Hydride reduction of NAD+ analogues by isopropyl alcohol: kinetics, deuterium isotope effects and mechanism

Article abstract of DOI:10.1021/jo800820u

(Chemical Equation Presented) Observed pseudo-first-order rate constants (k^obs) of the hydride-transfer reactions from isopropyl alcohol (i-PrOH) to two NAD⁺ analogues, 9-phenylxanthylium ion (PhXn ⁺) and 10-methylacridini

Full text of DOI:10.1021/jo800820u

VOLUME 73, NUMBER 13
July 4, 2008
Copyright 2008 by the American Chemical Society
Hydride Reduction of NAD+ Analogues by Isopropyl Alcohol:
Kinetics, Deuterium Isotope Effects and Mechanism
Yun Lu,* Fengrui Qu, Brian Moore, Donald Endicott, and William Kuester
Department of Chemistry, Southern Illinois UniVersity EdwardsVille, EdwardsVille, Illinois 62026
yulu@siue.edu
ReceiVed April 13, 2008
Observed pseudo-first-order rate constants (kobs) of the hydride-transfer reactions from isopropyl alcohol
(i-PrOH) to two NAD+ analogues, 9-phenylxanthylium ion (PhXn+) and 10-methylacridinium ion (MA+),
were determined at temperatures ranging from 49 to 82 °C in i-PrOH containing various amounts of AN
or water. Formations of the alcohol-cation ether adducts (ROPr-i) were observed as side equilibria. The
equilibrium constants for the conversion of PhXn+ to PhXnOPr-i in i-PrOH/AN (v/v ) 1) were determined,
and the equilibrium isotope effect (EIE ) K(i-PrOH)/K(i-PrOD)) at 62 °C was calculated to be 2.67. The
kH of the hydride-transfer step for both reactions were calculated on the basis of the kobs and K. The
corresponding deuterium kinetic isotope effects (e.g., KIEODH ) kH(i-PrOH)/kH(i-PrOD) and KIEꢀ-D6
)
H
kobs(i-PrOH)/kobs((CD3)2CHOH)), as well as the activation parameters, were derived. For the reaction of
PhXn+ (62 °C) and MA+ (67 °C), primary KIER-D (4.4 and 2.1, respectively) as well as secondary
H
KIEODH (1.07 and 1.18) and KIEꢀ-D6H (1.1 and 1.5) were observed. The observed EIE and KIEODH were
explained in terms of the fractionation factors for deuterium between OH and OH+(OHδ+) sites. The
observed inverse kinetic solvent isotope effect for the reaction of PhXn+ (kobs(i-PrOH)/kobs(i-PrOD) )
0.39) is consistent with the intermolecular hydride-transfer mechanism. The dramatic reduction of the
reaction rate for MA+, when the water or i-PrOH cosolvent was replaced by AN, suggests that the hydride-
transfer T.S. is stabilized by H-bonding between O of the solvent OH and the substrate alcohol OHδ+.
This result suggests an H-bonding stabilization effect on the T.S. of the alcohol dehydrogenase reactions.
Introduction
dehydrogenases.1–6 Searching for and understanding the hydride-
transfer reaction from alcohols to NAD+ models has thus
The chemistry of the metabolism of alcohols in living
organisms involves the hydride transfer from the R-H of alcohols
to the C-4 position of the pyridinium ring of the NAD+
coenzyme (1, R1 ) adenine dinucleotide, R2 ) CONH2).
Alcohols are oxidized to ketones or aldehydes, and NAD+ is
reduced to NADH (2, R1 and R2 are same as in NAD+). The
reversible biochemical conversion can be mediated by alcohol
(1) (a) Nagel, Z. D.; Klinman, J. P. Chem. ReV. 2006, 106, 3095–3118. (b)
Ramaswamy, S.; Eklund, H.; Plapp, B. V. Biochemistry 1994, 33, 5230–5237.
(2) Agarwal, P. K.; Webb, S. P.; Hammes-Schiffer, S. J. Am. Chem. Soc.
2000, 122, 4803–4812.
(3) Parkin, G. Chem. ReV. 2004, 104, 699–768.
(4) Holm, R. H.; Kennepohl, P.; Solomon, E. I. Chem. ReV. 1996, 96, 2239–
2314.
(5) Klinman, J. P. Crit. ReV. Biochem. 1981, 10, 39–78.
10.1021/jo800820u CCC: $40.75
Published on Web 06/11/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 4763–4770 4763

Lu et al.
become a target of biomimetic research of the alcohol dehy-
drogenase reaction.7
Thus far, biomimetic investigation of the enzymatic reaction
has used N-alkyl-3-substituted pyridinium cations (1, R2
)
CONH2 or CONR2) as NAD+ models. Since the hydride-transfer
reactions from alcohols to these model compounds are thermo-
dynamically unfavorable, reactive metal alcoholates containing
lithium,8,9 magnesium,10 and zinc11,12 have been used as hydride
sources. These reactions give mainly 1,4-dihydropyridine
products (2), but formation of the 1,6-dihydropyridine isomers
(3) was often observed. The nonenzymatic reactions, some of
which are less effective, are inconsistent with the regiospecific
enzymatic reaction in which only the former is produced.
We recently reported an effective thermal oxidation of
isopropyl alcohol (i-PrOH) via hydride transfer to another NAD+
model, 10-methylacridinium ion (MA+, 4).13 Since the central
pyridinium ring of MA+ is blocked by fused benzene rings,
only one electrophilic center, i.e., the 4-position of the central
ring, is available for the hydride ion donor to attack. Only the
hydride-transfer product, 9,10-dihydroacridan (MAH, 5), was
isolated, and the kinetics of the reaction were determined. It is
believed that this is the first kinetic determination of the hydride
transfer from a neutral alcohol to an NAD+ model. This reaction
can serve as a useful model of the dehydrogenase reaction.
We report herein our detailed kinetic investigation of the
hydride-transfer reaction from i-PrOH to MA+ and to another
more reactive analogue of MA+, 9-phenylxanthylium ion (6,
PhXn+). Like MA+, PhXn+ has a central pyrylium ring fused
with benzene rings on both sides. The effect of several variables
on the rate of the hydride-transfer reduction of both PhXn+ and
MA+ by isopropyl alcohol in either the parent alcohol solvent
or a mixed solvent containing the parent alcohol were studied.
