.
Chem. Soc. ReV. 2004, 33, 437–444. (f) McGarrigle, E. M.; Gilheany,
D. G. Chem. ReV. 2005, 105, 1563–1602. (g) Baleiza˜o, C.; Garcia, H.
Chem. ReV. 2006, 106, 3987–4043. (h) Matsumoto, K.; Saito, B.;
Katsuki, T. Chem. Commun. 2007, 3619–3627.
(10) Watanabe, A.; Uchida, T.; Irie, R.; Katsuki, T. Proc. Natl. Acad. Sci.
U.S.A. 2004, 101, 5737–5742.
(11) Dong, Z.; Karpowicz, R. J., Jr.; Bai, S.; Yap, G. P. A.; Fox, J. M.
J. Am. Chem. Soc. 2006, 128, 14242–14243.
(12) Wiznycia, A. V.; Desper, J.; Levy, C. J. Inorg. Chem. 2006, 45, 10034–
10036.
9
12394 J. AM. CHEM. SOC. 2009, 131, 12394–12405
10.1021/ja904635n CCC: $40.75  2009 American Chemical Society

Role of External Axial Ligands in Chirality Amplification
A R T I C L E S
Chart 1. Salen Ligands That Are Utilized in This Study
groups incorporated to the salicylidene rings are not bulky
enough to induce distortion of the planar N2O2 coordination
geometry.21 Nevertheless, it has been frequently proposed that
a subtle deviation from planarity of the conformation of these
catalysts is critically important in determining the course of
asymmetric reactions.2f Among very few studies on conforma-
tional properties, Chin et al. reported intriguing X-ray structures
for the CoIII(salen) complexes from (R,R)- and (S,S)-1 that bind
two chiral aziridine molecules as a transition-state analogue for
enantioselective epoxidation of olefins or ring-opening reactions
of epoxides.16 They found the salicylidene rings in CoIII(s-
alen)(aziridine)2 complexes are considerably distorted from
planarity, which they claim is responsible for the different
binding affinity between the CoIII(salen) complexes from (R,R)-
and (S,S)-1 for the chiral aziridine in solution. In contrast, Fox
et al. recently pointed out that the distortion of the salicylidene
rings that is directed by trans-cyclohexane-1,2-diamine is
unexpectedly small, by rigorously studying the solution con-
formations of their well-designed NiII(salen) complexes.18
Our continuing interest in higher-valent intermediates from
salen complexes22 and the quite puzzling conformational
properties of Jacobsen’s salen complexes prompted us to study
conformations of higher-valent Mn(salen) complexes from 1.
We recently reported a preparation of MnIV(salen)(N3)2 from
1, a rare example of a monomeric high-valent MnIV complex
in the salen platform.23 An X-ray crystal structure of
MnIV(salen)(N3)2 from (R,R)-1 adopts a nonplanar stepped
conformation with one of two salicylidene rings pointing upward
and the other pointing downward, although the one-electron
reduced MnIII(salen)(N3)(CH3OH) exhibits a planar structure.
To gain further understanding, we herein prepare and fully
characterize monomeric MnIV(salen)(L)2 bearing different ex-
ternal axial ligands (L ) Cl, NO3, and OCH2CF3) from (R,R)-1
and investigate their solid-state and solution conformations by
stereochemical properties is a salen-based supramolecule that
shows a solvent-dependent transformation between a triangular
macrocycle and a helical coordination polymer.14
Among salen ligands reported so far, (R,R)- or (S,S)-N,N′-
bis(3,5-di-tert-butylsalicylidene)-1,2-cyclohexanediamine (1, Chart
1), which was developed by Jacobsen and co-workers, is now
known as one of the “privileged chiral ligands”, and metal
complexes from 1 show high levels of enantioselectivity for
mechanistically unrelated reactions such as epoxidation of
olefins, ring-opening reaction of epoxides, and conjugate
addition of azide to unsaturated imides.15 In spite of outstanding
enantioselectivity generally seen for Jacobsen’s catalyst, the
origins of enantioselectivity are not fully understood and still
remain to be the subject of active research.2f,16-19 The confor-
mational properties of these chiral catalysts are of prime
importance, but very few studies on this subject have been
reported.16,18 Jacobsen’s ligand 1 possesses a trans-cyclohexane-
1,2-diamine linkage as the only chiral unit and salicylidene rings
modified by sterically demanding tert-butyl groups. Although
the C2 symmetrical trans-cyclohexane-1,2-diamine scaffold in
enantiopure form has served as a powerful stereodirecting group
for creating chiral molecular architechtures,20 X-ray structural
studies by Jacobsen et al. show that salen metal complexes from
1 usually adopt an almost planar structure and the tert-butyl
2
means of X-ray crystallography, circular dichroism, and H
NMR (nuclear magnetic resonance) spectroscopy, with the help
of quantum-chemical calculations. Quite interestingly,
MnIV(salen)(L)2 is shown to adopt a solution conformation of
an increasing chiral distortion in the order L ) Cl < NO3 , N3
< OCH2CF3. The present study clarifies an unexpected new
finding that the external axial ligands (L) play a critical role in
amplifying the chirality in trans-cyclohexane-1,2-diamine in
MnIV(salen)(L)2 to facilitate the formation of a chirally distorted
conformation, possibly a stepped conformation.
Results
Synthesis and Characterization of MnIV(salen)(L)2 from
Jacobsen’s Ligand and a Salen Ligand with Less Steric
Bulk. As summarized in Scheme 1, we prepared new
MnIV(salen)(L)2 from (R,R)-1, which are denoted as 1-L (where
L ) Cl, NO3, N3,23 and OCH2CF3). We also prepared
MnIV(salen)(N3)2 from (R,R)-N,N′-bis(5-methylsalicylidene)-1,2-
cyclohexanediamine ((R,R)-2, Chart 1) with less steric bulk,
which is denoted as 2-N3. The stream of O3 and O2 prepared
by UV-irradiation to the O2 gas was passed through the CH2Cl2
(13) Akine, S.; Taniguchi, T.; Nabeshima, T. J. Am. Chem. Soc. 2006, 128,
15765–15774.
(14) Heo, J.; Jeon, Y.-M.; Mirkin, C. A. J. Am. Chem. Soc. 2007, 129,
7712–7713.
(15) Yoon, T. P.; Jacobsen, E. N. Science 2003, 299, 1691–1693.
(16) Bobb, R.; Alhakimi, G.; Studnicki, L.; Lough, A.; Chin, J. J. Am.
Chem. Soc. 2002, 124, 4544–4545.
(17) (a) Fallis, I. A.; Murphy, D. M.; Willock, D. J.; Tucker, R. J.; Farley,
R. D.; Jenkins, R.; Strevens, R. R. J. Am. Chem. Soc. 2004, 126,
15660–15661. (b) Murphy, D. M.; Fallis, I. A.; Willock, D. J.; Landon,
J.; Carter, E.; Vinck, E. Angew. Chem., Int. Ed. 2008, 47, 1414–1416.
(18) Dong, Z.; Yap, G. P. A.; Fox, J. M. J. Am. Chem. Soc. 2007, 129,
11850–11853.
(21) (a) Pospisil, P. J.; Carsten, D. H.; Jacobsen, E. N. Chem.sEur. J.
1996, 2, 974–980. (b) Hansen, K. B.; Leighton, J. L.; Jacobsen, E. N.
J. Am. Chem. Soc. 1996, 118, 10924–10925. (c) Ready, J. M.;
Jacobsen, E. N. J. Am. Chem. Soc. 1999, 121, 6086–6087.
