Waste Management 34 (2014) 323–328
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Evaluation of laboratory-scale in-vessel co-composting of tobacco and apple waste Nina Kopcˇic´ ⇑, Marija Vukovic´ Domanovac, Dajana Kucˇic´, Felicita Briški Faculty of Chemical Engineering and Technology, University of Zagreb, Marulic´ev trg 19, 10000 Zagreb, Croatia
a r t i c l e
i n f o
Article history: Received 17 July 2013 Accepted 4 November 2013 Available online 28 November 2013 Keywords: Co-composting Apple waste Tobacco waste Oxygen Temperature Process modelling
a b s t r a c t Efficient composting process requires set of adequate parameters among which physical–chemical properties of the composting substrate play the key-role. Combining different types of biodegradable solid waste it is possible to obtain a substrate eligible to microorganisms in the composting process. In this work the composting of apple and tobacco solid waste mixture (1:7, dry weight) was explored. The aim of the work was to investigate an efficiency of biodegradation of the given mixture and to characterize incurred raw compost. Composting was conducted in 24 L thermally insulated column reactor at airflow rate of 1.1 L min1. During 22 days several parameters were closely monitored: temperature and mass of the substrate, volatile solids content, C/N ratio and pH-value of the mixture and oxygen consumption. The composting of the apple and tobacco waste resulted with high degradation of the volatile solids (53.1%). During the experiment 1.76 kg of oxygen was consumed and the C/N ratio of the product was 11.6. The obtained temperature curve was almost a ‘‘mirror image’’ of the oxygen concentration curve while the peak values of the temperature were occurred 9.5 h after the peak oxygen consumption. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Agro-food industries generate large quantities solid and liquid wastes which must be suitably managed before being discharged to the environment. The solid wastes, which originate from the food processing and agricultural industries, are suitable substrates for composting (Adhikari et al., 2008; Kim et al., 2008; Manios, 2004). Composting is acceptable solution because it reduces the volume of bulky solid waste and results with organic material becoming a stable end product (Haug, 1993; Insam and de Bertoldi, 2007). Successful composting requires meticulous attention to key parameters and that includes: (i) substrate nature parameters (e.g. C/N ratio, particle size, pH-value) and (ii) process parameters (e.g. temperature, moisture content (MC) and aeration rate), (Diaz and Savage, 2007; Haug, 1993; Rashad et al., 2010). Preferable C/N ratio of the composting substrate is between 25 and 30 (Diaz and Savage, 2007). Composting of the wastes with C/N ratio lower than 20 is feasible (Kumar et al., 2010), but with loss of nitrogen through ammonia volatilization during the process (Diaz and Savage, 2007). A pH-value between 5.5 and 8.5 is optimal for the compost microorganisms, but preferable values are between 6.5 and 7.5. Values out of the optimal range could inhibit microbial activity and initial lag in the composting rate could be expected leading to rate ⇑ Corresponding author. Tel.: +385 1 4597271; fax: +385 1 4597260. E-mail address: [email protected] (N. Kopcˇic´). 0956-053X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.wasman.2013.11.001
limitations (Sundberg et al., 2004; Yu and Huang, 2009). The particle size of the waste intended for composting is in the range of 10 mm (in forced aeration systems) to 50–100 mm (in passive aeration systems) (Neklyudov et al., 2008). If the particle size is too small, air recirculation through composting mass is inhibited, free air space in the system decreases and reduces oxygen diffusion (Vlyssides et al., 2008). Concerning those aspects, there are not many types of biodegradable waste that can be characterized as ‘‘ideal’’ substrates for composting and mixing with chemical and/ or bulking agents is often necessary (An et al., 2012; Iqbal et al., 2010; Yu and Huang, 2009). On the other hand, substrate with physical–chemical characteristics optimal for composting can be obtained by mixing two or more types of biodegradable waste (Diaz et al., 2002; Iqbal et al., 2010; Sundberg et al., 2011; Rashad et al., 2010). Apple waste, namely apple pomace has many applications, e.g. apple pomace as a substrate for the production of citric acid (Dhillon et al., 2011) and apple pulp as a biogas production source (Coalla et al., 2009), but it is rare as composting substrate. The main characteristics that makes fruit waste not preferable or not acceptable for composting are low pH-value (Dhillon et al., 2011) and high moisture content (van Heerden et al., 2002; Kim et al., 2008). On the other hand, tobacco solid waste is rich in nitrogen (Tso, 1990) and the consequence of that is a lower C/N ratio (Piotrowska-Cyplik et al., 2009). As reported in recent researches, tobacco waste from cigarette manufacturing (Briški et al., 2003; Briški et al., 2012) and tobacco waste briquettes (Piotrowska-Cyplik et al., 2009) was
324
N. Kopcˇic´ et al. / Waste Management 34 (2014) 323–328
successfully composted in the reactor systems. However, tobacco waste from primary tobacco production is a dry and dusty material consisted of very small particles and mechanical or manual stirring was necessary for efficient composting of that kind of waste in a packed bed reactor (Kopcˇic´ et al., 2013). Both apple and tobacco waste showed their biological potential for microbial degradation in fermentation and/or aerobic biodegradation processes but with certain disadvantages in their physical– chemical characteristics for efficient composting, as described above. An alternative to overcome such disadvantages and to recycle both wastes is the co-composting of selected wastes in the proper ratio (Fernández et al., 2008; Manios, 2004; Vlyssides et al., 2008). The aim of the work was to investigate feasibility of composting of the mixture of apple and tobacco solid waste in the laboratoryscale thermally insulated reactor. The study objectives were: (a) to determine physical–chemical properties of apple waste and tobacco waste, (b) to evaluate the efficiency of co-composting of selected wastes mixed in proper ratio, (c) to better understand correlation between oxygen uptake and temperature, and (d) to evaluate proposed mathematical model. 2. Materials and methods 2.1. Composting materials Mixture of apple waste and tobacco solid waste were used as a composting substrate. Apples were collected in autumn on orchard in northern part of Croatia. Apples were manually pealed and non-edible parts were chopped into pieces with sizes ranging from 1 to 50 mm, where about 70% of the particles were exceeded 20 mm. The apple waste used for the experiment was consisted of an apple skin, cores with seeds, stems and decaying parts of apple pulps and the waste was kept in the freezer at 18 °C until composting experiment. The tobacco waste was collected from the tobacco primary production in Hrvatski duhani d.d. Virovitica, Croatia. The waste was composed of tobacco leaf residues and very fine, powdery mixture of tobacco residues and soil particles. The tobacco waste was stored in the dry place until further use. Prior the experiment major physical–chemical properties of each waste material were determined: dry solids (DS), moisture
content (MC), volatile solids (VS), carbon (C) and nitrogen (N) content and pH-value. The waste mixture was designed to have a carbon to nitrogen ratio (C/N) of 25 and initial moisture content of 60% (w.w.). According to the obtained C/N ratio of each material and to the targeted value of C/N ratio of the composting substrate, apple and tobacco waste were mixed in the ratio of 1:7 (dry weight). Before composting, pH-value of the mixture was checked and water was added in order to obtain intended MC of the composting substrate. 2.2. Experimental setup The reactor was 24 L cylindrical, vertically positioned with a diameter of 190 mm and the height of 940 mm (Fig. 1). The reactor was of polyethylene and insulated with 50 mm thick polyurethane. Reactor was placed on the scale (M-3-1086 Sklad, Primjer, Zagreb, Croatia) in order to regularly check the changes of the mass of the substrate. The day before the experiment frozen apple waste was placed in a refrigerator at 4 °C for 24 h to defrost. The initial mixture was prepared of 5.4 kg of the tobacco waste and 3.7 kg of the defrosted apple waste. To ensure the intended moisture content, 5.7 L of tap water was added to the waste mixture. Before the experiment, initial pH-value, moisture content (MC) and C/N ratio of the mixture were determined. The reactor was filled from the top with 14.8 kg of solid waste mixture. The substrate was forcefully aerated with constant airflow rate set to 1.1 L min1 (equivalent to 0.31 L min1 kg1init.volat.solids) from the bottom of the reactor during 22 days of the experiment. The airflow rate was selected on the basis of the preliminary experiment (results not presented) as high enough to overcome the pressure drop through the composting mass and to obtain a constant outgoing gas flow for oxygen monitoring purposes. The trial length of the experiment of 22 days was considered representative of the overall composting process evolution. Upward aeration was provided by an air compressor (DE 50/204 FIAC, Italy) and output pressure of the air was set to 1.0 bar. Air flow rate was regulated by valve of an airflow-meter of measuring range of 0–2.313 L min-1 (Cole Parmer, USA,). In order to reduce and minimize drying of the substrate bed in the reactor, air was saturated with water by passing through Dreschel bottle (not presented in the scheme, Fig. 1) before entering the reactor.
