Application of external-cavity quantum cascade infrared lasers to nanosecond time-resolved infrared spectroscopy of condensed-phase samples following pulse radiolysis.
ABSTRACT Pulse radiolysis, utilizing short pulses of high-energy electrons from accelerators, is a powerful method for rapidly generating reduced or oxidized species and other free radicals in solution. Combined with fast time-resolved spectroscopic detection (typically in the ultraviolet/visible/near-infrared), it is invaluable for monitoring the reactivity of species subjected to radiolysis on timescales ranging from picoseconds to seconds. However, it is often difficult to identify the transient intermediates definitively due to a lack of structural information in the spectral bands. Time-resolved vibrational spectroscopy offers the structural specificity necessary for mechanistic investigations but has received only limited application in pulse radiolysis experiments. For example, time-resolved infrared (TRIR) spectroscopy has only been applied to a handful of gas-phase studies, limited mainly by several technical challenges. We have exploited recent developments in commercial external-cavity quantum cascade laser (EC-QCL) technology to construct a nanosecond TRIR apparatus that has allowed, for the first time, TRIR spectra to be recorded following pulse radiolysis of condensed-phase samples. Near single-shot sensitivity of DeltaOD <1 x 10(-3) has been achieved, with a response time of <20 ns. Using two continuous-wave EC-QCLs, the current apparatus covers a probe region from 1890-2084 cm(-1), and TRIR spectra are acquired on a point-by-point basis by recording transient absorption decay traces at specific IR wavelengths and combining these to generate spectral time slices. The utility of the apparatus has been demonstrated by monitoring the formation and decay of the one-electron reduced form of the CO(2) reduction catalyst, [Re(I)(bpy)(CO)(3)(CH(3)CN)](+), in acetonitrile with nanosecond time resolution following pulse radiolysis. Characteristic red-shifting of the nu(CO) IR bands confirmed that one-electron reduction of the complex took place. The availability of TRIR detection with high sensitivity opens up a wide range of mechanistic pulse radiolysis investigations that were previously difficult or impossible to perform with transient UV/visible detection.
- [Show abstract] [Hide abstract]
ABSTRACT: The linewidth of an external cavity quantum cascade laser is studied as a function of injection current and laser scan rate. The laser linewidth is inferred to be ca. 2.5 MHz from Lamb-dip spectra on a low pressure sample of NO and its variation with injection current is well modeled using literature values for the intrinsic material properties of the lasing medium. The laser linewidth measurements are corroborated by polarization spectroscopy studies as well as by analysis of hyperfine structure and cross-over resonances.Applied Physics B 112(2). · 1.78 Impact Factor
Application of External-Cavity Quantum Cascade Infrared
Lasers to Nanosecond Time-Resolved Infrared Spectroscopy
of Condensed-Phase Samples Following Pulse Radiolysis
DAVID C. GRILLS,* ANDREW R. COOK, ETSUKO FUJITA, MICHAEL W. GEORGE,
JACK M. PRESES, and JAMES F. WISHART
Chemistry Department, Brookhaven National Laboratory, Upton, New York 11973-5000 (D.C.G., A.R.C., E.F., J.M.P., J.F.W.);
and School of Chemistry, University of Nottingham, University Park, Nottingham, NG7 2RD, United Kingdom (M.W.G.)
Pulse radiolysis, utilizing short pulses of high-energy electrons from
accelerators, is a powerful method for rapidly generating reduced or
oxidized species and other free radicals in solution. Combined with fast
time-resolved spectroscopic detection (typically in the ultraviolet/visible/
near-infrared), it is invaluable for monitoring the reactivity of species
subjected to radiolysis on timescales ranging from picoseconds to seconds.
However, it is often difficult to identify the transient intermediates
definitively due to a lack of structural information in the spectral bands.
Time-resolved vibrational spectroscopy offers the structural specificity
necessary for mechanistic investigations but has received only limited
application in pulse radiolysis experiments. For example, time-resolved
infrared (TRIR) spectroscopy has only been applied to a handful of gas-
phase studies, limited mainly by several technical challenges. We have
exploited recent developments in commercial external-cavity quantum
cascade laser (EC-QCL) technology to construct a nanosecond TRIR
apparatus that has allowed, for the first time, TRIR spectra to be recorded
following pulse radiolysis of condensed-phase samples. Near single-shot
sensitivity of DOD , ,1 66 10? ?3has been achieved, with a response time of
, ,20 ns. Using two continuous-wave EC-QCLs, the current apparatus
covers a probe region from 1890–2084 cm? ?1, and TRIR spectra are
acquired on a point-by-point basis by recording transient absorption
decay traces at specific IR wavelengths and combining these to generate
spectral time slices. The utility of the apparatus has been demonstrated by
monitoring the formation and decay of the one-electron reduced form of
the CO2 reduction catalyst, [ReI(bpy)(CO)3(CH3CN)]þ þ, in acetonitrile
with nanosecond time resolution following pulse radiolysis. Characteristic
red-shifting of the m(CO) IR bands confirmed that one-electron reduction
of the complex took place. The availability of TRIR detection with high
sensitivity opens up a wide range of mechanistic pulse radiolysis
investigations that were previously difficult or impossible to perform with
transient UV/visible detection.
Index Headings: Nanosecond time-resolved infrared spectroscopy; TRIR;
Pulse radiolysis; External-cavity quantum cascade laser; EC-QCL.
The development1of flash photolysis in the late 1940s by
Norrish and Porter revolutionized the application of fast
transient spectroscopic measurements. Modern transient ab-
sorption spectroscopy is a technique in which an excitation
source (the pump) initiates a chemical process with a short pulse
of energy (e.g., laser light, ionizing radiation, electric discharge,
etc.), generating a high concentration of transient species.
Continuous-wave probe light passing through the sample at
selected wavelengths is used to monitor changes in transmit-
tance that occur after excitation, through the use of a fast
detector and a transient digitizer. Depending on the spectral
region being probed, this technique allows species with
lifetimes as short as a few hundred picoseconds to be
interrogated, provided the excitation pulse is short enough.
The detection of even shorter-lived transients down to the
femtosecond timescale can be achieved using a repetitive
sampling technique, with very short excitation and probe pulses
and an optical delay line,2or a method to multiplex optical delay
measurements.3Transient absorption spectroscopy has become
a very powerful tool for unraveling the kinetics and mechanisms
of a wide array of processes, ranging from electron/energy
transfer4and solvation/vibrational relaxation dynamics,5to
excited state reactivity6and photochemical reactions,7with the
most common probe wavelengths being in the ultraviolet (UV)/
visible/near-infrared (NIR). However, absorption bands in these
regions (particularly in the UV and visible) are often broad and
difficult to assign to specific structures of reaction intermediates
that would aid in mechanistic interpretation. Time-resolved
vibrational spectroscopy (TRVS), with either mid-IR absorption
or resonance Raman detection,6–8is a particularly powerful
approach since in contrast to the majority of UV/visible
measurements, it can provide rich structural information on
transient species through the appearance of narrow, highly
characteristic spectral bands. Thus, time-resolved infrared
(TRIR) and time-resolved resonance Raman (TR3) spectrosco-
py have become widely used techniques for the identification of
short-lived intermediates generated by pulsed laser excitation.
