Co-production of hydrogen and methane from herbal medicine wastewater by a combined UASB system with immobilized sludge (H₂ production) and UASB system with suspended sludge (CH₄ production)

Hydrogen production by biological methods has been the focus of research due to advantages such as less energy intensive and high efficiency of conversion to usable power. It has the potential to eliminate environmental deterioration that is derived from the utilization of conventional fossil fuels (Ren et al. 2010). Recently, many studies have been carried out for hydrogen production from various wastewaters, including food waste (Wang & Zhao 2009), cheese whey (Nikolaos et al. 2009), red canary grass (Lakaniemi et al. 2011) and molasses wastewater (Han et al. 2012). A large amount of herbal medicine wastewater (HMWW) characterized by high chemical oxygen demand (COD) and biological oxygen demand (BOD) can be generated in the developing countries. HMWW includes mainly sugar, anthraquinone, alkaloids, proteins, pigment, lignin and their hydrolysis products (Nandy & Kaul 2001). In traditional Chinese medicine factories, HMWW cannot be directly discharged into surface water bodies on account of its high toxicity and the difficulty of degradation. Generally, the anaerobic fermentation technology has been developed for the treatment of various high concentrated organic wastewaters. By applying suitable conditions (pH, hydraulic retention time (HRT), nutritive ratio and so on), HMWW can also be biological fermented for hydrogen production and achieve dual interests in the direction of wastewater treatment and energy recovery (Venkata et al. 2008). Sivaramakrishna et al. (2014) evaluated the hydrogen production from herbal wastewater by an enriched mixed slaughterhouse sludge. However, to our best knowledge, there is no paper on the co-production of hydrogen and methane using the HMWW as sole substrate via anaerobic fermentation.

The objective of this study was to develop an upflow anaerobic sludge bed (UASB) system with immobilized sludge of HMWW for hydrogen production at different organic loading rates (OLRs). For comparison, an identical UASB system with suspended sludge was also employed. The effluents of hydrogen fermentation were used for further continuous methane production at various HRTs, and the total energy recovery rate of hydrogen and methane production from HMWW was evaluated.

In this study, HMWW used as experimental substrate for sequential hydrogen production was collected from a local Chinese medicine factory (Harbin, China). As shown in Table 1, the concentration ratio of BOD/COD was just about 0.19, which suggested that HMWW was poor in biodegradability. The biodegradability of HMWW pumped into the immobilized sludge UASB reactor for hydrogen fermentation could be effectively improved.

The raw sludge was obtained from a sludge dewatering engine room in a local municipal wastewater treatment plant (Harbin, China) and screened to eliminate large particulate materials (≥1.5 mm). The total suspended solids (TSS), volatile suspended solids (VSS) and pH of raw sludge with 12.91 (±0.43) g L⁻¹, 8.35 (±0.12) g L⁻¹ and 4.5 (±0.2). For hydrogen production, after 30 d of aerobic cultivation using HMWW as the substrate in a sequencing batch reactor, acclimated seed sludge was inoculated into UASBs, to one of which was added granular activated carbon (GAC) as the support medium for sludge immobilization. The physical characteristics of the GAC are as follows: surface area 900–1,100 m² g⁻¹, bulk area 0.33–0.38 g mL⁻¹, hardness 95–99.9%, ash 3–6%, moisture content 5% and particle size 0.5–2 mm. The GAC was supplied with an average diameter of 1 mm at a volume (mL) to weight (g) ratio of 10:1 (Hainan Wen Chang Qiu Chi Activated Carbon Co. Ltd). After 24 h of cultivation in the UASB, immobilized sludge was formed outside and inside the GAC. In the case of methane production, the raw sludge without any pretreatment was inoculated into the methanogenic UASB reactor. Metabolism activities of methanogens during sludge cultivation were examined by detecting methane content in biogas from the methanogenic reactor. After enough cultivation, the hydrogen-producing sludge with suspended solids of 6.1 g L⁻¹ and VSS of 5.5 g L⁻¹ was inoculated into the UASBs. After sludge got enough enrichment over 3 months, the effluents of hydrogen fermentation began to be used as substrate for methanogens with addition of alkali solution to regulate pH.

COD, pH, TSS and VSS of HMWW were analyzed according to Standard Methods (APHA 1998). Total nitrogen (TN) was measured by a TN analyzer (ModelTNM-1, Shimadzu, Japan), and total phosphorus was measured by an ion chromatograph (model ICS-900, Dionex).

Hydrogen and methane were analyzed using a gas chromatograph (SC-7, Shandong Lunan Instrument Factory). The gas chromatograph was equipped with a thermal conductivity detector and a stainless steel column (2 m × 5 mm) filled with Porapak Q (50–80 meshes). Nitrogen was used as the carrier gas at a flow rate of 40 mL min⁻¹. Detection of metabolites in the fermentation solution was carried out using another gas chromatograph (GC 112, Shanghai Hogon Analytical Instrument Co.) with a flame ionization detector. A 2-m stainless steel column was packed with the support GDX-103 (60–80 meshes). The temperatures of injector, detector and initial column was raised from 60 °C, 60 °C and 50 °C to 220 °C, 220 °C and 180 °C, respectively. Nitrogen was used as the carrier gas at a flow rate of 30 mL min⁻¹. Biogas generated from each reactor was collected using a wet gas meter (Model LML-1, Changchun Filter Co. Ltd, Changchun, China) and analyzed everyday (Han et al. 2011). Effluent samples from each reactor were also collected for metabolite analyses during the overall period of reactor operation.

