328

JOURNAL OF BIOSCIENCE AND BIOENGINEERING © 2006, The Society for Biotechnology, Japan

Vol. 102, No. 4, 328–332. 2006

DOI: 10.1263/jbb.102.328

Effects of Temperature and Hydraulic Retention Timeon Anaerobic Digestion of Food Waste

Jung Kon Kim,1 Baek Rock Oh,2 Young Nam Chun,2 and Si Wouk Kim2*

Department of Bio Materials Engineering, Chosun University, 375 Seosuk-dong, Dong-gu, Gwangju 501-759,South Korea1 and Department of Environmental Engineering, BK21 Team for Biohydrogen Production,

Chosun University, 375 Seosuk-dong, Dong-gu, Gwangju 501-759, South Korea2

Received 22 May 2006/Accepted 18 July 2006

A modified three-stage methane fermentation system was developed to digest food waste effi-ciently. This system consisted of three stages: semianaerobic hydrolysis, anaerobic acidogenesisand strictly anaerobic methanogenesis. In this study, we examined the effects of temperature andhydraulic retention time (HRT) on the methanogenesis. Operation temperature was adjustedfrom 30°C to 55°C, and the HRTs ranged from 8 to 12 d. The rate of soluble chemical oxygen de-mand (sCOD) removal correlated with digestion time according to the first-order kinetic modeldeveloped by Grau et al. [Water Res., 9, 637–642 (1975)]. With liquor food waste, thermophilic di-gesters showed a higher rate of sCOD removal than mesophilic digesters. The rates of biogas andmethane production by thermophilic digesters were higher than those by mesophilic digesters re-gardless of HRT. Although maximum biogas production occurred when an HRT of 10 d was used,the methane yield was the highest in the reactor when an HRT of 12 d was used (223 l CH4/kgsCODdegraded). However, digestion stability decreased when an HRT of 8 d was used. The concen-tration of NH3-N generated in this experiment did not inhibit anaerobic digestion.

[Key words: anaerobic digestion, methane, temperature effect, organic waste, bioenergy]

Approximately 48 million tons of organic waste such asfood waste, livestock manure, vegetable waste and waste-water sludge is produced in Korea each year. This organicwaste is generally treated in sanitary landfills. However, thismethod of disposal causes secondary pollution of ground-water and soil. In particular, Korean food waste putrefiesbecause of its high water content, which makes its transportand storage difficult.

Anaerobic digestion has been suggested as an alternativemethod of removing the high-concentration organic waste.Several research groups have developed anaerobic digestionprocesses using different substrates (1–5). The advantagesof such processes over conventional aerobic processes are alow energy requirement for operation, a low initial invest-ment cost and a low sludge production. In addition, theanaerobic digestion process produces biogas, which can beused as a clean renewable energy source (6–8). In Europe,increasing numbers of biogas plants (BGPs) employinganaerobic digestion use food waste and manure as energysources. Currently, there are 19 BGPs in Denmark, 11 inGermany, and 10 in Sweden with additional plants underconstruction (9, 10).

Anaerobic digestion can be developed for different tem-perature ranges including, mesophilic temperatures of ap-proximately 35°C and thermophilic temperatures ranging

from 55 °C to 60°C (11). Conventional anaerobic digestionis carried out at mesophilic temperatures, that is, 35–37°C.However, the thermophilic temperature range is worth con-sidering because it will lead to give faster reaction rates,higher gas production, and higher rates of the destruction ofpathogens and weed seeds than the mesophilic temperaturerange. However, the thermophilic process is more sensitiveto environmental changes than the mesophilic process (12–14).

A modified three-stage methane fermentation system thatcan produce methane from easily biodegradable food wastewas developed in our laboratory (15, 16). This system usesa process that entails three stages: semianaerobic hydroly-sis/acidogenesis, strictly anaerobic acidogenesis, and strict-ly anaerobic methanogenesis. This separate-reactor systemefficiently decreases hydraulic retention time (HRT) by in-creasing the rates of hydrolysis, acidogenesis and methano-genesis without affecting pH, and shows a high methaneyield. In this study, we determined the optimum reactiontemperature and HRT of the methanogenesis, which is di-rectly related to methane production, with the aim of in-creasing food waste digestion efficiency using the three-stage methane fermentation system.

