Transition metal complexes coordinated by an NAD(P)H model compound and their enhanced hydride-donating abilities in the presence of a base

Article abstract of DOI:10.1002/chem.200401211

The ruthenium(II) and rhenium(I) complexes containing an NAD(P)H model compound, 1-benzyl-1,4-dihydronicotinamide (BNAH), as ligand, [Ru(tpy)(bpy)(BNAH)]²⁺ (1a) and [Re(bpy)(CO)₃(BNAH)] ⁺ (1b), were quantitatively produced

Full text of DOI:10.1002/chem.200401211

FULL PAPER
Transition Metal Complexes Coordinated by an NAD(P)H Model
Compound and their Enhanced Hydride-Donating Abilities in the Presence
of a Base
Atsuo Kobayashi,[c] Hideo Konno,[c, d] Kazuhiko Sakamoto,[c] Akiko Sekine,[b]
Yuji Ohashi,[b] Masashi Iida,[a] and Osamu Ishitani*[a]
Abstract: The ruthenium(ii) and rhe-
nium(i) complexes containing an
NAD(P)H model compound, 1-benzyl-
1,4-dihydronicotinamide (BNAH), as
ligand, [Ru(tpy)(bpy)(BNAH)]2+ (1a)
and [Re(bpy)(CO)3(BNAH)]+ (1b),
were quantitatively produced by the re-
action of the corresponding metal hy-
drido complexes with BNA+ (1-benzyl-
nicotinamidium cation). In the pres-
ence of base with pKa =8.9, 1a and 1b
have much greater reducing power
than “free” BNAH. The oxidation po-
tentials of 1a in the absence and the
presence of triethylamine were 0.55 V
and ꢀ0.04 V, respectively, versus Ag/
AgNO3, whereas that of “free” BNAH
was 0.30 V. Spectroscopic results clear-
ly showed that the base extracts a
proton from the carbamoyl group on
1a and 1b to give the deprotonated
BNAH coordinating to the transition-
being coordinated to the metal com-
plexes
[Ru(tpy)(bpy)(BNA+ꢀH+)]2+
(2a) and [Re(bpy)(CO)3(BNA+ꢀH+)]+
(2b); “free” BNAH and the protonat-
ed adducts 1a and 1b cannot act in this
way. X-ray crystallography was per-
metal
complexes
[Ru(tpy)(bpy)-
(BNAHꢀH+)]+
(3a) and [Re-
ꢀ
(bpy)(CO)3(BNAHꢀH+)] (3b); this
deprotonation underlies the enhance-
ment in reducing ability. The deproton-
ated forms 3a and 3b can efficiently
reduce other NAD(P) models to give
the corresponding 1,4-dihydro form, re-
sulting in the deprotonated BNA+
formed on the PF6 salt of 2a, and
showed that the deprotonated nitrogen
atom on the carbamoyl group coordi-
nates to the ruthenium(ii) metal center
with a bond length of 2.086(3) ꢀ. Infra-
red spectral data suggested that the de-
protonated carbamoyl group on the re-
duced forms 3a and 3b is converted to
the imido group, and that the oxygen
atom coordinates to the metal center.
Keywords: acidity
modes hydrides
transition metals
·
coordination
NADH ·
·
·
Introduction
[a] M. Iida, Prof. Dr. O. Ishitani
Department of Chemistry
Graduate School of Science and Engineering
Tokyo Institute of Technology, O-okayama 2-12-1
Meguro-ku 152-8551 (Japan)
Fax : (+81)3-5734-2655
E-mail: ishitani@chem.titech.ac.jp
It is known that the addition of metal ions promotes and ac-
celerates many chemical reactions.[1–2] In many cases the
metal ion coordinates as a Lewis acid to a lone pair or to p
electrons in a substrate, increasing the electrophilicity and
oxidation capability of the substrate. Metal ions also interact
with intermediates, such as radical anions, stabilizing them
and, consequently, influencing reaction rates.
A typical example of metal-ion effects occurs in the re-
duction of various substrates with coenzymes NAD(P)H or
their model compounds.[3] In some enzymic systems, zinc(ii)
ions are essential for the reduction of unsaturated com-
pounds with NAD(P)H.[4] Addition of bivalent metal ions
such as Mg2+ and Sc3+ dramatically accelerates the electron-
or hydride-transfer rate from the NAD(P)H analogues to
oxidants,[5–20] and detailed mechanistic studies show that in-
teraction of the oxidants and/or reduced intermediates with
the metal ions is an important cause of this. It has been re-
ported that NAD(P)H analogues also interact with metal
[b] Dr. A. Sekine, Prof. Dr. Y. Ohashi
Department of Chemistry and Material Science
Graduate School of Science and Engineering
Tokyo Institute of Technology, O-okayama 2-12-1
Meguro-ku 152-8551 (Japan)
[c] Dr. A. Kobayashi, Dr. H. Konno, Prof. Dr. K. Sakamoto
Graduate School of Science and Engineering
Saitama University, 255 Shimo-Okubo
Saitama 338-8570 (Japan)
[d] Dr. H. Konno
New address
National Institute of Advanced Industrial Science and Technology
(AIST)
16-1 Onogawa, Tsukuba 305-8569 (Japan)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2005, 11, 4219 – 4226
DOI: 10.1002/chem.200401211
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4219

ions.[5b,c,7,8a,21] The oxidation potential of a binary complex of
the NAD(P)H analogue with the metal ion is shifted posi-
tively relative to the “free” NAD(P)H analogue.[20] For ex-
ample, the complex of 1-benzyl-1,4-dihydronicotinamide
(BNAH, a typical NAD(P)H analogue) and Mg2+ is oxi-
dized at a potential 0.23 V more positive than “free”
BNAH. Consequently, the reducing abilities of the
NAD(P)H analogues themselves are lessened by complexa-
tion with the metal ions, although formation of a ternary
complex involving an NAD(P)H analogue, a metal ion, and
a oxidant has been assumed as another cause of the acceler-
ation.[5,19]
atom of the carbamoyl group of BNA+ coordinates the cen-
tral ruthenium ion, according to spectroscopic evidence and
X-ray crystallographic results as follows. The NMR signals
of the protons at the 2- and the 4-positions of the pyridini-
um cation in 2a were observed at 0.5 and 1.0 ppm downfield
from those in “free” BNA+, respectively, but the shifts in
other signals of the deprotonated BNA+ ligand were much
smaller (<0.3 ppm). The C=O stretching band of 2a was ob-
served at 67 cmꢀ1 lower than “free” BNA+, and the nCꢀ
N
band for 2a shifted up by 39 cmꢀ1 (Table 1). These shifts are
characteristic of deprotonated carbamoyl groups coordinat-
ing to metal complexes through the nitrogen atom.[22]
We report here the unique reactivities of NAD(P)H
model compounds coordinating to a ruthenium(ii) or rheniu-
m(i) complex.[21a] They have greater reducing abilities in the
presence of base than the corresponding “free” model com-
pound. Spectroscopic studies clearly show that deprotona-
tion from the carbamoyl group on the NAD(P)H model
compound is a key stage in this “unusual” phenomenon.
