• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Effect of a Starch-Rich Industrial Wastewater on the Acid-Phase Anaerobic Digestion Process

Posted on: Wednesday, 27 July 2005, 03:00 CDT

ABSTRACT:

This research investigated the effect of varying the starch- rich, industrial-wastewater component of mixtures with municipal wastewater fed to an anaerobic digester. A laboratory-scale, completely-mixed anaerobic digester was operated at an HRT of 30 h, an SRT of 10 d, and an ambient temperature of 21.5 1.5 C. The industrial-to-municipal ratios tested were 1:3, 1:1, 3:1, and 100% industrial by volume. Steady-state, acidogenic conditions were achieved for all runs, except 100% industrial. The pH was observed to drop substantially as the industrial constituent of the feed increased. Net volatile fatty acids (VFA) production reached a plateau of approximately 800 mg/L at ratios of 1:1 and higher, while volatile suspended solids (VSS) reduction steadily increased as the industrial component rose. The specific VFA and soluble chemical oxygen demand (SCOD) production rates leveled off at approximately 0.070 mgVFA/ mgVSS.d and 0.124 mgSCOD/mgVSS.d, respectively, for all the mixtures investigated, except for 100% industrial. In this latter case, both rates dropped dramatically. Finally, acetic and propionic acid concentrations fell as the industrial proportion of the mixture increased. This was compensated by a rise in butyric acid production. Water Environ. Res., 77, 366 (2005).

KEYWORDS: acid-phase, anaerobic digestion, starch wastewater, biological nutrient removal

doi: 10.2175/106143005X51941

Introduction

Starch wastewater is a major byproduct of potato processing industries and is high in chemical oxygen demand (COD), nitrogen (N), phosphorous (P), and volatile suspended solids (VSS). These four parameters have been shown to be reduced in a two-stage anaerobic-aerobic wastewater treatment process to values under current discharge limits (Abeling and Seyfried, 1993). The authors suggest that the role of the anaerobic pretreatment phase was to convert colloidal starch molecules into soluble COD [SCOD] required for nutrient (N and P) removal. This aim was largely achieved and control of the anaerobic process was excellent. Several other studies also encourage the notion that starch wastewater can be anaerobically digested. For example, one study investigating the conversion of cellulose, a carbohydrate closely related to starch, determined that an optimum pH of 5.6 and an hydraulic retention time (HRT) of 48 h were required to convert 44% of colloidal COD to SCOD products (i.e., volatile fatty acids [VFAs] and alcohols) (Chyi and Dague, 1994). A second study examined the digestion of cornstarch wastewater using two bench-scale reactor configurations (Kwong and Fang, 1996). Although both reactors removed 95.3 % COD at an HRT of 12 h with influent COD concentrations up to 45 g/L, both systems eventually failed because of solids washout when organic loads reached 90 g COD/L.d.

The first phase of the anaerobic digestion process is hydrolysis- acidogenesis, which leads to the production of VFAs (Fox and Pohland, 1994; Valentini et al., 1997). Within this phase, it is evident that general patterns of VFA production can be explored by examining the interaction between wastewater characteristics, environmental conditions (e.g., temperature and pH), reactor type, and operational parameters (e.g., HRT and solids retention time [SRT]) (Fongsatitkul et al., 1994; Jayaseelan and Matsuo, 1995). For example, one study focusing on the influence of pH indicated that neither hydrolysis nor acidogenesis were significantly affected by varying the pH between 4.3 and 5.2; however, pH levels ranging from 5.9 to 6.2 resulted in an approximately 30% decrease in the concentration of VFAs (Elefsiniotis et al., 1996). In another study, mesophilic digestion (T = 35C) resulted in better digestion of carbohydrates and lipids, but inferior organic nitrogen breakdown as compared to thermophilic digestion (T = 55C) (Kugelman and Guida, 1989). However, VFA production was observed to be highest at temperatures over 507deg;C.

Of particular interest is the influence of operational factors such as HRT and SRT. In one experiment on primary sludge, hydrolysis and acidogenesis rates steadily rose as the HRT increased from 6 to 12 h, while, at an HRT of 15 h, there was a moderate drop in both rates attributed to the slight presence of methanogenic bacteria (Elefsiniotis and Oldham, 1994a). Subsequently, the same authors held the HRT at 12 h, while the SRT was adjusted from 10 to 2Od (Elefsiniotis and Oldham, 1994b). They found that VFA production was independent over the SRT range studied; however, the proportion of VFAs in the soluble COD increased from 70% (10-d SRT) to 90% (20-d SRT), suggesting that, at the longer SRTs, more of the intermediate products (e.g., soluble amino acids) were converted to VFAs.

