Photochemical decomposition of N2O by Lyman-α radiation: Scientific basis for a chemical actinometer

Article abstract of DOI:10.1021/jp9093436

A novel IR method for measuring the kinetics of N₂O photodecomposition has been devised and used to calibrate the flux of Lyman-α (10.2 eV) radiation from a H₂/Ar microwave discharge lamp. The photodecomposition of N₂O occurs with a weak pressure dependence due to the operation of a wall effect consuming some photogenerated active oxygen species. This effect is removed by working at high N₂O pressures. The Lyman-α flux from the lamp is 1.28 ± 0.36 × 10¹⁵ photons cm^-2 s^-1.

Full text of DOI:10.1021/jp9093436

J. Phys. Chem. A 2010, 114, 3443–3448
3443
Photochemical Decomposition of N O by Lyman-r Radiation: Scientific Basis for a
2
Chemical Actinometer
Mahesh Rajappan, Michael B u¨ ttner, Charlie Cox, and John T. Yates, Jr.*
Department of Chemistry, UniVersity of Virginia, CharlottesVille, Virginia 22904
ReceiVed: September 29, 2009; ReVised Manuscript ReceiVed: January 13, 2010
A novel IR method for measuring the kinetics of N
calibrate the flux of Lyman-R (10.2 eV) radiation from a H
photodecomposition of N O occurs with a weak pressure dependence due to the operation of a wall effect
consuming some photogenerated active oxygen species. This effect is removed by working at high N
2
O photodecomposition has been devised and used to
2
/Ar microwave discharge lamp. The
2
2
O
1
5
-2 -1
pressures. The Lyman-R flux from the lamp is 1.28 ( 0.36 × 10 photons cm s .
The N O molecule is selective for Lyman-R excitation since
1
I. Introduction
2
3-17
the absorption cross section above 1400 Å is negligible.
The measurement of the light flux from vacuum UV lamps
is of importance in a number of areas including astrochemistry
and planetary atmosphere chemistry. Different measurement
methods are used to calibrate such lamps, including photoelec-
tron generation, photoionization of gases, and photodecompo-
sition of gases. All such methods require knowledge of
fundamental issues having to do with the photoexcitation of
solid surfaces, the photoionization of gas molecules, or the cross
section for photodissociation of a particular molecule used as
an actinometer. This paper concerns the photodissociation of
II. Experimental Section
A. Lyman-r Lamp and Actinometer Cell. Figure 1 shows
a schematic diagram of the quartz Lyman-R lamp and the
infrared cell containing N O(g), whose rate of photodecompo-
2
sition is used for calibration of the lamp. A microwave discharge
in an Ar + H (90% Ar, 10% H ) flowing mixture at a pressure
2
2
of 1.00 ( 0.02 Torr was employed to produce ultraviolet light.
The microwave power level was 60 W in all measurements.
2
the N
using a novel IR spectroscopic measurement method for the
determination of the rate of N O decomposition and hence the
2
O molecule by Lyman-R radiation (1216 Å, 10.2 eV)
The lamp had a cross sectional area of 1 cm . The quartz lamp
was sealed to a MgF window on the cell by a Viton O-ring
2
2
compressed between the lamp and the window. The cutoff for
this window was at 1100 Å,18 effectively removing possible Ar
emission lines. The external IR beam from a Bruker Tensor 27
FTIR spectrometer crossed the actinometer cell width (12 cm)
into an external optical system containing a liquid-nitrogen-
cooled HgCdTe detector. The cell was pumped with a separate
light flux from a Lyman-R discharge lamp. The paper serves as
a basis for understanding the factors influencing the photode-
composition of N
actinometry.
2
O as these factors influence its use in
A conventional method for the estimation of the Lyman-R
photon flux makes use of the photoionization of the NO
-8
turbomolecular pump and achieved a base pressure of 1 × 10
1
-3
Torr. The base pressure in the IR cell was measured with a
molecule in the gas phase.
In addition, silicon photodiode
-
3
-9
4
,5
cold cathode gauge with a range of 10 -10 Torr. Through
a leak valve, a differentially pumped quadrupole mass spec-
trometer (SRS Model RGA 200) could be used to sample the
gas within the cell at high pressure. The gas pressure in the IR
cell was measured with a Baratron capacitance manometer
detectors have been used by several groups. The sensitivity
of Si photodiodes is strongly dependent on changes in the
reflectivity at the oxide/silicon interface during prolonged
exposure to light. There are also a few studies where photo-
electrons are measured from metal surfaces to estimate the
-
2
5
,6
(MKS Model 627B) (10 -100 Torr), and the initial pressure
of N O in the range 1-10 Torr was adjusted to (0.01 Torr for
photolysis experiments. The N O was obtained from Liquid
photon flux from Lyman-R sources.
