Noise-immune cavity-enhanced optical heterodyne detection of HO2 in the near-infrared range

Article abstract of DOI:10.1021/jp301038r

Accurate measurements of the absolute concentrations of radical species present in the atmosphere are invaluable for better understanding atmospheric processes and their impact on Earth systems. One of the most interesting species is HO₂, the hydroperoxyl radical, whose atmospheric daytime levels are on the order of 10 ppt and whose observation therefore requires very sensitive detection techniques. In this work, we demonstrate the first steps toward the application of external-cavity diode-laser-based noise-immune cavity-enhanced optical heterodyne molecular spectroscopy (NICE-OHMS) to the detection of the HO₂ radical in the near-infrared range. Measurements of stable species and of HO₂ were made in a laboratory setting, and the possibilities of extending the sensitivity of the technique to atmospheric conditions are discussed.

Full text of DOI:10.1021/jp301038r

Article
pubs.acs.org/JPCA
Noise-Immune Cavity-Enhanced Optical Heterodyne Detection of
HO2 in the Near-Infrared Range
Claire L. Bell, Jean-Pierre H. van Helden, Tom P. J. Blaikie, Gus Hancock, Nicola J. van Leeuwen,
Robert Peverall, and Grant A. D. Ritchie*
Physical & Theoretical Chemistry Laboratory, Department of Chemistry, University of Oxford, South Parks Road, Oxford OX1 3QZ,
U.K.
ABSTRACT: Accurate measurements of the absolute concentrations of
radical species present in the atmosphere are invaluable for better
understanding atmospheric processes and their impact on Earth systems.
One of the most interesting species is HO2, the hydroperoxyl radical,
whose atmospheric daytime levels are on the order of 10 ppt and whose
observation therefore requires very sensitive detection techniques. In
this work, we demonstrate the first steps toward the application of
external-cavity diode-laser-based noise-immune cavity-enhanced optical
heterodyne molecular spectroscopy (NICE-OHMS) to the detection of
the HO2 radical in the near-infrared range. Measurements of stable
species and of HO2 were made in a laboratory setting, and the possibilities of extending the sensitivity of the technique to
atmospheric conditions are discussed.
by gas expansion), which is essentially laser-induced fluo-
rescence at low pressure,2−8 and PERCA (peroxy radical chain
amplification), whereby HO2 and RO2 catalyze the oxidation of
NO and CO into CO2 and NO2, which can then be
measured.9−13 Although FAGE specifically measures HO2
(indirectly through its conversion into OH), it suffers from
interference from RO2 radicals.14 PERCA traditionally meas-
ures the sum of all RO2 species, although we note that a very
recent work showed the possibility of selective measurement of
atmospheric peroxy radical concentrations of HO2 and RO2
using a denuding method based on PERCA.15 It is therefore
desirable to have a direct and speciated technique.
A highly attractive analytical method for trace gas sensing
that offers the required selectivity, can return absolute number
densities, and might have the required sensitivity levels is
absorption spectroscopy. In the ultraviolet (UV) range, where
optical techniques traditionally operate, the spectroscopy of
HO2 has been extensively studied, and important information
on reaction kinetics has been acquired in laboratory experi-
ments.16−22 Although the cross sections for electronic
excitation of HO2 are large (σ = 10−18 cm2), the absorption
spectrum of HO2 is broad and structureless, and there is
interference from other species, such as other peroxy radicals
and hydrogen peroxide, the main product of HO2 self-reaction.
The commercial development of diode lasers in the near-
infrared range opened up the opportunity of studying HO2 in
this range, and subsequent investigations into these narrower
and better-resolved absorption features and their use in
INTRODUCTION
■
The measurement of absolute concentrations of trace gases in
the atmosphere has far-reaching implications, as such
concentrations are crucial inputs for models that predict future
air quality and climate change. The numerous, varied, and
interlinked chemical cycles in the atmosphere are complex, and
for models to be refined and improved, comparison with
accurate data on the absolute concentrations of individual
species constitutes the best validation. Field measurements
therefore add tremendous value to theoretical modeling and
have led to the development of increasingly effective measure-
ment techniques.1 When measuring trace gases, particularly in
the field, in addition to achieving the necessary sensitivity, it is
also highly desirable that any detection system be able to
operate at ambient temperature and pressure and without the
need for extensive sample preparation. It is also an important
requirement that absolute concentrations of species can be
obtained.
Radical species, such as OH and HO2, control virtually all of
the daytime oxidative chemistry in the lower atmosphere and
are responsible for the transformation of primary emissions into
secondary pollutants.2 Important climate gases such as methane
and ozone are also intricately related to the concentration of
HO2; their atmospheric lifetimes depend on the concentration
of HO2, which also controls the production of acidic species,
with their well-known detrimental effects to the environment.
Although undoubtedly important, HO2 provides a range of
challenges when considering appropriate detection techniques
as atmospheric daytime levels of HO2 are very small [∼2.5 ×
108 molecules cm−3 ≡ 10 parts per trillion (ppt)] and so
require very sensitive detection techniques. Currently, HO2 is
measured by the indirect techniques FAGE (fluorescence assay
Received: February 1, 2012
Revised: April 18, 2012
Published: May 16, 2012
© 2012 American Chemical Society
5090
dx.doi.org/10.1021/jp301038r | J. Phys. Chem. A 2012, 116, 5090−5099

The Journal of Physical Chemistry A
Article
Figure 1. Experimental setup for the NICE-OHMS experiments detailing the ring cavity and a schematic diagram of the electronics. ECDL, external
cavity diode laser; EOM, electro-optic modulator; PD1, PD2, and PD3, photodiodes 1, 2, and 3, respectively; PZT1 and PZT2, piezoelectric
transducers.
