Composting in small laboratory pilots: Performance and reproducibility
G. Lashermes
a,1, E. Barriuso
a, M. Le Villio-Poitrenaud
b, S. Houot
a,⇑aINRA, UMR1091 Environment and Arable Crops (INRA, AgroParisTech), F-78850 Thiverval-Grignon, France
bVEOLIA Environment – Research and Innovation, F-78520 Limay, France
a r t i c l e i n f o
Article history:
Received 19 May 2011 Accepted 6 September 2011 Available online 6 October 2011
Keywords:
Compost
Biochemical fractionation Pilot
In-vessel composting reactor Sludge
a b s t r a c t
Small-scale reactors (<10 l) have been employed in composting research, but few attempts have assessed the performance of composting considering the transformations of organic matter. Moreover, composting at small scales is often performed by imposing a fixed temperature, thus creating artificial conditions, and the reproducibility of composting has rarely been reported. The objectives of this study are to design an innovative small-scale composting device safeguarding self-heating to drive the composting process and to assess the performance and reproducibility of composting in small-scale pilots. The experimental setup included six 4-l reactors used for composting a mixture of sewage sludge and green wastes. The performance of the process was assessed by monitoring the temperature, O2consumption and CO2emis- sions, and characterising the biochemical evolution of organic matter. A good reproducibility was found for the six replicates with coefficients of variation for all parameters generally lower than 19%. An intense self-heating ensured the existence of a spontaneous thermophilic phase in all reactors. The average loss of total organic matter (TOM) was 46% of the initial content. Compared to the initial mixture, the hot water soluble fraction decreased by 62%, the hemicellulose-like fraction by 68%, the cellulose-like fraction by 50% and the lignin-like fractions by 12% in the final compost. The TOM losses, compost stabilisation and evolution of the biochemical fractions were similar to observed in large reactors or on-site experi- ments, excluding the lignin degradation, which was less important than in full-scale systems. The repro- ducibility of the process and the quality of the final compost make it possible to propose the use of this experimental device for research requiring a mass reduction of the initial composted waste mixtures.
Ó2011 Elsevier Ltd. All rights reserved.
1. Introduction
Many studies have been conducted in pilot-scale reactors that enable easier tracking of the composting process than in full-scale plants (Mason and Milke, 2005). Large-scale reactors (10–300 l) frequently involve a self-heating phase, during which the compost temperature exceeds 60°C, depending solely on microbial heat production and ensuring a well-conducted composting process.
In such conditions, the simulation of the thermodynamic regime, including the thermophilic phase, cooling and maturation phases, should enable reproduction of many other parameters of full-scale composting systems, including biological activity and metabolic capacities (Ryckeboer et al., 2003; Sanz et al., 2006), moisture and water vapour transport, oxygen status and temperature (Mason and Milke, 2005).
Small-scale reactors (<10 l) have also been employed because they are easier to handle, less expensive and easier to control than large-scale reactors or full-scale systems (Petiot and de Guardia,
2004). They have been used to evaluate substrate compostability (Hu et al., 2009) or process suitability (Körner et al., 2003), define parameters for mathematical models (Sánchez Arias et al., 2011) and investigate the fate of specific compounds (Zenjari et al., 2006). Indeed, the miniaturisation of the process is required when the behaviour of pollutants is studied using radiolabeled chemicals due to the limited amount of necessary materials and the ability to control output gazes (Reid et al., 2002). Nevertheless, the experi- mental simulation of the composting at a small scale is not obvious because the mass of the organic matter involved in the process may not be large enough to reproduce heat generation and transfer and the resulting thermal inertia of full-scale systems (Mason and Milke, 2005). A small size of reactor may also limit potential sam- pling during the entire process (Hesnawi and McCartney, 2006). In small-volume reactors, a rapid decrease of temperature is usually observed because of the limited amounts of organic substrates and heat losses, contrasting with the slow and gradual decline in temperature of full-scale composting (Petiot and de Guardia, 2004).
On the other hand, very few studies have evaluated the realism of the composting process at a small scale, comparing the bio- chemical properties of organic matter before and after composting (Michel et al., 1995). Moreover, composting experiments at a small 0956-053X/$ - see front matterÓ2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.wasman.2011.09.011
⇑ Corresponding author. Tel.: +33 1 30 81 54 01; fax: +33 1 30 81 53 96.
E-mail address:[email protected](S. Houot).
1Present address: INRA, UMR614 Fractionation of AgroResources and Environment (INRA, URCA), F-51100 Reims, France.
Contents lists available atSciVerse ScienceDirect
Waste Management
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / w a s m a n
scale are often performed imposing a fixed temperature through- out the process that may create artificial or unrealistic conditions (Mason and Milke, 2005). Finally, the reproducibility of reactor- scale composting has been rarely reported (Schloss et al., 2000;
Petiot and de Guardia, 2004), although it conditions the potential use of the results for comparing different composting mixtures or conditions.