(22) (a) Kurahashi, T.; Kobayashi, Y.; Nagatomo, S.; Tosha, T.; Kitagawa,
T.; Fujii, H. Inorg. Chem. 2005, 44, 8156–8166. (b) Kurahashi, T.;
Kikuchi, A.; Tosha, T.; Shiro, Y.; Kitagawa, T.; Fujii, H. Inorg. Chem.
2008, 47, 1674–1686.
(19) Kemper, S.; Hroba´rik, P.; Kaupp, M.; Schlo¨rer, N. E. J. Am. Chem.
Soc. 2009, 131, 4172–4173.
(20) Bennani, Y. L.; Hanessian, S. Chem. ReV. 1997, 97, 3161–3195.
(23) Kurahashi, T.; Fujii, H. Inorg. Chem. 2008, 47, 7556–7567.
9
J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009 12395

A R T I C L E S
Kurahashi et al.
Scheme 1
solution of MnIII(salen)(Cl) from (R,R)-1 at 193 K to give 1-Cl
as a green powdery sample. 1-Cl dissolved in CH2Cl2 is stable
at room temperature for hours. 1-NO3 was prepared as a green
solid by the reaction of 1-Cl with 10 equiv of AgNO3 in CH2Cl2
for 3 h at 233 K. 1-NO3 in CH2Cl2 decomposes at room
temperature (t1/2 ≈ 20 min). 1-OCH2CF3 was prepared as a
brown solid by oxidation of MnIII(salen)(Cl) from (R,R)-1 with
1.0 equiv of m-CPBA (m-chloroperoxybenzoic acid) and a
subsequent ligand exchange by addition of 3 equiv of
CF3CH2ONa. 1-OCH2CF3 in CH2Cl2 is stable at room temper-
ature for hours. Reaction of MnIII(salen)(Cl) from (R,R)-2 at
233 K with 1 equiv of m-CPBA and a subsequent ligand
exchange by addition of 10 equiv of NaN3 gave 2-N3 as a green
solid. Repeated precipitation in CH2Cl2-pentane at 233 K three
times gave an analytically pure sample. 2-N3 in CH2Cl2 slowly
decomposes at room temperature (t1/2 ≈ 2 h).
Figure 1. X-band EPR spectra of (a) 1-OCH2CF3, (b) 1-NO3, and (c) 1-Cl
in frozen CH2Cl2-toluene (7:3) at 4 K. Signals that arise from ms ) (1/2
3
and (3/2 doublets of S )
/ are written in red and blue, respectively.
2
To confirm that 1-L and 2-N3 have a MnIV center, EPR
(electron paramagnetic resonance) spectra are measured at 4
K. The EPR spectrum of 1-OCH2CF3 at 4 K in Figure 1a
exhibits signals at g ) 3.56 and 2.00, which are typical of
Magnetic parameters: (a) E/D ≈ 0, (b) E/D ≈ 0.25, (c) E/D not determined.
Conditions: microwave frequency, 9.56 GHz; microwave power, 2.012 mW;
modulation amplitude, 7 G; time constant, 163.84 ms; conversion time,
163.84 ms.
22b,23,24
3
an S ) /2 spin system (E/D ≈ 0) from d3 MnIV.
The
and the weak signal at g ) 5.63 arises from the ms ) (3/2
doublets of an S ) 3/2 spin system. 1-Cl exhibits intense EPR
signals shown in Figure 1c, which is consistent with the
assignment as a mononuclear MnIV complex. The EPR
spectrum of 1-Cl contains multiple sets of signals, which may
indicate multiple MnIV species with different rhombicity in
solution or could be alternatively interpreted as arising from
the ms ) (1/2 and ms ) (3/2 doublets from a single MnIV
signal at g ) 2.00 displays a six-line hyperfine splitting due
to the I ) /2 55Mn nucleus. 1-NO3 shows one set of EPR
5
signals at g ) 5.04, 2.96, and 1.83 along with a weak signal
3
at g ) 5.63 (Figure 1b), which are also typical of an S ) /2
spin system (E/D ≈ 0.25) from d3 MnIV.22b,23,24 The signals
at g ) 5.04, 2.96, and 1.83 arise from the ms ) (1/2 doublets,
(24) (a) Bryliakov, K. P.; Babushkin, D. E.; Talsi, E. P. J. Mol. Catal. A
2000, 158, 19–35. (b) Adam, W.; Mock-Knoblauch, C.; Saha-Mo¨ller,
C. R.; Herderich, M. J. Am. Chem. Soc. 2000, 122, 9685–9691. (c)
Campbell, K. A.; Lashley, M. R.; Wyatt, J. K.; Nantz, M. H.; Britt,
R. D. J. Am. Chem. Soc. 2001, 123, 5710–5719.
species. 2-N3 shows EPR signals that are typical of an S )
3
/
spin system (E/D ≈ 0) from d3 MnIV (Figure S1,
2
Supporting Information).
9
12396 J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009

Role of External Axial Ligands in Chirality Amplification
A R T I C L E S
Table 1. Crystallographic Data for 1-Cl, 1-OCH2CF3, and 2-N3
1-Cl
1-OCH2CF3
2-N3
formula
Mr
C76H112Cl12Mn2N4O4 C44H62F6MnN4O4
C22H24MnN8O2
487.42
93
1681.06
93
879.92
93
T/K
cryst size/mm3 0.20 × 0.15 × 0.15 0.30 × 0.25 × 0.05 0.30 × 0.20 × 0.05
cryst syst
space group
cryst color
λ/Å
triclinic
P1 (No. 1)
dark red
0.710 70
0.746
monoclinic
P21 (No. 4)
black
0.710 70
0.355
orthorhombic
P212121 (No. 19)
brown
0.710 70
0.616
8.5312(4)
13.4133(7)
19.9001(10)
90.0000
90.0000
90.0000
2277.20(20)
4
µ/mm-1
a/Å
12.628(4)
12.917(5)
14.498(6)
82.04(3)
73.07(2)
65.24(2)
2053.9(13)
1
9.671(2)
23.135(5)
10.267(2)
90.0000(14)
95.5892(15)
90.0000(14)
2286.1(8)
2
b/Å
c/Å
R/deg
ꢀ/deg
γ/deg
V/Å3
Z value
Dcalc/g cm-3 1.359
1.275
1.422
GOF
1.093
0.897
0.0499
0.1418
0.057(15)
1.095
0.0381
0.0946
0.027(15)
a
0.0842
0.2474
0.04(5)
R1
Rw
b
Flack
a R1 ) Σ|Fo| - |Fc|/Σ|Fo|. b Rw ) [Σw(Fo - Fc2)2/Σw(Fo )2]1/2
.
2
2
Table 2. Selected Structural Parameters for 1-Cl, 1-OCH2CF3, and
2-N3
1-Cla
1-Clb
1-OCH2CF3
2-N3
R/deg
ꢀ/deg
191.4
173.6
154.5
170.3
0.017
154.7
158.7
0.009
158.5
160.9
0.005
Mn-N2O2 plane/Å 0.023
Mn-O1/Å
Mn-O2/Å
Mn-N1/Å
Mn-N2/Å
Mn-Lc/Å
1.867(10) 1.847(10) 1.873(2)
1.844(8)
1.992(8)
1.971(11) 1.988(11) 1.986(2)
2.300(2)
2.263(2)
1.321(16) 1.334(15) 1.329(3)
1.316(18) 1.325(16) 1.329(3)
1.292(19) 1.297(17) 1.281(4)
1.277(19) 1.295(18) 1.286(4)
1.415(19) 1.458(17) 1.445(4)
1.453(17) 1.404(14) 1.440(4)
1.8546(16)
1.8522(16)
1.9775(19)
1.9912(17)
1.992(2)
1.9836(19)
1.329(2)
1.334(2)
1.286(2)
1.286(2)
1.446(3)
1.445(3)
1.857(8)
1.965(8)
1.873(2)
1.986(2)
2.270(2)
2.283(2)
1.880(2)
1.866(2)
Mn-Lc/Å
O1-C2/Å
O2-C2′/Å
N1-C7/Å
N2-C7′/Å
C1-C7/Å
C1′-C7′/Å
O1-Mn-N2/deg
O2-Mn-N1/deg
Figure 2. X-ray crystal structures of (a) 1-Cl, (b) 1-OCH2CF3, and (c)
2-N3. Thermal ellipsoids represent the 50% probability surfaces. Hydrogen
atoms and a disordered tert-butyl group in 1-OCH2CF3 are omitted for the
sake of clarity. The angles R and ꢀ are defined as those between the planes
of the salicylidene rings and the least-squares plane of the N2O2 of the
salen ligand.