These include the effects of the presence of mono- and
perdeuterated isopropyl alcohol (i-PrOD, i-PrOH-2-d, and
i-PrOD-d7), protic acid, as well as that of varying the temper-
ature. The results including kinetic solvent effects, kinetic
temperature effects, and deuterium kinetic isotope effects at
positions of R-C, ꢀ-C, and alcohol O on the hydride transfer
from R-CH to both cations were used to elucidate the mechanism
of the reactions and to provide insights into the mechanism of
the alcohol dehydrogenase reactions.
cations, study of which has been ignored and undeveloped for
half a century. This kind of reaction last appeared in the literature
in the late 1950s when Bartlett14 and Franzen15 reported a
number of examples of triphenylmethanol and derivatives (as
their cations) abstracting hydride ion from aliphatic alcohols in
50% sulfuric acid aqueous solution. The function of the
concentrated acid was to completely convert the aryl alcohols
to their cationic forms so that they are activated toward hydride
abstraction from the alcohols and a direct monitoring of the
disappearance of the cation was possible for kinetic determina-
tions. However, the acid also deactivates the alcohols by
protonation, complicating the reaction system and thus constitut-
ing the major reason for the stagnant investigation of these
important reactions.16
Results
Reduction of PhXn+ by i-PrOH: Analogue to the MA+
Reduction. PhXn+BF4 (0.096 mmol) in 2 mL of acetonitrile
-
(AN) was added to 10 mL of i-PrOH at room temperature. The
yellow color of the PhXn+ disappeared immediately, implying
that the PhXn+ was consumed. The colorless product was
isolated and characterized to be the corresponding isopropyl
alcohol adduct, 9-isopropoxy-9-phenylxanthene (PhXnOPr-i, 7).
The same reaction solution was subjected to reflux conditions
(82 °C), and the progress of the reaction was monitored by TLC.
As the reaction time increased, PhXnOPr-i gradually disappeared
and another product formed. After about 6 h, PhXnOPr-i was
not detectable. The product was isolated and characterized to
be the hydride-transfer product 9-phenylxanthene (PhXnH, 8).
Reactions in which i-PrOH was replaced by either i-PrOD
or perdeuterated isopropyl alcohol (i-PrOD-d7) were carried out.
Only the latter reaction led to the incorporation of the deuterium
into the 9-position of the 4, indicating that (1) the source of the
transferring hydrogen was the R-H of the alcohol and (2) the
hydrogen did not exchange with the aromatic hydrogens of the
PhXn+, and hence, PhXnH was the primary net hydride-transfer
product. Our results indicate the involvement of an equilibrium
between PhXn+ and its isopropyl alcohol adduct (7) in the
overall irreversible hydride-transfer reaction.
At the same time, this work serves to provide insights into
the general hydride-transfer reaction from alcohols to organic
(6) Pettersson, G.; Eklund, H. Eur. J. Biochem. 1987, 165, 157–161.
(7) Bergquist, C.; Koutcher, L.; Vaught, A. L.; Parkin, G. Inorg. Chem. 2002,
41, 625–627.
The reaction is analogous to the irreversible hydride reduction
of MA+ by i-PrOH13 that we reported earlier. In the latter
reaction, the isopropyl alcohol adduct of MA+ (MAOPr, 9) was
(8) Shirra, A.; Suckling, C.-J. J. Chem. Soc., Perkin Trans. 2 1977, 759–
765.
(9) Ohnishi, Y.; Kitami, M. Tetrahedron Lett. 1978, 4035–4036.
(10) Kanomata, N.; Suzuki, M.; Yoshida, M.; Nakata, T. Angew. Chem., Int.
Ed. 1998, 37, 1410–1411.
(11) Kimura, E.; Shionoya, M.; Hoshino, A.; Ikeda, T.; Yamada, Y. J. Am.
Chem. Soc. 1992, 114, 10134–10137.
(14) Barlett, P. D.; McCollum, J. D. J. Am. Chem. Soc. 1956, 78, 1441–
1450.
(12) Walz, R.; Vahrenkamp, H. Inorg. Chim. Acta 2001, 314, 58–62.
(13) Lu, Y.; Endicott, D.; Kuester, W. Tetrahedron Lett. 2007, 48, 6356–
6359.
(15) Dr. rer. Nat. Franzen, V.; Katonah, Z. Chem.sZtg. 1957, 81, 205–206.
(16) Deno, N. C.; Peterson, H. J.; Saines, G. S. Chem. ReV. 1960, 60, 7–14,
and references cited therein.
4764 J. Org. Chem. Vol. 73, No. 13, 2008

Hydride Reduction of NAD+ Analogues by Isopropyl Alcohol
FIGURE 1. Kinetic scans determined by analysis of the reaction
aliquots, taken from the reaction solution of [PhXn+]0 ) 0.001 M in
i-PrOH/AN (v/v ) 1) at 62 °C, diluted 25 times with (a) pure AN and
(b) AN containing HClO4 (3 M). Reaction times for spectra 1- are
30, 120, 271, 506, and 1402 min, respectively.
also detected. That adduct is not as stable as PhXnOPr-i and
was not isolatable. This is apparently due to the higher stability
of its precursor cation MA+ as compared to PhXn+ with respect
to the corresponding adducts. According to the hydride affinity
values, MA+ is 13 kcal/mol more stable than PhXn+.17 The
formation of MAOPr-i was detected by its UV absorption (at
284 nm) and the NMR spectra in the mixture of MA+ and
i-PrOD-d7.13
Two possible pathways for the involvement of the adducts
in the corresponding irreversible hydride-transfer reactions are
described in eqs 1 and 2 where R+ represents the organic cations.
Note that in the intramolecular hydride-transfer pathway (2),
the adduct (ROPr-i) is an intermediate in the formation of the
hydride-transfer product, while in the intermolecular hydride-
transfer pathway (1) it is involved in a side equilibrium with
its precursor compounds. Also note that, in pathway (2), the
hydride-transfer step requires a four-membered ring transition
state (T.S.).
determining the decay of absorbance due to the cation in aliquots
at various time intervals. Spectral data showed that about 98%
of the PhXn+ was converted to the PhXnOPr-i adduct at 62 °C
upon addition of the cation in i-PrOH/AN (v/v ) 1). Spectra in
Figure 1a correspond with the aliquots diluted with pure
acetonitrile. Note that these spectra specially used for kinetic
analysis are different from those of the real reaction system in
that the PhXn+/PhXnOPr-i equilibrium system is shifted to some
extent to the side of the cation when the solution is diluted with
acetonitrile. The characteristic absorption band with λmax ) 373
nm is due to PhXn+ and that with λmax ) 283 nm is due to
both the PhXnOPr-i adduct and the PhXnH product, which are
known to absorb over the same wavelength range. The absor-
bance band at λmax ) 373 nm is unsuitable for monitoring the
conversion of the cation to PhXnH product. This is due to the
fact that the cation is in equilibrium with PhXnOPr-i. Further-
more, the absorbance band at λmax ) 283 nm is due to both the
PhXnOPr-i adduct and the PhXnH product, which prohibits its
use for monitoring the formation of PhXnH.
