TGA analysis of tobacco rob pyrolysis and release characteristics of noncondensable gas in a fixed-bed reactor

In this study, the size of tobacco rob (TR) particle was considered as a major factor in determining the mass loss in thermogravimetric analysis (TGA) and product yield and composition at different reactor temperatures in the fixed-bed reactor. The TGA results showed that the conversion rate increased and the activation energy (ranged from 53.29 to 58.25 kJ/mol) decreased with a decrease in particle size. The experiments demonstrated that fuel gas yield (from 0.76 to 0.82 Nm³/kg at 900 °C) increased with a decrease in particle size while char and tar yield decreased. Smaller particle sizes resulted in higher H₂ (25.68%) and CO (27.36%) contents. Minimizing the size of raw materials is an alternative method to improve the gas quality of TR pyrolysis. The increase of gas yield was attributed to the decomposition of char and tar vapor as temperature increased.     

Optimization of pretreatment conditions for enhanced sugar release

Present investigation was done to evaluate various algal genera found in water bodies of Varanasi city. The potential of any biomass for biofuels (bioalcohols, biohydrogen, etc.) production depends on the quantity of extractable sugar present in it. Acid (H₂SO₄) and alkali (NaOH) pretreatment were performed, and H₂SO₄ was chosen due to its nearly double yield as compared with alkaline pretreatment. Response surface methodology was utilized for the optimization of operating parameters such as treatment temperature, time, and acid concentration. Sugar yield up to 0.33 g/g of dry biomass was obtained using cyanobacterial biomass of Lyngbya limnetica, at 100°C, 59.19 min, and H₂SO₄ concentration of 1.63 M.     

The convective hot air drying of Lactuca sativa slices

The storage of fresh agricultural products is not easy because of its high moisture. Dehydration is an efficient preservation method. The investigation of drying modeling and transfer characteristics are important for selecting operating conditions and equipment design. The drying behavior of Lactuca sativa slices, with the thickness of 2 mm, was investigated at 60.0–80.0°C and 0.60–1.04 m sec^?1 velocity in a convective hot air drier. The mass transfer during the drying process was described using six thin drying models. The convective heat transfer coefficient α and mass transfer coefficient k_H were finally calculated. The results showed that the drying process could be separated into three stages including accelerating rate, constant rate, and falling rate period, which was influenced by hot air temperature and velocity, and the Modi?ed Page model agreed well with the experimental data. When the operating temperature was increased from 60.0°C to 80.0°C, α was found increased from 88.07 to 107.93 W·m^?2·K^?1, and k_H increased from 46.32 × 10^–3 to 68.04 × 10^–3 kg·m^?2·sec^?1·ΔH^?1. With the increase of air velocity from 0.60 to 1.04 m·sec^?1, α was increased from 78.85 to 101.35 W·m^?2·K^?1, and k_H was enhanced from 51.78 × 10^–3 to 65.85 × 10^–3 kg·m^?2·sec^?1·ΔH^?1.     

Enzymatic hydrolysis of banana stems (Musa acuminata): Optimization of process parameters and inhibition characterization

In current work, an optimum solid loading (solid: liquid = 1:20), pH (4.8), temperature (50°C), and enzyme dosing of 20 filter paper unit (amount of enzyme required to release 1 µmol of glucose as reducing sugar from filter paper in per mL per minute) were enumerated for enzymatic hydrolysis of banana stem using cellulase from Trichoderma reesei. Further, inhibition study on enzymatic hydrolysis of banana stem was investigated by the supplementation of monosaccharides (glucose, galactose, mannose, xylose, and arabinose), disaccharide (cellobiose), and inhibitors (acetic acid and furfural obtained from pre-enzymatic hydrolysis steps). Glucose and cellobiose showed inhibitory effect on enzymatic hydrolysis of pretreated banana stem at or above 8 g/L while galactose, mannose, and xylose showed a significant inhibitory effect at or above 4 g/L. Instead of inhibition, arabinose enhanced the enzymatic hydrolysis with increase in total reducing sugars. Acetic acid did not show any significant inhibition while furfural inhibited the system at a comparative low concentration of 2 g/L. Further, scanning electron microscopy analysis was performed to investigate the difference in ultra-structural morphology of raw biomass, pretreated biomass, and biomass obtained after enzymatic hydrolysis.     