Fig. 1. Schematic diagram of composting process: 1-air compressor and flow regulation; 2-composting reactor; 3-scale; 4-temperature probes; 5-data acquisition, monitoring and storage; 6-condensate collecting; 7-gas chromatograph.
N. Kopcˇic´ et al. / Waste Management 34 (2014) 323–328
Two temperature sensors were used; one placed at the middle of the reactor’s height (i.e. 470 mm above the base plate, on the central vertical axis) and the other at the air inlet of the reactor (Fig. 1). 2.3. Physical–chemical and microbiological monitoring and analyses Temperatures at the reactors inlet and in the middle of the reactor height were monitored by means of the RTD compact probes (±1%, Cole-Parmer, USA) connected to the computer by the 2-chanell RTD analogue input (SCC-RTD01, National Instruments, USA). Computer software used for that purpose was NI-DAQ 7, LabVIEW Win/PCI-6024/CB_68LP starter Kit, National Instruments, USA. Temperature data was programmed to be recorded and stored every 15 min. Oxygen content in outlet gas stream was recorded at preset times with gas chromatograph Multiple Gas Analyzer #2 (SRI, USA) was equipped with 1 m long Molecular sieve 5A column and thermal conductivity detector. The carrier gas was Helium 5.0 (Messer, Croatia) and gas analysis was carried out at helium flow rate of 20 mL min1 and at oven temperature of 60 °C. During the experiment most relevant physical–chemical and microbiological parameters were determined in samples of composting mass. Samples were taken daily from the top of the reactor at the middle point of the composting mass. At the beginning and at the end of the process mass of the substrate mixture/raw compost was determined and samples were taken in order to analyze the most relevant physical–chemical parameters. The pH-value was determined in a filtrate of a water extract (1:10 by weight) of the substrate sample after shaking for 30 min on the magnetic stirrer with pocket electrode (HI 98128, ±0.01, Hanna Instruments, Deutschland). The dry solids content (DS) (as well as MC) and volatile solids (VS) content were determined on duplicate samples at 105 °C and 550 °C, respectively, until the constant mass was obtained. The total nitrogen was determined as N-Kjeldahl with Digestor 2006 and distillation unit Kjeltec 2100, FOSS, USA. All analyses were carried out according to the Austrian S 2023 standard methods (ÖNORM, 1986.) for analysis of compost. Microflora was monitored through the estimation of mesophilic and thermophilic bacteria and fungi. Viable plate count was determined by the decimal dilution method and the results expressed as colony-forming units (CFU) of bacteria and fungi per gram of dry matter content (Briški et al., 2003). Mesophilic and thermophilic bacteria were enumerated on the nutrient agar (Biolife, Italy) after 24–48 h incubation at 37 °C and 50 °C, respectively. Mesophilic and thermophilic fungi were enumerated on the malt agar (Biolife, Italy) after 3 days incubation at 28 °C and 50 °C, respectively.
325
where mVS0 is initial mass of volatile solids (kg), T and Tin are temperatures (°C) of the substrate and of the air at reactor’s inlet, hw(T) is the heat of vaporization (kJ kg1) and HS(T) and HS0 are saturated humidities (kg kg1) at reactor and ambient temperature. The humidity of the air leaving the reactor (HS(T)) is calculated in two steps (Dalsenter et al., 2005). First the vapour pressure, pvp (Pa) is calculated according the Antoine equation:
pvp ¼ 133:322exp 18:3036 -
3816:44 ðT þ 273:15Þ 46:13
ð2Þ
The humidity of the air at saturation is then given by:
HS ¼ 0:62413
pvp p pvp
ð3Þ
where p is the total pressure (Pa) of the system and the factor 0.62413 is the ratio of the molecular weight of water to the average molecular weight of dry air. In the case of the air entering the bottom of the bed, T is replaced with Tin, and the vapour pressure (pvp) and humidity (HS0) of the entering air are calculated with the same equations. Eq. (1) was assigned to a mathematical model developed using the kinetic model and mass and energy balances and was reported in previous work (Briški et al., 2003). Modeling of the reaction kinetics was based on the simple process chemistry and other characteristics of the reactor selected for the experiment. Accordingly, an empirical equation was chosen as a kinetic model, Eq. (4):
rVS ¼ kðTÞ wVS expðntÞ þ wm VS
ð4Þ
where rVS is the degradation rate (d1), k(T) is the specific rate (d1), wVS is the mass fraction of volatile matter (kgVS kg1VS initial), t is the time (d) and n and m are adjustable parameters. To estimate dependence of the reaction rate on temperature, the empirical expression, Eq. (5), taken from the literature was applied (Haug, 1993):
h i kðTÞ ¼ k0 1:066T22 1:21T60
ð5Þ
Reactor was modelled as a continuous stirred tank reactor (CSTR) and several assumptions were used in developing the model (Briški et al., 2003): biodegradation of the substrate was slow compared to oxygen transfer through the boundary gas layer; oxygen concentration in the substrate bed was constant; airflow rate during composting was constant. It was also assumed that the process had been carried out under adiabatic conditions and that the released heat was proportional to the progress of biodegradation.
2.4. Statistical analysis
3. Results and discussion
Most of the results shown in present paper are the mean of two replicate analyses. Single determinations were made for temperature and oxygen concentration of outlet gases. Correlation coefficients of temperature and oxygen concentration of the outlet gases were expressed as Pearson’s r.
The main physical–chemical characteristics of selected wastes and prepared mixture are summarized in Table 1. Effect of freezing on microbial population and biological activity of the apple waste was not investigated but it was presumed that it was not affected (Pognani et al., 2012).
2.5. Mathematical model of the composting process The reaction enthalpy, (DHr) (kJ kgVS initial1) was calculated from experimental results by measuring the temperature of the composting mass during the reaction. The value of (DHr) was calculated using the following equation (Eq. (1)):
ðDHr Þ ¼
qa Q v mVS0 ð1 wVS Þ
Z
t
cpa ðT T in Þ þ hwðTÞ HSðTÞ HS0 dt
0
ð1Þ
Table 1 Physical–chemical characteristics of the waste and mixture used in the experiment. Substrate
Tobacco waste
Apple waste
Waste mixture
Mass of wet substrate, m0 (kg) Moisture content, MC (%, w.w) Volatile solids content, VS (%, d.w.) pH-value C/N ratio
5.4 5.8 57.3 6.8 19.7
3.7 80.1 96.9 4.6 185.6
14.8 62.0 62.8 6.1 22.9
log CFU (g-1 substrate)
12
60
10
50
8
40
6
30
4
20 mesophillic bacteria mesophillic fungi Temperature
2 0 0
2
4
6
8
10
termophillic bacteria termophillic fungi
Temperature (°C)
N. Kopcˇic´ et al. / Waste Management 34 (2014) 323–328
326
10 0
12
14
16
18
20
22
Time (d) Fig. 2. Temperature fluctuations and growth of total viable–cultivable mesophilic and thermophilic bacteria and fungi over time in composting mass.
24
10,0
22
9,0 8,0
18
7,0
16 14
6,0
pH-value
C/N ratio
20
12 C/N
10
5,0
pH
8
4,0 0
2
4
6
8
10
12
14
16
18
20
22
Time (d) Fig. 3. Changes of pH-value and carbon to nitrogen (C/N) ratio during 22 days of composting.
