JOURNAL OF CHEMICAL PHYSICS VOLUME 111, NUMBER 13 1 OCTOBER 1999

The structure and ground state dynamics of ArIH

A. McIntosh, Z. Wang, J. Castillo-Chara, R. R. Lucchese, and J. W. Bevan
Chemistry Department, Texas A&M University, College Station, Texas 77843-3255
R. D. Suenram
Optical Technology Division, National Institute of Standards and Technology, Gaithersburg, Maryland 20899
A. C. Legon
School of Chemistry, University of Exeter, Stocker Road, Exeter, England EX4 4QD
Received 20 May 1999; accepted 7 July 1999
The structure and ground state dynamics of the atomdiatom dimer interaction between Ar and HI
has been investigated by microwave and near infrared supersonic jet spectroscopy. Ab initio
molecular orbital calculations were used to provide greater insight into the nature of the interaction.
The ground state is shown to be in the isomeric form ArIH with R cm3.9975(1) ,
149.33(1) for the normal isotopomer and R cm3.9483(1) , 157.11(1) for ArID. The
potential surface from an ab initio molecular orbital calculation was scaled and shifted to yield a
nonlinear least-squares fit of the rovibrational state energies to the experimental data. The ground
state potential energy surface obtained in this manner has a barrier between the ArIH and ArHI
isomers of 88.5 cm1 with respect to the global minimum. Such calculations are also used to predict
the presence of localized states in the secondary minimum associated with isomers ArHI and
ArDI. Attempts to experimentally identify transitions associated with the latter were unsuccessful.
The ground state, ArIH isomeric structure, contrasts with the corresponding ground state of the
other members of the homologous series ArHX XF, Cl, and Br in which the Ar is bound to the
proton. 1999 American Institute of Physics. S0021-96069900837-5

INTRODUCTION ture of this dimer. There have been unpublished reports of

Investigations of the van der Waals complexes RgHX
(XF, Cl) 1,2 using high-resolution gas phase spectroscopy
have been particularly fruitful. Indeed, these species have
become prototypical systems for the investigation of
vibrationrotation-tunneling dynamics, the nature of vibra-
tional predissociation and the development of models for
characterization of the nature of intermolecular dynamics.
Such investigations are currently being expanded to include
multibody interaction forces in clusters of the form37
Rgn HX (XF, Cl) based on the availability of the pairwise
interactions. The substitution of Cl for F in the van der
Waals interactions of this type has been shown to produce
dramatic changes in the nature of the intermolecular poten-
tial energy surfaces and other properties.8 Consequently, sub-
stitution of Br or I in these species should produce further
dramatic changes. There have only been a restricted number
of high-resolution spectroscopic studies of ArHBr using
microwave9 and far infrared spectroscopy of 12 f Q(J)
branch10 and a recent characterization of the potential energy
surface in the 1 HBr stretching vibration of the dimer.11
A theoretical treatment of the ground intermolecular
potential12 based on the previous microwave study had pre-
dicted almost equal energy equilibrium isomers ArHBr and
ArBrH. The ground state of the complex was demonstrated
to be the ArHBr dimer with the isomer ArBrH being
10.88 cm1 above that ground state.
There have been no corresponding published high-
resolution studies of the interaction between Ar and HI that
permit an unequivocal assignment of the ground state struc-
attempts to observe the ArHI isomer using pulsed nozzle
FT Fourier transform microwave spectroscopy13 but they
proved unsuccessful. A different series of experiments has
also been completed on the photoinitiated chemical dynam-
ics in ArHBr and ArHI.14,15 These previous photodisso-
ciation studies of ArHI in a supersonic molecular beam
expansion did give proton distributions that were consistent
with interpretation of dissociation of an ArIH ground state.
More recently, there has been a REMPI resonantly en-
hanced multiphoton ionization (21) investigation of
ArHI.16 However, in this study, it was not possible to rota-
tionally resolve the spectrum and thus permit a differentia-
tion between a possible ground state having an ArHI or
ArIH isomeric form. A simultaneously performed equilib-
rium ab initio calculation did indicate that the preferred equi-
librium structure was ArIH. However, that investigation only
provided information about the equilibrium geometries and
did not provide evidence as to the nature of the ground rovi-
brational state of the complex. Furthermore, these investiga-
tions did not provide definitive evidence as to the tunneling
dynamics of the complex.
In this work, we shall report an analysis of the 1 near
infrared spectrum of the complex ArIH. This analysis was
then used to locate the quadrupole resolved rotational spec-
trum of ArIH recorded using a pulsed nozzle FT-
microwave spectrometer. It is expected that a corresponding
analysis of the deuterated isotopomer of ArID will contain
valuable information for characterization of the intermolecu-
lar potential so that isotopomer was also investigated. The