Although laser flash photolysis is the most frequently used
excitation source for time-resolved measurements, there are
many other means of initiating events for investigation by fast
spectroscopy. These include rapid mixing with stopped-flow
methods,9,10temperature-11–13and pH-jump14techniques, and
Pulse radiolysis, utilizing short, high-energy electron pulses
from accelerators, is a powerful method for rapidly adding
Received 13 January 2010; accepted 25 February 2010.
Volume 64, Number 6, 2010 APPLIED SPECTROSCOPY
? 2010 Society for Applied Spectroscopy
single positive or negative charges to molecules and ions.15–17
It is also among the most effective means for creating free
radicals. Indeed, there are cases when pulse radiolysis is the
only way to initiate a chemical process. For example,
molecules lacking any suitable chromophores for laser
excitation are conveniently studied by pulse radiolytic
methods. Redox-induced processes with metal complexes, in
which the photoexcited state of the complex is too short-lived
to be quenched by sacrificial electron donors or acceptors, can
also only be initiated by pulse radiolysis for time-resolved
investigations. Furthermore, even if a redox process can be
photoinitiated in the presence of a sacrificial electron donor or
acceptor, the resulting oxidized or reduced sacrificial agents
may interfere, either chemically or spectroscopically, with the
species under investigation; a problem that is often avoided in a
pulse radiolysis experiment. The reactions of radicals and
redox species generated by pulse radiolysis are important
throughout all areas of the chemical and biological sciences in
applications as varied as redox catalysis for solar energy
utilization,18–20the chemistry associated with advanced nuclear
energy systems,21materials for next-generation photovolta-
ics,22and studies of DNA damage23–25and other biologically
In a typical pulse radiolysis experiment on a dilute solution
([solute] , 100 mM), the solvent molecules are almost
exclusively ionized by the electron pulse, resulting in
secondary solvated electrons and various solvent radical
species. Depending on the nature of the solvent, and
sometimes on the controlled addition of other species to the
solution, the overall conditions can be either oxidizing or
reducing with varying degrees of oxidizing or reducing
potential, allowing precise control of the redox processes that
are initiated by the electron pulse.28However, despite the fact
that pulse radiolysis was developed more than 50 years
ago,29–35transient detection has been performed mainly by
measurement of absorption and emission in the UV/visible/
NIR, together with electrical and microwave conductivity,
and various magnetic resonance methods such as electron
paramagnetic resonance (EPR) and chemically induced
dynamic electron polarization (CIDEP).15,34,36–41These
approaches can yield excellent kinetic data but often provide
little direct information about the structures of the transient
species that are being detected.
As mentioned above, vibrational spectroscopy offers the
highly specific molecular and structural characterization that is
lacking, but has received only limited application in pulse
radiolysis. For example, there have been reports of using TR3
methods to study intermediates in pulse radiolysis, most
notably at the Risø National Laboratory in Denmark and the
Notre Dame Radiation Laboratory in the USA. Examples
include TR3spectroscopy of the triplet states of biological
molecules such as b-carotene and all-trans-retinal,42–44and
various other triplets and radical cations/anions of organic45–49
and inorganic species,50,51such as phenylthiyl, various semi-
quinones, phenoxyl, aniline, O3–, and (SCN)2–. However, this
approach has several challenges, most notably that (1)
resonance conditions are required to obtain satisfactory
signal-to-noise; (2) the Raman probe laser may not be innocent,
sometimes causing photochemistry; and (3) a large amount of
signal averaging is necessary for these measurements, which
requires large quantities of sample, particularly when following
non-reversible chemical reactions.
In principle, TRIR spectroscopy would be applicable to the
study of a much wider range of samples than TR3, particularly
since the IR probe light is nondestructive and in certain cases
TRIR can prove to be more sensitive than TR3spectroscopy.
However, until now the coupling of TRIR with pulse radiolysis
has been limited to only a handful of gas-phase studies. The
first of these were pioneering microsecond TRIR measurements
by Schwarz, performed over 30 years ago on gas-phase
ammoniated ammonium and oxonium hydrate ions generated
by pulse radiolysis.52,53In this early work, a globar was used as
the IR source with a dispersive IR spectrometer for detection.
This was followed by another group in the 1990s that used IR
diode-laser-based microsecond TRIR to probe the kinetics of
the gas-phase reactions of various small molecules and radicals
(e.g., NH2, CF3, CH3, NO, N2O, and FNO) initiated by pulse
radiolysis.54–60However, to the best of our knowledge there
have been no reports of pulse radiolysis-TRIR being applied to
condensed-phase chemistry, largely due to the many technical
challenges associated with such measurements. These include
the fact that (1) short path lengths are required for condensed-
phase TRIR (usually ?1 mm) due to strong IR absorption by
the solvent (resulting in smaller signal sizes), unlike in gas-
phase TRIR where path lengths of several meters can be
employed; (2) mid-IR detectors are less sensitive than UV/
visible/NIR detectors; (3) mid-IR molar absorptivities are
normally much smaller than those of UV/visible/NIR bands;
(4) the concentration of transient species generated by pulse
radiolysis is often much lower than in laser flash photolysis,
thus exacerbating the problems highlighted in (1) through (3);
(5) water is a commonly used solvent for pulse radiolysis but it
absorbs intensely throughout the mid-IR, requiring extremely
short path lengths (,100 lm); (6) until recently there has been
no high output power continuous-wave (cw) IR source
available that is tunable throughout the mid-IR region, and
(7) in a pulse radiolysis environment it is often necessary to
separate the sample from the detection system by a large
distance (several meters) to reduce electromagnetic interference
and exposure of the detection system to radiation; the long-
distance transport of mid-IR beams is non-trivial due to strong
absorptions by atmospheric water vapor and CO2 in certain
regions of the spectrum.