Finally, some research has recently been conducted on mixtures of municipal and starch wastewaters. Banerjee et al. (1999) carried out a study at an SRT of 10 days investigating the following three parameters: HRT (18 to 30 h), temperature (ambient to 35C), and the industrial-to-municipal feed waste ratio (1:1 and 100% municipal by volume). They concluded that in terms of maximizing VFA production, the optimal value for HRT was 30 h for a 1:1 mixture of municipal and starch wastewater at 30C. In a different study, Maharaj and Elefsiniotis (2001) increased the HRT from 30 to 60 h and decreased the temperature from ambient to 8C to investigate the effect on VFA production of the same wastewater combinations (i.e., 1:1 and 100% municipal by volume). In their study, it was shown that a 1:1 mixture resulted in the highest VFA production at an ambient temperature and a 30-h HRT.

Figure 1-Schematic of anaerobic digester system.

These latter two studies in particular suggest that, to maximize VFA production during the anaerobic digestion of a 1:1 mixture of municipal and starch-rich industrial wastewater, optimal values of 30 h and 10 d for HRT and SRT are required. The present research, therefore, extends the previous research by varying the starch wastewater-to-municipal wastewater ratio using these particular values of HRT and SRT. To determine the effect of the ratio on VFA production, this research experiments with wastewater ratios of 1:3, 1:1, 3:1, and 100% starch.

Experimental Methods

Configuration of Apparatus. A schematic of the laboratory-scale anaerobic digester system used in this research is shown in Figure 1. The anaerobic reactor was a plexiglass, cylindrical 4-L vessel (liquid volume of 3 L) with an internal diameter of 11.4 cm and a sampling port located 17 cm up from its base. The reactor was stirred continuously at a rate deemed sufficient by visual inspection to ensure solids were suspended at all times. The wastewater feed mixture was provided every 30 min for a duration of 1 min. Recycle of the biomass from the clarifier to the reactor coincided with feeding. The plexiglass clarifier had a modified Imhoff conical shape to aid in solids settling with the internal diameter of the clarifier being 10.8 cm and a liquid volume of 2 L. Further details of the experimental apparatus are presented elsewhere (Smith, 2002).

Table 1-Influent wastewater characteristics for all runs (industrial: municipal ratio by volume).

Source and characteristics of wastewater. Two wastewater types were involved in this study. The first was municipal primary concentrated sludge (10 000 mg/L < total solids (TS) < 63 000 mg/L) delivered every two weeks from the South End Pollution Control Centre in Winnipeg, Canada. The second wastewater consisted of a starch-rich, industrial sludge collected approximately every three weeks from the Old Dutch Food Company, also located in Winnipeg. The Old Dutch Food Company processes potatoes into potato chips; thus, the process water is very high in starch solids. The TS content of this latter sludge ranged from 35 000 to 1OO 000 mg/L. Both types of sludge were stored in a cold chamber at the University of Manitoba at 6C. Before use, they were diluted with tap water to a TS level of approximately 5000 mg/L (i.e., 0.5%). The reason for dilution was evidence that anaerobic contact processes perform better when solids levels are minimized (Hall, 1992). In addition, it has been found that VFA production and sludge settleability are enhanced during anaerobic digestion through the use of dilute wastewater (i.e., less than 10 000 mg/L volatile solids [VS]) (Skalsky and Daigger, 1995).

A cold-room study on the raw wastewater components (i.e., the undiluted sludges) indicated that VFA production did not cease at the low temperature encountered in the cold storage room.

However, observed increases in VFA content did not affect the final results, as these were evaluated as net production rather than as gross production. Table 1 shows the variation in some of the more important analytical parameters specific to this study. These characteristics are associated with the diluted influent wastewater for all five runs (A to E). As can be seen the variation in some parameters (VFAs and ammonia-nitrogen [NH^sub 3^-N]) is large and/ or even greater than the mean. This reflects both the natural variation in influent characteristics and the effect of continued degradation of organic m\atter in the cold storage room.

