Chemical Physics 331 (2007) 232–244
www.elsevier.com/locate/chemphys
VUV spectroscopy of CH3Cl and CH3I
S. Eden
a
a,*,1
, P. Limão-Vieira
a,2
, S.V. Hoffmann b, N.J. Mason
c
Department of Physics and Astronomy, University College London, Gower Street, London, WC1E 6BT, UK
b
Institute of Storage Rings, University of Aarhus, Ny Munkegade, Aarhus, Denmark
c
Department of Physics and Astronomy, Open University, Walton Hall, Milton Keynes MK7 6AA, UK
Received 17 August 2006; accepted 24 October 2006
Available online 6 November 2006
Abstract
High-resolution photoabsorption spectra of CH3Cl and CH3I are reported in the energy range 3.9–10.8 eV (320–115 nm). Several features are observed for the first time in the CH3Cl spectrum and a number of new assignments are proposed. For both molecules, the
present work provides the most reliable absolute cross-sections yet reported at energies above the dissociative A band transition.
Ó 2006 Elsevier B.V. All rights reserved.
Keywords: Photoabsorption; CH3Cl; CH3I; Electronic excitation; Vibrational excitation; Rydberg series
1. Introduction
CH3Cl and CH3I are important trace species in the terrestrial atmosphere. The principle sources of the species in
the atmosphere are oceanic emissions and biomass burning
[1,2], making them unusual among atmospheric halomethane gases whose major sources are anthropogenic [3].
CH3Cl is the most abundant of the atmospheric halomethanes and accounts for 15% of the free chlorine radicals
in the stratosphere [4,5]. Therefore, it plays a key role in
the destruction of ozone. Conversely, CH3I is believed to
be relatively environmentally benign because of its short
lifetime in the troposphere due to solar photolysis. Fahr
et al. [6] and Rattigan et al. [7] estimated photolysis lifetimes
of CH3I in the troposphere to be 4 and several sunlit days,
respectively, while its destruction rate due to reactions with
OH radicals at low altitudes has been shown to be around
*
Corresponding author. Tel.: +33 4 72 43 12 59; fax: +33 4 72 44 80 04.
E-mail address: [email protected] (S. Eden).
1
Also of Institut de Physique Nucléaire de Lyon, IN2P3-CNRS et
Université Claude Bernard Lyon 1, 43, Boulevard du 11 Novembre 1918,
69622 Villeurbanne Cedex, France.
2
Also of Laboratório de Colisões Atómicas e Moleculares, CEFITEC,
´ sica, FCT, Universidade Nova de Lisboa, P-2829-516
Departamento de Fı
Caparica, Portugal.
0301-0104/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.chemphys.2006.10.021
two orders of magnitude slower [8]. Monitoring the concentrations of CH3Cl and CH3I in the atmosphere requires a
detailed spectral database to be assembled.
The VUV photoabsorption cross-section of CH3Cl has
been studied by several previous authors. Early spectra
were reported by Price [9] and by Zobel and Duncan [10]
from 6.2 to 12.4 and 24.8 eV, respectively. Absolute
cross-sections were first measured by Tsubomura et al.
[11] from 6.60 to 7.75 eV and subsequently by Russell
et al. [12] in the energy range 6.20–11.16 eV. In each of
these works, the authors do not specify their energy resolution. The spectrum of Russell et al. [12] was revisited by
Raymonda et al. [13] who commented that the energy positions for sharp features observed in each spectrum are in
agreement to within ±1 meV. Robin [14] reviewed the spectra of Price [9], Zobel and Duncan [10], and Russell et al.
[12] in a broad discussion of the electronic excitation of
alkyl halides. Hochmann et al. [15] measured photoabsorption by CH3Cl using a McPherson Model 225 monchromator from 7.74 to 11.41 eV with an unspecified resolution.
The result of Hochmann et al. [15] was analysed further
by Felps et al. [16]. Truch et al. [17] reported the cross-section from 8.50 to 11.75 eV using a McPherson Model 231
monochromator. An indication of the resolution of the
spectrum of Truch et al. [17] is given by that fact that they
S. Eden et al. / Chemical Physics 331 (2007) 232–244
used a grating of 1200 lines per mm compared to 2000 at
the ASTRID facility. Olney et al. [3] derived the photoabsorption cross-section in the range 6–50 eV with a resolution of 50 meV from electron energy loss spectroscopy
(EELS) data taken using incident electrons of energy
between 6 and 8 keV. The cross-section for the absorption
of Lyman-a photons (10.120 eV) by CH3Cl was measured
by Vatsa and Volpp [18].
Recently Locht et al. [19] have published a detailed analysis of Rydberg structure observed in the photoabsorption
spectrum of CH3Cl from 6 to 25 eV. Their measurement
was made with a wavelength resolution of 0.1 nm, only
marginally less precise than the present result (0.075 nm).
In a separate article the authors report the analysis of the
vibrational structure in the energy range 7.5–10.5 eV [20].
Hitchcock and Brion [21] carried out EELS measurements
on both CH3Cl and CH3I using incident electrons of
2.5 keV and analysing scattered electrons in the energy loss
range 6–12 eV with a resolution of 80 meV. To our knowledge, the only low impact energy EEL spectra recorded for
CH3Cl are those of Nachtigallova et al. [22] measured with
a resolution of 60–80 meV using incident electrons of 0.05,
0.25, 0.65, and 20.05 eV. Their experimental work was augmented with electronic structure calculations made using
the single-excitation configuration interaction (CIS), complete-active-space self-consistent field (CASSCF), and
multi-reference MP2 (CASPT2) methods [23].
The earliest measurement of the VUV photoabsorption
spectrum of CH3I was carried out by Price [9] between 6.2
and 12.4 eV with an unspecified resolution. However, the
absolute cross-section was not reported until 1972 when
Boschi and Salahub [24] published results using a McPherson Model 225 monochromator in the energy range 3.7–
11.1 eV. Robin [14] cited the work of Price [9] and Boschi
and Salahub [24] in a broad discussion of the electronic
excitation of alkyl halides in which particular attention is
devoted to the Rydberg states of CH3I. Hochmann et al.
[15] measured the photoabsorption spectrum from 6.05 to
10.33 eV, again using a McPherson Model 225 spectrometer with an unspecified energy resolution. The results of
Hochmann et al. [15] were revisited by the same group in
two subsequent papers focussing on the Rydberg series
converging to the ionisation limits corresponding to the
removal of an iodine lone pair electron [16,25]. The photoabsorption spectrum from 6.888 to 10.972 eV has been
reported with a resolution of ±0.003 Å ( ± 0.1 meV at
9 eV) by Baig et al. [26]. This remarkable resolution was
achieved using a grating of 5000 lines per mm at the Synchrotron Radiation Laboratory of the University of Bonn.
The Rydberg transitions observed in the spectrum were discussed further by Dagata et al. [27,28]. However, the publications only show detailed plots and assignments for the
structure observed from 9.3 to 10.3 eV. Furthermore,
despite the exceptional precision of the measurement, the
authors neither listed the energies of the assigned Rydberg
peaks nor discussed the vibrational structure in detail.
Waschewsky et al. [29] reported the photoabsorption
233
cross-section in the ranges 3.70–5.28 and 6.11–6.33 eV,
claiming a maximum wavelength resolution of 0.1 nm
(3 meV at 6 eV). Photoabsorption measurements of the diffuse A band (4–6 eV) have also been carried out by Jenkin et al. [30], Fahr et al. [6], and Rattigan et al. [7]. Olney
et al. [31] derived the photoabsorption cross-section of
CH3I from 4 to 65 eV with a resolution of 50 meV from
EELS data taken using incident electrons of 8 keV. As
far as we are aware, no low impact energy electron energy
loss data for scattering from CH3I is available in the literature to clarify the assignment of optically forbidden
transitions.