The X-ray crystallographic structure of the NAD(P) model
compound coordinating to the ruthenium(ii) complex is also
reported for the first time.
Table 1. IR spectral data [cmꢀ1] measured in CD3CN at room tempera-
ture.
[a]
[b]
nC5
nC2
nC
=
O
=
C6
=
C3
1a
1680
1679
1680
1681
1687
1673
1684
–
1631
1614
1633
1612(sh)
1651
1590
1646
–
1530
–
1520
–
1578
1521
1550
1638
1640
1705
–
1308
–
1309
–
1300
–
1284
1427
1437
1388
[c]
1a + NEt3
1b
1b + NEt3
BNAH
1c
1563(sh)
–
1570
[d]
–
–
–
–
–
–
1d
2a
Results and Discussion
2b
–
–
–
–
BNA+
ꢀ
As a typical run, PF6 salts of 1-benzylnicotinamidium
[a] Medium peak, which is probably attributable to nC
.
See text.
=
N
cation (BNA+, 32 mmol) were added to a solution of [Ru-
[b] Medium peak, which is probably attributable to nCꢀN. See text. [c] In
the presence of 0.2m NEt3, 3a was quantitatively formed. [d] IR bands at-
tributed to 3b, which was formed by deprotonation of 1b in the presence
of 0.2m NEt3.
ꢀ
(tpy)(bpy)H]+PF6 (32 mmol) in acetonitrile, to give a 1:1
adduct 1a within one minute. This solution was quickly
added to a further solution of triethylamine (0.2m final con-
centration) and PF6 salts of 1-benzyl-N,N-diethylnicotina-
ꢀ
This identification is clearly supported by the X-ray crys-
ꢀ
tallographic analysis for the PF6 salt of 2a. Figure 1 shows
the ORTEP drawing of 2a, and selected bond lengths and
angles are summarized in Table 2. The structure clearly
shows that the nitrogen atom on the deprotonated carbamo-
yl group is directly bonded to the ruthenium center, since
only a single hydrogen atom bonded to the nitrogen atom
midium cation (BEt2NA+, 120 mmol) in acetonitrile under
an Ar atmosphere. The solution was kept at room tempera-
ture for 1 h, and the 1,4-dihydro product of BEt2NA+ (1,4-
BEt2NAH) was formed selectively; the position isomers of
1,4-BEt2NAH and the dihydro products of BNA+ were not
detected [Eq. (1)].
was found and the bond lengths of N6 C26 and O1 C26
were 1.315(4) and 1.234(4) ꢀ, respectively, which are typical
values for the carbamoyl group on NAD(P) model com-
pounds (1.32 and 1.23 ꢀ).[23] Both the N6 and C26 atoms on
the deprotonated carbamoyl group have trigonal planar
structures, and the torsion angle t[Ru-N6-C26-O1] was
4.2(5)8, showing that the lone pair of electrons on N6 and
ꢀ
ꢀ
From the reaction solution, a ruthenium complex coordi-
nated with deprotonated BNA+ (2a) was isolated at 75%
the C26 O1 double bond are highly conjugated in the same
ꢀ
manner as the carbamoyl group on “free” 1-methylnicotina-
mide, in which the bond angles around the carbamoyl nitro-
gen atom are 124, 116, and 1198.[23a] The pyridinium ring is
almost planar and the bond lengths on it are identical to
those on “free” 1-methylnicotinamide to within 0.01 ꢀ.[23]
The torsion angle t[C28-C27-C26-O1], which gives an indi-
cation of the conformation about the bond between the pyr-
idinium ring and the carbamoyl group, is 18.5(4)8, similar to
one of the two most stable conformations calculated theo-
yield. An electrospray ionization mass spectrum of 2a
showed essentially a single peak at m/z=352, attributable to
[Ru(tpy)(bpy)(BNA+ꢀH+)]2+. In 2a, the deprotonated N-
4220
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 4219 – 4226

Coordination Chemistry
FULL PAPER
These results clearly indicate that, specifically in the pres-
ence of NEt3, 1,4-BNAH coordinating to the ruthenium
complex has greater hydride-donor ability than “free” 1,4-
BNAH.
What is the crucial role of NEt3 in the hydride-transfer re-
actions of 1a to the NAD(P) models? Addition of NEt3 to a
1
solution of 1a in CD3CN caused H NMR spectral changes
1
that are summarized in Table 3 (3a), whereas the H NMR
Table 3. 1H NMR data of the NAD(P)H-model ligands in the complexes
and “free” NAD(P)H-model.
Chemical shift [ppm][a]
Figure 1. ORTEP view of the [Ru(tpy)(bpy)(BNA+-H+)]2+ molecule.
Probability of the thermal ellipsoid is 30%. PF6 counterions are omitted
for clarity.
