Sei sulla pagina 1di 9


discussions, stats, and author profiles for this publication at:

The evaluation of stability during the

composting of different starting materials:
Comparison of chemical and biological

Article in Chemosphere March 2011

DOI: 10.1016/j.chemosphere.2011.01.010 Source: PubMed


31 72

3 authors:

Marco Grigatti Luciano Cavani

University of Bologna University of Bologna


Claudio Ciavatta
University of Bologna


All content following this page was uploaded by Marco Grigatti on 10 July 2017.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
Chemosphere 83 (2011) 4148

Contents lists available at ScienceDirect

journal homepage:

The evaluation of stability during the composting of different starting materials:

Comparison of chemical and biological parameters
Marco Grigatti , Luciano Cavani, Claudio Ciavatta
Department of Agroenvironmental Sciences and Technologies, Alma Mater Studiorum University of Bologna, Viale G. Fanin, 40 40127 Bologna, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Three blends formed by: (i) food processing waste (CPFP), (ii) waste water sewage sludge (CPWW), and
Received 8 July 2010 (iii) their mixture (CPFP+WW), blended with tree pruning as bulking agent, were composted over 3 months.
Received in revised form 3 January 2011 During composting the blends were monitored for the main physicalchemical characteristics: temper-
Accepted 4 January 2011
ature, oxygen saturation level (O2%), pH, total and volatile solids, total organic carbon, and organic nitro-
Available online 28 January 2011
gen (Norg). In addition to the main parameters, the dissolved organic carbon (DOC), the inorganic nitrogen
and the Oxygen Uptake Rate (OUR) were monitored. All the mixtures easily reached a peak temperature
around 70 C, related to the lowest O2%. After 90 d, CPFP, CPFP+WW, and CPWW showed an organic matter
Food processing waste
Sewage sludge
mineralization of 43%, 35% and 33%, respectively; CPFP tted an exponential model while both CPFP+WW,
Composting and CPWW tted a logistic model. During composting an OUR reduction of 79%, 78% and 73% was regis-
Organic matter and nitrogen mineralization tered in CPFP, CPFP+WW, and CPWW, respectively; the OUR successfully tted the adopted exponential
Oxygen Uptake Rate model and well reected the stabilization process in time. The Norg recovery at the end of the process
was positive only in CPWW (11.6%). The DOC signicantly decreased during the composting process but
did not successfully t any model. The mineral nitrogen did not follow the typical pattern with NH 4 dis-
appearance and NO 
3 accumulation. Strong NO3 losses were evident in all blends, while NH4 accumula-

tions were detectable only in CPFP, and CPFP+WW. The NH 4 /NO3 ratio did not satisfactorily reect the
composting process over time. The comparison of the rst order (exponential) and logistic (sigmoidal)
models applied to the OUR and OM course highlights the role of mineral nitrogen as limiting factor during
composting of the more stabilized sludge.
2011 Elsevier Ltd. All rights reserved.

1. Introduction tion of unstable compost in soil are responsible for compost phyto-
toxicity (Iannotti et al., 1993).
Compost is a valuable resource as a soil fertilizer, providing a The application of chemical parameters to asses the stability
high content of macro- and micronutrients for crop growth, and of compost is a common practice in research on composting
represents a low-cost alternative to mineral fertilization (Eghball (Ciavatta et al., 1993; Govi et al., 1993; Chefetz et al., 1996;
and Power, 1999). A practical use of composted materials in Grigatti et al., 2007a). Some of these parameters are not consid-
agriculture requires good knowledge of its maturity and the level ered good indicators of compost stability because they poorly
of organic matter (OM) stabilization reached at the end of the com- describe the evolution of OM quality during composting; this is
posting process. The application of unstable OM on soil may pro- mainly due to the utilization of strong oxidants or strong bases
duce competition for oxygen between microbial biomass and which break down carbon not readily available for microorgan-
plant roots. This fact can deprive plant roots of oxygen and lead isms (Zmora-Nahum et al., 2005). The use of dissolved organic
to the production of H2S and NO 2 . Another problem is the possible carbon (DOC) as a stability parameter proved to be closer to
nitrogen starvation of plants as microorganisms scavenge soil N to the real request of energy by the microbial community during
make up for the decit resulting from the application of unstable composting (Zmora-Nahum et al., 2005). DOC generally contains
OM with high C/N ratio (Bernal et al., 1998). Moreover, the emis- organic compounds with different susceptibilities to microbial
sion of ammonia and the presence of other phytotoxic substances degradation and different phytotoxic properties. For this reason,
like phenolic compounds, ethylene oxide and low-molecular DOC composition may have an important role in determining

weight organic acids that are synthesized during the decomposi- the stabilization process. The NH 4 /NO3 ratio has also been pro-
posed to estimate the compost stability. At the end of the com-
posting process, the content of NO 3 should be higher than that
Corresponding author. Tel.: +39 (0)51 209 62 17; fax: +39 (0)51 209 62 03. of NH 4 , indicating that the process has been performed under
E-mail address: (M. Grigatti). adequate aeration conditions. Nitriers are a low competitive