Operation and Sampling. Before detailed sampling, the system acclimated for approximately two weeks to allow the anaerobic population to establish itself. No pH control was exercised during the entire study, and a fluctuation of more than 10% in pH and/or an obvious upward or downward trend were interpreted as signs of instability (non-steady-state) as per the practice of previous research (Elefsiniotis et al., 1996).

Figure 2-Variation in reactor pH for runs A, B, C, and D.

Figure 3-Variation in net volatile fatty acid production for runs A, B, C, and D.

As indicated, a total of five runs were conducted over the course of this research. The first four runs (A to D) involved varying the feed ratio, while the last run (E) was a replicate run to determine reproducibility. Each run was conducted over a period of approximately 45 to 70 d, according to the need to assess stability conditions. All runs were conducted at an HRT of 30 h and an ambient temperature of 21C 1.5C. The only variable between runs was the feed industrial-to-municipal ratio, as indicated in Table 1. The HRT was checked and maintained through the adjustment of the influent pump speed controller. A nominal SRT of 10 d was maintained by wasting quantities of reactor solids. Gas production and temperature were monitored on an approximate daily basis.

Influent, reactor, and effluent samples were taken approximately every 3 to 4 d, resulting in 12 to 16 sample sets during each run. Alkalinity, pH, TS, total suspended solids (TSS), VS, VSS, VFA, SCOD, NH^sub 3^-N, and total kjeldahl nitrogen (TKN) concentrations were evaluated for all samples according to Standard Methods (APHA et al., 1998). Again, more detail is presented in Smith (2002).

Quality Assurance. Each of the measurements conducted over the course of this research were done in duplicate to provide greater confidence in the results. Analytical error observed between these duplicates varied with certain parameters. For example, pH exhibited very small fluctuations (approximately 1 to 2%), while the largest fluctuations were approximately 20% (observed for some VFA and solids measurements). The other parameters, namely alkalinity, SCOD, NH^sub 3^-N, and TKN, displayed occasional errors not exceeding the 10% level.

Results

Stability of Reactor Operation. Anaerobic bacteria are sensitive to changes in influent conditions, and the reactor population will respond to these variations by shifting the proportion of its bacterial composition. The success or failure of these bacterial populations will express themselves through measures such as pH, VFA production, VFA speciation, and VSS conversion to SCOD. For instance, pH is an important indicator in anaerobic digestion stability, and Figure 2 shows the variation in pH over the course of the four main runs (A to D). In this study, pH fluctuations ranged from 2.4 to 5.8% of the mean value, indicating a reasonably stable system.

Volatile Fatty Acids Production and Speciation. Figure 3 illustrates the net VFA production patterns of runs A to D. As shown, all systems achieved significant net VFA production, ranging from approximately 200 to more than 1000 mg/L (as acetic acid). Table 2 displays the average net VFA production for runs A to D and their average specific VFA production rates. Net VFA values increased with increasing industrial content, peaking at approximately 800 mg/L in run B. Because VFAs are the principle end product of acidogenesis, it is safe to assume that hydrolysisacidogenesis was successful in each run.

Samples were analyzed not only for their overall VFA content, but also for their content of individual VFA compounds. Figure 4 illustrates the composition of reactor VFAs for runs A to D. The various VFAs that were identified were acetic acid, propionic acid, iso-butyric acid, n-butyric acid, iso-valeric acid, and n-valeric acid. As discussed later, the presence or absence of these compounds sheds some light on the prevailing metabolic pathways of digestion. Figure 4 also shows those VFAs detected by the previous two related studies for municipal-only feeds (Banerjee et al., 1999; Maharaj and Elefsiniotis, 2001). These results (and subsequent parameters) are for the set of operating conditions identical to the current study (i.e., HRT = 30 h, SRT = 10 d, and ambient temperature).

Evidence for Hydrolysis. The ability of a bacterial population to convert large, insoluble macromolecules into monomeric, soluble molecules (alternately called hydrolysis, solubilization, or liquefaction) can be measured through SCOD values. Table 3 presents the average net SCOD, the specific SCOD production rate, the VFAiSCOD ratio, and the average VSS reduction percentages. Results from Table 3 indicate a substantial amount of solubilization.

Table 2-Average net volatile fatty acid production and average specific volatile fatty acid production rates with varying industrial-to-municipal feed ratios.

Figure 4-Percent change in volatile fatty acid distribution as a function of industrial-to-municipal feed ratio.