R[b]
2
4
5
6
CH2(-Ph)
ꢀ
1a
1a+NEt3
1b
1b+NEt3
BNAH
1c[e]
CONH2
CONH
CONH2
CONH
CONH2
COCH3
COCH3
CONEt2
CONEt2
6.15
6.24
6.06
6.48
6.95
7.25
7.28
5.05
5.88
2.54
2.61
2.71
2.57
3.04
2.46
2.94
2.13
2.99
4.68
4.36
4.73
4.36
4.70
4.94
4.85
3.95
4.52
5.69
5.71
5.71
5.72
5.82
5.77
5.82
5.41
5.87
4.02
4.08
4.05
4.13
4.29
4.35
4.37
3.84
4.22
[c]
[d]
Table 2. Selected bond lengths [ꢀ] and angles [8] for [Ru(tpy)(bpy)-
(BNA+-H+)](PF6)2 (2a).
ꢀ
ꢀ
Ru1 N1
2.086(3)
2.070(3)
2.069(3)
1.948(3)
2.073(4)
Ru1 N6
2.086(3)
1.313(4)
1.232(4)
1.520(4)
0.82(4)
BAcPyH
1d[e]
ꢀ
ꢀ
Ru1 N2
N6 C26
ꢀ
ꢀ
Ru1 N3
O1 C26
BEt2NAH
ꢀ
ꢀ
Ru1 N4
C26 C27
[a] Numbers show the position on the dihydropyridine ring. Data were
obtained from CD3CN, and the residual protons of the solvent were used
as an internal standard. [b] Substituent group on the 3-position of the di-
hydropyridine ring. [c] Signals attributed to 3a, which was formed by de-
protonation of 1a in the presence of NEt3 (0.2m). [d] Signals attributed
to 3b, which was formed by deprotonation of 1b in the presence of NEt3
(0.2m). [e] Reference [21a].
ꢀ
ꢀ
Ru1 N5
N6 H
N1-Ru1-N2
N3-Ru1-N4
N4-Ru1-N5
N1-Ru1-N6
N2-Ru1-N6
N3-Ru1-N6
N4-Ru1-N6
N5-Ru1-N6
77.98(12)
79.29(13)
79.84(13)
94.28(13)
172.17(10)
93.72(13)
92.69(12)
85.08(12)
C26-N6-Ru1
O1-C26-N6
O1-C26-C27
N6-C26-C27
C26-N6-H6N
Ru1-N6-H6N
129.3(2)
125.4(3)
115.6(3)
119.0(3)
115(3)
115(3)
spectra of “free” 1,4-BNAH and other 1:1 adducts without a
carbamoyl moiety at the 3-position, namely 1c and 1d,[21]
were not affected by addition of NEt3. In other words, the
chemical shifts for the protons located close to the carbamo-
retically (1508 and 358).[24] The [Ru(tpy)(bpy)] group of 2a
has a similar structure to the reported [Ru(tpy)(bpy)L]+
ꢀ
type complexes (L=Iꢀ, HCO2 ),[25] that is, slightly distorted
octahedral.
The hydride transfer reaction from 1a to BEt2NA+ can
1
be followed by using H NMR and IR spectroscopy, which
shows that the reaction proceeds almost quantitatively, as
shown in Equation (1). On the other hand, “free” 1,4-
BNAH cannot reduce BEt2NA+ at ambient temperature,
even in the presence of NEt3 and [Ru(tpy)(bpy)(MeCN)]2+
.
yl group of the 1,4-BNAH ligand on the 2-, 4-, and 5-posi-
tions of the dihydropyridine ring were shifted by 0.07–
0.32 ppm. However the protons on the 6-position and on the
benzyl group showed much smaller shifts (<0.06 ppm) upon
addition of NEt3.
In the absence of NEt3, quantitative splitting of 1a proceeds
in MeCN to give “free” 1,4-BNAH and [Ru(tpy)(bpy)-
(MeCN)]2+ [Eq. (2)],[21a] but formation of 2a and 1,4-
BEt2NAH was not observed. A similar phenomenon was
observed by using a different NAD(P) model, the 1-benzyl-
3-acetyl-pyridinum cation (BAcPy+) instead of BEt2NA+.
The IR bands related to the carbamoyl group and its con-
ꢀ
jugated C2 C3 double bond of 1a also changed dramatically
Chem. Eur. J. 2005, 11, 4219 – 4226
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4221

O. Ishitani et al.
and Equation (4), the equilibrium constant K was deter-
mined as 0.5. Addition of bases with (pKa)W values=7.76
and (pKa)AN =15.9 (triethanolamine) did not change the
1H NMR and IR spectra of 1a. Figure 3 shows the depend-
Figure 2. IR spectral change of a solution of 1a (8.0 mm) in CD3CN
before (dotted line) and after (solid line) addition of NEt3 (0.2m).
*
Figure 3. Dependence of (- -) the yield of 2a produced by the reaction
*
as shown in Equation (1), (- -) the proportion of 3a in the equilibrium
~
between 1a and 3a as shown in Equation (3), and (- -) the equilibrium
with addition of NEt3, but no shift was observed for the nC5
=
C6
constant between 1a and 3a on the (pKa)W value of the conjugated acid
of an added base (0.2m):[26a]1,10-phenantroline (1.9); nicotinic acid
(3.47); pyridine (5.42); triethanolamine (7.76); 2-aminoethanol (8.9); di-
ethanolamine (9.5); triethylamine (10.72); diethylamine (10.93); piperi-
dine (11.5).
band, as shown in Figure 2. The nC2
band was shifted
=
C3
down by 17 cmꢀ1. The nC band, which gives as a strong
=
O
peak at 1530 cmꢀ1, and a medium peak at 1308 cmꢀ1, which
is probably attributable to the nCꢀ band in the absence of
N
NEt3, both disappeared, and a new medium absorption band
was observed at ~1563 cmꢀ1. These results imply that NEt3
acts as a base that takes on a proton from the carbamoyl
group of 1a to give 3a. The coordination mode of the de-
protonated 1,4-BNAH to the central ruthenium atom in 3a
should be different from 2a, in which the deprotonated ni-
trogen atom of the carbamoyl group of BNA+ coordinates
ence of the equilibrium constants of Equation (3) on (pKa)W
of the bases added;[26] from this result, the (pKa)AN value of
1a was estimated at 18.5ꢁ0.2. The coordination of 1,4-
BNAH to the ruthenium complex dramatically increases the
basicity of the carbamoyl group, since even the strongest
base used here, piperidine, for which (pKa)W =11.5 and
(pKa)AN =18.9, did not cause deprotonation from “free” 1,4-
BNAH.