Thus, this work seeks to design an experimental new device to perform realistic composting while reducing the pilot volume and the mass of composted wastes and to assess the performance and the reproducibility of this small-scale composting system regard- ing the chemical and biochemical transformations of organic mat- ter. A small-scale reactor system that enables self-heating in six parallel replicates was created, and it was tested using a classical waste mixture of sewage sludge and green waste.
2. Materials and methods
2.1. Composting reactor system
The composting system included six parallel reactors (C1–C6) (Fig. 1) to measure the reproducibility of composting. The reactors were 4-l glass cylinders with seals encapsulated with fluorinated ethylene propylene Teflon. To compensate heat losses during compost self-heating due to the high surface-area to volume ratio of the reactor (Petiot and de Guardia, 2004), the wall temperature of each reactor was controlled. Thus, each reactor had an external jacket through which water circulated from a thermostatic bath (Körner et al., 2003). Reactors were also wrapped with glass wool for additional insulation. A steel perforated screen was installed 5 cm above the reactor bottom (Michel et al., 1995). Leachates were recovered in the bottom section and could be sampled through an opening. The organic mixture was placed above the screen in a steel basket that was removable for compost sampling and mixing. Compressed air provided aeration through an opening in the bottom section and was distributed through a gas disper- sion tube. The air flow rate was adjusted daily and measured using a volumetric gas counter. The moisture losses due to heat
production and aeration were compensated by heating the sup- plied air by submerging the air pipes in the water baths, then humidifying the air through two successive bottles filled with water at the temperature of the thermostatic bath (Sánchez Arias et al., 2011).
Two temperature sensors were inserted in the middle of the compost (T1,Fig. 1) and close to the reactor wall (T2). A third tem- perature sensor was placed in the thermostatic bath (T3). Temper- ature values were recorded every 5 min using a data acquisition system (CR23X, Campbell Scientific, Loughborough, UK).
The CO2and O2contents in the exhaust air of the reactors were measured for 20 min every 3 h using gas analyser based on CO2
infrared absorption and O2electrochemical cell (Cristal 300, COS- MA, Igny, France). The exhaust gas was cooled and filtered before analysis. Automatically controlled solenoid valves ensured the suc- cessive delivery of the outlet air of the six reactors to the gas analyser.
For the maturation phase of the composting process, the organic material contained in each reactor was transferred to a 21-l her- metically-closed glass cell. Less instrumentation was required be- cause the microbial activity and the O2demand decreased, and the temperature usually remained steady during maturation.
2.2. Composting experiment 2.2.1. Composting mixture
An aerobically digested sewage sludge was mixed with branches, grass clippings, privet hedge trimmings and leaves to represent 20, 25, 15, 20 and 20% of the total dry mass (DM) of the initial mixture, respectively. These organic materials were cho- sen to stand for typical feedstock at composting plants handling sewage sludge and green waste (Brändli et al., 2005; Doublet et al., 2011). The composition and the compostability of the initial mixture were tested during preliminary experiments (results not shown). The branches were hand-cut into 4 cm pieces, and the leaves and privet trimmings were roughly ground to centimetric size. The equivalent of 287 g dry weight of the initial mixture was put into each of the 6 reactors.
T
T T T
T T T
T T T T T
gas analyzer O2, CO2 cooling filtration
exhaust air
solenoid valve data
acqui- sition, solenoid
valve
exhaust air composting
T3
T1
T2
T3
T1
T2
T3
T1
T2
T1
T2
T3
T1
T2
T3
T1
T2
T3
valve control
4-l basket
heating coil thermo-
couple
composting reactor
H O
C1 C2 C3 C4 C5 C6
leachates
air compressor air in
flow meter volumetric counter thermostated
bath
H2O
Fig. 1.Experimental setup used for the first phases of composting, including six small-scale reactors.
2.2.2. Composting procedure
The initial moisture content was brought up to 66% of wet weight by water spraying. During the first 6 days, the external temperature of the reactor double wall was maintained 1–2°C be- low that of the compost material to observe a natural self-heating of the compost, while compensating for large heat losses occur- ring through the external surface. This procedure was chosen to maintain a minimal net outflow of heat and simulate thermal inertia of a compost pile (Mason and Milke, 2005). Because tem- perature controls the composting process, once the thermophilic conditions were naturally established, the temperature was con- trolled to mimic a characteristic temperature profile of full-scale composting of such initial waste mixture: days 7–11, external heating was kept constant to ensure an internal temperature be- tween 60 and 65°C to maintain the thermophilic phase; days 12–41, external heating was gradually reduced to reach 28°C to simulate the cooling phase; days 42–83, composts were moved to the 21-l maturation cells and placed in a thermostatic room at 28 ± 1°C.