Considering these challenges, it is therefore clear that the
routine application of fast (sub-millisecond) TRIR to con-
densed-phase pulse radiolysis will require high radiolytic dose
electron pulses to maximize the concentration of transient
species generated, together with the use of high output power
IR probe sources. This combination will provide the necessary
sensitivity for condensed-phase pulse radiolysis-TRIR mea-
surements, enabling the elucidation of reaction mechanisms
that were previously difficult, or even impossible, to follow
with UV/visible detection. To the best of our knowledge, the
only other group currently making progress toward the
application of TRIR spectroscopy to condensed-phase pulse
radiolysis is in the CEA Laboratoire de Radiolyse (Saclay,
France), where a rapid-scan Fourier transform infrared (FT-IR)
based approach is being developed, currently with ;20–30 s
For nanosecond (and longer) time-resolution TRIR mea-
surements, the choice of IR source is typically limited to either
a globar or a lead-salt semiconductor IR diode laser. Although
very sensitive measurements have been obtained using a
globar-based dispersive TRIR spectrometer following UV/
Volume 64, Number 6, 2010
visible excitation,63,64such experiments require extensive
signal averaging, which is not amenable to the low repetition
rates of typical pulse radiolysis equipment. The CO gas laser
was extensively used as an IR laser source in many of the early
laser flash photolysis TRIR experiments65,66since it emits with
an extremely high output power (up to ;1 W) with a very
narrow (essentially monochromatic) linewidth. Thus, it is
potentially an ideal probe source for pulse radiolysis-TRIR
experiments. However, CO lasers have a limited spectral
window in the mid-IR (approximately 1600–2000 cm?1with a
cooled cavity)66and generate TRIR spectra with a fairly low
spectral resolution (discrete laser lines tunable in approximately
4 cm?1steps). Although lead-salt diode lasers are much more
versatile in that they are continuously tunable throughout the
entire mid-IR,7,66each laser head is restricted to a small region
of the mid-IR (?100 cm?1), and their output power is much
lower than a CO laser (?1 mW), putting a severe limitation on
the signal-to-noise ratio. In addition, they require inconvenient
cryogenic cooling for their operation. Despite these limitations,
IR diode lasers have become a very common probe source for
nanosecond laser flash photolysis-TRIR applications. However,
they do not have sufficient output power to extend this
versatility to condensed-phase pulse radiolysis. The comple-
mentary approach of time-resolved step-scan FT-IR is also a
very powerful and commonly used form of TRIR but could be
more challenging to couple with pulse radiolysis on the
nanosecond timescale, since it uses a globar IR source and
requires a large number of sample excitation pulses.67,68
The TRIR technique continually advances with new
technology and the development of the quantum cascade laser
(QCL) now offers such an opportunity.69,70Unlike diode
lasers, which emit photons upon electron-hole pair recombi-
nation across the material band gap, QCLs are unipolar devices
comprising a superlattice formed from a periodic series of thin
layers of two different semiconductors of varying material
composition that are grown by techniques such as metalorganic
chemical vapor deposition (MOCVD). Laser emission is
achieved via intersubband transitions in the repeated stack of
multiple quantum well heterostructures. Thus, rather than being
annihilated by electron-hole recombination, as in a traditional
diode laser, each electron injected into a QCL’s active region
cascades down a series of discrete energy levels within the
conduction band, emitting a photon at each stage. This cascade
effect results in much higher output powers than diode lasers
(e.g., up to 100 mW cf. , 1 mW for a typical mid-IR diode).
The other major advantage of QCLs is that they can be
designed to emit at almost any desired wavelength in the mid-
and far-IR by changing the thickness of the layers in the
heterostructure during the growth process. QCLs also require
only moderate thermoelectric cooling and are extremely
compact and portable devices.
Until recently, QCLs have been sold as either simple multi-
mode Fabry–Pe ´rot lasers (FP-QCLs) or as distributed feedback
lasers (DFB-QCLs), in which a grating is etched into the active
region to force the operation of the laser at a very specific
wavelength determined by the grating periodicity. As a
consequence, the DFB-QCL is a single-mode device, operating
at a single frequency that may be adjusted only slightly (up to
;10 cm?1) by changing the temperature of the active region.
This has proved to be extremely useful for high-resolution
spectroscopy of trace-gas molecules (e.g., NO, N2O, CO, NH3,
CH4 etc.) for a variety of possible applications including
remote atmospheric sensing of environmental gases and
pollutants, chemical sensing, and medical diagnostics.71
Spectroscopic detection techniques have varied from simple
direct IR absorption spectroscopy,72to more sophisticated
methods such as cavity ringdown spectroscopy,73cavity-
enhanced spectroscopy,74photoacoustic spectroscopy,75and
Faraday modulation spectroscopy.76However, performing IR
laser spectroscopy on multiple gases simultaneously or on
condensed-phase samples, which exhibit much broader ab-
sorption bands, requires an IR laser that is widely tunable.
Although there has been some application of FP- and DFB-
QCLs to the fixed-wavelength analysis of species dissolved in
aqueous solution (e.g., sulfite, sulfate, CO2, glucose, and
acetate), demonstrating the utility of high-power QCLs for
probing aqueous solutions,77–81these QCLs would not be
particularly useful as tunable IR sources for condensed-phase
TRIR measurements. Fortunately, the recent commercialization
of external-cavity (EC) technology for QCLs is set to
revolutionize the field of IR laser spectroscopy,82,83and we
believe it will be the key to enabling routine condensed-phase
pulse radiolysis-TRIR measurements.
External-cavity QCLs are extremely versatile, compact
lasers that emit with a very narrow linewidth and can be
continuously tuned over a wide range, often more than 650
cm?1of their center frequency.84In an EC-QCL, one of the
facets of the QCL chip is anti-reflection coated, thus defeating
the optical cavity action of the cleaved facets. A collimating
lens and a diffraction grating, typically in a Littrow
configuration, are arranged external to the QC chip to create
the optical cavity. This reduces the laser emission to a single
wavelength, which can then be tuned by simply rotating the
grating. Commercial EC-QCLs are currently available at
various center wavelengths ranging from 4.2 to 11.5 lm.82
The utility of EC-QCLs has already been demonstrated in IR
spectroscopic studies of gas-phase and solid samples85–88and
in preliminary TRIR studies of solutions following laser flash
photolysis.89It is therefore clear that they offer a unique
combination of high output power, narrow line width, and
wide, continuous tunability for use as an IR source in pulse
In this paper we describe the first coupling of pulse
radiolysis and nanosecond-TRIR spectroscopy to measure-
ments in the condensed phase. The experiments were
performed at the Laser-Electron Accelerator Facility
(LEAF)40at Brookhaven National Laboratory (BNL), using
two continuous wave EC-QCLs as IR probe sources. High
quality TRIR kinetic traces were obtained on a near single-
shot basis at multiple wavelengths across the tuning ranges
of the EC-QCLs in order to build up TRIR spectra. We
exemplify this approach by monitoring the pulse radiolytic
formation of the one-electron reduced rhenium complex
[ReI(bpy?-)(CO)3(CH3CN)]0(bpy ¼ 2,20-bipyridyl) from
[ReI(bpy)(CO)3(CH3CN)]þin acetonitrile and its subsequent
decay. [ReI(bpy)(CO)3(CH3CN)]þhas been shown to act as a
catalyst for the photocatalytic reduction of CO2to CO in the
presence of sacrificial electron donors, such as tertiary
amines, and other rhenium complexes as photosensitizers,90
and the one-electron reduced species is a key intermediate.
