Optimized Heating Rate and Soot-catalyst Ratio for Soot Oxidation over MoO3 Catalyst

MoO3 is now utilized as a promising catalyst due to its high activity and favorable mobility at low temperature. Its spectral data and surface microstructures were characterized by Fourier transform infrared spectra (FT-IR) and Field emission scanning electron microscope (FESEM). Thermo-analysis of the carbon black was performed over nano-MoO3 catalyst in a thermogravimetric analyzer (TGA) at various heating rates and soot-catalyst ratios. Through the analysis of kinetic parameters, we found that the heat transfer effect and diffusion effect can be removed by setting lower heating rates and soot-catalyst ratios. Therefore, a strategy for selecting proper thermogravimetric parameters were established, which can contribute to the better understanding of thermo-analytical process. Copyright © 2017 BCREC Group. All rights reserved


Introduction
Soot particulates in the exhaust from diesel engines have brought great harm to human health, which touches off a rapid technology development in emission reduction.Filters can capture soot particulates from the exhaust, but the filters must be regenerated periodically [1][2][3].This process can be continuous if a highperformance catalyst coated on the filter can make the device regenerate under practical conditions.It is very important to find some innovative catalysts which can reduce the combustion temperature of particulates.
Several catalytic systems based on transition metal oxides, single or mixed, have been explored for soot oxidation [4].Among them, MoO3 seems to be the most promising one.Recent researches have shown that MoO3 presents an extremely active performance for soot combustion reaction allowing for lower combustion temperatures around 500°C and higher selectivity to CO2 [3,5].This remarkable performance may be due to the favorable mobility of molybdenum oxide, either by the surface migration or melting and gas phase transportation with a high partial pressure of MoO3 [3].
Proper thermogravimetric parameters such as model gas composition, flow rate, crucible type, heating rate and soot-catalyst ratio are very important for catalytic oxidation experiments.Otherwise, inappropriate setup could make the kinetics-controlled regions deviate into diffusion-controlled regions [6,7].Especially, the heating rate exerts important effects on results of soot oxidation mainly due to the great influence of temperature on the oxidation reactivity [8].
Nevertheless, the catalyst performance was heavily affected by soot-catalyst ratios.There is no clear specification about soot-catalyst ratios, although many researchers have mixed small quantity of soot with large quantity of catalysts to assure quick combustion of soot.It is still difficult to define an optimal soot-catalyst ratio to avoid diffusion effects.
In this work, soot oxidation kinetics under the catalytic role of MoO3 was comprehensively analyzed at various heating rates as well as soot-catalyst ratios, based on Coats-Redfern method.With the assistance of kinetic parameters analysis, a strategy for the determination of optimized thermo gravimetric parameters was developed.

Sample preparation
A commercial carbon black supplied by Degussa Company was used as a model of diesel soot.It owns a specific surface area of 100 m 2 /g, with an average size of 25 nm and the density of 130 g/L.The carbon black has been traditionally applied to replace diesel soot in TG experiments due to its fine repeatability [6].The catalyst adopted in this study is a commercial nano-scale molybdenum trioxide MoO3 from Beijing DK Company (purity 99.9%).In general, the catalyst and soot were mixed in a tight contact state either or a loose contact state.The tight contact can be obtained by milling catalyst with soot together.However, the actual contact between the soot with catalyst in filters seems loose.So, the loose contact state was usually prepared for catalytic oxidation experiments.In this study, samples were prepared by mixing catalysts and the carbon black with ratios of 95:5, 90:10, and 85:15, respectively.The particulate concentrations are referred to as 5 wt%, 10 wt%, and 15 wt%.The loose contact was achieved by manually mixing catalysts with the carbon black in a mortar with a spatula for 3 minutes till a homogeneous state [8,9].

Catalyst Characterization
Thermogravimetric experiments were conducted in a TGA/DSC1 analyzer, while the Fourier transform infrared (FT-IR) spectra were recorded with a Nicolet Nexus 470 spectrometer.It can detect and recognize a variety of spectral data.The main technical parameters of the TGA/DSC1 analyzer are shown in Table 1.The initial quantity of sample was about 6 mg and the heating rates were set at 10, 15 and 20 °C/min, respectively.In accordance with the practical oxygen conditions in diesel exhaust, a simulation atmosphere of 13% O2/N2 in furnace was specified with a gas flow rate of 60 mL/min.All runs were shown in Table 2.

