Absolute rate constants for the reactions of germylene and dimethylgermylene with dimethylgermane: The deactivating effect of methyl groups in heavy carbenes

Article abstract of DOI:10.1016/S0009-2614(01)01257-X

Gas-phase rate constants for the title reactions have been obtained by laser flash photolysis at 297 K, by use of photoprecursors, 3,4-dimethyl-1-germacyclopent-3-ene for GeH₂ and pentamethyldigermane for GeMe₂. The values obtained were (k (cm³ molecule^-1 s^-1)): (2.38 ± 0.11) × 10¹⁰ for GeH₂, (2.26 ± 0.10) × 10^-13 for GeMe₂. These results show that the insertion reaction of GeMe₂ is 1050 times slower than that of GeH₂ into the Ge-H bonds of Me₂GeH₂. This is explained in terms of a general mechanism involving an intermediate H-bridged complex, applicable to both silylene and germylene insertions. For the GeMe₂ insertion, reactants are in equilibrium with the complex, which rearranges to the product in the rate controlling step.

Full text of DOI:10.1016/S0009-2614(01)01257-X

4 January 2002
Chemical Physics Letters 351 72002) 47±52
www.elsevier.com/locate/cplett
Absolute rate constants for the reactions of germylene
and dimethylgermylene with dimethylgermane: the
deactivating e€ect of methyl groups in heavy carbenes
a
Rosa Becerra , Mikhail P. Egorov , Irina V. Krylova ,
b
b
b
Oleg M. Nefedov , Robin Walsh
c,
*
a
Instituto de Quimica-Fisica `Rocasolano', C.S.I.C., C/ Serrano 119, Madrid 28006, Spain
N.D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Leninsky Prospekt 47, Moscow 117913,
Russian Federation
b
c
Department of Chemistry, University of Reading, Whiteknights, P.O. Box 224, Reading RG66AD, UK
Received 1 October 2001; in ®nal form 15 October 2001
Abstract
Gas-phase rate constants for the title reactions have been obtained by laser ¯ash photolysis at 297 K, by use
of photoprecursors, 3,4-dimethyl-1-germacyclopent-3-ene for GeH2 and pentamethyldigermane for GeMe2. The
values obtained were 7k ꢀcm3 moleculeÀ1 sÀ1†): ꢀ2:38 Æ 0:11†  10À10 for GeH2, ꢀ2:26 Æ 0:10†  10À13 for GeMe2.
These results show that the insertion reaction of GeMe2 is 1050 times slower than that of GeH2 into the GeAH
bonds of Me2GeH2. This is explained in terms of a general mechanism involving an intermediate H-bridged
complex, applicable to both silylene and germylene insertions. For the GeMe2 insertion, reactants are in equilib-
rium with the complex, which rearranges to the product in the rate controlling step. Ó 2002 Published by Elsevier
Science B.V.
1. Introduction
prototype species, SiH2 and GeH2, are important
in the mechanisms of chemical vapour deposition
7CVD) leading to formation of electronic device
materials [3,4].
Studies of the so-called `heavy carbenes', MR2,
where M ˆ Si, Ge and R ˆ H, Me are of funda-
mental interest because of the ubiquitous involve-
ment of these intermediates in the breakdown
mechanisms of organosilicon and organogerma-
nium compounds [1,2]. Moreover the particular
Gas-phase kinetic studies of silylenes [5] have
shown that SiMe2 is signi®cantly less reactive than
SiH2. For example, at 298 K, rate constants for the
SiAH insertion reactions of SiMe2 7with the
methylsilanes), are in the range 0:2 Â 10À12
À
5:5 Â 10À12 cm3 moleculeÀ1 sÀ1 [6], while those for
SiH2 lie in the range 2:5 Â 10À10 À 4:0 Â
10À10 cm3 moleculeÀ1 sÀ1 [7±10]. The decrease in
* Corresponding author. Fax: +44-118-931-6331.