The valence shell molecular orbital configuration of
CH3Cl and CH3I in the electronic ground state can be represented as (1a1)2(2a1)2(1e)4(3a1)2(2e)4: 1A1 [3,31]. Both
molecules can be considered to be pseudo-triatomic molecules of C3v symmetry, with the H atoms acting as H3
groups positioned at the centre of mass of the three atoms.
Accordingly, the vibrational modes listed in Table 1 include
CH3 stretching and deformation but no motion specific to
the individual C–H r bonds. Indeed only one symmetric
CH3 stretching vibration can be excited, the so-called
‘‘umbrella function’’ [34]. Comparison between the excitation energies of stretching modes indicates that the strongest bonds are those between the carbon atoms and
hydrogen groups. The fact that the CH3 bonds are apparently unaffected by the substitution of the iodine atom with
chlorine is unsurprising as C–X (X = I, Br, Cl) r orbitals
tend to be localised close to the electronegative halogens
and the larger halogens are situated relatively far from the
carbon atom. Therefore, we can expect minimal overlap
of CH3 and C–I or C–Cl orbitals. Furthermore, it is worth
noting that the C–X stretching excitation energies tend to be
lower for the more electronegative halogen species.
In this paper, we report the results of detailed analysis of
the photoabsorption spectra of CH3Cl and CH3I with
emphasis on providing a systematic assignment of the spectral features and absolute cross-sections in the energy range
3.9–10.8 eV of the VUV region.
2. Experimental
The present photoabsorption measurements were made
at the ASTRID facility, Aarhus University, Denmark.
Due to the high performance of the monochromator and
the stability of ASTRID synchrotron source, high-resolution spectra and low absolute cross-section errors can be
measured with great efficiency. The tuneable energy range
of the incident photons at the ISA photoabsorption facility
(3.9–10.8 eV) coincides with the solar visible–UV spectrum
which penetrates the stratosphere and troposphere
(<6.89 eV) [35] making the facility highly suitable for probing the photolysis of aeronomic molecules.
The experimental apparatus has been described in detail
elsewhere [36]. Synchrotron radiation is passed through a
static gas sample. A photo-multiplier is used to measure
the transmitted light intensity at 0.05 nm intervals and
234
S. Eden et al. / Chemical Physics 331 (2007) 232–244
Table 1
Motions associated with vibrational modes of excitation of CH3X (X = Cl, I) in the neutral ground state [32] and in the lowest energy ionic states [33]
Vibrational mode and symmetry
Description of motion
Energy in meV
CH3Cl ground
CH3Cl+
2
t1,
t2,
t3,
t4,
t5,
t6,
A1
A1
A1
E
E
E
CH3 s-stretch
CH3 s-deform
C–X stretch
CH3 d-stretch
CH3 d-deform
C–X bend
364
168
91
377
180
109
wavelength is selected using a toroidal dispersion grating
(2000 lines per mm). For wavelengths below 200 nm (energies above 6.20 eV), helium is flushed through the small
gap between the photomultiplier and the exit window of
the gas cell to prevent any absorption by air contributing
to the spectrum. The LiF entrance window filters out
higher order radiation before it can enter the cell. At longer
wavelengths, absorption by the air in the gap removes all
higher order radiation. The minimum and maximum wavelengths between which scans are performed, 115–320 nm
(10.8–3.9 eV), are determined by the transmission windows
of the gas cell and the grating range, respectively. A baratron capacitance manometer (MKS 390HA) is used to
measure the pressure up to a maximum of 1.4 mbar. The
sample pressure is varied to give maximum absorption
whilst ensuring that the transmitted signal never falls below
one tenth of the incident signal (i.e., avoiding saturation
when the transmitted intensity is close to zero). The synchrotron beam ring current is monitored throughout the
collection of each spectrum in order that spectra can be
corrected for any changes in incident photon flux during
the period of spectral accumulation. Absolute photoabsorption cross-sections are determined using the Beer–
Lambert law:
I t ¼ I 0 expðnrxÞ
ð1Þ
where It is the radiation intensity transmitted through the
gas sample, I0 is that through the evacuated cell, n the
molecular number density of the sample gas, r the absolute
photoabsorption cross-section, and x the absorption path
length (25 cm).
The energy scale is calibrated using SO2 since it has
clearly defined sets of sharp absorption peaks from 3.8 to
5.1 eV [37] and from 5.15 to 7.25 eV [38]. The energy resolution for the present results is calculated to be 0.07 nm,
corresponding to 3 meV at the midpoint of the energy
range studied. The error on the absolute cross-section measurements is estimated at ±5% [36,39]. Only when absorption by the sample is very weak (I0 It), does the error
increase significantly as a percentage of the measured
cross-section.
The CH3I and CH3Cl samples were purchased from
Aldrich Chemical Company Inc. and have a minimum purity of 99% and 99.5%, respectively. In both cases, the sam-
E3/2
–
133
79
–
193
110
CH3I ground
2
2
E1/2
–
–
83
–
190
106
CH3I+
364
155
66
379
178
109
E3/2
–
157
61
379
–
114
2
E1/2
368
154
–
–
–
114
2
A1
–
–
34
–
–
–
ple was introduced to the photoabsorption cell without
further purification or treatment. Comparisons of the present spectra with those of a number of possible contaminants (O2, CO2, H2O, N2) have revealed no evidence for
impurities.
3. Results and discussion
3.1. Spectroscopy of CH3Cl
3.1.1. Valence excitation of CH3Cl
The full range over which a non-zero absorption crosssection was measured in the present work is shown in
Fig. 1. The lowest energy feature rises to a maximum of
1.18 Mb at 7.27 eV and is shown in greater detail in
Fig. 2(a). This diffuse feature is assigned to the excitation
of an electron from the chlorine lone pair (n) to a C–Cl
antibonding (r*) MO following the established A band
analysis for methyl halides [14,40]. The local maximum
is reported to occur at 7.18, 7.21 and 7.3 eV by Zobel
and Duncan [10], Russell et al. [12], and Tsubomura
et al. [11], respectively. Russell et al. [12] initially attributed the feature to a C–Cl r ! r* transition but adopted
the standard n ! r* assignment in the subsequent publication [13].
The EELS results of Nachtigallova et al. [22] enabled the
authors to distinguish singlet and triplet states in the A
band. The singlet to singlet n ! r* transition was reported
at 7.14 eV, in good agreement with the local maximum
observed in the present and previous photoabsorption
spectra. The singlet to triplet n ! r* transition was measured at 6.74 eV [22]. As expected, no evidence is observed
for this strongly optically forbidden contribution to the A
band in the present spectrum.
The broad continuum underlying the Rydberg and
vibrational structure observed in the energy region 7.5–
8.6 eV (see Fig. 2(a)) strongly suggests dissociation, possibly via the assigned n ! 4s transitions. In particular, due
to the clear overlap of the continuum with the A band, dissociation may occur along the C–Cl bond due to relaxation
of a 4s electron into a r* MO.
