초록
In this study, a two-stage fermentation system to produce H<SUB>2</SUB> and CH<SUB>4</SUB> from Laminaria japonica was developed. In the first stage (dark fermentative H<SUB>2</SUB> production, DFHP), response surface methodology (RSM) with a Box-Behnken design (BBD) was applied for optimization of operational parameters, including cycle-frequency, HRT, and substrate concentration, using an intermittent-continuously stirred tank reactor (i-CSTR). Overall performance revealed that the degree of importance of the three variables in terms of H<SUB>2</SUB> yield is as follows: cycle-frequency > substrate concentration > HRT. In the confirmation test, H<SUB>2</SUB> yield of 113.1 mL H<SUB>2</SUB>/g dry cell weight (dcw) was recorded, corresponding with 96.3% of the predicted response value under desirable operational conditions (cycle-frequency of 17 hr, HRT of 2.7 days, and substrate concentration of 31.1 g COD/L). In the second stage, an anaerobic sequencing batch reactor (ASBR) and an up-flow anaerobic sludge blanket reactor (UASBr) were employed for CH<SUB>4</SUB> production from H<SUB>2</SUB> fermented solid state (HFSS) and H<SUB>2</SUB> fermented liquid state (HFLS), respectively. The CH<SUB>4</SUB> producing ASBR and UASBr showed a stable CH<SUB>4</SUB> yield and COD removal until a HRT of 12 days and OLR of 3.5 g COD/L/d, respectively. Subsequently, for recycling of CH<SUB>4</SUB> fermented effluent from the UASBr (MFE<SUB>UASBr</SUB>) as diluting water in DFHP, the tap water and MFE<SUB>UASBr</SUB> mixing ratio (T/M ratio) was optimized (a T/M ratio of 5:5) in a batch test using heat pretreated MFE<SUB>UASBr</SUB> at 90 <SUP>o</SUP>C for 20 min, resulting in the best performance. Although slight decreases of H<SUB>2</SUB> yield (7.6%) and H<SUB>2</SUB> production rate (3.5%) were recorded, 100% reduction of alkali addition was possible, indicating potential to maximize economic benefits. However, a drastic decrease of H<SUB>2</SUB> productivity and a change of liquid-state metabolites were observed with the use of non-heat pretreated MFE<SUB>UASBr</SUB>. These results coincided with those of the microbial analysis, where non-H<SUB>2</SUB> producing bacteria, such as Selenomonas sp., were detected. The results indicate that pretreatment of MFE<SUB>UASBr</SUB> may be required in order to recycle it in DFHP.