E-mail address: r.walsh@reading.ac.uk 7R. Walsh).
0009-2614/02/$ - see front matter Ó 2002 Published by Elsevier Science B.V.
PII: S0009-2614701)01257-X

48
R. Becerra et al. / Chemical Physics Letters 351 <2002) 47±52
reactivity has been attributed to the electron
withdrawing e€ect of the methyl groups in the
silylene [5,6].
stored in a transient recorder 7Datalab DL910)
interfaced to a BBC microcomputer. This was used
to average the decays of up to ®ve photolysis laser
shots 7at a repetition rate of 1 or 2 Hz). Typical
decay traces can be seen in our earlier publications
[11,14]. The averaged decay traces were processed
by ®tting the data to an exponential form using a
non-linear least-squares package. This analysis
provided the values for ®rst-order rate coecients,
We have recently begun a series of direct, time-
resolved, measurements of the rate constants of
germylenes, both GeH2 [11±13] and GeMe2 [14].
Until now these studies have mainly focussed on
GeH2, and indeed we have investigated its GeAH
insertion reactions with GeH4 [12] and Et3GeH
[13]. In our studies with GeMe2 we were unable to
detect the insertion reaction with Me3GeH [14],
although there is other evidence that GeMe2 can
undergo GeAH insertion [15]. In this Letter, we
report a reinvestigation of the question of the
GeAH insertion reaction of GeMe2, using
Me2GeH2 as substrate. Me2GeH2 was chosen as
substrate, since by analogy with SiMe2‡ Me2SiH2,
it appears to o€er the best prospect of obtaining a
measurable reaction rate. For comparison pur-
poses the reaction of GeH2 with Me2GeH2 was
also studied. Neither of these reactions has previ-
ously been studied.
k
obs, for removal of GeH2 and GeMe2 in the
presence of known partial pressures of Me2
GeH2.
The photoprecursors for the germylenes were
3,4-dimethyl-1-germacyclopent-3-ene 7DMGCP)
for GeH2 and pentamethyldigermane 7PMDG) for
GeMe2, although 1,1,2,2-tetramethyldigermane
7TMDG) was also used in a few experiments. The
monitoring lasers were a CW argon ion laser
7Coherent Innova 90-5) for GeMe2 and a single
mode dye laser 7Coherent 699-21) pumped by the
Ar ion laser for GeH2. Both germylenes were
1
detected via absorption in their strong Aꢀ B1†
1
arrowXꢀ A1† absorption bands, GeMe2 at 488 nm
[14] and GeH2 at 584.41 nm 717111:31 cmÀ1), a
strong vibration±rotation transition [11,16].
Gas mixtures for photolysis were made up
containing a small pressure of precursor 7DMG
CP, 2.6 mTorr; PMDG, 12.2 mTorr; TMDG, 10.0
mTorr), varying pressures of Me2GeH2 substrate
70±36 mTorr for the GeH2 studies; 0±30 Torr for
the GeMe2 studies) together with inert diluent,
SF6, at a total pressure of 10 Torr 7GeH2 studies
only).
2. Experimental
Germylene kinetic studies have been carried out
by the laser ¯ash photolysis technique, details of
which have been published previously [11,13,14].
Only essential and brief details are therefore in-
cluded here. Germylenes were produced by the 193
nm ¯ash photolysis of gaseous mixtures containing
suitable precursors using a Coherent Complex 100
exciplex laser, operating with ArF. Photolysis
pulses were ®red, at right angles, into a variable
temperature reaction vessel with demountable
windows which were regularly cleaned. The system
is a static reactor. Although a ¯ow system might
o€er some bene®ts, limited supplies of reagents
precluded this. Photolysis pulse energies were
typically 50±70 mJ with a variation of Æ5%. The
monitoring laser beam was multipassed 32 times
along the vessel axis, through the reaction zone, to
give an e€ective path length of 1.2 m. A portion of
the monitoring beam was split o€ before entering
the vessel for reference purposes. Light signals
were measured by a dual photodiode/di€erential
ampli®er combination and signal decays were
All gases used in this work were thoroughly
degassed prior to use. The organogermanium
compounds used in this work were synthesised by
literature methods described previously for
DMGCP [11] and PMDG [14]. TMDG was pre-
pared by a similar method to that of PMDG.