0045-6535/$ - see front matter 2011 Elsevier Ltd. All rights reserved.
42 M. Grigatti et al. / Chemosphere 83 (2011) 4148

community becoming effective when the temperature of the Table 1

mass under composting decreases and the readily degradable Main physicalchemical characteristics of raw materials.

substances (DOC and soluble N) are low (Veijalainen, 2007; Parameters Bulking Agro-industrial Waste water
Tontti et al., 2010). More recently, besides the CO2 evolution, agent (BA) waste mix (FP) sewage
many authors have studied the relationship between microbial sludge (WW)
activity and composting time by means of the OUR during OM TS (%) 55 30 51 24
degradation (Iannotti et al., 1994; Lasaridi and Stentiford, 1998; pH 6.84 7.28 10.22 7.94
Barrena et al., 2005). Tremier et al. (2005) used the respirometric EC (mS cm1) 0.93 1.10 1.90 0.90
technique to highlight the important role played by substrate VS (%) 41 36 78 38
TOC (mg g1) 252 226 681 277
biodegradability in compost evolution over time. In a previous
TKN (mg g1) 11.2 33.3 21.4 39.7
study, Grigatti et al. (2007b) proposed a novel test to measure C/N 23 7 33 7
stability by means of incubation in a liquid environment. This
Total heavy metals (mg kg1)
method assesses the stability level of the OM by means of indi- Cd <0.1 1.7 2.0 1.0
rect measuring of microbial respiration during incubation in a Cr 21 27 28 309
closed vessel in standardized and non-limiting conditions. The Cu 38 45 58 424
Ni 63 48 56 189
pressure course in time is converted into the cumulative oxygen
Pb 54 23 25 182
uptake in the linear part of the curve by means of the gas law Zn 87 123 200 1190
(PV = nRT) and further transformed into the OUR and expressed
on volatile solids (VS) as mmol O2 kg1 VS h1. The biological FP1: Starch processing waste; FP2: Poultry meat processing waste; TOC, TKN and
total heavy metals are expressed on TS basis.
parameters such as respirometric tests are expected to be the
most suitable measurements to evaluate the stability of the com-
posting materials because they are able to measure the extent at the waste water sewage sludge which was in any case within
which readily biodegradable OM has decomposed during the the normal range for this type of product.
composting process. The objectives of this work were, therefore,
the description of the chemical and biological changes during 2.2. Sampling and preparation of samples
the composting of different starting materials and the compari-
son of different parameters used with respect to their ability to Compost samples at different stages of stabilization (0, 15, 30,
evaluate the compost stability. 60, 90 d) were collected over a 3-month period. Two kg of mixed
samples were formed by mixing ve sub-samples taken from each
pile (IPLA, 1998). These were divided into fresh and dry samples,
2. Materials and methods the latter were dried in an air-forced oven at 40 C until a constant
weight was obtained and then sieved at 8 mm. TS were determined
2.1. Composting materials at 105 2 C until constant weight, the VS were determined on TS
at 550 C for 4 h. The dry sample was subsequently crushed using a
Starting materials were selected as representative for plant Tecator Cyclotec, 1093 (PBI, Sweden), until all the material passed
feeding. To this purpose, a food processing waste mix (FP) was through a 0.25 mm sieve to produce a wholemeal sample (IPLA,
prepared by mixing two lter-pressed sludges at 50% (v/v). 1998).
These came from food processing: starch production (FP1) and
poultry meat processing (FP2). In addition, a mixed waste water 2.3. Determination of temperature and oxygen saturation level
sewage sludge (WW) was formed by a combination of two sim-
ilar waste water sewage sludges at 50% (v/v), in order to make Temperature and oxygen saturation level (O2%) were deter-
up a more representative product. Tree pruning (branches and mined at a 30 cm depth (three replicates each pile) at the time of
leaves) chopped at < 5 cm was used as bulking agent (BA). sampling, by coupled platinum and polarographic sensors probe
Compost piles were prepared by means of a specialised compost (Bio.Ge.Co. model D09709, Italy).
mixing machine (DOPPSTAD TDU 320, Germany), mixing differ-
ent quantities of starting materials in order to obtain contrasting 2.4. Determination of pH and electrical conductivity
characteristics. The initial composting blends (v/v) were: CPFP
(BA 80% + FP 20%); CPWW (BA 80% + WW 20%); CPFP+WW (BA A suspension of 5 g of fresh sample and 50 mL of deionised
80% + FP 10% + WW 10%). In an external yard, three large piles water (d-H2O) was stirred for 30 min at 25 C. After ltration, the
(about 100 m3) were formed, one for each blend. The piles were pH was measured using a Micro TT 2022 (Crison, Spain) pH-meter
then divided into two smaller piles (about 30 m3 each) in order and the electrical conductivity (EC) using a SAT type (Halosis, Italy)
to have two replicates and arranged in the yard in a randomized conducimeter (IPLA, 1998).
block design. The piles were mixed up weekly for the rst
month to ensure good aeration conditions, and then monthly. 2.5. Determination of total nitrogen, mineral nitrogen and heavy
The main physicalchemical characteristics of the raw materials metals
(BA, FP1; FP2 and WW) are reported in Table 1. The BA, formed
by chopped tree branches and leaves, showed the typical charac- TKN was determined by the Kjeldahl method (Bremner, 1996)
teristics for this kind of product: TS: 55%, pH: 6.84, VS: 41%, on dried samples. The mineral nitrogen (NH 4 , NO2 and NO3 ) con-
TOC: 252 mg g1 and TKN: 11.2 mg g1 with a C/N ratio of 23. tent was determined in the liquid phase by means of an Autoana-
Compared to the BA, sludges generally showed lower TS; only lyzer III (Technicon) after an extraction in 2 M KCl (2 h; 120 rpm)
FP2 showed similar values (51%). At the same time, sludges and ltration on Whatman #42. The organic nitrogen (Norg) was
showed 2 to 3-fold TKN content (21.439.7 mg g1). All starting determined by subtracting the NH 4 N from the TKN. The total hea-
materials showed similar VS (40%) and TOC content, about vy metal content was determined by atomic absorption spectro-
250 mg g1; only FP2 had a 2-fold VS (78%) and 3-fold TOC photometry (GBC model 903, Australia) after digestion of 1 g of
(681 mg g1). This reected on C/N, which rose from about 7 dried sample with 20 mL of 70% HNO3 plus 5 mL of 37% HClO4
to 32. The heavy metals content was generally low, apart from (MiPAF, 2001).
M. Grigatti et al. / Chemosphere 83 (2011) 4148 43

