a kinetic study of anaerobic digestion of olive mill wastewater at mesophilic and thermophilic...

6
Environmental Pollution 88 (1995) 13-18 0 1995 Elsevier Science Limited Printed in Great Britain. All rights reserved ELSEVIER 0269-7491/95/$09.50 A KINETIC STUDY OF ANAEROBIC DIGESTION OF OLIVE MILL WASTEWATER AT MESOPHILIC AND THERMOPHILIC TEMPERATURES R. Borja,’ A. Martiqb C. J. Banks,c V. Alonsob & A. Chicab “Institute de la Grasa y sus Derivados (C.S.I.C.), Avda. Padre Garcia Tejero, 4, 41012-Sevilla, Spain bDepartamento de Ingenieria Quimica, Facultad de Ciencias, Avda. San Albert0 Magno s/n. 14004-Cdrdoba, Spain ‘Environmental Technology Centre, Department of Chemical Engineering, University of Manchester Institute of Science and Technology (UMIST), PO Box 88, Manchester UK, M60 I QD (Received 18 February 1994; accepted 28 July 1994) Abstract The kinetics of the anaerobic digestion of olive mill wastewater (OM W) was studied in the mesophilic and thermophilic ranges of temperature. Two completely mixed continuous Jlow bioreactors operating at 35’C and 55°C and with an average biomass concentration of 5.45 g VSS litre-’ were used. The thermophilic reactor worked satisfactorily between hydraulic retention times (HRT) of 10 to 40 days, removing between 94.6 and 84.4% of the initial chemical oxygen demand (COD). In contrast, the mesophilic reactor showed a marked decrease in sub- strate utilization and methane production at a HRT of 10 days. TVFA levels and the TVFA/alkalinity ratio were higher and close to the suggested limits for digester failure. The yield coeficient for methane production (1 CH, STP g-’ COD,,,,,! was 28% higher in the ther- mophilic process than in the mesophilic one. Macroenergetic parameters, calculated using Guiot ‘s kinetic model, gave yield coejicients for the biomass ( Y) of 0.18 (mesophilic) and 0.06 g VSS g-’ COD (ther- mophilic) and spectfic rates of substrate uptake for cell maintenance (m) of 0.12 (mesophilic) and 0.27 g COD g-’ VSS.day .’ (thermophilic). The experimental results showed the rate of substrate uptake (R,; g COD g-’ VSS.day-‘ ), correlated with the concentration of biodegradable substrate (Sb; g COD litre-‘ ), through an equation of the Michaelis-Menten type for the two temperatures used. Keywords: anaerobic digestion, olive mill wastewater, kinetics, mesophilic and thermophilic temperatures. NOTATION COD Chemical oxygen demand (g litre-‘). HRT Hydraulic retention time (days). K, K, Constants of the Michaelis equation (K, g COD g-’ VSSday-‘; K,, g COD litre-‘). m Specific rate of substrate uptake for cell main- tenance (g COD g-’ VSS.day-‘). MS Mineral solids (g litre-‘). MSS OMW 4G RG sb so, s Snon-biod STP t TVFA TS TSS V vs vss W W, x Y Mineral suspended solids (g litre-‘). Olive mill wastewater. Flow-rate of produced methane (litre CH, day- I). Rate of methane production (g COD g VSSday-‘). Rate of substrate uptake per unit mass of microorganisms in the reactor (g COD g VSS.day-I). Concentration of biodegradable substrate (g COD litre-‘). Incoming and outgoing substrate concentra- tions (g COD litre~‘). Concentration of non biodegradable substrate (g COD litre-‘). Standard temperature and pressure conditions. Time. Total volatile fatty acids (g litre ‘). Total solids (g litre- ‘). Total suspended solids (g litre-‘). Reactor volume (litre). Volatile solids (g litre- ). Volatile suspended solids (g litre-‘). Observed factor for the conversion of COD into biomass (g COD g-’ VSS). Methane equivalent of COD (g COD litre-’ CH,). Biomass concentration (g VSS litre-‘). Yield coefficient for the biomass (g VSS go’ COD). INTRODUCTION The manufacturing process of olive oil typically pro- duces an oily phase (20%), a solid residue (30%) and an aqueous phase (50%); the latter arises from the water content of the fruit and, when combined with that used to wash and process the olives, makes up the so-called ‘olive mill wastewater’ (OMW). This also contains soft tissues from olive pulp and a very stable oil emulsion. The use of a continuous process for oil extraction has been commonly used since 1972 and yields OMW 13