174.6(3)
172.8(4)
172.0(3)
173.6(4)
173.56(9) 175.07(7)
173.43(9) 175.47(7)
a Shown in Figure 2a. b Shown in Figure S3 (Supporting Infor-
mation). c L denotes Cl, OCH2CF3, or N3.
X-ray Crystal Structures of 1-Cl, 1-OCH2CF3, and 2-N3.
X-ray crystal structures of 1-Cl, 1-OCH2CF3, and 2-N3 were
successfully determined (Figure 2), although we could not obtain
a crystal of 1-NO3 that is suitable for the X-ray crystallography.
Table 1 summarizes the crystallographic data. The asymmetric
unit of 1-Cl contains two mononuclear complexes. One is shown
in Figure 2a, and the other is included in the Supporting
Information (Figure S3). The asymmetric unit of 1-OCH2CF3
and that of 2-N3 contain one mononuclear complex, which is
shown in Figure 2b and 2c, respectively. As shown in the
packing diagrams (Figure S2, Supporting Information), the
intermolecular Mn-Mn distances in the unit cell of 1-Cl and
2-N3 is 7.43 and 7.61 Å, respectively, while that of 1-OCH2CF3
is longer (16.26 Å), due to the steric bulk of the CF3CH2O
groups incorporated to the axial coordination sites. Table 2
compiles selected structural parameters. Table 2 also includes
parameters R and ꢀ, which are defined as angles between the
planes of the salicylidene rings and the least-squares plane of
the N2O2 of the salen ligand,2f as shown in Figure 2.
1-OCH2CF3 adopts a stepped conformation,25 which is similar
to that of 1-N3.23 2-N3 with less steric bulk also adopts a stepped
conformation, clearly indicating sterically demanding tert-butyl
groups in (R,R)-1 are not an indispensable prerequisite for the
formation of a chirally distorted stepped conformation. Notably,
the conformation of 1-Cl is much less distorted. As summarized
in Table 2, 1-OCH2CF3 and 2-N3 in a stepped conformation
bears shorter Mn-OCH2CF3 bonds (1.866 and 1.880 Å) and
Mn-N3 bonds (1.992 and 1.984 Å), while less-distorted 1-Cl
bears considerably longer Mn-Cl bonds (2.263-2.300 Å). But
the other structural parameters, including Mn-Osalen and
Mn-Nsalen bonds, are quite similar for all the MnIV(salen)
complexes.
Solution Studies of 1-L. Solution conformations of the
MnIV(salen) complexes are investigated by means of CD
(circular dichroism) spectroscopy. For the purpose of assigning
the CD bands from the ligand moiety of Mn(salen), we first
measured the UV-vis and CD spectra of the salen ligand
X-ray crystal structures in Figure 2 show all the MnIV(salen)
complexes have a planar N2O2 equatorial coordination sphere
with external ligands that occupy axial coordination sites.
(25) For proposed conformations of Mn(salen), see ref 2f.
9
J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009 12397

A R T I C L E S
Kurahashi et al.
Figure 5 shows CD spectra of 1-L (L ) Cl, NO3, N3,23 and
OCH2CF3) under exactly the same measurement conditions. The
intensity of CD in the region 200-300 nm, which is derived
from the salen ligand, is considerably increased in the order L
) Cl < NO3 , N3 < OCH2CF3. Because the UV-vis
absorptions in this region do not differ in intensity for all the
(R,R)-MnIV(salen) complexes, the CD change in this region is
solely ascribed to a conformational difference, rather than an
overlap with additional CD upon substitution of the axial
ligands. CD spectroscopy thus indicates 1-OCH2CF3 and 1-N3
in solution adopt a conformation in which the salen ligand is
more chirally distorted, as compared to 1-Cl and 1-NO3. This
is nicely in accord with the observation in the solid state, which
shows both 1-OCH2CF3 and 1-N3 adopt a chirally distorted
stepped conformation, in contrast to a relatively planar confor-
mation for 1-Cl. 2-N3 with less steric bulk also exhibits a much
more intense CD as compared with the starting MnIII(salen)(Cl)
in the region 200-300 nm (Figure S5, Supporting Information),
indicating 2-N3, which is crystallized as a stepped conformation,
also adopts a chirally distorted conformation in solution. Further
insight into the chiroptical properties was obtained from
variable-temperature CD measurements for 1-OCH2CF3 and
1-Cl (Figure 6). As the temperature of the sample solution is
decreased from 293 to 233 K, all the CD bands of 1-OCH2CF3
are increased. The same is true for 1-N3 as reported previously.23
This is indicative of an equilibrium between a chirally distorted
conformation at a lower energy level and a less chirally distorted
conformation at a higher energy level for 1-OCH2CF3 and 1-N3.
In contrast, in the case of 1-Cl, CD in the region 300-500 nm
is increased upon cooling the sample solution, but CD in the
region 200-300 nm, which reflects conformational changes of
the salen ligand, is not altered. The absence of the temperature-
dependent CD change in the region 200-300 nm for 1-Cl may
suggest every conformation that 1-Cl may adopt is at a similar
energy level.
Figure 3. (a) UV-vis and (b) CD spectra of (R,R)-1 (red line) and (R,R)-3
(blue line) in CH2Cl2 (0.6 mM) at room temperature.
(R,R)-1 and the tetrahydrosalen ligand (R,R)-3 (Chart 1). (R,R)-1
exhibits three UV-vis absorptions and the corresponding three
CD bands in the region 200-400 nm (Figure 3). It is thus
expected that MnIV(salen) complexes may also exhibit ligand-
derived UV-vis absorptions and CD bands in this region. In
contrast, (R,R)-3 without the azomethine moiety exhibits only
two UV-vis absorptions suggesting one of three UV-vis
absorptions in (R,R)-1 might be assigned as originating from
the azomethine moiety, while the other two might originate from
the phenolates. It is interesting to note that (R,R)-3 shows
negligibly weak CD as compared to (R,R)-1,26 suggesting the
azomethine moiety plays a role in inducing relatively strong
CD in (R,R)-1.