Mathematical modeling
Kinetic studies were performed in a thermogravimetric analyzer based on Coats-Redfern method [10].Usually, the change in extent of reaction (α) is used to research the solid state reactions kinetics: (1) In this equation, mo, m, dan m∞ are initial sample mass, sample mass at time t and sample mass at the end of experiment, respectively.Using extent of reaction, the rate of a solid state reaction can be usually described by: ( The temperature dependence of the reaction rate is usually described by the Arrhenius equation: (3) In this equation, A is the pre-exponential factor, E is the activation energy, T is the absolute temperature and R is the molar gas constant.The equation of f(a) = (1-a) n is the reaction model.Under non-isothermal conditions, in which a sample is heated at a constant rate, the explicit temporal in Equation ( 2) is eliminated through the trivial transformation: (4) In this equation, β is the constant heating rate.Finally, Coats-Redfern equations were obtained through integral transformation: When n≠1: (5) Plotting the left hand side of Equation ( 6), which include -ln(1-a) versus 1/T, gives E and A from the slope and intercept, respectively.

Catalyst Characterizations
Fourier transform infrared (FT-IR) spectra were recorded with a Nicolet Nexus 470 spectrometer.It can detect and recognize a variety of spectral data.The FT-IR spectrogram of MoO3 was shown in Figure 1.The corresponding O-H stretching modes are responsible for the broad bands at 3429 cm -1 and 1628 cm -1 , which indicate small amounts of water molecules absorbed by the surface of MoO3.The sharp strong peaks at 988 cm -1 and 1383 cm -1 are supposed to be the stretching modes of (Mo=O) groups.Two evident and broad bands formed at 875 cm -1 and 624 cm -1 can be confidently assigned to Mo-O-Mo groups.The results of fourier infrared analysis are basically identical with literature [11].
The structure and morphology of MoO3 were reflected by images from an S-4800 field emission scanning electron microscope (FESEM) operating at 15 kV.Before FESEM detecting, the sample was dusted on an adhesive conductive carbon paper attached on a brass mount and coated with platinum.
The FESEM image of MoO3 is presented in Figure 2. A large number of microspheres with diameters of 100-200 nm and nano-flakes with lattice fringes can be evidently observed.The nanosphers of MoO3 were prepared by the liquid phase deposition method.The MoO3 of spherical structure owns many particular physical attributes, such as great specific surface area, excellent thermal conductivity and mechanical flexibility which all bring about favorable catalytic performances [12,13].