Simulation and validation of a suitable model for thin layer drying of ginger rhizomes in an induced draft dryer

Variation in drying material and their biological differences, coupled with heat supply method in different dryers, makes mathematical modeling of drying complicated. Attempt was made to simulate a drying process and to identify best suitable model out of six selected drying models, for drying of ginger slices in a solar-biomass integrated drying system designed and developed for spice drying. Moisture content data were converted into the moisture ratio (MR) expressions and curve fitting with drying time for the selected drying models was analyzed. Sigma Plot software was used for nonlinear regression to the data obtained during drying and for modeling of drying curves. The suitability of the models was evaluated in terms of statistical parameters such as coefficient of determination (R²), mean percentage error (P), and standard error estimate. Drying air temperature was in the range of 47–55°C and air velocity was between 1.0 and 1.3 m s^?1. Ginger slices were dried from 88.13% to 7.65 ± 0.65% (wb) in 16 h. Trays were interchanged in a predetermined matrix sequence from 4 h onwards when moisture content was reduced to 60–70% (wb), for uniformity in drying. Highest value of R² (0.997), lowest value of SEE (0.020), and P value < 0.0001 established Page model as the best suitable model for the developed drying system. The predicted MRs were in good agreement with the experimental values and the effective moisture diffusivity for ginger was found to be 2.97 × 10^–7 m² s^?1.     

An effective approach to improve ethanol productivity by simultaneous saccharification and fermentation of rice straw

Fuel ethanol was produced using rice straw with the simultaneous saccharification and fermentation (SSF) method. The influence of cellulose liquefaction pretreatment and Fe²⁺ quantity on ethanol productivity was investigated in detail. At the same time, the optimized conditions including fermentation temperature, Fe²⁺ amount, yeast inoculation quantity, and the inoculated cellulose enzyme dosage in the SSF process were systematically investigated by analyzing fuel ethanol yield. The result indicated that fuel ethanol yield was 0.319 g per gram rice straw by SSF approach when appropriate amount of Fe²⁺ was added into the reaction system. The optimal technology parameters were: fermenting temperature of 36°C, Fe²⁺ amount of 4 mg · g^?1, inoculating proportion of 20%, cellulose enzyme of 20 IU · g^?1, and Pachysolen tannophilu/saecharomyces cerevisiae of 1:2 ratio. The ethanol yield under the best conditions was larger than that of the control group. We hope that this research can facilitate to achieve large-scale comprehensive utilization for rice straw.     

An Application of Remote Sensing Data in Mapping Landscape-Level Forest Biomass for Monitoring the Effectiveness of Forest Policies in Northeastern China

Monitoring the dynamics of forest biomass at various spatial scales is important for better understanding the terrestrial carbon cycle as well as improving the effectiveness of forest policies and forest management activities. In this article, field data and Landsat image data acquired in 1999 and 2007 were utilized to quantify spatiotemporal changes of forest biomass for Dongsheng Forestry Farm in Changbai Mountain region of northeastern China. We found that Landsat TM band 4 and Difference Vegetation Index with a 3 × 3 window size were the best predictors associated with forest biomass estimations in the study area. The inverse regression model with Landsat TM band 4 predictor was found to be the best model. The total forest biomass in the study area decreased slightly from 2.77 × 10⁶ Mg in 1999 to 2.73 × 10⁶ Mg in 2007, which agreed closely with field-based model estimates. The area of forested land increased from 17.9 × 10³ ha in 1999 to 18.1 × 10³ ha in 2007. The stabilization of forest biomass and the slight increase of forested land occurred in the period following implementations of national forest policies in China in 1999. The pattern of changes in both forest biomass and biomass density was altered due to different management regimes adopted in light of those policies. This study reveals the usefulness of the remote sensing-based approach for detecting and monitoring quantitative changes in forest biomass at a landscape scale.     

Experimental study on microalgae cultivation in novel photobioreactor of concentric double tubes with aeration pores along tube length direction