Upload: r-borja

Post on 22-Nov-2016

216 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: A kinetic study of anaerobic digestion of olive mill wastewater at mesophilic and thermophilic temperatures

Environmental Pollution 88 (1995) 13-18 0 1995 Elsevier Science Limited

Printed in Great Britain. All rights reserved

ELSEVIER 0269-7491/95/$09.50

A KINETIC STUDY OF ANAEROBIC DIGESTION OF OLIVE MILL WASTEWATER AT MESOPHILIC AND THERMOPHILIC

TEMPERATURES

R. Borja,’ A. Martiqb C. J. Banks,c V. Alonsob & A. Chicab “Institute de la Grasa y sus Derivados (C.S.I.C.), Avda. Padre Garcia Tejero, 4, 41012-Sevilla, Spain

bDepartamento de Ingenieria Quimica, Facultad de Ciencias, Avda. San Albert0 Magno s/n. 14004-Cdrdoba, Spain

‘Environmental Technology Centre, Department of Chemical Engineering, University of Manchester Institute of Science and Technology (UMIST), PO Box 88, Manchester UK, M60 I QD

(Received 18 February 1994; accepted 28 July 1994)

Abstract

The kinetics of the anaerobic digestion of olive mill wastewater (OM W) was studied in the mesophilic and thermophilic ranges of temperature. Two completely mixed continuous Jlow bioreactors operating at 35’C and 55°C and with an average biomass concentration of 5.45 g VSS litre-’ were used. The thermophilic reactor worked satisfactorily between hydraulic retention times (HRT) of 10 to 40 days, removing between 94.6 and 84.4% of the initial chemical oxygen demand (COD). In contrast, the mesophilic reactor showed a marked decrease in sub- strate utilization and methane production at a HRT of 10 days. TVFA levels and the TVFA/alkalinity ratio were higher and close to the suggested limits for digester failure. The yield coeficient for methane production (1 CH, STP g-’ COD,,,,,! was 28% higher in the ther- mophilic process than in the mesophilic one.

Macroenergetic parameters, calculated using Guiot ‘s kinetic model, gave yield coejicients for the biomass ( Y) of 0.18 (mesophilic) and 0.06 g VSS g-’ COD (ther- mophilic) and spectfic rates of substrate uptake for cell maintenance (m) of 0.12 (mesophilic) and 0.27 g COD g-’ VSS.day .’ (thermophilic).

The experimental results showed the rate of substrate uptake (R,; g COD g-’ VSS.day-‘), correlated with the concentration of biodegradable substrate (Sb; g COD litre-‘), through an equation of the Michaelis-Menten type for the two temperatures used.

Keywords: anaerobic digestion, olive mill wastewater, kinetics, mesophilic and thermophilic temperatures.

NOTATION

COD Chemical oxygen demand (g litre-‘). HRT Hydraulic retention time (days). K, K, Constants of the Michaelis equation (K, g

COD g-’ VSSday-‘; K,, g COD litre-‘). m Specific rate of substrate uptake for cell main-

tenance (g COD g-’ VSS.day-‘). MS Mineral solids (g litre-‘).