Analysis and Interpretation of Results

Interpretation of pH measurements. It has been reported that carbohydrates are normally more easily digested than lipids and proteins (Breure et al., 1986a and b). Therefore, one would expect to see a large degree of hydrolysis and acidogenesis in reactors fed with starch influent mixtures. Because VFAs are the major end product of hydrolysis-acidogenesis, a higher degree of hydrolysis- acidogenesis should be accompanied by a lower pH resulting from high acid concentrations. The influent pH (Table 1) was always at least 1 to 3 pH units higher than that found in the reactor (Figure 2), confirming the occurrence of hydrolysis-acidogenesis in all runs. As the high-starch, industrial portion of the wastewater increased from 25 to 100% by volume, the average pH dropped (Figure 2). This is supported by the fact that run D had the lowest buffering capacity as reflected in the smallest influent alkalinity value (Table 1). Finally, the low pH values encountered during this study resulted in no gas production in all runs, indicating successful suppression of methanogenesis (Kayhanian and Tchobanoglous, 1992).

Net Volatile Fatty Acids Production and Speciation. Figure 3 indicates that for runs A to C, the net VFA production was more consistent than run D. This latter run not only had wider fluctuations, but it also exhibited an increasing VFA trend rising from approximately 200 mg/L (as acetic acid) at the beginning of the run to values greater than 1200 mg/L. Consequently, the coefficient of variation for run D was 42%, compared to approximately 20% for runs A to C.

The explanation for the large VFA variation in run D is not clear, but it may be linked to the population dynamics within the reactor. That is, municipal wastewater is more complex and, therefore, would promote the growth of a more diverse microbial population as compared to that generated from the use of a starch wastewater. The result would be a more stable population of VFA producers in systems with a higher proportion of municipal wastewater. Another premise is that the large VFA variation in run D was connected to a problem in controlling run D's SRT. That is, the starch solids by themselves settled extremely well and, as the industrial component of the wastewater mixture increased (despite mixing), a concentration gradient appeared to be created inside the reactor making it difficult to get a representative sample. Because the sampling port was located halfway up the reactor (Figure 1), the reactor VSS was being continuously underestimated resulting in an SRT value further and further removed from the target value of 10 d. This eventually led to non-steady-state conditions. Inconsistency of this magnitude may render starch-only wastewater difficult to anaerobically treat, at least under conditions similar to this study. However, it is acknowledged that more vigorous mixing could promote a more uniform concentration and partially alleviate the difficulty in controlling SRT.

Table 3-Net and specific SCOD production rates and average percent VSS reduction over the spectrum of industrial-to-municipal feed ratios.

Table 2 clearly shows the large average values of VSS for the 100% municipal and 100% industrial runs. These are substantially higher than mixtures of the two. As indicated previously, the 100% municipal wastewater studies were completely different from the current set of experiments and, although operated under similar operating conditions (in terms of SRT, HRT, and temperature), were separated by a considerable period of time from the study reported here. Although the same influent TS value was targeted, there was no attempt in the current study to achieve reactor VSS concentrations of a similar magnitude to the 100% municipal runs. Because the steady-state reactor VSS value is a function of influent characteristics, it is conceivable that quite different influent conditions were experienced between this and the previous study. For example, the TSS and VSS of the 100% municipal wastewater could have been quite different from the present study in terms of the nature of the suspended solids (i.e., fraction of colloidal matter). This could ultimately result in a very different average reactor VSS value for the steady-state operation of each run.

As shown in Table 2, the municipal-only feed results from the related studies (Banerjee et al., 1999; Maharaj and Elefsiniotis, 2001) averaged to a low specific VFA production rate of 0.022 mgVFA/ mgVSS-d. In contrast, the specific VFA production rate of any of the mixtures in this research (except run D) were all similar at approximately 0.070 mgVFA/mgVSS-d. Thus, it can be concluded that any mixture of the starch improves the performance of the reactor in terms of the rate of VFA production.

Run D, however (fed solely with starc\h wastewater), only managed a rate of 0.018 mgVFA/mgVSS-d. Acknowledging the non-steady-state conditions, this rate is less than that of the municipal-only feed. Because the 100% municipal runs had large VSS values and small amounts of net VFAs, one would logically expect a low specific rate. In contrast, run D achieved a low specific VFA production rate, despite having a high net VFA production, primarily because of the large average VSS value. That is, the inability to maintain the SRT meant that when the net production was normalized to obtain the specific rate, the high VSS value lowered the specific rate to even less than that of the rate of the municipal feed only.