The airflow was set to 47 l h 1kg 1 DM of initial composting mixture during the first 8 days and then was reduced to 29 l h 1kg 1 DM of initial composting mixture during the 9–
41 day period because the microbial activity and O2demand were decreasing. During maturation, the cells were opened every 3 days to renew the atmosphere.
The composting mixtures were sampled three times: at the end of the thermophilic phase (day 13), at the beginning of the matu- ration phase (day 41) and at the end of the maturation phase (day 83). The entire mixtures were emptied and homogenised be- fore sampling. An average amount of 46 g DM and 43 g DM were sampled on days 13 and 41, respectively. At these sampling dates, milli-Q water (Millipore, USA) was sprayed onto the compost after sampling to keep the moisture content at 55–70% of wet weight.
Leachates were also collected, and their soluble C contents were analysed by infrared absorption of C–CO2produced after combus- tion at 680°C (Shimazu TOC, ASI 5000).
2.3. Composting performance and reproducibility 2.3.1. Biochemical characterisation of organic matter
The dry mass evolution was measured during composting. The initial mixtures were weighed before their introduction into the reactors, and the six composting mixtures were weighed before stirring and after sampling, on each sampling date. The mass losses related to sampling were accounted for in the overall mass balance of composting.
All samples of composting mixtures, feedstock and initial mix- tures were dried at 40°C and finely ground to a 1-mm particle size before analysis. Total organic matter (TOM) was determined by loss on ignition at 480°C. Total organic carbon (TOC) and N were determined by elementary microanalysis after additional grinding to 200
l
m and combustion on a CHN analyser (Carlo Erba NA 1500, Italy).The biochemical characteristics of the feedstock, the initial mix- ture and their evolution over composting were determined on 1- mm ground samples. The hot water soluble fraction (W100) was extracted with milli-Q water at 100°C for 30 min. An additional soluble fraction (SOL) was extracted with a neutral detergent for 1 h (Van Soest and Wine, 1967). The hemicellulose-like (HEM), cel- lulose-like (CEL) and lignin-like (LIC) fractions were sequentially obtained following the French standard XPU 44-162 (AFNOR, 2005) with a crude fibre extractor (Fibertec 1020, Foss). All fracti- onations were performed on 1 g DM samples using glass crucibles with coarse porosity (40–100
l
m). The biochemical fractions are expressed as g 100 g 1of TOM.2.3.2. Assessment of organic matter stabilisation
The biodegradabilities of the feedstock, initial mixture and com- posts were assessed through the measurement of organic C miner- alisation during incubation of soil-organic substrate mixtures under controlled laboratory conditions as described in the French standard XPU44-163 (AFNOR, 2009). This made it possible to com- pare the biodegradability and stability of all organic substrates in conditions considered optimal for micro-flora and in a reference soil. The incubations were performed in four replicates in hermet- ically closed jars placed at 28 ± 1°C in the dark. The incubated mix- tures were made with the equivalent of 25 g of dry soil. The soil came from a field experiment located in Feucherolles (Parisian Ba- sin, France). The soil was a Glossic Luvisol (WRB-FAO classification) with silt loam texture (% dry matter): 76 silt, 17 clay and 0.9 organ- ic C. All organic samples, dried at 40°C and 1-mm ground, were added based on the same amount of organic C (equivalent to 50 mg of TOC added). Mineral N was added at the beginning of the incubations to prevent limitation of organic matter decomposi- tion by low mineral N availability. Using Milli-Q water, the water content of the mixtures was adjusted to 240 g kg 1DM, the soil water content equivalent to field capacity, and this level was main- tained during the incubation period by adding water when neces- sary. The mineralised CO2 was trapped in 10 ml of 0.5 M NaOH replaced after 1, 3, 7, 14, 21, 28, 49, 70 and 91 days of incubation.
The trapped C–CO2was analysed by continuous flow colorimetric analyses (Continuous flow Skalar Analyzer, the Netherlands). The kinetics of total mineralised C was calculated by summing the C–
CO2respired between two sampling dates. Control incubations of soil without organic sample input were conducted to determine native soil organic C mineralisation. Consequently, the mineralisa- tion of the organic substrates was calculated as the difference be- tween C–CO2 produced in soil with and without the organic substrate input.
2.3.3. Assessment of composting reproducibility
The reproducibility of composting was evaluated calculating the mean, the standard deviation (mean ± SD) and the coefficient of variation (CV, in %) for the six composting replicates of composting parameters and analytical characteristics of the initial mixture, and the composts sampled after 13, 41 and 83 days of composting. The composting parameters included the following: the highest tem- perature reached by self-heating, the evolution of moisture con- tent, the losses in dry matter, the total organic C (TOC) and the total organic matter (TOM). The analytical characteristics assessed for variability were as follows: the organic matter, N and C con- tents, the biochemical fractions and the amount of CO2mineralised during 91-day soil incubation.