Nachtigallova et al. [22] calculated the C–Cl r ! r* singlet to triplet transition to occur at 9.15 and 9.18 eV,
respectively, by using CASPT2 and CIS methods. Accord-
S. Eden et al. / Chemical Physics 331 (2007) 232–244
235
140
Converging to 2E 3/2
5pa1
6pa1
2
Cross section (Mb, 10 cm )
4pa1
Converging to 2E 1/2
120
-18
100
4d
4pe
6sa1
80
7sa1
60
5s
40
4s
20
n→
σ* (C l)
5pe
6pe
n → σ*
0
6
6.5
7
7.5
8
8.5
9
9.5
10
10.5
11
Energy (eV)
Fig. 1. The ASTRID photoabsorption spectrum of CH3Cl.
ingly, the presently observed continuum extending from 8.8
to 9.6 eV is attributed to dissociative r ! r* valence excitations (see Fig. 2(b)). In agreement with Nachtigallova
et al. [22], we expect any direct evidence for the singlet to
triplet transition to be hidden by the Rydberg structure
around 9 eV.
3.1.2. Rydberg excitation of CH3Cl
To assign observed features arising from Rydberg series,
the standard equation is used:
En ¼ EI ðR=ðn dÞ2 Þ
ð2Þ
where En is the energy of the Rydberg state, EI the ionisation limit to which the series converges (this may be the
ionic ground state or an ionic excited state), R the Rydberg constant (13.61 eV), n the principal quantum number, and d the relevant quantum defect [41]. By highresolution He(I) photoelectron spectroscopy, Karlsson
et al. [33] determined the lowest adiabatic ionisation energies of CH3Cl to be 11.289 ± 0.003 eV and
11.316 ± 0.003 eV for the 2E3/2 and 2E1/2 ionic states,
respectively, corresponding to the removal of an electron
from the chlorine lone pair (n) with a spin–orbit energy
splitting of 27 ± 6 meV. Karlsson et al. [33] comment that
the energy splitting in the chlorine lone pair is markedly
lower than the predicted value of 73 meV due to the
quenching of spin–orbit coupling by vibronic coupling
in CH3Cl+. The quenching cited by Karlsson et al. [33]
is attributed to the Ham effect described by Sturge [42].
The adiabatic ionisation energies for the following 2A1
and 2E ionic states are reported to be 13.8 and 15.4 eV,
respectively [33].
The Rydberg transitions and the associated vibrational
structure observed in the photoabsorption spectrum of
CH3Cl have been investigated recently by Locht et al.
[19,20]. Therefore, the following discussion concentrates
upon the presently observed differences and new features,
while the reader is referred to previous publications
[19,20] for discussion of assignments of the other features.
The energies of peaks assigned to Rydberg transitions in
the present energy range are given in Table 2 and are
labelled in Fig. 2(a–c).
The most striking disagreement between the present
interpretation of the CH3Cl UV spectrum and that proposed by Locht et al. [19] lies in the n values of states
agreed to belong to particular Rydberg series. The expected
quantum defects for ns, np, and nd series corresponding to
the removal of an electron from an orbital localised on a Cl
atom to are 2.0, 1.7, and 1.0–1.3, respectively [41,43].
Accordingly, we attribute the lowest energy CH3Cl Rydberg peak to the n = 4 member of an ns series converging
to the 2E3/2 ionisation limit with a quantum defect of
2.04. This feature is assigned in the same way by Russell
et al. [12], Raymonda et al. [13], Robin [14], Olney et al.
[3], and Nachtigallova et al. [22]. Similarly, an example of
the ns (2E3/2) assignments made by Truch et al. [17] is that
of the peak at 10.42 eV as n = 6, corresponding to a quantum defect of 2.02. However, Locht et al. [19] suggest that
the peak corresponds to a 3s (2E3/2) transition with a quantum defect of 1.069. While acknowledging that their Rydberg interpretation often varies from that of Truch et al.
[17], Locht et al. [19] do not discuss why the proposed n
values and the corresponding quantum defects are systematically one unit lower than given elsewhere.
236
S. Eden et al. / Chemical Physics 331 (2007) 232–244
20
→CH 3Cl+ : 2E 3/2
Cross section (Mb, 10-18cm2)
Cross section (Mb, 10-18cm2)
15
10
υ
4s
1.4
ASTRID
1.2
1
Hubrich et al. 1977
0.8
Robbins 1976
0.6
→CH 3Cl+ : 2E 1/2
4s
υ
υ5
υ6
Simon et al. 1988
0.4
0.2
0
5.5
6
6.5
Energy (eV)
7
7.5
5
n→
σ* (CCl)
0
5.5
6.5
6
7.5
7
a
8
160
+
2
→ CH3Cl : E 3/2
υ3-0
140
υ
υ5
+
4pa1
120
Cross section (Mb, 10-18cm2)
9
8.5
Energy (eV)
+
2
→ CH3Cl : E 3/2
υ6
100
2
→ CH3Cl : E 1/2
υ
υ5
υ
υ5
4pa1
+
4pe
80
2
→ CH3Cl : E 1/2
υ
υ5
υ6
60
4pe
40
20
0
8.6
8.7
8.8
8.9
9
b
9.1
9.2
9.3
9.4
9.5
9.6
9.7
Energy (eV)
Fig. 2. Detail of the CH3Cl ASTRID VUV spectrum: (a) the A band n ! r* valence excitation and 4s Rydberg structure. Inset: comparison with previous
A band measurements [43,45,46]; (b) vibrational structure associated with 4p Rydberg transitions and (c) Rydberg structure in the high-energy part of
ASTRID range (over page).
The photoelectron bands associated with the 2E3/2 and
E1/2 ionic states [33] show clear vibrational structure.
The reported excitation energies for the ionic vibrational
modes are listed in Table 1 [33]. For promotion from a
non-bonding MO, the excitation energy for a vibrational
mode in a Rydberg state is generally expected to be close
to or slightly higher than that in the limiting ionic state.
Locht et al. [20] state that all the fine structure observed
from 7.5 to 10.5 eV can be described using three vibrational
modes with average excitation energies of 162 ± 3 meV,
104 ± 7 meV, and 77 ± 7 meV. This would suggest the
2
most probable assignments for the observed modes to be
t2 (CH3 deformation), t6 (CH3 rocking), and t3 (C–Cl
stretching). However, Locht et al. [20] demonstrate that
the conventional vibrational analysis is complicated by
Jahn–Teller symmetry distortions. On the basis of ab initio
optimised geometry calculations, Locht et al. [20] conclude
that the experimentally observed 77 ± 7 meV and
104 ± 7 meV modes are due to t6 (CH3 bending) and t5
(CH3 rocking coupled with C–Cl stretching), respectively.
The 162 ± 3 meV mode is simply labelled t and is considered to represent either t3 (anti-symmetric CH3 bending)
S. Eden et al. / Chemical Physics 331 (2007) 232–244
υ
2
Converging to E 3/2
100
6pa1
5pa1 υ5
2
υ6
Conevrging to E 1/2
υ5
Cross section (Mb, 10-18cm2)
5pa1
80
6sa1
υ6
5s
υ
υ5
40
4d
υ
6pa1
5s
60
237
6sa1
υ5
υ5
7sa1
υ
7sa1
υ6
4d
υ5
20
5pe
5pe
6pe
υ5
6pe
υ6
10.6
10.7
0
9.7
9.8
9.9
10
10.1
10.2
c
10.3
10.4
10.5
10.8
Energy (eV)
Fig. 2 (continued)
Table 2
Energy positions of features assigned to Rydberg series converging to the CH3Cl+ states 2E3/2 and 2E1/2
Rydberg analysis
Converging to 2E3/2 (11.289 eV)
Energy in eV
4sa1
5sa1
6sa1
7sa1
4pa1
5pa1
6pa1
4pe
5pe
6pe
4d
Converging to 2E1/2 (11.316 eV)
d
Present
Locht et al. [20]
7.754
9.77a
10.423
10.721
8.814
10.101
10.579
9.208
10.192
10.624
9.817
7.759
9.742
10.422
10.722
8.815
10.102
10.578
9.208
10.198
10.638
9.816
2.04
2.01
2.04
2.11
1.67
1.62
1.62
1.44
1.48
1.48
0.96
Energy in eV
d
Present
Locht et al. [20]
7.872
9.85a
10.445
10.749
8.894
10.117
10.61a
9.30a
10.272
10.69a
9.891
7.873
9.789
10.441
10.746
8.895
10.193
10.624
9.298
10.255
10.684
9.892
2.01
1.95
2.05
2.10
1.63
1.63
1.61
1.40
1.39
1.34
0.91
a
Indicates that the feature is diffuse and thus its energy position is subject to greater uncertainty. The quantum defect is calculated using the energy of
the feature (present data) and the adiabatic ionisation energy determined by Karlsson et al. [33]. Locht et al. [20] proposed assignments corresponding to
those given above except with ns = (n1)s, np = (n1)p, and nd = (n1)d.
or t4 (CH3 umbrella motion). The presently observed
photoabsorption features associated with Rydberg and
vibrational transitions are listed in Table 3. The structure
is broadly consistent with that observed by Locht et al.