The immobilized sludge system has been successfully applied to wastewater treatment in various bioreactors which possessed a quite stable and good COD reduction ability (Lee et al. 2008). It is known that COD is converted into intermediate metabolites significant in a hydrogen-producing system (Buitron et al. 2014); so the COD removal efficiency is lower than in the traditional anaerobic process. According to Figure 4, the system COD removal efficiency in the immobilized sludge reactor in this study stabilized in the range of 43.6 (±0.10)% to 50.1 (±0.09)%, but this was still higher than COD removal efficiency of the suspended sludge reactor, ranging from 25.8 (±0.70)% to 38.9 (±1.90)%. During the hydrogen-producing fermentation, the concentration ratio of BOD/COD could be improved from 0.19 to 0.45, which indicated the wastewater was more suitable for subsequent methanogenic fermentation. The biodegradability was effectively improved.

The UASB reactor was employed for methane production from effluents of hydrogen fermentation at HRT 35 h. The initial COD concentration of influent was kept at a constant level of 3.0 g L⁻¹.

As can be seen from Figure 6, the effluent pH in the methanogenic reactor was between 6.69 and 7.21 during the whole operation, within the optimum range of 6.0–8.0, which is suitable for the growth of methanogens (Liu et al. 1985; Hawkes et al. 2007). In the methanogenic reactor, the conversion of intermediate metabolites to methane by the methanogens realized the COD reduction. Compared with the effluent from hydrogen fermentation, a reduction of metabolite concentration could be observed for the methanogenic reactor. Sample analysis revealed that ethanol was converted totally and the acetate, propionate and butyrate concentration in the methanogenic effluent varied inconsistently at a very low level of 0.03–0.07 g L⁻¹, 0.04–0.07 g L⁻¹ and 0.02–0.06 g L⁻¹, respectively.

So far, many researchers have studied the feasibility of hydrogen production from various wastewaters, and subsequent methane production from hydrogen fermentation effluent. A comparison with previous researches can be seen in Table 3. A two-stage continuous stirred tank reactor (CSTR) system was employed for hydrogen and methane production from olive pulp (Koutrouli et al. 2009). The two-stage system obtained the maximum HPR of 0.86 mmol L⁻¹ hr⁻¹ and MPR of 2.10 mmol L⁻¹ hr⁻¹ at 7.5 h and 10 d of HRT, respectively. Wang & Zhao (2009) used food waste as a carbon source for hydrogen and methane production by the ‘SCRD-SCSTR’ system and they obtained the maximum hydrogen and methane production rates of 6.08 mmol L⁻¹ hr⁻¹ and 3.13 mmol L⁻¹ hr⁻¹, respectively. In addition, other types of two-stage system have also been set up for production of hydrogen and methane (Kyazze et al. 2006; Park et al. 2010). In our study, a maximum HPR of 10.0 mmol L⁻¹ hr⁻¹ at HRT 6 h was obtained from hydrogen fermentation of HMWW in the immobilized sludge UASB reactor. Effluents from hydrogen fermentation were converted into methane in the subsequent UASB reactor, reaching a maximum MPR of 5.49 mmol L⁻¹ hr⁻¹ at HRT 15 h. In the case of hydrogen and methane production, it could be comparable with previous works. Therefore, the ‘UASB-UASB’ system in this study showed excellent performance for production of hydrogen and methane.

It could be seen from Table 3 that the dual production of hydrogen and methane from the two-stage fermentation process was energetically favorable because energy recovery rate could be improved significantly once the two-stage process was adopted. Although the improved energy recovery rate could lead to additional economical benefits arising from simple down-stream processing, there is still much room for further development. One of the major bottlenecks of the two-stage fermentation process to realize commercial advancement is the lower hydrogen production, which must be improved. Matsuura et al. (2015) once improved hydrogen yield to 3.4 mol/mol glucose by blocking alcohol and acid formation pathways through the mutagenesis of Enterobacter cloacae. Some other methods were also adopted for hydrogen productivity, such as maintaining a low partial pressure of hydrogen (Intanoo et al. 2014), integrating the fermentative process with photosynthesis (Das & Veziroglu 2001) and optimum bioreactor design (Pisutpaisal et al. 2014). In our study, the obtained HPR of the immobilized sludge reactor is 19.9% higher than that of the suspended sludge reactor at OLR 24 g COD L⁻¹ d⁻¹. This result showed that the immobilized sludge rector is noticeable for hydrogen production. For comparison, a maximum HPR of 12.5 mmol L⁻¹ hr⁻¹ was obtained by Han et al. (2012) in an immobilized sludge CSTR system using molasses as substrate.

The hydrogen production performances of a UASB reactor with suspended and immobilized sludge from HMWW were investigated and compared at various OLRs (8–40 g COD L⁻¹ d⁻¹); the effluents from hydrogen fermentation were further converted into methane in a subsequent UASB reactor. Experimental data illustrated that the immobilized sludge system possessed better stability and higher production capacity than the suspended sludge system. At OLR 24 g COD L⁻¹ d⁻¹ of the hydrogen-producing phase and HRT 15 h of the methanogenic phase, the adopted two-stage fermentation process could improve significantly the energy recovery rate to 7.26 kJ L⁻¹ hr⁻¹ by the dual production of hydrogen and methane. Therefore, the ‘UASB-UASB’ system in this study showed excellent performance for production of hydrogen and methane.

This work was supported by the Projects in the National “Twelfth Five-Year” Plan for Science & Technology Support (2011BAD08B01-03), the Fundamental Research Funds for the Central Universities (2572014AB09) and the Natural Science Foundation of Heilongjiang Province (E201354).