Several VFA speciation patterns are apparent from an examination of Figure 4. Firstly, it is obvious that acetic acid is the dominant VFA product for all industrial-to-municipal feed mixture ratios. However, there is a sharp drop in its predominance at mixture ratios of 1:1 and greater. This pattern is repeated with propionic acid in the sense that any mixture of the starch wastewater substantially lowers the propionate proportion in the VFA distribution. The compound which seems to make up for the loss of both acetate and propionate is n-butyric acid. Municipal-only feed resulted in almost no n-butyrate production; however, its proportion jumped to approximately 35% at the higher starch ratios. The only other VFA compound that appeared in appreciable concentrations was n-valeric acid. Given the small production and variation between n-valerate numbers, it is not possible to make any definitive statement from these latter results.

Metabolism of Volatile Fatty Acids Production. Figure 4 implies that altering the industrial-to-municipal feed ratio has an effect on the degradation pathways used by anaerobic bacteria performing hydrolysis-acidogenesis. When no starch feed is included, acetic and propionic acid comprise over 95% of the end products. Because the anaerobic bacterial metabolism of carbohydrates, proteins, and lipids yields acetic acid, it is not surprising that acetic acid is the predominant VFA compound produced from all feed mixtures. As the proportion of starch wastewater in the feed is increased, n-butyric acid levels rise quickly, indicating that the metabolic pathways leading to the production of this acid are in use.

Butyrate and acetate production are often coincidental in mixed bacterial cultures. In addition, n-butyrate fermentation is one of the most energy-efficient uses of glucose, the major product of starch hydrolysis (Gottschalk, 1986). The increase in n-butyrate with starch feed content could, therefore, be explained by a microbial shift to the use of this more favorable metabolic pathway.

It was noted that propionic acid levels decreased as the starch content of the mixture increased. This suggests that propionate production is suppressed by the addition of carbohydrates. One explanation for this phenomenon may have to do with the relationship between lactic and propionic acid. Lactic acid is an intermediate metabolic product of glucose fermentation and the preferred substrate for propionate-producing bacteria according to the following reaction (Gottschalk, 1986):

3 lactate [arrow right] 2 propionate + acetate + CO2 (1)

Lactic acid bacteria have complex nutritional requirements that are likely provided for by municipal wastewater environments. As the starch component in the feed increases, the nutritional composition becomes more simplified, possibly leaving the nutritional requirements of lactic acid bacteria unsatisfied. It follows, then, that lactate production would drop, thereby dropping propionic acid production.

Solubilization and Acidogenesis Performance. The degree of solubilization in the reactor improved markedly with the addition of starch to the feed, as indicated by the average net SCOD results (Table 3). However, increasing the starch content beyond 25% seemed to have little effect. The average specific SCOD production rate for the same municipal wastewater is quite low as reported by others (Banerjee et al., 1999; Maharaj and Elefsiniotis, 2001). Table 3 indicates a substantial increase in the rate to an average value of 0.124 mgSCOD/mgVSS-d for runs A to C. The considerable drop in rate for run D is once again attributed to the normalizing effect of the large VSS reactor concentration. Probably the clearest evidence for the effect of increasing starch content on solubilization is revealed by the VSS reduction data. As can be seen, there was a steady increase in the percent VSS reduction to a value of approximately 72% in run C. The corresponding value for run D was not calculated because of the non-steady-state conditions. The performance of acidogenic bacteria can be gauged by the proportion of SCOD that is represented by VFAs. The VFAs were expressed as SCOD, and all VFA:SCOD ratios proved to be quite high, fluctuating at approximately the 100% level (Table 3). This indicates that all SCOD-contributing compounds in the reactor were present in the form of VFAs. This has also been observed in the other related studies (Banerjee et al., 1999; Maharaj and Elefsiniotis, 2001).

Table 4-Results of f-tests comparing runs B and E.

Feasibility of Replication. To ascertain the replicability of the experiment, run E was conducted at a 1:1 mixture ratio (i.e., the same ratio as run B). Table 4 summarizes the results for one parameter each that indicates acidogenesis and solubilization; namely, specific VFA production rate and VSS percent reduction. A t- test analysis to determine if there was a significant difference between the means of the two experimental runs was performed and the results are also shown in Table 4. They reveal that there was no significant difference between the two systems in terms of specific VFA production rate and percent VSS reduction at the 95% confidence interval. It can therefore be concluded that the findings of this study are reproducible, with the possible exception of the 100% industrial (run D), which could be altered if better mixing is obtained.