[20] and the vibrational modes are labelled accordingly as
t, t5, and t6 with average excitation energies of 159, 109,
and 78 meV, respectively.
Table 3 shows that a number of assignments suggested
in the present work do not agree with those made by Locht
et al. [20], particularly at the high energy end of the spectral
range. At higher energies the spectral resolution is lower,
the Rydberg peaks are more closely spaced, and the quantum defects are more affected by errors in the measurements for relevant adiabatic IE. It should also be noted
that any possible contributions due to traces of impurities
in the sample are extremely difficult to identify within such
dense and complex structure. Nonetheless, given the high
resolution and signal to noise ratio in the ASTRID spectrum, we believe that the present assignments represent
the most reliable analysis available for the high energy
Rydberg states of CH3Cl and their associated vibrational
structure.
3.1.3. Absolute photoabsorption cross-section of CH3Cl
There is considerable disagreement between the present
and previous absolute absorption cross-section measurements of CH3Cl in the VUV. In particular, the present
238
S. Eden et al. / Chemical Physics 331 (2007) 232–244
Table 3
CH3Cl features assigned to vibrational series
Energy in eV
Assignment
Locht et al. [20]
Present
7.912b
7.975
8.032
8.124
–
–
–
–
8.923
8.976
9.001
9.058
9.090
9.120
9.138
9.161b
9.166
9.252b
9.321
9.370
9.378
9.403
9.435
–
7.907
7.978
8.030
8.119
8.19a
8.26a
8.35a
8.722
8.921
8.975
9.010
9.057
9.090
9.120
9.14a
9.16a
9.18a
9.25a
9.322
–
9.379
9.40a
9.43a
–
4s(2E3/2) + t
4s(2E1/2) + t5
4s(2E1/2) + t
4s(2E1/2) + t + t6
4s(2E1/2) + 2t
4s(2E1/2) + 2t+ t6
4s(2E1/2) + 3t
4pa1 (2E3/2)t30
4pa1 (2E3/2) + t5
4pa1 (2E3/2) + t
4pa1 (2E1/2) + t5
4pa1 (2E1/2) + t
4pa1 (2E1/2) + t5 + t6
4pa1 (2E1/2) + 2t5
4pa1 (2E1/2) + t + t6
4pa1 (2E3/2) + 2t
4pa1 (2E1/2) + t + t5
4pa1 (2E3/2) + 2t + t5
4pe (2E3/2) + t5
4pe (2E3/2) + t
4pe (2E1/2) + t6
4pe (2E1/2) + t5
4pe (2E3/2) + 2t5
4pe (2E1/2) + t
Energy in eV
Assignment
Locht et al. [20]
Present
9.480
9.538
9.576
9.622
9.679b
9.934b
9.964
10.054
10.158
10.230b
–
10.254b
10.290b
10.308b
10.363
10.398b
10.486
10.501
10.522b
10.555b
10.609
10.696b
10.756b
9.475
9.541
9.57a
9.62a
9.67a
9.93a
9.967
10.051
10.16a
10.230
–
10.26a
10.289
10.311
10.362
10.401
10.485
10.49a
10.525
10.561
10.60a
10.70a
10.758
4pe (2E3/2) + t + t5
4pe (2E1/2) + t + t6
4pe (2E1/2) + t + t5
4pe (2E1/2) + 2t
4pe (2E1/2) + t + 2t5
4d (2E3/2) + t5
4d (2E1/2) + t6
4d (2E1/2) + t
4d (2E1/2) + t + t5
5pa1 (2E1/2) + t5
5pa1 (2E3/2) + t5
5pa1 (2E3/2) + t
5pa1 (2E3/2) + t5 + t6
5pe (2E3/2) + t5
5pa1 (2E3/2) + t + t5
5pa1 (2E3/2) + 2t
5pa1 (2E3/2) + 2t + t6
6s (2E3/2) + t6
6s (2E3/2) + t5
6s (2E1/2) + t5
6s (2E1/2) + t
6pe (2E3/2) + t6
6pa1 (2E1/2) + t
The Rydberg assignments listed above are given by Locht et al. [20] as (n1)s/pa1/pe/d.
a
Indicates that the feature is diffuse and thus its energy position is subject to greater uncertainty. The assignments listed with no corresponding energy
value are not directly observed but suggested by the following structure and may be hidden by another feature.
b
These features were assigned differently by Locht et al. [20].
maxima of the peaks at 7.872, 8.814, 10.192, and 10.624 eV
are around 50% higher than the corresponding measurements by Locht et al. [20], Truch et al. [17], and Raymonda
et al. [13]. Conversely, Vatsa and Volpp [18] measured an
absorption cross-section of 88 ± 6 Mb at 10.199 eV,
45% higher than those recorded at the ASTRID facility.
These variations cannot be attributed solely to differences
in spectral resolution and must therefore be explained in
terms of systematic errors.
Several high-pressure photoabsorption measurements
have been made in the key atmospheric energy region
below 6.89 eV. Robbins [43] reported the absorption spectrum from 5.64 to 7.13 eV. Hubrich et al. [45] measured the
cross-section in the energy range 7.85–5.28 eV at 298 and
208 K. Most recently, Simon et al. [46] recorded the spectrum from 5.74 to 7.13 eV at 295–210 K. Both Hubrich
et al. [45] and Simon et al. [46] reported minimal variation
in absorption cross-section with temperature. The measurements of Simon et al. [46] were carried out at a pressure of
27 mbar with a cross-section error of ±2%. The relevant
part of the present spectrum (recorded at 1 mbar) is compared to the results of Robbins [43], Hubrich et al. [45],
and Simon et al. [46] in Fig. 2(a) (inset). The plots are
observed to be in very close agreement with one another.
This suggests that the present measurement of the photoabsorption cross-section of CH3Cl can also be relied upon at
higher energies.
3.2. Spectroscopy of CH3I
3.2.1. Valence excitation of CH3I
The photoabsorption cross-section of CH3I recorded in
the present experiments is shown in Fig. 3. The spectrum is
characterised by the weak, diffuse A band lying below the
lowest energy Rydberg structure. The local maximum of
the feature is observed at 4.80 eV, in good agreement with
the measurement of the peak at 4.77 eV by Waschewsky
et al. [29]. Following the assignment proposed by Mulliken
[40] and confirmed in each subsequent analysis, the feature
is attributed to the excitation of an electron from the
HOMO centred on the iodine lone pair (n) to the LUMO
of C–I r* antibonding character.
No evidence is observed in the present work for spin–
orbit splitting in the A band. This suggests that the feature
is dominantly due to a single transition, assigned by Waschewsky et al. [29] to an excitation to the triplet r* state.
