JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 31 August 2007
Published in J Environ Qual 36:1512-1520 (2007)
DOI: 10.2134/jeq2006.0253
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Eklind, Y.
Right arrow Articles by Jönsson, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Eklind, Y.
Right arrow Articles by Jönsson, H.
Agricola
Right arrow Articles by Eklind, Y.
Right arrow Articles by Jönsson, H.
Related Collections
Right arrow Nitrogen
Right arrow Microbial Processes
Right arrow Municipal Waste

TECHNICAL REPORTS

Waste Management

Carbon Turnover and Ammonia Emissions during Composting of Biowaste at Different Temperatures

Ylva Eklinda, Cecilia Sundbergb,*, Sven Smårsb, Kristin Stegerc, Ingvar Sundhc, Holger Kirchmanna and Håkan Jönssonb

a Dep. of Soil Sciences, Swedish Univ. of Agricultural Sciences, P.O. Box 7014, SE-750 07 Uppsala, Sweden
b Dep. of Biometry and Engineering, Swedish Univ. of Agricultural Sciences, P.O. Box 7032, SE-750 07 Uppsala, Sweden
c Dep. of Microbiology, Swedish Univ. of Agricultural Sciences, P.O. Box 7025, SE-750 07 Uppsala, Sweden

* Corresponding author (cecilia.sundberg{at}bt.slu.se).

Received for publication June 30, 2006.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 Materials and Methods
 Results and Discussion
 Conclusions
 REFERENCES
 
The effects of different process temperatures (40, 55, and 67°C) during composting of source-separated household waste were studied in a 200 L compost reactor at an oxygen concentration of 16%. The overall decomposition measured as carbon mineralization, decomposition of different carbon constituents, and the dynamics of nitrogen mineralization and the microbial community, are reported. Ammonia emissions at 67°C were more than double those at lower temperatures, and they were lowest at 40°C. The decomposition rate, measured as CO2 emission, was highest at 55°C. Decomposition of crude fat was slower at 40°C than at 55 and 67°C. The peak in microbial biomass was largest in the run at 40°C, where substantial differences were seen in the microbial community structure and succession compared to thermophilic temperatures. Biowaste composting can be optimized to obtain both a high decomposition rate and low ammonia emissions by controlling the process at about 55°C in the initial, high-rate stage. To reduce ammonia emissions it seems worthwhile to reduce the temperature after an initial high-temperature stage.


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
Materials and Methods
 Results and Discussion
 Conclusions
 REFERENCES
 
COMPOSTING of biowaste (source-separated organic household waste) is increasing in Europe. Still, there is a need for more knowledge concerning the parameters influencing the composting process, to achieve efficient decomposition with minimal negative environmental impact.

The temperature of the compost affects both the dynamics of the microbial community (Strom, 1985) and the decomposition rate. Several studies have investigated the effect of temperature on the decomposition rate, but conclusions about the optimal temperature differ. Suler and Finstein (1977) found a higher decomposition rate around 50°C than at higher temperatures. Jeris and Regan (1973) showed maximum decomposition rates at 40 to 60°C for different raw materials. Campbell et al. (1990) and McKinley and Vestal (1984) found maximum decomposition at about 40°C, whereas Haug (1993) considered around 65°C to be optimal. Despite these differences, several reviews state that the fastest decomposition is to be expected at 52 to 60°C (Miller, 1993; Richard and Walker, 2006). However, since it is often difficult to regulate temperature, composting is often done at temperatures other than the microbial optimum. Small-scale composts (e.g., backyard composts) are often characterized by low temperatures, whereas large composting plants often have temperatures considerably higher than 60°C.