Electron Pulse Generation. The LEAF facility at BNL led
the application of photocathode electron gun accelerator
technology to picosecond pulse radiolysis studies of chemical
kinetics and has been described in detail previously.40Briefly,
in the normal operating mode of LEAF, 1–6 ps pulses of 266
nm laser light are used to excite photoelectrons from a
magnesium photocathode housed inside a 30 cm long resonant
cavity, radio frequency (RF) gun. The emitted electrons are
then accelerated to ;9 MeV by a ;15 MW pulse of RF power
from a klystron. Since the laser pulse is synchronized with the
RF power to produce the electron pulse near the peak field
gradient, the electron pulse length and intensity are a function
of the laser pulse properties. Therefore, electron pulses as short
as 5–7 ps with a typical charge of 5 nC are attainable
(corresponding to a dose of ;30 Gy in a 1 cm aqueous
sample). These processes are typically probed by UV, visible,
or NIR transient absorption spectroscopy on the few picosec-
ond timescale using a pulse-probe delay line40or optical fiber
single-shot methodology3or on the hundreds of picoseconds to
microseconds timescale by direct transient-digitization of the
signals from fast risetime detectors. However, in LEAF’s
alternative ‘‘macropulse’’ mode of operation, the photocathode
is irradiated by ;5 ns 266 nm laser pulses (Spectra Physics,
LAB-170), resulting in ;5 ns pulse width, 9 MeV electron
pulses, which have approximately an order of magnitude
greater radiolytic dose than the picosecond pulses. LEAF’s
macropulse is therefore ideally suited to TRIR experiments,
and furthermore, the 5 ns pulse width conveniently matches the
rise times of the fastest available mid-IR detectors (? ;10 ns).
Infrared Beam Geometry and Data Acquisition. Two
continuous wave EC-QCLs were obtained from Daylight
Solutions Inc. (Poway, CA),82which were continuously
tunable in the regions 1890–1960 cm?1(Model 21052-MHF;
40 mW at gain center) and 1996–2084 cm?1(Model 21049-
MHF; ,100 mW at gain center).? The Littrow configuration of
the external cavity provides a narrow linewidth (;0.003 cm?1)
and mode-hop-free tuning. The output power of the lasers
varies as a function of wavelength, dropping to zero at the
extremities of the tuning range and peaking near the center. In
the regular UV/visible transient absorption pulse radiolysis
experiments at LEAF, the sample is housed inside an
aluminum sample block at the end of the electron beamline
in order to reduce RF pick-up by the detector electronics. A
Faraday cup attached to the back of this housing is used to
measure the charge dose in each electron pulse. The probe light
(typically from a pulsed xenon lamp inside the accelerator
vault), is focused into the sample and imaged through a hole in
the wall onto a detector in the adjacent room. Extremely thin
mirrors (aluminum-coated 0.5 mm thick silica substrate) inside
the sample block transmit the electron beam with minimal
scattering, permitting a reverse collinear beam geometry to
maximize overlap of the electron and probe light beams in the
sample and minimize collection of Cerenkov light generated in
the sample by the electron beam. For the TRIR experiments
reported here, we made use of the same sample housing block
and some of the existing beam collection optics, with a few
modifications. The two EC-QCLs were interchangeably
mounted on a separate optical table inside the accelerator
vault, ;4 m from the sample. Using aluminum mirrors, the IR
beam was steered and focused through the sample and then
guided another 4.5 m into the adjacent room, where it was
refocused onto a liquid N2-cooled HgCdTe photovoltaic IR
detector (Kolmar Technologies, Inc., KMPV11-1-LJ2/239) by
a CaF2lens (20 cm focal length). The IR detector is equipped
with a built-in 20 MHz preamplifier with AC- and DC-coupled
outputs and has a response time of ,20 ns. Due to the high
output power of the EC-QCLs, it was necessary to attenuate the
IR beam with neutral density filters at some IR frequencies in
order to avoid saturation of the detector’s preamplifier. The
signals from the detector were digitized by the regular LEAF
data collection oscilloscope (LeCroy WaveMaster 8620A, 6
GHz),? with data acquisition being controlled by custom
LabVIEW software. Figure 1 shows a schematic diagram of the
experimental setup for pulse radiolysis-TRIR. It should be
noted that since we were working in the 5 lm region of the
mid-IR, where atmospheric water absorptions are less prob-
lematic, it was not necessary to transport the IR beam through
purged tubing. TRIR kinetic traces were acquired on a point-
by-point basis by tuning the EC-QCLs across their tuning
ranges in steps of approximately 2 to 4 cm?1. For each trace, I0
measurements of the IR light level were made before pulse
radiolysis using the DC-coupled output of the detector’s
preamplifier, and transient measurements after pulse radiolysis
were made using the AC-coupled output. Conversion to DOD
kinetic traces was achieved with Eq. 1, where AC represents
the raw kinetic trace from the AC-coupled output, and the
factor of two accounts for the fact that the transimpedance
amplification of the detector’s AC signal is twice that of its DC
DOD ¼ ?log10 1 þAC
the detection electronics used in the pulse radiolysis-TRIR experiments at
LEAF (not shown to scale, and focusing and other optics not shown). Dashed
line is the aluminum sample block housing and gray line is the IR beam. (EC-
QCL) External-cavity quantum cascade laser; (FC) Faraday cup; (M) 0.5 mm
thick aluminum-coated mirror; and (S) sample (IR flow cell).
Schematic representation of the beam geometry around the sample and
? EC-QCL technology continues to improve and the current versions of our
lasers now cover an even broader range (up to 140 cm?1each), with no
tuning gap between the two lasers. Indeed, one of our EC-QCLs has
recently been modified by Daylight Solutions to widen its tuning range to
? Such a high bandwidth oscilloscope is not required for nanosecond TRIR
measurements, but was used to minimize disruption to the regular LEAF
data acquisition setup.
Volume 64, Number 6, 2010
Igor Pro software was used to assemble TRIR spectra from
the combined traces and for kinetic analysis.
Sample Handling. [ReI(bpy)(CO)3(CH3CN)]þ[PF6]–was
prepared according to a literature procedure.91Acetonitrile
(Aldrich, anhydrous) was degassed with argon, stored in a
glovebox, and used without further purification. For a typical
experiment, 20 mL of a 1.5 mM acetonitrile solution of
[ReI(bpy)(CO)3(CH3CN)]þ[PF6]–was prepared in the glove-
box, where it was transferred into a special glass reservoir
vessel fitted with glass-to-kovar metal tube seals, Swagelok
fittings, and Teflon valves. The reservoir vessel was then
connected to a vacuum-tight flow system that has a connection
to a vacuum/gas manifold. Due to size constraints within the
LEAF sample housing block, it was necessary to construct a
miniature IR flow cell (see Fig. 2). This consists of a 1.1-in. w
3 1.5-in. h stainless steel plate with a 3/8-in. / hole drilled
through the center. Recessed shoulders surround the hole on
either side of the plate to accommodate two 0.5-in. /30.5 mm
thick sapphire windows. Holes with an inner diameter of 1/32
in. are drilled through the steel plate to allow liquid to be
flowed through the cell (dashed lines in Fig. 2) via standard
flat-bottom ¼-28 threaded ports at the top of the cell. Thin
Teflon gaskets (;100 lm) are placed on the recessed shoulders
to provide a seal for the sapphire windows, which are clamped
tightly in place using Thorlabs SM05-threaded cage plates (P/
N: SP02) on either side of the cell, which are themselves bolted
together via two through-holes. The design of the cell body
resulted in a fixed path length of 2 mm and tests showed that
the assembled cell was vacuum-tight. The remainder of the
flow system consisted of 1/16-in. PEEK tubing and a
recirculating gear pump (Micropump). The flow system tubing
was evacuated and refilled with argon to a pressure of 2 atm at
least three times before sealing the flow system and circulating
the solution cyclically through the flow cell for the TRIR
Step-Scan Fourier Transform Infrared Spectroscopy.
