X. Liu et al.
(Entry 8). It is suggested that the electron donating effect
depress the stability of the six-membered metal-ligand-
substrate transition state (TS, Fig. 2) [19, 20, 32, 33]. The
double metal complex in TS is confirmed by the elemental
analysis of the isolated Ru(II) complex with 3a when
mixing 1:1 ratio of Ru to ligand 3a in water at 60 °C.
However the electron withdrawing effect favors the for-
mation of hydrogen bonds in TS via increasing the nucle-
ophilicity of ligands to Ru(II). The electron effects on
activity are in accord with Noyori’s theoretical calculations
[32]. Moreover, by building a more crowded space around
the chiral center in the ligands, both steric hindrance and
electron donating effect on the ortho-position (ligand 3c,
entry 5) exerted a more negative influence on conversion
(72.4 %) and enantionselectivity (78.4 %, (S)-enantiomer).
When (S,S,S,S)-tetraaza ligand and p-cymene were used as
ancillaries, two (R)-configured Ru(II) centers were
obtained, in which two parts of the C2 symmetric ligand
were bound to metals independently. The transfer hydro-
genation via this (R)-configured TS with S conformation of
carbonyl carbon atom, resulting perhaps from the attractive
CH/p interaction between p-cymene ligand of the Ru-
complex and the aryl substituent in the substrate, afforded
(S)-enantiomer.
3-chloropropionphenone gave much lower yield but more
replacement by-product (propionphenone, 13.5 %) and the
reduction by-product (1-phenylpropanol, 52.6 %). It seems
that the nucleophilicity of the R group on substrates enables
the ketones to block the ligands to coordinate with the metal
center. In order to prove what was hypothesized, we tried the
3-N,N-dimethylpionphenone as a substrate. We failed to get
the expected alcohol but all propionphenone with its reduc-
tion outcome (1-phenylpropanol), which was in agreement
with the results in Wills’ and Qu’s experiments [34, 35].
4 Conclusions
Herein we report a new series of C2-symmetric bis(sulfo-
nyl) tetraaza ligands. By complexing with Ru(g6-p-
cymene)Cl2]2, the ligands exhibited good catalytic activi-
ties in the asymmetric transfer hydrogenation of aromatic
ketones employing sodium formate as the hydrogen source
in neat water at a modest temperature under insert gas
protection. A moderate to excellent conversions and overall
satisfying enantioselectivities of the chiral secondary
alcohols were obtained under ligand 3a and 3e. The
experiment results reveal that the electron donating effect
and steric effect corrode the reactivity of the ligands but the
electron withdrawing effect enhances the reactivity via
consolidating the sixed-membered transition state. How-
ever, given the applicability of the catalytic system, the
conditions for a wider range of aromatic ketones are still in
a further performance optimization.
Surprisingly, when nitro was imported in para-position
(3d), the yield of the expected 1-phenethylalcohol fallen to
22.0 % and a reversion of enantioselectivity was observed.
This was out of what we expected as only (S)-enantiomer
being the predominant product in the previous reactions.
The same reversion was caught in other substrates such as
entries 4, 9 and 19 exhibited in Table 2. It is speculated
that the delocalized p-bond of nitro causes this distinctness
of enantioselectivity and conversion by dragging the
hydrogen attached upon the prochiral carbon.
Acknowledgments We express our gratitude to the Public Benefit
Project of Zhejiang Science and Technology Department for financial
Support through Project No. 2012C21098 and the National Natural
Science Foundation of China (21101048).
Furthermore, the ATH of different aromatic ketones
(substrates) was explored under the same conditions as
previous. The effect of electron withdrawing had a positive
impact on the ATH over the electron donating groups on
substrates as shown in entries 2, 7, 12 and 15 in Table 2. It
may be attributed to the electron withdrawing effect which
giving more stable hydrogen bonds consisting in TS. Com-
paring the ATH of 2-chloroacetophenone and propionphe-
none with acetophenone under the same ligands (such as
entries 20 and 23 in Table 2, together with entry 8 in
Table 1), we found that a methyl or chlorine atom on a-
position of carbonyl group discouraged the performance of
ATH. This revealed that the bulkier around the carbonyl, the
lower stability of the TS. Notably, in entries 16–20, the
ATH of 2-chloroacetophenone had furnished acetophe-
none as a substitution by-product (9.8–20.5 %). But we
didn’t trace the reduction by-product (1-phenylalcohol),
indicating the priority of ATH to the side reaction of
chlorine atom replacement. In entry 24, the ATH of
References
1. Noyori R, Hashiguchi S (1997) Acc Chem Res 30:97–102
2. Noyori R, Yamakawa M, Hashiguchi S (2001) J Org Chem
66:7931–7944
3. Yamakawa M, Ito H, Noyori R (2000) J Am Chem Soc
122:1466–1478
4. Hashiguchi S, Fujii A, Takehara J, Ikariya T, Noyori R (1995) J
Am Chem Soc 117:7562–7563
5. Wang C, Wu XF, Xiao JL (2008) Chem Asian J 3:1750–1770
6. Hannedouche J, Clarkson GJ, Wills M (2004) J Am Chem Soc
126:986–987
7. Cheung FK, Hayes AM, Hannedouche J, Yim ASY, Wills M
(2005) J Org Chem 70:3188–3197
8. Hayes AM, Morris DJ, Clarkson GJ, Wills M (2005) J Am Chem
Soc 127:7318–7319
9. Morris DJ, Hayes AM, Wills
71:7035–7044
10. Cheung FK, Graharn MA, Minissi F, Wills M (2007) Organo-
metallics 26:5346–5351
M (2006) J Org Chem
123