3.1. Changes in physical–chemical and microbiological properties of the substrate during composting The temperature in the composting mass increased rapidly following a short lag phase, and temperatures were subsequently maintained in the thermophilic region for period typical of those occurring in laboratory-scale composting operations (Ghaly et al., 2006). This behavior is illustrated in Fig. 2, which shows the temperature profiles measured at the mid-point of the reactor. (i.e. 450 mm above the base plate, on the central vertical axis). The fluctuation of temperature recorded during the experiment showed common composting phases. At the start of the process the number of mesophilic and thermophilic bacteria in 1 g of the dry substrate was 5.0 106 and 3.8 106, respectively. After the first 12 h of the lag phase of the mesophilic microflora, just slight increase in number of thermophilic bacteria was observed indicating the lag phase of thermophilic microorganisms. Lower temperatures and lower pH-value during the first five days of composting were favorable for fungi and their activity was high (van Heerden et al., 2002; Ryckeboer et al., 2003). The waste mixture was predominantly colonized by mesophilic yeasts which were thermally inactivated in the second phase of the process. Temperature values above 45 °C occurred at day 5 and lasted for 8 days while peak value of 54.7 °C was reached at day 7. The number of mesophilic and thermophilic bacteria reached their maximum at day 6 and it was 2.9 109 and 1.1 109, respectively. After the peak temperature, the drop of total bacteria count was observed which is commonly reported in the literature (Fourti et al., 2008; Ryckeboer et al., 2003). After the day 13 the composting mass started to cool slowly when. yeasts were replaced with bacteria and the number of both bacteria and fungi tended to stabilize. By mixing apple and tobacco waste in the ratio 1:7 (d.w.) substrate with pH-value of 6.3 adequate for composting was obtained
(van Heerden et al., 2002). A decrease of pH was recorded for the given mixture during the first three days (Fig. 3) which (pH < 6.5) presumably inhibited thermophilic microorganisms and caused lag phase in the transition from mesophilic to thermophilic conditions (Bautista et al., 2011), as discussed above (Fig. 2). After that, day after day, the pH started to increase and levelled around the value 8.8 after day 11. Slight decrease of pH-value in the initial stage of the process was probably due to production of organic acids that resulted from fast degradation of easily degradable organic matter (e.g. carbohydrates which are present in apple waste in high concentration (Dhillon et al., 2011)) by mesophilic microorganisms. After day 4, when thermophilic phase occurred, pH increased due to ammonia evolution and nitrogen losses from the substrate (Bautista et al., 2011). The final pH-value of 8.8 is in agreement with the literature for the raw compost and further maturation would lead to decrease of pH (Kim et al., 2008). Fig. 3 also shows the variation of C/N ratio with time. The initial C/N ratio of the feed mixture was 23 and it was in the range of optimal values for composting (Adhikari et al., 2008). The C/N ratio was continuously dropping until the end of the experiment (day 22) when it was 11.6. It is well known that a C/N ratio below 20 is indicative of proper compost maturity, with ratio of 15 or less being preferred (van Heerden et al., 2002; Kim et al., 2008). The degree of degradation or the conversion of organic matter was calculated from the volatile solid contents taking into account the changes in mass of the material inside the reactor during the composting. During the process conversion of organic matter increased and the increase was more pronounced during the first mesophilic stage of the process. The composting of most substrates is characterized by an initial period of rapid degradation followed by longer period of slower degradation (Diaz and Savage, 2007; Diaz et al., 2002; Haug, 1993). At the end of the process conversion of volatile matter of 53.1% was calculated. It should be noted that values of pH, C/N ratio, MC and VS were determined in the samples taken from the middle point of the reactor. In order to evaluate the process more representatively sampling along the reactor height, including the bottom and the top of the composting mass, should be performed as in the work of Kopcˇic´ et al., 2013. At the end of the experiment the composting material was removed from the reactor. At the time, raw compost was granular and dark material of pleasant odour. The top of the composting mass was overgrown by actinobacteria what indicated the proper moisture and good oxygen supply in higher regions of the reactor. 3.2. Oxygen uptake The usual procedure in system design is the maintenance of aerobiosis in excess of that required for an efficient biodegradation in composting process. The recommended lower limit of the oxygen concentration in the composting mass was reported to be 5% (v/v) (Haug, 1993; Puyuelo et al., 2010). As shown in Fig. 4, that value was not recorded during the whole period of composting what indicated a proper aeration rate (Kim et al., 2008; Puyuelo et al., 2010). The highest microbial activity based on oxygen uptake is also well shown in Fig. 4. The cumulative oxygen pattern was derived from oxygen concentration profile using the ideal gas law. During the first 12 h the cumulative oxygen uptake curve indicated a lag phase of the process. Average value of cumulative oxygen uptake rate at this stage was 1.25 g O2 h1. After that, at day 1 oxygen content in the exhaust gas reached the lowest value of 13.9% (v/v) since the presence of readily biodegradable materials caused high oxygen consumption by mesophilic microorganisms. That led to an exponential growth of mesophilic microflora and the release of metabolic heat. As a consequence of it temperature in the composting mass increased (Fig. 2). The highest cumulative oxygen uptake rate was obtained at the day 1 (5.96 g O2 h1) and during
O2 concentration (%, v/v)
24
2000 1800
22
1600 20
1400
18
1200 1000
16
800 600
14 12
O2 concentration
400
cumulative O2
200
10
0 0
2
4
6
8
10
12
14
16
18
20
Cumulative O 2 consumption (g)
N. Kopcˇic´ et al. / Waste Management 34 (2014) 323–328
22
Time (d) Fig. 4. Oxygen concentration and cumulative oxygen consumption during 22 days of composting.
a
22
b
20
20
18
18
16
16
14
14
12
O2 concentration (%, v/v)
O2 concentration (%, v/v)
22
12 20
30
40
50
60
20
30
40
50
60
327
3.3. Modeling the composting process For an application of the model and its experimental verification, several parameters and process variables are required. Necessary data for application of the proposed model are given in Table 2. Based on known input variables presented in Table 2, proposed dependence of the degradation rate on temperature (Eq. (5)) and calculated reaction enthalpy (DHr), (Eq. (1)), the kinetic parameters n, m and k0 in Eq. (4) were estimated (Table 3). Direct nonlinear regression analysis based on the Nelder–Mead simplex method (Nelder and Mead, 1965) was performed to determine k0, m and n in the kinetic model. After optimal values of kinetic parameters were determined, the set of differential equations were numerically solved using Runge– Kutta method (DelBuono and Mastroserio, 2002). The theoretical curves were drawn together with the actual data plot in Fig 6. The graphical presentation shows that theoretical temperature is in agreement with experimentally obtained values. The theoretical conversion of volatile solids was over-estimated against the experimental data (Fig. 6, Table 3). The lower values of experimentally obtained conversion were probably due to inability of taking a representative sample in relatively small volume of the material. Namely, the difference in particle sizes of selected wastes and the non homogeneity of the mixture complicated sampling a small amount of the composting mass that had a same ratio of given wastes.
Temperature (°C) Table 2 Dimensions and characteristics of process and calculated reaction enthalpy, (DHr).
Fig. 5. Correlation between oxygen concentration and temperature presented: (a) in the real time (r = 0.7818, P < 0.0001); (b) when one set of experimental data (temperature or oxygen concentration) was time shifted for 9.5 h towards another (r = 0.8537, P < 0.0001).
14.80 3.53 0.31 1.01 1.30 20–23 3.5 3106
Table 3 Results of estimated parameters. n m k0 (d1) SDtemperature SDconversion
1.37 1.63 0.0121 0.043 2.46
SD; mean square deviation.
60
80
40 60 30 40 20 Temperature - experiment Temperature - model Conversion - experiment Conversion - model
10
20
0
Conversion of VS (%)
100
50
Temperature (°C)
the thermophilic phase when the peak value of 5.38 g O2 h1 at day 6 was reached. Those peak values occurred at temperatures of 38.2 and 54.5 °C, respectively, what is in agreement with the literature (Liang et al., 2003). Total oxygen uptake was calculated at the end of the experiment and it was 1.760 kg. Oxygen concentration curve (Fig. 4) was almost a ‘‘mirror image’’ of the temperature curve (Fig. 2) obtained over composting time. Temperature and oxygen concentration of the outlet gases were negatively correlated (Fig. 5a) with r = 0.7818, P < 0.0001 which was satisfying but not as high as expected (Ryckeboer et al., 2003). It was found that peak oxygen uptakes were recorded at days 1, 3 and 6 whilst temperature peak values were recorded 6.0, 12.0 and 6.2 h after those peaks, respectively. The temperature and oxygen concentration data were then time shifted one towards another for the set of thirteen time intervals (6.0–12.0 h, step 0.5 h) and the correlation coefficients were calculated for each time shift. The highest correlation coefficient (r = 0.8537, P < 0.0001) was calculated when one set of data was time shifted for 9.5 h towards another. The time shifted correlation between temperature and oxygen concentration is presented on Fig. 5b. Observed delay of temperature peak values could be explained by the methods and reactor type used in the experiment. Namely, oxygen concentration was measured in the outlet gas stream and it reflected biodegradation rate in whole composting mass, while the temperature was measured at the middle of the height of the reactor. Therefore, if the highest biodegradation had occurred locally in the lower or upper regions of the composting mass, the delay could be the time needed for thermal conduction and convection of released metabolic heat (de Guardia et al., 2012), what mainly depends on thermal properties of the substrate, and airflow rate.
m0 (kg) mVS0 (kg) QV (L min1 kgVS01) cpa (kJ kg1 K1) qa (kg m3) T (oC) cps (kJ kg1 K1) (DHr) (kJ kgVS01)
0 0
2
4
6
8
10
12
14
16
18
20
22
Time (d) Fig. 6. Experimental and model-predicted temperature values and conversion of volatile solids over time.