Potential Applications. On-site production of VFAs through anaerobic digestion of sludge is a viable alternative to external additions of carbon. In purposefully-designed wastewater treatment facilities, VFAs can be produced in sufficient quantities to provide a carbon source for both biological denitrification and biological phosphorus removal (Oldham et al., 1994). All the results of this study indicate that VFA production can be greatly enhanced through the addition of a starch wastewater into an anaerobic municipal wastewater digester. For example, net VFA production nearly doubled with an industrial-to-municipal ratio of 1:1 compared to municipal- only feeds. These findings should be of interest to municipalities with starch-producing food industries that face increasing cost and nutrient regulation pressures.

Conclusions

A number of conclusions can be drawn from this investigation. Acid-phase conditions were established in the reactor and methanogenesis was suppressed, as indicated by the continual production of VFAs, an acidic pH, and negligible gas production. It was noted that the larger the starch-rich industrial component of the mixture, the lower the pH. For the conditions studied in this research, specific VFA and SCOD production rates were maximized in the range of industrial-to-municipal ratios of 1:3 to 3:1. Net VFA production was also maximized within this range, but VSS reduction continued to increase with increasing starch content.

An increase in the industrial waste content of the feed was accompanied by a drop in the production of acetic and propionic acids, followed by a substantial increase in butyric acid production. The reactor containing 100% industrial feed did not reach steady-state operating conditions, most probably because of the excellent settling characteristics of the starch wastewater. This made SRT control difficult for that particular run. Finally, replication of this research appears to be readily feasible.

Acknowledgments

Credits. Financial support for this project was provided by the Natural Sciences and Engineering Research Council of Canada. The technical assistance of Ms. J. Tingley of the Environmental Engineering Laboratory at the University of Manitoba, Winnipeg, is gratefully acknowledged, as is the assistance of Ms. I. Maharaj.

References

Abeling, U.; Seyfried, C. F. (1993) Anaerobic-Aerobic Treatment of Potato-Starch Wastewater. Water Sd. Technol, 28 (2), 165.

American Public Health Association; American Water Works Association; Water Environment Federation (1998) Standard Methods for the Examination of Water and Wastewater, 20th Ed.; Washington, D.C.

Banerjee, A. K.; Elefsiniotis, P.; Tuhtar, D. (1999) The Effect of Addition of Potato-Processing Wastewater on the Acidogenesis of Primary Sludge under Varied Hydraulic Retention Time and Temperature, J. Biotechnoi, 72, 203.

Breure, A. M.; Beeftink, H. H.; Verkuijlen, J; van Andel, J. G. (1986a) Acidogenic Fermentation of Protein/Carbohydrate Mixtures by Bacterial Populations Adapted to One of the Substrates in Anaerobic Chemostat Cultures. Appl. Microbiol. Biotechnol., 23, 245.

Breure, A. M.; Beeftink, H. H.; Verkuijlen, J; van Andel, J. G. (1986b) Protein Degradation in Anaerobic Digestion: Influence of Volatile Fatty Acids and Carbohydrates on Hydrolysis and Acidogenic Fermentation of Gelatin. Appl. Microbiol. Biotechnol, 24, 426.

Chyi, Y. T.; Dague, R. R. (1994) Effects of Paniculate Size in Anaerobic Acidogenesis Using Cellulose as a Sole Carbon Source. Water Environ. Res., 66 (5), 670.

Elefsiniotis, P.; Oldham, W. K. (1994a) Effect of HRT on Acidogenic Digestion of Primary Sludge. J. Environ. Eng., 120 (3), 645.

Elefsiniotis, P.; Oldham, W. K. (1994b) Anaerobic Acidogenesis of Primary Sludge: The Role of Solids Retention Time. Biotechnol. Bioeng., 44, 7.

Elefsiniotis, P.; Wareham, D. G.; Oldha\m, W. K. (1996) Paniculate Organic Carbon Solubilization in an Acid-Phase Upflow Anaerobic Sludge Blanket System, Environ. Sd. Technol., 30, 1508.