studying HO2 kinetics have been published.23−37 The HO2
absorption bands in the near-infrared range are a vibronic
obtained at a pressure of 25 Torr.42 In this method, light is
trapped in a high-finesse optical resonator consisting of two
highly reflective mirrors containing the sample under
investigation and undergoes repeated reflections such that the
effective optical path length is on the order a few kilometers in
a tabletop setup. Although the basic technique of cw-CRDS is
well-established and has been confirmed to be one of the most
convenient and useful sensitive spectroscopic methods regularly
reaching sensitivities on the order of 10−9 cm−1,43 it is
insufficient for the purpose of HO2 detection under
atmospheric conditions, which requires an absorption coef-
ficient of ∼10−12 cm−1 to be measured; for reliable
quantification, even an order of magnitude lower would be
needed. The true potential of optical cavity-based cw laser
absorption spectroscopy is realized, however, when the
radiation is coupled into an optical cavity that is locked in
resonance with the laser wavelength. There are multiple
advantages in conducting spectroscopy within fully resonant
cavities, including increased power levels on the detectors,
which lead to a reduction in the shot-noise limit; excitation of
single stable cavity modes, which leads to a narrower
distribution of ring-down times; and increased data acquisition
rates. Sensitivities on the order of 10−11−10−12 cm−1 Hz−1/2 are
expected and, indeed, have been attained in a handful of elegant
experimental methods.43
However, where truly ultrasensitive detection is required, as
is the case for the atmospheric detection of HO2, the potential
that the noise-immune cavity-enhanced optical heterodyne
molecular spectroscopy (NICE-OHMS) technique offers is
unrivalled. Pioneered by Ye and Hall,44 this technique requires
that the laser be locked on resonance with the cavity, utilizing
the increase in path length and signal, while also frequency
modulating the input radiation. In fm-NICE-OHMS, a specific
case of frequency-modulation spectroscopy (FMS) is per-
formed in which the modulation frequency is carefully selected
to exactly match that of the cavity free spectral range or
multiples thereof. This leads to the “noise-immunity” feature of
2
2
progression in the low-lying A′ ← A″ electronic transition
(∼7017 cm−1), along with the first vibrational overtone in the
OH stretch (2ν1) centered at 6649 cm−1. It is this latter band
that we probe in this work, because, although the spectral range
offers smaller line strengths than the mid-infrared and UV
ranges,38,39 it does offer advantages through the availability of
cheap, commercially available continuous-wave (cw) laser
sources such as the external-cavity diode laser (ECDL)
(1480−1540 nm) used in this work. The commercial
importance of diode lasers in the telecommunications and
consumer electronics industries drives their ongoing develop-
ment, resulting in lasers being available at a wide range of
wavelengths (375−430 and 630−2700 nm) with suitable
powers and narrow linewidths. The introduction of cw
quantum cascade lasers in the past decade has now enabled
expansion into the mid-infrared range as has been recently
reported for HO2 probing the ν3 band, which has absorption
cross sections 1 order of magnitude larger than in the near-
infrared range.40,41
Previous works have reported a variety of diode-laser-based
experiments in the near-infrared range within the 2ν1 band to
detect HO2, including Taatjes and Oh using wavelength-
modulation spectroscopy (WMS)25 and DeSain et al. using
two-tone frequency-modulation spectroscopy (TTFMS) to
measure the line positions with high resolution.26 In most of
these studies, heterodyne detection techniques such as WMS
and TTFMS were used to measure small absorbances, often in
combination with multipass cells, such as Herriott-type flow
cells, to increase the absorption path length. In studies by
Ibrahim et al. on air-broadening coefficients34 and by Thiebaud
et al. on detailed high-resolution measurements of absorption
cross sections,35 continuous-wave cavity ring-down spectrosco-
py (cw-CRDS) was used to monitor time-resolved HO2
concentrations. With this technique, a detection limit for
HO2 of 3.2 × 109 cm−3 for a 1 Hz detection bandwidth was
5091
dx.doi.org/10.1021/jp301038r | J. Phys. Chem. A 2012, 116, 5090−5099

The Journal of Physical Chemistry A
Article
NICE-OHMS; as the frequency components of the FM triplet
(carrier frequency and sidebands) are all separated by the free
spectral range, each is transmitted through the cavity in the
same way (i.e., maintaining the resonance with a cavity mode),
with each component being affected identically by the cavity. As
the fm-NICE-OHMS signal arises from the beating between
the carrier and the sidebands, the signal is noise-immune to
jitter in the laser frequency, which is responsible for most of the
sensitivity-limiting noise in the signal. The use of modulation
techniques transposes the signal to high frequency (a few
hundred megahertz) to achieve a reduction in the noise. On top
of this, a “slower” wavelength-modulation spectroscopy (WMS)
can be applied to further improve the sensitivity (wm-NICE-
OHMS), for example, by altering a piezoelectric transducer on
one of the cavity mirrors.
power incident on the cavity was 1 mW. The wavelength of
the laser was monitored with a wavemeter (Burleigh WA-
1000), and a frequency scale for the data was defined by
monitoring the laser output with a spectrum analyzer (10 GHz
Melles Griot 1300−1500 nm, adjusted for 0.81 GHz free
spectral range).
The ring cavity was formed by four high-reflectivity mirrors
(1 in. in diameter, radius of curvature of r = −1 m) in a bow-tie
configuration with a half-round trip length, L, of 137 cm and a
cavity free spectral range (FSR) of 109.5 MHz. The specified
reflectivity of the input and output coupling mirrors was 0.9994
(Newport 10CV00SR.70T), and that of the other two mirrors
was 0.9997 (Newport 10CV00SR.70F). The two highest-
reflectivity mirrors were mounted on piezoelectric transducers
(PZTs): PZT1 (PiezoMechanik HPSt 150/20-15/55) was used
for scanning the cavity length, and PZT2 (Physik Instrumente
GmbH P-622.10 L) was used for the modulation of the cavity
length for WMS. The vacuum chamber housing the ring cavity
was evacuated to 10−4 Torr before it was filled with a given
pressure of gas, monitored with a capacitance manometer
(MKS instruments 722A11TBA2FA, 0−10 Torr). To avoid
etalons, wedged windows (antireflection coated for 1050−1600
nm) were used on the vacuum chamber.
Electronic Components. The electronic components were
numerous and finely tuned to the power and frequency of the
laser source. For the locking of the laser frequency to a cavity
TEM00 mode and the FM frequency to twice the cavity FSR,
three frequencies were generated by two voltage-controlled
oscillators (VCOs, Mini-Circuits JTOS-300). Signals were
generated at 219 and 244 MHz (used for the FSR locking)
and mixed and filtered to produce the 25 MHz difference
frequency used for locking the laser to the cavity. The laser
frequency was locked to a TEM00 mode of the cavity using the
Pound−Drever−Hall (PDH) method,51 with EOM1 used to
produce sidebands at 25 MHz on the laser frequency. The
laser light reflected from the cavity was directed onto a fast
photodetector (PD1) (New Focus 1611) with a bandwidth of 1
GHz, amplified (Mini-Circuits ZFL-500LN), and then
demodulated at 25 MHz using a double balance mixer (Mini-
Circuits ZAD-1) to generate the PDH error signal. This error
signal was processed by a fast (megahertz) locking circuit with a
bandwidth of ∼2 MHz and fed back to the laser injection
current for primary locking of the laser frequency to the cavity.
With the laser locked to the cavity, the cavity length was
changed such that the laser frequency was scanned over a
preselected range by applying a sinusoidal voltage ramp, at a
frequency of 1 Hz, to PZT1. A slow locking circuit followed the
change in cavity length and generated a signal that was fed back
to the laser PZT input for the ECDL, to maintain the lock of
the laser to the cavity while scanning the cavity length. The
bandwidth of the laser when locked was necessarily less than
∼50 kHz. A feed-forward loop was implemented in the locking
scheme to reduce the load placed on the locking electronics
and to increase the scan rate for which mode-hop free tuning of
the laser was maintained.
Although NICE-OHMS has yielded unprecedented sensitiv-
ities of 10−14 cm−1 Hz−1/2, we note that such performance has
been observed only in measurements of sub-Doppler saturated
line shapes.45 Other practitioners have shown that the
technique becomes limited by noise associated with etalon
structures that become evident only when scanning over
hundreds of megahertz, as is generally necessary for making
observations of molecular species.46−48 For example, the Axner
group used fiber-laser-based NICE-OHMS for Doppler-
broadened detection of C2H2 and CO2 around 1531 nm
using a cavity with a finesse of 4800 with a minimum detectable
47
absorption of 3.5 × 10−11 cm−1 Hz−1/2 and most recently
reported a sensitivity of 1.8 × 10−11 cm−1 Hz−1/2 49 We recently
.
reported the performance of an ECDL-based NICE-OHMS
spectrometer and obtained a sensitivity of 4 × 10−11 cm−1
Hz−1/2 50 In this article, we present the application of this near-
.
infrared NICE-OHMS spectrometer to the detection of the
HO2 radical. Spectra from stable species of atmospheric
relevance, namely, CO2, CH4, and N2O, are used to illustrate
the performance of the spectrometer and to demonstrate that
the technique is capable of measuring spectral parameters. We
then present the first application of the NICE-OHMS system as
an accurate and direct technique for measuring the
concentration of the hydroperoxyl radical HO2 in a laboratory
setting.