2
There are also a few reports of the use of chemical
4
,7-11
2
actinometry for Lyman-R intensity calibration.
Here, one
Carbonic Specialty Gas Corporation (99.9% purity) and was
freeze-pump thawed prior to use. Infrared spectroscopy and mass
spectroscopy confirmed the gas purity. The measured volume
of the IR cell and its connecting tubing was 1.9 L. The mixing
is interested primarily in knowing the efficiency for all of the
elementary processes involved in destroying the molecule. We
summarize in Table 1 the output of Lyman-R lamps as measured
by different methods over a period of 47 years.
time for the photodecomposition product, NO with N
Torr pressure along the length of the cell, was estimated to be
from 3.4 to 34 s employing known collision cross sections.
2
O at 1-10
Other actinometers for Lyman-R flux measurement have
4
,11
involved the photolysis of an O
2
ice to make O
3
.
In this
1
9,20
example, the authors have assumed that the solid-phase quantum
yield (1.92) to produce O is the same as the gas-phase value.
Also, others have used the photolysis of CO for chemical
actinometry, but at the Lyman-R wavelength, the absorption
Because the time between spectral acquisitions was 600 s, it
was assumed that all measurements were made with complete
mixing of gases in the cell.
3
2
coefficient is quite low (1.97 cm- Torr ).
1
-1 12
The spectrum of a Lyman-R lamp similar to ours has been
7
published and is reproduced in Figure 2. In addition, the
absorption cross sections for N
2
O are shown in the inset to
*
To whom correspondence should be addressed. E-mail: johnt@
Figure 2, as published by others.1
3-17
It may be seen that
virginia.edu.
1
0.1021/jp9093436  2010 American Chemical Society
Published on Web 02/15/2010

3
444 J. Phys. Chem. A, Vol. 114, No. 10, 2010
Rajappan et al.
TABLE 1: Lyman-r Photon Flux Determination Reported in Literature Using Various Methods
relevant details of the VUV lamp
method of photon flux calibration
photon flux
reference
6
2 × 10 photons s-1
1
5
He-38% H
.15 Torr
2
gas mixture at
three different methods: (1) volume
photoelectric effect using Ni
0
photocathode, (2) oxygen actinometry,
(
3) oxalic acid actinometry
3 × 10 photons s-1
1
4
7
Ar-10% H
Torr
2
gas mixture at
relative peak heights of emission line on
monochromator with reference to
1
He-10% Kr lines 1236 and 1165 Å
calibrated using CO actinometry
2
1
4
He-2% H
2
gas mixture at 1 Torr
CO
2
actinometry made in a similar way
7.6 × 10 photons s-1
8
9
to that of H. Okabe
15
expanded Ar plasma jet admixed
with H
thermal detector similar to Coblentz
radiometer
2 × 10 photons cm-2 s-1
2
1 × 10 photons cm-2 s-1
1
5
10
5
H
2
discharge lamp
2
CO actinometry
5 × 10 photons cm-2 s-1
14
Ar-10% H
Torr
2
gas mixture at
photocurrent from aluminum grid placed
in the photon stream; this photocurrent
was calibrated against a commercial
photodiode
1
5 × 10 photons cm-2 s-1
1
4
11
4
Ar-10% H
Torr
2
gas mixture at
oxygen actinometry
1
5 × 10 photons cm-2 s-1
14
H
2
gas at 1000 Torr; Lyman-R
NIST silicon photocathode as well as
oxygen actinometry
emission at 1216 Å accounts
for 5% of the total intensity in
the wavelength range of
1
000-2000 Å
1 × 10 photons cm-2 s-1
1
5
this work
Ar-10% H
Torr
2
gas mixture at
2
N O actinometry
1
Lyman-R emission at 1216 Å accounts for almost all absorption
photon will always destroy a single N
absorption length in the primary dissociation process. The active
fragment species produced are also able to destroy N
molecules in secondary processes, and the quantum yield for
both the primary and secondary N O destruction processes has
been measured as 1.46 ( 0.04 N
O/photon.21
2
O molecule in a short
between 1000 and 2000 Å.