Microalgae have been identified as a superior feedstock for biodiesel production and varied tubular photobioreactors are developed for high efficient and scale-up microalgae cultivation. This article presented a novel concentric double tubes using aeration through radial pores along the length direction of inner tube. Experiments on microalgae cultivation were carried out in the novel photobioreactor, and two control groups including concentric double tubes with axial aeration at both ends and common tubular. The biomass productivity of novel photobioreactor increased by 43.6% and 107.4%, respectively, compared with concentric double tubes with axial aeration at both ends and common tubular without aeration. The values of pH shifted from 7.5 to 9.0 for novel photobioreactor, but 7.5 to 8.8 for common tubular, and 7.5 to 9.6 for concentric double tubes with axial aeration. The dissolved oxygen concentration fluctuated between 6.0 and 7.0 mg·L^?1 for novel photobioreactor, but rose from 6.6 to 10.2 mg·L^?1 for the common tubular, and 6.9 to 8.1 mg·L^?1 for the concentric with axial aeration. Results show that the aeration style of novel photobioreactor can make efficient local mixing and maintain smaller range of pH and lower level of dissolved oxygen in case of higher biomass concentration. Moreover, compared with the two control groups, the novel concentric double tubes have advantages on the light/dark cycle frequency, which may be benefit for microalgae cultivation. The novel concentric double tubes presented in this work can give some inspiration for high efficiency microalgae cultivation.     

Application of lean flammability limit study and large eddy simulation to burner development for an oxy-fuel combustion system

Previously, we developed a model to predict lean flammability limit L and flame propagation velocity Sb for pulverized coal. In the present paper, we have extended the model to apply it in development of oxy-fuel combustion systems. The basic model consists of two particles. One of the two particles burns first, then, the other particle is ignited by the heat of combustion of the one burning particle. We analyzed at what distance the first burning particle could ignite the next particle, and how fast the first burning particle could ignite the next particle. The model was verified both for air and oxy-fuel combustion conditions. Next, a method to support burner development was examined by using the model. Local Sb and L near the ignition points of the burner could be analyzed from the concentration and temperature profiles of CFD results. Flame stability was judged by the calculated Sb and L profiles, and past results of blow-off limits obtained with actual- and pilot-scale experiments. A DS^®T-burner was developed by Hitachi Power Europe, and installed at the Schwarze Pumpe pilot plant. Combination of the technique and large eddy simulation was applied to confirmation of the system.     

Relationship between liquid depth and the acoustic generation of hydrogen: design aspect for large cavitational reactors with special focus on the role of the wave attenuation

Understanding the effect of the liquid depth (z) on the acoustic generation of hydrogen is highly required for designing large-scale sonoreactors for hydrogen production because acoustic cavitation is the central event that initiates sonochemical reactions. In this paper, we present a computational analysis of the liquid-depth effect on the generation of H₂ from a reactive acoustic bubble trapped in water irradiated with an attenuating sinusoidal ultrasound wave. The computations were made for different operating conditions of frequency (355–1000 kHz), acoustic intensity (1–5 W/cm²), and liquid temperature (10–30°C). The contribution of the acoustic wave attenuation on the overall effect of depth was appreciated for the different conditions. It was found that the acoustic generation of hydrogen diminished hardly with increasing depth up to z = 8 m, and the depth effect was strongly operating parameter-dependent. The sound wave attenuation played a crucial role in quenching H₂ yield, particularly at higher z. The reduction of the H₂ yield with depth was more pronounced at higher frequency (1000 kHz) and lower temperature (10°C) and acoustic intensity (1 W/cm²). The attenuation of the sound wave may contribute up to 100% in the overall reductive effect of depth toward H₂ production rate. This parameter could be imperatively included when studying all aspects of underwater acoustic cavitation.     

Application of response surface methodology for the optimization of biodiesel production from yellow mustard (Sinapis alba L.) seed oil

In the present study, response surface methodology (RSM) involving central composite design (CCD) was applied to optimize the reaction parameters of biodiesel production from yellow mustard (Sinapis alba L.) seed oil during the single-step transesterification process. A total of 30 experiments were designed and performed to determine under the effects of variables on the biodiesel yield such as methanol to oil molar ratio (2:1–10:1), catalyst concentration (0.2–1.0 wt.% NaOH), reaction temperature (50–70°C), and reaction time (30–90 min). The second order polynomial model was used to predict the biodiesel yield and coefficient of determination (R²) was found to be at 0.9818. The optimum biodiesel yield was calculated as 96.695% from the model with the following reaction conditions: 7.41:1 of methanol to oil molar ratio, 0.63 wt. % NaOH of catalyst concentration, 61.84°C of reaction temperature, and 62.12 min of reaction time. It is seen that the regression model results were in agreement with the experimental data. The results showed that RSM is a suitable statistical technique for optimizing the reaction parameters in the transesterification process in order to maximize the biodiesel yield.     

