(
)
292
U.N. Alexander et al.rChemical Physics Letters 299 1999 291–295
sure the reaction rate constants for GeH2 prepared
by photolysis of PhGeH3 with the aim of identifying
the mechanism responsible for the anomalous rate
constants found for this precursor. Identification of
the mechanism that leads to anomalous rate con-
stants for PhGeH3 would add confidence that the
values reported for DMGCP are indeed accurate.
While the experimental technique used in our study
is similar in most respects to that used by BBENW,
our method incorporates two features, a flow cell and
background subtraction, that have the potential to
eliminate possible sources of error that might ac-
count for the discrepancy reported in BBENW’s
work.
tensity fluctuations. The differential amplifier output
Ž
is recorded using a digital oscilloscope Hewlett
.
Packard 54510A . Typically between 64 and 128
traces are averaged at each reactant pressure, and a
minimum of three such averaged traces are recorded
and analysed. The experiment is repeated at a num-
ber of reactant pressures. To minimise non-germy-
lene contributions to the signal, two sets of decay
traces are recorded, one with the laser tuned to a
Ž
.
GeH2 absorption signal and one with the laser
Ž
.
detuned from GeH2 background . The background
is subtracted from the signal to provide the GeH2
decay signal. The method of analysis of the GeH2
decay signal is discussed when presenting the results.
The precursor, reactant and a buffer gas continu-
Using PhGeH3 we have reproduced the reaction
rate constants reported by BBENW for GeH2 using
Ž
ously flow through the reaction cell 1 m long=25
Ž .
.
Ž
.
DMGCP as a precursor, indicating: 1 that PhGeH3
mm dia. at room temperature 295 K . The flow rate
and excimer laser repetition rate are adjusted to
ensure that reaction product concentrations are negli-
gible. The gas mixture is controlled using mass flow
is a suitable GeH2 precursor for kinetic measure-
ments under appropriate experimental conditions; and
Ž .
2 that BBENW’s assertion that their DMGCP data
provide the more reliable set of rate constants is
correct. We find that photolysis of PhGeH3 at wave-
lengths of 193 and 248 nm leads to the same values
for the GeH2 rate constants, indicating that vibra-
tional relaxation is not affecting the measured values.
We here report the details of this investigation.
Ž
.
controllers MKS 1159B and 2159B . The rate con-
stants reported here were obtained with a total cell
pressure of 10 Torr. The PhGeH3 pressure was
typically 10 mTorr for experiments using 193 nm
photolysis and 25 mTorr using 248 nm photolysis.
The reactant pressure was typically varied in the
Ž
.
range 20–120 mTorr. N2 99.9%, BOC and SF6
Ž
.
99.9%, BOC were used as buffer gases. Reaction
2. Experimental details
rate constants were found to be independent of the
The experimental system is based on that used in
buffer gas used.
w
x
our extensive studies of methylene kinetics 5–7 .
An excimer laser, in this case operating at 193 or
248 nm, photolyses a precursor to produce GeH2.
The GeH2 concentration is followed by time-re-
solved laser absorption using a single-mode cw ring
dye laser tuned to a band in the GeH2 absorption
spectrum. We have used the 74GeH2 band at
17111.31 cmy1, the same band used by BBENW,
which provides a monitor of the population in the
BBENW used 193 nm as their PhGeH3 photoly-
sis wavelength. In order to provide a direct compari-
son with their experiment we have also used this
wavelength. However, for i-butylene and trimethylsi-
lane, the two reactants with the largest discrepancy
between the reported rate constants using DMGCP
vs. PhGeH3 as precursors, we have used 248 nm as
an additional photolysis wavelength. Should vibra-
tionally excited GeH2 be produced in the photolysis
step, its collisional relaxation provides a growth
Ž
Ž
..
ground vibrational state designated 0, 0, 0 . The
probe laser is multipassed through the cell to achieve
a path length of 4 m, thereby increasing the absorp-
tion signal. To further increase the signal to noise
ratio, a portion of the probe laser is split off prior to
the reaction cell and used as a reference. The signal
and reference dye laser beams are imaged onto sepa-
rate photodiodes whose outputs are processed by a
differential amplifier, largely removing dye laser in-
Ž
.
mechanism for the GeH2 0, 0, 0 and might explain
the observations of BBENW. Such circumstances
can be illuminated using different photolysis wave-
lengths: a difference in the GeH2 vibrational popula-
tions will be produced using different photolysis
photon energies and so different GeH2 removal rate
constants would be observed. This effect has been
w x
reported for singlet methylene kinetics 8 .