Since the equilibrium between the adduct and PhXn+ favors
the formation of the latter under sufficiently acidic conditions,
spectral kinetic scans that reflect the genuine conversion from
PhXn+ to PhXnH were recorded for the reaction aliquots diluted
with acetonitrile containing HClO4 (Figure 1b). The [HClO4]
in AN was 3 M, sufficiently high that the concentration of adduct
was negligible. The observed pseudo-first-order rate constant
(kobs) for the formation of PhXnH was thus computed according
to the decay of PhXn+ at 373nm. Here, kobs ) -d[PhXn+]T/dt
) d[PhXnH]/dt where the total [PhXn+] measured, denoted as
[PhXn+]T, is the sum of [PhXn+] and [PhXnOPr-i] in the
reaction solution.
As stated in the Introduction, previous kinetic studies of the
hydride-transfer reactions from alcohols to carbocations were
carried out in concentrated aqueous acid solutions.14 Our method
allows the determination of the rates of these reactions under
neutral aprotic conditions, avoiding the complications due to
the protonation of the alcohols by concentrated acids.
It should be mentioned that searching for the evidence to
discriminate between the two possible mechanisms for the
general hydride-transfer reaction from alcohols to organic
cations started as early as when these reactions were first
discovered. However, an answer to this mechanistic question
had never been found.14,16 For example, although Bartlett and
co-workers studied the kinetics of the hydride-transfer reactions
from aliphatic alcohols to carbocations derived from the
corresponding arylcarbinols by treating them as intermolecular
reactions, they concluded that “the kinetic data cannot exclude
the possibility that the actual hydride transfer which is preferred
in these systems is a intramolecular one between the two
R-positions of the mixed ether ion.”14
Determination of the Rates of the Hydride-Transfer
Reaction from i-PrOH to PhXn+ and to MA+: Method,
Kinetic Solvent Effect, Kinetic (Solvent) Isotope Effect,
and ꢀ-Secondary Kinetic Isotope Effect. The UV-vis spectral
kinetic scans were recorded for the reaction of PhXn+ (0.001M)
in i-PrOH/AN (v/v ) 1) over a range of temperatures by
Data illustrating the effect of mono- and perdeuterated
isopropyl alcohol (i-PrOD, i-PrOH-2-d, and i-PrOD-d7) (entries
1-2 to 1-5), temperature (entries 1-1, 1-2, 1-6, and 1-7) and
solvent (entries 1-8 and 1-9) on the kobs of the reaction are
summarized in Table 1. When AN solvent was replaced by
i-PrOH or the nonpolar solvent cyclohexane, no hydride-transfer
(17) The hydride affinities (-∆GH-(R+)) of the two cations are 76 kcal/mol
(MA+) and 89 kcal/mol (PhXn+), respectively, see: (a) Cheng, J. -P.; Lu, Y.;
Zhu, X.; Mu, L. J. Org. Chem. 1998, 63, 6108–6114. (b) Cheng, J. -P.; Handoo,
K. L.; Parker, V. D. J. Am. Chem. Soc. 1993, 115, 2655–2660.
J. Org. Chem. Vol. 73, No. 13, 2008 4765

Lu et al.
TABLE 1. Effect of D-Substitution in Isopropyl Alcohol, the
Temperature, and the Solvent on the Observed Pseudo-First-Order
Rate Constants for the Hydride-Transfer Reaction from Isopropyl
compared to the reaction in i-PrOH containing 4.7% water (v/
v) (entries 1-13 vs 1-10). In order to further understand the
observed kinetic solvent effect, and because we have the rate
data (entry 1-15) of the reaction in i-PrOH containing 4.7%
water at 82 °C in hand for comparison,13 the rate of the reaction
in i-PrOH containing 4.7% AN at 82 °C was determined (entry
1-14). A reduction in rate (1.4-fold) resulted from the change
of the cosolvent from water to AN was observed as well.
The method used to determine the rate constants of the MA+
reaction was the same as that used in the kinetic determinations
of the PhXn+ reaction described above. The decay of absorbance
with time at 358 nm due to MA+ was recorded and used to
determine the rate of the reaction.
a
Alcohol to PhXn+ and MA+
b
entry
T (°C)
reaction conditions
reduction of PhXn+ (v/v ) 1c)
i-PrOH/AN
i-PrOH/AN
i-PrOH-2-d/AN
i-PrOD/AN
i-PrOD-d7/AN
i-PrOH/AN
i-PrOH/AN
i-PrOH/cyclohexane
i-PrOH
kobs × 103 (min-1
)
1-1
1-2
1-3
1-4
1-5
1-6
1-7
1-8
1-9
69
62
62
62
62
56
48
62
62
4.49 ( 0.08
2.13 ( 0.09
0.480 ( 0.016
5.53 ( 0.14
1.11 ( 0.01
1.11 ( 0.02
0.755 ( 0.069
n.r.d
n.r.d
The Effect of [H+] on the Kinetics of the Reactions of
PhXn+ and MA+: An Opposite Effect. Since acid H+ is
involved in both of the assumed pathways (1) and (2), an
examination of the hydride-transfer mechanism for the reactions
of both PhXn+ and MA+ would need to take into account the
effect of [H+] upon the kinetics of the reactions. Since acid
favors the conversion of the adduct to the free cation, it is
expected to facilitate pathway (1) but to inhibit pathway (2).
This is apparent from the rate equations (3) and (4) for the
formation of the hydride-transfer product (R-H) derived from
reduction of MA+ (in alcohol containing 4.7% H2Oe)
1-10
1-11
1-12
1-13
1-14
1-15
67
67
67
67
82
82
i-PrOH
i-PrOH-2-d
i-PrOD
i-PrOH/AN (v/v )1)
i-PrOH/4.7% AN (v/v)
i-PrOH
0.711 ( 0.013
0.341 ( 0.063
0.602 ( 0.025
0.134 ( 0.018
1.28 ( 0.21
1.74 ( 0.17f
a [PhXn+]0
)
1
mM; [MA+]0
)
0.186 mM. b Based on three
no hydride-transfer
determinations. c For mixed solvents. d n.r.
)
reaction; only adduct production was observed. e v/v, unless otherwise
indicated; Or D2O for the use of i-PrOD. Note that the initial
concentration of MA+ (0.186 mM) is different from that (0.398 mM)
used in our previous paper.13 f Reference.13
pathways (1) and (2), respectively. Note that eqs 3 and 4 are
+
only suitable when the protonation of i-PrOH to form i-PrOH2
,
which does not undergo hydride-transfer reaction, is negligible.
The rate of the reaction (1) increases with increasing [H+] until
[H+] . K [i-PrOH] conditions under which the rate of the
reaction no longer changes with [H+] and kobs is equal to the
rate constant of the hydride-transfer step (kH), while that of the
reaction (2) is expected to decrease with increasing [H+].
reaction was observed. Spectral studies indicate that under these
conditions PhXn+ appears to be completely converted into the
corresponding isopropyl alcohol adduct PhXnOPr-i.