The formation of gases with undesirable environmental effects, such as ammonia, methane, and nitrous oxide, is affected by temperature, but the mechanisms are only partly understood. Ammonia volatilization is the most important process of nitrogen loss during composting (Martins and Dewes, 1992; Beck-Friis et al., 2001). Ammonia losses depend on the microbial formation of ammonium, but also on the chemical equilibria between ammonium, liquid ammonia, and gaseous ammonia, which are temperature-dependent. It is therefore expected that ammonia emissions increase at higher composting temperatures. However, since ammonia emissions from composts are also strongly affected by other factors such as pH (Dewes, 1996), aeration rate (Hong et al., 1998), C/N ratio (Kirchmann and Witter, 1989; Eiland et al., 2001), and availability of substrate carbon (Barrington et al., 2002), no studies, to our knowledge, have yet been able to isolate the effect of temperature.

Earlier research in our group has shown that it is possible to speed up composting considerably with a start-up strategy in which the temperature is not allowed to increase above 37°C for the first few days, until the pH increases above 5 (Smårs et al., 2002). This can be explained by a faster breakdown of short-chain organic acids, which inhibit the composting process when temperature increases above the mesophilic range (Sundberg et al., 2004).

In the present study, we investigated the composting process at three temperatures: 40, 55, and 67°C. The changes in temperature that occur during the composting process contribute to a succession of different microbial communities over time (McKinley and Vestal, 1985; Miller, 1993; Klamer and Bååth, 1998). Therefore, we expected that the active regulation of temperature during the initial stage of high decomposition would select for different microbial communities. Together with the varying physicochemical conditions, this was expected to influence the decomposition rate and ammonia emissions.

The overall objective of our reactor composting research is to investigate the effects of different composting conditions on the decomposition dynamics, including gaseous emissions, and on the microbial community succession. Earlier work has described composting under different oxygen conditions (Beck-Friis et al., 2003; Steger et al., 2005) and with different start-up strategies (Beck-Friis et al., 2001; Smårs et al., 2002). The present study focuses on the effects of different composting temperatures (40, 55, and 67°C) in a well-oxygenated composting system. The importance of temperature for overall decomposition, decomposition of different carbon constituents, and ammonia emissions is investigated. In addition, analyses of phospholipid fatty acids (PLFA) were used to determine whether temperature affected the dynamics of the microbial biomass and community structure.


   Materials and Methods
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results and Discussion
 Conclusions
 REFERENCES
 
Experimental Setup
We composted biowaste at 40, 55, and 67°C using a 200-L compost reactor (Fig. 1 , described in detail by Smårs et al., 2001) in which temperature and oxygen levels were controlled independently. The raw material was source-separated organic household waste, mainly food waste, collected on arrival at the central composting plant in Uppsala, Sweden on one occasion in February 1995. After sorting out impurities, the material was homogenized, frozen, and stored at –20°C for later use. A detailed analysis of the chemical composition of the waste was performed by Eklind et al. (1997). Before composting, the waste was slowly thawed and mixed with wheat straw as a structural amendment. The mixture was homogenized and sampled for physical, chemical, and microbial analyses (3–10 samples depending on analysis), and finally placed in the compost reactor. The mixture had an initial C/N ratio of 22:1 (Table 1). The set point for oxygen in the compost gas was 16% and the set point for water content in the compost media was 65% in all experiments.


Figure 1
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 1. A schematic diagram of the compost reactor.

 

View this table:
[in this window]
[in a new window]

  		Table 1. Composition of substrates used in the composting runs.

 
A start-up strategy was used in which the objective was to minimize the time of the initial acidic phase. The material self-heated until the temperature reached 37°C. Then the reactor was cooled to keep the temperature at 37 to 40°C while the pH was still low (Smårs et al., 2002). When the pH of the condensate from cooled compost gases increased above 5, the set point temperature was changed for each experiment to either 40°C (one run, "Run40"), 55°C (three runs, "Run55a;b;c") or 67°C (three runs, "Run67a;b;c"). Due to this start-up procedure, all seven runs were in principle replicates until the time of set point change, which was also chosen as the reference time point (Fig. 2 ). In Run55c and Run67c the temperature was lowered successively after the initial 9 d. The experiments were run for between 7 and 59 d.