2
At the N O gas densities employed in this work, all Lyman-R
2
O
photons are absorbed in a short path length compared to the
cell length. Hence, the Stark-Einstein concept may be em-
ployed, where, in the short absorption path length, a single
2
2
Figure 1. Calibration apparatus for N O photodecomposition by Lyman-R radiation.
2

Photochemical Decomposition of N2O by Lyman-R Radiation
J. Phys. Chem. A, Vol. 114, No. 10, 2010 3445
Figure 2. Hydrogen discharge lamp spectrum using an Ar-H
2
(90:
2
Figure 4. Optimization of total pressure of the Ar-H gas mixture in
7
1
0) gas mixture at a pressure of 1 Torr. The inset shows the absorption
order to maximize the lamp output. The power was measured using a
thermopile detector containing a sapphire window which cuts off at
>170 nm and therefore does not measure the Lyman-R input.
cross sections of N O at room temperature (1000-2000 Å) obtained
2
13-17
from various reports in the literature.
All reports agree that the
cross section is negligible in the wavelength range of 1500-2000 Å.
Figure 5. Variation of light intensity (as measured using a thermopile
detector) with RF power. The thermopile became very noisy above 50
W of microwave power (dashed line).
Figure 3. Integrated absorption versus pressure of N
bands ν and ν . These calibration curves are applied to the experimental
value of absorbance for photon flux determination.
2
O for vibrational
1
3
B. Measurement of the Kinetics of N
sition-Spectral Measurements. As shown in an example in
Figure 3, the integrated IR spectrum of the N O(g) in several
spectral regions was employed to measure its partial pressure.
The integrated spectral data from pure N O was accurately
calibrated against the measured gas pressure for the four
vibrational modes of N O, ν , ν , 2ν , and 2ν . During the
photochemical experiments, conversion of the integrated ab-
sorbance to partial pressure of N O was carried out for the four
spectral regions employed, where the IR bands for N O were
observed using data like that shown in Figure 3. Each spectrum
2
O Photodecompo-
2
for various Ar-H pressures and (2) by near optimization of
the microwave power applied to the lamp. This detector
responds only to radiation above 170 nm and is used only to
judge the relative lamp output.
2
2
III. Results and Discussion
2
1
3
1
2
A. Overall Photochemical Reaction. Extensive studies in
the literature2
1-25
2
agree that the photochemistry of N O using
2
Lyman-R radiation can be explained by a process involving two
primary photochemical steps followed by thermally activated
2
-
1
processes involving an active species reaction with N
O + hν f N + O
N O + hν f NO + N
2
O
was obtained by averaging 500 interferograms at 2 cm
resolution, and a small correction for the background spectrum
in the empty cell was applied to each spectrum. Integration
of the IR bands was carried out using three different methods:
N
(1)
(2)
2
2
(1) the Bruker FTIR OPUS program, (2) a method deconvoluting
2
the peak profile into multiple peaks and fitting each of them to
Gaussian or Lorentzian functions followed by summing up of
the areas, and (3) a trapezoid approximation wherein the peak
profile is vertically sliced into smaller trapezoids followed by
summing up of all of the areas. The three methods were found
to agree within (5%.
1
2
1
2
1
2
1
2
O + N O f N + O2
(3)
(4)
2
2
1
2
1
2
C. Optimization of the Lamp Output. As shown in Figures
O + N O f NO
2
4
and 5, a thermopile detector (type 2M, Dexter Research
Center, Inc.) ∼5 cm from the MgF
2
window in a N atmosphere
was used to optimize the output of the Lyman-R lamp in two
2
The species produced photochemically may exist in different
electronic states. In addition, the active species produced from
ways, (1) by maximization of the thermopile-detected output

3
446 J. Phys. Chem. A, Vol. 114, No. 10, 2010
Rajappan et al.
the processes in reactions 1 and 2 can in principle reach the
walls of the cell and react at these surfaces according to reactions
5
and 6
O + wall f adsorbed species
N + wall f adsorbed species
(5)
(6)
In the absence of the wall effect, the net reaction will be
3
2
1
2
3
N O + 2hν f N + N + O + 2NO
(7)
2
2
2
2
and 0.67 photons/N O (decomposed by both the photochemical
and subsequent reaction pathways) will be required based on
the postulated elementary steps. This is in excellent agreement
with measurements of N
an efficiency of 0.69 ( 0.02 photons/N
Should all active intermediate species be consumed by
the walls, the net reaction 8 will result, and 1 photon/
2
O photodesorption at 123.6 nm, where
Figure 6. Spectral evolution of N
2
O, NO and NO
2
features during
O. P N O ) 10 Torr. The NO vibrational
band became stronger after irradiation ceased due to the reaction
between NO and O formed during irradiation.
