Ultratrace kinetic measurements of the reduction of methylene blue

Article abstract of DOI:10.1021/ja028129r

The kinetics of methylene blue reduction by ascorbic acid in acetonitrile was investigated by cavity ring-down spectroscopy. Because of our high sensitivity we were able to use very low concentrations (1-10 nM) of the dye. Under these conditions, we observed a second-order loss of dye as well as a competing back reaction with dissolved oxygen. The use of an inexpensive diode laser and a relatively simple setup should make ultratrace kinetic studies more accessible. Copyright

Full text of DOI:10.1021/ja028129r

Published on Web 01/11/2003
Ultratrace Kinetic Measurements of the Reduction of Methylene Blue
Alexander J. Hallock, Elena S. F. Berman, and Richard N. Zare*
Department of Chemistry, Stanford UniVersity, Stanford, California 94305-5080
Received August 13, 2002 ; E-mail: zare@stanford.edu
Methylene blue (MB+) is a widely studied molecule currently
under investigation for its properties in solar energy,1 HIV
infectivity,2 and hydride-transfer reactions.3 MB+ is known to accept
a hydride in either a concerted or stepwise manner depending on
the reaction partner and conditions.4 Previous work has shown that
the kinetics for reduction in water containing ascorbic acid is first
order in each reagent.5,6 We report the second-order reduction of
methylene blue to leucomethylene blue (MBH) by ascorbic acid
(H2A) in acetonitrile:
We carry out these measurements using cavity ring-down
spectroscopy (CRDS). Previously, we have demonstrated that this
Figure 1. The CRDS apparatus. A diode laser, switched by an AOM, builds
technique can monitor absorbing species in a variety of organic
solvents7 and follow concentration changes on the microsecond time
scale for absorbing species present in solution at the level of nano-
grams to picograms per milliliter. Without the use of CRDS such
measurements would not have been possible. Moreover, because
we employ a small, inexpensive diode laser as the light source,
this type of measurement is easy to implement and should have
wide applicability to the investigation of the kinetics and mechanism
of organic reactions in a regime that was hitherto inaccessible.
CRDS employs highly reflective mirrors to form an optical
cavity. Laser light fills the cavity with radiant energy and then is
shut off. The intensity of the light is then recorded as it leaks out
of the cavity through the back mirror (the cavity ring-down signal).
The time constant of the exponential ring-down, τ, depends on the
characteristics of the cavity. The insensitivity to laser intensity
combined with the multipass nature of the method provides an
increase in sensitivity that is typically orders of magnitude greater
than a traditional absorption measurement of the same solution
sample.8 When an absorber is present in the cavity, it contributes
to the decay of the light intensity, and the value of τ is reduced.
We measure the change in the decay time, ∆τ, with and without
an absorber present in the cavity. Quantities such as the length of
the cavity and the exact reflectivity of the mirrors then cancel out.
Only knowledge of the molecular absorption coefficient ꢀ, which
is readily obtained by a calibration curve or conventional UV-vis
absorption measurements, is required to convert ∆τ to absolute
concentration.
up light in a cavity containing the solution to be measured. The light is
then switched off, and the decay is measured by a photomultiplier and
processed by oscilloscope and computer to extract the decay constant τ.
Figure 2. Data from a reaction of 3.0 nM MB+ with 2.5 µM ascorbic
acid. The derived rate law is clearly a better description than a first-order
loss. Data do not start at t0 because the solutions are well mixed outside
the cavity and then introduced into the chamber.
investigate the reduction of MB+ to MBH by H2A in acetonitrile
with microsecond resolution.
To simplify the kinetics, a several 1000-fold excess of ascorbic
acid was maintained. Under these conditions, it was expected that
a simple, pseudo-first-order disappearance of MB+ would be
observed, as had been seen for the same reaction in water.5,6 Our
data in acetonitrile clearly demonstrate more complex kinetics
(Figure 2). Careful analysis of the reaction over a range of [MB+]
and [H2A] points to a second-order reaction at early times coming
to equilibrium at later times.
The simplest model we could devise that fits our observations
was a second-order loss of MB+ coupled to a first-order regeneration
of MB+ from the product MBH. MBH is known to return to MB+
in the presence of dissolved oxygen.9 With initial [MB+] in the
1-10 nM range, the concentration of H2A is much in excess as is
the amount of dissolved O2. Thus, we write
Figure 1 shows the experimental setup. An acousto-optic
modulator (AOM) chops the output of a diode laser on a time scale
of Hz to MHz. A solution containing MB+ from the chloride salt
(1-10 nM) and H2A (1-10 µM) in acetonitrile at room temperature
fills a 23-cm optical cavity that is endcapped with two mirrors
whose reflectivity (99.98% at 655 nm, in air) has been checked
not to change with solute or reaction product concentration. The
ring-down trace is recorded by a photomultiplier and fit to an
exponential function to obtain τ. We utilized this method to
d[MB+]
) - k′f[MB+]2 + k′r[MBH]
(1)
dt
9
1158
J. AM. CHEM. SOC. 2003, 125, 1158-1159
10.1021/ja028129r CCC: $25.00 © 2003 American Chemical Society