ꢀ
ꢀ
the ruthenium metal (i.e., -C(=O) NH Ru), as described
above, since the IR bands related to the deprotonated car-
bamoyl groups are very different in these two complexes
(Figure 2 and Table 1). The absence of the nC band and
the medium peak at 1530 cmꢀ1 in the case of=O3a suggests
that the deprotonated carbamoyl group is converted to the
imide form and the oxygen atom coordinates the ruthenium,
K ¼ ½3 aꢂ½BaseHþꢂ=½1 aꢂ½Baseꢂ
ð4Þ
Figure 3 also shows the dependence of the proportion of
3a at equilibrium with 1a in the presence of various bases
(0.2m), and dependence of the yields of 2a, in the reduction
of BEt2NA+ (30mm) with 1a (8.0mm) and base (0.2m) upon
the pKa values of the added bases.[27] Similar dependences of
the equilibria and the reactivities on the added base were
also observed by using BAcPy+ instead of BEt2NA+ as a
hydride acceptor. These results strongly imply that hydride
reduction of the NAD(P) models proceeds by 3a, but not
1a, which dissociates to 1,4-BNAH and the solvento complex.
This claim is also supported by the observation that the
adducts of [Ru(tpy)(bpy)H]+ with BAcPy+ (1c, Tables 1
and 3), in which the substituent at the 3-position has no dis-
sociative proton, could not reduce BEt2NA+ even in the
presence of piperidine, and quantitatively dissociated to the
corresponding 1,4-dihydro product and the solvato complex.
How much does the coordination to the ruthenium com-
plex and deprotonation from the carbamoyl group influence
the redox property of 1,4-BNAH? Figure 4 shows cyclic vol-
ꢀ ꢀ
that is, -C(=NH) O Ru [Eq. (3)]. The new band at
1563 cmꢀ1 is probably from the C=N vibrational band.[22]
The protonated (1a) and deprotonated (3a) forms are in
equilibrium with each other in the presence of NEt3, and
also in the presence of other strong bases with (pKa)W =8.9
and (pKa)AN =17.5 (2-aminoethanol), where (pKa)W and
(pKa)AN are pKa values measured in aqueous solution[26a]
and in an acetonitrile,[26b,c] respectively. For example, the
ratio of 1a and 3a (total concentration 8.0mm) in MeCN
was 7:93 in the presence of 0.2m NEt3; by using these data
4222
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 4219 – 4226

Coordination Chemistry
FULL PAPER
of 1a.[28] Note that 3a is oxidized at a potential 0.36 V more
negative even than “free” 1,4-BNAH. This is consistent with
the greater hydride-donating ability of 3a relative to that of
“free” 1,4-BNAH. We believe that this electrochemical data
(Table 4) is the first clear evidence that complexation with a
Lewis acid enhances the reducing power of the NAD(P)H
model with the assistance of a base.
Table 4. Oxidation potentials of the complexes and “free” BNAH[a]
.
Eopx [V vs Ag/AgNO3][b]
Eopx [V vs Ag/AgNO3][b]
1a
1b
BNAH
0.55, 0.86, 0.98[c]
0.60, 1.0
0.32
3a[d]
ꢀ0.04
0.13
0.30
3b[d]
BNAH[d]
[a] Cyclic voltammograms were taken in CH3CN containing 0.1m
Et4NBF4 at a 0.2 Vsꢀ1 scan rate, using a glassy-carbon working electrode,
a Pt counter electrode, and a Ag/AgNO3 (0.1m) reference electrode
under an Ar atmosphere. [b] Peak potentials for irreversible oxidation
ox
1=2
waves. [c] E for the quasi-reversible oxidation process. [d] In the pres-
ence of 0.2m NEt3.
The adduct of 1,4-BNAH with [Re(bpy)(CO)3] complex
Figure 4. Cyclic voltammograms of A) 1,4-BNAH, B) 1a, and C) 3a
(3.4 mm) in MeCN containing Et4NBF4 (0.1m) under an Ar atmosphere
at a scan rate of 200 mVsꢀ1. In the case of C), the solution contained
0.2m NEt3, and the wave at ~0.3 V is attributed to oxidation of NEt3.
1b is also deprotonated from the carbamoyl moiety by addi-
tion of base to give 3b, for which the IR and H NMR spec-
tral changes are similar to those of 1a (Tables 1 and 3).
1
Also, IR stretching bands of the CO ligands in the depro-
tonated form 3b were observed at 2011 and 1907 cmꢀ1
;
tammograms (CV) of 1,4-BNAH (A), 1a (B), and 3a (C)
measured in MeCN or MeCN-NEt3 (0.2m). An irreversible
anodic wave was observed at 0.32 V for oxidation of “free”
1,4-BNAH in MeCN, and addition of NEt3 did not signifi-
cantly affect the oxidation potential. On the other hand, the
CV of 1a shows two irreversible anodic waves, attributable
to oxidation of the 1,4-BNAH ligand at peak potentials of
0.55 and 0.86 V, and one quasi-reversible RuIII/RuII wave at
0.98 V. Unfortunately, reversible waves for the first two oxi-
dations could not be observed even at high scan rates up to
2000 Vsꢀ1. Even after an anodic scan up to 1.2 V, the follow-
ing cathodic scan did not show a wave attributable to the
BNA+ adduct 2a, which has redox potential for RuII/RuIII
of 0.33 V. Although interpretation of the CV of NAD(P)
model compounds is still incomplete,[3b] it is clear that their
oxidation potentials strongly reflect their reduction abilities.
The NAD(P)H model compounds are known to form 1:1
complexes with metal ions, including Mg2+. It has been re-
these values are lower by 19 and 8 cmꢀ1 than those found
for 1b. Deprotonation from the carbamoyl group coordinat-
ing to the central rhenium(i) ion should increase the charge
density of the metal ion. This causes increase of p-back don-
ation from the carbonyl ligands, and weakens the CO bonds.