EXPERIMENTAL DETAILS
■
Optical Setup. A schematic of the experimental setup is
shown in Figure 1. The light source was an ECDL (Sacher
Lasertechnik LION, linewidth < 2 MHz) tunable in the range
of 1480−1540 nm with a maximum output power of 3.2 mW
providing a mode-hop free tuning range of 56 GHz using
current coupling. The laser light was first directed through an
optical isolator (Isowave I-15-UHP-4 1550 nm) to prevent
optical feedback before passing through two beam-shaping
lenses, which reshaped the elliptical beam output from the laser
to a predominantly Gaussian-shaped beam, mimicking the
TEM00 mode of the cavity. Two electro-optic modulators
(EOMs) were situated after these lenses: EOM1 [Nova phase
electro-optic phase modulator (EO-PM-NR-C3)] to modulate
the light for locking the laser frequency to a TEM00 mode of
the cavity and EOM2 [Quantum Technologies (TWAP 10-
1350-1650 nm)] for modulation at the cavity free spectral range
and, hence, postcavity FM detection. To isolate the alignment
of the beam into the cavity from adjustments to the laser and
EOMs and also to provide a clean, predominantly Gaussian-
shaped beam, the light was fed through a single-mode
polarization preserving fiber (OZ Optics). The resulting
To perform fm-NICE-OHMS, EOM2 frequency-modulated
the laser light at ∼219 MHz to generate sidebands at 2 times
the FSR from the carrier laser frequency. However, the FSR of
the cavity changes slightly during a frequency scan over an
absorption line as the length of the cavity changes, and the
noise-immune aspect of NICE-OHMS requires a good match
between the modulation frequency and the FSR of the cavity.
Hence, the modulation frequency was locked to the FSR using
5092
dx.doi.org/10.1021/jp301038r | J. Phys. Chem. A 2012, 116, 5090−5099

The Journal of Physical Chemistry A
Article
Table 1. Reactions Occurring during HO2 Generation and Their Corresponding Rate Constants
reaction
Cl2 → Cl + Cl
rate constant
7 × 10−3
units
ref
R1
s−1
this work
53
R2
5.5 × 10−11
9.1 × 10−12
3.5 × 10−11
1.0 × 10−11
2.0 × 10−11
1.4 × 10−12
6.6 × 10−10
4.1 × 10−13
1.8 × 10−12
7.3 × 10−11
2.9 × 10−11
cm3 molecule−1 s−1
cm3 molecule−1 s−1
cm3 molecule−1 s−1
cm3 molecule−1 s−1
cm3 molecule−1 s−1
cm3 molecule−1 s−1
cm3 molecule−1 s−1
cm3 molecule−1 s−1
cm3 molecule−1 s−1
cm3 molecule−1 s−1
cm3 molecule−1 s−1
Cl + CH3OH → CH2OH + HCl
CH2OH + O2 → HO2 + H2CO
HO2 + Cl → HCl + O2
R3
53
R4
53
R5
53
HO2 + Cl → OH + ClO
R6
54
HO2 + CH2OH → H2CO + H2O2
HO2 + HO2 → H2O2 + O2
Cl + CH2OH → H2CO + HCl
H2O2 + Cl → HCl + HO2
OH + H2O2 → HO2 + H2O
H2CO + Cl → HCO + HCl
CH2OH + Cl2 → ClCH2OH + Cl
R7
53
R8
55
R9
53
R10
R11
R12
53
53
56
the DeVoe and Brewer method:52 A fraction of the amplified
signal from PD1 was split off (Mini-Circuits ZFSC-2-1-S) and
demodulated with a double balance mixer (Mini-Circuits ZAD-
1) at 244 MHz (the sum of the cavity FSR and the PDH
locking frequency) to produce an error signal for locking the
FM frequency to the cavity FSR. This error signal was fed into a
proportional−integral locking circuit to adjust the tuning
voltage on the VCO and thus the modulation frequency of
EOM2 to match twice the FSR.
the necessity to maintain the laser lock to the cavity over the
entire data acquisition period.
HO2 Generation and Kinetic Modeling. HO2 was
generated within the optical cavity through Cl-atom reactions
with CH3OH and O2, which were initiated by the continuous-
wave photolysis of Cl2 by UV lamps
Cl2 + hν → Cl + Cl
(1)
Cl + CH3OH → CH2OH + HCl
(2)
To produce the fm-NICE-OHMS signal, the ac output of a
second photodetector PD2 (New Focus 1611) was amplified
(Mini-Circuits ZKl-1R5 and ZFL500HLN) and high-pass-
filtered (Mini-Circuits SHP-175), before being demodulated at
∼219 MHz with a double balance mixer (Mini-Circuits ZDM-
1W-S). This signal was passed through a 100 Hz low-pass filter,
recorded using a digital oscilloscope (Lecroy 9304), and stored
on a computer. Changing the demodulation phase allowed
either the absorption or dispersion FMS signal to be acquired.
Following the generation of the fm-NICE-OHMS signal, the
detection sensitivity can be further improved by applying an
additional modulation to the cavity in the form of WMS. The
cavity length and, through the locking circuit, the laser
frequency were dithered using PZT2 at a low frequency of
60 Hz with a frequency excursion of 0.1 GHz. To produce this
wm-NICE-OHMS signal, the fm-NICE-OHMS signal was
demodulated at 60 Hz using a lock-in amplifier (EG&G 7265
DSP). Demodulation of the signal occurred with a time
constant of 50 ms, corresponding to a bandwidth of 3.2 Hz.
Again, the processed signals were recorded with the
oscilloscope and passed to a computer for analysis.
The maximum sensitivity that could be achieved was limited
by residual amplitude modulation (RAM) resulting from
imperfections in the EOMs and etalons before the cavity,
whereby some of the radio frequency applied through the EOM
induces amplitude modulation, leading to undesirable noise on
the detected signal. Etalon effects were minimized with the use
of wedged windows, careful alignment of the optical
components, and using optical fibres with angled ends. A
simple method of removing a background variation would be to
implement a background subtraction technique, whereby the
signals are taken both with and without the absorber present in
the cavity and then subtracted from one another to remove the
background variation. However, this approach was not possible
with this NICE-OHMS setup, given the drift in the laser, and
CH2OH + O2 → HO2 + H2CO
(3)
A more complete reaction scheme, including all of the major
reactions involved in the generation and loss of HO2 with their
rate constants, is listed in Table 1. Detailed modeling of the
HO2 production and reaction scheme was undertaken to
predict the concentrations of HO2 and of potentially interfering
species that also absorb around 1.5 μm, such as H2CO. For the
modeling, an estimate of the rate constant for Cl2 photolysis
was needed (reaction 1) for our particular experimental
configuration.