MSS OMW 4G

RG

sb

so, s

Snon-biod

STP t TVFA TS TSS V vs vss W

W,

x Y

Mineral suspended solids (g litre-‘). Olive mill wastewater. Flow-rate of produced methane (litre CH, day- I). Rate of methane production (g COD g ’ VSSday-‘). Rate of substrate uptake per unit mass of microorganisms in the reactor (g COD g ’ VSS.day-I). Concentration of biodegradable substrate (g COD litre-‘). Incoming and outgoing substrate concentra- tions (g COD litre~‘). Concentration of non biodegradable substrate (g COD litre-‘). Standard temperature and pressure conditions. Time. Total volatile fatty acids (g litre ‘). Total solids (g litre- ‘). Total suspended solids (g litre-‘). Reactor volume (litre). Volatile solids (g litre- ‘). Volatile suspended solids (g litre-‘). Observed factor for the conversion of COD into biomass (g COD g-’ VSS). Methane equivalent of COD (g COD litre-’ CH,). Biomass concentration (g VSS litre-‘). Yield coefficient for the biomass (g VSS go’ COD).

INTRODUCTION

The manufacturing process of olive oil typically pro- duces an oily phase (20%), a solid residue (30%) and an aqueous phase (50%); the latter arises from the water content of the fruit and, when combined with that used to wash and process the olives, makes up the so-called ‘olive mill wastewater’ (OMW). This also contains soft tissues from olive pulp and a very stable oil emulsion.

The use of a continuous process for oil extraction has been commonly used since 1972 and yields OMW

13

Page 2: A kinetic study of anaerobic digestion of olive mill wastewater at mesophilic and thermophilic temperatures

14 R. Borja, A. Martin, C. J. Banks, V. Alonso, A. Chica

at about 40°C (Alba et al., 1982). The warm waste with its high organic load make anaerobic digestion a feasi- ble alternative for its purification both at mesophilic (around 35°C) and thermophilic (around SY’C) temper- atures.

Comparisons of organic matter removal efficiency in anaerobic digesters treating different types of high strength wastes at mesophilic and thermophilic temper- atures are not uncommon (Hashimoto, 1983; Kandler, 1983; Dhavises et al., 1985; Sriprasertsak et al., 1985; Vincent et al., 1985; Demharter & Pfeiffer, 1988; Lema et al., 1988; Marchaim, 1988; Sales et al., 1988). The general conclusion is that operation at thermophilic temperatures offers several potential advantages, (Buhr, 1977), which include: increased reaction rates with respect to the destruction of organic solids; increased efficiency with respect to the fraction of organic solids destroyed; improved solids-liquid separation and in- creased destruction of pathogenic organisms. Disad- vantages, on the other hand, may include: high energy requirement for heating and poor process stability. Where the wastewater is warm, as is the case with OMW, the first of these is nullified as the small ener- getic difference can easily be supplied by the produced gas.

The aim of the work was to compare anaerobic digestion of OMW at both thermophilic and mesophilic temperatures. This was achieved by evaluating the kinetic constants of the processes, the volumetric rates of substrate uptake, percentages of organic matter removed and yield coefficients of methane production.

MATERIALS AND METHODS

Equipment Two completely-mixed bioreactors, consisting of two 1 . 1Zlitre glass vessels with an integral settling zone designed to prevent loss of the biomass in the effluent, were used. The settling arrangement also allowed for venting biogas and formed an air lock to prevent air from entering the reactor. The reactors were maintained in thermostatic chambers, one at 35°C (mesophilic reac- tor) the other at 55°C (thermophilic reactor). The reactors were fed continuously by means of a peristaltic pump. The biogas produced was passed through a solution of NaOH to remove CO2 yielding directly the volume of methane produced, this was then collected by water displacement. The biomass was agitated with the aid of a mechanical stirring system working at 120 rpm.

Olive mill wastewater The OMW was obtained from a factory using a contin- uous process. The properties of the wastewater are given in Table 1, the values given are the means of four determinations for each parameter.