Figure 2
View larger version (20K):
[in this window]
[in a new window]

  		Fig. 2. Temperature in the composting material. Time = 0 when the set point temperature was changed from 37°C to 40, 55, or 67°C, respectively. Short-term temperature fluctuations in the figure were caused by decreases in temperature during turning and sampling.

 
The reactor was thermally insulated by 25 mm foam rubber in two layers. The airflow rate in the reactor was constant at about 200 m3 h–1. Most of this airflow was recirculated, but to maintain the oxygen content at 16%, a regulator-controlled proportion of the cooled gas was replaced with fresh air (electrically controlled valve Badger Meter/Research Control 807/HH-500, Germany; airflow meter Meriam 50MJ10-10, USA). The amount of off-gas was measured with an integrating gas flow meter (Gallus 2000; TC, Denmark). Temperature was measured at four different heights in the reactor and several more in the gas stream and between the insulating material and the reactor. Temperature was controlled by a heater in the air circulation loop and a cooler in a separate loop (Fig. 1). Part of the compost gas was circulated through the cooler in which a condensate was formed.

The composting material was mixed by turning and rotating the reactor, initially daily but less frequently when the rate of decomposition had decreased. On each turning occasion, triplicate samples of the material were taken with rotation of the reactor between samples. After sampling, condensate was recycled to the reactor, in the same proportion, or less than, that with which the cooled gas was recycled to the reactor since the previous mixing. Thus the amount of ammonia and other water-soluble substances permanently removed from the process was the amount associated with the air exchange necessary to maintain the oxygen concentration. The reported NH3 emissions are the net emissions associated with the outgoing gas; the ammonia returned with the condensate is not included. To keep the water content of the compost as close to the set point (65%) as possible, the recycled condensate was often supplemented by appropriate amounts of deionized water.

Temperature, O2, and emissions of CO2 were measured automatically every 5 min during the experiment. Furthermore, the weight of the reactor contents and its dry matter content were measured in the beginning and at the end of the experiments. The dry matter loss for a certain period of time was calculated by assuming it was proportional to the CO2 emission and compensating for the samples removed from the reactor for chemical and microbial analyses.

Physical, Chemical, and Microbial Analyses
Temperature in the compost material was registered with temperature transducers (Pt-100, Heraeus). CO2 concentration in the compost gas was measured by a gas analyzer with IR-detector (Binos 4b.2, Germany), and O2 concentration with a paramagnetic oxygen gas analyzer (Servomex 1131, Great Britain). NH3 in the gas was measured with a photoacoustic multi-gas monitor (Type 1312, Innova, Denmark). NH3 in the condensate was analyzed by flow injection analysis (FiaStar 5010 analyzer, Tecator, Sweden). The pH was measured directly in the condensate and in the compost material after mixing with H2O at a weight ratio of 1:5 and shaking for 1 h.

Dry matter content was determined after 14 h at 105°C, and ash content after 4 h at 550°C. The contents of sugar, starch, crude fat, hemicellulose, cellulose, and lignin in the material were monitored. Sugar, expressed as sucrose, was determined enzymatically (Steen and Larsson, 1986). Starch was determined enzymatically with the thermostable amylase and amyloglucosidase assays (Bengtsson and Larsson, 1990). For determination of crude fat, samples were boiled with HCl and extracted with petroleumether (Anonymous, 1998). Cellulose, hemicellulose, and lignin were determined by the detergent system and permanganate oxidation of lignin as outlined by Van Soest (1982) (the NDF/ADF method). Short-chain organic acids were determined by HPLC after water extraction. Total organic carbon was measured by dry combustion and IR determination of evolved CO2 (LECO analyzer, USA). Total N was measured by the Kjeldahl method on wet, thawed material (Bremner and Mulvaney, 1982).

The microbial biomass and community structure was monitored by analysis of microbial PLFA according to Steger et al. (2003).