The difference in electrochemical properties between 1b
and 3b is similar to the ruthenium complexes; the oxidation
potential of 3b was negatively shifted by 0.47 V relative to
1b (Table 4).
Although these data strongly indicate that the coordinat-
ing structures of 1b and 3b are similar to the respective
ruthenium(ii) complexes 1a and 3a, deprotonation from 1b
is a much slower process than from 1a. For example, 80 mi-
nutes was required after mixing 1b and NEt3 (0.2m) to
reach equilibrium between 1b and 3b, though equilibrium
between 1a and 3a was reached within a minute, and split-
ting of 1b to “free” BNAH and the solvent complex occur-
red competitively in this condition (Figure 5 and Scheme 1).
The equilibrium constants (K) between 1b and 3b in the
presence of a base were smaller than for the ruthenium
complex 1a and 3a. For example, in the presence of 0.2m
NEt3, K=2ꢂ10ꢀ2 and (pKa)AN =19.9 for the rhenium com-
plexes, and K=0.6 and (pKa)AN =18.5ꢁ0.2 for the rutheni-
um complexes. The difference in charge of the central metal
ions, that is, the stronger acidity of the ruthenium complex,
is probably one of the causes of this difference. The local
concentration of NEt3, which has a donor number (DN)[29]
of 61 compared to 14.1 for the solvent MeCN, would be
higher around the ruthenium(ii) complex than around the
rhenium(i) complex. In fact, using the more basic solvent
ported that the oxidation potential of 1,4-BNAH is shifted
[3b,20]
positively by ~0.2 V upon complexation with Mg2+
.
It is
therefore likely that the coordination of 1,4-BNAH to the
central ruthenium(ii) metal in 1a causes the 0.2 V positive
shift of the oxidation potential. Clearly, s donation from the
carbamoyl group of 1,4-BNAH to RuII should lower the re-
ducing power of the 1,4-BNAH ligand in 1a.
Addition of NEt3 (0.2m) caused a dramatic negative shift
of the oxidation potential of the complex, as shown in Fig-
ure 4C; a first irreversible anodic wave, attributable to the
oxidation of 3a, was observed at ꢀ0.04 V versus Ag/AgNO3,
which is 0.59 V more negative than the first oxidation wave
Chem. Eur. J. 2005, 11, 4219 – 4226
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4223

O. Ishitani et al.
Experimental Section
General procedures: 1H NMR spectra were recorded on a Bruker ARX-
400 NMR spectrometer (400 MHz) at 258C. Infrared absorption spectra
were recorded by JASCO FT/IR-610 spectrometers with a MCT detector
at ambient temperature. Electrospray ionization mass (ESI MS) spectra
were obtained by using a Mariner mass spectrometer (Applied Biosys-
tems) with a 100 V nozzle potential. Cyclic voltammetry was carried out
in an acetonitrile containing tetraethylammonium tetrafluoroborate
(0.1m) as a supporting electrolyte, using an ALS/CHI CHI-620 electro-
chemical analyzer with a carbon working electrode (3 mm diameter), a
Ag/AgNO3 (0.01m) reference electrode, and a Pt counter electrode. A
BAS C2 cell stand and a platinum (10 mm diameter) working electrode
were used for rapid-scan cyclic voltammetry.
Materials: Acetonitrile was distilled over P2O5 three times and over
CaH2 immediately before use. Dimethylformamide (DMF) was dried
over molecular sieves 4 ꢀ and distilled at reduced pressure (~20 Torr)
before use. The compound RuCl3·3H2O was kindly supplied by Kojima
Chemical Co. The hydrido complexes: [Ru(tpy)(bpy)H](PF6)·0.5H2O,[25a]
and [Re(bpy)(CO)3H],[33] the PF6 salt of the NAD(P) model compounds:
BNA+,[34] BEt2NA+,[21c,d,35] BAcPy+,[21c,d] and the corresponding 1,4-dihy-
droforms: 1,4-BNAH,[34] 1,4-BEt2NAH,[21c,d,35] and 1,4-BAcPyH[21c,d] were
prepared according to the literature methods.
*
*
Figure 5. Reaction of 1b (- -) with NEt3 (0.2m); yields of 3b (- -) and
~
BNAH (- -) based on 1b used.
In situ preparation of 1a–d: A solution of an NAD(P) model (8.0mm) in
CD3CN (0.5 mL) was added to a solution (0.5 mL) a hydrido complex
(8.0mm) in CD3CN (0.5 mL) in an NMR tube under an argon atmos-
phere. After purging with argon for 1 min, the NMR tube was sealed.
1H NMR and ESI MS data of [Ru(tpy)(bpy)(BNAH)]2+ (1a), [Re-
(bpy)(CO)3(BNAH)]+ (1b), [Ru(tpy)(bpy)(BAcPyH)]2+ (1c), and [Ru-
(tpy)(bpy)(BEt2NAH)]2+ (1d) are reported in the Supporting Informa-
tion of reference [21] (http://pubs.acs.org.).
[Ru(tpy)(bpy)(BNA+ꢀH+)](PF6)2 (2a): A hexafluorophosphate salt of
BNA+ (60 mmol) was added to
a solution of [Ru(tpy)(bpy)H]-
(PF6)·0.5H2O (30 mmol) and NEt3 (0.2m) in acetonitrile (2.0 mL). The so-
lution was kept under dim light for 20 min at ambient temperature, then
evaporated under reduced pressure at ambient temperature. The black
precipitates were collected by filtration, washed twice with diethyl ether,
and purified by column chromatography on aluminum oxide (Merck,
eluent toluene/CH3CN 2:1). The main dark purple band was collected,
and the solvent was removed using a rotary evaporator. The brown
powder was recrystallized using CH3CN/diethyl ether, then in vacuo; the
typical yield was 75%. 1H NMR (400 MHz, CD3CN): d=9.57 (d, J=
5.2 Hz, 1H), 8.78 (s, 1H), 8.59 (d, J=8.0 Hz, 1H), 8.49 (d, J=6.0 Hz,
1H), 8.41 (d, J=8.0, Hz, 1H), 8.34 (t, J=8.2 Hz, 2H), 8.30 (d, J=8.4 Hz,
2H), 8.22 (t, J=8.0 Hz, 1H), 8.02 (t, J=8.0 Hz, 1H), 7.84 (t, J=8.0 Hz,
1H), 7.84 (t, J=8.2 Hz, 5H), 7.79 (d, J=0.4 Hz, 2H) 7.77 (d, J=5.4 Hz,
1H), 7.68 (t, J=5.5 Hz, 1H), 7.39 (m, 3H), 7.29 (m, 2H), 7.21 (t, J=
7.0 Hz, 2H), 7.16 (d, J=5.0 Hz, 1H), 6.85 (t, J=5.0 Hz, 1H), 5.56 (s,
2H), 4.69 ppm (s, 1H); elemental analysis calcd (%) for
C38H32F12N7OP2Ru: C 45.93, H 3.25, O 9.87; found: C 45.86, H 3.23, O
9.50.