328
N. Kopcˇic´ et al. / Waste Management 34 (2014) 323–328
4. Conclusions The apple waste was proved to be an adequate composting cosubstrate with the tobacco waste originating from the tobacco primary production. Temperature, evolution, oxygen consumption and changes in micro flora during the composting experiment showed regular composting performance. The C/N ratio of the raw compost was 11.6 with the pH value of 8.8 and during the process 53.1% of volatile solids were degraded. Accordingly, the mixture prepared of an apple and tobacco waste (1:7, d.w.) proved suitable for the composting process with regard to the physical– chemical and microbiological properties. The temperature curve was almost a ‘‘mirror image’’ of the oxygen concentration curve obtained over composting time, but with time delay of peak values of 9.5 h. The highest cumulative oxygen uptake rate was obtained at the day 1 (5.96 g O2 h1) and during the thermophilic phase. Total oxygen consumption during the experiment was 1.760 kg. Evaluation of selected mathematical model showed good prediction of temperature but over prediction of the conversion of VS during the most of the composting time. The over prediction of the conversion of VS was assigned to an experimental error because of the taking a representative sample in relatively small volume of non homogeneous material was disabled. Temperature measuring and sampling along the reactor height, including the bottom and the top of the composting mass, should be investigated in future works. Those findings would be useful for process modelling and optimization as well as for reactor selection and/or designing. Furthermore, stabilization and maturation of the obtained raw compost should be examined along with more detailed physical– chemical analysis to characterize and evaluate the quality of the final product.
Acknowledgement This work is a part of research project 125-1251963-1968 supported by Ministry of Science, Education and Sport, Republic of Croatia.
References Adhikari, B.K., Barrington, S., Martinez, J., King, S., 2008. Characterization of food waste and bulking agents for composting. Waste Manage. Res. 28, 795–804. An, C.-J., Huang, G.-H., Yao, Y., Sun, W., An, K., 2012. Performance of in-vessel composting of food waste in the presence of coal ash and uric acid. J. Hazard. Mater. 203–204, 38–45. Bautista, J.M., Kim, H., Ahn, D.-H., Zhang, R., Oh, Y.-S., 2011. Changes in physicochemical properties and gaseous emissions of composting swine manure amended with alum and zeolite. Korean J. Chem. Eng. 28, 189–194. Briški, F., Kopcˇic´, N., C´osic´, I., Kucˇic´, D., Vukovic´, M., 2012. Biodegradation of tobacco waste by composting: Genetic identification of nicotine-degrading bacteria and kinetic analysis of transformations in leachate. Chem. Pap. 66, 1103–1110. Briški, F., Gomzi, Z., Horgas, N., Vukovic´, M., 2003. Aerobic composting of tobacco solid waste. Acta Chim. Slov. 50, 697–714. Coalla, H.L., Fernández, J.M.B., Morán, M.A.M., Bobo, M.R.L., 2009. Biogas generation apple pulp. Bioresour. Technol. 100, 3843–3847. Dalsenter, F.D.H., Viccini, G., Barga, M., Mitchell, D.A., Krieger, N., 2005. A mathematical model describing the effect of temperature variations on the kinetics of microbial growth in solid-state culture. Process Biochem. 40, 801– 807. DelBuono, N., Mastroserio, C., 2002. Explicit methods based on a class of four stage fourth order Runge–Kutta methods for preserving quadratic laws. J. Comput. Appl. Math. 140, 231–243.
Dhillon, G.S., Brar, S.K., Verma, M., Tyagi, R.D., 2011. Utilization of different agroindustrial wastes for sustainable bioproduction of citric acid by Aspergillus niger. Biochem. Eng. J. 54, 83–92. Diaz, L.F., Savage, G.M., 2007. Factors that Affect the Process. In: Diaz, L.F., de Bertoldi, M., Bidlingmaier, W. (Eds.), Compost Science Technology. Elsevier, Amsterdam, pp. 49–64. Diaz, M.J., Madejón, J.E., López, F., López, R., Cabrera, F., 2002. Optimization of the rate vinasse/grape marc for co-composting process. Process Biochem. 37, 1143– 1150. Fernández, F.J., Sánchez-Arias, V., Villaseñor, J., Rodríguez, L., 2008. Evaluation of carbon degradation during co-composting of exhausted grape marc with different biowastes. Chemosphere 73, 670–677. Fourti, O., Jedid, N., Hassen, A., 2008. Behaviour of main microbiological parameters and of enteric microorganisms during the composting of municipal solid wastes and sewage sludge in a semi-industrial composting plant. Am. J. Environ. Sci. 4, 103–110. de Guardia, A., Petiot, C., Benoist, J.C., Druilhe, C., 2012. Characterization and modelling of the heat transfers in a pilot-scale reactor during composting under forced aeration. Waste Manage. 32, 1091–1105. Ghaly, A.E., Alkoaik, F., Snow, A., 2006. Thermal balance of invessel composting of tomato plant residues. Can. Biosyst. Eng. 48, 6.1-6.11. Haug, R.T., 1993. The Practical Handbook of Compost Engineering. Lewis, Boca Raton. van Heerden, I., Cronje, C., Swart, S.H., Kotze, J.M., 2002. Microbial, chemical and physical aspects of citrus waste composting. Bioresour. Technol. 81, 71–76. Insam, H., de Bertoldi, M., 2007. Microbiology of the Composting Process. In: Diaz, L.F., de Bertoldi, M., Bidlingmaier, W. (Eds.), Compost Science Technology. Elsevier, Amsterdam, pp. 26–45. Iqbal, M.K., Shafiq, T., Ahmed, K., 2010. Characterization of bulking agents and its effects on physical properties of compost. Bioresour. Technol. 101, 1913–1919. Kim, J.-D., Park, J.-S., In, B.-H., Kim, D., Namkoong, W., 2008. Evaluation of pilot-scale in-vessel composting for food waste treatment. J. Hazard. Mater. 154, 272–277. Kopcˇic´, N., Vukovic´ Domanovac, M., Ðakovic´, Z., Briški, F., 2013. Composting of tobacco dust in different types of reactors. Chem. Biochem. Eng. Q. 27, 57–64. Kumar, M., Ou, Y.-L., Lin, J.-G., 2010. Co-composting of green waste and food waste at low C/N ratio. Waste Manage. 30, 602–609. Liang, A., Das, K.C., McClendon, R.W., 2003. The influence of temperature and moisture contents regimes on the aerobic microbial activity of a biosolids composting blend. Bioresour. Technol. 86, 131–137. Manios, T., 2004. The composting potential of different organic solid wastes: experience from the island of Crete. Environ. Int. 29, 1079–1089. Neklyudov, A.D., Fedotov, G.N., Ivankin, A.N., 2008. Intensification of composting processes by aerobic microorganisms: a review. Appl. Biochem. Microbiol. 44, 6–18. Nelder, J.A., Mead, R., 1965. A simplex method for function minimization. The Comput. J. 7, 308–313. ÖNORM, 1986. S 2023. Analytical Methods and quality Control for Wastecompost. Austrian Standardization Institute, Vienna. Pognani, M., Barrena, R., Font, X., Sánchez, A., 2012. Effect of freezing on the conservation of the biological activity of organic solid wastes. Bioresour. Technol. 104, 832–836. Puyuelo, B., Gea, T., Sánchez, A., 2010. A new control strategy for the composting process based on the oxygen uptake rate. Chem. Eng. J. 165, 161–169. Piotrowska-Cyplik, A., Olejnik, A.P., Cyplik, J., Czarnecki, Z., 2009. The kinetics of nicotine degradation, enzyme activities and genotoxic potential in the characterization of tobacco waste composting. Bioresour. Technol. 100, 5037– 5044. Rashad, F.M., Saleh, W.D., Moselhy, M.A., 2010. Bioconversion of rice straw and certain agro-industrial wastes to amendments for organic farming systems: 1. Composting, quality, stability and maturity indices. Bioresour. Technol. 101, 5952–5960. Ryckeboer, J., Mergaert, J., Coosemans, J., Deprins, K., Swings, J., 2003. Microbiological aspects of biowaste during composting in a monitored compost bin. J. Appl. Microbiol. 94, 127–137. Sundberg, C., Franke-Whittle, I.H., Kauppi, S., Yu, D., Romantschuk, M., Insam, H., Jönsson, H., 2011. Characterisation of source-separated household waste intended for composting. Bioresour. Technol. 102, 2859–2867. Sundberg, C., Smårs, S., Jönsson, H., 2004. Low pH as an inhibiting factor in the transition from mesophilic to thermophilic phase in composting. Bioresour. Technol. 95, 145–150. Tso, T.C., 1990. Production, Physiology, and Biochemistry of Tobacco Plant. IDEALS Inc., Bektsville, Maryland. Vlyssides, A.G., Barampouti, E.M.P., Mai, S.T., 2008. Physical characteristics of olive stone wooden residues: possible bulking material for composting process. Biodegradation 19, 209–214. Yu, H., Huang, G.H., 2009. Effects of sodium acetate as a pH control amendment on the composting of food waste. Bioresour. Technol. 100, 2005–2011.