Me2GeH2 was prepared by the LiAlH4 reduction
of Me2GeBr2 in dried n-Bu2O solution. The crude
product was puri®ed by vacuum distillation at )95
°C. All compounds were subjected to low tem-
perature distillation prior to use and were checked
for purity by GC. Purities were DMGCP 7>90%),
PMDG 7>93%) TMDG 792.5%) and Me2GeH2
798:0 Æ 0:5%). SF6 7no GC detectable impurities)
was from Cambrian Gases.

R. Becerra et al. / Chemical Physics Letters 351 <2002) 47±52
49
3. Results
Preliminary checks showed that values for the
least-squares values of the slopes of these lines, are
as follows:
GeH2 ‡ Me2GeH2;
decay constants, kobs, were not dependent on the
exciplex laser energy or the number of photolysis
shots. For each germylene, GeH2 itself, or GeMe2,
a set of runs was carried out in which the
Me2GeH2 substrate pressure was varied over a
suitable range in order to explore the systematic
dependence of kobs upon it. The results of these
experiments are shown in Fig. 1, where the main
graph shows this dependence on a common pres-
sure 7Torr) scale. However, because the reaction of
GeH2 is so fast, the results for these experiments
are also shown as an inset using a mTorr pressure
scale. The kobs data used in these plots were often
themselves averages of more than one set of 5-shot
averages 7up to four sets in some cases). The sets of
results for both germylene species can be seen to
show good linear ®ts, which support second-order
kinetics. It is also worth noting that, for the
GeMe2 reaction, data points using TMDG as
precursor are consistent with the majority for
which PMDG was used as the precursor. This is
the ®rst use of TMDG as a GeMe2 precursor. The
second-order rate constants, obtained from the
k ˆ ꢀ2:38 Æ 0:11†  10À10 cm3 moleculeÀ1 sÀ1
GeMe2 ‡ Me2GeH2;
k ˆ ꢀ2:26 Æ 0:10†  10À13 cm3 moleculeÀ1 sÀ1
These reactions are not likely to be pressure de-
pendent. For GeH2 ‡ Me2GeH2, the rate constant
is either at or close to its upper limit of pressure
dependence both by analogy with SiH2 ‡ Me2SiH2
[7,9,10], and the fact that it is close to the maxi-
mum collisional rate. For GeMe2 ‡ Me2GeH2 the
lack of a pressure dependence was indicated by the
fact that the total pressure was variable 7dependent
on ‰Me2GeH2Š) during the experiments and no
curvature in the second-order plot was found. The
low rate constant obtained for the latter reaction,
must strictly be regarded as an upper limit, be-
cause of the slight possibility of GeMe2 reacting at
close to the collision rate with an impurity. This is
unlikely since most of the ca 2% impurity in
Me2GeH2 is Me3GeH, a molecule of similar re-
activity.
N
Fig. 1. Second-order plots for reactions of GeMe2 ‡ Me2GeH2 7ꢁ, GeMe2 from PMDG;
; GeMe2 from TMDG) and
GeH2 ‡ Me2GeH2 7d). The plot for the latter is also shown inset with a di€erent substrate pressure scale.

50
R. Becerra et al. / Chemical Physics Letters 351 <2002) 47±52
4. Discussion
icant activation e€ect in the substrate germane.
Thus methyl-for-H substitution
7GeH2 ‡
4.1. Rate constants and comparisons
Me2GeH2 compared with GeH2 ‡ GeH4) gives a
factor of 4.36 or, in fact, 8.7 on a per GeAH bond
basis. This we have noted before for ethyl-for-H
substitution [12,13]. These methyl substituent ef-
fects and reaction selectivities can be understood in
terms of a mechanism involving an intermediate
complex.