To photolyze Cl2, five UV lamps were fitted in the top of the
vacuum chamber, 2 cm above the bow-tie cavity, beneath the
Perspex lid. The spectrum of one of the lamps (Philips Sylvania
TLD F18W black light blue tubes) was recorded using a Jobin
Yvon spectrometer (CP200), and in combination with
knowledge of the power of one lamp (18 W), this allowed
the number of photons emitted (per nanometer per second) to
be calculated. Considering the five-lamp configuration inside
the vacuum housing and the assumption that one-fourth of the
radiation from each lamp was incident on the sample based on
the geometry of the vacuum chamber, the flux of photons, I0(λ)
(cm−2 s−1), was thus estimated at 2 cm below the lamps, that is,
at the beam path in the cavity. In the case of weak absorption,
as here, the Beer−Lambert law can be applied to calculate the
number of photons absorbed in 1 cm−3 in 1 s
Iabs(λ) = I0(λ)ε(λ)cl
(4)
where ε(λ) is the wavelength-dependent absorption cross
section of chlorine; c is the concentration of Cl2; and l is the
interaction length, taken to be 1 cm. Integrating this expression
over the wavelength range of interest results in the number of
photons absorbed per cubic centimeter per second, which
corresponds to the rate of Cl2 dissociated (assuming that the
5093
dx.doi.org/10.1021/jp301038r | J. Phys. Chem. A 2012, 116, 5090−5099

The Journal of Physical Chemistry A
Article
quantum yield for photodissociation is unity). The rate
constant for chlorine-atom production, k, is defined as d[Cl]/
dt = k[Cl2] and is therefore equal to 2∫ ∞0 I0(λ) ε(λ) dλ; this
gives a rate constant for these experimental conditions of 7 ×
10−3 s−1. For a similar experiment, the rate constant for
chlorine-atom production was measured as 1.35 × 10−3 s−1 for
one lamp,57 corresponding to a value of 6.7 × 10−3 s−1 if five
lamps were used. We note that the value determined here,
which uses the assumption that a geometrical factor of one-
fourth of the photons emitted interact with Cl2, does not
control the HO2 concentration; modeling with the photolysis
rate constant increased by a factor of four led to the HO2
concentration increasing by only 1%, because both the
production and loss rates of HO2 are mainly dependent on
species formed through photolysis.
The HO2 peak absorption cross sections in the 1.5-μm range
are typically around σ = 1 × 10−19 cm2,35 and to ensure that
detection of HO2 was successful and well within the capabilities
of the NICE-OHMS spectrometer, an initial minimal
absorption coefficient, αmin, of 10−7 cm−1 was determined.
The required concentration of HO2 molecules was ∼8 × 1011
molecules cm−3, and kinetic modeling was undertaken, with
initial concentrations of reactants chosen to generate HO2 at
this concentration. The simulation models continuously flowing
products and reactants, with the entire gas volume in the
vacuum chamber (36800 cm3) being replaced every 120 s.
Simulations were performed with and without wall-loss
processes, with the effective rate constants estimated using
the method of Chantry,58 and their effects are discussed later.
The steel and Perspex vacuum assembly housing the ring
cavity was first evacuated to 10−4 Torr before being filled to 760
Torr with N2. After the lights had been switched on, the
chamber was evacuated to a pressure of 40 Torr, and the flows
of constituent gases were started. Flows of 18.2 sccm (standard
cubic centimeters per minute) of 1% Cl2 in Ar (certified purity
of 1.00% in argon balance), 3.7 sccm of O2 (BOC, certified
purity of 99.5%), and 1719 sccm of Ar (BOC Pureshield,
99.998% purity) were controlled by mass flow controllers
(Tylan FC-260), with the flow of 1.83 sccm of CH3OH vapor
being controlled through a calibrated needle valve. Pressures
inside the vacuum chamber were measured with two
capacitance manometers (MKS Instruments, 0−100 and 0−
760 Torr), with the total pressure in the cell of reactants and
buffer gas being controlled actively with a valve in the exit path
before the pump. With the flows and lamps on, the locking
procedure was implemented, as outlined above.
through the cavity can be represented by the modified Beer−
Lambert law
I0(ν) − I(ν)
I(ν)
α(ν)L
1 − R
=
(5)
where I(ν) and I0(ν) are the transmitted intensities with and
without an absorbing species present, respectively; α(ν) is the
frequency-dependent absorption coefficient (the product of the
absorption cross section and the absorber number density); L is
the cavity length; and R is the geometric mean of the mirror
reflectivity.43
For the determination of the cavity finesse, 25 Torr of
methane was introduced into the cavity, and the CEA spectrum
was recorded for an unassigned and weak CH4 transition at
6595.90 cm−1 with an integrated absorption cross section, σint,
of 1.115 × 10−25 cm2 cm−1 obtained from the HITRAN
database.59 Recent high-resolution CRDS measurements by
Campargue et al.60 gave very similar results for the line
strengths. Figure 2 shows a typical locked CEAS trace for 25
Figure 2. Typical locked CEAS spectrum of 25 Torr of CH4 at
6595.90 cm−1.
Torr of CH4 along with a Voigt fit with the Gaussian
component fixed to the calculated full width at half-maximum
(fwhm) Doppler width of 611 MHz appropriate for a sample at
298 K. This CEA spectrum was acquired by scanning over 5
GHz at a scan rate of 1 Hz; the signal was acquired with 30
averages over an acquisition time of 30 s. The returned
Lorentzian component of the fit was 91
3 MHz (fwhm),
which does not agree well with the calculated value for self-
broadening of 138 MHz using the self-broadening parameter
reported in the HITRAN database for transitions in this
range.59 However, widespread values are reported in the
literature for the self-broadening of CH4 transitions in various
vibrational bands,61,62 and our value is within these limits. A
RESULTS AND DISCUSSION
■
Spectrometer Characterization. In this section, observa-
tions of a selection of weak transitions of methane, carbon
dioxide, and nitrous oxide across 52 cm−1 (from 6595 to 6647
cm−1) are presented, demonstrating (i) the calibration of the
technique, (ii) the broad operating range of the spectrometer,
and (iii) its applicability for returning spectroscopic parameters
such as pressure-broadening parameters.
mean mirror reflectivity R of 0.9985
0.0001, and thus a
finesse of 2100 200, was calculated from the integrated area,
A, of the locked CEAS data, and the determined sensitivity was
6.5 × 10−8 cm−1. The moderate sensitivity is partly due to the
fact that, in locked CEAS, any phase and frequency
modulations are converted into residual amplitude modulation,
which increases the noise on the signal; the NICE-OHMS
technique seeks to minimize these effects by utilizing
modulation spectroscopy within a fully resonant cavity.
Broad Wavelength Tunability and Pressure Broadening.
To analyze the performance of the spectrometer over a broad
wavelength range and to demonstrate that the technique is
capable of measuring spectral parameters, this section details
Locked CEAS. Calibration of the spectrometer requires the
finesse of the optical cavity to be determined. To this end, the
transmission of the cavity was monitored by a fast photo-
detector (PD2), and its dc output recorded in a locked cavity-
enhanced absorption spectroscopy (CEAS) experiment with
the same data acquisition system but with the electronics for
fm-NICE-OHMS switched off. For CEAS, the transmission
5094
dx.doi.org/10.1021/jp301038r | J. Phys. Chem. A 2012, 116, 5090−5099

The Journal of Physical Chemistry A
Article
observations on CO2 measurements and the measurement of
pressure-broadening parameters for CH4 and N2O. Figure 3
Figure 3. fm- and wm-NICE-OHMS signals for the P34(e) transition
in the ν1 + ν2 + 2ν3 band at 6646.58 cm−1 for 8.1 (fm-NICE-OHMS)
and 0.2 Torr (wm-NICE-OHMS) of CO2.
shows fm- and wm-NICE-OHMS signals for the P34(e)
transition in the ν1 + ν2 + 2ν3 band of CO2 at 6646.58 cm−1
with the highest sensitivity achieved with this measurement
being 2 × 10−10 cm−1. This is 2 orders of magnitude
improvement in the sensitivity over locked CEAS. The signal
was acquired with six averages over an acquisition time of 10 s.