Solution conformations of 1-L were also investigated with
NMR spectroscopy. Due to the paramagnetic high-spin d3 MnIV
center, MnIV(salen) complexes show multiple 1H NMR signals
at largely shifted positions.22b,23,24a,27 As shown in Figure S6
(Supporting Information), 1-L commonly exhibits three down-
field-shifted signals and two upfield-shifted signals, except for
1-NO3 that exhibits an additional signal at -40.4 ppm. For the
purpose of assignment of these signals, we employed 2H NMR
for 2H-labeled salen complexes. A 2H-labeled Jacobsen’s salen
ligand, (R,R)-1-d, was prepared from phenol-d6. 1H and 2H NMR
indicate that 2H atoms are selectively incorporated to the
phenolate rings (80% D) as well as the tert-butyl groups (7%
D) in (R,R)-1-d (Figure S7, Supporting Information). As clearly
seen from 2H NMR spectra in Figure 7, 1-L-d commonly
exhibits one downfield-shifted and one upfield-shifted signal,
which are unambiguously assigned as arising from the 4/4′H
and 6/6′H of the phenolate rings (for numbering of the aromatic
rings, see Scheme 1).28 Then, the signals within the diamagnetic
region are assigned as arising from the tert-butyl groups. Since
the paramagnetic shifts of the tert-butyl groups are mainly
induced by dipolar magnetic interaction, a broad signal that is
more shifted could be assigned as arising from the tert-butyl
groups at 3/3′ positions close to the paramagnetic Mn center,
Figure 4 shows UV-vis spectra of 1-L (L ) Cl, NO3, N3,23
and OCH2CF3). All the MnIV(salen) complexes commonly
exhibit two UV-vis absorptions of similar intensity in the region
200-300 nm. These are most probably derived from the salen
ligand, in comparison with the UV-vis spectrum of (R,R)-1
shown in Figure 3. Another remarkable feature is the absorption
at >600 nm, which is shifted by substitution of the axial ligand
L. In the case of 1-OCH2CF3, the corresponding absorption may
be shifted to the region of <600 nm. These absorptions are most
probably assigned as the phenolate-to-MnIV charge transfer band.
1-N3 exhibits an absorption at 439 nm, which was previously
assigned as the N3-to-MnIV charge transfer band.23
(27) Bonadies, J. A.; Maroney, M. J.; Pecoraro, V. L. Inorg. Chem. 1989,
28, 2044–2051.
1
(26) Weak CD for (R,R)-3 is not due to racemization during the preparation,
because (S,S)-3 shows CD of the opposite sign to (R,R)-3 (Figure S4,
Supporting Information).
(28) We had incorrectly assigned the H NMR signals from 1-N3 in our
previous report,23 which leads to an incorrect interpretation that two
phenolate rings in 1-N3 are not equivalent in solution.
9
12398 J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009

Role of External Axial Ligands in Chirality Amplification
A R T I C L E S
Figure 4. UV-vis spectra of (a) 1-Cl, (b) 1-NO3, (c) 1-N3, and (d) 1-OCH2CF3 in CH2Cl2 (0.3 mM) at 233 K.
Figure 5. Circular dichroism spectra of (a) 1-Cl, (b) 1-NO3, (c) 1-N3, and (d) 1-OCH2CF3 in CH2Cl2 (0.3 mM) at 233 K.
1
while the other sharp signal could be assigned as arising from
the tert-butyl groups at 5/5′ positions, which are more distant
from the Mn center. H NMR signals at approximately -40
ppm and <20 ppm could be tentatively assigned as arising from
9
J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009 12399

A R T I C L E S
Kurahashi et al.
the azomethine proton and from the cyclohexane protons,
respectively, based on the spin polarization mechanism from
MnIV center.
2
As shown in Figure 7, H NMR signals for the 4/4′H and
6/6′H of the phenolate rings in 1-L-d appear outside the normal
diamagnetic region, due to the paramagnetic effect from the
Mn center through the π-system of the salen ligand. Paramag-
netic shifts of the 4/4′H and 6/6′H of the phenolate rings in
1-L-d increase in the order L ) OCH2CF3 < N3 , NO3 ≈ Cl.
This might be partly due to a difference in interaction between
the Mn d-orbital and the phenolate π-orbital; the paramagnetic
shift would be decreased with the distortion of the phenolate
rings from the planar conformation. Another point to be noted
is a 2H NMR signal from the tert-butyl groups at 5/5′ positions,
which shows increasing paramagnetic shifts in the order L )
OCH2CF3 < N3 < Cl < NO3. Paramagnetic shifts of the tert-
butyl groups via a dipolar interaction would be decreased with
the distortion of the phenolate rings from the planar conforma-
tion, depending on the geometric relationship between the
paramagnetic metal center and the shifted nucleus, as schemati-
cally shown in Figure S8.29 The smaller shifts for 1-OCH2CF3
and 1-N3 as compared with 1-NO3 and 1-Cl may imply
1-OCH2CF3 and 1-N3 may adopt a stepped conformation in
solution as well as in the solid state. It is also interesting to
note that the extent of signal broadening for one of two tert-
butyl groups differs depending on the external axial ligand,
suggesting a different conformational equilibrium. Although the
X-ray crystal structure shows 1-Cl adopts a planar conformation
with unequivalent phenolate rings, the signals for the 4/4′H and
6/6′H of the phenolate rings in 1-Cl-d as well as 1-OCH2CF3-d
and 1-N3-d in a C2-symmetric stepped conformation is not
resolved even at 233 K. This is possibly because fluctuation of
the phenolate rings occurs at a rate comparable to the NMR
time scale. This is in clear contrast with Fox’s Ni(salen) system
bearing a large substituent at the 3/3′ positions, which was
reported to show a chemical exchange with coalescence at 253
K.18
Figure 6. Circular dichroism spectra of (a) 1-Cl and (b) 1-OCH2CF3 in
CH2Cl2 (0.3 mM) at 293 (black line), 273 (green line), 253 (blue line), and
233 K (red line).
Quantum-Chemical Study of Circular Dichroism from a
Salen Complex. To gain an understanding of circular dichro-
ism and solution conformations of a salen complex, quantum-
chemical calculations are carried out using the SAC-CI
(symmetry adapted cluster-configuration interaction)
method.30 First, we carried out a quantum-chemical calcula-
tion for the metal-free salen ligand that is composed of (R,R)-
trans-cyclohexane-1,2-diamine and unsubstituted salicylidene
rings and compared the theoretical result with the experi-
mental observation to confirm the validity of the present
calculation. A DFT (density functional theory) calculation
at the B3LYP/6-311+G(d) level of theory gives the optimized
geometry (Figure S9, Supporting Information), for which
SAC-CI calculations are carried out. Table 3 summarizes the
calculated circular dichroism for the metal-free salen ligand.
Molecular orbitals that are related to the transition are shown
in Table S1 and Figure S10 (Supporting Information). The
calculation indicates three sets of CD-active absorptions,
which are fully consistent with the experimental observation
in Figure 3. Furthermore, the present quantum-chemical
calculation well reproduces the experimental wavelengths
Figure 7. 2H NMR (76.65 MHz) spectra of (a) 1-Cl-d, (b) 1-NO3-d, (c)
1-N3-d, and (d) 1-OCH2CF3-d in CH2Cl2 (40 mM) at 233 K. The signal
designated with an asterisk comes from residual CHDCl2 and is referenced
to 5.32 ppm.
(29) Ming, L.-J. In Physical Methods in Bioinorganic Chemistry, Spec-
troscopy and Magnetism; Que, L. , Jr., Ed.; University Science Books:
CA, 2000; p 375.
(30) (a) Nakatsuji, H. Chem. Phys. Lett. 1978, 59, 362–364. (b) Nakatsuji,
H. Chem. Phys. Lett. 1979, 67, 329–333.
9
12400 J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009

Role of External Axial Ligands in Chirality Amplification
A R T I C L E S
Table 3. Calculated Circular Dichroism for the Salen Ligand with
SAC-CI Calculationsa
symmetryb
wavelength/nm
rotatory strength
1B
1A
2A
2B
3B
3A
303.0
301.0
252.8
251.5
232.1
224.2
-565.2
376.2
45.7
-41.6
-217.1
250.8
a An optimized structure of the salen ligand with a DFT calculation,
for which SAC-CI calculations are carried out, is shown in Figure S9.
b Assignment of the transitions that are responsible for the CD is shown
in Table S1.