The main experimental purpose was to measure
the rate constant for the GeMe2 ‡ Me2GeH2 re-
action at 298 K for the ®rst time which has been
achieved. Additionally the rate constant for
GeH2 ‡ Me2GeH2 was obtained for the ®rst time.
There is little doubt that these reactions occur via
GeAH bond insertion 7CAH insertion is known
not to occur [11,14] and GeAC insertion is im-
probable by analogy with the lack of SiAC inser-
tion into SiMe4 [17]). These rate constants are
compared with one another and also with the rate
constants of some other related reactions in Table
1. This brings out the dramatic rate increase for
GeH2 relative to GeMe2 of 1050 in this GeAH
insertion reaction. SiH2 is similarly faster than
SiMe2 in the analogous SiAH insertion reaction,
but only by a factor of 60. Thus methyl-for-H
substitution is more deactivating for a germylene
than a silylene.
4.2. The intermediate complex mechanism
We have shown in earlier work that the rates of
silylene [5,6,8] and germylene insertion [12,13]
processes can be explained by a mechanism in-
volving an intermediate complex. For the present
work the mechanism may be written:
R2Ge ‡ HGeHMe2
1
2
ꢀ R2Ge Á Á Á H Á Á Á GeHMe2 ! R2HGeGeHMe2
À1
ꢀR ˆ H; Me†
These comparisons also bring out two other
features. First, germylene insertions into GeAH
bonds are slower than silylene insertions into
SiAH bonds. This we have noted before [13], but it
can now be seen that the relative rates are highly
variable. For example, the comparison of
SiMe2 ‡ Me2SiH2 with GeMe2 ‡ Me2GeH2 gives
a factor of 24, much larger than the value of 1.4 for
the comparison of SiH2 ‡ Me2SiH2 with
GeH2 ‡ Me2GeH2. The rate ratio for SiH2 ‡ SiH4
relative to GeH2 ‡ GeH4 is 8.4 and lies in between.
Clearly GeMe2 is signi®cantly less reactive than
SiMe2. Secondly the methyl groups have a signif-
For the GeH2 insertion, the rate constant is rela-
tively high, suggesting that the overall process is
largely controlled by step 71) the molecular en-
counter, and the rearrangement of the complex,
step 72), is fast compared with redissociation, step
7)1). This is more like the silylene analogue, and
can be understood in terms of the substrate methyl
group e€ects, viz. that methyl substitution lowers
the barrier 7E2) for the second step. Exactly this
e€ect is found in both the reaction series SiH2‡
MenSiH4Àn 7n ˆ 1±3) [7±10] and SiMe2 ‡ Men
SiH4Àn ꢀn ˆ 1±3† [6]. It is worth noting that ab
initio calculations [13] have shown that for the
Table 1
Comparison of gas-phase rate constants for GeAH insertion 7GeMe2; GeH2) and SiAH insertion 7SiMe2; SiH2) at 298 K
Reaction
k
krel
Ref.
7cm3 moleculeÀ1 sÀ1
)
GeMe2 ‡ Me2GeH2
GeH2 ‡ Me2GeH2
SiMe2 ‡ Me2SiH2
SiH2 ‡ Me2SiH2
GeH2 ‡ GeH4
2:3 Â 10À13
2:4 Â 10À10
5:5 Â 10À12
3:3 Â 10À10
5:5 Â 10À11
4:6 Â 10À10
1a
This work
This work
1050
1a
[6]
[7]
60
1a
8.4
[13]
[8]
SiH2 ‡ SiH4
a Reference reaction 7of each pair).