   Results and Discussion
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
Conclusions
 REFERENCES
 
As the same temperature control strategy was used for all runs during the start-up phase, they were initially replicates, until the pH increased above 5 in the condensate and the set point temperature was changed to 40, 55, or 67°C, respectively. This occurred 1.6 to 3.3 d after the material was placed in the reactor.

Homogeneous process conditions were maintained in the composting reactor by good insulation of the reactor, vigorous circulation of the process gas and frequent mixing of the material. The spatial temperature variation was less than 0.7°C within the compost material, and the variation in the oxygen content in the compost gas was less than 0.3% (Smårs et al., 2001).

Carbon Dioxide Emissions
Decomposition of the biowaste was fastest at 55°C (Table 2; Fig. 3 and 4 ). Decomposition tended to be faster at 67°C than at 40°C, but the difference was not significant. The peak CO2 emission rate was 17.1% of initial C d–1 at 55°C, but markedly lower at both 40 and 67°C (Table 2). The accumulated CO2–C release during the first 6 d at 40 and 67°C was 74 and 76% of that measured at 55°C, respectively.


View this table:
[in this window]
[in a new window]

  		Table 2. Carbon dioxide emissions at 40, 55, and 67°C.

 

Figure 3
View larger version (21K):
[in this window]
[in a new window]

  		Fig. 3. Carbon dioxide emissions in Run40, Run55a, and Run67a. Time = 0 when the set point temperature was changed from 37°C to 40, 55, or 67°C, respectively.

 

Figure 4
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 4. Cumulative carbon dioxide emissions in all runs at 40, 55, and 67°C. Arrows indicate start of successive decrease in temperature set point in Run55c and 67c. Time = 0 when the set point temperature was changed from 37°C to 40, 55, or 67°C, respectively.

 
Our peak CO2 emission rates were between 70 and 110% of those reported for laboratory scale experiments with real and synthetic food waste seeded with compost or microbial inoculum (Nakasaki et al., 1990; Nakasaki et al., 1998; Nakasaki et al., 2005; Chang et al., 2006). The use of seed to give a rapid start probably results in faster decomposition rates. In other studies (Richard, 1997; VanderGheynst et al., 1997), peak CO2 emission rates were 30 to 70% lower than in our experiments, which may be explained by larger particle sizes. However, comparisons are difficult, because the obtained peak values depend on the length of the time intervals for determination of peak decomposition rates.

Decomposition of Carbon Constituents
Cellulose constituted the largest fraction of the waste/straw mixture, followed by hemicellulose (Fig. 5 , Run67a given as an example). The sugar fraction constituted only about 3% of volatile solids (VS) and was almost completely decomposed during the first few days, suggesting that this carbon source was of little importance for the period of highest activity after the temperature was allowed to increase.


Figure 5
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 5. Decomposition of different carbon constituents in Run67a. Time = 0 when the set point temperature was changed from 37°C to 40, 55, or 67°C, respectively. Bars indicate standard deviation of triplicate samples.

 
Starch decomposition also started rapidly, but 55 to 70% of this fraction still remained at the time of set point change. Thereafter, the decomposition of the starch fraction was fastest at 55°C and slowest at 40°C (Fig. 6 ).


Figure 6
View larger version (30K):
[in this window]
[in a new window]

  		Fig. 6. Decomposition of starch, crude fat, hemicellulose, and cellulose in Run40, Run55c, and Run67a-c. Values are relative, and each fraction set to 100% at time = 0; when the set point temperature was changed from 37°C to 40, 55, or 67°C, respectively. Bars show standard deviation of triplicate samples.