We found that this was the optimal acquisition time to provide
the maximum signal-to-noise ratio, limited by longer-term laser
drift problems. It should be emphasized that the position of the
CO2 transition is 40 cm−1 from the spectral position where
CH4 data were taken previously, which resulted in a sensitivity
of 3 × 10−11 cm−1 for wm-NICE-OHMS.50 The response of the
laser to the locking is not constant across the broad spectral
range accessible with the laser, resulting in a varying amplitude
modulation across the spectral range and a different level of
noise. The highest sensitivities achieved with the system occur
in the spectral range around 6596 cm−1, for which the system
was initially designed and optimized. From measurements
across the tuning range of the laser, it is clear that optimizing
the electronics at each spectral range is particularly important
and can markedly improve the sensitivity attained; yet this is
time-consuming and often not possible when designing a field
instrument for practical application. However, despite this
variation in sensitivity across wavelength ranges, we note that
the ability to apply NICE-OHMS to various transitions across
different ranges of the spectrum while still achieving
sensitivities of around 10−10 cm−1 is of great benefit when
designing a spectrometer that can meet a wide range of
detection requirements.
Figure 4. Plots of signal height versus pressure of CO2 gas for fm-
NICE-OHMS and wm-NICE-OHMS data.
then used to determine pressure-broadening coefficients for the
unassigned transition of CH4 at 6610.06 cm−1 broadened by
helium. Twenty-five Torr of CH4 was added to the evacuated
cavity, after which increasing amounts of He were added, from
2 to 300 Torr. At each pressure, fm-NICE-OHMS data were
recorded at both the absorption and dispersion phases. The FM
data were fitted using a Voigt profile, fixing the fwhm of the
Gaussian component of the fit to the theoretical value of 611
MHz and allowing the half-width at half-maximum (hwhm) of
the Lorentzian component of the fit to vary. From this fitting
routine, the Lorentzian width of each absorption profile was
obtained. As demonstrated by the typical trace showing the
data (black line) and fit (red line) in Figure 5, the signals of
both the absorption (top) and dispersion (bottom) phases for
25 Torr of CH4 broadened by 150 Torr of helium are well-
represented by Voigt profiles transformed into FM space.
During the data acquisition routine, only one phase of the data
(absorption or dispersion) can be acquired per scan, with the
operator selecting the phase. Hence, for the data presented
here, at each pressure, both the absorption and dispersion
signals were acquired separately, but during the same lock. The
time delay between each pair of measurements was around 20 s.
In Figure 6, the Lorentzian component of each fit from the
absorption spectra is plotted against pressure, with a straight
line fitted to the points. From the graph, a pressure-broadening
coefficient of 1.5 0.1 MHz Torr−1 (hwhm) is obtained, which
corresponds well with the literature values of 1.3−1.7 MHz
Torr−1 for transitions within the 2ν3 band of methane at 1.65
μm.63,64 The figure shows again a y-axis offset of 44 5 MHz
attributable to self-broadening that is consistent with the locked
CEAS experiment for 25 Torr of pure CH4.
To verify the linearity of the technique, the relationship
between the amplitude of the absorption feature and the
pressure of added gas was recorded for both the fm-NICE-
OHMS and wm-NICE-OHMS signals with increasing pressures
of CO2. In both cases, the amplitudes increased linearly with
increasing pressure of gas, with the results being shown in
Figure 4. The nonzero intercept gives an indication of the noise
inherent in the measurements and will determine the minimal
sensitivity achievable with the system. fm-NICE-OHMS was
A similar procedure was undertaken to retrieve pressure-
broadening coefficients of N2O broadened by the addition of
helium. Figure 7 shows the fm-NICE-OHMS absorption signal
5095
dx.doi.org/10.1021/jp301038r | J. Phys. Chem. A 2012, 116, 5090−5099

The Journal of Physical Chemistry A
Article
vary. Figure 8 shows the plot of the Lorentzian hwhm
component of each fit against pressure, with a straight line
Figure 8. Lorentzian hwhm as a function of pressure for 0.42 Torr of
N2O broadened by He.
yielding a pressure-broadening coefficient of 2.2
0.1 MHz
Figure 5. Simultaneous fitting for the (top) absorption and (bottom)
dispersion spectra of 25 Torr of CH4 at 6610.063 cm−1 broadened by
150 Torr of helium.
Torr−1, in excellent agreement with the literature value of 2.1
MHz Torr−1 for the P(26) transition in the same ν3 band.65
Spectroscopic Detection of HO2. Having shown the
calibration and broad wavelength tunability of our NICE-
OHMS spectrometer with stable species, the final part of this
article focuses on the detection of the hydroperoxyl radical
HO2 in a laboratory setting. Specifically, the near-infrared
radiation of the laser probes the first vibrational overtone in the
OH stretch, 2ν1 (centered at 6648.9 cm−1). The two absorption
features selected for investigation in this work were assigned by
DeSain et al.26 as the P1(12) transition at 6623.32 cm−1,
q
consisting of two transitions due to spin-rotation splitting, and
the qP2(10) transition at 6623.57 cm−1, consisting of four
transitions due to spin-rotation splitting. The transitions are
ΔKa
labeled according to
ΔJK″ (N″) where N is the rotational
a
angular momentum quantum number, J is the sum of the
rotational and spin angular momentum quantum numbers
given by N 1/2, and Ka is the quantum number for rotation
about the a axis of the molecule as defined in ref 26.
Figure 6. Lorentzian hwhm from the absorption spectra as a function
of pressure for 25 Torr of CH4 broadened by He.
The ECDL was scanned with a sinusoidal ramp of 1.8 V
peak-to-peak amplitude at a frequency of 0.532 Hz, giving a
scan range of 12 GHz. Figure 9 shows the fm-NICE-OHMS
spectra of HO2 at 6623.32 and 6623.57 cm−1, with peak
absorption cross sections of σ = 9.7 × 10−20 cm2 and σ = 1.3 ×
10−19 cm2, respectively.35 In the case of HO2, we were able to
subtract a background measurement by just turning off the UV
lamps, which resulted in a HO2-free background after 10 s. No
absorption features due to methanol were observed in the
spectral range, but the effect of etalons within the setup on the
spectra is clearly visible. A fit to the 6623.32 cm−1 absorption
feature is shown in red in Figure 9a, but indicates a problem.