Inoculum The reactors were inoculated with biomass from an olive mill wastewater storage and evaporation pond after some dilution and neutralization; this was neces-

Table 1. Features ad composition of Olive Mill Was&water (OMW) from a continuous production plant0

PH 5.2 Soluble chemical oxygen demand (COD) (g litre-‘) 52.5 TS (g litre-I) 45.0 MS (g litre-‘) 11.0 VS (g litre-‘) 34.5 TSS (g litre-‘) 13.8 MSS (g litre-‘) 3.8 VSS (g litre-‘) 10.0 Volatile acidity (acetic acid, g litre-‘) 0.15 Alkalinity (CaCO,, g litre-‘) 0.60 Nitrogen (NH,+, mg litre-‘) 90 Total phenolic compounds (c&eic acid, mg litre-‘) 750 o-diphenols (caffeic acid, mg litre-‘) 95

‘Values are averages of four determinations; the differences between the observed values were less than 3% in all cases.

sary so as to avoid the highly acidic conditions which result from anaerobic fermentations within the pond. The contents in total (TSS), volatile (VSS) and mineral suspended solids (MSS) of the inoculum was: 13.0, 10.4 and 2.6 g litre-‘, respectively.

Experimental procedure The bioreactors were initially charged with 750 ml of distilled water and 250 ml of inoculum. Acclimatisation of the digesters was achieved by incrementing the vol- ume of OMW loaded from 10 to 100 ml over a two- month period. During this time the digesters were operated in a fill and draw mode with fresh feed being added when gas production had stopped.

This preliminary acclimatisation step was followed by a series of continuous flow experiments using feed flow-rates of filtered wastewater of between 28 and 112 ml day’; these correspond to hydraulic retention times (HRT) in the range 40 to 10 days. Each experiment was run for 20 days to ensure steady-state conditions were achieved. The biomass concentration (estimated according to the recommendations of Chen et al., 1985) in both reactors ranged between 5.4 and 5.5 g VSS litre-r (mean value = 5.45 g litre-‘) and was virtually identical during the course of the experiments.

Steady-state was assumed when the substrate concen- tration in the digester did not vary by more than lt2% when measured daily over a period of 5-10 days. Dur- ing these periods, daily methane production, volatile fatty acids, pH and chemical oxygen demand (COD) of the effluents were measured.

Chemical analyses The effluent characteristics were determined using Stan- dard Methods for the Examination of Water and Wastewater (APHA, 1985).

The total phenol content was determined by the Folin-Ciocalteau method, while odiphenols were assayed with sodium molybdate and sodium nitrite (Vazquez et al., 1974). Volatile fatty acids were determined by gas chromatography (Martin et al., 1993).

Page 3: A kinetic study of anaerobic digestion of olive mill wastewater at mesophilic and thermophilic temperatures

Anaerobic digestion of olive mill wastewater 15

Table 2. Methane production, etlhent COD and reactor pH for the various HRTs used at the two temperatures”

HRT (days)

Mesophilic Thermophilic

Methane COD Methane COD (litre day-‘) (g litre ‘) (htre day-‘) (g litre-‘)

40 0.355 3.95 0.430 2.85 35 0.405 4.20 0.490 3.00 30 0.475 4.45 0,580 3.20 25 0.565 4.95 0.680 3.50 20 0.710 5.80 0.830 4.05 15 0.790 8.05 1.050 5.05 10 0.650 24.15 1.460 8.20

“Methane production and effluent COD are averages of six determinations on samples taken over six days at the end of each experimental run. The differences between the values observed were less than 2% and 3%, respectively.

RESULTS AND DISCUSSION

Operational parameters The results obtained from the reactors at mesophilic and thermophilic temperatures, under steady-state con- ditions for different HRT, are summarized in Table 2. The pH in both reactors remained within the optimal working range for anaerobic digesters (66-7.8) at HRTs between 15 and 40 days. However, an HRT of 10 days caused acidification in both reactors, an increased COD content in the efffuents, and decreased methane production; these were more accentuated in the mesophilic reactor. Figure 1 shows the variation of the total volatile fatty acid concentration (TVFA) as a function of the HRT. It can be seen that pH values re- mained virtually constant for HRT values over 15 days for both digesters, whilst TVFA concentration in-

450

TVFA G 350

(ms/t)

250

2m_ u

creased with a decreasing HRT’s. At an HRT less than 15 days the TVFA concentration increased sharply, this was concomitant with the decrease in pH, and was more pronounced in the mesophilic reactor. Although complete failure was not observed in the mesophilic process, mean production was severely depressed at the short HRT (Table 2). In contrast, high gas yields and process stability were always observed in the ther- mophilic process.