3. Results and discussion
3.1. Composting process 3.1.1. Temperature profiles
The temperature of the composting mixture rose soon after beginning the experiment and reached 69 ± 4°C within 2 to 4 days, corresponding to an average increase rate of 12°C day 1(Fig. 2).
This corresponded to natural self-heating of the organic mixture since the temperature of the external jacket of the reactor was maintained 1–2°C below the temperature measured in the middle of the composting mixture. Comparable increases of temperature rise have been reported for similar sludge-based mixtures in larger composting devices (de Guardia et al., 2008) and on full-scale plants (Jouraiphy et al., 2005). However, the initial mixtures largely influence the temperature increaseviatheir nutrient balance and their content in easily biodegradable organic fractions (Haug,
1993). Indeed, much lower rates of temperature increase were ob- served during the composting of green wastes on full-scale plants (Hannon and Mason, 2003) and of green waste-biowaste mixtures in large-scale reactor (Francou et al., 2008). The low coefficient of variation (6%) of the maximum temperature attested of the high reproducibility of the self-heating for such given organic mixture (Clark et al., 1977).
To avoid the rapid decrease in temperature observed in small- scale reactors (Petricet al., 2009), the temperature was set to a gi- ven profile once the thermophilic conditions were established, to approach temperature dynamics of on-site composting. The tem- perature profile extended the thermophilic stage over the three minimum days recommended for pathogens sanitation (Baby et al., 2005) and then drove gradually the temperature from ther- mophilic to the cooler temperatures usually observed during mat- uration. The duration of the cooling phase from day 12 to 41 was comparable with those observed in large-scale reactor (Oleszczuk, 2006; de Guardia et al., 2008) but was largely shorter than re- corded on full-scale composting with similar organic mixtures (Jones and Westmoreland, 1998; Jouraiphy et al., 2005). The proce- dure of temperature control makes it possible to simulate the ther- modynamic regime of a composting process which is likely to condition many other parameters of composting such as biological activity and microbial population dynamics, moisture conditions (Ryckeboer et al., 2003; Mason and Milke, 2005) although it may have partly limited the reproduction of full-scale composting.
3.1.2. Concentration in CO2and O2in the exhaust gas
The O2concentration in the outlet gas slightly diminished over the first two weeks and remained constant thereafter, providing evidence of aerobic conditions (Fig. 2). The composting mixtures were turned twice during the process preventing the formation of preferential air pathways (Petiot and de Guardia, 2004). In- creases of CO2concentration in the outlet gas were detected during the first week, then after sampling on day 13, when the compost was homogenised and humidified. The variations in O2 and CO2
concentrations in the exhaust gas were very reproducible among the six reactors and confirmed the activation of endogenous micro- bial populations (Körner et al., 2003). No further variations of CO2
and O2concentrations were detected after two weeks of compost- ing most likely because of the important dilution although the air- flow was decreased during the cooling phase. A wide range of aeration rates leads to successful composting and after preliminary tests, the used aeration rates were halfway between the optimal aeration rate of 120 l h 1kg 1DM, proposed byYamada and Kaw- ase (2006)and the optimal range of 8.5–16.6 l h 1kg 1DM, pro- posed by de Guardia et al. (2008) during the composting of sludge with bulking agents.
3.1.3. Moisture content
The moisture content tended to decrease due to the combina- tion of high temperature levels and aeration during the thermo- philic phase and was controlled by applying water at each sampling date and by humidifying the inlet air. The initial moisture content (66% of wet weight) was reduced in all experiments to reach an average moisture content of 41 ± 7% (CV = 11%) of the wet weight, remaining above the minimum moisture content of 40% suggested byHaug (1993)for optimal composting conditions (Table 1). However, the composting process may have been tempo- rarily limited by low moisture content in a few cases due to com- post drying by the airflow although optimal moisture contents were restored in all cases by water addition on sampling dates.