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Evaluation of laboratory-scale in-vessel co-composting of tobacco and apple waste Nina Kopcˇic´ ⇑, Marija Vukovic´ Domanovac, Dajana Kucˇic´, Felicita Briški Faculty of Chemical Engineering and Technology, University of Zagreb, Marulic´ev trg 19, 10000 Zagreb, Croatia
a r t i c l e
i n f o
Article history: Received 17 July 2013 Accepted 4 November 2013 Available online 28 November 2013 Keywords: Co-composting Apple waste Tobacco waste Oxygen Temperature Process modelling
a b s t r a c t Efficient composting process requires set of adequate parameters among which physical–chemical properties of the composting substrate play the key-role. Combining different types of biodegradable solid waste it is possible to obtain a substrate eligible to microorganisms in the composting process. In this work the composting of apple and tobacco solid waste mixture (1:7, dry weight) was explored. The aim of the work was to investigate an efficiency of biodegradation of the given mixture and to characterize incurred raw compost. Composting was conducted in 24 L thermally insulated column reactor at airflow rate of 1.1 L min1. During 22 days several parameters were closely monitored: temperature and mass of the substrate, volatile solids content, C/N ratio and pH-value of the mixture and oxygen consumption. The composting of the apple and tobacco waste resulted with high degradation of the volatile solids (53.1%). During the experiment 1.76 kg of oxygen was consumed and the C/N ratio of the product was 11.6. The obtained temperature curve was almost a ‘‘mirror image’’ of the oxygen concentration curve while the peak values of the temperature were occurred 9.5 h after the peak oxygen consumption. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Agro-food industries generate large quantities solid and liquid wastes which must be suitably managed before being discharged to the environment. The solid wastes, which originate from the food processing and agricultural industries, are suitable substrates for composting (Adhikari et al., 2008; Kim et al., 2008; Manios, 2004). Composting is acceptable solution because it reduces the volume of bulky solid waste and results with organic material becoming a stable end product (Haug, 1993; Insam and de Bertoldi, 2007). Successful composting requires meticulous attention to key parameters and that includes: (i) substrate nature parameters (e.g. C/N ratio, particle size, pH-value) and (ii) process parameters (e.g. temperature, moisture content (MC) and aeration rate), (Diaz and Savage, 2007; Haug, 1993; Rashad et al., 2010). Preferable C/N ratio of the composting substrate is between 25 and 30 (Diaz and Savage, 2007). Composting of the wastes with C/N ratio lower than 20 is feasible (Kumar et al., 2010), but with loss of nitrogen through ammonia volatilization during the process (Diaz and Savage, 2007). A pH-value between 5.5 and 8.5 is optimal for the compost microorganisms, but preferable values are between 6.5 and 7.5. Values out of the optimal range could inhibit microbial activity and initial lag in the composting rate could be expected leading to rate ⇑ Corresponding author. Tel.: +385 1 4597271; fax: +385 1 4597260. E-mail address: [email protected] (N. Kopcˇic´). 0956-053X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.wasman.2013.11.001
limitations (Sundberg et al., 2004; Yu and Huang, 2009). The particle size of the waste intended for composting is in the range of 10 mm (in forced aeration systems) to 50–100 mm (in passive aeration systems) (Neklyudov et al., 2008). If the particle size is too small, air recirculation through composting mass is inhibited, free air space in the system decreases and reduces oxygen diffusion (Vlyssides et al., 2008). Concerning those aspects, there are not many types of biodegradable waste that can be characterized as ‘‘ideal’’ substrates for composting and mixing with chemical and/ or bulking agents is often necessary (An et al., 2012; Iqbal et al., 2010; Yu and Huang, 2009). On the other hand, substrate with physical–chemical characteristics optimal for composting can be obtained by mixing two or more types of biodegradable waste (Diaz et al., 2002; Iqbal et al., 2010; Sundberg et al., 2011; Rashad et al., 2010). Apple waste, namely apple pomace has many applications, e.g. apple pomace as a substrate for the production of citric acid (Dhillon et al., 2011) and apple pulp as a biogas production source (Coalla et al., 2009), but it is rare as composting substrate. The main characteristics that makes fruit waste not preferable or not acceptable for composting are low pH-value (Dhillon et al., 2011) and high moisture content (van Heerden et al., 2002; Kim et al., 2008). On the other hand, tobacco solid waste is rich in nitrogen (Tso, 1990) and the consequence of that is a lower C/N ratio (Piotrowska-Cyplik et al., 2009). As reported in recent researches, tobacco waste from cigarette manufacturing (Briški et al., 2003; Briški et al., 2012) and tobacco waste briquettes (Piotrowska-Cyplik et al., 2009) was
324
N. Kopcˇic´ et al. / Waste Management 34 (2014) 323–328
successfully composted in the reactor systems. However, tobacco waste from primary tobacco production is a dry and dusty material consisted of very small particles and mechanical or manual stirring was necessary for efficient composting of that kind of waste in a packed bed reactor (Kopcˇic´ et al., 2013). Both apple and tobacco waste showed their biological potential for microbial degradation in fermentation and/or aerobic biodegradation processes but with certain disadvantages in their physical– chemical characteristics for efficient composting, as described above. An alternative to overcome such disadvantages and to recycle both wastes is the co-composting of selected wastes in the proper ratio (Fernández et al., 2008; Manios, 2004; Vlyssides et al., 2008). The aim of the work was to investigate feasibility of composting of the mixture of apple and tobacco solid waste in the laboratoryscale thermally insulated reactor. The study objectives were: (a) to determine physical–chemical properties of apple waste and tobacco waste, (b) to evaluate the efficiency of co-composting of selected wastes mixed in proper ratio, (c) to better understand correlation between oxygen uptake and temperature, and (d) to evaluate proposed mathematical model. 2. Materials and methods 2.1. Composting materials Mixture of apple waste and tobacco solid waste were used as a composting substrate. Apples were collected in autumn on orchard in northern part of Croatia. Apples were manually pealed and non-edible parts were chopped into pieces with sizes ranging from 1 to 50 mm, where about 70% of the particles were exceeded 20 mm. The apple waste used for the experiment was consisted of an apple skin, cores with seeds, stems and decaying parts of apple pulps and the waste was kept in the freezer at 18 °C until composting experiment. The tobacco waste was collected from the tobacco primary production in Hrvatski duhani d.d. Virovitica, Croatia. The waste was composed of tobacco leaf residues and very fine, powdery mixture of tobacco residues and soil particles. The tobacco waste was stored in the dry place until further use. Prior the experiment major physical–chemical properties of each waste material were determined: dry solids (DS), moisture
content (MC), volatile solids (VS), carbon (C) and nitrogen (N) content and pH-value. The waste mixture was designed to have a carbon to nitrogen ratio (C/N) of 25 and initial moisture content of 60% (w.w.). According to the obtained C/N ratio of each material and to the targeted value of C/N ratio of the composting substrate, apple and tobacco waste were mixed in the ratio of 1:7 (dry weight). Before composting, pH-value of the mixture was checked and water was added in order to obtain intended MC of the composting substrate. 2.2. Experimental setup The reactor was 24 L cylindrical, vertically positioned with a diameter of 190 mm and the height of 940 mm (Fig. 1). The reactor was of polyethylene and insulated with 50 mm thick polyurethane. Reactor was placed on the scale (M-3-1086 Sklad, Primjer, Zagreb, Croatia) in order to regularly check the changes of the mass of the substrate. The day before the experiment frozen apple waste was placed in a refrigerator at 4 °C for 24 h to defrost. The initial mixture was prepared of 5.4 kg of the tobacco waste and 3.7 kg of the defrosted apple waste. To ensure the intended moisture content, 5.7 L of tap water was added to the waste mixture. Before the experiment, initial pH-value, moisture content (MC) and C/N ratio of the mixture were determined. The reactor was filled from the top with 14.8 kg of solid waste mixture. The substrate was forcefully aerated with constant airflow rate set to 1.1 L min1 (equivalent to 0.31 L min1 kg1init.volat.solids) from the bottom of the reactor during 22 days of the experiment. The airflow rate was selected on the basis of the preliminary experiment (results not presented) as high enough to overcome the pressure drop through the composting mass and to obtain a constant outgoing gas flow for oxygen monitoring purposes. The trial length of the experiment of 22 days was considered representative of the overall composting process evolution. Upward aeration was provided by an air compressor (DE 50/204 FIAC, Italy) and output pressure of the air was set to 1.0 bar. Air flow rate was regulated by valve of an airflow-meter of measuring range of 0–2.313 L min-1 (Cole Parmer, USA,). In order to reduce and minimize drying of the substrate bed in the reactor, air was saturated with water by passing through Dreschel bottle (not presented in the scheme, Fig. 1) before entering the reactor.