Time-resolved step-scan FT-IR experiments with pulsed 355
nm laser excitation were performed on a Bruker IFS 66/S
apparatus that has previously been described in detail.92The
same flow system as described above was used, but with a
commercially available CaF2 IR flow cell (Harrick Scientific
RESULTS AND DISCUSSION
Infrared Cell Window Material and Sensitivity of the
Pulse Radiolysis-Time-Resolved Infrared Apparatus. Our
goal was to demonstrate acquisition of TRIR kinetic traces on
the nanosecond timescale following pulse radiolysis of
condensed-phase samples, on a single- or near single-shot
basis. Single-shot sensitivity is necessary since the LEAF
accelerator operates at a low repetition rate (typically ?1 Hz
during this kind of automated data acquisition) and samples
subjected to pulse radiolysis are gradually decomposed by the
electron pulses. It was first necessary to find a suitable window
material that would transmit mid-IR light without generating
artifacts in the transient IR signals recorded after pulse
radiolysis. We performed TRIR tests on an empty KBr cavity
cell (Spectra-Tech, Inc.; P/N: Z11231-3; 10 mm total thickness
of KBr), on two 2 mm thick CaF2windows, and on two 0.5
mm thick sapphire windows. Kinetic traces were recorded in
emission mode (i.e., by recording the signal intensity from the
AC-coupled detector output without IR probe light) following
pulse radiolysis of the different window materials placed inside
the LEAF sample housing block.
For the KBr cavity cell, a large transient emission signal was
produced immediately upon pulse radiolysis, decaying with a
lifetime of ;50 ns. In addition, the cell developed intense blue
color centers. Better results were obtained with the CaF2
windows, with the radiation-induced IR emission signal being
the windowwere observed after a shortperiodofpulseradiolysis.
The best results were obtained with the sapphire windows, which
produced negligible IR emission upon pulse radiolysis and no
detectable color centers. Since 0.5 mm thick sapphire has
sufficient transmittance in the 5 lm region, we used the sapphire
windows in the IR flow cell for the TRIR measurements.
To test the sensitivity of the new TRIR apparatus, we tuned
one of the EC-QCLs to an IR frequency near the center of its
tuning range (1938 cm?1) and recorded a TRIR kinetic
absorption trace with the empty IR flow cell in the sample
housing block. This trace was obtained as the average of four
individual traces and it is clear from Fig. 3 that TRIR signal
levels on the order of DOD , 13 10?3can easily be detected
with this apparatus following pulse radiolysis, on a near single-
Time-Resolved Infrared Spectroscopy Following Pulse
Radiolysis of [ReI(bpy)(CO)3(CH3CN)]þin Acetonitrile.
The reduction of CO2into renewable fuels and chemicals is an
important field of research, since using CO2 captured from
industrial emissions may help to mitigate the effects of global
warming and our dwindling supplies of fossil fuels.93,94
Unfortunately, being the final product of combustion, CO2is
an extremely stable molecule, requiring large amounts of
energy and the use of special catalysts to convert it into reduced
forms. Using sunlight to drive these reactions in so-called
‘‘artificial photosynthetic’’ processes is an extremely attractive
option, since it is a plentiful and clean source of energy; more
energy strikes the earth from the Sun in one hour than an entire
year’s worth of global energy consumption. However,
developing catalysts that can efficiently harness solar energy
and promote multi-electron CO2 reduction reactions is
paramount to the success of such a strategy.
of the miniature home-built IR flow cell used in the pulse radiolysis-TRIR
experiments at LEAF. Dashed lines represent internal holes and recesses.
Front and side views of the stainless steel body (1.1-in. w31.5-in. h)
[ReI(bpy)(CO)3(CH3CN)]þhas previously been shown to act
as a catalyst in the photocatalytic reduction of CO2to CO in the
presence of sacrificial electron donors (e.g., triethanolamine or
triethylamine) and another rhenium complex as a photosensitiz-
er.90Its one-electron reduced form, [ReI(bpy?-)(CO)3(CH3CN)]0is
a key intermediate in the catalytic cycle. However, the precise
mechanism of its reactivity following the reduction step remains
unknown and pulse radiolysis offers a unique opportunity to
investigate this reactivity in the absence of sacrificial electron
donors. Therefore, in a preliminary effort to determine whether we
can monitor the formation of the one-electron reduced species by
TRIR following pulse radiolysis, we performed pulse radiolysis
with TRIR detection on [ReI(bpy)(CO)3(CH3CN)]þin argon-
saturated acetonitrile, using LEAF’s macropulse for excitation.
This was also a convenient system with which to demonstrate the
pulse radiolysis-TRIR experiment because we could easily
characterize the one-electron reduced intermediate in a separate
laser flash photolysis TRIR experiment, allowing a direct
comparison with the pulse radiolysis experiment (see below).
It has already been established that pulse radiolysis of
acetonitrile solutions results in highly reducing conditions due
to the formation of an equilibrium mixture of solvated
electrons (consisting of an electron trapped in a cavity of
several CH3CN molecules) and the acetonitrile radical anion
transparent in the m(CO) IR region, it is an ideal solvent for
our TRIR measurements. Figure 4a shows three TRIR spectra
recorded in the m(CO) region at time delays of 80 ns, 3.1 ls,
and 13 ls after pulse radiolysis of [ReI(bpy)(CO)3(CH3CN)]þ
in argon-saturated acetonitrile. Bleaching of the two ground
state m(CO) IR bands at 1936 and 2040 cm?1is clearly
observed, together with an instantaneous formation of two
new transient bands shifted to lower wavenumbers at 1904
and 2014 cm?1. It was not possible to cover the regions
between the transient bands and on the low-frequency side of
the 1904 cm?1band due to the available tuning ranges of our
EC-QCLs. The red-shift of the m(CO) bands of the transient
species relative to the ground state is characteristic of the
?–.95Thus, since acetonitrile is also suitably
formation of the one-electron reduced complex, [ReI
(bpy?-)(CO)3(CH3CN)]0, in which there is increased electron
density at the rhenium center, resulting in more p-back
bonding into the p* anti-bonding orbitals of the CO ligands
and a weakening of the C[O bonds. It has previously been
shown that the extra electron in the one-electron reduced form
of this type of rhenium complex resides in the p* anti-bonding
orbitals of the bpy ligand.96By comparing the intensities of
the bleach bands in the TRIR spectrum recorded immediately
after pulse radiolysis (Dt ¼ 35 ns) with those of the IR bands
of the ground state before pulse radiolysis, we estimate that
each electron pulse generates ;55 lM of reduced complex.