3.2. Evolution of compost characteristics 3.2.1. Organic matter loss during composting
Dry matter losses mainly occurred during the first 13 days but varied among composting replicates at this stage of composting, with a mean loss of dry matter of 31 ± 9% (Table 1) and a coefficient of variation of 29%. The average dry matter loss reached 40 ± 7%
over the entire composting process and variability among com- posting replicates decreased to 17%. The extent of dry matter loss was similar to observed during sewage sludge composting on a full-scale composting site over 110 days (Sánchez-Monedero et al., 1999). Similar patterns were observed for TOM, with a loss of 46 ± 6% (CV = 12%) of the initial TOM during the overall com- posting process. Total organic carbon loss of 42 ± 6% (CV = 14 %) of initial TOC at the end of the composting was similar to the C1
0 10 20 30 40 50 60 70
80 C2 C3
O2 (%) and CO2 (%) 0 2 20 25 30 35 40
C4
0 10 20 30 40
Temperature (°C)
0 10 20 30 40 50 60 70
80 C5
Composting time (days)
0 10 20 30 40
C6
0 10 20 30 40
0 2 20 25 30 35 40 O2
CO2 T3
T1
Fig. 2.Evolution of the temperature profile (°C) in the centre of the compost (T1) and in the thermostatic bath (T3), percentages of O2and CO2in the outlet gas flow during the first 41 days of composting in the six composting reactors C1–C6.
TOC loss (39%) observed byDoublet et al. (2011)during the com- posting of sewage sludge in 170-l reactors over an identical period.
An expected increase in N content was observed in all reactors due to lower N loss by volatilisation compared to TOC lossviaminerali- sation during composting. The C:N ratio thus decreased during composting (Table 1). The final C:N ratio had a coefficient of vari- ation of 12% indicating that the chemical evolution among the six replicated composting experiments was reproducible. Losses of C within the leachates only accounted for 0.1–1.6% of the initial TOC and were considered as negligible.
3.2.2. Organic matter stabilisation during composting
The microbial degradation of the most easily biodegradable or- ganic compounds during composting led to organic matter
stabilisation. The proportion of easily biodegradable organic C in the initial mixture reached 52% of TOC and decreased regularly in the composts sampled after 13, 41 and 83 days (36 ± 3, 35 ± 6 and 22 ± 2% of TOC, respectively;Fig. 3). The proportion of biode- gradable organic C had CV varying between 9 and 17% among the six composts that attested a good reproducibility of the organic matter stabilisation over composting. After 83 days of composting, 56 g of organic C in each reactor had been mineralised (calculated fromTable 1), and 17 g remained as easily biodegradable in the fi- nal compost. The total of 73 g of organic C corresponded to the bio- degradable organic C present in the initial mixture. The most recalcitrant fraction of organic C remained constant throughout the entire process (58 g of organic C) representing 48% of TOC in the initial mixture and 78% of TOC in the final composts. Such Table 1
Chemical and biochemical characteristics of the initial mixture and their evolution during composting: mean, standard deviation (SD) and coefficient of variation (CV) calculated considering the six replicated composting C1–C6.
Initial mixture Compost sample
Day 0 Day 13 Day 41 Day 83
Mean ± SD CV (%) Mean ± SD CV (%) Mean ± SD CV (%)
DM (% wet weight) 37 51 ± 18 36 55 ± 14 25 59 ± 7 11
TOM (% dry weight) 87.8 82.6 ± 1.2 2 81.6 ± 1.4 2 78.5 ± 1.3 2
TOC (% dry weight) 45.5 45.5 ± 0.7 1 46.4 ± 1.5 3 43.7 ± 1.0 2
N (% dry weight) 2.9 3.2 ± 0.2 8 2.9 ± 0.5 17 3.6 ± 0.3 9
C:N 15.7 14.4 ± 1.3 9 16.8 ± 4.5 27 12.2 ± 1.5 12
W100 (% of TOM) 25 16 ± 1 6 14 ± 2 13 17 ± 1 6
SOL (% of TOM) 17 20 ± 3 17 21 ± 2 10 22 ± 4 17
HEM (% of TOM) 18 12 ± 4 31 11 ± 2 17 11 ± 3 29
CEL (% of TOM) 23 28 ± 2 7 29 ± 3 10 21 ± 4 19
LIC (% of TOM) 17 24 ± 1 4 26 ± 1 5 29 ± 3 12
LIC: (CEL + HEM) 0.4 0.6 ± 0.1 9 0.6 ± 0.1 15 0.9 ± 0.3 27
Total DM per reactor (g) 287 198 ± 26 13 185 ± 14 8 173 ± 20 11
DM, dry matter; TOM, total organic matter; TOC, total organic carbon; N, total N; W100, water soluble fraction in boiling water; SOL, soluble fraction in neutral detergent;
HEM, hemicellulose-like fraction; CEL, cellulose-like fraction; LIC, lignin-like fraction.
10 20 30 40 50 60 70
Initial mixture 13 day compost 41 day compost 83 day compost C1
g 100g-1 TOC)
C2 C3
0
C-CO2 mineralized (g
0 20 40 60 80
0 10 20 30 40 50 60 70 C4
0 20 40 60 80
C5
0 20 40 60 80
C6
Incubation time (days)
Fig. 3.Kinetics of organic C mineralisation of the initial mixture and of the six composting replicates (C1 to C6) sampled after 13, 41 and 83 days of composting during incubation in soil, expressed in g 100 g1of compost-TOC. Values are the means ± standard errors of four replicated incubations.
organic matter stabilisation was comparable to the results ob- served in reactors with larger volume or on industrial plants (Ber- nal et al., 1998; Tognetti et al., 2008; Doublet et al., 2011).