0021-9606/99/111(13)/5764/7/$15.00 5764 1999 American Institute of Physics

Downloaded 07 Jun 2002 to 128.104.69.200. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/jcpo/jcpcr.jsp
J. Chem. Phys., Vol. 111, No. 13, 1 October 1999
former permits an unequivocal confirmation of the rovibra-
tional assignment. The latter, in addition, permits determina-
tion of the ground state structure of the dimer as the isomeric
form ArIH. Furthermore, this data will be used in conjunc-
tion with the results of ab initio calculations to estimate the
energy of the ground state of the ArHI isomer relative to
the global ground state. In particular, the present calculations
included a determination of the full intermolecular potential,
modified to obtain agreement with experimental data, and a
calculation of the rovibrational dynamics on that surface.
These calculations thus represent a much more detailed the-
oretical study of this system than previously reported.16 Fur-
thermore, the determined semiempirical potential energy sur-
face obtained in this combined experimental and theoretical
study was used to predict several transitions involving inter-
molecular vibrations which are as yet unobserved.
EXPERIMENT
The supersonic jet spectra of the 1 band of ArIH was
first recorded in the 4.48 m spectral region using a tunable
diode laser cw continuous wave supersonic slit jet spec-
trometer at Texas A&M University with an InSb detector.
This spectrometer was capable of operating as a high fre-
quency wavelength modulation spectrometer using lock-in
detection of the second derivative signal up to 1 MHz.17 The
spectrometer has been shown to have an ultimate sensitivity
consistent with a minimum absorbance of two parts in 107 or
better. The supersonic jet expansion was formed from a gas-
eous reservoir consisting of typically 1% HI in Ar as carrier
gas and sustained at 2.0105 Pa. The expansion was formed
through a 12 cm long slit with 25 m width and all spectra
recorded with a single pass of the laser beam. The diode laser
was a IIVI lead salt diode laser centered at 2230 cm1. The
transition frequencies in the spectrum of the ArIH complex
were calibrated to an estimated absolute accuracy of 0.001
cm1 or better using standard frequencies of a simulta-
neously recorded spectrum of N2O. 18 Repeated efforts to ini-
tially observe the spectrum of the alternate isomeric form,
ArHI, proved unsuccessful.
The microwave spectra of both ArIH and ArID were
recorded in the frequency range 8 GHz18 GHz using a
newly developed mini pulsed nozzle FT microwave spec-
trometer at the National Institute of Standards and Technol-
ogy at Gaithersburg, Maryland. This spectrometer, including
its instrumental resolution 2 kHz and its sensitivity, has
been described in detail elsewhere.19 Thus the following de-
scription will focus on factors relevant to the current study.
The supersonic pulsed nozzle expansions used in the current
experiments were carried out with a 1 mm nozzle from both
static and flowing gaseous reservoirs consisting of typically
1% HI in Ar as carrier gas and sustained at 1.3510 5 Pa to
2.70105 Pa. In certain cases 10% Ar and 1% HI were
mixed in a He/Ne carrier particularly for attempted broad-
band experimental searches 1 GHz1.5 GHz for the
ArHI isomeric form of the dimer. Such searches also
proved unsuccessful and will be the subject of further dis-
cussion below. Corresponding searches for the isotopomer
ArDI were equally unsuccessful.