On the other hand, between HRT’s of 15 to 40 days the TVFAalkalinity ratio was found to be constant (0.10 and 0,12 for the mesophilic and thermophilic pro- cesses, respectively) and lower than the failure limit (0.3-0.4) value (Fannin, 1987). However, at a HRT of 10 days, a considerable increase of this ratio was observed in the mesophilic reactor (0,225) but not in the thermophilic reactor (0.105).

One of the major contributing factors to the failure of the mesophilic digester at a 10 day HRT is the build up of longer chain volatile fatty acids. This is indicative of carbon flow through to methane being interrupted by an inhibition of the hydrogen producing acetogenic bacteria, which are primarily responsible for the break- down of these compounds to acetic acid. If hydrogen is not effectively being removed from solution by the activity of the autotrophic methanogens then inhibition of this reaction takes place. Failure at the IO-day HRT in the mesophilic reactor might therefore be attributed directly to a failure at some point in this chain of reac- tions, either directly or by feedback inhibition.

Biodegradability Between 15 and 40 days, COD removal decreased slightly from 92.5 to 84.7% for the mesophilic reactor and 94.6 to 90.4% for the thermophilic reactor. At an HRT of 10 days a marked decrease in efficiency was

/ I , I I I / / I

1

5 10 15 20 25 30 35 40 45 50

HRT (days)

9

6

7

PH CJ

6

Fig. 1. Variation of the total volatile fatty acid (TVFA) concentration (0, mesophilic; 0, thermophilic) and pH values (V, mesophilic; v, thermophilic) with hydraulic retention time (HRT).

Page 4: A kinetic study of anaerobic digestion of olive mill wastewater at mesophilic and thermophilic temperatures

16 R. Borja, A. Martin, C. J. Banks, V. Alonso, A. Chica

In S

(g COD/I)

2.5

1lHRT (days-l)

Fig. 2. Estimation of the fraction of non-biodegradable or- ganic matter contained in the waste. (@, mesophilic process;

v, thermophilic process).

observed in the mesophilic reactor (54%), but the decrease was less severe in the thermophilic process (84.4%). In addition, the thermophilic process gave slightly higher reductions than the mesophilic process and this difference increased with decreasing HRT. COD’s do not distinguish between biodegradable and non-biodegradable substrate, but considering the com- plexity of the wastewater treated, it is likely that a frac- tion of this was not biodegradable; this has been estimated in order to perform the subsequent calcula- tions. Figure 2 shows a graphical estimation of the amount of non-biodegradable substrate based on the linear relationship between In (S)experimental and l/(HRT) (Martin et al., 1991). Least-squares fitting of the two variables for both data sets gave an intercept of In(S) = 0.69, which corresponds to an infinite HRT, giving the concentration of non-biodegradable substrate (Snon_biod, = 2 g COD litre-‘).

Methane yield coefficient The experimental data listed in Table 2 and the influent substrate concentration were used to determine the methane yield coefficient for both cases. By fitting the (daily methane production, g COD added) value pairs to a straight line, the average yield coefficients under standard temperature and pressure (STP) conditions were found to be 0.29 and 0.23 litre CH, STP g-l CODadded for the thermophilic and mesophilic reactors, respectively.