2
1
O was found.
0
2
the photodecomposition of N
2
2
2
2
N
2
O(decomposed) will be required.
2
N O + 2hν f N + NO + N(wall) + O(wall)
2
2
(8)
Thus, depending upon whether the active intermediates are
allowed to react with N O or with the walls, we have an
uncertainty in consumption of photons by N O ranging from
.67 to 1 photon/N O(decomposed) using the net reactions
assumed. By combining the wall effect and the homogeneous
reaction effect in different ratios, the efficiency of N
consumption by photons can range from 0.67 to 1 photon/
2
2
0
2
2
O
N O(decomposed) using the net reactions 1-8 in various
2
combinations.
As will be shown later, a method to eliminate or substantially
reduce the capture of active intermediate species on the cell
walls has been developed in this work. Thus, we will assume
an efficiency of N
O in accordance with the argument above and the literature.21
B. Kinetics of Photodecomposition of N O. Figure 6 shows
a sequence of IR spectra obtained during 8 h of exposure of
2
O destruction corresponding to 0.67 photons/
N
2
2
0
N O to Lyman-R radiation at an initial pressure of P N O ) 10.00
2
2
Torr. IR spectra were recorded at 10 min intervals. Selected
samples of the spectra along with the assignments of the
fundamental vibrational modes, overtones, and combination
bands are shown in Figure 6. The slow disappearance of the
Figure 7. NO and NO
2
production rate during and after the photo-
0
decomposition of N O at P N O ) 10 Torr.
2
2
2
N O bands is accompanied by the appearance of a NO band
-1
-1
centered at 1875 cm , and an NO
2
band centered at 1617 cm .
2
decrease in the partial pressure of N O, using both infrared
Upon interruption of the irradiation, it was found (Figure 7)
modes, the measured rate of partial pressure decrease is identical
to within 0.7%. Upon interruption of the irradiation, the partial
-
1
that a dark reaction to produce NO
2
(1617 and 2907 cm ) was
observed to occur over hours, in accordance with the known
pressure of N
2
O becomes constant, although the data obtained
26
slow reaction of NO and O
the irradiation conditions in this experiment, most NO
2
. This observation shows that under
formed
for the ν mode shows an apparent decrease of 0.12 Torr
3
2
(-1.3%) over the 8 h in the dark, which may be a spectroscopic
artifact, as indicated in the caption to Figure 8. Similar
measurements at 1 and 5 Torr of initial N
were also performed using the four strongest vibrational modes.
The linearity and smoothness of the N O partial pressure
during irradiation is photodecomposed. In addition, we note that
traces of adsorbed OH species are deposited on the KBr IR
windows of the cell. This is probably due to the photodecom-
position of traces of water vapor present in the cell.
2
O partial pressure
2
Figure 8 shows the variation of the partial pressure of N
calculated from the integrated IR absorbance using two N
2
O
O
curves obtained consistently throughout this work indicate that
the Lyman-R lamp delivers a constant light output over many
hours of operation.
2
vibrational modes, ν
and for 8 h after irradiation is discontinued, beginning with 10.00
Torr of N O. As deduced from the linear behavior of the
1
3
and ν , during an 8 h irradiation period
2
C. Pressure Changes During N O Photodissociation. Fig-
2
ure 9 shows a plot of the total pressure change which occurs in

Photochemical Decomposition of N2O by Lyman-R Radiation
J. Phys. Chem. A, Vol. 114, No. 10, 2010 3447
Figure 8. Determination of photon flux from the slope of the N
partial pressure versus time for vibrational bands ν and ν . The -1.3%
change in the integrated absorbance of the ν mode after the lamp was
turned off is attributed to a small cooling effect on the integrated
absorbance of this mode.
2
O
1
3
Figure 10. Photon fluxes determined using 1, 5, and 10 Torr initial
3
partial pressures of N
asymptotic dependence of the apparent photon flux on the N
pressure. At high N O pressures, wall effects removing excited species
are minimized. The two error bars represent the range of random errors
in the spectroscopically determined rates of N O consumption using
the various mode frequencies and N O pressures.
