Use of broadband, continuous-wave diode lasers in cavity ring-down spectroscopy for liquid samples.
ABSTRACT Cavity ring-down spectroscopy (CRDS) is an extremely sensitive absorption technique that has been applied primarily to gas samples, which are characterized by having narrow absorption features. Recently, CRDS has also been applied to liquid samples, which have broad absorption features. The use of small inexpensive diode lasers as light sources for liquid samples is demonstrated. The low cost coupled with the ease and technical straightforwardness of application gives this technique wide appeal.
- [Show abstract] [Hide abstract]
ABSTRACT: Continuous-wave (cw) cavity-ringdown (CRD) spectroscopy provides a highly sensitive way to measure optical absorption by observing the decay rate of light from a high-finesse optical cavity containing the sample of interest (usually gas-phase molecules). In rapidly swept cw-CRD spectroscopy, optical build-up and subsequent ringdown decay are initiated by rapidly sweeping the cavity length or the wavelength of the monochromatic tunable cw laser radiation, thereby establishing and interrupting optical resonance between the laser light and the longitudinal-mode frequencies of the cavity. We review the experimental methodology and applications of this technique, indicating its advantages and prospects for spectroscopic sensing.Chemical Physics Letters 01/2011; 512(1):1-20. · 2.15 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Plasma-cavity ringdown spectroscopy (P-CRDS) was first introduced for trace elemental measurements in 1997. During the last decade, this technique has rapidly advanced from the initial demonstration of the basic concept to the latest (prototype) instrumentation. P-CRDS has been demonstrated to be a powerful technique for ultra-sensitive elemental and isotopic analysis of trace metals, radionuclides, and isotopes existing in solid, vapor, and/or liquid phases. Utilizing various plasma sources, such as an inductively coupled plasma (ICP), a microwave-induced plasma (MIP), and a tube-shaped MIP, and different laser sources, namely, pulsed and continuous wave lasers, several elements and isotopes, including Pb, 238U, 235U, Hg, Sr, and Mn, as well as the fine structures of Hg, have been measured with P-CRDS, and the detection limits have ranged from μg mL−1 to pg mL−1 levels. This paper presents a critical review of the P-CRDS technique with an emphasis on its application in elemental and isotopic analysis. Some critical issues encountered in the application of P-CRDS and its technological development during the last ten years, such as sensitivity, spectral interference, plasma optimization, evolution of plasma sources, laser beam behavior in the plasma, laser source effects, and instrument configurations are discussed. Technological and scientific barriers to be overcome for eventual instrumentation in the field of atomic/isotopic spectrometry, including (1) portable UV laser sources, (2) novel plasma sources, and (3) detection of multiple analytes, are also discussed.Journal of Analytical Atomic Spectrometry 01/2007; 22(11). · 3.40 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: A new application of incoherent broad-band cavity enhanced absorption spectroscopy (IBBCEAS) to weak transitions in solution through a very straightforward modification of commercially available double-beam UV/VIS absorption spectrometers is reported. The improved sensitivity of the new approach is demonstrated on basis of the weak Franck-Condon inhibited absorption of the fifth C-H stretch overtone in liquid benzene. The theoretical limits of the enhancement of the signal-to-noise ratio of IBBCEAS in comparison with single pass absorption experiments are discussed for a set of given experimental cavity parameters. The optical loss properties of a typical transparent cuvette window in the cavity are also discussed.Review of Scientific Instruments 01/2005; 76. · 1.60 Impact Factor
Volume 57, Number 5, 2003APPLIED SPECTROSCOPY
q 2003 Society for Applied Spectroscopy
Use of Broadband, Continuous-Wave Diode Lasers in Cavity
Ring-Down Spectroscopy for Liquid Samples
A. J. HALLOCK, E. S. F. BERMAN, and R. N. ZARE*
Department of Chemistry, Stanford University, Stanford, California 94305-5080
Cavity ring-down spectroscopy (CRDS) is an extremely sensitive
absorption technique that has been applied primarily to gas sam-
ples, which are characterized by having narrow absorption features.