Observed kinetic isotope effects (KIEobs) at 62 °C were
calculated and are listed in Table 2. The KIER-Dobs in first column
reflects the kinetic isotope effect of the hydride-transfer process
since the adduct formation equilibrium is not affected by the
presence of D at the 2(or R)-position of the alcohol. The inverse
KIE (row 1, column 2) represents the solVent KIE since i-PrOH
d[R - H] ⁄ dt ) -d[R+]T ⁄ dt
) kH[R+]T[i-PrOH][H+] ⁄ ([H+] +
K[i-PrOH])
obs
acts as both reactant and cosolvent. The inverse KIEOD is
) kobs[R+]T[i-PrOH]
d[R - H] ⁄ dt ) -d[R+]T ⁄ dt
(3)
unlikely to originate from the facile O-H bond cleavage, with
which the rate-determining C-H bond breaking cannot compete.
The observed solvent KIEODobs should involve the isotope effect
on the adduct formation equilibrium which involves O-H bond
cleavage to form solvated H+ (see the following paragraphs and
the Discussion section). The last KIED8obs (row 1, column 3) is
an overall value derived from the reaction of perdeuterated
isopropyl alcohol. With the assumption that KIE is multiplica-
tive, the ꢀ-secondary KIE observed upon deuteration of the two
CH3 positions of i-PrOH during the PhXn+ reaction was
calculated to be equal to 1.1 (row 1, column 4). Note that this
KIE only applies to the hydride-transfer step since the presence
of CD3 groups would not be expected to affect the adduct
formation equilibrium.
) kHINTRA[R+]T[i-PrOH]⁄([H+] + K[i-PrOH]) (4)
The effects of added [H+] () [HBF4]) on the rate of the
reaction of PhXn+ in i-PrOH/AN (v/v ) 1) and on that of MA+
in i-PrOH containing 4.7% of water (v/v) were determined
(Table 3, entries 3-1-3-2 for the PhXn+ reaction, entries
3-6-3-8 for the MA+ reaction). The effect of the [H+] on the
rate of the PhXn+ reaction in i-PrOH containing 2.5% water
(v/v), whose rate was not measurable under neutral conditions,
was also determined (entries 3-3-3-5). Since under the condi-
tions studied the concentration range of the H+ added is limited,
a thorough investigation of the acid effect on the kinetics of
the reaction is restricted. The current data, however, clearly show
that an increase of [H+] accelerates the hydride-transfer reaction
from i-PrOH to PhXn+ but inhibits the rate of the reaction to
MA+.
Equilibrium Constant (K), Thermodynamic Parameters,
and the Equilibrium Isotope Effect (EIE). In order to
understand the effect of the adduct PhXnOPr-i formation
equilibrium on the overall rate constant for the corresponding
reaction (kobs) and on the observed inVerse KIEODobs, the
equilibrium constants (K) defined in eqs 1 and 2 at different
temperatures in i-PrOH (or i-PrOD)/AN (v/v ) 1) were
determined using a spectral method based on the extinction
We have reported a deuterium KIED8obs of 3.8 for the hydride
reduction of MA+ at 67 °C in i-PrOH (or i-PrOD-d7) containing
4.7% (v/v) H2O (or D2O).13 We have now determined the rates
of the reaction with i-PrOH-2-d and with i-PrOD in the same
reaction medium at 67 °C (Table 1, entries 1-10-1-12). The
obs
obs
KIER-D ) 2.1 and KIEOD ) 1.18 were derived, and the
ꢀ-secondary kinetic isotope effect (KIEꢀ-D6obs) is thus calculated
to be 1.5 (Table 2, in row 2). Note that here the corresponding
obs
solvent KIE value (KIEOD ) 1.18) is normal rather than
inverse (0.39) as was observed in the PhXn+ reaction.
The rate constant of the reaction of i-PrOH with MA+ in
i-PrOH/AN (v/v ) 1) at 67 °C was also determined (entry 1-13).
The rate of the reaction is dramatically diminished (5.3-fold)
4766 J. Org. Chem. Vol. 73, No. 13, 2008

Hydride Reduction of NAD+ Analogues by Isopropyl Alcohol
a
TABLE 2. Kinetic Isotope Effects on the Hydride-Transfer Reactions from i-PrOH to PhXn+ and MA+
b
KIEꢀ-D6 b, c(i-PrOH vs (CD3)2CHOH)
obs
obs
obs
obs
KIER-D
(i-PrOH vs i-PrOH-2-d)
KIEOD b(i-PrOH vs i-PrOD)
reduction of PhXn+ in i-PrOH/AN (v/v ) 1) at 62 °C
0.39 1.9
reduction of MA+ in i-PrOH containing 4.7% water at 67 °Cd
1.18
3.8e
KIED8 b(i-PrOH vs i-PrOD-d7)
4.4
2.1
1.1
1.5
a [PhXn+]0 ) 1 mM; [MA+]0 ) 0.186 mM. b Definitions of KIE can be found in the context. c Equals KIED8obs/[(KIER-Dobs)(KIEODobs)] d Note that the
initial concentration of the MA+ (0.186 mM) is different from that (0.398 mM) used in our previous paper.13 e From ref 13.
TABLE 3. Effect of Added [H+] on the Observed
would be reasonable to assume that the equilibrium between
MA+ and its isopropyl alcohol adduct would have an equilib-
Pseudo-First-Order Rate Constants for the Hydride-Transfer
a
Reaction from i-PrOH to PhXn+ and MA+
rium constant (K) of a similar order of magnitude.
b
entry
reaction conditions
reduction of PhXn+
kobs x103 (min-1
)
Intermolecular Mechanism (1) of the Hydride Reduc-
tion of PhXn+. It appears clear that all of the results obtained
from the reaction of PhXn+ with i-PrOH are consistent with
the intermolecular reaction pathway (1).
in i-PrOH/AN (v/v ) 1) at 62 °Cc
3–1
3–2
[H+] ) 0 M
[H+] ) 0.001 M
2.13 ( 0.09d
7.13 ( 0.01
in i-PrOH containing 2.5% water (v/v) at 62 °Ce
Hydride-transfer reaction was not observed in pure i-PrOH
and in i-PrOH/cyclohexane (v/v)1) where all the PhXn+
appeared to be converted to the adduct. The same result was
observed when basic K2CO3 was added to the i-PrOH/AN
system. The presence of the base facilitates the formation of
the adduct. All these results suggest that the presence of the
free cation is a prerequisite for the hydride-transfer reaction to
occur. The function of the polar solvent, AN, may be to enhance
ionization of the adduct to the PhXn+ cation. This is also
consistent with the observed catalytic effect of H+ on the
reaction. The greater than unity EIE value (2.67) and the inverse
3–3
3–4
3–5
[H+] ) 0.0098 M
[H+] ) 0.098 M
[H+] ) 0.196 M
0.264 ( 0.001
8.98 ( 0.38
34.60 ( 0.30
reduction of MA+
in i-PrOH containing 4.7% water (v/v) at 82 °Cf
3–6
3–7
3–8
[H+] ) 0 M
1.74 ( 0.17g
1.18 ( 0.14
0.574 ( 0.006
[H+] ) 0.050 M
[H+] ) 0.196 M
a [H+] ) [HBF4]; note that it is the concentration of the H+ added
rather than the real [H+] in the reaction system. b Based on two
c
determinations unless otherwise indicated. [PhXn+]0 ) 1 mM. d Same
e
f
as entry 1-2. [PhXn+]0 ) 0.398 mM. [MA+]0 ) 0.398 mM. g Same
obs
KIEOD (0.39) indicate that more free cation exists in the
as entry 1-15.
i-PrOD in which the reaction rate is higher, lending strong
support for mechanism (1) (also see the Discussion section).