Downloaded 07 Jun 2002 to 128.104.69.200. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/jcpo/jcpcr.jsp
Structure and dynamics of ArIH 5765
The hydrogen iodide used in both experiments was cat-
egorized as 98% pure and obtained from Matheson, Glouc-
ester, Mass. The broadband FTIR Fourier-transform infra-
red spectrum of the gas shows the presence of HCl, HBr,
and H2O as impurities. In the infrared laser studies of the
dimer, this gas was used without further purification, after
removal of the first cut of gas taken from the cylinder. How-
ever, for the microwave studies, the impurities proved more
troublesome. In this case, not only was a first cut removed,
but the hydrogen iodide was also distilled from the gas cyl-
inder after the cylinder had been pumped at liquid nitrogen
temperatures. Despite this procedure, residual water contin-
ued to provide some problems and microwave spectra attrib-
uted to (H2O220,21 and H2OHI22 dimers were both simul-
taneously recorded with that ArIH. The corresponding
sample of deuterium iodide used in observation of the ArID
spectrum was obtained from Cambridge Isotopes, Cam-
bridge, Mass. Prior to use, the sample of DI in the jet expan-
sion was subjected to a similar procedure as that used to
purify HI. It is pertinent to note that the maximum deuterium
enrichment observed in the dimer was 80% due to partial
isotopic exchange in the gas lines.
THEORETICAL CALCULATIONS
We have also studied the ArHI complex using ab initio
methods so that we can probe the molecular tunneling dy-
namics of the dimer. To study this system, we have com-
puted the interaction energy on a grid of points where we
have kept the HI bond length fixed at its equilibrium value23
of r e 1.609 16 . Nine values of the distance between the
Ar and I atoms were considered, R3.5 , 3.75 , 4.0 ,
4.25 , 4.5 , 4.75 , 5.0 , 5.5 , and 6.0 , and there
were seven values of the angle HIAr considered
0 , 30 , 60 , 90 , 120 , 150 , and 180 leading to
a total of M 63 points calculated. At each point, the energy
was computed using second order MllerPlesset perturba-
tion theory MP2 using modified versions of the medium
sized polarized basis sets developed by Sadlej.24,25 The basis
set for Ar was obtained following the procedure described by
Sadlej24 using the HartreeFock wave function of the neutral
atom to determine the polarization function contractions and
starting with the basis set McLean and Chandler.26 The basis
sets were modified by uncontracting the last polarization
function and adding a basis function of one higher angular
momentum with the same exponents as were part of the
function which was uncontracted. In the basis set of I, the
function that was uncontracted was the last d type contracted
Gaussian so that two f functions were then added to the basis
set. Thus for H we added d functions with exponents 0.1027
and 0.0324, for Ar we added f functions with exponents of
0.1454 and 0.0509, and for I we added f functions with ex-
ponents 0.091 353 and 0.033 851. The final contraction
scheme was then 6.4.2/3.3.2 for H, 13.10.4.2/7.5.3.2 for
Ar, and 19.15.12.8/11.9.7.5 for I. The interaction energies
were then corrected for the basis set superposition error us-
ing the counterpoise method of Boys and Bernardi.27 All
electronic structure calculations were performed using the
28
GAUSSIAN electronic structure package.
5766 J. Chem. Phys., Vol. 111, No. 13, 1 October 1999 McIntosh et al.