3.2.3. Organic matter transformation during composting
The evolution in the distribution of organic matter among the biochemical fractions was similar in all reactors, with all CV rang- ing from 6 to 19%, except for HEM fraction which evolution was poorly reproducible (Table 1). The stabilisation of compost organic matter during composting was explained by the increase of the LIC fraction and the decrease of the CEL, HEM and the W100 fractions (Fig. 4). The average losses were 62 ± 5, 31 ± 17, 68 ± 7, 50 ± 11 and 12 ± 13% of the initial W100, SOL, HEM, CEL and LIC fractions, respectively. The W100, SOL and HEM fractions rapidly decreased during the first weeks, then remained constant as observed with larger composting devices (Francou et al., 2008; Doublet et al., 2011), whereas the decrease in the CEL fraction was mostly observed after 41 days during the maturation phase. The hot water extracted readily available C and N sources for the microbial bio- mass (Said-Pullicino et al., 2007) which were degraded first. The lag phase in the degradation of the CEL fraction has previously been observed and was explained by the presence of a C source more easily metabolised in the earlier stages of composting (Eiland et al., 2001) and by the inhibition of the activity of the cellulolytic microorganisms at high temperatures (Jouraiphy et al., 2005). The degradation of CEL and HEM fractions exceeded 50% of the initial contents as most often observed, regardless of the given scale of composting (Jouraiphy et al., 2005; Cayuela et al., 2006; Alburquer- que et al., 2009; Doublet et al., 2011). The LIC fraction tended to de- crease slightly or remained stable, indicating the resistance of this fraction to biodegradation. However, lignin degradation occurred during composting and thermophilic micro-fungi are the most important lignin degraders with an optimal activity around 50°C (Tuomela et al., 2000) and lignin degradation increased with the duration of the thermophilic phase. In the small-scale pilots, the LIC fraction degradation reached similar extent than in 170-l reactors (Francou et al., 2008; Doublet et al., 2011). It remained lower than observed during the composting of sewage sludge on full-scale composting plants probably because of the shorter dura- tion of the thermophilic phase (Jouraiphy et al., 2005; Cayuela et al., 2006). The LIC:(CEL + HEM) ratio increased with compost or- ganic matter stabilisation during the process (Francou et al., 2008) (Table 1). The degradation of the HEM and CEL fractions and/or the oxidation of the LIC fraction may have fedde novothe water solu-
ble and SOL fraction balancing their decrease by mineralisation and explaining the observed relative stability of the fractions after 13 days of composting (Sánchez-Monedero et al., 1999; Said-Pulli- cino et al., 2007; Francou et al., 2008).
The organic matter evolution during composting was thus con- sistent with previous observations in larger reactors and even in full-scale composting plants. However in industrial plants, com- post screening at the end of the thermophilic phase separates the most stabilised organic matter from the coarsest non degraded fractions mainly composed of CEL and LIC fractions (Doublet et al., 2011) and accentuates the decrease of these two fractions in the final screened compost. In our case, no screening was real- ized because of the initial rather small size of the organic mixture and the low amounts of organic matter in the small-scale reactors.
It could thus be a limitation to the simulation of the industrial composting process at small-scale, but it was difficult to evaluate because the screening effect has been rarely studied in the litera- ture. Another possible limitation of the composting simulation in laboratory devices is the lack of colonization by macrofauna ob- served on full-scale composting plants although the influence of macrofauna on organic matter evolution during maturation re- mained poorly documented.
4. Conclusions
The challenge to minimise the volume and the waste mass in a small-scale composting system conducted to build an innovative experimental setup including six instrumented reactors function- ing in parallel. This work aimed to confirm that this device repro- duced the composting process and that the final compost was comparable to those generated by full-scale composting plants.
The performance and reproducibility of composting was investi- gated on a mixture of sewage sludge and green wastes standing as example of classically used initial waste mixture. The establish- ment of thermophilic conditions by self-heating succeeded in all reactors. The losses in TOM over the whole process were similar to large-scale reactors or on-site experiments. A classical stabilisa- tion of compost organic matter corresponding to the enrichment in the LIC fraction and a decrease in the biodegradable fractions was observed. Nevertheless, the lignin degradation was less important than in full-scale systems probably related with the shorter thermophilic phase, essential for lignin degradation, and the lack of screening that sorts coarse LIC and CEL-rich fractions.