TABLE 4. Equilibrium Constants (K), Percentage Conversion,
and Equilibrium Isotope Effect (EIE) for the Equilibration of
PhXn+ with PhXnOPr-i in i-PrOH (or i-PrOD)/AN (v/v ) 1)
The KIER-Dobs (4.4) observed for the hydride-transfer process
in the reduction of PhXn+ at 62 °C is more likely consistent
with the intermolecular mechanism that has a linear or slightly
bent T.S. In general, KIE is lower in a hydrogen transfer reaction
with a bent T.S. than that with a linear T.S.19 The intramolecular
mechanism requires a severely bent four-membered ring T.S.,
implying that a smaller KIE would be observed. The observed
KIER-Dobs of 4.4 at 62 °C renders the latter mechanism less likely.
It is of interest to compare the magnitude of the observed value
with the KIE (4.1 at 52 °C) for the intermolecular hydride
transfer from MAH (5) to 1-benzyl-3-cyanoquinolinium cation
in acetonitrile recently reported by us.20
entry
T (°C)
K
% conversion
EIEa
4–1
4–2
4–3
4–4
69
62
56
49
0.00881
0.0119
0.0108
0.0123
98
98
99
99
2.67
a EIE ) K(i-PrOH)/K(i-PrOD).
coefficient of the PhXn+ determined (Table 4). Interestingly, K
does not change significantly within the temperature range from
49 to 69 °C. Using the van’t Hoff equation, a small negative
standard enthalpy change (∆H° ) -12.7 kJ/mol) and a negative
standard entropy change (∆S° ) -76 J/mol K) were derived.
The large negative ∆S° may be due to the formation of the
adduct and the well-solvated proton in the equilibrium, which
are expected to possess fewer degrees of freedom. Under the
reaction conditions, the degree of conversion of the PhXn+ to
the adduct was calculated to be about 98%. The equilibrium
isotope effect (EIE ) K(i-PrOH)/K(i-PrOD)) was calculated to
be 2.67 (Table 4), indicating that under these conditions the
equilibrium concentration of the PhXn+ is higher in i-PrOD than
in i-PrOH.
The degree of conversion of MA+ to MAOPr-i under the
reaction conditions, i.e., in i-PrOH containing 4.7% of water,
was too low to obtain an equilibrium constant. The equilibrium
constant for the interconversion between 9-phenyl-10-methyl
acridinium ion and its methanol adduct in methanol has been
previously reported to be as small as 1.5 × 10-8 at 22 °C.18 It
The observed normal ꢀ-secondary KIE (KIEꢀ-D6obs ) 1.1) is
consistent with the intermolecular hydride-transfer reaction
mechanism with a T.S. in which positive charge is developing
on the R-C-OH group.21 This type of KIE is believed to
originate from the T.S. stabilization effect caused by the
hyperconjugation between the ꢀ-C-H σ orbital and the p/or π
orbital of the R-C
OH group developing the positive
charge.21,22 Compared to the C-D bond, the more easily
weakened C-H bond provides better orbital overlap for
21,22
hyperconjugation, and promotes a normal KIE.
However
(19) More O’Ferrall, R. A., J. Chem. Soc. B 1970785, and references cited
therein
(20) Lu, Y.; Zhao, Y.; Handoo, K. L.; Parker, V. D. Org. Biomol. Chem.
2003, 1, 173–181.
(21) (a) Thompson, M. S.; Meyer, T. J. J. Am. Chem. Soc. 1982, 104, 4106–
4115. (b) Angelis, Y. S.; Hatzakis, N. S.; Smonou, I.; Orfanopoulos, M.
Tetrahedron Lett. 2001, 42, 3753–3756.
(22) (a) Jewett, J. G.; Dunlap, R. P. J. Am. Chem. Soc. 1968, 90, 809–810.
(b) Kresge, A. J.; Preto, R. J. J. Am. Chem. Soc. 1967, 89, 5510–5511. (c)
Karabatsos, G. J.; Sonnichsen, G. C.; Papaioannou, C. G.; Scheppele, S. E.;
Shone, R. L. J. Am. Chem. Soc. 1967, 89, 463–465. (d) Deraniyagala, S. A.;
Adediran, S. A.; Pratt, R. F. J. Org. Chem. 1995, 60, 1619–1625.
(18) Zhou, B.; Kano, K.; Hashimoto, S. Bull. Chem. Soc. Jpn. 1988, 61,
1633–1640.
J. Org. Chem. Vol. 73, No. 13, 2008 4767

Lu et al.
TABLE 5. Pseudo-First-Order Rate Constants (kH) for the
Intermolecular Hydride-Transfer Step in the Reaction of PhXn+ in
i-PrOH(D)/AN (v/v ) 1) as a Function of Temperaturea
dehydrogenase. In addition to this reactivity, alcohols also
function as nucleophiles when treated with NAD+ analogues
in less selective nonenzymatic reactions. The latter reactivity
of the alcohol results in the possible formation of the alcohol
adduct of the NAD+ models. In this study, the adducts
(PhXnOPr-i and MAOPr-i) were formed in side equilibria,
complicating the study of the dehydrogenase model reactions
between isopropyl alcohol and the two NAD+ analogues PhXn+
and MA+. All of the results are consistent with the intermo-
lecular hydride-transfer mechanism (1) (from i-PrOH to the
cations). The alternative intramolecular hydride-transfer mech-
anism (2) requires a four-membered ring T.S., rendering the
mechanism less likely. The adduct PhXnOPr-i is isolatable but
readily decomposed into its precursor cation in polar solvents
or under low acid concentration conditions. In contrast, the
adduct MAOPr-i is extremely unstable such that the formation
of this adduct is insignificant even under the conditions where
i-PrOH is the major solvent component (i-PrOH containing 4.7%
b
H
entry Reaction conditions [H+]eq (mM) kH (min-1
)
KIEOD
5–1
5–2
5–3
5–3
5–4
69 °C (i-PrOH)
62 °C (i-PrOH)
62 °C (i-PrOD)
56 °C (i-PrOH)
49 °C (i-PrOH)
0.982
0.979
0.968c
0.987
0.988
0.250
0.196
0.184
0.0835
0.0595
1.07
a
c
[PhXn+]0 ) 1 mM. b KIEOD ) kH(i-PrOH)/kH(i-PrOD). [D+]eq.