FIG. 1. Pulsed nozzle FT microwave spectra of resolved quadrupole transitions of J76 transitions in a ArIH and b ArIH, and ArID recorded with
broadband scans from 14.0 GHz to 15.0 GHz. The former spectrum was recorded with a supersonic jet expansion of Ar and HI. The latter spectrum was
recorded with an expansion of Ar and DI, the presence of ArIH in the spectrum reflects partial isotope exchange involving the labile DI isotope.

The computed interaction energies were then fit to an
analytical form using the two-dimensional interpolation tech-
nique proposed recently by Ho and Rabitz.29 In this tech-
nique, the points are fitted using a linear combination of a set
of fitting functions which are obtained from the formalism of
the reproducing kernel of the Hilbert space. In the current
study, we used the fitting functions
Q k R,y q n2,m6
1 R,R k q n2
2 y,y k ,
where the point (R k ,y k ) is the kth computed point, y is re-
lated to the angle HIAr through y(1cos )/2, and
where the two one-dimensional kernels are defined as29
and a more widely spaced group of FF transitions. The
assignment was based on a model calculation using the
ground state rotational constant determined from the initial
rovibrational analysis, an assumed structure for the dimer
and the available quadrupole constant 0 for monomeric HI.
This structure was based on estimated changes in the R cm
distance, an elongation of 0.4 , and increase of 6
relative to previous studies of ArBrH.11,12 The assignment
1
was rapidly confirmed by prediction of other quadrupole-
resolved lines associated with other rotational transitions.
The estimated accuracy of the measured transitions is 2 kHz.
The spectral data was then fitted to a Hamiltonian

q n2,m6
1 R,R
1 7
R
14
1
7 R
9 R 2
HH R H Q H SR ,

that is the standard expression for the rotational Hamiltonian

5

and of an effective semirigid linear molecule. To the standard

q n2
2 y,y 1y y 2y
2
y 1 1 y
3 y. 3
expression for the matrix of H Q in the coupled basis set F
IJ, we added diagonal elements for J J(J1) where
J is a coefficient describing the J dependence of the cou-
Thus the final potential is represented by the functional form pling constant (I). During the fit it was found that an ad-
M ditional term, J J 2 (J1) 2 , though small empirically im-
V ab initio R,
k1
k Q k R, 1cos /2 , 4 proved the standard deviation, so it was also added. H SR is
the spin-rotation Hamiltonian, of which only the diagonal
where the expansion coefficients k are determined by re- elements 21 M bb (F(F1)I(I1)J(J1) were used.
quiring the V ab initio reproduce the M computed points. The The resulting constants were determined from a diagonaliza-
rovibrational states were then computed using the variational tion of the Hamiltonian matrix. The errors on the constants
method described in detail previously.3032 reflect the uncertainty of two standard deviation in the fit.
The determined molecular constants are given in Table I.
Corresponding transitions for the deuterated isomer, ArID
RESULTS
were similarly identified based on a refined structure. A seg-
A segment of the J76 rotational transition of ment of the J76 transition is shown in Fig. 1b with an
ArIH is shown in Fig. 1a. The resolved quadrupole com- approximately 50% D labelled sample. This spectrum also
ponents have a characteristic group of F1F transitions shows quadrupole-resolved transitions of the corresponding

Downloaded 07 Jun 2002 to 128.104.69.200. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/jcpo/jcpcr.jsp
J. Chem. Phys., Vol. 111, No. 13, 1 October 1999 Structure and dynamics of ArIH 5767

TABLE I. The rotational constants obtained fron the microwave analysis of

ArIH and ArID.

ArIH ArID

B MHz 1034.08324 1052.359810

D kHz 10.5968 6.94318
H Hz 1.684 2.8110
a (MHz) 1114.402 1409.404
J (kHz) 80.810 21.14
JJ (Hz) 2710
M bb (kHz) 1.5618 1.72
MHz 0.0020 0.0015