Determination of the macroenergetic parameters Since biomass is retained in the bioreactors, the hy- draulic retention time is short relative to the cellular re- tention time. Under these conditions, the system cannot reach a steady state with respect to the solid phase (re- actor biomass), yet steady-state conditions can be applied to both the liquid phase and the soluble substrate. However, the changes in the biomass concentration (X) over relatively long time intervals are negligible compared

to the biomass load of the reactor, thus they have no significant affects on the system dynamics. In this pseudo-steady-state, in solving the mass balances for the soluble and solid phases it is generally taken that dS/dt = 0 (i.e. that no substrate accumulation occurs) while dX/dt f 0. It can also be considered that the vol- umetric rate of biomass generation remains constant throughout the pseudo-steady-state interval (Guiot et al., 1989). Under these conditions, the following equa- tion is obtained (Guiot et al., 1989)

R, = (l-WY)R, + WmY (1)

where R, is the rate of methane production (g COD g-’ VSS.day’); R, is the rate of substrate uptake per unit mass of microorganisms in the reactor (g COD g-’ VSS.day’); Y is the true yield coefficient of the biomass (g VSS g-’ COD) and m is the specific rate of substrate uptake for cell maintenance (g COD g-r VSS.day-I). W is the observed factor for the conversion of COD into biomass (g COD g-r VSS). The value of this factor W = 1.41 g COD g-’ VSS was taken from the literature (Guiot et al., 1989). R, and RG were calculated using the following equations

R, = (SO - S)/[X(HRT)] (2)

R, = W,qdXV (3)

where S, and S are the incoming and outgoing substrate con- centrations (g COD lit&), qc is the flow rate of produced methane (1 CI-I,, day’), V is the reactor volume and W, is the methane equivalent of COD. The values of W, were caleu- lated from the methane production rate and the removed COD (Table 2), and were determined to be 3.6 g COD lit& CI-I.+ and 3.1 g COD line’ CH, for the mesophilic and ther- mophilic process, respectively.

As W, Y and m are constant for a given system, plotting R, against R, will yield a straight line. Even though the working conditions are not strictly steady-state, the slow evolution of the systems allows this assumption to be made. Figure 3 illustrates such a plot and as the

0.9, , , , , , , , I ,

0.8

= d

0.7

ti 0.6 P ,” 0.5 0” 0 0.4 3 & 0.3

0.2

0.1

0.0’ ’ ’ ’ ’ ’ ’ ’ ’ ’ 0.0 0.1 0.2 Lt.3 0.4 0.5 0.6 0.7 0.0 0.9

R, (g COD/ g VSS. day)

Fig. 3. Variation in the rate of methane production, R,, as a function of the rate of substrate uptake, R,. (0. mesophilic

process; v, thermophilic process).

Page 5: A kinetic study of anaerobic digestion of olive mill wastewater at mesophilic and thermophilic temperatures

Anaerobic digestion of olive mill wastewater 17

P 0.8 8 0 s li?

0.4

0.2

0.0 0 12 3 4 5 6 7

s, (gcoofl)

Fig. 4. Variation of the rate of substrate uptake, R,, as a function of the concentration of biodegradable substrate, S,,.

(0, mesophilic process; v, thermophilic process).

points fit a straight line for the two temperatures stud- ied, this strongly suggests the validity of the proposed model. By fitting the data to a linear function, using the least-squares method, the slope and intercept of the line can be calculated. Once W was known, the macroenergetic parameters Y and m could be deter- mined and gave values of Y = 0.18 g VSS g-’ COD (mesophilic) and 0.06 g VSS g-’ COD (thermophilic) and m = 0.12 g COD g ’ VSS.day-‘(mesophilic) and 0.27 g COD g ’ VSS.day ’ (thermophilic). These values are comparable to those reported in the literature (Young & McCarty, 1967; Lawrence & McCarty, 1969; Van den Berg, 1977; Shieh et al., 1985; Yoda et al., 1987; Henze & Harramoes, 1989; Borja et al., 1993; Martin et al., 1993).

Derivation of the kinetic equation In the above equations, S denotes the concentration of biodegradable substrate. Thus, the experimental values given in Table 2 must be corrected by subtracting the fraction of non-biodegradable substrate (2 g COD litre-‘, in both cases).