MATERIALS AND METHODS

Design of anaerobic methanogenic fermentor The lab-scalemethanogenic digester used was a cylindrical-type reactor (200 [i.d.]×350 [h] mm) equipped with sampling ports, a gas valve and pres-

* Corresponding author. e-mail: swkim@chosun.ac.krphone: +82-62-230-6649 fax: +82-62-225-6040

EFFECTS OF TEMPERATURE AND HRT ON ANAEROBIC DIGESTIONVOL. 102, 2006 329

sure gauze. The total volume was 11 l and the working volume was8 l (Fig. 1).

Chemical properties of reaction mixture The three-stagemethane fermentation system developed in this laboratory used aprocess that entailed a primary semianaerobic hydrolysis/acido-genic stage, a secondary anaerobic acidogenic stage, and a tertiarystrictly anaerobic methanogenic stage. The system was operatedto digest food waste rapidly. Table 1 shows the characteristics ofKorean food waste. These characteristics indicate that the foodwaste is a good source for methane fermentation.

In this study, the acid fluid eluted from the secondary acido-genic fermentor was used as the substrate and the methanogenicfluid eluted from the methanogenic fermentor was used as the in-oculum. The microorganisms used in the acidogenic fermentationwere Clostridium species and those used in the methanogenic fer-mentation were a mixture of methanogenic bacteria isolated fromlandfill soil and cow manure. The pH, soluble chemical oxygen de-mand (sCOD), ammonium concentration and C/N ratio of the sub-strate were 4.5, 17,000 mg/l, 380 mg/l and 15.8, whereas those ofthe inoculum were 8.2, 2600 mg/l, 1300 mg/l and 2.0, respectively.

Operation conditions In the batch reaction, 8 l of a mixtureof the substrate and inoculum (1:1) was placed in the methano-genic reactor. Each reactor was operated at increasing tempera-tures ranging from 30°C to 55°C at intervals of 5°C. The changesin pH, sCOD, and gas production were monitored. The effect oftemperature the kinetics of the process was analyzed using the fol-lowing first-order model developed by Grau et al. (17).

S/S0=exp−(k

1X

0t/S

0) (1)

Here, S is the concentration of organic matter at any fermentationtime (g-COD l–1), S

0 is the initial substrate concentration (g-COD

l–1), k1 is a first-order kinetic constant (d–1), and X

0 is the initial

concentration of microorganisms expressed as grams of volatilesuspended solids per liter (VSS l–1). Assuming that X

0 was con-

stant during the experiment, which means that k1X

0 is a constant

(k′1, g-COD l–1d–1), Eq. 1 was simplified to

S/S0 = exp −(k′

1t/S

0) (2)

The kinetic constant of the first-order model was calculated bytransforming Eq. 1 to

−ln(S/S0) = pt (3)

where p is equal to k′1/S

0.

Because p and S0 are constants under the experimental condi-

tions studied, a plot of −ln(S/S0) against t yielded a straight line. k′

1

was determined from the slope (p, d–1) of the line of the best fit us-ing the least-squares method as

k′1 = pS

0 (4)

In the semicontinuous experiment, the working volume was 8 l(volume is changes depending on HRT) and fresh substrate wasplaced in the digester each day. Operation temperature was changedfrom 40°C to 55°C at intervals of 5°C, and HRT was set at 8, 10and 12 d.

Analytical methods The following parameters were exam-ined: pH, sCOD, and ammonium (NH

3-N) concentration. sCOD

was measured using the sealed tube method (sample size = 2 ml)(DR/4000U spectrophotometer; Hach, Loveland, CO, USA) afterfiltration (Whatman GF/C). NH

3-N was determined using NH

3 sen-

sitive electrodes (model 95-12; Thermo Orion, Beverly, MA, USA),connected to a pH meter (710Aplus pH/ISE Meter; Thermo Orion).

The amount of gas produced from the methanogenic digesterwas monitored using a wet gas meter (W-NK-5; Shinagawa, Tokyo),and gas composition was analyzed by gas chromatography (GC-14B; Shimadzu, Kyoto) using a packed column (100/120 mesh,1/8′′×10 ft, Haysep D; Alltech, Columbia, SC, USA). The temper-atures of the column, injector, and detector (thermal conductivity)were 100°C, 50°C and 150°C, respectively.