R. Becerra et al. / Chemical Physics Letters 351 <2002) 47±52
51
prototype reactions the complex H2Ge Á Á Á H Á Á Á
GeH3 is less strongly bound than the complex
H2Si Á Á Á H Á Á Á SiH3. This means that step 7)1) oc-
curs more readily in the germylene case, meaning
that germylene complexes are intrinsically more
likely to redissociate and therefore anything which
lowers E2 will have a more marked e€ect on
germylene insertions than on silylene insertions. A
more detailed analysis of this will given in a future
publication describing a systematic study of the
kinetics of the reactions GeH2 ‡ EtnGeH4Àn
7n ˆ 1±3) [18].
zwitterionic complexes R2Si Á Á Á OMe2. Kinetic
studies [20] give binding energy of ca
a
37 kJ molÀ1 for Me2Si Á Á Á OMe2, whereas kinetic
studies [21] supported by theory [22] suggest a
binding energy of ca 84 kJ molÀ1 for
H2Si Á Á Á OMe2. This is further supported by cal-
culations [23] on analogous water complexes,
R2Si Á Á Á OH2. If the same considerations apply to
R2Ge Á Á Á H Á Á Á GeHMe2, then the weakness of this
complex, when R ˆ Me, will lower the binding
energy, which corresponds to EÀ1, relative to
R ˆ H. Such a lowering could account for the
switch to step 72) being fully rate determining and
the high overall retardation e€ect for GeMe2‡
Me2GeH2.
For the GeMe2 insertions, the rate constant is
so low, that it suggests that step 72) is fully rate
determining, i.e., k2 ꢂ kÀ1. Baggott et al. [6] have
analysed the factors which a€ect the ratio k2=k in
There have been no theoretical calculations on
the insertion of GeMe2 into GeAH bonds. How-
ever, DFT and MP2 studies by Su and Chu [24,25]
show that 7a) GeMe2 inserts less readily than
GeH2 into the CAH bond of CH4, although both
reactions have high activation barriers [24], and 7b)
GeMe2 inserts more readily into the SiAH bond of
SiH4 7Ea ˆ 66 kJ molÀ1) than into the CAH bond
of CH4 7Ea ˆ 164 kJ molÀ1) [25].
À1
the case of SiMe2 insertions and shown that rate
factors of 10À3 can arise between di€erent cases as
a result of rate controlling step switching from 71)
to 72). For the SiMe2 ‡ Me2SiH2 case at 298 K the
situation is in between. Thus while step 72) in this
case is rate determining, it does not produce the
highest retardation e€ect. For GeMe2 ‡ Me2
GeH2, however, the measured factor is close to
10À3. Thus this reaction is close to one extreme
limit of behaviour. This will be the case when
E
ꢃ E2. The implication of this is that this re-
Acknowledgements
À1
action should have an overall activation energy
close to zero. We plan to test this. The underlying
reason why the insertion reactions of both SiMe2
and GeMe2 are so much slower than those of their
SiH2 and GeH2 prototypes has been attributed, in
the silicon case [5,6], to the electron withdrawing
ability of the methyl groups. This arises from the
fact that C is more electronegative than Si. It was
argued that orbital contraction in SiMe2 7and
therefore also GeMe2) would result and therefore
require shorter range, closer contact with sub-
strates for reaction to take place. However, al-
though this may be a contributing factor we are
now more inclined to believe that it is due to the
weakness of initial bond making in the complex.
Methyl groups are known to stabilise SiMe2. It has
a higher divalent state stabilisation energy 7DSSE)
value 7128 kJ molÀ1) than SiH2 794 kJ molÀ1) [19].
The more stable the silylene, the weaker is likely to
be the bonding in the intermediate complex. Some
support for this comes from a comparison of the
We thank the following: INTAS-RFBR 7pro-
ject IR-97-1658), NATO 7project PST.CLG.
975368). R.B. also thanks the DGICYT 7Spain)
for support under projects PB98-0537-C02-01 and
BQU2000-1163-C02-01. M.P.E., I.V.K. and
O.M.N. also thank RFBR 7projects 01-03-32630
and 00-03-32630). R.B. also thanks EPSRC 7pro-
ject C7410). We thank Sergey Boganov for helpful
comments and additionally Keith King and War-
ren Lawrance for an advance copy of their paper.