The weakness of the feature is consistent with an excitation
of singlet to triplet character as identified by Kitajima et al.
[47] for the CF3X (X = Cl, Br, I) A bands.
Waschewsky et al. [29] expect pre-dissociation to occur
in the B band beginning at 6.16 eV and shown in detail
in Fig. 4(a). The structure observed in this energy range
is dominated by sharp peaks and shows no evidence for a
background continuum. Therefore, although the Rydberg
transition at 6.165 eV may be interpreted to just coincide
S. Eden et al. / Chemical Physics 331 (2007) 232–244
239
600
B
Cross Section (Mb)
1.5
Cross Section (Mb, 10-18cm2)
500
400
C
ASTRID
1
Rattigan et al.
0.5
CH3I
0
2
4
4.5
5
5.5
E 3/2
6
+
CH3I
+
2
E 1/2
Energy (eV)
300
D
200
100
A band
σ
n→
0
3
4
5
6
7
8
9
10
11
Energy (eV)
Fig. 3. The ASTRID photoabsorption spectrum of CH3I. Inset: detail showing the A band compared with the measurements of Rattigan et al. [7].
with the high-energy extreme of the dissociative n ! r*
transition (the A band), the photoabsorption spectrum
does not include evidence for pre-dissociation in the B
band.
Robin [14] and Boschi and Salahub [24] state that the
broad background continuum observed below 9.54 eV,
the lowest ionisation energy, is due to valence transitions
of r ! r* character. As the LUMO is universally recognised as being C–I antibonding, the Rydberg states above
7.4 eV may be pre-dissociative along the C–I bond. This
understanding is generally consistent with the r ! r*
assignments proposed for CH3Cl, CF3Cl, CF3Br, and
CF3I [10].
3.2.2. Rydberg excitation of CH3I
The convergence limits 9.5381 and 10.1648 eV calculated by Baig et al. [26] with an error of ±0.0002 eV for
the Rydberg series identified in their very high-resolution
photoabsorption measurements are integrated into Eq.
(2) to verify the present Rydberg assignments. These ionisation energies correspond to a spin–orbit energy splitting
of 627 meV. Accordingly Karlsson et al. [33] measured
the adiabatic ionisation energies using He(I) photoelectron
spectroscopy to be 9.540 ± 0.004 eV and 10.168 ± 0.004 eV
for the 2E3/2 and 2E1/2 ionic states, respectively, associated
with the removal of an electron from the iodine lone pair
(n).
The Rydberg states of CH3I in the present photoabsorption result have been assigned in a generally consistent way in the earlier publications. The two major
peaks shown in Fig. 4(a) are attributed to the n = 6 members of ns series converging to the ionisation limits associated with the 2E3/2 and 2E1/2 ionic states. The differences
between the various ns series assignments can be consid-
ered to be purely due to the resolution of the respective
photoabsorption spectra and the precision to which the
relevant adiabatic ionisation energies were known. Similarly, following the work of Robin [14], the sharp peak
at the beginning of the D band (7.306 eV) has been recognised as being due to the n ! 6p transition belonging to a
series converging to the 2E3/2 limit. Accordingly, Hochmann et al. [15] proposed an np series with a similar
quantum defect converging to the 2E1/2 limit as well as
suggesting assignments for the lowest energy members
(n = 5) of nd series converging to both 2E limits. These
series were similarly assigned by Baig et al. [26] on the
basis of their high-resolution spectrum. The exceptionally
high resolution achieved at the University of Bonn is
reflected in the fact that the feature at 9.253 eV (n ! 9d
(2E3/2), Fig. 4(c)) is clearly shown by Baig et al. [26] to
be split into two sharp peaks.
The Rydberg assignments recommended in the present
work are given in Table 4, and marked on Fig. 4(a–c).
The features from 9.3 to 10.3 eV are assigned in agreement
with Baig et al. [26]. Those at lower energies are attributed
on the basis of the quantum defects of the higher members
of the series identified by Baig et al. [26]. In most cases,
these assignments are consistent with those shown in the
plots of Olney et al. [31]. Like Baig et al. [26], Olney
et al. [31] do not list the energies or quantum defects corresponding to their assignments, while the ns and np series
converging to the 2E3/2 ionisation limit lie outside the
energy range shown by Baig et al. [26]. In these cases, we
assign the peaks in agreement with Hochmann et al. [15],
as the identified structure corresponds to quantum defects
similar to those for the equivalent 2E1/2 series. However,
we do not agree with n > 7 assignments of Hochmann
et al. [15] for the ns and np series converging to the 2E1/2
240
S. Eden et al. / Chemical Physics 331 (2007) 232–244
a
800
υ1 υ 4
υ2-0
υ 1+υ 6
nυ 2
nυ 6
nυ 3
700
ASTRID x 25
υ 5+υ 2
υ5
υ 2-0
υ5+υ 3
υ1
7
7.1
υ 1+υ 2
υ4
υ 1+υ 3
nυ2
nυ6
nυ 3
600
Cross Section (Mb, 10-18cm2)
ASTRID
υ 4+υ 2
υ 4+υ2
500
6s
(2E 1/2)
6s
(2E 3/2)
400
300
200
100
0
6
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
7.2
7.3
Energy (eV)
b
350
nυ1
Cross Section (Mb, 10-18cm2)
300
6p
(2E 3/2)
250
nυ2
6p (2E 1/2)
nυ6
υ2
υ3
7s (2E 3/2)
200
nυ2
υ2+υ6
2υ2+υ3
150
nυ6
2
5d ( E 3/2)
7p
(2E 3/2)
υ2
nυ3
100
50
0
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
8
8.1
8.2
8.3
8.4
8.5
Energy (eV)
Fig. 4. Detail of the CH3I ASTRID VUV spectrum: (a) the B and C bands associated with n ! 6s transitions of series converging to 9.5381 and
10.1648 eV; (b) Rydberg and vibrational structure from 7.2 to 8.5 eV, including the D band and (c) Rydberg and vibrational from 8.4 to 10.4 eV, including
the 2E ionisation limits (over page).
ionisation limit. Similarly, we disagree with the nd (2E1/2)
series represented in the plots of Olney et al. [31].
There is clear evidence for vibrational structure associated with the Rydberg transitions discussed above. Price
[9] reports an energy separation between vibrational features in the B and C bands of 148 meV which the author
attributes to symmetrical CH3 deformation, t2 (Table 1).
Boschi and Salahub [24] report series with an average
energy spacing between features of 136 meV and assign
the structure in agreement with Price [9]. Felps et al. [16]
use the higher resolution spectrum first reported by Hochmann et al. [15] to analyse the vibrational structure in the B
and C bands. The corresponding part of the present spec-
trum is shown in detail in Fig. 4(a), while the assignments
proposed by Felps et al. [16] and those recommended in the
present work are listed in Table 5. Felps et al. [16] describe
the structure in terms of four separate series, labelled a1–4.
The present assignments are generally in agreement with
the a2 and a4 series, which include the previously recognised t2 progressions [9,24]. The average excitation energies
in the present analysis are 330, 134, 59, 356, 158, and
108 meV for the t1–t6 modes, respectively, quite close to
their ground state equivalents (see Table 1). Thus these
assignments are consistent with the supposition that the
geometry of the molecule is not significantly deformed in
the 6s (2E) excited states. The excitation energies for the
S. Eden et al. / Chemical Physics 331 (2007) 232–244
241
300
7p
8p
9p
8s
9s
CH3I+
10
2
10
11d 13 15...
250
Cross Section (Mb, 10-18cm2)
6d
7d
8d
E 3/2
9d
10d
12 14
10 11 12...