Figure 8. Planar and stepped conformations of the NiII(salen) model which
is created from the X-ray structure of 1-Cl and 1-OCH2CF3, respectively.
These structures are utilized for quantum chemical calculations with the
SAC-CI method. The arrows indicate the directions of the twist of the
phenolates to calculate CD for conformations with twisted phenolates.
Figure 9. Calculated CD spectra for (a) the planar conformation (black
line) and (b) the stepped conformation (black line) of the Ni(salen) model
shown in Figure 8. Quantum-chemical calculations are also carried out for
the conformations in which the phenolates are twisted to a lesser (blue line)
and greater extent (red line) along one of the molecular vibration modes,
as shown by the red arrows in Figure 8.
with an error of less than ∼30 nm. We also carried out a
quantum-chemical calculation for hypothetical structures in
which two benzene rings are placed in exactly the same
manner as the stepped conformation from 1-OCH2CF3 and
the planar conformation from 1-Cl (Figure S11, Supporting
Information), to consider exactly what gives rise to CD from
the metal-free salen ligand. Irrespective of arrangements of
the benzene rings, a theoretical calculation gives no CD-
active absorption, indicative of an important role of the
azomethine moiety for relatively strong CD from the salen
ligand. This theoretical result well accounts for the experi-
mental observation that the tetrahydrosalen ligand, (R,R)-3,
without the azomethine moiety exhibits only weak CD as
compared to the salen ligand, (R,R)-1 (Figure 3).
We then carried out quantum-chemical calculations for
circular dichroism from planar and stepped conformation models
in which Mn is replaced by Ni for ease of computation (Figure
8). As shown in Figure 9, the stepped conformation model
created from the X-ray structure of 1-OCH2CF3 exhibits more
intense CD than the planar conformation model that is created
from 1-Cl in the range 180-300 nm (depicted in a black line).
These CD bands are attributed to transitions associated with
the salen ligand. We further consider a possibility that the
calculated CD intensity may reflect a twist of the phenolates,
rather than the difference between the planar and stepped
conformations. Then, CD spectra are also calculated for the
conformations in which the phenolates are twisted along one
of the molecular vibration modes as shown by red arrows in
Figure 8. Even though the planar conformation model is altered
from the starting conformation (black line) to those with twisted
phenolates to a lesser (blue line) and greater extent (red line),
the CD from the planar conformation model remains weaker
than the CD from the stepped conformation model (Figure 9).
Therefore, it is securely indicated that a metal salen complex
in a stepped conformation exhibits substantially more intense
CD than a metal salen complex in a planar conformation.
Discussion
Biologically important chiral structural motifs such as helices
of DNA and proteins are built up from asymmetric carbon atoms
of point chirality. Therefore, synthetic molecules with such
higher-order chirality has been of great interest and extensively
studied from both fundamental and practical viewpoints.31 It
has been shown that noncovalent interactions such as hydrogen
bonding play an important role to create and maintain such
higher-order chirality. In the case of Jacobsen’s salen complexes,
the formation of a chiral conformation, if any, is similarly
directed by the asymmetric carbon atoms in trans-cyclohexane-
1,2-diamine as the only chiral unit. However, Jacobsen’s salen
complexes apparently lack such noncovalent interactions that
could support a chiral conformation, according to the previous
design criteria. In addition, Fox et al. rigorously investigated
9
J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009 12401

A R T I C L E S
Kurahashi et al.
solid-state and solution conformations of their well-designed
NiII(salen) complexes and unambiguously showed that trans-
cyclohexane-1,2-diamine is only a weak director to induce a
chiral conformation in a metal salen complex.18 Thus, there has
been a good reason to assume that Jacobsen’s salen complex
may not adopt a distinct chiral conformation in solution,
although exquisite enantioselectivity by Jacobsen’s salen com-
plexes has been frequently ascribed to a chiral conformation of
some kind or another.2f
We show herein that the solution of 1-L exhibits CD of
increasing intensity in the order L ) Cl < NO3 , N3 < OCH2CF3
in the range 200-300 nm (Figure 5). Because UV-vis spectra
in this region do not differ to any significant extent (Figure 4),
the CD bands of different intensity cannot be a result of an
overlap with additional CD upon substitution of the axial ligands
L but are securely attributed to a conformational difference. It
is thus indicated 1-L adopts a solution conformation of an
increasing chiral distortion depending on the axial ligands in
this order. In accordance with the observation from CD
spectroscopy, 2H NMR spectroscopy also indicates a difference
of solution conformations similarly depending on the external
axial ligands. The most remarkable structural difference in 1-L
is a Mn-L bond length, which becomes shorter in parallel with
the degree of a chiral distortion in the order L ) Cl (2.263-2.300
Å) > N3 (1.958-2.006 Å)23 > OCH2CF3 (1.866-1.880 Å), while
the other structural parameters such as Mn-Osalen/Nsalen bond
lengths are not altered. The order of a chiral distortion in 1-L
is also in parallel with the pKa values of HL (HCl, -7;32 HNO3,
-1.4;32 HN3, 4.72;33 CF3CH2OH, 11.434), which reflect the
ability of L- to dissociate in solution. These observations
indicate a rather unexpected finding that the formation of a
chirally distorted conformation is significantly facilitated by
external axial ligands that could coordinate to the Mn center
from both sides along the axial direction with a shorter Mn-L
bond length and a lower degree of dissociation in solution. In
light of the present finding, a failure to form a chirally distorted
conformation in five-coordinate and four-coordinate metal salen
complexes such as MnIII(salen)(L)23 and NiII(salen)18 as reported
previously might be ascribed to the absence of such external
axial ligands. Although we could not obtain definitive evidence
for solution conformations, 1-OCH2CF3, 1-N3,23 and 2-N3
exhibiting more intense CD are crystallized as a stepped
conformation without exception, while the crystal of 1-Cl
exhibiting relatively weak CD contains a planar conformation.
In addition, quantum-chemical calculations indicate that a
stepped conformation exhibits more intense CD than a planar
conformation. Therefore, a chirally distorted conformation that
is formed in the solution of 1-L more preferentially in the order
Figure 10. A model that is reproduced from the X-ray crystal structure of
1-OCH2CF3.
L ) Cl < NO3 , N3 < OCH2CF3 would be most probably a
stepped conformation.25
One possible explanation for the role of axial ligands is
steric repulsion between the trans-cyclohexane-1,2-diamine
moiety as the only chiral center and the axial ligands coming
close to the Mn center, which bring the axial ligands to one
asymmetric direction and then puts down the salicylidene
rings. As shown in Figure 10, the methine hydrogen atom of
trans-cyclohexane-1,2-diamine, which is closest to the axial
ligands, is considered the most important steric factor. But
the methine hydrogen atom is expected to rather bring the
OCH2CF3 or N3 group to the direction that disturbs the
formation of a stepped conformation, and hence the pos-
sibility of steric repulsion is ruled out. It is also interesting
to note that 2-N3 bearing much less sterically demanding
salicylidene rings similarly adopts a chirally distorted
conformation in solution, which means steric interaction
between the external axial ligand and the tert-butyl groups
is not a major factor for chiral distortion. Our hypothesis
from the present work is that tight binding of the external
axial ligands from both sides along the axial direction
confines the movement of the Mn ion along this direction,
which may serve to enhance the energy difference among
conformers. In the case of tightly bound ligands such as
OCH2CF3 and N3, the stability of the stepped conformation
might be significantly enhanced, leading to an increased
distribution of this conformer relative to the others in solution,
which results in the appearance of intense circular dichroism
and also crystallization exclusively as a stepped conformer.