 
The decomposition of crude fat was variable during the initial 1.6 to 3.3 d, when all runs were replicates, with 65 to 90% of the fat still undecomposed immediately before the increase in temperature. Decomposition of fat was subsequently slower at 40°C compared to 55 and 67°C (Fig. 6). An increase in availability at temperatures exceeding the melting point of parts of the crude fat fraction can probably explain the faster decomposition at higher temperatures. Horwath and Elliott (1996) found that combined mesophilic and thermophilic conditions (25°/55°C) during incubation of ryegrass straw favored the breakdown of lipids and the decomposition of cellulose and lignin, as compared to only mesophilic conditions (25°C).

Hemicellulose and cellulose decomposition was slow during the first days of the experiments, amounting to 0 to 15% and 0 to 3% of the initial amount for hemicellulose and cellulose, respectively (Fig. 5). The results of the three runs at 67°C differed considerably, making comparisons with the other temperatures difficult. Still, hemicellulose degradation was faster at 55° than at 40°C, and cellulose degradation was faster at 55°C than at both 40 and 67°C (Fig. 6). Thus, 55°C seems to be the most favorable temperature for decomposition of these carbon fractions. Hemicellulose was degraded faster than cellulose at 67°C, but at the lower temperatures there was no difference.

Decomposition of the lignin fraction was slow, amounting to 0 to 20% of the initial amount during the first 15 d (e.g., Run67a in Fig. 5). Again, the results from the three replicate runs at 67°C differed and there were also large variations between replicate samples. Thus, no definite conclusion could be drawn on the impact of temperature on lignin degradation. In a study with radiolabeled synthetic lignin, Tuomela et al. (2001) found a relatively high lignin decomposition at 35 and 50°C, and very low decomposition at 58°C, indicating that lignin-degrading organisms in compost may be inactivated above 50°C.

With the extraction scheme used, about 25% of the organic matter remained unclassified, which is not really satisfactory. Proteins and pectins, which were not accounted for, probably constituted a substantial part of the unclassified organic matter. Moreover, the NDF/ADF method does not seem to be an ideal method to study decomposing materials, since it is not known how the humic substances formed interfere with the different fiber fractions. It is possible that some of the organic matter formed during composting is extracted together with the solubles and thus erroneously identified as hemicellulose, or oxidized by the permanganate and thus identified as lignin.

pH, Short-Chain Organic Acids, and Microbial Biomass
The biowaste/straw mixture had a pH of 5.5 (standard deviation [SD] = 0.06) at the start of the experiments, and the pH decreased to a minimum of 5.2 (SD = 0.33) after 1.9 d (SD = 0.8). Thereafter, the pH in the material increased rapidly to above 8 (Fig. 7 ). The pH in the condensate generally showed a similar trend, but was somewhat lower than in the corresponding compost material (data not shown) (Smårs et al., 2002).


Figure 7
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 7. Development of pH in compost material in Run40, Run55a, and Run67a. Time = 0 when the set point temperature was changed from 37°C to 40, 55, or 67°C, respectively. Bars indicate standard deviation of triplicate samples.

 
The concentration of short-chain organic acids (mainly lactic and acetic acid) in the compost increased from an initial value of about 50 mg g–1 to a maximum of about 80 mg g–1 when the pH was at minimum. The organic acid concentrations decreased to less than 10 mg g–1 when the pH increased above 8 and remained close to zero thereafter. In general, changes in the concentrations of short-chain organic acids closely followed the changes in pH, which confirms earlier observations that these acids control the pH in the early phase of composting (Reinhardt, 2002; Beck-Friis et al., 2003).

In later stages, the pH was between 8 and 9 in all composts. In Run40, the pH decreased to 8.2 after a maximum of 9.0 (Fig. 7), and in one of the composts at 55°C (Run55c), the pH decreased from 8.5 to 8.2 (data not shown). The decrease in pH in both Run40 and Run55c coincided with a decrease in NH4+–N but not with any increase in NO3––N (data not shown), indicating that it was caused by NH4+ immobilization into the microbial biomass, a process where protons are released (Beck-Friis et al., 2003). Ammonia emissions were low during these specific time periods, and could not explain the decrease in NH4+–N.