5d
200
6d
5d+υ2
7d
8d
6d+υ2
9d
7d+υ2
8d+υ2 9d+υ2
10d+υ2...
7s
8s
150
9s
10s
+
11s
CH3I
2
E 1/2
100
50
7p
8p
9p
10p
11p
0
8.4
86
8.8
9
9.2
c
9.4
9.6
9.8
10
10.2
10.4
Energy (eV)
Fig. 4 (continued)
Table 4
Energy positions and quantum defects of features assigned to Rydberg series converging to the CH3I+ states 2E3/2 and 2E1/2
Rydberg analysis
Converging to 2E3/2 (9.540 eV)
Energy in eV
6s
7s
8s
9s
10s
11s
6p
7p
8p
9p
10p
11p
5d
6d
7d
8d
9d
10d
11d
12d
13d
14d
15d
16d
Converging to 2E1/2 (10.168 eV)
d
Present
Previous
6.165
8.022
8.710
9.007
9.17a
–
7.306
8.429
8.869
9.090
9.215
–
7.820
8.610
8.949
9.143
9.253
9.315
9.364
9.396
9.421
9.439
9.454
–
6.162
8.020
8.707
9.010
9.164
–
7.302
8.429
8.862
9.090
9.232
–
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[31]
–
[15]
[15]
[15]
[15]
[15]
[15]
[15]
[15]
[15]
[15]
3.99
4.00
3.95
3.94
3.92
–
3.53
3.50
3.49
3.49
3.51
–
2.19
2.17
2.20
2.13
2.09
2.19
2.16
2.22
2.22
2.29
2.28
–
Energy in eV
d
Present
Previous
6.777
8.652
9.329
9.634
9.793
9.89a
7.996
9.047
9.490
9.713
9.840
9.923
8.536
9.273
9.593
9.755
9.867
9.939
9.987
10.023
10.047
10.064
10.08a
10.092
6.775
8.652
[26]
[26]
[26]
[26]
7.932
9.050
[26]
[26]
[26]
[26]
–
–
[26]
[26]
[26]
[26]
[26]
[26]
[26]
[26]
[26]
[26]
[15]
[15]
[15]
[15]
4.00
4.00
3.96
3.93
3.95
3.96
3.51
3.51
3.51
3.51
3.52
3.49
2.11
2.09
2.12
2.23
2.23
2.23
2.24
2.18
2.22
2.34
2.28
2.27
a
Indicates that the feature is diffuse and thus its energy position is subject to greater uncertainty. The quantum defects are calculated using the ionisation
limit identified by Baig et al. [26]. Energies are not given for features identified by Olney et al. [31] and Baig et al. [26] because the authors only report their
assignments as labels on figures. The present nd 2E1/2 assignments are not in agreement with those proposed by Hochmann et al. [15] or by Olney et al. [31].
present B and C analysis are used to propose assignments
for the vibrational structure around the 7s (2E3/2) transition
(Table 6 and Fig. 4(b)). The structure observed around the
6p (2E) and 5d (2E3/2) transitions suggests vibrational series
with average excitation energies of 337 (t1), 150 (t2), 55
(t3), and 93 meV (t6). This is consistent with the suggestion
242
S. Eden et al. / Chemical Physics 331 (2007) 232–244
Table 5
Energy positions of the vibrational structure associated with the n ! 6s excitations belonging to series converging to the CH3I+ states 2E3/2 and 2E1/2
Felps et al. [16]
Present
Energy in
eV
Vibrational
analysis
Energy in
eV
–
6.102
6.144
6.163
6.195
6.223
6.232
6.267
6.298
6.316
6.337
6.360
6.403
6.434
–
a1,
a2,
a2,
a1,
a2,
a2,
a2,
a2,
a2,
a1,
a2,
a2,
a2,
6.468
6.489
6.521
6.568
–
6.618
a1, t4
a2, t1
a2, t4
a2, 3t2
–
a2, t1 + t2 or a4,
t20
a2, t4 + t2
a4, t30
6.652
6.707
a
b
t00
t20 + t2
t00
t6
t3
t3-0 + t2
t6
t2
t5
t2 + t6
2t6
t2 + t6
2t2
Felps et al. [16]
Present
Vibrational
analysis
Energy in
eV
Vibrational analysis
Energy in
eV
Vibrational
analysis
6.010
6.103b
–
6.165
6.205
6.227
–
6.270
6.302
–
6.342
6.363
6.406
6.437
6s (2E3/2), t20
Unassigned
–
6s (2E3/2), t00
Unassigned
t3
–
t6
t2
–
Unassigned
t2 + t3
t2 + t6
2t2
6.751
6.772
6.799
6.812
6.831
6.875
6.885
6.906
6.933
6.967
6.984
7.018
7.039
7.070
–
6.777
6.805
–
6.835
–
6.89a
6.909
6.936
6.969
6.987
7.02a
7.045
7.066
–
6s (2E1/2), t00
Unassigned
–
t3
–
t6
t2
t5
t2 + t3
t5 + t3
t2 + t6
2t2
t2 + t5
6.476
6.493
6.525
6.572
6.60a
6.621
Unassigned
t1
t4
3t2
t1 + t6
6s (2E1/2), t20
7.104
7.123
7.163
7.173
7.193
7.231
a4, t20 + t2
a4, t00
a3, t6
a4, t20 + t2 + t3
a4, t3
a4, t6
a4, t20 + 2t2 & a4, 2t3
a4, t2
a4, t5 or a3, t2 + t6
a4, t2 + t3
a3, t2 + t5
a4, t20 + 3t2
a4, 2t2
a4, t2 + t5 or a3, t4 or a3,
2t2 + t6
a4, t1
a4, t4
a4, t1 + t3
a4, 3t2
from np series
a4, t1 + t2
7.109
7.128
7.17a
7.177
7.198
7.236
t1
t4
t1 + t3
3t2
Unassigned
6s (2E1/2), t1 + t2
6.665
6.711
6s (2E3/2), t4 + t2
4t2
7.252
from np series
7.257
t4 + t2
Indicates that the feature is diffuse and thus its energy position is subject to greater uncertainty.
This feature is extremely weak and therefore must be considered to be doubtful with regards to possible noise or trace impurities.
Table 6
Energy positions of the vibrational features associated with CH3I Rydberg
peaks lying above 7.3 eV
Table 7
Energy ranges in which CH3I Rydberg transitions belonging to series
converging to the 2A1 ionic state may be expected to occur
Energy in eV
Series converging to 11.949 eVa
7.360
7.402
7.458
7.491
7.585
7.61a
7.642
7.98a
7.880
7.93a
7.966
8.146
8.130
8.157
8.23a
8.266
Vibrational analysis
2
6p ( E3/2) + t3
t6
t2
2t6
3t6
2t2
t1
2t1
5d (2E3/2) + t3
2t3
t2
6p (2E1/2) + t2
7s (2E3/2) + t6
t2
2t6
t2 + t6
Energy in eV
Vibrational analysis
8.299
8.335
8.360
8.68a
9.42b
9.75b
9.91a
10.02b
10.09b
10.138
10.175
10.200
10.221
10.238
10.247
2t2
3t6
2t2 + t3
5d (2E1/2) + t2
6d + t2
7d + t2
8d + t2
9d + t2
10d + t2
11d + t2
12d + t2
13d + t2
14d + t2
15d + t2
16d + t2
a
Indicates that the feature is diffuse and thus its energy position is
subject to greater uncertainty.
b
Accompanies features which coincide energetically with another suspected Rydberg feature.
by Robin [14] that a short t2 series is observed in the D
band with an energy separation which is only a few meV
lower than the equivalent excitations in the neutral ground
state. With the exception of the diffuse feature at 7.02 eV
ns
np
nd
a
Energy range in eV
n=5
n=6
n=7
–
–
9.38–10.62
7.75–8.87
8.87–9.77
10.70–11.18
10.21–10.53
10.53–10.84
11.21–11.45
The adiabatic energy for the formation of the 2A1 ionic state [33].