Recently, CRDS has also been applied to liquid samples, which have
broad absorption features. The use of small inexpensive diode lasers
as light sources for liquid samples is demonstrated. The low cost
coupled with the ease and technical straightforwardness of appli-
cation gives this technique wide appeal.
Index Headings: Absorption; Cavity ring-down spectroscopy;
CRDS; Diode laser; Liquids.
Cavity ring-down spectroscopy (CRDS) is a relatively
new technique heralded as an extremely sensitive mea-
sure of absorption.1It has generally been applied to mol-
ecules in the gas phase with excellent results.2CRDS
employs highly re?ective mirrors to form an optical cav-
ity. Laser light is allowed to ?ll the cavity and is then
shuttered. A photomultiplier or photodetector is used to
detect the light as it leaks out of the cavity through the
back mirror. The intensity of the detected light, the ring-
down signal, decays exponentially with time constant t,
which depends only on the characteristics of the cavity
and any absorber within the cavity. Insensitivity to ?uc-
tuations in input laser intensity and a massive increase in
effective pathlength owing to the multipass nature of the
method provide the large increase in sensitivity relative
to a traditional absorption measurement. Almost all
CRDS studies have been performed on molecules in the
gas phase. We report here a promising approach for car-
rying out cavity ring-down spectroscopy in liquid sam-
ples cheaply and easily. Because of the large linewidths
of absorbers in liquids, we are able to develop a much-
simpli?ed approach to CRDS measurements.
Of critical concern is the linewidth of the laser com-
pared with the linewidth of the absorption feature of in-
terest. Use of a laser with a linewidth wider than or even
comparable to that of the absorption feature produces sys-
tematic errors in the concentrations of absorber calculated
Received 21 November 2002; accepted 10 January 2003.
* Author to whom correspondence should be sent.
using a simple exponential-?t model to the data.3Of the
two possible solutions to this problem, most experimen-
talists have chosen to use narrow-linewidth lasers rather
than more complicated ?tting routines.4Absorption line-
widths in the gas phase are typically 0.1 cm21or narrow-
er, so a very narrow linewidth laser is necessary.
Typical light sources used for gas-phase CRDS are
thus either expensive pulsed lasers or very narrowband,
continuous-wave (cw) diode lasers. The tunable pulsed
lasers typically cover an entire bench top, require massive
amounts of electricity to run, and cost several hundred
thousand dollars. The narrowband diode lasers cost com-
paratively less (several thousand dollars) but require so-
phisticated instrumentation to couple the light into the
Recently, we have been successful in adapting CRDS
for use in liquids with a conventional Nd : YAG pumped
dye laser.6Absorption linewidths in the liquid phase are
much wider than those in the gas phase, on the order of
1000 cm21. Consequently, in the current work we were
able to employ a hand-held, broadband, inexpensive
($500) diode laser capable of running on just a few volts
of electricity. In addition, little sophisticated instrumen-
tation was needed to couple the laser light into the cavity
other than an optical chopping device.
The laser beam may follow many stable trajectories in
an optical cavity. These are generally divided into axial
and transverse modes. Axial modes are different frequen-
cies that ‘‘?t’’ in the cavity; a standing wave solution
requires a half-integer number of wavelengths to ?t be-
tween the boundary conditions imposed by the mirrors.
Transverse modes are different cross-sectional shapes. In
the case of cylindrical symmetry, as in our experiment,
they are described by Gaussian–Laguerre polynomials.
These transverse modes are the well-known TEMpl
Each of the different modes supported by the optical
cavity has a slightly different ring-down time. Neverthe-
less, under many experimental conditions the ring-down
time constant, t, can be well approximated by:
I 5 Ioexp(2t/t)
where I is the intensity of the transmitted light, t is time,
and t is the decay constant.8By carefully directing the
laser light into the cavity, we are able to excite mostly
Volume 57, Number 5, 2003
655 nm. The falling edge is expanded to show the exponential decay
of the ring-down signal.