Fig. 1. Schematic diagram of composting process: 1-air compressor and flow regulation; 2-composting reactor; 3-scale; 4-temperature probes; 5-data acquisition, monitoring and storage; 6-condensate collecting; 7-gas chromatograph.
N. Kopcˇic´ et al. / Waste Management 34 (2014) 323–328
Two temperature sensors were used; one placed at the middle of the reactor’s height (i.e. 470 mm above the base plate, on the central vertical axis) and the other at the air inlet of the reactor (Fig. 1). 2.3. Physical–chemical and microbiological monitoring and analyses Temperatures at the reactors inlet and in the middle of the reactor height were monitored by means of the RTD compact probes (±1%, Cole-Parmer, USA) connected to the computer by the 2-chanell RTD analogue input (SCC-RTD01, National Instruments, USA). Computer software used for that purpose was NI-DAQ 7, LabVIEW Win/PCI-6024/CB_68LP starter Kit, National Instruments, USA. Temperature data was programmed to be recorded and stored every 15 min. Oxygen content in outlet gas stream was recorded at preset times with gas chromatograph Multiple Gas Analyzer #2 (SRI, USA) was equipped with 1 m long Molecular sieve 5A column and thermal conductivity detector. The carrier gas was Helium 5.0 (Messer, Croatia) and gas analysis was carried out at helium flow rate of 20 mL min1 and at oven temperature of 60 °C. During the experiment most relevant physical–chemical and microbiological parameters were determined in samples of composting mass. Samples were taken daily from the top of the reactor at the middle point of the composting mass. At the beginning and at the end of the process mass of the substrate mixture/raw compost was determined and samples were taken in order to analyze the most relevant physical–chemical parameters. The pH-value was determined in a filtrate of a water extract (1:10 by weight) of the substrate sample after shaking for 30 min on the magnetic stirrer with pocket electrode (HI 98128, ±0.01, Hanna Instruments, Deutschland). The dry solids content (DS) (as well as MC) and volatile solids (VS) content were determined on duplicate samples at 105 °C and 550 °C, respectively, until the constant mass was obtained. The total nitrogen was determined as N-Kjeldahl with Digestor 2006 and distillation unit Kjeltec 2100, FOSS, USA. All analyses were carried out according to the Austrian S 2023 standard methods (ÖNORM, 1986.) for analysis of compost. Microflora was monitored through the estimation of mesophilic and thermophilic bacteria and fungi. Viable plate count was determined by the decimal dilution method and the results expressed as colony-forming units (CFU) of bacteria and fungi per gram of dry matter content (Briški et al., 2003). Mesophilic and thermophilic bacteria were enumerated on the nutrient agar (Biolife, Italy) after 24–48 h incubation at 37 °C and 50 °C, respectively. Mesophilic and thermophilic fungi were enumerated on the malt agar (Biolife, Italy) after 3 days incubation at 28 °C and 50 °C, respectively.
325
where mVS0 is initial mass of volatile solids (kg), T and Tin are temperatures (°C) of the substrate and of the air at reactor’s inlet, hw(T) is the heat of vaporization (kJ kg1) and HS(T) and HS0 are saturated humidities (kg kg1) at reactor and ambient temperature. The humidity of the air leaving the reactor (HS(T)) is calculated in two steps (Dalsenter et al., 2005). First the vapour pressure, pvp (Pa) is calculated according the Antoine equation:
pvp ¼ 133:322exp 18:3036 -
3816:44 ðT þ 273:15Þ 46:13
ð2Þ
The humidity of the air at saturation is then given by:
HS ¼ 0:62413
pvp p pvp
ð3Þ
where p is the total pressure (Pa) of the system and the factor 0.62413 is the ratio of the molecular weight of water to the average molecular weight of dry air. In the case of the air entering the bottom of the bed, T is replaced with Tin, and the vapour pressure (pvp) and humidity (HS0) of the entering air are calculated with the same equations. Eq. (1) was assigned to a mathematical model developed using the kinetic model and mass and energy balances and was reported in previous work (Briški et al., 2003). Modeling of the reaction kinetics was based on the simple process chemistry and other characteristics of the reactor selected for the experiment. Accordingly, an empirical equation was chosen as a kinetic model, Eq. (4):
rVS ¼ kðTÞ wVS expðntÞ þ wm VS
ð4Þ
where rVS is the degradation rate (d1), k(T) is the specific rate (d1), wVS is the mass fraction of volatile matter (kgVS kg1VS initial), t is the time (d) and n and m are adjustable parameters. To estimate dependence of the reaction rate on temperature, the empirical expression, Eq. (5), taken from the literature was applied (Haug, 1993):
h i kðTÞ ¼ k0 1:066T22 1:21T60
ð5Þ
Reactor was modelled as a continuous stirred tank reactor (CSTR) and several assumptions were used in developing the model (Briški et al., 2003): biodegradation of the substrate was slow compared to oxygen transfer through the boundary gas layer; oxygen concentration in the substrate bed was constant; airflow rate during composting was constant. It was also assumed that the process had been carried out under adiabatic conditions and that the released heat was proportional to the progress of biodegradation.
2.4. Statistical analysis
3. Results and discussion
Most of the results shown in present paper are the mean of two replicate analyses. Single determinations were made for temperature and oxygen concentration of outlet gases. Correlation coefficients of temperature and oxygen concentration of the outlet gases were expressed as Pearson’s r.
The main physical–chemical characteristics of selected wastes and prepared mixture are summarized in Table 1. Effect of freezing on microbial population and biological activity of the apple waste was not investigated but it was presumed that it was not affected (Pognani et al., 2012).
2.5. Mathematical model of the composting process The reaction enthalpy, (DHr) (kJ kgVS initial1) was calculated from experimental results by measuring the temperature of the composting mass during the reaction. The value of (DHr) was calculated using the following equation (Eq. (1)):
ðDHr Þ ¼
qa Q v mVS0 ð1 wVS Þ
Z
t
cpa ðT T in Þ þ hwðTÞ HSðTÞ HS0 dt
0
ð1Þ
Table 1 Physical–chemical characteristics of the waste and mixture used in the experiment. Substrate
Tobacco waste
Apple waste
Waste mixture
Mass of wet substrate, m0 (kg) Moisture content, MC (%, w.w) Volatile solids content, VS (%, d.w.) pH-value C/N ratio
5.4 5.8 57.3 6.8 19.7
3.7 80.1 96.9 4.6 185.6
14.8 62.0 62.8 6.1 22.9
log CFU (g-1 substrate)
12
60
10
50
8
40
6
30
4
20 mesophillic bacteria mesophillic fungi Temperature
2 0 0
2
4
6
8
10
termophillic bacteria termophillic fungi
Temperature (°C)
N. Kopcˇic´ et al. / Waste Management 34 (2014) 323–328
326
10 0
12
14
16
18
20
22
Time (d) Fig. 2. Temperature fluctuations and growth of total viable–cultivable mesophilic and thermophilic bacteria and fungi over time in composting mass.
24
10,0
22
9,0 8,0
18
7,0
16 14
6,0
pH-value
C/N ratio
20
12 C/N
10
5,0
pH
8
4,0 0
2
4
6
8
10
12
14
16
18
20
22
Time (d) Fig. 3. Changes of pH-value and carbon to nitrogen (C/N) ratio during 22 days of composting.