With the Gaussian width constrained to the Doppler-
broadened value [426 MHz (fwhm)], the returned Lorentzian
component is close to zero, implying no substantial pressure
broadening at 40 Torr of Ar. Thiebaud et al.35 measured a
fwhm for this transition as 660 MHz at 50 Torr He, and we
would expect at least a similar pressure-broadening contribution
under our conditions. The HO2 measurements were taken
toward the end of this study, when the diode laser was starting
to exhibit multimode behavior, and the calibrated frequency
scale (obtained from the spectrum analyzer traces) might be
inaccurate. Although the results discussed above and taken
Figure 7. Typical absorption fm-NICE-OHMS signal of 0.42 Torr of
N2O in 50 Torr of He, with a fit to the data and the associated
residuals plot.
(black) with the corresponding fit (red) for 0.42 Torr of N2O
broadened by 50 Torr of helium for the R28(e) transition in
the ν3 band. For the signals broadened by helium, again each
absorption-based fm-NICE-OHMS trace was fitted with the
Gaussian width fixed to the calculated value of 366 MHz, and
the Lorentzian component of the Voigt profile was allowed to
5096
dx.doi.org/10.1021/jp301038r | J. Phys. Chem. A 2012, 116, 5090−5099

The Journal of Physical Chemistry A
Article
used a peak cross section for HO2 obtained for 50 Torr of He,35
and this might change somewhat for 40 Torr of Ar. The
concentration of HO2 was determined to be 6 × 1011 cm−3,
close to the expected value for HO2 of 5.8 × 1011 cm−3 based
on the simulations. Using this calibrated HO2 concentration
and the signal-to-noise ratio of 38 gave a minimum detection
sensitivity of 1.8 × 10−9 cm−1, and a minimum detectable
concentration of ∼4 × 1010 radicals cm−3 is inferred.
Comparing this value with those given in the Introduction, it
can be seen that the sensitivity reported here is an order of
magnitude lower than those reported for fm-NICE-OHMS
previously. These measurements were conducted toward the
end of the lifetime of the laser source, which was suffering
degradation from repeated modulation over several years. In
addition, the setup was also not working under optimal
conditions, as one of the piezoelectric transducers received an
erroneously large voltage that reduced its performance and
meant that no wm-NICE-OHMS could be performed at this
stage.
As a further confirmation that the species produced and
detected from the photolysis experiments in the cavity was
HO2, two kinetic analyses were undertaken. Analysis of the
time taken for the HO2 signal to disappear following the
cessation of the chlorine flow and analysis of the time taken for
the HO2 signal to disappear following switch-off of the UV
lamps were investigated. The input flow rates for the flow
controllers were set so that the residence time in the cavity was
120 s. Once all of the flows were on and the laser was locked
and stable with fm-NICE-OHMS data being collected on HO2,
the chlorine flow was ceased. As the lamps remained on during
this time, the photolysis reaction of Cl2 continued, and it is this
reaction that drove the formation of HO2 until all of the Cl2
was used or pumped away. The laser remained locked, with all
of the other flows on, and data from scans over the same HO2
qP1(12) transition at 6623.32 cm−1 were recorded at
approximately 20-s intervals. These data are shown in Figure
10 and have been offset in the y direction for clarity. It can be
Figure 9. fm-NICE-OHMS spectra of HO2 on the (a) qP1(12)
transition at 6623.32 cm−1 and (b) qP2(10) transition at 6623.57 cm−1,
each with a fit to the data shown in red.
earlier in the study for other molecules returned sensible
linewidths and pressure-broadening parameters, we concluded
that our frequency calibration might be in error for the HO2
q
measurements. For the P2(10) transition at 6623.57 cm−1, a
greater linewidth [Gaussian component of 576
15 MHz
(fwhm)] was obtained (Figure 9b), appearing to show the
effects of the unresolved spin-rotation splitting. The ratio of
linewidths of these two features, 1.36, is very similar to that
found by Thiebaud et al. (1.35) and would be consistent with
our present results showing the effect of an incorrect frequency
scale.
Table 1 shows that formaldehyde is a stable photolysis
product and that it absorbs weakly in the 6623 cm−1 range. A
spectrum of pure H2CO (50 mTorr) showed absorption
features that did not coincide with those in Figure 9 but that
would be observed in the frequency range scanned if the
concentration was above the measured H2CO detection limit of
6 × 1013 cm−3. Modeling using the purely gas-phase kinetic
scheme of Table 1 predicted the concentration of H2CO to be
2 × 1014 cm−3, a concentration that should have been
observable. The absence of such absorption features suggests
that the modeled concentration is too high, and inclusion of a
wall-loss term58 was found to decrease the concentration below
the detection limit for values of the sticking coefficient above
10−5. Formaldehyde is known to interact strongly with
surfaces,66 and values of >10−5 are to be expected.
The data were calibrated against a known concentration of
CH4 in the same spectral range and taken at a similar time to
the HO2 measurement. The methane transition chosen was at
6623.19 cm−1, with a peak absorption cross section of 1.69 ×
10−23 cm2 under Doppler-broadened conditions. With both
data sets being taken at the same phase (absorption), the
relative heights of the signals and the peak absorption cross
sections were compared for the calibration. We note that we
Figure 10. fm-NICE-OHMS data recorded over 2 min following the
cessation of chlorine flow (data offset in the y direction for clarity).
seen from the data that, over the 120 s following the cessation
of the chlorine flow, the concentration of HO2 dropped steadily
as the remaining chlorine in the cavity was pumped out.
The measurement of the decay of the HO2 absorption
feature after the lamps were switched off is shown in Figure 11.
Error bars are shown in the figure; however, the error bars are
small and so cannot be clearly seen. Also shown are two
simulations of the decay starting from a HO2 concentration of 5
× 1011 cm−3: one from the kinetic scheme shown in Table 1
5097
dx.doi.org/10.1021/jp301038r | J. Phys. Chem. A 2012, 116, 5090−5099

The Journal of Physical Chemistry A
Article
modulated triplet can be minimized by using appropriate types
of EOMs.49 Furthermore, it was shown that using an EOM with
active feedback significantly reduces drifts and noise in the
system, as was also previously shown by Bell et al.50 Higher
powers would also improve the signal size with an associated
reduction in the shot-noise limit for the setup. The current
sensitivity was achieved with a cavity of relatively low finesse,
which could be improved by the use of higher-reflectivity
mirrors; it should be noted, however, that this will change the
locking response of the cavity. Although the implementation of
a narrow-linewidth EDFL would increase the ease of locking,
the spectral coverage limitations of such laser sources might
hamper the application of such a laser in the atmospheric-
pressure-broadened regime.
AUTHOR INFORMATION
Corresponding Author
■
Figure 11. HO2 concentration versus time after the UV lamps were
turned off, together with the modeled HO2 concentrations under two
conditions, namely, with (dotted blue line) and without (solid red
line) a wall-loss term (see text).
*E-mail: grant.ritchie@chem.ox.ac.uk. Phone: +44 (0)1865
285723. Fax: +44 (0)1865 275410.
Notes
The authors declare no competing financial interest.
and one including the maximum possible wall-loss rate (when
diffusion to the walls is the rate-limiting step and wall-loss
coefficients are relatively high). Although the simulation
without wall loss provides a better fit, it is clear that the
simulations are in good agreement with the experimental data.
ACKNOWLEDGMENTS
■
This work was conducted under the EPSRC Programme Grant
EP/E019765/1: New frontiers in quantitative infra-red to
ultraviolet spectroscopy using diode and quantum cascade
lasers. The authors gratefully acknowledge the New Zealand
Foundation for Research, Science and Technology (Contract
UOXX0501) for supporting N.J.v.L. C.L.B. is grateful to the
NERC for a studentship. We are particularly grateful to a
reviewer for making most helpful suggestions and improve-
ments to this article.