The composting process in the small-scale reactors was highly reproducible for maximum temperature reached through self- heating, dry matter and organic matter losses and compost organ- ic matter stabilization and biochemical transformations. Despite the inherent difficulties of the miniaturisation of the composting process, the small-scale reactors and the composting procedure performed representative composting. It could be used in further studies on the degradation of organic pollutants during compost- ing using radio-labelled molecules requiring small amounts of ini- tial mixtures.
Acknowledgements
This work was financed by the ADEME (French Environment and Energy Management Agency) and the INRA (French National Institute for Agricultural Research). The experiments were funded by Veolia Environment, Research and Development. We would like to thank Christophe Labat and Guillaume Bodineau for their help in the experiment setup, Valérie Bergheaud and Valérie Dumeny for their assistance in sampling and Véronique Etievant for her collab- oration in the laboratory analyses. We would like to thank Suzette Tanis-Plant for her discussion and English editorial advice.
25
10 8 9
17
13 12 12
18
8 76
23
18 17
11 17
16
15
15
35 40
46
0 10 20 30 40 50 60 70 80 90 100
0 13 41 83
Composting time (days) g 100g-1 of initial TOM
Mineralized OM LIC
CEL HEM SOL W100
Fig. 4.Average evolution of the total organic matter (TOM) distribution in the Van Soest fractions for the six composting replicates (C1–C6), expressed in g 100 g 1of initial total organic matter (TOM). The proportion of mineralised OM was calculated from the mass balance during composting (W100, hot water soluble fraction; SOL, soluble fraction in neutral detergent; HEM, hemicellulose-like fraction; CEL, cellulose-like fraction; LIC, lignin-like fraction). Error bars represent the standard deviation.
References
AFNOR, 2005. Norme Française NFU 44-162. Amendements organiques et supports de culture – Fractionnement biochimique et estimation de la stabilité biologique. AFNOR, Paris, France.
AFNOR, 2009. Norme Française NFU 44-163. Amendements organiques et supports de culture – Détermination du potentiel de minéralisation du carbone et de l’azote. AFNOR, Paris, France.
Alburquerque, J.A., Gonzalvez, J., Tortosa, G., Baddi, G.A., Cegarra, J., 2009. Evaluation of ‘‘alperujo’’ composting based on organic matter degradation, humification and compost quality. Biodegradation 20, 257–270.
Baby, R.E., Cabezas, M.D., Labud, V., Marqui, F.J., de Reca, N.E.W., 2005. Evolution of thermophilic period in biosolids composting analyzed with an electronic nose.
Sensors and Actuators: B-Chemical 106, 44–51.
Bernal, M.P., Sánchez-Monedero, M.A., Paredes, C., Roig, A., 1998. Carbon mineralization from organic wastes at different composting stages during their incubation with soil. Agriculture, Ecosystems & Environment 69, 175–189.
Brändli, R.C., Bucheli, T.D., Kupper, T., Furrer, R., Stadelmann, F.X., Tarradellas, J., 2005. Persistent organic pollutants in source-separated compost and its feedstock materials – A review of field studies. Journal of Environmental Quality 34, 735–760.
Cayuela, M.L., Sánchez-Monedero, M.A., Roig, A., 2006. Evaluation of two different aeration systems for composting two-phase olive mill wastes. Process Biochemistry 41, 616–623.
Clark, C.S., Buckingham, C.O., Bone, D.H., Clark, R.H., 1977. Laboratory scale composting – techniques. Journal of the Environmental Engineering Division – ASCE 103, 893–906.
de Guardia, A., Petiot, C., Rogeau, D., 2008. Influence of aeration rate and biodegradability fractionation on composting kinetics. Waste Management 28, 73–84.
Doublet, J., Francou, C., Poitrenaud, M., Houot, S., 2011. Influence of bulking agents on organic matter evolution during sewage sludge composting; consequences on compost organic matter stability and N availability. Bioresource Technology 102, 1298–1307.
Eiland, F., Leth, M., Klamer, M., Lind, A.M., Jensen, H.E.K., Iversen, J.J.L., 2001. C and N turnover and lignocellulose degradation during composting of Miscanthus straw and liquid pig manure. Compost Science & Utilization 9, 186–196.
Francou, C., Linères, M., Derenne, S., Le Villio-Poitrenaud, M.L., Houot, S., 2008.
Influence of green waste, biowaste and paper-cardboard initial ratios on organic matter transformations during composting. Bioresource Technology 99, 8926–
8934.
Hannon, J.B., Mason, I.G., 2003. Composting of green waste shredded by a crush/cut roller versus a low speed counter-rotating shear shredder. Compost Science &
Utilization 11, 61–71.
Haug, R.T., 1993. The Practical Handbook of Compost Engineering. Lewis Publishers, Boca Raton, FL.
Hesnawi, R.M., McCartney, D.M., 2006. Impact of compost amendments and operating temperature on diesel fuel bioremediation. Journal of Environmental Engineering and Science 5, 37–45.
