- Global warming is becoming an increasingly important environmental concern and
CO₂ is considered as the major cause of global warming. Creating useful applications
for CO₂ would generate alternatives to merely venting CO₂ to the atmosphere, and
decreasing the carbon intensity of human activities. Among various methods of CO₂
utilization, conversion of CO₂ to value-added chemical products is the most attractive.
In this study, microtechnology and application of corona discharge are combined and
introduced into CO₂ reduction process.
Experiments were conducted before the design and manufacture of the microreactor
to determine the configuration of the reactor, products of reaction and effect of active
volume ratio: (1) it is proved that through-gas corona discharge are much simpler to
implement than through-liquid, especially those with high solubility of CO₂; (2) the
V-I curve for the corona discharge reactors is characterized by a transition from a high
voltage low current (which we call spark discharge) to a low voltage high current state
(the corona discharge state), with this transition used as a diagnostic of the reactor
operation in the corona discharge mode even when direct observation of the discharge
is not practical; (3) the products for the reduction of dry CO₂ are carbon monoxide
(CO) and oxygen (O₂), while for the reduction of wet CO₂, methane (CH4) is also formed in addition to these two products; (4) a larger active volume ratio results in
higher conversion of CO₂ to products.
Based on the experimental results listed above, two multi-discharge microscale-based
corona reactors were designed and manufactured. We found that for a needle-to-plate
gap of 110μm, at the voltage of 0.840kV and current of 0.62mA, a flow of CO₂ and
H₂O mixture (flow rate of CO₂ = 50sccm, CO₂-to-H₂O molar ratio = 1:2) can result in
5.5~6% conversion of CO₂ (with 40~50% conversion within the active volume of the
reactor) with energy efficiency of 85~95%. The influence of the three main factors,
namely the power applied to the reactor (specific points of the V-I curve used), flow
rate of CO₂, and CO₂-to-H₂O molar ratio, on the performance and energy efficiency
of the reactor were investigated. It was found that (1) the glow regime (or corona
regime) is the optimal operation regime for this process from both conversion and
energy efficiency perspectives, with higher current in this regime resulting in higher
CO₂ conversion; (2) lower flow rate of CO₂ can result in higher conversion with lower
energy efficiency, and conversely, the highest energy efficiency is achieved at the
highest flow rate; (3) the conversion of CO₂ increases as the CO₂-to-H₂O molar ratio
decreases, but the highest energy efficiency is achieved when this ratio matches the
ratio of stoichiometric numbers.
A numerical model of the process reflecting the geometry, momentum balance,
material balance and kinetics inside the reactor was developed to help understand the
chemical reaction process. The reaction scheme was modelled as being driven by the
initial cleavage of a CO₂ or H₂O molecule caused by collision with energetic electrons
to produce CO+O or OH+H, followed by cascaded spontaneous reactions that yield
the products. The reaction kinetics were approached as following pseudo-Arrhenius
laws with a pre-exponential term k₀ and an exponential term, but rather than
modelling the exponential term as dependent on temperature, the term was modelled
as depending on the applied electrical potential in the corona discharge (i.e., rather than dependent on e^(-EA/RT) it was modelled as dependent on e^(-EA/βVF), where β is an
effectiveness parameter, V is the applied electrical potential to the discharge, and F is Faraday's constant). A refined parameter n that defines the fraction of CO intermediate that either further fragment or remain as CO in the product stream, termed the kinetic parameter in this work, was also employed.
Optimization of the numerical model was applied to extract kinetic parameters for the CO₂ reduction process in the multi-point corona discharge with good agreement between simulated results and experimental data. The values of the final refined parameters were: for the initial dissociation of CO₂, the final values of the refined parameters were k₀,₁ = 3.543±0.071×10¹⁰sec⁻¹ and the electrical potential effectiveness parameter was β₁ = 2.181±0.044×10¹, and for the dissociation of H₂O, the parameters were k₀,₁ = 1.266±0.025×10⁸sec⁻¹, β4 = 8.403±0.168, and the kinetic parameter n = 0.50±0.010.