Reforming of CO2-containing natural gas with steam in AC gliding arc discharge for hydrogen production

In this study, a low-temperature gliding arc discharge system was used to investigate the effects of steam content and operational parameters, i.e. total feed flow rate, applied voltage, and input frequency, on the reforming performance of CO₂-containing natural gas. The addition of steam less than 10% was found to enhance both selectivities for hydrogen and carbon monoxide with a significant reduction of power consumptions. The highest hydrogen selectivity and yield with the lowest power consumptions were achieved at a steam content of 10 mol%, a total feed flow rate of 100 cm³/min (corresponding to a very short residence time of 1.37 s), an applied voltage of 13.5 kV and an input frequency of 300 Hz. Under these optimum conditions, the oxidative dehydrogenation reactions to produce hydrogen were dominant with the minimum activities of all undesired reactions of cracking and coupling. Both of the short residence time (1.37 s) with a low bulk temperature (lower than 200°C) in the plasma zone under atmospheric pressure, indicate that the studied plasma reactor (gliding arc system) is superior to conventional catalytic reactors for reforming of natural gas.     

Adequacy of Visually Classified Particle Count Statistics From Regional Stream Habitat Surveys1

Abstract: Streamlined sampling procedures must be used to achieve a sufficient sample size with limited resources in studies undertaken to evaluate habitat status and potential management‐related habitat degradation at a regional scale. At the same time, these sampling procedures must achieve sufficient precision to answer science and policy‐relevant questions with an acceptable and statistically quantifiable level of uncertainty. In this paper, we examine precision and sources of error in streambed substrate characterization using data from the Environmental Monitoring and Assessment Program (EMAP) of the U.S. Environmental Protection Agency, which uses a modified “pebble count” method in which particle sizes are visually estimated rather than measured. While the coarse (2?) size classes used in EMAP have little effect on the precision of estimated geometric mean (D_gm) or median (D₅₀) particle diameter, variable classification bias among observers can contribute as much as 0.3?, or about 15‐20%, to the root‐mean‐square error (RMSE) of D_gm or D₅₀ estimates. D_gm and D₅₀ estimates based on EMAP data are nearly equal when fine sediments (<2 mm) are excluded, but otherwise can differ by up to a factor of 2 or more, with D_gm < D₅₀ for gravel‐bed streams. The RMSE of reach‐scale particle size estimates based on visually classified particle count data from EMAP surveys, including variability associated with reoccupying unmarked sample reaches during revisits, is up to five to seven times higher than that reported for traditional measured pebble counts by multiple observers at a plot scale. Nonetheless, a variance partitioning analysis shows that the ratio of among site to revisit variance for several EMAP substrate metrics exceeds 8 for many potential regions of interest, suggesting that the data have adequate precision to be useful in regional assessments of channel morphology, habitat quality, or ecological condition.     

Impact assessment of a hydroelectric project on the flora in the Western Himalayan region based on vegetation analysis and socio-economic studies

This study provides an overview of the impacts of a proposed hydroelectric power project in the Western Himalayan region in India, using a primary database on floristic diversity and vegetation analysis. The remote sensing data revealed that in the submergence zone only mixed deciduous forest that occupies 807.5 ha area and has a wood biomass volume of 4,027,503 m³ is likely to be lost due to impoundment. A total of 165 plant species found in the submergence zone also occur in the influence and free draining catchment area of the project. In the influence zone of the project area only one tree species (Acer oblongum) is found under conservation threat category, which is also present in the free draining catchment of the project. The project affected population (6716 people) residing in the submergence and influence zone depend upon the surrounding forests for fuel wood, fodder, wild edibles etc., and most likely they will settle in the nearby areas, thus mounting more pressure on residual forests of the influence zone for various forest products. Further, from the vegetation analysis it is evident that several tree species (e.g. Lannea coromandelica, Terminalia alata, T. bellerica etc.), may face more pressure from exploitation as they provide a number of useful products and are represented in lower numbers in the forests of the project area. To compensate for the loss of various goods and services provided by the forests falling in the submergence zone and to offset the increased pressure of the project affected families on the forests of influence zone, a biodiversity management plan is suggested incorporating socio-economic considerations.     