H
the KIE value cannot be used to distinguish between the two
mechanisms (1) and (2) since hyperconjugation is also present
involving the ꢀ-C-H σ orbital and the p/or π orbital of the
developing polar R-C
O group in the T.S. of the concerted
intramolecular hydride-transfer process (2). This analysis also
applies to the reaction of MA+ in which a normal ꢀ-secondary
obs
KIE (KIEꢀ-D6 )1.5) was observed as well.
obs
H
Mechanism of the Hydride Reduction of MA+ and the
Relevant kH and KIEODH. It is reasonable to suggest that the
hydride-transfer reduction of the more stable MA+ follows the
intermolecular mechanism (1) as well. An examination of the
data in Table 3, however, shows that kobs decreases as [H+]
increases, opposite to the effect observed for the PhXn+ reaction.
This, at first sight, conflicts with mechanism (1). However, it
is likely that the side equilibrium to form the unstable MAOPr-i
adduct is insignificant in the overall reaction so that the
introduction of acid lowers the rate of the reaction by H-bonding
to and/or protonation of i-PrOH, lowering the hydride donation
ability of i-PrOH.
of water or AN). The observed primary KIER-D ()KIER-D
)
for both reactions indicates that the hydride-transfer step is rate-
determining. The observed ꢀ-secondary kinetic isotope effect
obs
(KIEꢀ-D6 ) KIEꢀ-D6H) is consistent with a hydride-transfer
T.S. in which positive charge has developed on the R-C-OH
group of the alcohol moiety. The net hydride-transfer reaction
of PhXn+ (kH) is faster than that of the reaction of MA+ (kH )
kobs) by ∼3 orders of magnitude. For example, the former
reaction in i-PrOH/AN (v/v ) 1) at 69 °C is about 1800 times
faster than that of the latter in the same medium at 67 °C (entry
5-1 vs entry 1-13).
Kinetic solvent isotope effects have played an important role
in determining the acid-catalyzed reaction mechanisms.23 Many
studies of the kinetic effect brought about by a change of solvent
from H2O to D2O have been reported.24,25 Generally, reactions
involving a pre-equilibrium formation of the conjugate acid
(XH+) of the substrate (X) (eq 5) were found to be faster in
D2O than in H2O, resulting in inverse kinetic solvent isotope
effects. This has been explained in terms of the known greater
acid dissociation constants of weak acids in H2O than in D2O,
resulting in higher equilibrium concentrations of the conjugate
acid of the substrate in D2O.25,26 The observations are limited
to the condition that the subsequent rate-determining step does
not involve a proton or deuteron transfer, which accompanies
a primary KIE that masks the equilibrium isotope effect
contribution to the observed KIE.27 In this study, a change of
solvent/reactant from i-PrOH to i-PrOD resulted in a rate
increase for the hydride-transfer reaction from isopropyl alcohol
to PhXn+. The latter is consistent with the kinetic effect for a
change of solvent from H2O to D2O in the acid-catalyzed
reactions (5), lending strong support for the intermolecular
The very small K value results in K[i-PrOH] , [H+] and
rate eq 3 then reduces to; d[MAH]/dt ) –d[MA+]/dt )
kH[MA+][i-PrOH] ≈ kobs[MA+][i-PrOH]. This leads to the
suggestion that in this case, kobs can be considered to be the
rate constant for the hydride-transfer process from i-PrOH to
MA+, i.e., kobs ) kH. Therefore, the kinetic isotope effect for
the hydride-transfer step accompanying deuteration at the OH
group of the alcohol is KIEOD ) kH(i-PrOH)/kH(i-PrOD) )
H
kobs(i-PrOH)/kobs(i-PrOD) ) KIEOD ) 1.18. The activation
obs
parameters for the hydride-transfer step have been reported for
this reaction under similar conditions.13
Derivation of kH, KIEODH, and Activation Parameters
for the PhXn+ Reaction. The rate constants for the hydride-
transfer step (kH(i-PrOH) and kH(i-PrOD)) in the PhXn+ reaction
at different temperatures in i-PrOH(D)/AN (v/v ) 1) were
calculated by transforming the corresponding kobs, K, [i-
PrOH(D)], and [H(D)+]eq into kobs ) kH[H+]/([H+] + K[i-
PrOH]) derived from the rate eq (3) (Table 5). The [H(D)+]eq
is expected to remain constant throughout most of the reaction
according to the stoichiometric eq 1. Using the Arrhenius and
Eyring equations, the activation parameters of the hydride-
hydride-transfer mechanism (1). Our observation of the inverse
obs
KIEOD
(0.39) can be explained in terms of the greater
transfer process were derived; EaH ) 71.5 kJ/mol, ∆HqH ) 68.6
equilibrium constant (K(i-PrOH)) for formation of the adduct
between PhXn+ and i-PrOH as compared to that involving
i-PrOD (K(i-PrOD)) so that the cation necessary for the hydride-
transfer reaction is present in higher concentration in the i-PrOD
system. Although isotopic substitution of the alcohol O-H
results in a normal isotope effect on the subsequent hydride-
kJ/mol, ∆SqH ) -57.3 J/mol ·K. The corresponding KIEOD
)
H
kH(i-PrOH)/kH(i-PrOD) was calculated to be 1.07, a normal
secondary KIE that reflects the O-D KIE on the hydride-transfer
H
step. The KIE value is consistent with the secondary KIEOD
of 1.18 for the hydride-transfer reaction of MA+ (see the
preceding section).
(23) (a) Kresge, A. J. Pure Appl. Chem. 1964, 8, 243–258. (b) Gold, V.
Pure Appl. Chem. 1963, 7, 141–143. (c) Mitton, C. G.; Gresser, M.; Schowen,
R. L. J. Am. Chem. Soc. 1969, 91, 2045–2047.
Discussion and Conclusions
(24) Nelson, W. E.; Bulter, J. A. V. J. Chem. Soc. 1938, 957.
(25) Wiberg, K. B. Chem. ReV. 1955, 55, 713–743, and references cited
therein.
Alcohols containing R-H act specifically as hydride donors
when oxidized by NAD+ coenzyme mediated by alcohol
4768 J. Org. Chem. Vol. 73, No. 13, 2008

Hydride Reduction of NAD+ Analogues by Isopropyl Alcohol
H
transfer step, the corresponding secondary KIEOD (1.07) is
of the positive charge developing in the T.S. The O-D isotope
effect (resulting from the direct conjugation effect) on the
hydride-transfer process is expected to be greater than the
ꢀ-C-D isotope effect (resulting from the hyperconjugation
effect). This seems to be the case, at least in our systems. Note
masked by the EIE (2.67) of the side-equilibrium in the overall
obs
KIE of the reaction (KIEOD ) KIEODH/EIE).
