936
J. Chem. Phys., Vol. 111, No. 3, 15 July 1999
King, Herring, and Eden
ϩ
Ϫ
→
X(0–2) band but, in general, the In –I pair yield shows
little influence from B→X transitions. In the same vein, the
2
In(6 S ) yield spectrum exhibits virtually no structure as-
1
/2
signable to A–X transitions. One concludes, therefore, that,
in the spectral region studied here, A–X transitions offer
insufficient energy to photodissociate InI ͑yielding the
2
In(6s S1/2) species͒ but metal–halide ion pairs are pro-
duced, whereas the opposite is true if one drives B←X tran-
sitions. When exciting B←X(vЈ–vЉ) bands in the 395–410
2
nm region, production of In(6s S ) predominates over the
1/2
formation of ion pairs. Since the excitation of the A–X(2,0)
ϩ
Ϫ
transition produces In –I ion pairs but not the neutral spe-
2
cies In(6 S), one readily finds that the lower limit for the
ϩ
Ϫ1
XO dissociation energy is ϳ25 143 cm . This estimate is
Ϫ1
FIG.
In(ns S1 np P
10.
Fluorescence
suppression
spectrum
of
ϳ100 cm smaller than, but consistent with, the lower limit
2
2
2
2
,nd D
←6s S , 25рnр15) transitions ob-
/2,
1/2,3/2
3/2,5/2
1/2
suggested in the last section.
2
2
tained by monitoring In(6 S1 →5 P ) spontaneous emission at 410.2
/2
1/2
The value for De proposed here, 25 340Ϯ90 cmϪ1,
agrees well with the estimate by Vempati and Jones of
2
nm. In this wavelength region, production of the 6 S state requires the
1/2
11
absorption of three photons.
ϩ
Ϫ1
D (XO )ϭ25 296 cm , assuming the InI ground state to be
0
ϩ
covalently bonded and the AO , B1, and C1 states are de-
rived from In( P )ϩI( P3/2). As discussed in Ref. 11, this
value for the XO dissociation energy is consistent with
measurements from flame photometry ͑ϳ25 880 cm , Ref.
2
2
multiphoton ionization spectrum of the Column III B alkyls
3
/2
ϩ
Al2͑CH3͒6, Ga͑CH3͒3, and In͑CH3͒3 that were reported by
Hackett, John, and Mitchell,2
1–23
who demonstrated that the
Ϫ1
Ϫ1
absorption of 4–6 photons in the visible completely removes
the methyl ligands and yields a metal ion. Thus, the mul-
tiphonon ionization ͑MPI͒ spectrum of the alkyl in the vis-
ible has the signature of solely the metal atom. In the case of
8
͒, the upper limit of 27 630 cm determined from photo-
ionization experiments ͑Ref. 7͒, and the range in values pro-
Ϫ1
posed in Ref. 1 ͑21 700–27 430 cm ͒. It should also be
11
noted that the estimates of D (X) by Vempati and Jones
0
2
In͑CH ͒ , two-photon ionization of the In 6 S state, reso-
3
3
1/2
were drawn from Morse potential extrapolations and, as the
authors mention, ‘‘...should be considered as upper limits.’’
2
nantly enhanced by the np P states (13рnр30) was
J
observed2 but, to our knowledge, the single-photon forbid-
2
2
2
den, md D3
←6s S
transitions have not been re-
/2,5/2
1/2
ported previously.
C. Fluorescence suppression spectra:
2
2
2
2
S1/2 , P
, D
—6s S transitions of In
1/2,3/2
3/2,5/2
1/2
IV. CONCLUSIONS
For excitation wavelengths beyond ϳ410 nm, populat-
ing the 6 S state of In from InI requires the absorption of
2
2
The relative, wavelength-dependent yield of In(6s S1/2)
from InI, and the InI ground state (XO ) dissociation energy
1
/2
ϩ
three photons. Since the threshold wavelength for photoion-
2
ization of the 6s S1 level by two photons of equal energy
have been measured by excitation spectroscopy and laser-
induced fluorescence in the spectral region that overlaps the
/2
2
is 448 nm, excitation spectroscopy of In(6s S1/2) in the
ϩ
ϩ
Ͼ448 nm region is a convenient tool for examining high
AO , B1←XO transitions of the molecule ͑390–431 nm͒.
Seven previously unreported B←X bands have been ob-
served and the absence of vibrational bands for which vЈ
Ͼ6 suggests that the B1 state is predissociated. The pres-
ence or absence of specific A←X or B←X bands in the
Rydberg states of atomic In while simultaneously probing
2
In(6s S1/2) production. Figure 10 shows the fluorescence
suppression spectrum for indium in the 450–470 nm region
2
2
that was recorded by monitoring 6s S1 →5p P emis-
/2
1/2
2
sion at 410.2 nm. The strongest features correspond to the
In(6 S ) excitation spectrum or photoionization spectra
1/2
2
2
ϩ
dipole-allowed np P1
←6s S transitions, and princi-
permit the XO ground-state dissociation energy to be deter-
/2,3/2
1/2
Ϫ1
pal quantum numbers nу25 are readily observable. Also, the
mined to be D ϭ25 340Ϯ90 cm , which agrees with the
e
2
fine structure splitting in the P states has not been reported
lower value suggested in Ref. 11—the value of D to be
e
ϩ
ϩ
previously for the nϭ15–17 terms of the series, but the
values measured in these experiments ͑shown in Fig. 10͒ are
in agreement with the expression given by Neijzen and
expected if: ͑1͒ XO is covalent, and ͑2͒ the AO , B1, and
C1 states are all correlated, in the separated atom limit, with
2
2
In(5p P3 )ϩI( P3/2). Deducing De from direct spectro-
/2
2
0
Ϫ3
Ϫ1
D o¨ nszelmann: ⌬T ϭ5880n*
͑in cm ͒, where ⌬T is
scopic evidence, rather than from the extrapolation of a
Morse potential vibrational series, yields a value more accu-
rate than those available previously, due in part to the fact
that the result proposed here is not dependent upon the struc-
n
n
2
the P
splitting and n* is the effective quantum num-
/2–3/2
1
ber.
Also noticeable in Fig. 10 are single-photon transitions
2
2
ϩ
terminating on the ns S1 (15рnр25) and md D
ture of the XO state near the dissociation limit. The
/2
3/2,5/2
2
(
13рmр25) states of the atom. These transitions were
In(6s S1/2) relative yield data, particularly in the 395–407
found to quickly vanish as the laser pulse energy was de-
creased, but were observed for principal quantum numbers as
low as mϭ9. The spectrum of Fig. 10 is reminiscent of the
nm region, exhibit a strong dependence upon the B-state ro-
tational quantum number, which reflects the fact that the
photodissociation process is near threshold in this spectral
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.105.215.146 On: Wed, 24 Dec 2014 05:46:49