The total PLFA concentration, an indicator of microbial biomass, increased during the period of maximal CO2 emission at all three temperatures, but most dramatically when the process was run at 40°C (Fig. 8 ). The large increase between Day 7 and 17 in Run40 coincided with the decrease in the concentration of NH4+–N, further supporting our hypothesis that ammonia immobilization into microbial biomass caused the decrease in pH. Microorganisms generally show decreased biomass yield at higher temperatures, at least in the range above the temperature for maximal growth rate (Coultate and Sundaram, 1975; Pirt, 1975). Furthermore, negative relationships between temperature and microbial biomass yield in complex microbial associations have also been found in aerobic wastewater treatment systems (Bérubé and Hall, 2000; LaPara et al., 2000). The fact that the PLFA composition in our composting run at 40°C differed from those at the higher process temperatures (see below) implies that selection of different microbial communities may also have contributed to the differences in growth yields, and therefore to higher microbial biomass at 40°C.


Figure 8
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 8. The sum of phospholipid fatty acid (PLFA) concentrations, indicating microbial biomass, in Run40, Run55c, and Run67c. Arrows indicate the start of stepwise decrease in temperature set point in Run55c and 67c. Time = 0 when the set point temperature was changed from 37°C to 40, 55, or 67°C, respectively. Bars indicate standard deviation of triplicate samples.

 
Similar to the findings of a previous study (Steger et al., 2005), starch and fat were the main substrates during the period of rapid growth in microbial biomass in the thermophilic processes, whereas the decomposition of these two fractions was slower at 40°C (Fig. 6). The decomposition of mainly cellulose and hemicellulose led to further growth in microbial biomass in the 40°C run (Fig. 6 and 8), but the decomposition of these substrates apparently did not lead to comparable biomass increases in the thermophilic runs.

The fatty acid composition of the phospholipids differed substantially at 40°C compared to the 55 and 67°C processes. On Day 12–14, the total contribution of the sum of all monounsaturated fatty acids with 16 to 19 carbon (except 18:1w9), plus cy17:0 and cy19:0, was larger in Run40 than in the runs at higher temperatures (12% at 40°C, 5% at 55°C, and 2% at 67°C), whereas the opposite was found for the total contribution of all iso- and anteiso-branched fatty acids with 15 to 19 carbon plus 15:0 and 17:0 (25% at 40°C, 33% at 55°C, and 41% at 67°C). Monounsaturated fatty acids occur in high amounts in many gram-negative bacteria, whereas branched fatty acids are typical for gram-positive bacteria (Lechevalier and Lechevalier, 1988; Pinkart et al., 2002). Thus, this major difference in the PLFA composition shows that populations of gram-negative bacteria constituted a larger proportion of the microbial biomass in the mesophilic run, whereas gram-positive bacteria and/or thermophiles were more important in the thermophilic runs.

Ammonia Emissions and Carbon/Nitrogen Ratios
Ammonia emissions were considerably larger at 67°C than at the lower temperatures (Table 3, Fig. 9 –10; please observe the different time scale used in the table). After 11 d of composting, more than double the amount of ammonia had been released from composts at 67°C compared to 55°C. The difference in ammonia losses was also reflected in the nitrogen remaining in the compost, with smaller amounts found at 67°C. Consequently, the C/N ratio in the material was higher after composting at 67°C.


View this table:
[in this window]
[in a new window]

  		Table 3. Nitrogen content, ammonia emissions, carbon dioxide emissions, and C/N ratios during composting at different temperatures.

 

Figure 9
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 9. Ammonia emissions during composting in Run40, Run55a, and Run67a. Thick lines represent ammonia captured in condensate, thin lines also includes ammonia left in the compost gas after condensation.