ArIH isomer, which occurred in this frequency region and

is present due to incomplete deuteration of the sample. The
corresponding molecular parameters of this isomer are also
given in Table I. For this fit the center frequencies of the
resolved doublet transitions associated with deuterium quad-
rupole splitting were used. An interesting observation is that
the B 0 for ArIH is less than the corresponding value in
ArID, which is attributed to a coupling interaction. The
quadrupole constants of the ArIH and ArID were deter-
mined to be 1114.402 MHz and 1409.404 MHz,
where the uncertainties quoted represent two standard devia-
tions. These values were used to determine the correspond-
ing angles of 149.33 and 157.11, respectively, given in Fig.
2, respectively, using the familiar expression a
0(cos2 )1/2 under the assumption that the presence of
the Ar atom does not modify the charge distribution at I in
the equilibrium form of the complex. The R cm distance was
then determined by optimizing the structure to the deter-
mined rotational constants, also shown in Fig. 2.
A segment of the R(7) to R(11) branches of the rovi-
brationally resolved spectrum of the v 1 in ArIH is shown in
FIG. 3. Segment of the R(7) to R(11) transitions of 1 ArIH recorded
using high frequency wavelength modulation supersonic jet spectrometer in
the frequency range 2228.5 cm1 2228.95 cm1.
Fig. 3. The quadrupole structure associated with these tran-
sitions has collapsed and is not resolved in the current ex-
periments. It is pertinent to note that this spectrum is rela-
tively weak and is detected at the sensitivity limits of the
spectrometer. This is a consequence of the rovibrational ab-
sorption cross section of HI being two orders of magnitude
smaller than corresponding cross sections in HBr. This spec-
trum lacks a Q branch and is characteristic of the
band. It is interesting to note that the band origin is red
shifted by 1.549113 cm1, indicating that the excited state
dissociation energy of this complex is slightly stronger than
that of the corresponding ground state. A similar effect has
been observed for the Ar 79BrH and Ar 81BrH isomers
where the corresponding red shifts were 1.461 1130 cm1
and 1.461 7429 cm1, respectively. In contrast, the corre-
sponding ArHBr complexes were blue shifted 0.336 6511
cm1 and 0.339 7012 cm1 for the 79 and 81 isomers.
Observed P(J) and R(J) transitions were used to fit the
expression
0 B J J 1 D J J 2 J 1 2
B J J 1 D J J 2 J 1 2 . 6
The rotational assignment can be rapidly confirmed by deter-
mining lower state combination differences for these transi-
tions and comparing with the corresponding values based on
the microwave analysis of ArIH reported above. The spec-
trum of the complex was recorded from 2226 cm1 to

TABLE II. The 1 rovibrational constants obtained from the infrared analy-
sis of ArIH.

Constant Value (cm1)

0 2228.032 574
B 0.034 656 478
D 2.96(32)107
FIG. 2. Ground state structures of ArIH and ArID. The R cm values are H 2.1(34)1011
based on rotational constants determined from analysis of the two isoto- 0.000 47
pomers.

Downloaded 07 Jun 2002 to 128.104.69.200. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/jcpo/jcpcr.jsp
5768 J. Chem. Phys., Vol. 111, No. 13, 1 October 1999 McIntosh et al.

TABLE III. Predicted spectroscopic constants from V ab initio and V scaled .

Also given are the experimental data used in the fit with the estimated
uncertainities due to the use of the same potential for both isotopic species
and the dimensionless standard deviations DSD of the two sets of pre-
dicted constants from the experimental data.

V ab initio V scaled Experiment Uncertainty

2 1
ArIH B 0 /10 cm 3.292 3.454 3.449 0.01
ArIH D J /107 cm1 3.43 3.54 3.53 0.01
ArIH P 1 (cos ) 0.799 0.817
ArIH P 2 (cos ) 0.585 0.603 0.610 0.004
ArIH D /106 45.8 40.6 44.3 1.5
ArID B 0 /102 cm1 3.345 3.505 3.510 0.01
ArID D J /107 cm1 2.09 2.34 2.33 0.03
ArID P 1 (cos ) 0.909 0.912
ArID P 2 (cos ) 0.754 0.761 0.771 0.004
ArID D /106 11.3 11.2 15.5 1.5
DSD 10.40 1.81

2230 cm1 and transition frequencies associated with P(4)