Fongastitkul, P.; Mavinic, D. S.; Lo, K. V. (1994) A Two-Phased Anaerobic Digestion Process: Concept, Process Failure and Maximum System Loading Rate. Water Environ. Res., 66 (3), 243.

Fox, P.; Pohland, F. G. (1994) Anaerobic Treatment Applications and Fundamentals: Substrate Specificity During Phase Separation. Water Environ. Res., 66 (5), 716.

Gottschalk, G. (1986) Bacterial Metabolism, 2nd ed.; Springer- Verlag: New York, 359.

Hall, E. R. (1992) Anaerobic Treatment of Wastewater in Suspended Growth and Fixed Film Processes, In Design of Anaerobic Processes for the Treatment of Industrial and Municipal Wastes, Malina, J. F., Jr.; Pohland, F. G. (Eds.) Technomic Publishing Company, Inc.: Lancaster, Pennsylvania, 41.

Jeyaseelan, S.; Matsuo, T. (1995) Effects of Phase Separation in Anaerobic Digestion on Different Substrates. Water Sd. Technol., 31 (9), 153.

Kayhanian, M.; Tchobanoglous, G. (1992) Pilot Investigations of an Innovative Two-Stage Anaerobic Digestion and Aerobic Composting Process for the Recovery of Energy and Compost from the Organic Fraction of MSW. Proceedings of the International Symposium on Anaerobic Digestion of Solid Waste, Venice, Italy, April 14-17, 181.

Kugelman, I. J.; Guida, V. G. (1989) Comparative Evaluation of Mesophilic and Thermophilic Anaerobic Digestion. Re. No. EPA/600/S2- 89/001, Cincinnati, Ohio.

Kwong, T. S.; Fang, H. H. P (1996) Anaerobic Degradation of Cornstarch in Wastewater in Two Upflow Reactors. J. Environ. Eng., 122 (1), 9.

Maharaj, I. J.; Elefsiniotis, P. (2001) The Role of HRT and Low Temperature on the Acid-Phase Anaerobic Digestion of Municipal and Industrial Wastewaters, Biores. Technol., 76, 191. Oldham, W. K.; Abraham, K.; Dawson, R. N.; McGeachie, G. (1994) Primary Sludge Fermentation Design and Optimization for Biological Nutrient Removal Plants. In Nutrient Removal from Wastewaters, Horan, N. J.; Lowe, P.; Stentiford, E. I. (Eds.) Technomic Publishing Co., Inc.: Lancaster, Pennsylvania, 187.

Skalsky, D. S.; Daigger, G. T. (1995) Wastewater Solids Fermentation for Volatile Acid Production and Enhanced Biological Phosphorous Removal. Water Environ. Res., 67 (2), 230.

Smith, M. O. (2002) Acid-Phase Anaerobic Digestion: Effect of Varying Industrial-to-Municipal Feed Ratio and Suitability of Effluent for Batch-Culture Denitrification. M. Sc. Thesis. University of Manitoba, Winnipeg, Canada, October.

Valentini, A.; Garuti, G.; Rozzi, A; Tilche, A. (1997) Anaerobic Degradation Kinetics of Paniculate Organic Matter: A New Approach. Water Sd. Technol., 36 (6/7), 239.

Authors. Panagiotis Elefsiniotis is a senior lecturer with the Department of Civil and Environmental Engineering at the University of Auckland, Auckland, New Zealand. At the time of this research, he was an associate professor in the Department of Civil and Geological Engineering at the University of Manitoba, Winnipeg, Canada. David G. Wareham is a senior lecturer with the Department of Civil Engineering, University of Canterbury, Christchurch, New Zealand. Marcus O. Smith is an environmental consultant with Wardrop Engineering Inc., Winnipeg, Canada. At the time of this research, he was a M.Sc. candidate in the Department of Civil and Geological Engineering at the University of Manitoba, Winnipeg, Manitoba, Canada. Correspondence should be addressed to Dr. P. Elefsiniotis, Department of Civil and Environmental Engineering, Private Bag 92019, Auckland, New Zealand; e-mail: t.elefsiniotis@auckland.ac.nz.

Submitted for publication March 3, 2003; revised manuscript submitted January 28, 2004; accepted for publication February 19, 2004.

The deadline to submit Discussions of this paper is December 15, 2005.

Copyright Water Environment Federation Jul/Aug 2005

Source: Water Environment Research

More News in this Category