2
O. A second-order polynomial fit shows a weak
2
O partial
2
2
2
Figure 9. Pressure changes in the chamber during and after the
photodecomposition of N O.
2
2
the cell during and after the photodecomposition of N O. An initial
linear pressure increase is observed to diminish in slope as the
irradiation continues. When the lamp is turned off, an immediate
decreasing trend in pressure is observed, which continues for at
least 8 h. The increase in pressure is caused by two effects: (1) the
2
Figure 11. Absorption length of Lyman-R radiation at various N O
pressures in the cell.
2 2 2
production of N , NO, and NO by the photodecomposition of N O
and (2) warming of the gas by radiation from the Lyman-R source.
The decrease in pressure following 8 h of irradiation is caused by
explained by the operation of a wall effect which depends on
pressure, as described below.
two or more effects: (1) reaction of NO product with O
to produce NO , as observed by IR spectroscopy (Figure 7) and
2) cooling of the gas following the interruption of irradiation. The
shape of the pressure curve in Figure 9 can be approximately
explained using either reaction 7 or reaction 8, followed by NO
2
product
2
E. Wall Effect on N
gentle increase in the apparent yield of the photodecomposition
of N O as the pressure rises from 1 to 10 Torr is consistent
2
O Photolysis. The observation of a
(
2
2
with the presence of a wall effect in the cell. Such effects have
not been studied in other chemical actinometers. The cell is
composed of stainless steel (whose surface is known to be
production in the dark, combined with about 18 K of heating during
irradiation and subsequent cooling following irradiation. The
combination of reactions and heating/cooling as contributors to the
pressure changes in Figure 9 invalidate the use of total pressure
2
7
primarily Cr
2
O
3
) and MgF
2
in the form of the window which
O pressure is increased
admits the Lyman-R radiation. As the N
2
change as a means for Lyman-R lamp calibration by N
todecomposition.
D. Calibration of the Lyman-r Lamp using the Decom-
position of N O as Measured by IR Spectroscopy. Using the
initial rate of consumption of N O as measured during the first
stage of photochemical destruction and the efficiency ratio of
.67 photons/N O(consumed) as discussed above, the photon
yield of the Lyman-R lamp was measured at three initial
pressures of N O (1, 5, and 10 Torr). The results are plotted in
2
O pho-
from 1 to 10 Torr, the absorption length of the Lyman-R
radiation will decrease, as shown in the scale drawing in Figure
11. At 10 Torr of pressure, the photoactivation region occupies
2
only a few mm of length in the N
and the active species will be formed with essentially only a
MgF surface nearby. As the N O pressure is reduced to 1 Torr,
2 2
O from the MgF window,
2
2
2
0
2
both the photon absorption length and the excited species
diffusion rate increase and excited gas phase species will be
able to interact with the stainless steel walls of the cell more
readily. If we assume that the oxide film of the stainless steel
2
Figure 10, using the four strongest IR absorption bands as
shown. It is found that the apparent photon flux approaches a
is more active than MgF
then under low pressure conditions, reactions 3 and 4 will
contribute less to N O decomposition, causing a reduction in
2
for reaction with the excited species,
maximum near 10 Torr of N
2
O pressure. The weak dependence
on pressure in the apparent photon flux of the lamp may be
2

3
448 J. Phys. Chem. A, Vol. 114, No. 10, 2010
Rajappan et al.
consumption of N O per photon and hence a reduction in the
apparent photon yield of the Lyman-R lamp. In the high-pressure
limit (10 Torr), the wall reactions 5 and 6 will be minimized,
2
Coleman (Department of Chemistry, Wellesley College) who
critically reviewed the manuscript for us.
References and Notes
2
and an efficiency near 0.67 photons/N O will be achieved.
(
(
(
1) Chubb, T. A.; Friedman, H. ReV. Sci. Instrum. 1955, 26, 493.
2) Bollinger, J. J.; Pipkin, F. M. ReV. Sci. Instrum. 1981, 52, 938.
3) Watanabe, K.; Marmo, F. F.; Inn, E. C. Y. Phys. ReV. 1953, 91, 1155.
Hence, for the measurement of the photon yield of the Lyman-R
lamp, we focus on the 10 Torr measurement, giving a photon
1
5
-2 -1
yield of 1.28 ( 0.36 × 10 photons cm s .