In the case of weakly bound ligands such as NO3 and Cl,
the energy difference among conformers remains very small,
and then the solution of salen complexes bearing such axial
ligands contains an almost equimolar mixture of multiple
conformers, which could account for the circular dichroism
of low intensity even at lower temperatures and broadened
2H NMR signals.
(31) For example, see: (a) Mizutani, T.; Yagi, S.; Morinaga, T.; Nomura,
T.; Takagishi, T.; Kitagawa, S.; Ogoshi, H. J. Am. Chem. Soc. 1999,
121, 754–759. (b) Inai, Y.; Tagawa, K.; Takasu, A.; Hirabayashi, T.;
Oshikawa, T.; Yamashita, M. J. Am. Chem. Soc. 2000, 122, 11731–
11732. (c) Kurta´n, T.; Nesnas, N.; Koehn, F. E.; Li, Y.-Q.; Nakanishi,
K.; Berova, N. J. Am. Chem. Soc. 2001, 123, 5974–5982. (d) Kano,
K.; Hasegawa, H. J. Am. Chem. Soc. 2001, 123, 10616–10627. (e)
Lintuluoto, J. M.; Borovkov, V. V.; Inoue, Y. J. Am. Chem. Soc. 2002,
124, 13676–13677. (f) Maeda, K.; Morino, K.; Okamoto, Y.; Sato,
T.; Yashima, E. J. Am. Chem. Soc. 2004, 126, 4329–4342. (g) Yu, J.;
RajanBabu, T. V.; Parquette, J. R. J. Am. Chem. Soc. 2008, 130, 7845–
7847.
Conclusion
We herein prepare a series of MnIV(salen)(L)2 complexes
bearing different external axial ligands (L) from Jacobsen’s salen
ligand and clarify a rather unexpected finding that the external
axial ligands play a pivotal role in effectively transmitting the
chirality at the bridging diamine moiety to the phenolate moiety.
(32) Bell, R. P. The Proton in Chemistry, 2nd ed.; Cornel University Press:
Ithaca, NY, 1973.
(33) Bjerrum, J.; Schwarzenbach, G.; Sille´n, L. G. Stability Constants of
Metal-Ion Complexes; Chemical Society: London, 1958.
(34) Ballinger, P.; Long, F. A. J. Am. Chem. Soc. 1959, 81, 1050–1053.
9
12402 J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009

Role of External Axial Ligands in Chirality Amplification
A R T I C L E S
using the crystal structure visualization and exploration program,
Mercury (ver. 2.2), provided by the Cambridge Crystallographic
Data Center.
The present new finding would provide valuable insight to create
metal salen complexes with unique stereochemical properties.
Quantum-Chemical Calculations. Quantum-chemical calcula-
tions were performed using the Gaussian03 program package.41
The optimized structures were calculated using the B3LYP/6-
311+G(d) level of theory. Circular dichroism was calculated with
the SAC-CI method.
Experimental Section
Caution! All the operations for preparing 1-Cl, including
filtration of the product, should be carried out in a fume hood. A
volatile substance, probably Cl2, which is generated during the
preparation, causes serious damage to eye, nose, and mouth.
Instrumentation. UV-vis spectra were recorded in spectro-
photometric grade CH2Cl2 in a quartz cell (l ) 0.1 cm) on an
Agilent 8453 spectrometer (Agilent Technologies) equipped with
a USP-203 low-temperature chamber (UNISOKU). CD spectra were
recorded in a quartz cell (l ) 0.1 cm) on a J-720W spectropola-
rimeter (JASCO) equipped with a USP-203 low-temperature
chamber (UNISOKU). Measurements were done under the follow-
ing conditions: scan rate, 100 nm min-1; bandwidth, 5.0 nm;
response, 4 s; resolution, 1 nm. EPR spectra were recorded for 30
µL of the 10 mM solution in a quartz cell (d ) 5 mm) on an E500
continuous wave X-band spectrometer (Bruker) with an ESR910
helium-flow cryostat (Oxford Instruments). Measurements were
made under the following conditions: microwave frequency, 9.56
GHz; microwave power, 2.012 mW; modulation amplitude, 7 G;
time constant, 163.84 ms; conversion time, 163.84 ms. 500 and
400 MHz NMR spectra were measured on an LA-500 and LA-400
spectrometer (JEOL), respectively. 1H NMR chemical shifts in
CD2Cl2 and CDCl3 were referenced to CHDCl2 (5.32 ppm) and
CHCl3 (7.24 ppm), respectively. 13C NMR chemical shifts in CDCl3
were reported relative to CHCl3 (77.0 ppm). Elemental analyses
were conducted on a CHN corder MT-6 (Yanaco). High-resolution
mass spectra were measured with the JMS-777 V mass spectrometer
(JEOL).
X-ray Crystallography. Measurements were made on a
Rigaku/MSC Mercury CCD diffractometer equipped with graph-
ite monochromated Mo KR radiation (λ ) 0.710 70 Å). Data
were collected at 93 K under a cold nitrogen stream. All crystals
were mounted on a glass fiber using epoxy glue. The images
were processed with the CrystalClear program (ver. 1.3.5).35 The
structures were solved by the direct method using SHELXS-9736,37
and refined by full-matrix least-squares procedures on F2 using
SHELXL-97,36,38 with the CrystalStructure software package
(ver. 3.8.2).39 Anisotropic refinement was applied to all non-
hydrogen atoms. Hydrogen atoms were placed at the calculated
positions and refined with isotropic parameters. Corrections for
Lorentz polarization effects and absorption were performed. The
Flack parameters40 were calculated to confirm the absolute
configuration. In the case of 1-OCH2CF3, one of four tert-butyl
groups was disordered and was treated as such during the
refinement procedure. The Checkcif program gives an A-level
Materials. Anhydrous and spectrophotometric grade CH2Cl2
and other anhydrous solvents were purchased from Kanto or
Wako and were utilized as received. CD2Cl2 and CDCl3 were
purchased from ACROS and were passed though aluminum oxide
just before use. NaH (65% dispersion in mineral oil) was
purchased from Wako. NaBH4 was purchased from Nacalai.
AgNO3 (99.9999%), CF3CH2OH (99.5%), (R,R)-/(S,S)-N,N′-
bis(3,5-di-tert-butylsalicylidene)-1,2-cyclohexanediaminoman-
ganese(III) chlorides, and ReBr(CO)5 were purchased from
Aldrich and were used as received. m-CPBA (m-choloroper-
oxybenzoic acid) was purchased from Nacalai and purified by
washing with phosphate buffer. The purity of m-CPBA was
checked with iodometry. (R,R)-1,2-Cyclohexanediamine and
5-methylsalicylaldehyde were purchased from Tokyo Chemical
Industry and utilized as received. Manganese(II) acetate tet-
rahydrate was purchased from Wako. Phenol-d6 (98%) was
purchased from Cambridge Isotope Laboratories, Inc.
