Optimization of Carbon Dioxide Reduction in Biohythane Using an Innovative Water Scrubber

The major problem of biohythane production from palm oil mill effluent is the high carbon dioxide (CO2) content. In this study, an innovative water scrubber system for upgrading biohythane has been experimentally investigated. Biohythane composing of ∼53.34% of CH4, ∼39.12% of CO2, and ∼7.54% of H2 was simulated regarding the composition of biohythane in the lab scale. Response surface methodology (RSM); a 5-level, 3-factor, central composite design (CCD), was employed to optimize three important parameters (biohythane flow-rate, water flow-rate, and operating time) in order to minimize the CO2 content in the biohythane production. As a result, CO2 concentration decreased with the increase of both the operating time and water flow-rate but inversely proportion to the biohythane flow-rate, which led to higher CO2 absorption by water. The optimal condition regarding the maximum value of CO2 reduction was found at: 3 Nl/min of biohythane flow-rate, 16 Nl/min of water flow-rate, and 9 min of operating time, thereby yielding 77.6% of CO2 reduction.


Introduction
Currently, an interest in the eco-friendly energy from renewable sources seems to be one of the most important issues for any country because fossil fuels are being depleted. Biogas is one of the renewable energy sources used as a fuel to generate heat such as for cooking, for space heating, and for generating electricity. The biogas can be produced from renewable sources, e.g. cow dung, other animal waste, and palm oil mill effluent.
In palm oil milling industries, solid biomasses are used as materials to supply heat (steam) and electricity for milling processes. A study in 2017 revealed that Thailand had a total of 66 palm oil mills, of which 30 mills installed biogas plants. The total gas production was estimated at 315,000 m 3 /day. Recent development in palm oil mill effluent (POME) biogas production was produced via a 2-stage digestion process in which hydrogen and methane gases are separately generated in the first and second stages, respectively. Thus, this is a good opportunity for using biohythane as power energy production. However, the raw composition of biohythane consists of approximately 35-38% of carbon , and the rests are 53-55% of methane (CH4), and 5-10% of hydrogen (H2) [1]. Thus, it is very necessary to remove CO2 contained in the biohythane in order to upgrade its quality. Jeong et al [3] reported that the biogas 9th TSME-International Conference on Mechanical Engineering (TSME-ICoME 2018) IOP Conf. Series: Materials Science and Engineering 501 (2019) 012003 IOP Publishing doi:10.1088/1757-899X/501/1/012003 2 power plant is characterized as relatively low efficiency, especially for internal combustion engine because the low flame speed of CH4 was blamed for adverse effects [4]. Fortunately, hydrogen was used as compliment additive to improve the combustion of biogas and the mixture was named hythane [5]. For these reasons, the upgrading of biohythane is ultimately realized.
In order to enhance both CH4 and H2 of the biohythane, a water scrubbing absorption method was employed for CO2 removal. According to this conventional concept, raw gases were fed into a packedbed column from the bottom into the water scrubber tank. Then, pressurized water was sprayed from the top. The absorption is thus counter-current flow. The advantage of water scrubbing is that it is a simple technology [6,7]. However, the disadvantage is that the water contaminated by CO2 is needed to be regenerated. In this study, an innovative water scrubbing concept was proposed. Both the pressurized raw gases and water were mixed before entering through the venturi device, which was installed at the top of the water scrubber tank, and they subsequently flowed into the tumbler area in side the tank in the downward direction, thereby absorbing CO2 contained in the biohythane.
The objective of this present work is to study the optimization of three parameters; biohythane flowrate, water flow-rate and operating time in order to reduce the CO2 concentration that contained in the biohythane. Response surface methodology (RSM) with a 5-level and 3 factor central composite design (CCD) was employed to optimize these three parameters [8]. The goal is to prognosticate the proper response surface models of the relationship according to the reduction of CO2 and these parameters.

Materials
Biohythane was formed by the mixing of CH4, H2, and CO2 via gas mixture. Impurities contained in the real biohythane, such as H2S and ammonia (NH3), were ignored due to its negligible amount. CH4 was obtained from natural gas (CNG) for vehicles from the gas stations. Therefore, it was anticipated that the presence of some constituents such as propane and butane in the mixture would have slightly increased the heating value than the intended biohythane. Both the commercial H2 (95% purity) and CO2 (99.99% purity) were purchased form Linde, while the CNG used in this experiment was obtained from the gas station in Tasae District, Chumphon province, Thailand. Certificate of CNG compositions, listed in Table 1, was provided by the gas separation plant, and its composition mainly composed of 73.8 % of CH4 and 17.1% CO2. This data was also utilized in the calculations of simulated biohythane compositions.  Figure 1. It mainly consists of the fuel supply system (H2, CO2, and CH4(CNG)), water scrubber tank and water system. The key part of biohythane upgrading is the water scrubber tank. It composes of a vertical cylindrical tank (22 cm diameter) with two layers inside (total height: 29 cm and total volume:11 L) according to Figure 2.
The water scrubber tank is made of a metallic alloy (~2 mm wall thickness). It is pierced at the bottom to allow the passage of an outlet tube from which the water with CO2 is drained after experiments. One pressure gage is positioned at the top of the tank for measuring the pressure inside the water scrubber tank, while the gas outlet port is placed at the top of the cylindrical part. H2, CO2, and CH4 (CNG) were mixed via gas mixture under pressure (1.8 bar) to be simulated as biohythane. The simulated biohythane was then fed into the scrubber tank along with water. To control gas flow-rate, H2, CO2, and CH4 mass flow controllers (Kofloc model 3660 ±1% full scale) were installed between the gases supply and water scrubber tank to control the composition of simulated biohythane. To complete mixing, a receiver tank was installed after a gas mixing unit. For safety concern, an anti-backfire valve was mounted at the H2 fuel pipe line. An ultrasonic flow meter was used to measure the water flow-rate, which the reading was confirmed by the Rotameter. The operating temperature was controlled at ~20-25 °C (ambient temperature) throughout experiment. The mixing of biohythane and water was sprayed to the tumbler inside the scrubber tank (Figure 2b), thus producing bubbles and thereby absorbing CO2 by water. The CO2-rich water was drained through the outlet port after experiments. The experiments were carried out according to the experimental design matrix ( 3). After experiments, the sampling biohythane was collected by gas sampling bags and then analysed by gas chromatography (model: GC-2014, Shimadzu).