Figure 4 shows a plot of R, vs S, pairs for the mesophilic and thermophilic reactors, where St, is the concentration of biodegradable substrate (g COD 1itre-I). The shape of the curves suggests that both processes conform to a Michaelis-Menten type equation

R, = KS&K, + S,) (4)

which is confirmed by the straight lines obtained by plotting l/R, against l/S, (Fig. 5)

l/R, = (l/K) + (K,IK)(lIS,) (5)

By using a weighted least-squares parameter-estima- tion procedure in a non-linear multivariate model (Valko & Vajda, 1989) the following values for the kinetic parameters K and K, in eqn (4) were obtained: K = 0.69 g COD g-’ VSS.day-’ (mesophilic) and 1.45 g

.5

0.0 ' I I I I I I 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

11s b (g COD/l) -’

Fig. 5. Lineweaver-Burk plot of rate of substrate uptake per unit mass microorganisms against biodegradable substrate concentration, expressed as reciprocal values. (0, mesophilic

process; v, thermophilic process).

COD g-’ VSS.day-’ (thermophilic) and K, = 2.7 g COD litre-’ and 4.3 g COD litre-‘. Substitution of these values into eqn (4) allowed the theoretical rate of substrate uptake to be determined. The deviations obtained were lower than 10% in all cases, which suggests that the proposed model can be used to predict the behaviour of these types of reactors very accurately.

ACKNOWLEDGEMENTS

The authors wish to express their gratitude to the ‘De- partamento de Postgrado y Especializacion de1 C.S.I.C. (Spain)’ and ‘Junta de Andalucia’ for their financial support.

REFERENCES

Alba, J., Mufioz, E. & Martinez, J. (1982). Obtencion de aceite de oliva. Empleo de productos que facilitan su ex- traccicn. Alimentaria, 25-55.

American Public Health Association. (1985). Standard Meth- ods for the Examination of Water and Wastewater (16th edn), APHA, Washington, D.C., USA.

Bolte, J. P., Hill, D. T. Wood, T. H. (1985). Anaerobic diges- tion of screened swine waste liquids in suspended particle- attached growth reactors. ASAE, Paper no. 85-3085.

Borja, R., Martin, A. Garrido, A. (1993). Anaerobic diges- tion of black-olive wastewater. Biores. Technol., 45, 27-32.

Buhr, H. 0. (1977). The thermophic anaerobic digestion pro- cess. Water Res., 11, 129-43.

Chen, S. J., Li, C. T. & Shieh, W. K. (1985). Performance evaluation of the anaerobic fluid&d bed system: II. Biomass holdup and characteristics. J. Chem. Technol. & Biotechnol., 35, 183-90.

Demharter, W. & PfeifIer, W. (1988). Sewage sludge digestion and disiiection. A comparison of mesopbilic and thermophilic pro- cesses. Proc. 5th Int. Symp. on Anaerobic Digestion, Bologna, Italy, ed. A. Tilche & A. Roti, Monduzi, 699702.

Dhavises, G., Sriprasersak, P., Tanaka, T., Taniguchi, M. & Oi, S. (1985). Mesophilic and thermophilic methane fer- mentation of agrowastes and grasses. J. Ferment. Technol., 63, 45-9.

Page 6: A kinetic study of anaerobic digestion of olive mill wastewater at mesophilic and thermophilic temperatures

18 R. Borja, A. Martin, C. J. Banks, V. Alonso, A. Chica

Fannin, K. F. (1987). Start-up, operation, stability and con- Martin, A., Borja, R. & Chica, A. (1993). Kinetic study of an trol. In Anaerobic digestion of Biomass, ed. D. P. anaerobic fluidized bed system used for the purification of Chynoweth & R. Isaacson. Elsevier Applied Science, Lon- fermented olive mill wastewater. J. Chem. Technol. & don, pp. 171-96. Biotechnol., 56, 155-62.