to P(19), R(3) to R(21) recorded. In the fittings of these
transitions to the energy expression given in Eq. 6 the
ground state is fixed to the previously and precisely deter-
mined values of B 0 and D 0J available from microwave spec-
troscopy. The resulting fit is given in Table II. The estimated
uncertainties reflect two standard deviation from the fit. Ab-
solute frequency accuracy of the measured transitions was
0.001 cm1.
Using the ab initio calculations described above, we
have predicted the rovibrational energy levels for both iso-
meric systems, ArIH and ArID, assuming that the same
fixed r potential energy surface. We have also considered a
modified version of the potential of the form
V scaled R, c 1 V ab initio c 2 Rc 3 , , 7
where the three parameters, c 1 , c 2 , and c 3 were varied to
obtain the best fit of the spectroscopic constant to the avail-
able experimental data. This modified potential is a type of
morphed potential as discussed by Meuwly and Hutson.33
The spectroscopic constants computed with both of these
potentials are given in Table III. The rotational constants of
the HI molecule used in the kinetic energy term in the
Hamiltonian were those of the free diatomic, 6.341 965 cm1
for HI and 3.223 077 2 cm1 for DI. The uncertainties used
to obtain the fits given in Table III were based primarily on
an estimate of the error in the predicted spectroscopic con-
FIG. 4. A contour plot of the ground state potential of ArIH as a function
of and R. The plot indicates a global minimum consistent with the struc-
ture ArIH and a secondary minimum associated with the isomeric equilib-
rium form ArHI. The contour labels are in units of cm1 .
stants due to the use of the same potential for the DI and HI
forms. The changes due to the shift in the center of mass of
the diatomic are included in the calculation, however, the
changes in the intermolecular potential due to the different
vibrational state of the diatomic are not included.
The values of the parameters for the optimized scaled
potential were c 1 1.0400.004, c 2 0.9870.008, and c 3
0.140.03 . The potential V scaled is given in Fig. 4. Here
we can see that the global minimum is at the 180 ArIH
geometry and there is a local minimum at the 0 ArHI
geometry which has much lower curvature in the direction of
the bending motion. The minima are at V scaled(R3.79 ,
180)220.0 cm1 and V scaled(R4.49 , 0)
172.7 cm1. There is also a secondary minimum at
V scaled(R4.18 , 44)170.5 cm1. The saddle point
between the ArIH and ArHI geometries occurs at
V scaled(R4.22 , 98)131.5 cm1. In Table IV we
have given the spectroscopic constants for a number of low-
lying states for this potential. All of the energies are relative

TABLE IV. Predicted spectroscopic parameters for unobserved states from the V scaled potential. The values of
for the IH and ID isotopomers are relative to the energies of the ArIH and ArID ground states, respectively.

/cm1 B/102 cm1 D/107 cm1 P 1 (cos ) P 2 (cos ) q/102 cm1

ArIH stretch 23.09 3.256 3.97 0.661 0.506

ArIH bend 42.67 2.964 2.78 0.270 0.212 0.410
ArHI ground state 7.89 2.960 1.50 0.638 0.350
ArHI stretch 29.85 2.913 1.15 0.575 0.320
ArHI bend 26.71 3.090 2.30 0.245 0.057 0.025
ArID stretch 25.18 3.322 3.07 0.848 0.653
ArID bend 34.69 3.380 2.64 0.628 0.389 0.211
ArDI ground state 15.19 2.915 2.23 0.772 0.461
ArDI stretch 34.35 2.699 329.73 0.649 0.341
ArDI bend 28.85 3.054 2.66 0.506 0.116 0.003