CONCLUSIONS AND OUTLOOK
■
This work has demonstrated the use of an external-cavity
diode-laser-based NICE-OHMS system and its application to
detection of a range of species of atmospheric importance. The
spectrometer was successfully used to make the first measure-
ments of the HO2 radical species in a laboratory setting,
marking a first step in applying this ultrasensitive spectrometer
to real atmospheric scenarios. One of the key features of any
detection technique is that it yields a direct and speciated
solution. It has been demonstrated that the NICE-OHMS
spectrometer can provide this for HO2.
REFERENCES
■
(1) Heard, D. E. Annu. Rev. Phys. Chem. 2006, 57, 191−216.
(2) Heard, D. E.; Pilling, M. J. Chem. Rev. 2003, 103, 5163−5198.
(3) Clemitshaw, K. C. Crit. Rev. Environ. Sci. Technol. 2004, 34, 1−
108.
(4) Heard, D. E. Stratospheric chemistry and composition: Hydroxyl
radical. In Encyclopedia of Atmospheric Sciences; Holton, J. R., Pyle, J. A.,
Curry, J. A., Eds.; Elsevier: Amsterdam, 2003; pp 2185−2193.
(5) Monks, P. S. Chem. Soc. Rev. 2005, 34, 376−395.
(6) Parra, J.; George, L. A. Appl. Opt. 2009, 48, 3355−3361.
(7) Dusanter, S.; Vimal, D.; Stevens, P. S.; Volkamer, R.; Molina, L.
T. Atmos. Chem. Phys. 2009, 9, 1665−1685.
(8) Whalley, L. K.; Furneaux, K. L.; Goddard, A.; Lee, J. D.; Mahajan,
A.; Oetjen, H.; Read, K. A.; Kaaden, N.; Carpenter, L. J.; Lewis, A. C.;
Plane, J. M. C.; Saltzman, E. S.; Wiedensohler, A.; Heard, D. E. Atmos.
Chem. Phys. 2010, 10, 1555−1576.
(9) Cantrell, C. A.; Stedman, D. H.; Wendel, G. J. Anal. Chem. 1984,
56, 1496−1502.
(10) Wood, E. C.; Cohen, R. C. Fluorescence Methods. In Analytical
Techniques for Atmospheric Measurement; Heard, D. E., Ed.; Blackwell
Publishing: Oxford, U.K., 2006; Chapter 4, pp 189−228.
(11) Clemitshaw, K. C.; Carpenter, L. J.; Penkett, S. A.; Jenkin, M. E.
J. Geophys. Res. 1997, 102, 25405−25416.
(12) Green, T. J.; Reeves, C. E.; Brough, N.; Edwards, G. D.; Monks,
P. S.; Penkett, S. A. J. Environ. Monit. 2003, 5, 75−83.
(13) Liu, Y.; Morales-Cueto, R.; Hargrove, J.; Medina, D.; Zhang, J.
Environ. Sci. Technol. 2009, 43, 7791−7796.
(14) Fuchs, H.; Bohn, B.; Hofzumahaus, A.; Holland, F.; Lu, K. D.;
Nehr, S.; Rohrer, F.; Wahner, A. Atmos. Meas. Tech. 2011, 4, 1209−
1225.
The experimental sensitivity limit achieved for detection of
HO2 was 1.8 × 10−9 cm−1 corresponding to ∼4 × 1010 radicals
cm−3, which is not sufficiently sensitive to achieve detection of
atmospheric levels. There is significant and real potential that
these limits can be met with reasonably attainable improve-
ments to the experimental system. In the current setup, there
are two main limitations: the laser source and the mirrors. With
improvements to the laser source to return the system to its
optimal operating conditions, NICE-OHMS offers a viable,
speciated, and compact spectrometer for application to
atmospheric remote sensing. Although having a broad range,
with a tunable ECDL allowing greater flexibility in the selection
of the detection wavelength, the relatively large linewidth (2
MHz) of the laser locking system adds a further level of
experimental complexity to the system, and maintaining the
lock of the laser to the cavity during the wavelength scanning
for data acquisition is difficult. With the implementation of an
improved laser source, such as an erbium-doped fiber laser
(EDFL), this situation would be improved, as the much
narrower linewidth of the laser would allow it to be locked
more easily to the cavity. In combination with fiber-coupled
EOMs, this could lead to a compact and mobile fiber-based
NICE-OHMS system. In a recent work, it was shown that the
background signals originating from an imbalance of the phase-
(15) Miyazaki, K.; Parker, A. E.; Fittschen, C.; Monks, P. S.; Kajii, Y.
Atmos. Meas. Tech. 2010, 3, 1547−1554.
5098
dx.doi.org/10.1021/jp301038r | J. Phys. Chem. A 2012, 116, 5090−5099

The Journal of Physical Chemistry A
Article
(16) Jenkin, M. E.; Cox, R. A.; Hayman, G. D.; Whyte, L. J. J. Chem.
Soc., Faraday Trans. 2 1988, 84, 913−930.
(17) Crowley, J. N.; Simon, F. G.; Burrows, J. P.; Moortgat, G. K.;
Jenkin, M. E.; Cox, R. A. J. Photochem. Photobiol. A 1991, 60, 1−10.
(18) Jemi-Alade, A. A.; Lightfoot, P. D.; Lesclaux, R. Chem. Phys. Lett.
1992, 195, 25−30.
(19) Tomas, A.; Villenave, E.; Lesclaux, R. J. Phys. Chem. A 2001,
105, 3505−3514.
(20) Nielsen, O. J.; Johnson, M. S.; Wallington, T. J.; Christensen, L.
K.; Platz, J. Int. J. Chem. Kinet. 2002, 34, 283−291.
(21) Christensen, L. E.; Okumura, M.; Sander, S. P.; Salawitch, R. J.;
Toon, G. C.; Sen, B.; Blavier, J. F.; Jucks, K. W. Geophys. Res. Lett.
2002, 29, 1299.
(51) Drever, R. W. P.; Hall, J. L. H.; Kowalski, F. V.; Hough, J.; Ford,
G. M.; Munley, A. J.; Ward, H. Appl. Phys. B: Lasers Opt. 1983, 31,
97−105.
(52) DeVoe, R. G.; Brewer, R. G. Phys. Rev. A: At., Mol., Opt. Phys.
1984, 30, 2827−2829.
(53) Sander, S. P.; Abbatt, J.; Barker, J. R.; Burkholder, J. B.; Friedl, R.
R.; Golden, D. M.; Huie, R. E., Kolb, C. E., Kurylo, M. J.; Moortgat, G.
K.; Orkin, V. L.; Wine, P. H. Chemical Kinetics and Photochemical Data
for Use in Atmospheric Studies, Evaluation No. 17; JPL Publication 10-6;
Jet Propulsion Laboratory: Pasadena, CA, 2011; available at http://
jpldataeval.jpl.nasa.gov.
(54) Tsang, W. J. J. Phys. Chem. Ref. Data 1987, 16, 471−508.
(55) Pagsberg, P.; Munk, J.; Sillesen, A.; Anastasi, C. Chem. Phys. Lett.
1988, 146, 375−381.