(assigned here to n ! 6s (2E1/2) + t2 + t6), the features
assigned to sequence bands by Felps et al. [16] are not visible in the present spectrum (Table 5).
Whereas the a2 and a4 series suggested by Felps et al.
[16] begin at the intense 6s (2E) transitions, the a1 series
begins at an extremely weak feature at 6.103 eV while the
origin of the a3 series (6.700 eV) is neither visible in the
present spectrum nor in that reported by Felps et al. [16].
Therefore, we consider that the present data does not provide sufficient evidence for the short a1 and a3 series,
removing possible explanations for the clear peak at
6.205 eV, as well as the weaker features at 6.342, 6.476,
and 6.805 eV. The energy difference between the 6.205 eV
peak and the 6s (2E3/2) transition is distinctly lower than
the excitation energy of any of the ground state vibrational
modes. This suggests that analysis based upon comparisons
S. Eden et al. / Chemical Physics 331 (2007) 232–244
with the ground and ionic states may be inadequate to
describe the vibrational structure associated with the 6s
(2E) excited sates of CH3I. In particular, Jahn–Teller symmetry distortions analogous to those proposed by Locht
et al. [20] to account for the vibrational series observed in
the spectrum of CH3Cl may have a significant effect on
the spectrum of CH3I. Indeed the low-energy Rydberg
excitations in VUV spectrum of CH3I demonstrate the
characteristic sharp and strong t00 peaks and the following
vibrational features of successively weaker intensity which
are often associated with geometrical differences between
initial and final states [20]. Theoretical studies are required
to clarify the present understanding of the Rydberg state
vibrational excitation of CH3I.
Due to the high density of Rydberg peaks, it is very difficult to identify vibrational series above 8.4 eV in the
present spectrum. Nonetheless, the high resolution available to Baig et al. [26] enabled the authors to assign single
quanta of t2 (CH3 symmetric deformation) excitations following the peaks of the nd (2E1/2) series up to and beyond
n = 16. The features are also visible in the present work
although they are clearly identifiable only above the adiabatic 2E1/2 ionisation energy. The closely spaced Rydberg
peaks above the first ionisation limit prevent the possible
observation of any similar structure associated with the
nd (2E3/2) series. The nd (2E1/2) + t2 assignments are listed
in Table 6 and shown Fig. 4(c). In accordance with the proposed vibrational structure associated with the 6p (2E) and
5d (2E3/2) transitions, the average energy spacing between
t2 peaks and the corresponding t00 transitions is
152 meV, very close to the excitation energy of the mode
in the neutral electronic ground state and in the ionic
2
E1/2 state (see Table 1) and inconsistent with the t2 excitation energy of 134 meV observed in the vibrational structure associated with the 6s (2E) and 7s (2E3/2) transitions.
This suggests that the nd and np states may maintain the
symmetry of the neutral ground state molecule to a greater
extent than the ns states.
A number of features observed in the CH3I VUV spectrum remain unassigned. In particular, we have not been
able to propose excitations with confidence to account
for the clear features at 7.384, 7.523, 7.687, 7.860, 8.778,
8.806, 8.929, and 8.968 eV (see Fig. 4(b) and (c)). Possible
assignments include Rydberg transitions belonging to series converging to limits associated with ionic excited states.
Karlsson et al. [33] give the third lowest adiabatic ionisation energy as 11.949 ± 0.007 eV, associated with a 2A1
state and the removal of an electron from the 3a1 orbital
of the molecule in the neutral ground state. The authors
do not discuss the expected nature of the vacated orbital.
However, comparisons with CH3F [33] and CF3X
(X = Cl, Br, I) [48] lead us to expect the highest occupied
molecular orbital below the iodine lone pair to be of C–I
r bonding character. In the cases of CH3F, CF3Cl, and
CF3I, the bonding orbitals between the carbon atom and
the single halogen atoms have a1 symmetry. This is also
true of the orbital vacated to form the CH3I+ 2A1 state.
243
As discussed earlier, the C–I r orbital is understood to
be localised near to the strongly electronegative iodine
atom. Therefore, we expect the quantum defect scheme:
d = 3.9 4.2 (ns), d = 3.5 3.9 (np), and d = 1.8 2.7
(nd) [41,43] for the vacation of an electron from an
iodine-localised MO to apply to Rydberg series converging
to any of the three lowest ionisation limits of CH3I. Table 7
shows the energy ranges which correspond to the required
quantum defects for series converging to 11.949 eV. It
seems therefore plausible to expect up to 5 Rydberg peaks
within the present energy range belonging to series converging to the limit associated with the 2A1 ionic state.
However, the lack of sharp features between 10.2 and
10.8 eV shown in Figs. 3 and 4(e) suggests that any ns
and np series must be weak and therefore are unlikely to
account for the unidentified features. Similarly, we do not
have sufficient evidence assign a specific peak to a 5d
(2A1) transition.
3.2.3. Absolute photoabsorption cross-section of CH3I
The magnitude of the absolute cross-section measured at
energies above the A band has not been discussed in the literature. Baig et al. [26] do not show intensity units in their
photoabsorption plots while the low resolution of Boschi
and Salahub [24], Hochmann et al. [15], and Olney et al
[31] severely limit any useful comparisons between the
cross-sections of the sharp CH3I peaks. Therefore, a meaningful comparison can only be made between the work of
Waschewsky et al. [29] and the present absolute photoabsorption cross-sections. Waschewsky et al. [29] report the
maximum of the 6s (2E3/2) peak to be 450 Mb, in fairly
good agreement with the present measurement of 547
Mb. The greater cross-section observed in the present spectrum may be attributed to the slightly higher resolution
achieved at the ASTRID facility.
The absolute cross-section of the broad A band, however, is not affected significantly by differences in resolution
and has been investigated by a number of groups due to its
atmospheric importance. Fig. 3 (inset) shows that the present cross-section is in very good agreement (±10%) with
the measurements of Rattigan et al. [7], as well as with
those of Waschewsky et al. [29] and of Jenkin et al. [30].
Conversely, the cross-sections recorded by Fahr et al. [6]
at 295 K are around 20% greater than the present cross-sections, while Boschi and Salahub [24] report the band maximum to be 0.50 Mb, compared to the present maximum of
1.16 Mb.
4. Conclusions
Several features are observed for the first time in the
CH3Cl spectrum and a number of new assignments are suggested. For both molecules, the present work provides the
most reliable absolute cross-sections yet reported at energies above the dissociative A band transition. Although
the present CH3I photoabsorption spectrum is not the
highest resolution measurement to have been carried out,
244
S. Eden et al. / Chemical Physics 331 (2007) 232–244
the full spectrum of Baig et al. [26] is not available in the
literature. Therefore, we observe photoabsorption features
which, although doubtless evident in the University of
Bonn result, have never previously been reported and add
to our understanding of the molecule. Theoretical studies
are required to determine any symmetry distortions in the
Rydberg states of CH3I and thus clarify the interpretation
of the associated vibrational structure.
Acknowledgements
The authors wish to acknowledge the European Commission for access to the ASTRID facility at the University
of Aarhus, Denmark through the Access to Research Infrastructure action of the Improving Human Potential programme. Some of this work forms part of the EU
network programme EPIC, HPRN-CT-2002-00179. PLV
acknowledges the honorary research fellow position at
University College London and the visiting fellow position
at The Open University, UK. SE, PLV, and NJM acknowledge the support of the UK funding councils: EPSRC,
NERC, and CLRC during the course of this research.