F. Summary of Results. The results of this study may be
summarized as follows:
(4) Cottin, H.; Moore, M. H.; Nilan, Y. B. Astrophys. J. 2003, 590, 874.
(
5) Westley, M. S.; Baragiola, R. A.; Johnson, R. E.; Baratta, G. A.
Nature 1995, 373, 405.
6) Warneck, P. Appl. Opt. 1962, 1, 721.
(7) Okabe, H. J. Opt. Soc. Am. 1964, 54, 478.
(
(
1) A novel IR method has been used to follow the
consumption of N O during exposure to almost pure Lyman-R
radiation from a H /Ar lamp with optimized output.
2) A weak pressure dependence of the apparent photon flux
(
(
8) Davis, D.; Braun, W. Appl. Opt. 1968, 7, 2071.
9) Yamashita, I. Jpn. J. Appl. Phys. 1975, 14, 70.
2
2
(
(
10) Weber, P.; Greenberg, J. M. Nature 1985, 316, 403.
11) Baratta, G. A.; Leto, G.; Palumbo, M. E. Astron. Astrophys. 2002,
(
3
2
1
84, 343.
from the Lyman-R lamp has been discovered, indicating that a
wall effect involving the consumption of active O intermediate
species is present, occurring primarily on the stainless steel cell
(
12) Inn, E. C. Y.; Watanabe, K.; Zelikoff, M. J. Chem. Phys. 1953,
1, 1648.
(13) Hitchcock, A. P.; Brion, C. E.; Van der Wiel, M. J. Chem. Phys.
980, 45, 461.
walls. This effect is reduced as the N
the 10 Torr range, and at 10 Torr, the surface exposed to the
O excitation regime becomes mainly MgF
3) N O photodecomposition is accompanied by the slow
reaction of the product O with product NO to produce NO
The reaction to make NO occurs more efficiently in the dark.
4) The photon flux of Lyman-R radiation is estimated to be
2
O pressure is raised into
(
(
14) Chan, W. F.; Cooper, G.; Brion, C. E. Chem. Phys. 1994, 180, 77.
15) Selwyn, G.; Podolske, J.; Johnston, H. S. Geophys. Res. Lett. 1977,
N
2
2
.
4, 427.
(
(
16) Hubrich, C.; Stuhl, F. J. Photochem. 1980, 12, 93.
17) Nicolet, M.; Peetermans, W. Ann. Geophys. 1972, 28, 751.
(
2
2
2
.
(18) Duncanson, A.; Stevenson, R. W. H. Proc. Phys. Soc., London 1958,
72, 1001.
19) Hirschfelder, J. O.; Curtiss, C. F.; Bird, R. B. Molecular Theory of
2
(
(
Gases and Liquids; Wiley: New York, 1954.
(20) Massman, W. J. J. Atmos. EnViron. 1998, 32, 1111.
15
-2 -1
1
.28 ( 0.36 × 10 photons cm
s
in the optimized H
2
/Ar lamp
(
21) Groth, W. E.; Schierholz, H. Planet. Space Sci. 1959, 1, 333.
employed. Previous measurements in other laboratories yield
15
-2 -1
(22) Hampson, R. F., Jr.; Okabe, H. J. Chem. Phys. 1970, 52, 1930.
Lyman-R fluxes in the range of 0.3-2 × 10 photons cm s .
(
(
23) Gilpin, R.; Welge, K. H. J. Chem. Phys. 1971, 55, 975.
24) McEwan, M. J.; Lawrence, G. M.; Poland, H. M. J. Chem. Phys.
1
974, 61, 2857.
Acknowledgment. We thank the support by the Center for
Chemistry of the Universe through the NSF Centers for Chemical
Innovation Program Award CHE-0847919. We acknowledge with
sincere thanks the contributions made by Professor Lester Andrews
(
(
25) Okabe, H. J. Chem. Phys. 1967, 47, 101.
26) Henriques, F. C., Jr.; Duncan, A. B. F.; Noyes, W. A., Jr. J. Chem.
Phys. 1938, 6, 518.
(27) Cho, B.; Choi, E.; Chung, S.; Kim, K.; Kang, T.; Park, C.; Kim,
B. Surf. Sci. 1999, 439, L799.
(Department of Chemistry, University of Virginia), Dr. Allan
Laufer (NIST), and Professors Chris Arumainayagam and William
JP9093436