 
The differences in nitrogen dynamics between 40 and 55°C were, however, not as large, and more difficult to quantify since the run at 40°C was not replicated and two of the runs at 55°C were very short. Ammonia emissions were smaller at 40°C, but this was partly balanced by the slower carbon dynamics. This is reflected in the similar C/N ratios in Run40 and Run55c on Day 7 (Table 3). However, when 59% of the initial carbon had been emitted as carbon dioxide (after 11 and 24 d in Run55c and Run40 respectively), the accumulated ammonia emissions at 55°C were more than twice as large as at 40°C. In our experiments, the start of ammonia emissions coincided with the time of pH change from acidic to alkaline (Fig. 7, 9, and 10 ). This is characteristic for composting (Elwell et al., 2002; Beck-Friis et al., 2003) and is due to the pH-dependent equilibrium between the ammonium ion and ammonia.

Formula


Figure 10
View larger version (21K):
[in this window]
[in a new window]

  		Fig. 10. Cumulative ammonia emissions in the composting runs. Arrows indicate the start of the stepwise decrease in temperature set point in Run55c and 67c. Time = 0 when the set point temperature was changed from 37°C to 40, 55, or 67°C, respectively.

 
To our knowledge, such large differences in ammonia emissions from thermophilic composts (more than double at 67°C compared to 55°C) have not been documented before. However, a positive relationship between temperature and ammonia emissions should be expected, since ammonia emissions are controlled by the equilibrium between dissolved and gaseous ammonia.

Formula

This equilibrium is highly temperature-dependent; the solubility of ammonia in water at 40°C is double that at 60°C (Aylward and Findlay, 1994). Frequent aeration of the compost will increase the loss of ammonia from water films on the compost particle surfaces, and increased aeration has been shown to increase ammonia emissions from composts (Elwell et al., 2002). In our reactor, air was circulated at a high rate to maintain uniform process conditions, and ammonia-containing condensate was removed when the compost gas was cooled. Condensate was recycled to the reactor in the same proportion as the proportion of cooled gas recycled to the reactor. On average 50% of the condensate was returned.

Temperature and aeration rate are thus two major factors affecting ammonia emissions from composts. These two factors are closely linked, since increased aeration is normally the only way to achieve a lower temperature in a composting process of a given size. Another way to achieve lower temperatures without increasing the aeration rate is to reduce the pile size, which increases surface cooling. This was shown to reduce ammonia emissions when composting swine manure (Fukumoto et al., 2003). When the temperature was reduced gradually from 67 to 40°C (Run67c), ammonia emissions decreased considerably compared to a continued treatment at 67°C (Run67a, Table 3). The temperature reduction in Run55c did not result in a corresponding decrease in ammonia emissions.

Ammonia emissions from the three replicate runs at 67°C were very similar (Fig. 10). This indicates that the results are highly reproducible. However, the variation in the nitrogen content between samples of the starting material was large (coefficient of variation 15.9%). This high variation leads to uncertainty in the mass balances, i.e., when the ammonia losses and the remaining nitrogen are described as a percentage of the initial nitrogen content (Table 3). Nevertheless, uncertainty regarding the initial nitrogen content cannot explain the trend that the unaccounted nitrogen losses seen in the mass balances increased with time (these can be calculated from data in Table 3). A possible explanation for these unaccounted losses of nitrogen is the formation and loss of N2 by denitrification, which has been shown to occur under a wide range of conditions during composting (Körner and Stegmann, 2004). Nitrous oxide emissions accounted for less than 1% of the nitrogen emissions (data not shown).