Hu, Z., Lane, R., Wen, Z., 2009. Composting clam processing wastes in a laboratory- and pilot-scale in-vessel system. Waste Management 29, 180–185.
Jones, F.W., Westmoreland, D.J., 1998. Degradation of nonylphenol ethoxylates during the composting of sludges from wool scour effluents. Environmental Science & Technology 32, 2623–2627.
Jouraiphy, A., Amir, S., El Gharous, M., Revel, J.-C., Hafidi, M., 2005. Chemical and spectroscopic analysis of organic matter transformation during composting of sewage sludge and green plant waste. International Biodeterioration &
Biodegradation 56, 101–108.
Körner, I., Braukmeier, J., Herrenklage, J., Leikam, K., Ritzkowski, M., Schlegelmilch, M., Stegmann, R., 2003. Investigation and optimization of composting processes-test systems and practical examples. Waste Management 23, 17–
26.
Mason, I.G., Milke, M.W., 2005. Physical modelling of the composting environment:
a review. Part 1: reactor systems. Waste Management 25, 481–500.
Michel Jr., F.C., Reddy, C.A., Forney, L.J., 1995. Microbial degradation and humification of the lawn care pesticide 2, 4-dichlorophenoxyacetic acid during the composting of yard trimmings. Applied and Environmental Microbiology 61, 2566–2571.
Oleszczuk, P., 2006. Influence of different bulking agents on the disappearance of polycyclic aromatic hydrocarbons (PAHs) during sewage sludge composting.
Water Air Soil Pollution 175, 15–32.
Petiot, C., de Guardia, A., 2004. Composting in a laboratory reactor: a review.
Compost Science & Utilization 12, 69–79.
Petric, I., Sestan, A., Sestan, I., 2009. Influence of initial moisture content on the composting of poultry manure with wheat straw. Biosystems Engineering 104, 125–134.
Reid, B.J., Fermor, T.R., Semple, K.T., 2002. Induction of PAH-catabolism in mushroom compost and its use in the biodegradation of soil-associated phenanthrene. Environmental Pollution 118, 65–73.
Ryckeboer, J., Mergaert, J., Vaes, K., Klammer, S., De Clercq, D., Coosemans, J., Insam, H., Swings, J., 2003. A survey of bacteria and fungi occurring during composting and self-heating processes. Annals of Microbiology 53, 349–410.
Said-Pullicino, D., Erriquens, F.G., Gigliotti, G., 2007. Changes in the chemical characteristics of water-extractable organic matter during composting and their influence on compost stability and maturity. Bioresource Technology 98, 1822–1831.
Sánchez Arias, V., Fernández, F.J., Rodríguez, L., Stentiford, E.I., Villaseñor, J., 2011.
Kinetics of forced aerated biodegradation of digested sewage sludge-reed mixtures at different temperatures. Journal of Environmental Management.
doi:10.1016/j.jenvman.2011.05.014.
Sánchez-Monedero, M.A., Roig, A., Cegarra, J., Bernal, M.P., 1999. Relationships between water-soluble carbohydrate and phenol fractions and the humification indices of different organic wastes during composting. Bioresource Technology 70, 193–201.
Sanz, E., Prats, D., Rodríguez, M., Camacho, A., 2006. Effect of temperature and organic nutrients on the biodegradation of linear alkylbenzene sulfonate (LAS) during the composting of anaerobically digested sludge from a wastewater treatment plant. Waste Management 26, 1237–1245.
Schloss, P.D., Chaves, B., Walker, L.P., 2000. The use of the analysis of variance to assess the influence of mixing during composting. Process Biochemistry 35, 675–684.
Tognetti, C., Mazzarino, M.J., Laos, F., 2008. Compost of municipal organic waste:
effects of different management practices on degradability and nutrient release capacity. Soil Biology and Biochemistry 40, 2290–2296.
Tuomela, M., Vikman, M., Hatakka, A., Itävaara, M., 2000. Biodegradation of lignin in a compost environment: a review. Bioresource Technology 72, 169–
183.
Van Soest, P.J., Wine, R.H., 1967. Use of detergents in the analysis of fibrous feeds.
IV: determination of plant call-wall constituents. Journal of the Association of Official Analytical Chemists 50, 50–55.
Yamada, Y., Kawase, Y., 2006. Aerobic composting of waste activated sludge: kinetic analysis for microbiological reaction and oxygen consumption. Waste Management 26, 49–61.
Zenjari, B., El Hajjouji, H., Baddi, G.A., Bailly, J.R., Revel, J.C., Nejmeddine, A., Hafidi, M., 2006. Eliminating toxic compounds by composting olive mill wastewater–
straw mixtures. Journal of Hazardous Materials 138, 433–437.