C O MMU N I C A T I O N S
The only source of MBH comes from MB+, so we write that
d[MB+]
) - k′f[MB+]2 + k′r([MB+]o - [MB+]) (2)
dt
This expression can be integrated under the constraint that the rate
constants must be real and positive:
2
a - k′r
2k′f
2ak′f[MB+]o
Figure 3. Comparison of the reaction of 1 nM MB+ with 3.1 µM ascorbic
acid in the presence and absence of dissolved oxygen. A plot of conc-1 vs
time is linear for a second-order reaction. At early times, both cases are
similar, but in the presence of oxygen the reaction bends toward equilibrium.
Noise increases with time in an inverse concentration plot.
[MB+]t )
+
(3)
b exp(at) - 2k′f2[MB+]o
2
Here
and
k4
a ) k′ (k′ + 4k′ [MB+] )
(4)
(5)
MB• + MBH+• 98MBH + MB+
(11)
x
r
r
f
o
generates MBH product. Thus, the rate expression for formation
of product, still without taking the back reaction into account,
becomes:
b ) a2 + 2k′f [MB+]o2 + a(k′r + 2k′f[MB+]o)
2
Figure 2 shows the fit of this model to our data.
d[MBH]
) k4(k1[MB+])(k1k2[MB+]) ) k′[MB+]2 (12)
As a check, a log-log plot of k′f vs [H2A] gives a line of slope
of 0.96, essentially unity. This result confirms the pseudo-first-
order dependence imposed by the excess of ascorbic acid. It follows
dt
which is second order in [MB+]. As a test of our model, a reaction
was run in a solution where the solvent was 1% water in acetonitrile.
The native absorption of water at 650 nm prevents us from using
higher concentrations. Nevertheless, sufficient water is present to
act as the proton source for the proton-transfer reaction. First-order
kinetics were observed under these conditions, as expected.
We have demonstrated the use of cavity ring-down spectroscopy
to follow the kinetics of organic reactions whose species are present
in only nanomolar concentrations. This technique may be used to
overcome common difficulties such as extremely poor solubility
or limited quantity of a molecule under study. The straightforward
laser system and experimental setup appear to make this type of
analysis suitable for many applications. Our investigation of
methylene blue reduction kinetics differs from past experiments in
two important respects. First, our increased sensitivity has allowed
us to use much smaller dye concentrations so the relative concentra-
tion of dissolved oxygen is significant. With greater dye concentra-
tion it was necessary to bubble oxygen through solutions to see
the reverse reaction.9 Second, the aprotic solvent acetonitrile
profoundly alters the reaction mechanism, making it second-order
rather than first order in methylene blue.
that the forward rate constant is kf ) k′/[H2A] ) (8.3 ( 1.6) × 103
f
M-1 s-1 and the reverse is k′ ) 86 ( 69 s-1 where the errors
r
represent one standard deviation. The large uncertainty in the reverse
rate likely arises from a combination of factors. The O2 concentra-
tion varies from solution to solution as we did not attempt to control
it. Moreover, the portion of the curve that most affects k′ is late in
r
the reaction when the signal is much closer to the background noise.
Consequently, we believe that we have only determined a range
for kr′ and that it is possible the dependence is more complex than
first order in [MBH]. The forward rate constant, however, is well
defined. As a test of our model, a reaction was run in a solution
that was well sonicated to remove as much of the dissolved oxygen
as possible. The resulting reaction should show a simple second-
order disappearance of MB+ because the reverse reaction should
be negligible. Figure 3 shows a plot of [MB+]-1 vs time that should
be linear in the case of second-order kinetics. Clearly, removal of
oxygen leads to the behavior predicted by our model.
We propose the following mechanism for the second-order
formation of MBH, which is consistent with our kinetic data and
previously reported results.5,6
In water:
Acknowledgment. A.J.H. is grateful for a Veach Memorial
Fellowship and E.S.F.B. is grateful for an NSF Graduate Research
Fellowship. This work is supported by a grant from the Office of
Naval Research (Grant N00014-00-1-0364).
k1
9
MB+ + H2A
8 MB• + H2A+•
(6)
(7)
(8)
(9)
k2
MB• + H2A+• 98MBH+• + HA•
References
k′2
MB• + H3O + 98 MBH+• + H2O
(1) Jain, S.; Dangi, G.; Vardia, J.; Ameta, S. Int. J. Energy Res. 1999, 23,
71-77.
k3
(2) Owada, T.; Yamada, T.; Abe, H.; Hirayama, J.; Sediguchi, S.; Ikebuchi,
K. J. Med. Virol. 2000, 62, 421-425.
MBH+• + HA•98MBH + HA+
(3) Kirkor, E. S.; Scheeline, A. Eur. J. Biochem. 2000, 267, 5014-22.
(4) Liu, Y.; Yamamoto, S.; Sueishi, Y. J. Photochem. Photobiol., A 2001,
143, 153-159.
and the corresponding rate expression for product formation is:
(5) Mowry, S.; Ogren, P. J. Chem. Educ. 1999, 76, 970-973.
(6) Snehalatha, T.; Rajanna, K. C.; Saoprakash, P. K. J. Chem. Educ. 1997,
74, 228-233.
d[MBH]
) k1k′k3[MB+] ) k[MB+]
(10)
(7) Hallock, A. J.; Berman, E. S. F.; Zare, R. N. Anal. Chem. 2002, 74, 1731-
2
1743.
dt
(8) Berden, G.; Peeters, R.; Meijer, G. Int. ReV. Phys. Chem. 2000, 19, 565-
607.
In acetonitrile, proton transfer from the solvent is not available,
so the proton-transfer step occurs much more slowly. The lifetime
of the MB• intermediate significantly lengthens and the reaction
(9) Resch, P.; Field, R. J.; Schneider, F. W. J. Phys. Chem. 1989, 93,
8181-8186.
JA028129R
9
J. AM. CHEM. SOC. VOL. 125, NO. 5, 2003 1159