3.1. Changes in physical–chemical and microbiological properties of the substrate during composting The temperature in the composting mass increased rapidly following a short lag phase, and temperatures were subsequently maintained in the thermophilic region for period typical of those occurring in laboratory-scale composting operations (Ghaly et al., 2006). This behavior is illustrated in Fig. 2, which shows the temperature profiles measured at the mid-point of the reactor. (i.e. 450 mm above the base plate, on the central vertical axis). The fluctuation of temperature recorded during the experiment showed common composting phases. At the start of the process the number of mesophilic and thermophilic bacteria in 1 g of the dry substrate was 5.0 106 and 3.8 106, respectively. After the first 12 h of the lag phase of the mesophilic microflora, just slight increase in number of thermophilic bacteria was observed indicating the lag phase of thermophilic microorganisms. Lower temperatures and lower pH-value during the first five days of composting were favorable for fungi and their activity was high (van Heerden et al., 2002; Ryckeboer et al., 2003). The waste mixture was predominantly colonized by mesophilic yeasts which were thermally inactivated in the second phase of the process. Temperature values above 45 °C occurred at day 5 and lasted for 8 days while peak value of 54.7 °C was reached at day 7. The number of mesophilic and thermophilic bacteria reached their maximum at day 6 and it was 2.9 109 and 1.1 109, respectively. After the peak temperature, the drop of total bacteria count was observed which is commonly reported in the literature (Fourti et al., 2008; Ryckeboer et al., 2003). After the day 13 the composting mass started to cool slowly when. yeasts were replaced with bacteria and the number of both bacteria and fungi tended to stabilize. By mixing apple and tobacco waste in the ratio 1:7 (d.w.) substrate with pH-value of 6.3 adequate for composting was obtained
(van Heerden et al., 2002). A decrease of pH was recorded for the given mixture during the first three days (Fig. 3) which (pH < 6.5) presumably inhibited thermophilic microorganisms and caused lag phase in the transition from mesophilic to thermophilic conditions (Bautista et al., 2011), as discussed above (Fig. 2). After that, day after day, the pH started to increase and levelled around the value 8.8 after day 11. Slight decrease of pH-value in the initial stage of the process was probably due to production of organic acids that resulted from fast degradation of easily degradable organic matter (e.g. carbohydrates which are present in apple waste in high concentration (Dhillon et al., 2011)) by mesophilic microorganisms. After day 4, when thermophilic phase occurred, pH increased due to ammonia evolution and nitrogen losses from the substrate (Bautista et al., 2011). The final pH-value of 8.8 is in agreement with the literature for the raw compost and further maturation would lead to decrease of pH (Kim et al., 2008). Fig. 3 also shows the variation of C/N ratio with time. The initial C/N ratio of the feed mixture was 23 and it was in the range of optimal values for composting (Adhikari et al., 2008). The C/N ratio was continuously dropping until the end of the experiment (day 22) when it was 11.6. It is well known that a C/N ratio below 20 is indicative of proper compost maturity, with ratio of 15 or less being preferred (van Heerden et al., 2002; Kim et al., 2008). The degree of degradation or the conversion of organic matter was calculated from the volatile solid contents taking into account the changes in mass of the material inside the reactor during the composting. During the process conversion of organic matter increased and the increase was more pronounced during the first mesophilic stage of the process. The composting of most substrates is characterized by an initial period of rapid degradation followed by longer period of slower degradation (Diaz and Savage, 2007; Diaz et al., 2002; Haug, 1993). At the end of the process conversion of volatile matter of 53.1% was calculated. It should be noted that values of pH, C/N ratio, MC and VS were determined in the samples taken from the middle point of the reactor. In order to evaluate the process more representatively sampling along the reactor height, including the bottom and the top of the composting mass, should be performed as in the work of Kopcˇic´ et al., 2013. At the end of the experiment the composting material was removed from the reactor. At the time, raw compost was granular and dark material of pleasant odour. The top of the composting mass was overgrown by actinobacteria what indicated the proper moisture and good oxygen supply in higher regions of the reactor. 3.2. Oxygen uptake The usual procedure in system design is the maintenance of aerobiosis in excess of that required for an efficient biodegradation in composting process. The recommended lower limit of the oxygen concentration in the composting mass was reported to be 5% (v/v) (Haug, 1993; Puyuelo et al., 2010). As shown in Fig. 4, that value was not recorded during the whole period of composting what indicated a proper aeration rate (Kim et al., 2008; Puyuelo et al., 2010). The highest microbial activity based on oxygen uptake is also well shown in Fig. 4. The cumulative oxygen pattern was derived from oxygen concentration profile using the ideal gas law. During the first 12 h the cumulative oxygen uptake curve indicated a lag phase of the process. Average value of cumulative oxygen uptake rate at this stage was 1.25 g O2 h1. After that, at day 1 oxygen content in the exhaust gas reached the lowest value of 13.9% (v/v) since the presence of readily biodegradable materials caused high oxygen consumption by mesophilic microorganisms. That led to an exponential growth of mesophilic microflora and the release of metabolic heat. As a consequence of it temperature in the composting mass increased (Fig. 2). The highest cumulative oxygen uptake rate was obtained at the day 1 (5.96 g O2 h1) and during
O2 concentration (%, v/v)
24
2000 1800
22
1600 20
1400
18
1200 1000
16
800 600
14 12
O2 concentration
400
cumulative O2
200
10
0 0
2
4
6
8
10
12
14
16
18
20
Cumulative O 2 consumption (g)
N. Kopcˇic´ et al. / Waste Management 34 (2014) 323–328
22
Time (d) Fig. 4. Oxygen concentration and cumulative oxygen consumption during 22 days of composting.
a
22
b
20
20
18
18
16
16
14
14
12
O2 concentration (%, v/v)
O2 concentration (%, v/v)
22
12 20
30
40
50
60
20
30
40
50
60
327
3.3. Modeling the composting process For an application of the model and its experimental verification, several parameters and process variables are required. Necessary data for application of the proposed model are given in Table 2. Based on known input variables presented in Table 2, proposed dependence of the degradation rate on temperature (Eq. (5)) and calculated reaction enthalpy (DHr), (Eq. (1)), the kinetic parameters n, m and k0 in Eq. (4) were estimated (Table 3). Direct nonlinear regression analysis based on the Nelder–Mead simplex method (Nelder and Mead, 1965) was performed to determine k0, m and n in the kinetic model. After optimal values of kinetic parameters were determined, the set of differential equations were numerically solved using Runge– Kutta method (DelBuono and Mastroserio, 2002). The theoretical curves were drawn together with the actual data plot in Fig 6. The graphical presentation shows that theoretical temperature is in agreement with experimentally obtained values. The theoretical conversion of volatile solids was over-estimated against the experimental data (Fig. 6, Table 3). The lower values of experimentally obtained conversion were probably due to inability of taking a representative sample in relatively small volume of the material. Namely, the difference in particle sizes of selected wastes and the non homogeneity of the mixture complicated sampling a small amount of the composting mass that had a same ratio of given wastes.
Temperature (°C) Table 2 Dimensions and characteristics of process and calculated reaction enthalpy, (DHr).
Fig. 5. Correlation between oxygen concentration and temperature presented: (a) in the real time (r = 0.7818, P < 0.0001); (b) when one set of experimental data (temperature or oxygen concentration) was time shifted for 9.5 h towards another (r = 0.8537, P < 0.0001).
14.80 3.53 0.31 1.01 1.30 20–23 3.5 3106
Table 3 Results of estimated parameters. n m k0 (d1) SDtemperature SDconversion
1.37 1.63 0.0121 0.043 2.46
SD; mean square deviation.
60
80
40 60 30 40 20 Temperature - experiment Temperature - model Conversion - experiment Conversion - model
10
20
0
Conversion of VS (%)
100
50
Temperature (°C)
the thermophilic phase when the peak value of 5.38 g O2 h1 at day 6 was reached. Those peak values occurred at temperatures of 38.2 and 54.5 °C, respectively, what is in agreement with the literature (Liang et al., 2003). Total oxygen uptake was calculated at the end of the experiment and it was 1.760 kg. Oxygen concentration curve (Fig. 4) was almost a ‘‘mirror image’’ of the temperature curve (Fig. 2) obtained over composting time. Temperature and oxygen concentration of the outlet gases were negatively correlated (Fig. 5a) with r = 0.7818, P < 0.0001 which was satisfying but not as high as expected (Ryckeboer et al., 2003). It was found that peak oxygen uptakes were recorded at days 1, 3 and 6 whilst temperature peak values were recorded 6.0, 12.0 and 6.2 h after those peaks, respectively. The temperature and oxygen concentration data were then time shifted one towards another for the set of thirteen time intervals (6.0–12.0 h, step 0.5 h) and the correlation coefficients were calculated for each time shift. The highest correlation coefficient (r = 0.8537, P < 0.0001) was calculated when one set of data was time shifted for 9.5 h towards another. The time shifted correlation between temperature and oxygen concentration is presented on Fig. 5b. Observed delay of temperature peak values could be explained by the methods and reactor type used in the experiment. Namely, oxygen concentration was measured in the outlet gas stream and it reflected biodegradation rate in whole composting mass, while the temperature was measured at the middle of the height of the reactor. Therefore, if the highest biodegradation had occurred locally in the lower or upper regions of the composting mass, the delay could be the time needed for thermal conduction and convection of released metabolic heat (de Guardia et al., 2012), what mainly depends on thermal properties of the substrate, and airflow rate.
m0 (kg) mVS0 (kg) QV (L min1 kgVS01) cpa (kJ kg1 K1) qa (kg m3) T (oC) cps (kJ kg1 K1) (DHr) (kJ kgVS01)
0 0
2
4
6
8
10
12
14
16
18
20
22
Time (d) Fig. 6. Experimental and model-predicted temperature values and conversion of volatile solids over time.