To confirm that the TRIR spectra observed after pulse
radiolysis are due to the expected reduction product, we
performed a laser flash photolysis TRIR experiment on
[ReI(bpy)(CO)3(CH3CN)]þin acetonitrile in the presence of
0.5 M triethylamine (TEA) as a sacrificial electron donor.
Photoexcitation of the rhenium complex into its metal-to-ligand
charge transfer (MLCT) excited state results in rapid reductive
quenching by electron transfer from TEA, forming the long-
lived (on the order of seconds) reduced complex, [ReI
(bpy?-)(CO)3(CH3CN)]0.97Figure 4b shows a time-resolved
step-scan FT-IR spectrum from this experiment, recorded 500
ns after 355 nm laser excitation. The spectrum is almost
identical to the pulse radiolysis-TRIR spectra apart from the
intensity ratio of the high- and low-frequency m(CO) bands,
which is likely a result of the different spectral resolutions in
the two experiments, thus supporting our interpretation of the
pulse radiolysis-TRIR data.
In the pulse radiolysis experiment, the two transient bands
decay and the two bleach bands recover to ;15% and ;35%
of their initial intensities, respectively, on the 90 ls timescale
pulse radiolysis of 1.5 mM [ReI(bpy)(CO)3(CH3CN)]þin acetonitrile saturated
with 2 atm of argon. Smooth lines are multi-peak Voigt curve fits of the data.
(b) Time-resolved step-scan FT-IR spectrum recorded 500 ns after 355 nm laser
flash photolysis of 1.4 mM [ReI(bpy)(CO)3(CH3CN)]þin acetonitrile saturated
with 2 atm of argon, in the presence of 0.5 M triethylamine as a sacrificial
electron donor. This is the original step-scan FT-IR spectrum (not curve-fitted).
(a) EC-QCL TRIR spectra recorded at the specified time delays after
radiolysis of the empty sapphire window IR flow cell. (*) Small transient
absorption signal (FWHM ; 30 ns) that gradually builds up on the signal with
prolonged use of the same sapphire windows.
EC-QCL TRIR kinetic trace recorded at 1938 cm?1after pulse
Volume 64, Number 6, 2010
(see Fig. 5). The kinetic decay traces clearly exhibit bi-
exponential behavior, with the average lifetimes of the two
components of the transient decay being s1¼2.4 6 0.2 ls and
s2¼23.0 6 3 ls and those of the bleach recovery being s1¼
2.8 6 0.3 ls and s2¼ 28.8 6 3 ls. This was unexpected,
since the one-electron reduced species prepared by electro- or
photochemical methods is stable on the seconds timescale.
However, it is likely that in the pulse radiolysis experiment
the one-electron reduced complex is oxidized by various
cations and radicals present in solution, particularly by
CH3CN?þ, resulting in a faster recovery of the starting
complex by more than one possible pathway. Future
experiments will examine the radiolytic dose dependence of
the decay kinetics, together with the possible use of additives
that will scavenge the CH3CN?þcations (e.g., aniline).98Since
the reaction of the related one-electron reduced complex
[ReI(dmb?-)(CO)3(CH3CN)]0(dmb ¼ 4,40-dimethyl-2,20-bi-
pyridine) in acetonitrile with CO2is known to be very slow,97
future work will also involve the use of the pulse radiolysis-
TRIR technique to investigate the reactivity of other rhenium
complexes (synthetically modified for solubility in alkanes),
toward CO2 in isooctane solution following one-electron
reduction. This would be aided by the use of more EC-QCLs
covering a wider region of the mid-IR.
Pulse radiolysis is an extremely powerful method for
rapidly generating one-electron reduced and oxidized species
and probing their subsequent reactivity with time-resolved
spectroscopic methods. However, transient detection has been
mainly limited to the UV/visible/NIR regions, where
structural information on the species being probed is often
lacking in the spectra. TRIR spectroscopy offers the structural
specificity that is lacking and is applicable to a wide array of
samples, but until now it has only been used to study a few
gas-phase reactions following pulse radiolysis, due to several
Taking advantage of recent technological developments in IR
laser technology in the form of high-power, continuous-wave,
tunable external-cavity quantum cascade lasers, we have
constructed a nanosecond TRIR detection apparatus for the
investigation of condensed-phase samples subjected to excitation
by pulse radiolysis. This is the first demonstration of fast TRIR
detection following pulse radiolysis of condensed-phase sys-
tems. Using two EC-QCLs we were able to cover a probe region
spanning 1890–2084 cm?1, which could easily be expanded in
the future through the purchase of a suite of additional EC-QCLs
and/or a CO gas laser. The sensitivity of the apparatus is on the
order of DOD , 1 3 10?3(after four averaged shots), with a
response time of ,20 ns. A preliminary TRIR experiment has
demonstrated its utility for obtaining high quality TRIR spectral
and kinetic data on the nanosecond timescale following pulse
radiolysis, by monitoring the formation and decay of the one-
electron reduced CO2 reduction photocatalyst [ReI
(bpy?-)(CO)3(CH3CN)]0in acetonitrile solution.
Since TRIR spectroscopy offers rich information about the
structures of transient species, the availability of routine TRIR
spectroscopic detection for pulse radiolysis now opens up the
possibility of studying the mechanisms of a wide range of
redox processes that were previously difficult, or even
impossible, to study with time-resolved UV/visible detection
methods. Examples include unraveling the mechanistic reac-
tivity of biologically important nitrogen oxide species such as
nitroxyl (HNO) and NO–, which lack any UV/visible
absorption bands, and the investigation of transition metal-
based redox catalysts for solar energy conversion and other
catalytic applications. The identification of transient interme-
diates following the radiolysis of ionic liquids and conventional
solvents, which will be critical for the development of
advanced nuclear fuel cycles, will also be greatly aided by
the application of TRIR methods following pulse radiolysis.
This work was performed at Brookhaven National Laboratory and was
funded under contract DE-AC02-98CH10886 with the U.S. Department of
Energy and supported by its Division of Chemical Sciences, Geosciences &
Biosciences, Office of Basic Energy Sciences. We are grateful for discussions
with Dr. Sergei Lymar (BNL) regarding possible pulse radiolysis-TRIR
experiments with HNO and NO–. MWG gratefully acknowledges the receipt of
a Royal Society Wolfson Merit Award.
1. R. G. W. Norrish and G. Porter, Nature (London) 164, 658 (1949).
2. G. R. Fleming, Chemical Applications of Ultrafast Spectroscopy (Oxford
University Press, Oxford, 1986).
3. A. R. Cook and Y. Z. Shen, Rev. Sci. Instrum. 80, 073106 (2009).
4. K. Henbest and M. A. J. Rodgers, ‘‘Photochemical Techniques’’, in
Electron Transfer in Chemistry, M. A. J. Rodgers, Ed. (Wiley-VCH,
Weinheim, 2001), vol. 1, p. 558.
