towards the utilization of wet biomass gasification in supercritical water - on the enerby efficiency and char formation
Riza Yukananto is a PhD student in the Thermal Engineering (TE) research group. His supervisor is prof.dr.ir. G. Brem from the Faculty of Engineering Technology (ET).
Wet biomass, which contains over 80 wt-% of water, is considered to be a cheap type of biomass with an abundant availability. Its utilization is beneficial since it can be converted into biofuels with a wide range of further applications, and can account toward reducing the problem of overflowing biowaste streams. The most used method to process wet biomass is to decompose it via biochemical conversion (i.e. using microorganism) to produce biogas. This biochemical method, however, is hindered by its most important limitation which is the long residence time necessary for complete conversion of the wet biomass. Taking this into consideration, the utilization of fast thermochemical conversion processes becomes very appealing. However, due to the high moisture content, the wet biomass requires a significant amount of energy for drying before it can be converted in conventional thermochemical processes (i.e. gasification, torrefaction, pyrolysis, combustion). This makes it hard to achieve an energy efficient process. Consequently, utilization of a hydrothermal conversion process, such as the Supercritical Water Gasification (SCWG), in which water is used as a reactant is preferred.
The SCWG process is typically done at temperatures above 773 K and pressures of 25 MPa. The SCWG process can be used to produce either hydrogen rich gas or methane rich gas, and its composition can be fine-tuned by altering the temperature of the gasification process to suit the targeted application. During the heating up of wet biomass, however, unwanted side reactions may occur leading to char formation. This char can deposit on the walls of the reactor or heat exchanger causing plugging of the system. One of the solution for char reduction is to develop a fast heating method of the wet biomass by e.g. mixing the relatively cold wet biomass stream with hot supercritical water via direct injection. The aim of this thesis is to gain a fundamental understanding of this process and reveal the key parameters that have a major impact on char reduction in the direct injection system. Additionally, the impact of applying this novel approach on the overall thermal efficiency of the process is also examined.
The investigation on the influence of the direct injection system on the thermal efficiency, is addressed by developing a system model using UniSIM. The UniSIM model is applied for both situations: the direct injection system and the conventional premixed SCWG system, while glycerol is used as feed for both systems. It is found that the implementation of direct injection significantly reduces the efficiency with approximately 8% - 23% in comparison to the premixed system. Subsequently, the feedstock is then replaced with sewage sludge that is modelled by means of a surrogate fuel which incorporates three different compounds (i.e. acetic acid, diketene, propanone and benzene). Accordingly, it is found that a thermal efficiency of approximately 62% can be achieved when using a feedstock with 20 wt-% dry matter content. However, using a feedstock with 8 wt-% dry matter content leads to a thermal efficiency of only 10%. A further sensitivity analysis of several key operating parameters on the system efficiency has been carried out and this resulted in the proposal of an optimum operating window. Accordingly, the optimum reactor temperature is between 843 – 873 K. The ratio of the hot supercritical water flow to the total reactor feed stream is approximately 0.4 - 0.5. Furthermore, the total dry matter content contained in the reactor feed stream should not be more than 14 wt-%, due to expected difficulties in pumping. Based on this operating window, a pinch analysis is conducted for a further optimization of the system. The results show that when the system operates with a ratio of hot supercritical water to the total reactor feed stream of 0.4, the system can achieve a thermal efficiency of 22% and 50%, when the system operates with a total dry matter content of 8 wt-% and 12 wt-%, respectively.
The main objective of the present research is to gain insight in the effect of the biomass heating rate on the char formation. This is addressed by developing a Computational Fluid Dynamic (CFD) model of the SWG process. Three stages in the CFD modelling can be considered: 1) a single-phase model with glycerol feedstock; 2) a single-phase model with glucose feedstock; 3) and a multi-phase model with glucose feedstock. The first CFD model is developed to simulate the gasification of a non-char producing compound, i.e. glycerol. The gasification takes place in a straight tubular reactor with a tee connection near the two inlets (i.e. main tube with water and the feed injection tube). By mixing the relatively cold biomass with hot supercritical water, a fast heating rate of the feed can be obtained. The performance of the selected turbulence model and the expanded Arrhenius formulation to describe the kinetics are assessed via a validation with experimental data from literature. Subsequently, the CFD model is used to examine the flow behavior during the gasification process, showing that in cases where the mass flowrate is low, the gravitational force plays a major role in enhancing the mixing and heat transfer. Additionally, it is shown that this tranquil flow due to the low mass flowrate can lead to of flow recirculation in the tee connection, which may result in a gradual partial heating of the wet biomass that comes from the injection tube.
The second CFD model is developed for glucose gasification in a helical tubular reactor with a tee connection that bridges the two inlets (i.e. main tube and injection tube) together. The aim of this investigation is to examine the influence of several key operating parameters (e.g. the flowrate ratio of cold feed and hot supercritical water stream) on the char formation behavior of glucose. Initially, the CFD model has been developed with the assumption that glucose is directly converted into either gas or char via two different reaction path. Experimental validation of the numerical results showed that the model is capable of giving a good char yield prediction at low temperatures (i.e. 623K). However, at higher temperatures, noticeable discrepancies for the char yield prediction are noticed. Nevertheless, the trend of the char formation as function of temperature is well simulated. The third CFD model includes a Euler-Lagrange formulation and a complex five competing reactions scheme for the formation of gas and char. The implementation of the discrete phase approach is done to mimic the biomass particle behavior (evaporation and devolatilization) during gasification. The model proves to be very accurate in predicting the char yield at high temperatures (up to 693K), and it can also capture the general trend of the gas yield of the gasification process. The subsequent sensitivity analysis shows that implementing the direct injection of cold biomass into a hot supercritical water at a temperature of 723 K reduces the char yield with approximately 27% in comparison to the premixed system. Reducing the flow rate ratio of hot supercritical water and cold biomass from 4:1 to 1:1 (ratio of hot supercritical water to the total reactor feed stream of 0.8 to 0.5) leads to an increase in the char yield by 25%. This is mainly because of the fact that less energy is available to provide a fast heat-up of the cold biomass.
Finally, in an effort to account for realistic dimensions of a SCWG reactor, the model is applied to a reactor with a larger diameter (e.g. 8mm instead of 1 mm). Through this investigation, it is found that enlarging the reactor diameter significantly affects the char formation process. It is shown that increasing the reactor diameter to 8 mm may lead to an increase of the char yield with 107% for a temperature of 723 K. This char formation increase is mainly obtained due to the extended residence time of the biomass in the low temperature region. An optimum mixing of the feed and the preheated water is required to avoid a non-uniform temperature distribution inside the reactor tube is. Therefore an optimal injector geometry that can contribute to a more uniform temperature field and less char formation has been investigated. Three different injector designs are proposed (i.e. 90° wall injection, 45° wall injection and central injection). It is found that injecting the glucose feed in the middle of the reactor via central injection has the best performance in comparison to the other designs. Central injection of glucose leads to a more prominent swirling pattern inside the reactor that enhances the mixing and heat transfer, and reduces the char yield with 25%. Additionally, a further char yield reduction can be achieved by preheating the relatively cold biomass feed to about 473 K.
The data and knowledge regarding the process obtained in this research provide reliable means to predict the gasification performance and therefore may facilitate the further development of this technology. The conversion of wet biomass via gasification in supercritical water can definitely be done reliable, safe and in an energy efficient manner. Further research is recommended to increase the energy efficiency of the SCWG system, to develop a laboratory set-up for detailed validation of the CFD model, and to further improve the CFD model with a discrete char particle model to describe the char or salts deposition on the wall.