2.6. Total, dissolved organic carbon and total phenols means of a logistic (sigmoidal) model of the type: y = a/[1 + ex-
p((t  t0)/b)] in which y represents the potential OM mineraliza-
Total (TOC) and dissolved organic carbon (DOC) were measured tion at time t (%), a represent the maximum mineralization (%), b is
using dichromate oxidation methods (MiPAF, 2001). DOC was the mineralization rate (d), t is the composting time (d), and t0 is
isolated as follows: 20 g of sample was extracted with 100 mL of the point of inection (d). The OUR data from each single sampling
d-H2O shaken at 110 rpm for 2 h (Grigatti et al., 2004). The suspen- were analyzed by means of an exponential model of the type:
sion was centrifuged at 5000 rpm for 30 min, ltered through a ym y0 aekt , in which: ym represents the OUR over time (mmol
0.8 lm lter (mixed cellulose esters, Millipore, USA). The total O2 kg1 VS h1); y0 is the potential OUR (mmol O2 kg1 VS h1); a is
water soluble phenols were determined on the same solution. Phe- the intercept; k is the respiration rate constant (d1); t is the com-
nolic compound concentration was determined by the FolinCio- posting time (d). Nonlinear regressions were tted using the Sigma
calteau reagent and gallic acid as standard (Slinkard and Plot 8.0 (SPSS Inc.) software.
Singleton, 1977).