Scheme 1. Reaction of 1b in the presence of NEt3.
DMF (DN=26.6) instead of MeCN caused K between 1b
and 3b in the presence of 0.2m NEt3 to decrease to 2ꢂ10ꢀ3.
Finally, it is noteworthy that the hydride transfer from 3a
to the NAD(P) models is also interesting as a model of
“transhydrogenation” reactions between NAD and NADH
by the enzyme transhydrogenase.[30] Only the 1,4-dihydro-
form of the NAD(P)H models is produced by 3a as de-
scribed above,[31] and this is the same product distribution as
the enzymic reaction, whereas most other reported transhy-
drogenation reactions between NAD and NADH models
give the 1,6- and/or 1,2-dihydro form as byproduct(s).[32]
ꢀ
Crystal structure determination: A single dark-red crystal of the PF6 salt
of 2a was obtained by slow diffusion of diethyl ether into a solution of
the complex in acetonitrile. Diffraction data was collected on a Rigaku
RAPID imaging-plate diffractometer with graphite-monochromated
MoKa radiation (l=0.71073 ꢀ) with a crystal of size 0.25ꢂ0.10ꢂ0.01 mm
at 293 K. The intensity data were corrected for Lorentz and polarization
effects. Absorption correction[36] was applied. The structure was solved
by the direct method using the TEXSAN program,[37] and was refined on
F2 by using full-matrix least-square procedures implemented in the
SHELXL-97 program.[38] No secondary extinction corrections were ap-
plied. In the least-square refinements, all non-hydrogen atoms were re-
Conclusion
Acidity of the carbamoyl group on the BNAH increases
dramatically by coordination to the ruthenium(ii) or rheniu-
m(i) complex. Addition of base ((pKa)W >8.9) gives relative-
ly stable deprotonated BNAH coordinating to the metal
complex. These deprotonated BNAH Ru and Re com-
plexes have much stronger hydride-donating abilities and
are oxidized at more negative potentials than “free” BNAH.
The 1:1 adduct of the NAD(P) model and the ruthenium
complex has been isolated for the first time and its structure
successfully determined by X-ray crystallographic analysis.
II
I
ꢀ
ꢀ
ꢀ
fined with anisotropic displacement parameters. PF6 appeared to be dis-
ordered, but we did not succeed in separating the disordered structures.
There are no peaks beyond 0.33 ꢀꢀ3 except for those around PF6ꢀ. The
crystallographic data are as follows: [Ru(tpy)(bpy)(BNA+-H+)](PF6)2],
Mr =962.75, monoclinic, P21/c, a=14.281(18), b=26.02(3), c=
11.678(10) ꢀ, b=110.48(5)8, V=4069(7) ꢀ3, Z=4, 1calcd =1.572 gcmꢀ3, F-
4224
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 4219 – 4226

Coordination Chemistry
FULL PAPER
(000)=1936, T=298 K, R=0.044 for 6855 reflections [I>2s(I)], R=
0.061 for all data.
Chem. 1999, 20, 145–165; d) U. Eisner, J. Kuthan, Chem. Rev. 1972,
72, 1–42; e) R. Lavilla, J. Chem. Soc. Perkin Trans. 1 2002, 1141–
1156.
CCDC-255988 contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[4] C.-I. Branden, H. Jonvall, H. Eklund, B. Furugren, in The Enzymes
3rd ed. Vol. 11 (Ed.: P. D. Boyer), Academic Press, New York, 1975,
pp. 103–190.
[5] a) A. Ohno, S. Yasui, R. A. Gase, S. Oka, U. K. Pandit, Bioorg.
Chem. 1980, 9, 199–211; b) A. Ohno, H. Yamamoto, T. Okamoto, S.
Oka, Y. Ohnishi, Bull. Chem. Soc. Jpn. 1977, 50, 1535–1538; c) A.
Ohno, S. Yasui, S. Oka, Bull. Chem. Soc. Jpn. 1980, 53, 2651–2654.
[6] D. J. Creighton, D. S. Sigman, J. Am. Chem. Soc. 1971, 93, 6314–
6316.
ꢀ
Reaction of an BEt2NA+ with 1d in the presence of piperidine: PF6
salts of BAcPy+ (32 mmol) were added to a solution of [Ru(tpy)(bpy)H]+
(PF6ꢀ) (32 mmol) in acetonitrile (2 mL), to give a 1:1 adduct 1d within
1 min. This solution was added to a solution of piperidine (0.2m final con-
ꢀ
centration) and PF6 salts of BEt2NA+ (120 mmol) in acetonitrile under
an Ar atmosphere. The solution was kept at room temperature for 1 h,
and 1,4-BAcPyH was formed quantitatively; however, reduction products
of BEt2NA+ were not detected at all.
[7] O. Ishitani, M. Ihama, Y. Miyauchi, C. Pac, J. Chem. Soc. Perkin
Trans. 1 1985, 1527–1531.
ꢀ
Determination of the equilibrium constants between 1 and 3: Both PF6
[8] a) S. Fukuzumi, N. Nishizawa, T. Tanaka, Chem. Lett. 1983, 1755–
1758; b) S. Fukuzumi, Y. Fujii, T. Suenobu, J. Am. Chem. Soc. 2001,
123, 10191–10199; c) S. Fukuzumi, O. Inada, N. Satoh, T. Suenobu,
H. Imahori, J. Am. Chem. Soc. 2002, 124, 9181–9188.
[9] a) S. Shinkai, T. C. Bruice, J. Am. Chem. Soc. 1972, 94, 8258; b) S.