Ring-up and ring-down signals taken in pure acetonitrile at
acousto-optic modulator. An iris selects the ?rst-order beam. A photo-
multiplier records the decay, which is sent to an oscilloscope and com-
puter for processing.
Light from the laser is focused by a lens and chopped by an
acid (0.33 mM) in a solution of 20% toluene/80% acetonitrile monitored
at 655 nm. The ?t shows an exponential loss of absorber concentration
The reduction of methylene blue (3.0 nM MB1) by ascorbic
the TEM00transverse mode. This is the circular mode that
exhibits the least amount of diffraction loss. Various
schemes have been employed to excite a single axial
mode of an optical cavity,5to average over multiple axial
modes,2or to excite off-axis to avoid cavity mode struc-
ture with a concomitant loss of intensity.9Our system is
the type that averages over many axial modes because
the linewidth of the laser is much larger than the spacing
between cavity axial modes. In this respect our setup is
entirely equivalent to those using a pulsed system.
Although the power of the laser does not affect the
exponential decay constant t, it does affect the signal-to-
noise ratio on the detector. One might therefore expect
that a relatively low-power diode laser would be unsuit-
able for CRDS. However, long pulses may constructively
interfere over multiple re?ections to store power in the
cavity. Figure 1 shows intensity building up in our cavity
and then ringing down.
The diode laser (Power Technology Inc., PM series,
655 nm) is driven by 5 V at 120 mA to produce approx-
imately 35 mW of power. The laser is quite small, 12.60
3 50.93 mm, and is plugged into a normal wall socket.
We used an acousto-optic modulator (AOM; Brimrose
TM 200-58-658) to switch the light on and off to perform
the experiments. The optical cavity is constructed from
two highly re?ective mirrors (Newport Supermirrors, R
ø 99.9%). Figure 2 shows the setup in more detail. We
separated the ?rst-order modulated beam (approximately
5 mW) with an iris and directed this beam into the optical
cavity. The maximum switching rate is determined not
by the AOM controller but rather by the amount of time
needed for intensity to build up in the cavity and then
ring down for each measurement. Because we work pri-
marily in regions where decay constants are on the order
of 100 ns (due to absorption by the solvent), we can
switch as fast as 1 MHz. This signal-sampling rate is in
sharp contrast to the typical pulsed setup, which typically
runs at speeds on the order of 10 Hz. This massive in-
crease in repetition rate is one of the sources of increased
sensitivity with time for cw-CRDS. Narrowband cw di-
ode lasers also enjoy this increase in repetition rate. With
this particular experimental setup, we are limited by the
data transfer rate of our oscilloscope (LeCroy LT372) to
a data acquisition rate of 10 kHz. In future work this
limitation will be eliminated by use of analog signal pro-
cessing. In future work we also expect to address the
issue that the current diode laser operates at a single
wavelength, either by employing a tunable diode or by
coupling many diodes at once using ?ber optics.
RESULTS AND DISCUSSION
In this study we use a broadband, continuous-wave di-
ode laser as the light source for CRDS. The measured
ring-down lifetime varies by less than 2% (st/t) over
several hours. This variation translates to a minimum de-
tectable change in absorption coef?cient of 1 part in 3 3
106cm21(one s). This corresponds to detecting ppb to
ppt for good absorbers. This detection limit is only slight-
ly higher than the limit reported for a standard CRDS
setup with a pulsed dye laser under similar conditions.5
Changes in experimental setup, particularly solvent and
wavelength, have a large impact on the detection limit;
therefore, future work is expected to improve the detec-
tion limit greatly.