5. S. T. Roberts, K. Ramasesha, and A. Tokmakoff, Acc. Chem. Res. 42,
6. J. R. Schoonover and G. F. Strouse, Chem. Rev. 98, 1335 (1998).
7. K. McFarlane, B. Lee, J. Bridgewater, and P. C. Ford, J. Organomet.
Chem. 554, 49 (1998).
8. D. C. Grills and M. W. George, ‘‘Fast and Ultrafast Time-Resolved Mid-
infrared Spectrometry Using Lasers’’, in Handbook of Vibrational
Spectroscopy, J. M. Chalmers and P. R. Griffiths, Eds. (John Wiley and
Sons, Chichester, 2002), vol. 1, pp. 677–692.
9. B. Chance, Rev. Sci. Instrum. 22, 619 (1951).
10. M. S. Hargrove, ‘‘Ligand Binding with Stopped-Flow Rapid Mixing’’, in
Protein-Ligand Interactions: Methods and Applications, G. U. Nienhaus,
Ed. (Humana Press, Totowa, 2005), pp. 323–341.
11. J. V. Beitz, G. W. Flynn, D. H. Turner, and N. Sutin, J. Am. Chem. Soc.
92, 4130 (1970).
12. R. B. Dyer, F. Gai, and W. H. Woodruff, Acc. Chem. Res. 31, 709 (1998).
13. M. Gruebele, J. Sabelko, R. Ballew, and J. Ervin, Acc. Chem. Res. 31, 699
14. S. V. Lymar and J. K. Hurst, J. Am. Chem. Soc. 117, 8867 (1995).
radiolysis of 1.5 mM [ReI(bpy)(CO)3(CH3CN)]þin acetonitrile saturated with 2
atm of argon. Black curves are bi-exponential fits of the data.
TRIR kinetic traces recorded at 2014 and 2040 cm?1after pulse
15. L. K. Patterson, ‘‘Instrumentation for Measurement of Transient Behavior
in Radiation Chemistry’’, in Radiation Chemistry: Principles and
Applications, Farhataziz and M. A. J. Rodgers, Eds. (VCH, New York,
1987), pp. 65–96.
16. Y. Tabata, Pulse Radiolysis (CRC Press, Boca Raton, FL, 1990).
17. J. F. Wishart, ‘‘Accelerators for Ultrafast Phenomena’’, in Radiation
Chemistry: Present Status and Future Trends, C. D. Jonah and B. S. M.
Rao, Eds. (Elsevier Science, Amsterdam, 2001), vol. 87, pp. 21–35.
18. C. Creutz, H. A. Schwarz, J. F. Wishart, E. Fujita, and N. Sutin, J. Am.
Chem. Soc. 113, 3361 (1991).
19. E. Fujita, J. F. Wishart, and R. van Eldik, Inorg. Chem. 41, 1579 (2002).
20. D. E. Polyansky, D. Cabelli, J. T. Muckerman, T. Fukushima, K. Tanaka,
and E. Fujita, Inorg. Chem. 47, 3958 (2008).
21. J. F. Wishart and I. A. Shkrob, ‘‘The Radiation Chemistry of Ionic Liquids
and its Implications for their use in Nuclear Fuel Processing’’, in ACS
Symposium Series Vol. 1030: Ionic Liquids: From Knowledge to
Application, N. V. Plechkova, R. D. Rogers, and K. R. Seddon, Eds.
(American Chemical Society, Washington, D.C., 2009), pp. 119–134.
22. S. Asaoka, N. Takeda, T. Lyoda, A. R. Cook, and J. R. Miller, J. Am.
Chem. Soc. 130, 11912 (2008).
23. M. Adinarayana, E. Bothe, and D. Schulte-Frohlinde, Int. J. Radiat Biol.
54, 723 (1988).
24. S. Steenken, Chem. Rev. 89, 503 (1989).
25. S. Steenken, S. V. Jovanovic, M. Bietti, and K. Bernhard, J. Am. Chem.
Soc. 122, 2373 (2000).
26. O. V. Gerasimov and S. V. Lymar, Inorg. Chem. 38, 4317 (1999).
27. L. J. Hayward, J. A. Rodriguez, J. W. Kim, A. Tiwari, J. J. Goto, D. E.
Cabelli, J. S. Valentine, and R. H. Brown, J. Biol. Chem. 277, 15923
28. G. V. Buxton, ‘‘An overview of the radiation chemistry of liquids’’, in
Radiation Chemistry: From Basics to Applications in Material and Life
Sciences, M. Spotheim-Maurizot, M. Mostafavi, T. Douki, and J. Belloni,
Eds. (EDP Sciences, Les Ulis, France, 2008), pp. 3–16.
29. Comment made by M. Sangster on p. 112 of Discuss. Faraday Soc. 17, 90
30. J. P. Keene, Nature (London) 188, 843 (1960).
31. M. S. Matheson and L. M. Dorfman, J. Chem. Phys. 32, 1870 (1960).
32. R. L. McCarthy and A. Maclachlan, Trans. Faraday Soc. 56, 1187 (1960).
33. E. J. Hart and J. W. Boag, J. Am. Chem. Soc. 84, 4090 (1962).
34. R. H. Schuler, Radiat. Phys. Chem. 43, 417 (1994).
35. R. H. Schuler, Radiat. Phys. Chem. 47, 9 (1996).
36. A. D. Trifunac, R. G. Lawler, D. M. Bartels, and M. C. Thurnauer, Prog.
React. Kinet. 14, 43 (1986).
37. I. A. Shkrob and A. D. Trifunac, Radiat. Phys. Chem. 50, 227 (1997).
38. D. W. Werst and A. D. Trifunac, Acc. Chem. Res. 31, 651 (1998).
39. F. C. Grozema, R. Hoofman, L. P. Candeias, M. P. de Haas, J. M.
Warman, and L. D. A. Siebbeles, J. Phys. Chem. A 107, 5976 (2003).
40. J. F. Wishart, A. R. Cook, and J. R. Miller, Rev. Sci. Instrum. 75, 4359
41. A. Saeki, T. Kozawa, S. Kashiwagi, K. Okamoto, G. Isoyama, Y. Yoshida,
and S. Tagawa, Nucl. Instrum. Methods Phys. Res., Sect. A 546, 627
42. R. F. Dallinger, J. J. Guanci, W. H. Woodruff, and M. A. J. Rodgers, J.
Am. Chem. Soc. 101, 1355 (1979).
43. N. H. Jensen, R. Wilbrandt, P. B. Pagsberg, A. H. Sillesen, and K. B.
Hansen, J. Am. Chem. Soc. 102, 7441 (1980).
44. R. Wilbrandt and N. H. Jensen, J. Am. Chem. Soc. 103, 1036 (1981).
45. G. N. R. Tripathi, J. Chem. Phys. 74, 6044 (1981).
46. G. N. R. Tripathi and R. H. Schuler, J. Chem. Phys. 81, 113 (1984).
47. G. N. R. Tripathi and R. H. Schuler, Chem. Phys. Lett. 110, 542 (1984).
48. G. N. R. Tripathi, S. Qun, D. A. Armstrong, D. M. Chipman, and R. H.
Schuler, J. Phys. Chem. 96, 5344 (1992).