X + H+ y z XH+ slow8 products
(5)
rapid
H
that the KIEꢀ-D6 values obtained in this work are a result of
the cumulative effect of six ꢀ-C-D bonds.
H
The EIE and KIEOD can be explained in terms of fraction-
One feature of interest is the rate reduction of the MA+
reaction (∼1.4-fold) when the cosolvent was changed from water
(4.7%) to AN (entries 1-14 vs 1-15). When AN was increased
to 50% (v/v), accompanied with a 2-fold decrease in the alcohol
concentration, the rate of the reaction decreased by as large as
5.3-fold (entries 1-10 vs 1-13). Furthermore, in AN without
i-PrOH as cosolvent, for example, [i-PrOH] ) 0.04 M in AN,
no reaction was observed over a period of days. This strongly
suggests that the protic solvents, either i-PrOH or water, strongly
stabilize the T.S. of the hydride-transfer process through
H-bonding interactions with partial positively charged H of the
alcohol moiety (Scheme 1). on the OH group of the alcohol
moiety is highly developed in the T.S. Such a late T.S. is
ation factors (ꢁ) for deuterium between ROH and ROH2+ sites
and between ROH and its hydride-transfer T.S. R
OHδ+
sites, respectively.28 In these systems, ꢁ can be defined to be
the D/H ratio in positively charged OL+ or OLδ+ bonds (L )
H or D) divided by the D/H ratio in neutral OL bonds, i.e., ꢁ
28,29
+
) (D/H)OL /(D/H)OL
.
A ꢁ ) 0.69, determined for the
hydronium ion in aqueous solution, has been commonly assigned
to all positively charged OL+ bonds.28,29 The less than unity ꢁ
value suggests a weaker O-L+ bond than is an O-L bond. It
also means that, in a deuterated solvent at room temperature,
D has only 69% preference compared with H in the undeuterated
solvent for placement in OL+. In another sense, it also reflects
that H2O (or ROH) is a stronger base than is D2O (or ROD).
Thus, considering the equilibrium PhXn+ + 2 i-PrOL S
PhXnOPr-i + i-PrOL2+, and assuming that D and H have equal
opportunity of being in i-PrOL, the EIE (K(i-PrOH)/K(i-PrOD))
of the equilibrium can be estimated to be 12/(0.69)2 ) 2.1, not
far from the value of 2.67 determined from this work. Note
that in our equilibrium system L+ may exist in a form other
than i-PrOL2+, e.g. (i-PrOL)nL+, plus our experimental condi-
tions (60 °C in i-PrOH/AN (v/v ) 1)) are different from those
used to obtain the 0.69 value, a quantitative comparison between
the two EIE values (2.1 and 2.67) may thus be meaningless.
H
consistent with the observed small primary KIER-D (2.1) for
the reaction of MA+ and the relatively small normal secondary
H
H
KIEꢀ-D6 ) 1.5 for the same reaction as well as KIEꢀ-D6
)
1.1 for the reaction of PhXn+. Note that a ꢀ-secondary KIE as
large as 3.5 has been observed in the oxidation of isopropyl
alcohol by a ruthenium(IV) complex.21a
SCHEME 1. Proposed Hydride Transfer T.S. Stabilized by
i-PrOH through H-Bonding
H
Nevertheless, our larger than unity EIE and KIEOD values
correspond with the equilibrium process from OL to OL+ and
the activation process from OL to [OLδ+]q, respectively. Also,
that KIEODH is masked by EIE in the overall KIE of the reaction
may be relevant to the fact that the O-L bond in [OLδ+ q
]
(carrying a partial positive charge) is stronger than that in OL+
(bearing a full positive charge). Importantly, to the best of our
knowledge, this is the first observation of KIEODH in the hydride-
transfer reaction involving alcohols as hydride donors. It is
highly likely that the KIEOD should also be observed in other
types of reactions in which the T.S.’s contain a center of partial
positive charge R to a hydroxyl group.
This T.S. structural pattern can be used to understand the
unexpected relatively small rate ratio for the hydride-transfer
reaction of PhXn+ and MA+ (e.g., ∼1.8 × 103 in i-PrOH/AN
30
(v/v)1) at 69 °C), in spite of the large (∼13 kcal/mol
)
difference in thermodynamic stability of the two cations.
Differences in the degree of H-bonding stabilization effect on
the T.S.’s most likely diminish the difference in T.S. energies
for the two hydride-transfer processes. This is interestingly quite
consistent with the facile biological NAD+ oxidation of alcohols
where alcohol hydroxyl hydrogen apparently interacts favorably
in the T.S. by H-bonding with the hydroxyl oxygen of an amino
acid residue (e.g., serine) in the active site of alcohol
dehydrogenases.1b,2 Our results reflect the significance of the
kind of base catalysis in lowering the T.S. energy of the
biological hydride-transfer processes.
H
H
The observed KIEꢀ-D6 and KIEOD are worthy of further
comment. The results show that both ꢀ-C-D and O-D
connected to the R-C of the alcohol exert isotope effects on the
rate of the hydride transfer from R-C-H to the cations. The
former is a result of a hyperconjugation effect involving the
σ orbital of the ꢀ-C-H bond and the p/or π orbital of the
R-C
OH group where a partial positive charge is developing
in the T.S., altering the ꢀ-C-H bond vibration frequency. The
latter is a result of direct resonance effect involving a partial
positive charge being delocalized onto the O-H group of the
It should be mentioned that the hydride reduction of an NAD+
model BNA+ (1, R1 ) benzyl, R2 ) CONH2) by alkyl alcohols
including i-PrOH catalyzed by various Zn(II) complexes has
been studied and found to be inefficient.7,11,12 The addition of
the Zn(II) complexes was meant to imitate the complexed zinc
environment in the active site of the alcohol dehydrogenases.
T.S. through the C
O bond formation, affecting the vibration
frequency of the O-H bond in it. Both effects are a consequence
(26) (a) Bunton, C. A.; Shiner, V. J., Jr J. Am. Chem. Soc. 1961, 83, 42–47.
(b) Bunton, C. A.; Shiner, V. J., Jr J. Am. Chem. Soc. 1961, 83, 3207–3214, and
references cited therein.
(27) Bunton, C. A.; Shiner, V. J., Jr J. Am. Chem. Soc. 1961, 83, 3214–
3220.