Downloaded 07 Jun 2002 to 128.104.69.200. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/jcpo/jcpcr.jsp
J. Chem. Phys., Vol. 111, No. 13, 1 October 1999
FIG. 5. Minimum energy path energy profile for the ArHI described in
text illustrating a global minimum associated with the isomer ArIH and
the prediction of two secondary minima with a bound state associated with
the isomeric form ArHI.
to the ground state energy that for this potential has a value
of 146.43 cm1 for ArIH and 158.87 cm1 for ArID.
The difference in the curvature in the bending motion can
also be seen in the energy required for exciting the bending
mode. In the ArIH minimum, this energy is 42.74 cm1
whereas in the ArHI minimum the energy is 18.89 cm1.
The contour plot of the potential energy surface shown
in Fig. 4 can be contrasted with the corresponding plots of
other members of the homologous series ArHX XF, Cl,
and Br. The differences can be rationalized in terms of dis-
persive forces being relatively greater than the dipole
induced-dipole forces thus giving rise to the ground state
ArIH structure. In the case of the other members of the
homologous series the opposite occurs. Another interesting
characteristic of the potential of the ArIH complex can be
seen in the influence of the electron distribution centered at
the H atom being outside the effective Pauling radius of the
halogen. Its effect is reflected in the development of an ad-
ditional minimum in the potential contour plot. Figure 5 il-
lustrates other aspects of the potential. The dashed line
shows the minimum energy path in the potential as a func-
tion of each angle for fixed bending angle. The solid line is
obtained by fixing the angle and calculating the energy of the
ground intermolecular stretching state. The energy levels of
the bending states of the two isomers ArIH and ArHI are
shown. We can see that the inclusion of the zero-point en-
ergy of the intermolecular stretching mode has only a small
effect on the height of the barrier in the bending potential.
We further note that both of the bending states are above
the potential energy barrier separating the isomers ArIH
and ArHI.
CONCLUSIONS
We have recorded the infrared spectrum of the 1 vibra-
tion in ArIH in a supersonic slit jet expansion. This infor-
Downloaded 07 Jun 2002 to 128.104.69.200. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/jcpo/jcpcr.jsp
Structure and dynamics of ArIH 5769
mation was used to predict and then to observe the micro-
wave spectrum of the ground vibrational state of the isomer
using a pulsed nozzle FT microwave spectrometer. The latter
studies have been extended to include the isotopomer, Ar
ID. The rotational constants and quadruple constants
determined for the ArIH and ArID isotopomers are con-
sistent with R cm3.9975(1) , 149.33(1), and R cm
3.9483(1) , 157.11(1). The increase of the angle
on deuteration is consistent with a change in zero point en-
ergy corresponding reduction of the average amplitude of the
proton displacement. It is interesting to note that these values
can be compared with the corresponding values of ArBrH
and ArBrD, R cm3.917(1) , 143.72(1) and R cm
3.850(1) , 150.11(1). Attempts to observe the tran-
sitions associated with alternative isomeric species ArHI
and ArDI were not successful. The ground state of this
dimer thus is in contrast to other members of the homolo-
gous series ArHX (XF, Cl, Br) in which the ground states
have the isomeric form ArHX. A combination of the avail-
able experimental data with ab initio calculations has been
used to model the rovibrational tunneling dynamics. This
modeling yields a potential energy surface in which the glo-
bal minimum has the structure ArIH. The model potential
energy surface predicts a secondary minimum with a bound
state corresponding to the isomeric form ArHI that is 7.6
cm1 less stable than the ground state. We emphasize that
this latter value is an approximate value and needs to be
refined when further experimental information becomes
available.
The conclusions presented here can also be compared to
the results of previous investigations. We are now able to
confirm the ground state ArIH structure inferred from the
proton translational energy distributions in UV ultraviolet
photolysis experiments using the high-n Rydberg time-of-
flight technique by Wittig and co-workers.14,15 In addition
our estimated D 0 143 cm1 for the ArIH system is some-
what deeper than the value of 93 cm1 determined from the
photolysis experiments of Heaven and co-workers.16 Our
value for D 0 would imply that the sequence of D 0 values for
the homologous series of complexes ArHX is strictly in-
creasing, where the dissociation energies previously deter-
mined are D 0 101 cm1 for ArHF,34 D 0 115 cm1 for
ArHCl,35,36 and D 0 121 cm1 for ArHBr.12,35
ACKNOWLEDGMENTS
We are grateful to the National Science Foundation for
supporting this research. A.L.M. wishes to thank the Robert
A. Welch Foundation for financial support in form of the
award of predoctoral fellowship. The support of the Texas
A&M University Supercomputing Facility is also acknowl-
edged. ACL thanks the EPSRC of the United Kingdom for
the award of a Senior Fellowship. Dale Brugh is also thanked
for his assistance in recording the spectra. Throughout the
course of this work, certain name brands were necessary to
adequately describe the experimental parameters. The use of
these names should not be construed as an endorsement by
NIST in any way.
5770 J. Chem. Phys., Vol. 111, No. 13, 1 October 1999 McIntosh et al.