Assessment of biomass and lipid productivity and biodiesel quality of an indigenous microalga Chlorella sorokiniana MIC-G5

Generation of biodiesel from microalgae has been extensively investigated; however, its quality is often not suitable for use as fuel. Our investigation involved the evaluation of biodiesel quality using a native isolate Chlorella sorokiniana MIC-G5, as specified by American Society for Testing and Materials (ASTM), after transesterification of lipids with methanol, in the presence of sodium methoxide. Total quantity of lipids extracted from dry biomass, of approximately 410–450 mg g^?1 was characterized using FTIR and ¹H NMR. After transesterification, the total saturated and unsaturated fatty acid methyl esters (FAMEs) were 43% and 57%, respectively. The major FAMEs present in the biodiesel were methyl palmitate (C16:0), methyl oleate (C18:1), and methyl linoleate (C18:2), and the ¹H NMR spectra matched with criteria prescribed for high-quality biodiesel. The biodiesel exhibited a density of 0.873 g cm^–3, viscosity of 3.418 mm² s^?1, cetane number (CN) of 57.85, high heating value (HHV) of 40.25, iodine value of 71.823 g I₂ 100 g^?1, degree of unsaturation (DU) of 58%, and a cold filter plugging point (CFPP) of –5.22°C. Critical fuel parameters, including oxidation stability, CN, HHV, iodine value, flash point, cloud point, pour point, density, and viscosity were in accordance with the methyl ester composition and structural configuration. Hence, C. sorokiniana can be a promising feedstock for biodiesel generation.     

Modeling of Vertical Solute Dispersion in a Sediment Bed Enhanced by Wave‐Induced Interstitial Flow1

Abstract: Mass (solute) transport in a stream or lake sediment bed has a significant effect on chemical mass balances and microbial activities in the sediment. A “1D vertical dispersion model” is a useful tool to analyze or model solute transfer between river or lake water and a sediment bed. Under a motionless water column, solute transfer into and within the sediment bed is by molecular diffusion. However, surface waves or bed forms create periodic pressure waves along the sediment/water interface, which in turn induce flows in the pores of the sediment bed. The enhancement of solute transport by these interstitial periodic flows in the pores has been incorporated in a 1D depth‐dependent “enhanced dispersion coefficient (D_E).” Typically, D_E diminishes exponentially with depth in the sediment bed. Relationships have been developed to estimate D_E as a function of the characteristics of sediment (particle size, hydraulic conductivity, and porosity) and pressure waves (wave length and height). In this paper, we outline and illustrate the calculation of D_E as well as the penetration depth (d_p) of the flow effect. Sample applications to illustrate the computational procedure are provided for dissolved oxygen transfer into a stream gravel bed and release of phosphorus from a lake bed. The sensitivity of the results to input parameter values is illustrated, and compared with the errors obtained when interstitial flow is ignored. Maximum values of D_E near the sediment surface can be on the order of 1 cm²/s in a stream gravel bed with standing waves, and 0.001 cm²/s in a fine sand lake bed under progressive surface waves, much larger than molecular diffusion coefficients.     

Modeling of the denitrification/anaerobic digestion process of salmon fishery wastewater in a biofilm tubular reactor

The literature has paid scarce attention to the modeling of the denitrification-anaerobic digestion process in packed bed biofilm tubular reactors used to treat wastewater. The present study obtained a steady-state model for industrial salmon fishery wastewater treatment in a biofilm tubular reactor, including pH as a variable and the effect of biomass on hydrolysis. The axial profile of the reactor components and process efficiency were predicted with deviations below 6%. The optimal operating zone for the process was found at hydraulic retention time (HRT)>1.5d and inlet protein concentration (S(prot,0))<3000 mgTOCL(-1). Based on our results, we concluded that the removal of organic matter and nitrogen compounds depended mainly on HRT. The effluent pH was mainly affected by the C/N ratio, where a decrease increases pH. Organic matter removal was related with the anaerobic digestion process, while denitrification influenced mostly nitrate and nitrite removal.     

Determination of dissolved oxygen limitation in aerobic biofilm reactors

The concentration of dissolved oxygen (DO) strongly influences the performance of aerobic biofilm reactors because organic oxidation is limited by the availability of oxygen. However, it is not necessary to maintain a high DO level in the reactors in order to overcome this limitation. Excessive aeration wastes energy. Therefore, the determination of the onset of DO limitation against organic substrate removal in aerobic biofilm reactors is important for their effective operation. This study is aimed at developing an expression to determine the onset of DO limitation and hence to control the aeration system. The expression developed is as follows: , where S_b and C_b are the bulk concentrations of organic substrate and DO, respectively; D_ws and D_wc are the diffusion coefficients of organic substrate and oxygen in the reactors respectively; and R_b is an overall ratio of oxygen consumption to organic substrate removal in the reactors. The latter is the key parameter in the equation, and is determined by the characteristics of the substrate, biofilm, and reactor. In order to measure the value of R_b, the authors have developed a micro-biofilm reactor. The value of R_b was determined to be 0.13 (mg O₂ mg⁻¹ COD_cr) for glucose removal with this reactor. The equation has, subsequently, been verified with data from batch and continuous experiments.     