328
N. Kopcˇic´ et al. / Waste Management 34 (2014) 323–328
4. Conclusions The apple waste was proved to be an adequate composting cosubstrate with the tobacco waste originating from the tobacco primary production. Temperature, evolution, oxygen consumption and changes in micro flora during the composting experiment showed regular composting performance. The C/N ratio of the raw compost was 11.6 with the pH value of 8.8 and during the process 53.1% of volatile solids were degraded. Accordingly, the mixture prepared of an apple and tobacco waste (1:7, d.w.) proved suitable for the composting process with regard to the physical– chemical and microbiological properties. The temperature curve was almost a ‘‘mirror image’’ of the oxygen concentration curve obtained over composting time, but with time delay of peak values of 9.5 h. The highest cumulative oxygen uptake rate was obtained at the day 1 (5.96 g O2 h1) and during the thermophilic phase. Total oxygen consumption during the experiment was 1.760 kg. Evaluation of selected mathematical model showed good prediction of temperature but over prediction of the conversion of VS during the most of the composting time. The over prediction of the conversion of VS was assigned to an experimental error because of the taking a representative sample in relatively small volume of non homogeneous material was disabled. Temperature measuring and sampling along the reactor height, including the bottom and the top of the composting mass, should be investigated in future works. Those findings would be useful for process modelling and optimization as well as for reactor selection and/or designing. Furthermore, stabilization and maturation of the obtained raw compost should be examined along with more detailed physical– chemical analysis to characterize and evaluate the quality of the final product.
Acknowledgement This work is a part of research project 125-1251963-1968 supported by Ministry of Science, Education and Sport, Republic of Croatia.
References Adhikari, B.K., Barrington, S., Martinez, J., King, S., 2008. Characterization of food waste and bulking agents for composting. Waste Manage. Res. 28, 795–804. An, C.-J., Huang, G.-H., Yao, Y., Sun, W., An, K., 2012. Performance of in-vessel composting of food waste in the presence of coal ash and uric acid. J. Hazard. Mater. 203–204, 38–45. Bautista, J.M., Kim, H., Ahn, D.-H., Zhang, R., Oh, Y.-S., 2011. Changes in physicochemical properties and gaseous emissions of composting swine manure amended with alum and zeolite. Korean J. Chem. Eng. 28, 189–194. Briški, F., Kopcˇic´, N., C´osic´, I., Kucˇic´, D., Vukovic´, M., 2012. Biodegradation of tobacco waste by composting: Genetic identification of nicotine-degrading bacteria and kinetic analysis of transformations in leachate. Chem. Pap. 66, 1103–1110. Briški, F., Gomzi, Z., Horgas, N., Vukovic´, M., 2003. Aerobic composting of tobacco solid waste. Acta Chim. Slov. 50, 697–714. Coalla, H.L., Fernández, J.M.B., Morán, M.A.M., Bobo, M.R.L., 2009. Biogas generation apple pulp. Bioresour. Technol. 100, 3843–3847. Dalsenter, F.D.H., Viccini, G., Barga, M., Mitchell, D.A., Krieger, N., 2005. A mathematical model describing the effect of temperature variations on the kinetics of microbial growth in solid-state culture. Process Biochem. 40, 801– 807. DelBuono, N., Mastroserio, C., 2002. Explicit methods based on a class of four stage fourth order Runge–Kutta methods for preserving quadratic laws. J. Comput. Appl. Math. 140, 231–243.
Dhillon, G.S., Brar, S.K., Verma, M., Tyagi, R.D., 2011. Utilization of different agroindustrial wastes for sustainable bioproduction of citric acid by Aspergillus niger. Biochem. Eng. J. 54, 83–92. Diaz, L.F., Savage, G.M., 2007. Factors that Affect the Process. In: Diaz, L.F., de Bertoldi, M., Bidlingmaier, W. (Eds.), Compost Science Technology. Elsevier, Amsterdam, pp. 49–64. Diaz, M.J., Madejón, J.E., López, F., López, R., Cabrera, F., 2002. Optimization of the rate vinasse/grape marc for co-composting process. Process Biochem. 37, 1143– 1150. Fernández, F.J., Sánchez-Arias, V., Villaseñor, J., Rodríguez, L., 2008. Evaluation of carbon degradation during co-composting of exhausted grape marc with different biowastes. Chemosphere 73, 670–677. Fourti, O., Jedid, N., Hassen, A., 2008. Behaviour of main microbiological parameters and of enteric microorganisms during the composting of municipal solid wastes and sewage sludge in a semi-industrial composting plant. Am. J. Environ. Sci. 4, 103–110. de Guardia, A., Petiot, C., Benoist, J.C., Druilhe, C., 2012. Characterization and modelling of the heat transfers in a pilot-scale reactor during composting under forced aeration. Waste Manage. 32, 1091–1105. Ghaly, A.E., Alkoaik, F., Snow, A., 2006. Thermal balance of invessel composting of tomato plant residues. Can. Biosyst. Eng. 48, 6.1-6.11. Haug, R.T., 1993. The Practical Handbook of Compost Engineering. Lewis, Boca Raton. van Heerden, I., Cronje, C., Swart, S.H., Kotze, J.M., 2002. Microbial, chemical and physical aspects of citrus waste composting. Bioresour. Technol. 81, 71–76. Insam, H., de Bertoldi, M., 2007. Microbiology of the Composting Process. In: Diaz, L.F., de Bertoldi, M., Bidlingmaier, W. (Eds.), Compost Science Technology. Elsevier, Amsterdam, pp. 26–45. Iqbal, M.K., Shafiq, T., Ahmed, K., 2010. Characterization of bulking agents and its effects on physical properties of compost. Bioresour. Technol. 101, 1913–1919. Kim, J.-D., Park, J.-S., In, B.-H., Kim, D., Namkoong, W., 2008. Evaluation of pilot-scale in-vessel composting for food waste treatment. J. Hazard. Mater. 154, 272–277. Kopcˇic´, N., Vukovic´ Domanovac, M., Ðakovic´, Z., Briški, F., 2013. Composting of tobacco dust in different types of reactors. Chem. Biochem. Eng. Q. 27, 57–64. Kumar, M., Ou, Y.-L., Lin, J.-G., 2010. Co-composting of green waste and food waste at low C/N ratio. Waste Manage. 30, 602–609. Liang, A., Das, K.C., McClendon, R.W., 2003. The influence of temperature and moisture contents regimes on the aerobic microbial activity of a biosolids composting blend. Bioresour. Technol. 86, 131–137. Manios, T., 2004. The composting potential of different organic solid wastes: experience from the island of Crete. Environ. Int. 29, 1079–1089. Neklyudov, A.D., Fedotov, G.N., Ivankin, A.N., 2008. Intensification of composting processes by aerobic microorganisms: a review. Appl. Biochem. Microbiol. 44, 6–18. Nelder, J.A., Mead, R., 1965. A simplex method for function minimization. The Comput. J. 7, 308–313. ÖNORM, 1986. S 2023. Analytical Methods and quality Control for Wastecompost. Austrian Standardization Institute, Vienna. Pognani, M., Barrena, R., Font, X., Sánchez, A., 2012. Effect of freezing on the conservation of the biological activity of organic solid wastes. Bioresour. Technol. 104, 832–836. Puyuelo, B., Gea, T., Sánchez, A., 2010. A new control strategy for the composting process based on the oxygen uptake rate. Chem. Eng. J. 165, 161–169. Piotrowska-Cyplik, A., Olejnik, A.P., Cyplik, J., Czarnecki, Z., 2009. The kinetics of nicotine degradation, enzyme activities and genotoxic potential in the characterization of tobacco waste composting. Bioresour. Technol. 100, 5037– 5044. Rashad, F.M., Saleh, W.D., Moselhy, M.A., 2010. Bioconversion of rice straw and certain agro-industrial wastes to amendments for organic farming systems: 1. Composting, quality, stability and maturity indices. Bioresour. Technol. 101, 5952–5960. Ryckeboer, J., Mergaert, J., Coosemans, J., Deprins, K., Swings, J., 2003. Microbiological aspects of biowaste during composting in a monitored compost bin. J. Appl. Microbiol. 94, 127–137. Sundberg, C., Franke-Whittle, I.H., Kauppi, S., Yu, D., Romantschuk, M., Insam, H., Jönsson, H., 2011. Characterisation of source-separated household waste intended for composting. Bioresour. Technol. 102, 2859–2867. Sundberg, C., Smårs, S., Jönsson, H., 2004. Low pH as an inhibiting factor in the transition from mesophilic to thermophilic phase in composting. Bioresour. Technol. 95, 145–150. Tso, T.C., 1990. Production, Physiology, and Biochemistry of Tobacco Plant. IDEALS Inc., Bektsville, Maryland. Vlyssides, A.G., Barampouti, E.M.P., Mai, S.T., 2008. Physical characteristics of olive stone wooden residues: possible bulking material for composting process. Biodegradation 19, 209–214. Yu, H., Huang, G.H., 2009. Effects of sodium acetate as a pH control amendment on the composting of food waste. Bioresour. Technol. 100, 2005–2011.