RESULTS AND DISCUSSION

With liquor food waste, the thermophilic digester gave astable performance, which was superior to that of the meso-philic digester. Figure 2A shows sCOD as a function oftemperature for the batch reaction (S0= 9800 mg-COD/l).sCOD decreased with increasing digestion time. The maxi-mum sCOD removal efficiency (83%) was obtained fromthe digester operated at 50°C. The removal rate at 45°C wasalmost the same as that at 50°C. However, the rate at 55°Cdecreased to 77%. Removal rates also gradually decreasedat temperatures lower than 40°C. Figure 2B shows a plot of−ln(S/S0) as a function of digestion time (d) for some tem-peratures. A plot of the experimental data, as indicated byEq. 3, gave a straight line with the intercept at almost zero.The kinetic constant k′1 was obtained using Eq. 4. The k′1values were 0.14 (30°C), 0.35 (35°C), 0.38 (40°C), 0.6(45°C), 0.61 (50°C), and 0.42 (55°C). sCOD removal effi-ciency increased with temperature up to 50°C. Chen et al.(18) reported that there is a kinetic advantage in the diges-tion of livestock manure at thermophilic temperatures overthat at mesophilic temperatures. However, the advantages ofcarrying out the digestion at temperatures higher than 55°Care insignificant.

Figure 3 shows the levels of biogas productions at differ-ent temperatures in the batch reaction. The a mount of bio-gas produced from the reactors at 45°C and 50°C increasedwith decreasing sCOD. Consequently, it was found that astabilization period is needed when digestion is carried outat temperatures lower than 45°C or higher than 50°C. Thisresult suggests that the activity of the methanogenic bacteriaused in this study depends on to operation temperature.Therefore, a thermophilic temperature <55°C is more effec-tive for biogas production than a mesophilic temperature.Bouallagui et al. (19) also reported higher biogas produc-

FIG. 1. Schematic of lab-scale methane digester.

TABLE 1. Elemental analysis of collected food waste

Solidcontent

(%)

VS/TS(%)

COD/VS

Elemental composition (%)

C H O N

12.38 89.3 0.71 47.8 6.1 40.9 5.2

KIM ET AL. J. BIOSCI. BIOENG.,330

tion using an experimental thermophilic digester than thoseusing psychrophilic and mesophilic digesters.

Figure 4 shows the HRT-dependent removal efficienciesof sCOD by the semicontinuous reaction. The mean influentsCOD was 14,500 mg/l. When the HRT was 10 d, the sCODsof the digesters operated at 40°C, 45°C, and 50°C decreased

to 3480, 3361, and 3240 mg/l, respectively. The rates ofsCOD removal were 76%, 77%, and 78% at 40°C, 45°C,and 50°C, respectively (Fig. 4A). The rate at 55 °C was sim-ilar to that at 40°C. Regardless of temperature, the pattern ofsCOD removal using an HRT of 12 d was similar to thatusing an HRT of 10 d, even though sCOD removal efficiencyusing an HRT of 12 d was better (Fig. 4B). As previouslystated, sCOD removal efficiency increased with increasingtemperature up to 50°C, regardless of whether the HRT was10 d or 12 d. These results are in accordance with thosereported by Mackie and Bryant (20). They mentioned thatthe thermophilic digestion of cattle waste is more stablethan the mesophilic digestion at different HRTs. However,when HRT was decreased to 8 d, pH decreased significantlyand no sCOD removal or biogas production was observed(data not shown).

Table 2 shows the temperature- and HRT-dependent totalbiogas and methane productions in the semicontinuous re-action. At 50°C, the maximum amount of biogas was pro-duced when an HRT of 10 d was used, whereas methanewas generated efficiently when an HRT of 12 d was used.On the other hand, the lowest biogas production was ob-served at 55°C with methane contents ranging from 54% to60%, depending on HRT.

Table 3 shows the calculated methane yields according to

FIG. 2. (A) Temperature-dependent changes in sCOD in anaerobicmethane digesters in batch reaction. (B) Grau model plot of tempera-ture-dependent changes in sCOD.