Guiot, S. R., Podruzny, M. F. & McLean, D. D. (1989). As- sessment of macroenergetic parameters for an anaerobic upflow biomass bed and filter (UFB) reactor. Biotechnol. & Bioengng, 34, 1277-88.

Hashimoto, A. G. (1983). Conversion of straw-manure mix- tures to methane at mesophilic and thermophilic tempera- tures. Biotechnol. & Bioengng, 25, 185-200.

Henze, M. Harremoes, H. (1989). Literature review: Anaero- bic treatment of wastewater in fixed film reactors. IA WPR, Specialized Seminar, 16-18 June Copenhagen, Denmark, pp. l-94.

Kandler, 0. (1983). Efficiency and stability of methane fer- mentation of wastes at mesophilic and thermophilic tem- peratures. CHEMTECH Symp. on Recent Advances in Biotechnology, 18(11), 57-65.

Lawrence, A. L. & McCarty, P. L. (1969). Kinetics of methane fermentation in anaerobic treatment. J. WPCF, 41, Rl-R7.

Sales, D., Romero, L. I., Valcarcel, M. J., Perez, L. & Mar- tinez de la Ossa, E. (1988). Thermophilic and mesophilic anaerobic digestion of wine-distillery wastewaters. Proc. 5th Znt. Symp. on Anaerobic Digestion, Bologna, Italy, ed. A. Tilche & A. Rozzi, Monduzzi, 575-8.

Shieh, W. K., Chun, T. L. & Chen, S. J. (1985). Performance evaluation of the anaerobic fluidised bed system: III. Pro- cess Kinetics. J. Chem. Technol. & Biotechnol,, 35, 229-34.

Sriprasertsak, P., Dhavises, G. & Oi, S. (1985). Mesophilic and thermophilic methane fermentations of slop waste. J. Ferment. Technol., 63(6), 567-73.

Valko, P. & Vajda, S. (1989). Advanced Scient@c Computing in Basic with Applications in Chemistry, Biology and Phar- macology. Elsevier, Amsterdam, The Netherlands.

Van den Berg, L. (1977). Effect of temperature on growth and activity of methanogenic culture utilizing acetate. Can. J. Microbial., 23, 898-902.

Lema, J. M., Soto, M., Mendez, R. & Blazquez, R. (1988). Comparison of mesophilic and thermophilic anaerobic fil- ters treating very high saline wastewater. Proc. 5th Znt. Symp. on Anaerobic Digestion, Bologna,.Italy, ed. A. Tilche & A. Rozzi, Monduzzi, 547-9.

Vazquez, A., Maestro, R. & Graciani, E. (1974). Compo- nentes fenolicos de la aceituna. II. Polifenoles de1 alpechin. Grasas y Aceites, 25, 341-5.

Marchaim, U. (1988). Thermophilic anaerobic digestion sys- tern for slaughterhouse wastes: An economical and ecologi- cal solution. Proc. 5th Int. Symp. on Anaerobic Digestion, Bologna, Italy, ed. A. Tilche & A. Rozzi, Monduzzi, 555-6.

Martin, A., Borja, R., Garcia, I. & Fiestas, J. A. (1991). Ki- netics of methane production from olive mill wastewater. Process Biochem., 26, 101-7.

Vicent, T., Paris, J. M., Lema, J. M. & Ibanez, E. (1985). Thermophilic anaerobic treatment of an industrial wastew- ater: Start-up and stability studies. Biotechnol. & Bioengng, 15, 599-609.

Yoda, M., Shin, S. W., Watanabe, A., Kitagawa, M. & Miyaji, Y. (1987). Anaerobic fluidized bed treatment with a steady biofilm. Water Sci. Technol., 19, 287-98.

Young, J. C. & McCarty, P. L. (1967). The anaerobic filter for waste treatment. Proc. 22nd Znd. Waste ConjI, Purdue University, Lafayette, Indiana, pp. 559-74.