Experimental design.
Response surface methodology (RSM), with a 5-level and 3-factor central composite design (CCD), was employed in order to optimize the reduction of CO2 in biothythane. Biohythane flow-rate (̇), water flow-rate (̇), and time (t) are three independent variables that were used to find the optimization of the reduction of CO2 in the biohythane after scrubbing process. Regarding the three independent variables, the axial parameter (αx) is 1.68 (for rotatable CCD), which was calculated by equation (1)  (1) where αx is the axial parameter for rotatability, and k is the number of variables. The experiments were considered for 5 levels of the independent variable identified as -1.68, -1, 0, 1, and 1.68 according to Table 2.  Table 3 summarizes the experimental design matrix generated by Multiple regression software according to the range of independent variables in Table 2 and experimental results for 17 experimental runs in a water scrubber. Experiments were carried to determine the reduction of CO2 in the biohythane. The composition of simulated biohythane, by volume, was kept constant at 7.3% of H2, 53.7% of CH4, and 39.0% of CO2. After completing experiments, it was reported that the reduction of CO2 was in the range of 15.6-60.7 % by volume. Noted that no amount of the water contained in the product gases after scrubbing was detected.

Response surface models of results and statistical analyses
After considering the results of the effect of three independent variables regarding biohythane flow-rate, water flow-rate, and operating time on the reduction of CO2, response surface models were applied to analyse the data in Table 3 by utilizing a multiple regression model to fit a second-order polynomial equation according to equation (2).
The equation (2) shows the relationship between the CO2 reduction values and the other independent variables in the form of squared model.
Where CR is the reduction of CO2 (%), ̇ is biohythane flow-rate, ̇ is water flow-rate, t is time, and is the coefficient value.
The model was considered by the probability of error value (p-value) in the detailed Table 4. The pvalue was employed to exam the statistical significance of each regression coefficient. When their pvalues are higher than 0.1, at 90% confidence level, the independent variables in a model were considered as negligible effects. In contrast, when their p-values are lower than 0.1, the independent variables were considered as significant parameters.
After considering the p-values, the negligible parameters in equation (2) were suppressed, thereby resulting in a new second-order polynomial equation according to equation (3). = 0 + 1 (̇) + 2 (̇) + 3 (̇2) + 4 ( 2 ) + 5 (̇3) + 6 ( 3 ) (3) Table 4 shows coefficient values, p-values, R 2 , and R 2 adjusted of equation (3) after suppressing the negligible parameters. According to this table, the operating time became to be the most significant term in the response models because of its p-value being lowest (0.04682). Therefore, it is very necessary to keep this parameter for calculating the reduction of CO2 in biohythane by water scrubbing. Note that 1 and 5 were kept in the equation (3), although their P-values were higher than 0.1 because their coefficient values are important to be considered for the calculation in the reduction of CO2.  Figure 3 shows contour plots of the influence of biohythane flow-rate, water flow-rate, and operating time on the reduction of CO2. It was found that an increase in the operating time with the reduction of biohythane flow-rate has a positive effect on CO2 reduction (Figure 3a). The percentage of CO2 reduction reached ~60% at the operating time range 7-9 min while injecting biohythane in the range of 3-4 Nl/min. This implies that increasing operating time with decreasing biohythane flow-rate showed a positive impact on CO2 reduction. As Figure 3b illustrated, the CO2 reduction increased significantly when increasing water flow-rate, e.g. from 20% at ~9 Nl/min to 50% at 15 Nl/min, while injecting biohythane in the range of 5-11 Nl/min. This can be explained by the fact that higher water flow-rate leads to higher CO2 absorption by water. Moreover, increasing water flow-rate while increasing operating time in the range of 4-8 min favours the reduction of CO2 (40-70%), according to Figure 3c. For these variations, it can be summarised that increasing both water flow-rate and operating time while reducing the biohythane flow-rate inherently promoted the reduction of CO2.