(30) In addition to using the hydride affinity values of the two cations from
ref 17, the stability difference of the two cations can also be estimated from
their pKR+ values: pKR+(PhXn+) ) 0.81: Arnett, E. M.; Flowers, R. A.; Lundwig,
R. T.; Meekhof, A. E.; Ealek, S. A. J. Phys. Org. Chem. 1997, 10, 499-513,
and pKR+(MA+) ) 10.1: Bunting, J. W.; Chew, V. S. F.; Abhyankar, S. B.;
Goda, Y. Can. J. Chem. 1984, 62, 351-354.
(28) Alvarez, R.; Schowen, R. T. Isotopes in Organic Chemistry; Elsevier:
New York, 1987; Vol. 7, Chapter 1, and references cited therein.
(29) Arrowsmith, C. H.; Guo, H.-X.; Kresge, A. J. J. Am. Chem. Soc. 1994,
116, 8890–8894.
J. Org. Chem. Vol. 73, No. 13, 2008 4769

Lu et al.
153-155 °C; 1HNMR (δ, CD3Cl) 6.95-7.45 (m, 13H), 3.42-3.55
(septet, 1H), 0.73-0.82 (d, 6H).
The Lewis acid Zn(II) complexes are believed to catalyze the
hydride transfer from alcohol to NAD+ by the formation of a
reactive zinc alkoxide intermediate through Lewis acid catalyzed
deprotonation of the alcohol.1–6 Our observation of the hydride
reduction of MA+ by isopropyl alcohol, in which the adduct
formation is insignificant, provides a model reaction for possible
further study of the role of the Zn(II) complexes in the biological
hydride-transfer reaction.
In conclusion, the kinetics, the deuterium isotope effects and
the mechanism of the hydride-transfer reaction from isopropyl
alcohol to two stable organic cations (NAD+ analogues) were,
for the first time, thoroughly studied in the parent alcohol
medium containing acetonitrile or water. Both C-D and O-D
connected to the R-C of the alcohol molecule exert isotope
effects on the rates of the hydride transfers from the R-C-H to
the cations. The observed inVerse kinetic solvent isotope effect
excludes the alternative intramolecular mechanism (2) for the
hydride-transfer reaction. This model study also imitates the
H-bonding stabilization effect on the T.S. of the alcohol
dehydrogenase reaction. These reactions can be useful alcohol
dehydrogenase model reactions, thus helping us understand this
important biological hydride-transfer reaction and, further, the
function of the Zn(II) ion in the active site of the alcohol
dehydrogenases.
Formation of the hydride-Transfer Product of PhXn+
(PhXnH, 8). Instead of the separation of the adduct, the above
reaction solution was subjected to reflux conditions (∼82 °C). After
about 6 h, TLC results showed that the adduct PhXnOPr-i
disappeared and the hydride-transfer product PhXnH was formed.
The reaction solution was allowed to reflux for another 2 h, and
the i-PrOH and AN were removed in a rotary evaporator. Extraction
of the residue with CH2Cl2 gave PhXnH with a quantitative yield:
mp 140-142 °C (lit.33 mp 145 °C); 1HNMR (δ, CD3Cl) 6.94-7.32
(13H), 5.22-5.28 (s, 1H).
Determination of the Equilibrium Constant (K). The equilib-
rium constant (K) of the PhXnOPr-i formation equilibration was
determined spectroscopically (UV-vis) by immediately recording
the absorbance at 373 nm (ꢀ ) 30503) attributable to PhXn+ upon
adding various concentration of PhXn+ to i-PrOH/AN (v/v ) 1) at
different temperatures. Two parallel determinations were performed.
General Kinetic Determination Procedure. A certain amount
of the stock solution of PhXn+ in AN and MA+ in water or in AN
was added to 8 mL i-PrOH containing AN or water in a well sealed
10 mL reaction vial thermostatted in a water bath of certain
temperature (<70 °C). About 300 µL of the reaction aliquots were
periodically taken into a sample vial precooled in an ice-bath. The
samples were kept in a freezer (∼-20 °C) until six to eight reaction
aliquots within 1-3 half-lives of the reaction were collected. The
aliquots were analyzed by diluting certain amount of them in AN
containing 3 M HClO4 (for the reaction of PhXn+) and in AN
containing 30 mM of the Lewis acid Zn(OTf)2 (for the reaction of
MA+) and determining the corresponding UV spectra. The absor-
bance (Abs) decays with time at 373 nm (PhXn+) or at 358nm
(MA+) were recorded. The obtained Abs-t data were fitted into
the first-order integrated rate equation, -ln(Abs) ) kt + constant,
and the slope of the linear plot of -ln(Abs) - t was taken as the
pseudo-first-order rate constant (kobs) of the reaction. The linearity
of the plots was usually of a regression coefficient of R2 > 0.996.
The kinetic determinations were repeated at least three times in
most cases, and the standard deviations were reported.
Experimental Section
General Procedures. 9-Phenylxanthylium ion (PhXn+BF4-, 6)31
and 10-methylacridinium ion (MA+ClO4-, 4)32 were synthesized
according to the published procedures. The structures were
characterized by comparing their 1H NMR data and melting point
values with those reported. Commercially available deuterated
isopropyl alcohol (i-PrOD, i-PrOH-2-d, and i-PrOD-d7) and the
normal isopropyl alcohol were all purified before use by distillation
over K2CO3. Acetonitrile was redistilled twice.
The procedure used to determine the rate constants of the MA+
reaction at refluxing temperature of i-PrOH (82 °C) has been
reported in the Supporting Information section of our previous
communication.13
Formation of the Isopropyl Alcohol Adduct of the PhXn+
-
(PhXnOPr-i, 7). PhXn+BF4 (0.02 g) was dissolved in 2 mL of
AN followed by the addition of 10 mL of i-PrOH at room
temperature. The yellow color of the reaction solution disappeared
immediately. The solution was allowed to stay for another 10 min,
and the isopropyl alcohol and acetontrile were evaporated off in a
rotary evaporator. The addition of K2CO3 in the reaction hinders
the decomposition of the adduct in the separation process. The
PhXnOPr-i of quantitative yield was extracted with CH2Cl2: mp
Acknowledgment. This work was supported by the Petro-
leum Research Fund administrated by the American Chemical
Society (ACS-PRF: GB-44844). We thank Professors Vernon
D. Parker and Donald Bethell for helpful discussions.
JO800820U
(31) Dauben, H. J., Jr.; Honnen, L. R.; Harmon, K. M. J. Org. Chem. 1960,
25, 1442–1445.
(33) Matthews, W. S.; Bares, J. E.; Bartmess, J. E.; Bordwell, F. G.;
Cornforth, F. J.; Drucker, G. E.; Margolin, Z.; McCallum, R. J.; McCollum,
G. J.; Vanier, N. R. J. Am. Chem. Soc. 1975, 97, 7006–7014.
(32) Fukuzumi, S.; Koumitsu, S.; Hironaka, K.; Tanaka, T. J. Am. Chem.
Soc. 1987, 109, 305–316.
4770 J. Org. Chem. Vol. 73, No. 13, 2008