1
K. R. Leopold, G. T. Fraser, S. E. Novick, and W. Klemperer, Chem. Rev. 18
G. Guellachvilli and K. N. Rao, Handbook of Infrared Standards Aca-
94, 1807 1994. demic, New York, 1986.
2 19
cf. contributions reviewed in J. M. Hollas and D. Philips, Jet Spectroscopy R. D. Suenram, J. U-. Grabow, A. Zuban, and I. Leonov, Rev. Sci. In-
and Molecular Dynamics Blackie Academic and Professional, 1995. strum. 70, 2127 1999.
3 20
T. Schroder, R. Schinke, S. Liu, Z. Bacic, and J. W. Moskowitz, J. Chem. E. Zwart, J. J. ter Meulen, W. L. Meerts, and L. H. Coudert, J. Mol.
Phys. 103, 9228 1995. Spectrosc. 147, 27 1991.
4 21
A. McIlvoy, R. Lascola, C. M. Lovejoy, and D. J. Nesbitt, J. Chem. Phys. L. H. Coudert, F. J. Lovas, R. D. Suenram, and J. Hougen, J. Chem. Phys.
95, 2636 1991. 87, 6290 1987.
5
A. McIlvoy and D. J. Nesbitt, J. Chem. Phys. 97, 6044 1992. 22
A. L. McIntosh, T. Walther, R. R. Lucchese, and J. W. Bevan, R. D.
6
A. Ernesti and J. M. Hutson, Faraday Discuss. 97, 119 1994. Suenram, and A. C. Legon unpublished.
7 23
S. M. Cybulski, M. M. Szczesniak, and G. Chalasinski, J. Chem. Phys. K. P. Huber and G. Herzberg, Molecular Spectra and Molecular Structure
101, 10708 1994. IV. Constants of Diatomic Molecules Van Nostrand Reinhold, New York,
8
T. D. Klots and H. S. Gutowsky, J. Chem. Phys. 91, 63 1989. 1979.
9
M. R. Keenan, E. J. Campbell, T. J. Balle, L. W. Buxton, T. K. Minton, P. 24
A. J. Sadlej, Theor. Chim. Acta 79, 123 1991.
D. Soper, and W. H. Flygare, J. Chem. Phys. 72, 3070 1980. 25
A. J. Sadlej, Theor. Chim. Acta 81, 339 1992.
10
D. W. Firth, M. A. Dvorak, S. W. Reeve, R. S. Ford, and K. R. Leopold, 26
A. D. McLean and G. S. Chandler, J. Chem. Phys. 72, 5639 1980.
Chem. Phys. Lett. 168, 161 1990. 27
S. F. Boys and F. Bernardi, Mol. Phys. 19, 553 1970.
GAUSSIAN 94, M. J. Frisch et al. Gaussian, Inc., Pittsburgh PA, 1995.
11 28
J. Han, A. L. McIntosh, C. L. Hartz, R. R. Lucchese, and J. W. Bevan,
Chem. Phys. Lett. 265, 209 1997. 29
T.-S. Ho and H. Rabitz, J. Chem. Phys. 104, 2584 1996.
12
J. M. Hutson, J. Chem. Phys. 91, 4455 1989. 30
J. Han, A. L. McIntosh, Z. Wang, R. R. Lucchese, and J. W. Bevan,
13
A. C. Legon unpublished. Chem. Phys. Lett. 265, 209 1997.
14 31
C. Jacques, L. Valachovic, S. Ionov, E. Bohmer, Y. Wen, J. Segall, and C. A. McIntosh, A. M. Gallegos, R. R. Lucchese, and J. W. Bevan, J. Chem.
Wittig, J. Chem. Soc., Faraday Trans. 89, 1419 1993. Phys. 107, 8327 1997.
15 32
J. Segall, T. Wen, R. Singer, C. Wittig, A. Garcia-Vela, and R. B. Gerber, A. Quinones, G. Bandarage, J. W. Bevan, and R. R. Lucchese, J. Chem.
Chem. Phys. Lett. 207, 504 1993. Phys. 97, 2209 1992.
16
T. Suzuki, H. Katayanagi, and M. C. Heaven, J. Phys. Chem. 101, 6697 33
K. Meuwly and J. M. Hutson, J. Chem. Phys. 110, 8338 1999.
1997. 34
J. M. Hutson, J. Chem. Phys. 96, 6752 1992.
17
A. L. McIntosh, Z. Wang, R. R. Lucchese, and J. W. Bevan in prepara- 35
J. M. Hutson, Annu. Rev. Phys. Chem. 41, 123 1990.
tion. 36
J. M. Hutson, J. Chem. Phys. 89, 4550 1988.

Downloaded 07 Jun 2002 to 128.104.69.200. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/jcpo/jcpcr.jsp