49. G. N. R. Tripathi and Y. L. Su, J. Phys. Chem. A 108, 3478 (2004).
50. R. Wilbrandt, N. H. Jensen, P. Pagsberg, A. H. Sillesen, K. B. Hansen, and
R. E. Hester, Chem. Phys. Lett. 60, 315 (1979).
51. Y. L. Su and G. N. R. Tripathi, Chem. Phys. Lett. 188, 388 (1992).
52. H. A. Schwarz, J. Chem. Phys. 67, 5525 (1977).
53. H. A. Schwarz, J. Chem. Phys. 72, 284 (1980).
54. J. T. Jodkowski, E. Ratajczak, A. Sillesen, and P. Pagsberg, Chem. Phys.
Lett. 203, 490 (1993).
55. A. Sillesen, E. Ratajczak, and P. Pagsberg, Chem. Phys. Lett. 201, 171
56. H. Meunier, P. Pagsberg, and A. Sillesen, Chem. Phys. Lett. 261, 277
57. P. Pagsberg, A. Sillesen, J. T. Jodkowski, and E. Ratajczak, Chem. Phys.
Lett. 252, 165 (1996).
58. P. Pagsberg, A. Sillesen, J. T. Jodkowski, and E. Ratajczak, Chem. Phys.
Lett. 249, 358 (1996).
59. P. Pagsberg, E. Bjergbakke, E. Ratajczak, and A. Sillesen, Chem. Phys.
Lett. 272, 383 (1997).
60. P. Pagsberg, J. T. Jodkowski, E. Ratajczak, and A. Sillesen, Chem. Phys.
Lett. 286, 138 (1998).
61. S. Le Cae ¨r, G. Vigneron, J. P. Renault, and S. Pommeret, Chem. Phys.
Lett. 426, 71 (2006).
62. S. Le Cae ¨r, G. Vigneron, J. P. Renault, and S. Pommeret, Radiat. Phys.
Chem. 76, 1280 (2007).
63. T. Yuzawa, C. Kato, M. W. George, and H. O. Hamaguchi, Appl.
Spectrosc. 48, 684 (1994).
64. S. Srivastava, J. P. Toscano, R. J. Moran, and D. E. Falvey, J. Am. Chem.
Soc. 119, 11552 (1997).
65. M. Poliakoff and E. Weitz, Adv. Organomet. Chem. 25, 277 (1986).
66. M. W. George, M. Poliakoff, and J. J. Turner, Analyst 119, 551 (1994).
67. W. Uhmann, A. Becker, C. Taran, and F. Siebert, Appl. Spectrosc. 45, 390
68. P. Y. Chen and R. A. Palmer, Appl. Spectrosc. 51, 580 (1997).
69. J. Faist, F. Capasso, D. L. Sivco, C. Sirtori, A. L. Hutchinson, and A. Y.
Cho, Science (Washington, D.C.) 264, 553 (1994).
70. C. Gmachl, F. Capasso, D. L. Sivco, and A. Y. Cho, Rep. Prog. Phys. 64,
71. A. A. Kosterev and F. K. Tittel, IEEE J. Quant. Electron. 38, 582 (2002).
72. G. Wysocki, A. A. Kosterev, and F. K. Tittel, Appl. Phys. B-Lasers Opt.
80, 617 (2005).
73. A. A. Kosterev, A. L. Malinovsky, F. K. Tittel, C. Gmachl, F. Capasso, D.
L. Sivco, J. N. Baillargeon, A. L. Hutchinson, and A. Y. Cho, Appl. Opt.
40, 5522 (2001).
74. Y. A. Bakhirkin, A. A. Kosterev, C. Roller, R. F. Curl, and F. K. Tittel,
Appl. Opt. 43, 2257 (2004).
75. J. P. Lima, H. Vargas, A. Miklos, M. Angelmahr, and P. Hess, Appl. Phys.
B-Lasers Opt. 85, 279 (2006).
76. H. Ganser, M. Horstjann, C. V. Suschek, P. Hering, and M. Murtz, Appl.
Phys. B-Lasers Opt. 78, 513 (2004).
77. B. Lendl, J. Frank, R. Schindler, A. Muller, M. Beck, and J. Faist, Anal.
Chem. 72, 1645 (2000).
78. W. B. Martin, S. Mirov, and R. Venugopalan, Appl. Spectrosc. 59, 881
79. A. Lambrecht, T. Beyer, K. Hebestreit, R. Mischler, and W. Petrich, Appl.
Spectrosc. 60, 729 (2006).
80. S. Schaden, A. Dominguez-Vidal, and B. Lendl, Appl. Spectrosc. 60, 568
81. S. Schaden, A. Dominguez-Vidal, and B. Lendl, Appl. Phys. B-Lasers Opt.
86, 347 (2007).
83. E. Takeuchi, K. Thomas, and T. Day, ‘‘Quantum Cascade Lasers:
Applications multiply for external-cavity QCLs’’, in Laser Focus World
(2009), vol. 45, Issue 1.
84. C. Peng, G. P. Luo, and H. Q. Le, Appl. Opt. 42, 4877 (2003).
85. Q. Wen and K. H. Michaelian, Opt. Lett. 33, 1875 (2008).
86. G. Wysocki, R. Lewicki, R. F. Curl, F. K. Tittel, L. Diehl, F. Capasso, M.
Troccoli, G. Hofler, D. Bour, S. Corzine, R. Maulini, M. Giovannini, and J.
Faist, Appl. Phys. B-Lasers Opt. 92, 305 (2008).
87. A. Karpf and G. N. Rao, Appl. Opt. 48, 5061 (2009).
88. C. W. Van Neste, L. R. Senesac, and T. Thundat, Anal. Chem. 81, 1952
89. J. A. Calladine and M. W. George, Spectrosc. Eur. 21, 6 (2009).
90. H. Takeda, K. Koike, H. Inoue, and O. Ishitani, J. Am. Chem. Soc. 130,
91. J. V. Caspar and T. J. Meyer, J. Phys. Chem. 87, 952 (1983).
92. D. C. Grills, R. van Eldik, J. T. Muckerman, and E. Fujita, J. Am. Chem.
Soc. 128, 15728 (2006).
93. A. J. Morris, G. J. Meyer, and E. Fujita, Acc. Chem. Res. 42, 1983 (2009).
94. M. D. Doherty, D. C. Grills, J. T. Muckerman, D. E. Polyansky, and E.
Fujita, Coord. Chem. Rev., paper in press (2010).
95. I. A. Shkrob and M. C. Sauer, J. Phys. Chem. A 106, 9120 (2002).
96. T. Scheiring, A. Klein, and W. Kaim, J. Chem. Soc.-Perkin Trans. 2, 2569
97. Y. Hayashi, S. Kita, B. S. Brunschwig, and E. Fujita, J. Am. Chem. Soc.
125, 11976 (2003).
98. S. Nad and H. Pal, J. Chem. Phys. 116, 1658 (2002).
Volume 64, Number 6, 2010