FIG. 3. Biogas production profiles at different temperatures throughbatch reaction.

FIG. 4. Temperature-dependent sCOD removal efficiencies of dif-ferent anaerobic methane digesters by continuous reaction. (A) HRT-10 d; (B) HRT-12 d.

EFFECTS OF TEMPERATURE AND HRT ON ANAEROBIC DIGESTIONVOL. 102, 2006 331

the results shown in Table 2. Although the maximum biogasproduction occurred at 50°C with a 10 d-HRT, the methaneyield was highest (223 l CH4/kg sCODdegraded) in the reactorwith a 12 d-HRT. Ahn and Forster (12) reported that a highspecific methane yield could be obtained in a thermophilicdigester with a long HRT. In many studies, researchers haveexamined the potential of thermophilic digesters with themain technical concern being not whether a thermophilic di-gestion system will operate, but whether it has actual advan-tages over a mesophilic digestion system for any type ofwastewater. Overall, the results show that there is a distinctadvantage in considering the treatment of food waste by

anaerobic digestion in the thermophilic range with a 10-dHRT.

As shown in Fig. 5, NH3-N concentration increased from380 to approximately 1000 mg/l at all the temperatures testedthrough semicontinuous anaerobic digestion. Many studieshave reported the inhibitory effects of a high free ammonia(NH

3) concentration on the metabolism of methanogens (21–

23). However, NH3-N concentration can be ignored for thedigestion of food waste because the NH3-N concentrationgenerated in this experiment did not inhibit anaerobic diges-tion significantly.

Currently, we are exploiting ways to improve the anaer-obic digestion of Korean food waste using a three-stagemethane fermentation system developed in this laboratory.The above results show that thermophilic digestion with a12-d HRT in the tertiary methanogenic digester is the bestway to increase the performance of anaerobic digestion andfood waste digestion efficiency.

ACKNOWLEDGMENTS

This work was supported by grant no. R01-2006-000-10355-0from the Basic Research Program of the Korean Science &Engineering Foundation.

REFERENCES

1. Lastella, G., Testa, C., Cornacchia, G., Notornicola, M.,Voltasio, F., and Sharma, V. K.: Anaerobic digestion ofsemi-solid organic waste: biogas production and its purifica-tion. Energy Convers. Manage., 43, 63–75 (2002).

2. Gallert, C., Henning, A., and Winter, J.: Scale-up of anaer-obic digestion of the biowaste fraction from domestic wastes.Water Res., 37, 1433–1441 (2003).

3. Tada, C. and Sawayama, S.: Photoenhancement of biogasproduction from thermophilic anaerobic digestion. J. Biosci.Bioeng., 98, 387–390 (2004).

4. Choi, D. W., Lee, W. G., Lim, S. J., Kim, B. J., Chang,H. N., and Chang, S. T.: Simulation on long-term operationof an anaerobic bioreactor for Korean food wastes. Biotechnol.Bioprocess Eng., 8, 23–31 (2003).

5. Nakamura, Y. and Mtui, G.: Anaerobic fermentation ofwoody biomass treated by various methods. Biotechnol.Bioprocess Eng., 8, 179–182 (2003).

6. Bouallagui, H., Ben Cheikh, R., Marouani, L., and Hamdi,M.: Mesophilic biogas production from fruit and vegetablewaste in a tubular digester. Bioresour. Technol., 86, 85–89(2003).

7. Uemura, S. and Harada, H.: Treatment of sewage by aUASB reactor under moderate to low temperature conditions.Bioresour. Technol., 72, 275–282 (2000).

8. Milan, Z., Sanchez, E., Weiland, P., Borja, R., Martin, A.,and Ilangovan, K.: Influence of different natural zeoliteconcentrations on the anaerobic digestion of piggery waste.Bioresour. Technol., 80, 37–43 (2001).

9. Albihn, A., Norin, E., and Lindberg, A.: Regulation andpresent situation in Sweden concerning hygienic and environ-mental safety for anaerobic digestion of waste, p. 73–79. InBöhm, R. and Wellinger, A. (ed.), Hygienic and environmen-tal aspects of anaerobic digestion: legislation and experiencesin Europe. IEA Workshop, Stuttgart-Hohenheim (1999).