Experimental validation of in situ CO2 capture with CaO during the low temperature combustion of biomass in a fluidized bed reactor

A novel concept for capturing CO₂ from biomass combustion using CaO as an active solid sorbent of CO₂ is discussed and experimentally tested. According to the CaO/CaCO₃ equilibrium, if a fuel could be burned at a sufficiently low temperature (below 700 °C) it would be possible to capture CO₂ “in situ” with the CaO particles at atmospheric pressure. A subsequent step involving the regeneration of CaCO₃ in a calciner operating at typical conditions of oxyfired-circulating fluidized combustion would deliver the CO₂ ready for purification, compression and permanent geological storage. Several series of experiments to prove this concept have been conducted in a 30 kW interconnected fluidized bed test facility at INCAR-CSIC, made up of two interconnected circulating fluidized bed reactors, one acting as biomass combustor-carbonator and the other as air-fired calciner (which is considered to yield similar sorbent properties than those of an oxyfired calciner). CO₂ capture efficiencies in dynamic tests in the combustor-carbonator reactor were measured over a wide range of operating conditions, including different superficial gas velocities, solids circulation rates, excess air above stoichiometric, and biomass type (olive pits, saw dust and pellets). Biomass combustion in air is effective at temperatures even below the 700 °C, necessary for the effective capture of CO₂ by carbonation of CaO. Overall CO₂ capture efficiencies in the combustor-carbonator higher than 70% can be achieved with sufficiently high solids circulation rates of CaO and solids inventories. The application of a simple reactor model for the combined combustion and CO₂ capture reactions allows an efficiency factor to be obtained from the dynamic experimental test that could be valuable for scaling up purposes.     

Effect of feeding strategy and COD/sulfate ratio on the removal of sulfate in an AnSBBR with recirculation of the liquid phase

The objective of this work was to analyze the effect of the interaction between feeding strategy and COD/sulfate ratio on the removal efficiency of sulfate and organic matter from a synthetic wastewater. An anaerobic sequencing batch reactor with recirculation of the liquid phase and containing immobilized biomass on polyurethane foam (AnSBBR) was used. The AnSBBR with a total volume of 3.7 L, treated 2.0 L synthetic wastewater in 8-h cycles at 30 ± 1 °C and was inoculated with anaerobic biomass from a UASB. Two feeding strategies were assessed: (a) batch and (b) batch followed by fed-batch. In strategy (a) the reactor was fed in 10 min with 2 L wastewater containing sulfate and carbon sources. In strategy (b) 1.2 L wastewater (containing only the sulfate source) was fed during the first 10 min of the cycle and the remaining 0.8 L (containing only the carbon source) in 240 min. The COD/sulfate ratios assessed were 1 and 3. Based on these values and on the concentrations of organic matter (0.5–11.25 gCOD/L) and sulfate (0.5 and 2.5 gSO₄^2?/L), the sulfate and organic matter loading rates applied equaled 1.5 and 4.5 gSO₄^2?/L d for sulfate and 1.5, 4.5 and 13.5 gCOD/L d for organic matter. After stabilization of the system time profiles were run of monitored parameters (COD, sulfate, sulfide and sulfite). In general, the reactor showed to be robust for use in the anaerobic treatment of wastewaters containing sulfate. Gradual feeding (strategy b) of the carbon source favored sulfate reduction, resulting in sulfate removal efficiencies of 84–98% and organic matter removal efficiencies of 48–95%. The best results were observed under COD/sulfate ratio equal to 1 (loading rates of 1.5 and 4.5 gSO₄^2?/L d for sulfate, and 1.5 and 4.5 gCOD/L d for organic matter). When COD/sulfate ratio was 3 (loading rates of 1.5 and 4.5 gSO₄^2?/L d for sulfate, and 4.5 and 13.5 gCOD/L d for organic matter) the effect of feed mode became less significant. These results show that the strategy batch followed by fed-batch is more advantageous for COD/sulfate ratios near the stoichiometric value (0.67) and higher organic matter and sulfate concentrations.