Synthesis of (R,R)-3. To the solution of (R,R)- or (S,S)-1 (1 g,
1.83 mmol) in anhydrous tetrahydrofuran (8 mL) and anhydrous
methanol (8 mL) were added 10 equiv of NaBH4 (691.8 mg, 18.3
mmol) at 273 K. The resulting solution was stirred at room
temperature for 3 h, and then the solvent was removed by rotary
evaporation. The residue was dissolved in CH2Cl2 (100 mL). The
organic layer was washed with distilled water (50 mL × 3) and
then a saturated NaCl aqueous solution (50 mL × 1). The organic
layer was dried over anhydrous MgSO4. The solvent was removed
by rotary evaporation, and the residue was purified by recrystallizing
from hot methanol to give (R,R)- or (S,S)-3 (540 mg, 0.98 mmol)
1
in 54% yield as a colorless crystalline solid: H NMR (400 MHz,
CDCl3) δ 1.2-1.3 (m, 4H), 1.27 (s, 18H), 1.36 (s, 18H), 1.6-1.8
(m, 2H), 2.1-2.3 (m, 2H), 2.4-2.6 (m, 2H), 3.96 (AB, J ) 13.6
Hz, 4H), 6.85 (d, J ) 2.4 Hz, 2H), 7.19 (d, J ) 2.4 Hz, 2H), 10.6
(brs, 2H); 13C NMR (100.4 MHz, CDCl3) δ 24.1, 29.6, 30.7, 31.7,
34.1, 34.8, 50.8, 59.9, 122.3, 123.0, 123.1, 136.0, 140.6, 154.4;
HRMS (FAB) m/z calcd for C36H59N2O2 [M + H]+ 551.4577, found
551.4570. Anal. Calcd for C36H58N2O2: C, 78.49; H, 10.61; N, 5.09.
Found: C, 78.51; H, 10.67; N, 5.09.
Synthesis of 1-Cl. The stream of O3 and O2 prepared by UV-
irradiation to the O2 gas was passed through a solution of
MnIII(salen)(Cl) from (R,R)-1 (200 mg, 0.31 mmol) in anhydrous
CH2Cl2 (7 mL) at 193 K. After 1.5 h, the cold solution was filtered
and the filtrate was stirred at room temperature for 20 min.
Anhydrous pentane (90 mL) was added to the filtrate, and then the
resulting solution was stirred at 233 K for 10 min to give a green
precipitate. This was filtered off, washed with pentane, and dried
in Vacuo. Recrystallization from CH2Cl2 (2 mL) and pentane (50
mL) at 253 K gave analytically pure 1-Cl (82 mg, 0.12 mmol) in
39% yield. The crystal suitable for the X-ray crystallographic
analysis was obtained by crystallizing 1-Cl (10 mg) in CH2Cl2 (0.5
mL) and pentane (10 mL) at 253 K. Anal. Calcd for
C36H52Cl2MnN2O2 ·(H2O)0.2: C, 64.13; H, 7.83; N, 4.15. Found: C,
64.11; H, 7.65; N, 4.25.
j
alert to suggest another space group, P1, for 1-Cl, but this was
confirmed not true because 1-Cl bearing the (R,R)-trans-
cyclohexane-1,2-diamine moiety is incompatible with an inver-
sion center. The Checkcif program points an A-level alert on a
suspected C-H bond for 1-OCH2CF3, but this is an artifact of
the disordered tert-butyl group. The structural parameters
including R, ꢀ, and the Mn-N2O2-plane distance were obtained
(35) (a) CrystalClear 1.3.5 SP2; Rigaku and Molecular Structure Corp.:
The Woodlands, TX, 2004. (b) Pflugrath, J. W. Acta Crystallogr. 1999,
D55, 1718–1725.
(36) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112–122.
(37) Sheldrick, G. M. SHELXS-97, Program for Crystal Structure Solution;
University of Go¨ttingen: Go¨ttingen, Germany, 1997.
(38) Sheldrick, G. M. SHELXL-97, Program for Refinement of Crystal
Structures; University of Go¨ttingen: Go¨ttingen, Germany, 1997.
(39) (a) CrystalStructure 3.8.2: Crystal Structure Analysis Package, version
3.8.2; Rigaku and Rigaku/MSC: The Woodlands, TX, 2007. (b)
Watkin, D. J.; Prout, C. K.; Carruthers, J. R.; Betteridge, P. W.
Crystals, Issue 10; Chemical Crystallography Laboratory: Oxford,
U.K., 1996.
Synthesis of 1-NO3. To the solution of 1-Cl (112.3 mg, 0.167
mmol) in anhydrous CH2Cl2 (4 mL) were added 10 equiv of AgNO3
(284.4 mg, 1.67 mmol) at 233 K. The resulting solution was stirred
for 3 h at 233 K. Then, the cold solution was filtered to remove
silver salts, and the cold filtrate kept at 233 K was passed through
a membrane filter (Cosmonice Filter S, pore size 0.45 µm, diameter
13 mm, Nacalai). Anhydrous pentane (50 mL) was added to the
(41) Frisch, M. J. Gaussian03, revision C.03; Gaussian, Inc.: Wallingford,
(40) Flack, H. D. Acta Crystallogr. 1983, A39, 876–881.
CT, 2004.
9
J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009 12403

A R T I C L E S
Kurahashi et al.
Scheme 2
filtrate at 233 K to give a green precipitate. This was filtered off,
washed with pentane, and dried in Vacuo. Recrystallization from
CH2Cl2 (2 mL) and pentane (20 mL) at 253 K gave analytically
pure 1-NO3 (39.5 mg, 0.055 mmol) in 33% yield. Anal. Calcd for
C36H52MnN4O8 ·(H2O)1.5: C, 57.59; H, 7.38; N, 7.46. Found: C,
57.34; H, 7.02; N, 7.56.
Synthesis of 1-OCH2CF3. To a solution of MnIII(salen)(Cl)
from (R,R)-1 (495.5 mg, 0.78 mmol) in anhydrous CH2Cl2 (20
mL) was added 1.0 equiv of m-CPBA (134.6 mg, 0.78 mmol)
at 233 K. After the mixture was stirred for 30 min, 3 equiv of
NaH (65% dispersion in mineral oil, 86.4 mg, 2.34 mmol)
dissolved in CF3CH2OH (2 mL) were added to the reaction
mixture at 233 K. After the mixture was stirred for 1 h at 233
K, the solvent was removed in Vacuo at 253 K. The resulting
solid was dissolved in CH2Cl2 (10 mL) at room temperature,
and the solution was filtered through a filter paper and then a
membrane filter (Millex-FG, pore size 0.20 µm, diameter 25 mm,
Millipore). The solvent was removed by rotary evaporation. The
residue was dissolved in a small amount of CH2Cl2-pentane
(1:10), and the addition of pentane (∼100 mL) at 233K gave
analytically pure 1-OCH2CF3 (384 mg, 0.48 mmol) in 62% yield
as a reddish precipitate. Crystals suitable for the X-ray crystal-
lographic analysis were obtained by crystallizing 1-OCH2CF3
(10 mg) in CH3CN (4 mL) at 253 K. Anal. Calcd for
C40H56F6MnN2O4: C, 60.22; H, 7.07; N, 3.51. Found: C, 60.04;
H, 7.00; N, 3.63.
FG, pore size 0.20 µm, diameter 25 mm, Millipore). Addition of
anhydrous pentane (10 mL) at 233 K gave a green precipitate. The
product was redissolved in CH2Cl2 (10 mL) at 233 K and was
precipitated by the addition of pentane (8 mL). Precipitation from
CH2Cl2 (10 mL) and pentane (8 mL) was repeated three times. The
precipitate was collected by vacuum filtration and dried in vacuo
at room temperature. The product was obtained as a green powder
in 21% yield (46.6 mg, 0.096 mmol). Anal. Calcd for
C22H24MnN8O2 ·(CH3OH)0.3: C, 53.89; H, 5.11; N, 22.54. Found:
C, 54.30; H, 5.02; N, 22.10.