10. Sahlstrom, L.: A review of survival of pathogenic bacteria inorganic waste used in biogas plants. Bioresour. Technol., 87,161–166 (2003).

11. Sanchez, E., Borja, R., Weiland, P., Travieso, L., and

TABLE 2. Comparisons of biogas productions dependingon temperature and HRT by continuous reaction

using anaerobic methane digesters

Digestion condition Gas production (l/d) Methanecontent

in biogas(%)

Temp(°C)

HRT(d)

Biogas Methane

40 10 7.3 4.5 61.612 6.1 4.0 65.6

45 10 8.7 5.5 63.212 7.4 4.9 66.2

50 10 10.4 6.7 64.412 8.6 5.8 67.4

55 10 6.8 3.7 54.412 5.6 3.3 58.9

TABLE 3. Methane yields in continuous digestion

Digestion condition Methane gas yield(l CH4/kg sCODdegraded)Temp (°C) HRT (d)

40 10 14512 154

45 10 17712 187

50 10 21612 223

55 10 11912 129

FIG. 5. Temperature-dependent NH3-N concentration changes indifferent anaerobic methane digesters by continuous reaction.

KIM ET AL. J. BIOSCI. BIOENG.,332

Martin, A.: Effect of substrate concentration and temperatureon the anaerobic digestion of piggery waste in a tropical cli-mate. Process Biochem., 37, 483–489 (2001).

12. Ahn, J.-H. and Forster, C. F.: A comparison of mesophilicand thermophilic anaerobic upflow filters treating paper-pulp-liquors. Process Biochem., 38, 256–261 (2002).

13. El-Mashad, H. M., van Loon, W. K. P., and Zeeman, G.: Amodel of solar energy utilisation in the anaerobic digestion ofcattle manure. Biosyst. Eng., 84, 231–238 (2003).

14. Kim, M., Ahn, Y.-H., and Speece, R. E.: Comparative proc-ess stability and efficiency of anaerobic digestion; mesophilicvs. thermophilic. Water Res., 36, 4369–4385 (2002).

15. Kim, S. W., Park, J. Y., Kim, J. K., Cho, J. H., Chun, Y. N.,Lee, I. H., Lee, J. S., Park, J. S., and Park, D. H.: Develop-ment of a modified three-stage methane production processusing food wastes. Appl. Biochem. Biotechnol., 84–86, 731–741 (2000).

16. Kim, J. K., Cho, H. C., Lee, J. S., Hahm, K. S., Park, D. H.,and Kim, S. W.: Mass production of methane from foodwastes with concomitant wastewater treatment. Appl. Biochem.Biotechnol., 98–100, 753–764 (2002).

17. Grau, P., Dohanyos, M., and Chudoba, J.: Kinetics ofmulticomponent substrate removal by activated sludge. Water

Res., 9, 637–642 (1975).18. Chen, Y., Varel, V. H., and Hashimoto, A. G.: Effect of tem-

perature on methane fermentation kinetics of beef-cattle ma-nure. Biotechnol. Bioeng. Symp., 10, 325–339 (1980).

19. Bouallagui, H., Haouari, O., Touhami, Y., Ben Cheikh, R.,Marouani, L., and Hamdi, M.: Effect of temperature on theperformance of an anaerobic tubular reactor treating fruit andvegetable waste. Process Biochem., 39, 2143–2148 (2004).

20. Mackie, R. I. and Bryant, M. P.: Anaerobic digestion of cat-tle waste at mesophilic and thermophilic temperatures. Appl.Microbiol. Biotechnol., 43, 346–350 (1995).

21. Angelidaki, I. and Ahring, B. K.: Thermophilic anaerobicdigestion of livestock waste: The effect of ammonia. Appl.Microbiol. Biotechnol., 38, 560–564 (1993).

22. Zeeman, G., Wiegant, W. M., Koster-Treffers, M. E., andLettinga, G.: The influence of the total ammonia concen-tration of the thermophilic digestion of cow manure. Agric.Wastes, 14, 19–35 (1985).

23. Sawayama, S., Tada, C., Tsukahara, K., and Yagishita, T.:Effect of ammonium addition on methanogenic community ina fluidized bed anaerobic digestion. J. Biosci. Bioeng., 97,65–70 (2004).