(56) Tyndall, G. S.; Orlando, J. J.; Kegley-Owen, C. S.; Wallington,
T. J.; Hurley, M. D. Int. J. Chem. Kinet. 1999, 31, 776−784.
(57) Burrows, J. P.; Tyndall, G. S.; Moortgat, G. K. J. Phys. Chem.
1985, 89, 4848−4856.
(22) Stone, D.; Rowley, D. M. Phys. Chem. Chem. Phys. 2005, 7,
2156−2163.
(23) Johnson, T. J.; Wienhold, F. G.; Burrows, J. P.; Harris, G. W.;
Burkhard, H. J. Phys. Chem. 1991, 95, 6499−6502.
(24) Fink, E. H.; Ramsay, D. A. J. Mol. Spectrosc. 1997, 185, 304−324.
(25) Taatjes, C. A.; Oh, D. B. Appl. Opt. 1997, 36, 5817−5821.
(26) DeSain, J. D.; Hob, A. D.; Taatjesb, C. A. J. Mol. Spectrosc. 2003,
219, 163−169.
(58) Chantry, P. J. J. Appl. Phys. 1987, 62, 1141−1148.
(59) Rothman, L. S.; Jacquemart, D.; Barbe, A.; Benner, D. C.; Birk,
M.; Brown, L. R.; Carleer, M. R.; Chackerian, C., Jr.; Chance, K.;
Coudert, L. H.; Dana, V.; Devi, V. M.; Flaud, J.-M.; Gamache, R. R.;
Goldman, A.; Hartmann, J.-M.; Jucks, K. W.; Maki, A. G.; Mandin, J.-
Y.; Massie, S. T.; Orphal, J.; Perrin, A.; Rinsland, C. P.; Smith, M. A.
H.; Tennyson, J.; Tolchenov, R. N.; Toth, R. A.; Vander Auwera, J.;
Varanasi, P.; Wagner, G. J. Quant. Spectrosc. Radiat. Transfer 2005, 96,
139−204.
(27) Christensen, L. E.; Okumura, M.; Sander, S. P.; Friedl, R. R.;
Miller, C. E.; Sloan, J. J. J. Phys. Chem. 2004, 108, 80−91.
(28) Kanno, N.; Tonokura, K.; Tezaki, A.; Koshi, M. J. Phys. Chem. A
2005, 109, 3153−3158.
(29) Kanno, N.; Tonokura, K.; Tezaki, A.; Koshi, M. J. Mol. Spectrosc.
2005, 229, 193−197.
(30) Christensen, L. E.; Okumura, M.; Hansen, J. C.; Sander, S. P.;
Francisco, J. S. J. Phys. Chem. A 2006, 110, 6948−6959.
(31) Taatjes, C. A. J. Phys. Chem. A 2006, 110, 4299−4312.
(32) Thiebaud, J.; Fittschen, C. Appl. Phys. B: Lasers Opt. 2006, 85,
383−389.
(60) Campargue, A.; Wang, L.; Liu, A. W.; Hu, S. M.; Kassi, S. Chem.
Phys. 2010, 373, 203−210.
(61) Predoi-Cross, A.; Brown, L. R.; Malathy Devi, V.; Brawley-
Tremblay, M.; Chris Benner, D. J. Mol. Spectrosc. 2005, 232, 231−246.
(62) Smith, M. A. H.; Chris Benner, D.; Predoi-Cross, A.; Malathy
Devi, V. J. Quant. Spectrosc. Radiat. Transfer 2010, 111, 1152−1166.
(33) Suzaki, K.; Kanno, N.; Tonokura, K.; Koshi, M.; Tsuchiya, K.;
Tezaki, A. Chem. Phys. Lett. 2006, 425, 179−184.
(34) Ibrahim, N.; Thiebaud, J.; Orphal, J; Fittschen, C. J. Mol.
Spectrosc. 2007, 242, 64−69.
́
(63) Zeninari, V.; Parvitte, B.; Courtois, D.; Kapitanov, V. A.;
Ponomarev, Yu. N. Appl. Phys. B: Lasers Opt. 2001, 72, 953−959.
(64) Yamano, D.; Sakamato, Y.; Yabushita, A.; Kawasaki, M.; Morino,
I.; Inoue, G. Appl. Phys. B: Lasers Opt. 2009, 97, 523−528.
(35) Thiebaud, J.; Crunaire, S.; Fittschen, C. J. Phys. Chem. A 2007,
111, 6959−6966.
(65) Nakayama, T.; Fukuda, H.; Sugita, A.; Hashimoto, S.; Kawasaki,
M.; Aloisio, S.; Morino, I.; Inoue, G. Chem. Phys. 2007, 334, 196−203.
(66) Dahnke, H.; Von Basum, G.; Kleinermanns, K.; Hering, P.;
(36) Aluculesei, A.; Tomas, A.; Schoemaecker, C.; Fittschen, C. Appl.
Phys. B: Lasers Opt. 2008, 92, 379−385.
Murtz, M. Appl. Phys. B: Lasers Opt. 2002, 75, 311−316.
̈
(37) Tang, Y.; Tyndall, G. S.; Orlando, J. J. J. Phys. Chem. A 2010,
114, 369−378.
(38) Zahniser, M. S.; Stanton, A. C. J. Chem. Phys. 1984, 80, 4951−
4960.
(39) Zahniser, M. S.; McCurdy, K. E.; Stanton, A. C. J. Phys. Chem.
1989, 93, 1065−1070.
(40) Miyano, S.; Tonokura, K. J. Mol. Spectrosc. 2011, 265, 47−51.
(41) Sakamoto, Y.; Tonokura, K. J. Phys. Chem. A 2012, 116, 215−
222.
(42) Thiebaud, J.; Parker, A.; Fittschen, C.; Vincent, G.; Zahraa, O.;
Marquaire, P.-M. J. Phys. Chem. C 2008, 112, 2239−2243.
(43) van Helden, J. H.; Peverall, R.; Ritchie, G. A. D. Cavity
enhanced techniques using continuous wave lasers. In Cavity Ring-
Down Spectroscopy: Techniques and Applications; Engeln, R.; Berden,
G., Eds; Wiley-Blackwell: Oxford, U.K., 2009; Chapter 2, pp 27−56.
(44) Ye, J.; Ma, L.-S.; Hall, J. L. J. Opt. Soc. Am. B 1998, 15, 6−15.
(45) Ma, L.-S.; Ye, J.; Dube, P.; Hall, J. L. J. Opt. Soc. Am. B 1999, 16,
2255−2268.
(46) van Leeuwen, N. J.; Wilson, A. C. J. Opt. Soc. Am. B 2004, 21,
1713−1721.
(47) Schmidt, F.; Foltynowicz, A.; Ma, W.; Lock, T.; Axner, O. Opt.
Express 2007, 15, 10822−10831.
(48) Foltynowicz, A.; Wang, J.; Ehlers, P.; Axner, O. Opt. Express
2010, 18, 18580−18591.
(49) Foltynowicz, A.; Silander, I.; Axner, O. J. Opt. Soc. Am. B 2011,
28, 2797−2805.
(50) Bell, C. L.; Hancock, G.; Peverall, R.; Ritchie, G. A. D.; van
Helden, J. H.; van Leeuwen, N. J. Opt. Express 2009, 17, 9834−9839.
5099
dx.doi.org/10.1021/jp301038r | J. Phys. Chem. A 2012, 116, 5090−5099