Catalytic Activity
A repeatability test was run at the scheme E* in Table 2, with a heating rate of 15 °C/min and the carbon black concentration of 10 wt%.The thermogravimetric results for scheme E* are shown in Figure 3.There was a very small fraction of weight loss before 250 °C, normally due to the evaporation of moisture.In view of this, only the data above 400 °C are presented.The carbon black was mainly oxidized at the temperature interval from 450 °C to 600 °C.Still, there is a bit of weight loss above 700 °C, perhaps because of the sublimation of molybdenum [3].A good repeatability in the three curves indicates that the initial quantity of sample, heating rate, gas flow, and furnace atmosphere are all in an appropriate range.
Kinetic parameters resulted from TG and DTG data are shown in Table 3.The values of activation energy for samples with various catalyst fractions lied between 116.3 kJ/mol and 155.3 kJ/mol, and that for the carbon black was 171.8 kJ/mol at 15 °C/min.Pre-exponential factors in a logarithm form were located within 8.6 and 16.0.Both kinetic parameters of the activation energy and the pre-exponential factor were well conformed with those values found in literatures [6,14].
TG and DTG data for the 100 wt% carbon black at various heating rates of 10 °C/min, 15 °C/min and 20 °C/min are illustrated in Figure 4.The carbon black began to lose weight quickly above 500 °C.With the increase in heating rate, a hysteresis phenomenon appeared, perhaps because the oxidation rate of the carbon black at low temperature was greatly influenced by the heat and mass transfer effects the heat transfer of the carbon black from outside to inside and the intermediate overflowing of volatile products from internal to surfaces [6].A high heating rate may result in a great temperature gradient from internal to surface for the carbon black, as well as from the carbon black surface to the furnace chamber, which lowers the weight loss rate of the carbon black.As a re-  * y is the ln((-ln(1-a))/T 2 ) , and x is the 1/T sult, TG, and DTG profiles shift to the high temperature area.Figures 5 and 6 respectively present the tendency of the logarithmic of pre-exponential factor and activation energy with the heating rate.In general, the pre-exponential factor indicates the collision frequency of activated molecules.The greater pre-exponential factor means more effective collisions between molecules.As shown in Figure 5, the pre-exponential factor slightly increases with the growth of heating rate from 10 °C/min to 15 °C/min at the same particulate concentration, but it dramatically increases at the heating rate of 20 °C/min.It may be easily supposed that with the increase in the heating rate, more carbon blacks participate in reacting per unit time and the molecules' collisions are growing.Besides, more carbon blacks were accumulated to be oxidized at high temperature due to the previous lower level of decomposition.Therefore, no apparent inflection point on DTG was perceived owing to insufficient oxidation at a rapid heating rate [14,15].
In Figure 6, the activation energy has little variance with the heating rate at the same particulate concentration.Through a comprehensive consideration for pre-exponential factor and activation energy, the influence of heating rate within 10-15 °C/min on kinetic parameters is extremely weak.It suggests that the reaction conditions hardly reach the heat-transfer limitations.Therefore, the low heating rate is proper for catalytic oxidation in thermogravimetric analyzer [16].
TG and DTG features of different particulate concentrations at the heating rate of 15 °C/min are shown in Figure 7. TG and DTG profiles of the carbon black containing catalysts shift to the low temperature region by a larger step.Further, for samples of the carbon black with MoO3 catalysts, the thermogravimetric features transfer slightly to the low temperature with an obvious increase in catalyst content.Oxidation parameters of the carbon black catalyzed by MoO3 at different particulate concentrations are presented in Table 4.As compared with the neat carbon black, the  Considering the influence of the particulate concentration on kinetic parameters, Figure 8 and Figure 9 respectively show the trend of pre-exponential factor and activation energy changing with the particulate concentration.Both the pre-exponential factors in the logarithm form and activation energy keep in the same low level at the particulate concentrations of 5 wt% and 10 wt%, while that at the particulate concentration of 15 wt% are particularly high, which indicates an apparent increase in the transfer resistance, maybe owing to poor effects of mass transfer at high particulate concentrations.In view of this, the combustion process will be greatly affected by the oxygen diffusion and perhaps some particulates can not contact sufficiently with oxygen.In addition, a very low contact rate between the catalyst and the carbon black resulted from higher particulate concentration may make the heat transfer difficult at rapid heating.Mean-while, pyrolytic products can not spread out in time and non-catalytic oxidation is relatively increased, which consequentially indicates a higher activation energy.Therefore, suiting current oxidation conditions in this study, the appropriate particulate concentration should no more than 10 wt% to reduce the deviation of experiment [6,17,18].
In order to minimize the errors of the kinetic parameters, it is necessary to reduce the resistance of mass transfer as well as diffusion effects.As a result of current study, catalytic oxidation experiments should be performed at lower particulate concentrations between 5 wt% and 10 wt% and lower heating rates between 10 °C/min and 15 °C/min).

Conclusions
Soot oxidation under catalysts is greatly affected by the specified thermogravimetric parameters.The estimated kinetic parameters revealed little influence of the heating rate on the activation energy.But the pre-exponential factor was increased slightly with the growth in the heating rate, especially a large rise at the heating rate of 20 °C/min.Both activation energy and the pre-exponential factor presented a high level resulted from diffusion effects at a high soot-catalyst ratio of 15 wt%.Therefore, for catalytic oxidation studies, TG experiments should be performed at a heating rate lower than 15 °C/min and the particulate concentrations no more than 10 wt% in order to obtain the accurate and reliable values of the pre-exponential factor and the activation energy.
Thereby, based on the reasonability evaluations on the pre-exponential factor and activation energy of each experiment scheme, a strat-  egy for determining the optimized thermogravimetric parameters was built up.This strategy can assist to reduce or avoid the heat transfer resistance and diffusion effect due to inappropriate experimental conditions.Overall, the significance of the results and comprehensive understanding was improved with this strategy for soot oxidation.

Figure 3 .
Figure 3. Sample mass vs. temperature for the Scheme E

Figure 8 .Figure 9 .
Figure 8. Logarithm of pre-exponential factor ln(A) as a function of particulate concentration

Table 2 .
Experimental variables for TGA experiments

Table 3 .
Kinetic parameters for each experimental scheme

Table 4 .
Catalytic oxidation parameters at various particulate concentrations