What is the Optimal Composting Temperature?
Our results may give some useful indications concerning a possible temperature optimum for biowaste decomposition. First, 67°C is above the optimum temperature, but it is not too high for efficient decomposition. Second, decomposition was much more efficient at 55°C than at 40°C. These results are in line with the common view, derived from studies with different kinds of substrates, that the mid to upper 50°C range is optimal for composting (Miller, 1993). It also agrees with the results of Nakasaki et al. (1987), using data from Bach et al. (1984), on composting dewatered sewage sludge that showed high decomposition rates at 56 and 62°C but significantly lower rates at 37, 43, and 70°C. Our results contrast with studies that found lower temperature optima when composting bark and sewage sludge (McKinley and Vestal, 1984; Campbell et al., 1990; Liang et al., 2003). This indicates a difference in optimum compost temperatures between food wastes that contain easily degradable compounds, and more resistant substrates. However, such a conclusion is not consistent with the results of Richard (1997) who found no difference in the optimum temperatures for decomposition of food waste and anaerobically digested sewage sludge. Besides the substrate composition, other factors such as oxygen concentration (Richard, 1997) and pH (Sundberg et al., 2004) may influence the optimal composting temperature in specific cases. Moreover, the different composting phases can be expected to have different temperature optima, since both the substrate composition and the microflora change during the process.

However, rapid decomposition is not the only criteria in the selection of an optimal composting temperature. Our data suggest that the biowaste composting process can be optimized to obtain both a high decomposition rate and low ammonia emissions by controlling the process at about 55°C.

Another important goal is to minimize emissions of other gases with harmful effects on the environment, such as N2O and CH4, which are also affected by the compost temperature (Hellman et al., 1997). In addition, the waste material must be exposed to a sufficiently high temperature for a sufficiently long period of time to ensure an effective inactivation of pathogens (Haug, 1993). On the other hand, the promotion of beneficial microorganisms that are antagonistic to plant pathogens is favored by a low temperature phase for colonization (Hoitink and Fahy, 1986). Finally, it can be economically advantageous to run a process at a temperature above the biological optimum due to the costs of aeration for cooling (Keener et al., 2002).

Applicability of Results
The experimental reactor was constructed to enable fundamental studies of the influence of different factors on the composting process. To this end, temperature and oxygen, which are interdependent when both are controlled by aeration, were decoupled by recycling cooled process gas, as described previously. The reactor used in our study also fulfilled another important prerequisite, namely homogeneous process conditions. Our results may be used to understand and describe what happens in various parts of an entire large- or small-scale compost.

In most composting systems used in practice, temperature and oxygen concentration are not controlled independently, since they both depend on aeration and decomposition rates. This implies that in a given composting system, a certain temperature will be linked to a specific oxygen concentration in the process gas. The relationship between temperature and oxygen concentration depends on the conductive and radiative heat losses from the compost and will therefore differ between composts of different sizes. Nevertheless, results from our experimental reactor may also be applied more directly to certain kinds of composting systems used in practice. These are processes at the same temperature and oxygen concentration and where gas losses, including vapor, are only due to the air exchange needed to maintain the oxygen concentration. For the oxygen concentration used in our experiments (16% O2) this would be the case when the conductive-radiative cooling is large (e.g., in small composts) or when cooled air and condensate is recycled in composts of any size.


   Conclusions
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Conclusions
REFERENCES
 
Composting at 55°C gave the highest decomposition rate, measured as CO2 emission, compared to 40 and 67°C. Ammonia emissions at 67°C were more than double those at lower temperatures, and they were higher at 55°C than at 40°. Thus, to optimize biowaste composting to obtain both a high decomposition rate and low ammonia emissions, it is recommended to steer the process to about 55°C. To reduce ammonia emissions, it seems worthwhile to reduce the temperature after an initial high-temperature stage.


   ACKNOWLEDGMENTS
 
Financial support for this investigation was provided by The Swedish Council for Forestry and Agricultural Research and the Swedish University of Agricultural Sciences. We would like to thank Dr. Åsa Jarvis for close cooperation in planning and evaluation of the composting experiments.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Conclusions
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Conclusions
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Eklind, Y.
Right arrow Articles by Jönsson, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Eklind, Y.
Right arrow Articles by Jönsson, H.
Agricola
Right arrow Articles by Eklind, Y.
Right arrow Articles by Jönsson, H.
Related Collections
Right arrow Nitrogen
Right arrow Microbial Processes
Right arrow Municipal Waste

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome