160 / 390
Reaction of Si with CuCl
CuCl was found to effectively catalyze the direct process reaction and significantly
reduce the length of the induction period.
18
Its reaction with silicon is seen as a key
reaction and a model for forming catalytically active surface species in the direct
process.
19
The thermodynamics and kinetics aspects of the Si and CuCl reaction has
been extensively studied. It is proposed that the reaction pathway go through reactions
10 and 11 to generate SiCl
4
and copper silicides (Cu
x
Si, x = 3, 5, etc., depending on Si
and CuCl ratio).
20-22
This manuscript reports new methods for investigating this
reaction.
Si + 4CuCl ––> 4Cu + SiCl
4
(10)
Si + xCu ––> Cu
x
Si (x = 3, 5, etc., depending on Si and CuCl ratio)
(11)
Experimental details
Reaction mass was loaded into a reactor with a constant flow of N
2
passing through.
The reactor was heated up to and maintained at 320 °C. The gas products were
separated isothermally by an Agilent 7890 GC instrument with 100 °C column
temperature and 3.6 mL/min helium flow. A thermal conductivity detector (TCD) was
used. Peak identifications were confirmed by injecting known standards. Based on the
N
2
flow rate (45 sccm, 3.4 g/hr) and N
2
-to-SiCl
4
peak-intensity ratio, SiCl
4
formation
rate was calculated. The reaction was also performed in a chemisorption analyser
(Micromeritics AutoChem II) for better temperature control. The SiCl
4
product was
monitored by a mass spectrometer (MKS-Cirrus) with cycle-time of 0.1 minute. In
these experiments, ~0.5 g reaction mass was used and N
2
flow rate was set at 40
sccm.
Results and discussion
Figure 1 shows a GC chromatogram of the gas product mixture from the reaction of
Si with CuCl. The GC method described in the experimental details allowed good
separation of all the products exiting the reactor. The GC cycle time was slightly
shorter than 2 minutes. We found that increasing the column temperature or flow rate
did not significant decrease cycle time. On the other hand, the N
2
and HCl peaks co-
eluted at higher temperatures. SiCl
4
was found to be the main product, accounting for
greater than 95% of the mixture excluding N
2
. There were also small amounts of HCl
and HSiCl
3
. We speculate that HCl was generated from reaction of SiCl
4
with trace
amount of moisture left in the reaction mass. The HSiCl
3
could be generated from
reaction of HCl gas with Si. Less HCl and HSiCl
3
were generated from the reaction
mass with longer drying time, confirming that moisture was involved in formation of
these by-products. The finding here supports previous proposals that SiCl
4
is the main
gas phase product from the reaction of Si with CuCl.
20
152