References
[1] Z. Rappoport, Y. Apeloig 7Eds.), The Chemistry of
Organic Silicon Compounds, vol. 2, Wiley, Chichester,
1998.
[2] S. Patai 7Ed.), The Chemistry of Organic Germanium, Tin
and Lead Compounds, vol. 1, Wiley, Chichester, 1995.
[3] J.M. Jasinski, S.M. Gates, Acc. Chem. Res. 24 71991) 9.
[4] C. Isobe, H. Cho, J.E. Crowell, Surf. Sci. 295 71993) 117.

52
R. Becerra et al. / Chemical Physics Letters 351 <2002) 47±52
[5] R. Becerra, R. Walsh, Kinetics and mechanisms of silylene
[14] R. Becerra, S.E. Boganov, M.P. Egorov, V.Ya. Lee, O.M.
Nefedov, R. Walsh, Chem. Phys. Lett. 250 71996) 111.
[15] E.C-L. Ma, D.P. Paquin, P.P. Gaspar, J. Chem. Soc.,
Chem. Commun. 71980) 381.
reactions: a prototype for gas-phase acid/base chemistry,
in: R.G. Compton, G. Hancock 7Eds.), Research in
Chemical Kinetics, vol. 3, Elsevier, Amsterdam, 1995, p.
263 7Chapter 6).
[16] A. Campargue, R. Escribano, Chem. Phys. Lett. 315 71999)
397.
[6] J.E. Baggott, M.A. Blitz, H.M. Frey, R. Walsh, J. Am.
Chem. Soc. 112 71990) 8337.
[17] R. Becerra, R. Walsh, Int. J. Chem. Kinet. 31 71999) 393.
[18] R. Becerra, S.E. Boganov, M.P. Egorov, I.V. Krylova,
O.M. Nefedov, R. Walsh, to be published.
[7] J.E. Baggott, H.M. Frey, P.D. Lightfoot, R. Walsh, I.M.
Watts, J. Chem. Soc. Faraday Trans. 86 71990) 27.
[8] R. Becerra, H.M. Frey, B.P. Mason, R. Walsh, M.S.
Gordon, J. Chem. Soc. Faraday Trans. 91 71995) 2723.
[9] I.W. Carpenter, Ph.D. Thesis, University of Reading, 1996.
[10] I.W. Carpenter, R. Walsh 7in preparation).
[11] R. Becerra, S.E. Boganov, M.P. Egorov, O.M. Nefedov, R.
Walsh, Chem. Phys. Lett. 260 71996) 433.
[19] R. Becerra, R. Walsh, Thermochemistry 7Chapter 4), p.
153 in Ref. [1].
[20] J.E. Baggott, M.A. Blitz, H.M. Frey, P.D. Lightfoot, R.
Walsh, Int. J. Chem. Kinet. 24 71992) 127.
[21] U.N. Alexander, K.D. King, W.D. Lawrence, Phys. Chem.
Chem. Phys. 3 72001) 3085.
[12] R. Becerra, S.E. Boganov, M.P. Egorov, O.M. Nefedov, R.
Walsh, Mendeleev Commun. 71997) 87.
[22] M.W. Heaven, G.F. Metha, M.A. Buntine, J. Phys. Chem.
A 105 72001) 1185.
[13] R. Becerra, S.E. Boganov, M.P. Egorov, V.I. Faustov,
O.M. Nefedov, R. Walsh, J. Am. Chem. Soc. 120 71998)
12657.
[23] S. Su, M.S. Gordon, Chem. Phys. Lett. 204 71993) 306.
[24] M-D. Su, S-Y. Chu, J. Am. Chem. Soc. 121 71999) 4229.
[25] M-D. Su, S-Y. Chu, J. Phys. Chem. A 103 71999) 11011.