Given the high repetition rate, our new methodology
can be used to follow processes with time resolution as
?ne as microseconds. We have been able to monitor the
kinetics of reactions in which an absorber is either created
or destroyed under a variety of conditions. Figure 3
shows the decay in concentration of methylene blue as it
undergoes a reduction in the presence of ascorbic acid to
form leucomethylene blue, which does not absorb. We
feel that the capability to monitor reactions at nanomolar
concentrations with microsecond time resolution is very
promising. In general, the study of reactivity over a broad
over several minutes. The ?t is to a simple diffusion model. The oscil-
lations, therefore, are not related to the diffusion process but are be-
lieved to be the result of photothermal beam steering caused by an
inhomogeneous absorber distribution.
Molecules of methylene blue are diffusing into the probe beam
to switch the laser. A higher-resolution picture of a falling edge of one
such ring-down pro?le shows an exponential ?t from which t may be
Waveform produced by allowing feedback from re?ected light
range of concentration and time regimes is important be-
cause elucidation of kinetic details often contributes
greatly to mechanistic understanding.
Because CRDS is a line-of-sight technique, it can also
be used to record events with spatial resolution. Methy-
lene blue added to a volume of acetonitrile diffuses into
the probe beam. Figure 4 shows the diffusion process
over a few minutes. The ?t was produced by equations
that describe the diffusion of an instantaneous point
source of heat in a metallic cylinder,10which have been
easily adapted to describe diffusion of molecules in so-
It is possible to simplify the experiment even further
by removing the optical switch necessary to create reg-
ular pulses of light. In this case, light that does not couple
into the optical cavity re?ects directly back into the laser,
causing chaotic instability in the lasing.11We can then
take advantage of this self-pulsation-type effect.12Each
time the laser stops producing light, a ring down can be
measured as before. Under our conditions, the switching
occurs at random times on the order of once a millisec-
ond. This treatment does not adversely affect the laser in
any permanent way. It makes the apparatus simpler and
less expensive by removing the need to purchase and
align a fast optical switch. Typical results are shown in
We are able to record cavity ring-down spectra in liq-
uid samples utilizing an inexpensive, broadband cw diode
laser. This device, unsuitable for gas-phase measure-
ments, is readily adapted to liquids because of the intrin-
sically large absorption features in the latter. We are able
to follow reactions at very low concentrations. This study
demonstrates an inexpensive, easy to implement, and sen-
sitive form of cavity ring-down spectroscopy that can be
applied to monitor absorbing species in liquid samples.
A. J. H. is grateful for a Veach Memorial Fellowship and E. S. F. B.
is grateful for an NSF Graduate Research Fellowship. This work is
supported by a grant from the Of?ce of Naval Research (Grant N00014-
00-1-0364) and by Hamamatsu Corp.
1. A. O’Keefe and D. A. G. Deacon, Rev. Sci. Instrum. 59, 2544
2. G. Berden, R. Peters, and G. Meijer, Int. Rev. Phys. Chem. 19, 565
3. J. T. Hodges, J. P. Looney, and R. D. van Zee, Appl. Opt. 35, 4112
4. A. P. Yalin and R. N. Zare, Laser Phys. 12, 1065 (2002).
5. B. A. Paldus, C. C. Harb, T. G. Spence, B. Wilke, J. Xie, J. S.
Harris, and R. N. Zare, J. Appl. Phys. 83, 3991 (1998).
6. A. J. Hallock, E. S. F. Berman, and R. N. Zare, Anal. Chem. 74,
7. H. Kogelnik and T. Li, Proc. IEEE 54, 1312 (1966).
8. P. Zalicki and R. N. Zare, J. Chem. Phys. 102, 2708 (1995).
9. J. B. Paul, L. Lapson, and J. G. Anderson, Appl. Opt. 40, 4904
10. H. S. Carslaw and J. C. Jaeger, Conduction of Heat in Solids (Ox-
ford University Press, New York, 1986).
11. G. C. Dente, P. S. Durkin, K. A. Wilson, and C. E. Moeller, IEEE
J. Quantum Electron. 24, 2441 (1988).
12. K. Otsuka and H. Iwamura. Phys. Rev. A 28, 3153 (1983).