Shinkai, T. C. Bruice, Biochemistry 1973, 12, 1750–1759.
[10] M. Shirai, T. Chishina, M. Tanaka, Bull. Chem. Soc. Jpn. 1975, 48,
1079–1080.
salts of BNA+ (8.0mm) and hydrido complex (8.0mm) were dissolved in
CD3CN (0.5 mL), and the solution was quickly added to a solution of
amine (0.2m final concentration) in CD3CN (0.5 mL). These procedures
were carried out in an NMR tube under an argon atmosphere. The reac-
tion was monitored by using 1H NMR spectroscopy. After equilibria be-
tween 1a and 3a, and 1b and 3b were achieved, the ratio of concentra-
tion was determined from the area average of the 1H NMR peaks at d=
6.15 and 4.68 ppm (1a); d=6.24 and 4.36 ppm (3a); d=6.06 and
4.73 ppm (1b); and d=6.48 and 4.36 ppm (3b). These data and Equation
(4) were used for determination of the equilibrium constants.
[11] R. A. Gase, G. Boxhoorn, U. K. Pandit, Tetrahedron Lett. 1976, 17,
2889–2892.
[12] U. K. Pandit, H. van Dam, J. B. Steevens, Tetrahedron Lett. 1977, 18,
913–916.
[13] K. Watanabe, R. Kawaguchi, H. Kato, Chem. Lett. 1978, 255–258.
[14] R. A. Gase, U. K. Pandit, J. Am. Chem. Soc. 1979, 101, 7059–
7064.
Acknowledgement
[15] A. J. De Koning, H. J. Alberts-Jansen, J. Boersma, G. J. M. Van der
Kerk, Recl. Trav. Chim. Pays-Bas 1980, 99, 316–321.
[16] J. B. Steevens, U. K. Pandit, Tetrahedron 1983, 39, 1395–1400.
[17] J. F. J. Engbersen, A. Koudijs, H. C. V. D. Plas, J. Org. Chem. 1990,
55, 3647–3654.
We acknowledge helpful discussion of Dr. Chyongjin Pac and Prof. Shu-
nichi Fukuzumi (Osaka University). We also thank the Kojima Chemical
Company for a generous gift of pure RuCl3·3H2O. This work was partial-
ly supported by a Grant-in Aid for Scientific Research on Priority Areas
(417) from the Ministry of Education, Culture, Sports, Science, and Tech-
nology (MEXT) of the Japanese Government.
[18] N. Kanomata, T. Nakata, Angew. Chem. 1997, 109, 1263–1266;
Angew. Chem. Int. Ed. Engl. 1997, 36, 1207–1211.
[19] S. Yasui, A. Ohno, Bioorg. Chem. 1986, 14, 70–96.
[20] S. Fukuzumi, S. Koumitsu, K. Hironaka, T. Tanaka, J. Am. Chem.
Soc. 1987, 109, 305–316.
[1] For reviews see: a) S. Fukuzumi, in Electron Transfer in Chemistry
(Ed.: V. Balzani), Wiley-VCH, Weinheim, 2001, pp. 43–67; b) H.
Yamamoto, Lewis Acid Chemistry: A Practical Approach, Oxford
University Press, Oxford, 1999; c) S. V. Kessar, P. Singh, Chem. Rev.
1997, 97, 721–737; d) K. Mizuno, Y. Otsuji, Top. Curr. Chem. 1994,
169, 301–346; e) Selectivities in Lewis Acid Promoted Reactions
(Ed.: D. Schinzer), Kluwer, Boston, 1989.
[21] a) A. Kobayashi, R. Takatori, I. Kikuchi, H. Konno, K. Sakamoto,
O. Ishitani, Organometallics 2001, 20, 3361–3363; b) H. Konno, K.
Sakamoto, O. Ishitani, Angew. Chem. 2000, 112, 4227–4229; Angew.
Chem. Int. Ed. 2000, 39, 4061–4063; c) H. C. Lo, O. Buriez, J. B.
Kerr, R. H. Fish, Angew. Chem. 1999, 111, 1524–1527; Angew.
Chem. Int. Ed. 1999, 38, 1429–1432; d) H. C. Lo, C. Leiva, O.
Buriez, J. B. Kerr, M. M. Olmstead, R. H. Fish, Inorg. Chem. 2001,
40, 6705–6716; e) P. S. Wagenknecht, E. J. Sambrisk, Recent Res.
Dev. Inorg. Chem. 2003, 3, 35–50; f) P. S. Wagenknecht, J. M.
Penney, R. T. Hembre, Organometallics 2003, 22, 1180–1182.
[22] a) R. J. Balahura, R. B. Jordan, J. Am. Chem. Soc. 1970, 92, 1533–
1539; b) D. P. Powell, A. Woollins, Spectrochim. Acta Part A 1985,
41, 1023–1033.
[23] a) G. R. Freeman, C. E. Bugg, Acta Crystallogr. Sect. B 1974, 30,
431–443; b) C. I. Branden, I. Lindqvist, M. Zeppezauer, Ark. Kemi
1968, 30, 41–50.
[24] J. L. Coubeils, B. Pullman, P. Courriere, Biochem. Biophys. Res.
Commun. 1971, 44, 1131–1134.
[25] a) H. Konno, A. Kobayashi, K. Sakamoto, F. Fagalde, N. E. Katz, H.
Saitoh, O. Ishitani, Inorg. Chim. Acta 2000, 299, 155–163; b) M. A.
Billadeau, W. T. Pennington, J. D. Petersen, Acta Crystallogr. Sect. C
1990, 46, 1105–1107.