Synthesis of (R,R)-N,N′-Bis(5-methylsalicylidene)-1,2-cyclo-
hexanediamine. 5-Methylsalicylaldehyde (11.35 g, 83.4 mmol) in
anhydrous EtOH (80 mL) was heated to reflux (∼363 K) until
dissolution was achieved, and then (R,R)-1,2-cyclohexanediamine
(4.76 g, 41.7 mmol) dissolved in anhydrous EtOH (20 mL) was
added. The resulting solution was stirred at reflux (∼363 K) for
1 h, and then the solution was cooled at 277 K to give a yellow
precipitate. The precipitate was collected by vacuum filtration and
dried in vacuo at 373 K. The product was obtained as a yellow
Synthesis of 2,4-Di-tert-butylphenol-d. Preparation of 2H-
labeled Jacobsen’s salen ligand, (R,R)-1-d, is outlined in Scheme
2. 2,4-Di-tert-butylphenol-d was synthesized from commercially
available phenol-d6 according to the procedure of Nishiyama and
Sonoda42 with a slight modification to the amount of t-BuCl to
improve the yield of the desired product. ReBr(CO)5 (51.3 mg, 0.13
mmol) and t-BuCl (6.8 mL, 62.5 mmol) were added to a solution
of phenol-d6 (1.25 g, 12.5 mmol) in 1,2-dichloroethane (23 mL).
The resulting mixture was stirred at 357 K for 30 min, and then
the solvent was removed by rotary evaporation. The residue was
purified by silica flash column chromatography (CH2Cl2/hexane )
2:5) to give 2,4-di-tert-butylphenol-d (1.95 g, 9.4 mmol) in 75%
1
powder in 77% yield (11.24 g, 32.1 mmol). H NMR (400 MHz,
CDCl3) δ 1.4-2.0 (m, 8H), 2.19 (s, 6H), 3.2-3.3 (m, 2H), 6.77
(d, J ) 8.4 Hz, 2H), 6.91 (d, J ) 1.6 Hz, 2H), 7.02 (dd, J ) 1.6,
8.4 Hz, 2H), 8.18 (s, 2H), 13.07 (s, 2H); 13C NMR (100.4 MHz,
CDCl3) δ 20.2, 24.1, 33.1, 72.7, 116.4, 118.3, 127.6, 131.5, 132.8,
158.6, 164.6; HRMS (FAB) m/z calcd for C22H27N2O2 [M + H]+
351.2073, found 351.2065. Anal. Calcd for C22H26N2O2: C, 75.40;
H, 7.48; N, 7.99. Found: C, 75.54; H, 7.57; N, 7.99.
1
Synthesis of (R,R)-N,N′-Bis(5-methylsalicylidene)-1,2-cyclo-
hexanediaminomanganese(III) Chloride. (R,R)-N,N′-Bis(5-me-
thylsalicylidene)-1,2-cyclohexanediamine (5.00 g, 14.3 mmol) in
anhydrous EtOH (40 mL) was heated to reflux (363 K) until
dissolution was achieved, and then manganese acetate tetrahydrate
(10.49 g, 42.8 mmol) was added. The resulting mixture was stirred
at reflux for 1 h, and then the solvent was removed by rotary
evaporation. The residue dissolved in CH2Cl2 (300 mL) was washed
with 1 M HCl (100 mL × 3) and brine (100 mL × 1). After drying
over MgSO4, the solvent was removed by rotary evaporation. To
the residue redissolved in CH3OH was added a 1 M HCl aqueous
solution to give a precipitate. The precipitate was collected by
vacuum filtration and dried in vacuo at 373 K. The product was
obtained as a brown powder in 39% yield (2.46 g, 5.61 mmol).
Anal. Calcd for C22H24ClMnN2O2: C, 60.21; H, 5.51; N, 6.38.
Found: C, 60.18; H, 5.47; N, 6.40.
Synthesis of 2-N3. To a solution of (R,R)-N,N′-bis(5-methyl-
salicylidene)-1,2-cyclohexanediaminomanganese(III) chloride (201.6
mg, 0.459 mmol) dissolved in anhydrous CH2Cl2 (10 mL) and
anhydrous CH3OH (10 mL) was added 1 equiv of m-CPBA (75%,
105.7 mg, 0.459 mmol) at 233 K. After the mixture was stirred for
1 h at 233 K, 10 equiv of NaN3 (298.7 mg, 4.59 mmol) in H2O (1
mL) were added. The resulting solution was stirred at 233 K for
1 h, and then the solvent was removed in vacuo at 253 K. The
residue dissolved in CH2Cl2 (10 mL) at 233 K was passed through
a pad of Celite and subsequently through a membrane filter (Millex-
yield as a viscous oil. H NMR and mass spectrometry indicate
90% of the 2H at the ortho position and 20% of the 2H at the meta
position were replaced with 1H under the reaction conditions (Figure
S12, Supporting Information).
Synthesis of 3,5-Di-tert-butyl-2-hydroxybenzaldehyde-d. The
procedure of Jacobsen and Gao43 was employed except for the
purification step. 2,4-Di-tert-butylphenol-d (1.71 g, 8.2 mmol) and
hexamethylenetetramine (HMT, 2.30 g, 16.4 mmol) in glacial acetic
acid (4 mL) were heated at 403 K for 3 h. 33% (w/w) aqueous
H2SO4 (4 mL) was added to the mixture at 373 K, and the mixture
was stirred at this temperature for 1 h. The resulting mixture was
allowed to cool to room temperature and extracted with diethyl
ether (150 mL). The extract was washed with brine (100 mL × 3)
and then dried over MgSO4. The solvent was removed by rotary
evaporation. The crude product was purified by silica flash column
chromatography (CH2Cl2/hexane ) 1:4) to give 3,5-di-tert-butyl-
2-hydroxybenzaldehyde-d (940 mg, 4 mmol) in 48% yield as a pale-
1
2
yellow solid. H NMR and mass spectra show the H atoms that
are incorporated at the meta position of the phenolate rings remain
intact under the reaction conditions (Figure S13, Supporting
Information).
(42) Nishiyama, Y.; Kakushou, F.; Sonoda, N. Bull. Chem. Soc. Jpn. 2000,
73, 2779–2782.
(43) Larrow, J. F.; Jacobsen, E. N.; Gao, Y.; Hong, Y.; Nie, X.; Zepp,
C. M. J. Org. Chem. 1994, 59, 1939–1942.
9
12404 J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009

Role of External Axial Ligands in Chirality Amplification
A R T I C L E S
Synthesis of (R,R)-1-d. (1R,2R)-1,2-Diaminocyclohexane (225.3
mg, 1.97 mmol) in EtOH (5 mL) was added to a solution of 3,5-
di-tert-butyl-2-hydroxybenzaldehyde-d (932.5 mg, 3.95 mmol) in
EtOH (30 mL). The resulting mixture was heated at 353 K for 1 h
and then allowed to cool to room temperature. The precipitate was
collected by vacuum filtration and dried in vacuo at 353 K. The
product was obtained as a yellow solid in 89% yield (965.2 mg,
1.75 mmol). 1H and 2H NMR spectroscopy shows 2H was
incorporated to the phenolate rings (80% D) as well as the tert-
butyl groups (7% D) (Figure S7, Supporting Information). ESI-
MS spectrometry shows the product contains a mixture of molecules
that are labeled with 2 to 8 2H atoms (Figure S7, Supporting
Information).
Acknowledgment. We thank Mr. Seiji Makita (IMS) for
elemental analysis and measurements of high-resolution mass
spectrometry. This work was supported by grants from the Japan
Science and Technology Agency, CREST and the Japan Science
Promotion Society, the Global COE program.
Supporting Information Available: The X-ray crystallo-
graphic file in the CIF format for 1-Cl, 1-OCH2CF3, and 2-N3.
Figures S1-S13, Table S1, and complete ref 41. This material
is available free of charge via the Internet at http://pubs.acs.org.
JA904635N
9
J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009 12405