References
[1] J.H. Seinfeld, S.N. Pandis, Atmospheric Chemistry and Physics, John
Wiley and Sons Inc., New York, 1998.
[2] N. Bell, L. Hsu, D.J. Jacob, M.G. Schultz, D.R. Blake, J.H. Butler,
D.B. King, J.M. Lobert, E. Maier-Reimer, J. Geophys. Res. – Atmos.
107 (D17) (2002) 4340.
[3] T.N. Olney, G. Cooper, W.F. Chan, G.R. Burton, C.E. Brion, K.H.
Tan, Chem. Phys. 205 (1996) 421.
[4] M.A.K. Khalil, in: P. Febian, O.N. Singh (Eds.), The handbook of
environmental chemistry, Part 4E, Reactive Halogen Compounds in
the Atmosphere, 52, Springer, Berlin, 1999.
[5] A. Fell, Sources and effects of air pollution 2000, http://lorien.ncl.ac.uk/ming/cleantech/pollut1.htm.
[6] A. Fahr, A.K. Nayak, R. Huie, Chem. Phys. 199 (1995) 275.
[7] O.V. Rattigan, D.E. Shallcross, R.A. Cox, J. Chem. Soc., Faraday
Trans. 93 (1997) 2839.
[8] P. Marshall, A. Misra, R.J. Berry, Chem. Phys. Lett. 265 (1997) 48.
[9] W.C. Price, J. Chem. Phys. 4 (9) (1936) 539.
[10] C.R. Zobel, A.B.F. Duncan, J. Am. Chem. Soc. 77 (1955) 2611.
[11] H. Tsubomura, K. Kimura, K. Kaya, J. Tanaka, S. Nagakura, Bull.
Chem. Soc. Jpn. 37 (1964) 417.
[12] B.R. Russell, L.O. Edwards, J.W. Raymonda, J. Am. Chem. Soc. 95
(7) (1973) 2130.
[13] J.W. Raymonda, L.O. Edwards, B.R. Russell, J. Am. Chem. Soc. 96
(6) (1974) 1708.
[14] M.B. RobinHigher Excited States of Polyatomic Molecules, I,
Academic, New York, 1974.
[15] P. Hochmann, P.H. Templet, H.-T. Wang, S.P. McGlynn, J. Chem.
Phys. 62 (7) (1975) 1588.
[16] W.S. Felps, P. Hochmann, P. Brint, S.P. McGlynn, J. Mol. Spectrosc.
59 (1976) 355.
[17] D.T. Truch, D.R. Salomon, D.A. Armstrong, J. Mol. Spectrosc. 78
(1979) 31.
[18] R.K. Vatsa, H.-R. Volpp, Chem. Phys. Lett. 340 (2001) 289.
[19] R. Locht, B. Leyh, A. Hoxha, H.W. Jochims, H. Baumgartel, Chem.
Phys. 272 (2001) 259.
[20] R. Locht, B. Leyh, A. Hoxha, D. Dehareng, H.W. Jochims, H.
Baumgartel, Chem. Phys. 272 (2001) 277.
[21] A.P. Hitchcock, C.E. Brion, J. Electron Spectrosc. Relat. Phenom. 13
(1978) 193.
[22] D. Nachtigallova, D.E. Love, K.D. Jordan, J. Phys. Chem. 100 (1996)
5642.
[23] R. Locht, B. Leyh, A. Hoxha, D. Dehareng, K. Hottmann, H.W.
Jochims, H. Baumgartel, Chem. Phys. 272 (2001) 293.
[24] R.A. Boschi, D.R. Salahub, Mol. Phys. 24 (2) (1972) 289.
[25] H.T. Wang, W.S. Felps, G.L. Findley, A.R.P. Rau, S.P. McGlynn, J.
Chem. Phys. 67 (9) (1977) 3940.
[26] M.A. Baig, J.P. Connerade, J.A. Dagata, S.P. McGlynn, J. Phys. B:
At. Mol. Phys. 14 (2) (1981) L25.
[27] J.A. Dagada, G.L. Findley, S.P. McGlynn, J.-P. Connerade, M.A.
Baig, Phys. Rev. A 24 (5) (1981) 2485.
[28] J.A. Dagada, M.A. Scott, S.P. McGynn, J. Chem. Phys. 88 (1) (1988) 9.
[29] G.C.G. Waschewsky, R. Horansky, V. Vaida, J. Phys. Chem. 100
(1996) 11559.
[30] M.E. Jenkin, T.P. Murrells, S.J. Shalliker, G.D. Hayman, J. Chem.
Soc., Faraday Trans. 89 (3) (1993) 433.
[31] T.N. Olney, G. Cooper, C.E. Brion, Chem. Phys. 232 (1998) 211.
[32] T. Shimanouchi, Tables of Molecular Vibrational Frequencies
Consolidated, vol. I, National Bureau of Standards, 1972, 1-160.
[33] L. Karlsson, R. Jadrny, L. Mattsson, F.T. Chau, K. Siegbahn,
Physica Scripta 16 (1977) 225.
[34] P. Felder, Chem. Phys. 143 (1990) 141.
[35] W.B. Demore, S.P. Sandler, D.M. Golden, R.F. Hampson, M.J.
Kurylo, C.J. Howard, A.R. Ravishankara, C.E. Kolb, M.J. Molina,
Chemical kinetics and photochemical data for use in stratospheric
modelling, Evaluation Number 12, Jet Propulsion Laboratory Publication 97-4, 1997.
[36] S. Eden, P. Limáo-Vieira, S.V. Hoffmann, N.J. Mason, Chem. Phys.
323 (2006) 313.
[37] A.C. Vandaele, P.C. Simon, J.M. Guilmot, M. Carleer, R. Colin, J.
Geophys. Res. – Atmos. 99 (D12) (1994) 25599.
[38] D.E. Freeman, K. Yoshino, J.R. Esmond, W.H. Parkinson, Planet.
Space Sci. 32 (1984) 1125.
[39] A. Giuliani, J. Delwiche, S.V. Hoffmann, P. Limáo-Vieira, N.J.
Mason, M.J. Hubin-Franskin, J. Chem. Phys. 119 (7) (2003) 3670.
[40] R.S. Mulliken, J. Chem. Phys. 8 (1940) 382.
[41] C. Sandorfy (Ed.), The Role of Rydberg States in Spectroscopy and
Photochemistry, Kluwer Academic Publishers, Dordrecht, 1999.
[42] M.D. Sturge, Solid State Physics 20 (1967) 92.
[43] A. Giuliani, F. Motte-Tollet, J. Delwiche, J. Heinesch, N.J. Mason,
J.M. Gingell, I.C. Walker, N.C. Jones, M.-J. Hubin-Franskin, J.
Chem. Phys. 110 (21) (1999) 10307.
[43] D.E. Robbins, Geophys. Res. Lett. 3 (1976) 213, erratum op. cit. 3
(1976) 757.
[45] C. Hubrich, C. Zetzsch, F. Stuhl, Ber. Bunsenges. Phys. Chem. 81
(1977) 437.
[46] P.C. Simon, D. Gillotay, N. Vanlaethem-Meuree, J. Wisemberg, J.
Atmos. Chem. 7 (1988) 107.
[47] M. Kitajima, R. Suzuki, H. Tanaka, L. Pichl, H. Cho, Nukleonika 48
(2) (2003) 89.
[48] T. Cvitaš, H. Güsten, L. Klasine, J. Chem. Phys. 67 (1977) 2687.