[26] Because the (pKa)AN values of some bases used in this study have
not been reported, (pKa)w values are used in Figure 3. For (pKa)w,
see: a) D. R. Lide, Handbook of Chemistry and Physics, 73rd ed.,
CRC Press, Boca Raton, 1992. For (pKa)AN, see: b) K. Izutsu, Acid-
Base Dissociation Constants in Dipolar Aprotic Solvents, Blackwell
Scientific, Oxford, 1990; c) J. F. Coetzee, G. R. Padmanabhan, J.
Am. Chem. Soc. 1965, 87, 5005–5010.
[2] a) P. Renaud, M. Gerster, Angew. Chem. 1998, 110, 2704–2722;
Angew. Chem. Int. Ed. 1998, 37, 2562–2579; b) S. Itoh, H. Kumei, S.
Nagatomo, T. Kitagawa, S. Fukuzumi, J. Am. Chem. Soc. 2001, 123,
2165–2175; c) H. Ohtsu, S. Fukuzumi, Chem. Eur. J. 2001, 7, 4947–
4953; d) N. Asao, T. Asano, T. Ohishi, Y. Yamamoto, J. Am. Chem.
Soc. 2000, 122, 4817–4818; e) S. Fukuzumi, K. Ohkubo, Chem. Eur.
J. 2000, 6, 4532–4535; f) C. L. Mero, N. A. Porter, J. Am. Chem.
Soc. 1999, 121, 5155–5160; g) M. P. Sibi, J. Ji, J. B. Sausker, C. P. Jas-
perse, J. Am. Chem. Soc. 1999, 121, 7517–7526; h) Y. Yamamoto, S.
Onuki, M. Yumoto, N. Asao, J. Am. Chem. Soc. 1994, 116, 421–422;
i) S. Fukuzumi, N. Satoh, T. Okamoto, K. Yasui, T. Suenobu, Y.
Seko, M. Fujitsuka, O. Ito, J. Am. Chem. Soc. 2001, 123, 7756–7766;
j) S. Fukuzumi, K. Yasui, T. Suenobu, K. Ohkubo, M. Fujitsuka, O.
Ito, J. Phys. Chem. A 2001, 105, 10501–10510; k) S. Fukuzumi, H.
Mori, H. Imahori, T. Suenobu, Y. Araki, O. Ito, K. M. Kadish, J.
Am. Chem. Soc. 2001, 123, 12458–12465; l) K. Mizuno, N. Ichinose,
Y. Otsuji, J. Org. Chem. 1992, 57, 1855–1860; m) T. Tamai, K.
Mizuno, I. Hashida, Y. Otsuji, J. Org. Chem. 1992, 57, 5338–5342;
n) K. Mizuno, K. Terasaka, M. Ikeda, Y. Otsuji, Tetrahedron Lett.
1985, 26, 5819–8522.
[3] For reviews see: a) C. Pac, O. Ishitani, Photochem. Photobiol. 1988,
48, 767–785; b) S. Fukuzumi, T. Tanaka, in Photoinduced Electron
Transfer Part C (Eds.: M. A. Fox, M. Chanon), Elsevier, New York,
1988, pp. 578–635; c) S. Yasui, M. Okamura, M. Fujii, Rev. Heteroat.
Chem. Eur. J. 2005, 11, 4219 – 4226
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4225

O. Ishitani et al.
[27] “Free” 1,4-BNAH and the solvento ruthenium complex were
formed in the cases of low yields of 1,4-BEt2NAH, and total yields
of 1,4-BNAH and 1,4-BEt2NAH were almost 100% in all the cases.
[28] The strong peak observed at > 0.21 V in Figure 4C is attributable
to the oxidation of NEt3.
e) N. Bodor, R. Pearlman, J. Am. Chem. Soc. 1978, 100, 4946–4953;
f) H. Minato, T. Ito, M. Kobayashi, Chem. Lett. 1977, 13–16; g) J.
Ludowieg, A. Levy, Biochemistry 1964, 3, 373–378.
[33] D. H. Gibson, X. Yin, J. Am. Chem. Soc. 1998, 120, 11200–11201.
[34] D. Mauzerall, F. H. Westheimer, J. Am. Chem. Soc. 1955, 77, 2261–
2264.
[35] J. P. Collman, P. S. Wagenknecht, N. S. Lewis, J. Am. Chem. Soc.
1992, 114, 5665–5673.
[36] T. Higashi, Abscor—Empirical Absorption Correction Based on
Fourier Series Approximation, Rigaku Corporation, Tokyo, Japan,
1995.
[29] V. Gutmann, E. Wychera, Inorg. Nucl. Chem. Lett. 1966, 2, 257–
260.
[30] a) A. S. Pietro, N. O. Kaplan, S. P. Colowick, J. Biol. Chem. 1955,
212, 941–952; b) J. B. Jackson, S. A. White, P. G. Quirk, J. D. Venn-
ing, Biochemistry 2002, 41, 4173–4185; c) S. P. Colowick, N. O.
Kaplan, E. F. Neufeld, M. M. Ciotti, J. Biol. Chem. 1952, 195, 95–
105; d) L. Ernster, C. P. Lee, Methods Enzymol. 1967, 10, 729–738.
[31] The mechanism of the regioselective transhydrogenation has not yet
been clarified.
[32] a) P. Eikeren, D. J. Grier, J. Am. Chem. Soc. 1977, 99, 8057–8060;
b) T. J. Von Bergen, T. Mulder, R. A. Van der Veen, R. M. Kellogg,
Tetrahedron 1978, 34, 2377–2383; c) T. J. Von Bergen, T. Mulder,
R. M. Kellogg, J. Am. Chem. Soc. 1976, 98, 1960–1962; d) Y. Oh-
nishi, S. Tanimoto, Tetrahedron Lett. 1977, 18, 1909–1912; corrigen-
dum: Y. Ohnishi, S. Tanimoto, Tetrahedron Lett. 1977, 18, 2918;
[37] M. S. Corporation, TEXSAN, Program for the Crystal Structure,
The Woodlands, TX 77381-5209, (USA), 2000.
[38] G. M. Sheldrick, SHELXS-97, Program for the Crystal Structure,
University of Gçttingen, Gçttingen (Germany), 1997.
Received: November 26, 2004
Revised: March 2, 2005
Published online: May 2, 2005
4226
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 4219 – 4226