3. Results
2.7. Determination of the OUR

3.1. Initial physicalchemical characteristics of mixtures

The OUR was determined as reported in Grigatti et al. (2007b).
Briey, an amount of sample corresponding to 2 g of VS was placed
The main characteristics of the composting mixtures are re-
in a 1 L Schott ask, adding 180 mL of d-H2O water and 10 mL of a
ported in Table 2 and Figs. 13. At the mixing time (0 d), CPFP
solution rich in nutrients to the bottle, together with 0.2 mL of a
showed the highest TS (46.4%), TOC (325 mg g1), and OUR
micronutrient solution. Ten mL of phosphate buffer (pH 7.00)
(79 mmol O2 kg1 VS h1), whereas it showed a background level
was also added. The incubation was run in the dark for 168 h at
for NH4 . On the contrary, CPWW showed the lowest TS (37.5%),
25 C, shaken at 120 rpm in the presence of N-allylthiourea
TOC (282 mg kg1) and OUR (41 mmol O2 kg1 VS h1) and the
(50 mg L1) as nitrication inhibitor; the pressure course in the
highest NH 4 N (196 mg kg
) and EC (1.6 mS cm1). CPFP+WW gen-
bottle headspace was recorded continuously by means of the Oxi-
erally showed intermediate levels for all parameters apart from
Top device (WTW GmbH, Germany). The CO2 was trapped as
Norg (15.9 mg g1), which was slightly higher than other blends
formed by Sodalime (Merk); the pressure drop in this condition
(14.5 mg g1 on average). NO 3 N (54 mg kg
), DOC (11.4 mg g1)
is only due to the oxygen consumption because of the microbial 1
and total phenols content (415 mg GA eq kg ) were similar in all
2.8. Data analysis
3.2. Evolution of the composting blends
The physicalchemical data (three replicates), were analyzed by
means of Two Way ANOVA, according to compost type, compost- The data showed that the temperature and the O2 saturation le-
ing time and their interaction. Mean separation was done using vel were inuenced only by the composting time; the three blends
the Fisher least signicant difference (LSD) test (a = 0.05), using showed a peak value of 66 C after 2 weeks, then the temperature
the Sigma Stat 2.03 (SPSS Inc.) software. The pH, TS and VS data decreased regularly down to 35 C at the end of the process; a few
were ln-transformed, before analysis. The OM mineralization was degrees higher than the atmospheric temperature in July (Fig. 1a).
calculated as: OMmin = [(VSt0  VSt1)/VSt0] 100, and analyzed over The O2 saturation level followed a similar pattern, the lower value
time by means of an exponential model of the type: ym (81% of the atmospheric level) corresponding to the temperature
y0 1  ekt , in which ym represents the OM mineralization over peak (Fig. 1b). The O2 saturation then increased regularly up to
time (%), y0 the potential OM mineralization (%), k the mineraliza- the same values as the composting start (90%). The TS content
tion rate constant (d1) and t is the composting time (d); and by varied according to the mixtures and time; TS followed the

Table 2
Changes in the main physicalchemical characteristics during composting of different mixtures.

Compost Time (days) TS (%) pH (log[H+]) EC (mS cm1) TOC (mg g1) VS (%) Norg (mg g1) C/N DOC (mg g1) TPh (mg GA eq kg1)
CPFP 0 46.4 7.4 1.12 325 56.7 14.5 22.4 12.4 401.2
15 55.7 7.2 1.10 262 45.1 12.8 20.5 8.3 329.0
30 55.8 7.6 1.03 255 43.4 11.0 23.2 7.1 293.2
60 62.3 7.6 1.02 253 35.0 14.3 17.7 4.7 215.0
90 63.3 7.5 1.02 187 32.4 13.8 13.6 5.2 250.2
CPFP+WW 0 43.2 8.0 1.25 302 55.7 15.9 19.0 11.3 407.8
15 50.4 7.6 0.76 290 50.8 12.5 23.2 7.1 285.5
30 51.9 7.3 0.90 263 49.1 14.5 18.1 7.4 308.2
60 60.4 7.3 0.93 258 37.4 13.9 18.6 4.2 206.3
90 60.9 7.3 0.93 213 36.2 15.0 14.2 5.5 274.3
CPWW 0 37.5 7.6 1.60 282 57.3 14.5 19.4 10.6 438.8
15 49.7 7.3 0.52 272 50.0 14.9 18.3 10.1 456.7
30 51.9 7.2 0.69 266 51.3 15.3 17.4 5.8 280.3
60 58.0 7.1 0.70 263 42.6 14.5 18.1 5.8 246.2
90 58.5 7.1 0.70 226 38.3 16.1 14.0 5.8 282.3
** *** *
Compost LSD 1.95 n.s. n.s. n.s. LSD 2.7 LSD 0.68 n.s. n.s.
** *** *** * *** ***
Time LSD 2.52 n.s. LSD 10.18 LSD 3.5 LSD 0.85 LSD 4.5 LSD 62.9
*** **
Compost  time n.s. n.s. n.s. LSD 17.62 n.s. LSD 1.47 n.s. n.s.

signicant at P 6 0.10.
Signicant at P 6 0.05.
Signicant at P 6 0.01.
Signicant at P 6 0.001.
44 M. Grigatti et al. / Chemosphere 83 (2011) 4148

80 (a) 50 CPFP (a)


OM mineralization (%)
Temperature C

External temperature



20 0

LSD0.05 (b) 75

OUR (mmol O2 kg-1 VS h-1)

O2 (% of atmospheric)



80 15

0 15 30 45 60 75 90
0 15 30 45 60 75 90
Days of composting Days of composting
Fig. 2. Organic matter mineralization (a) and OUR (b), during composting of
Fig. 1. Temperature (a) and oxygen saturation level (b) during composting of
different blends. The parameters of the tted function of OUR are shown in Table 3.
different blends. The external temperature is reported as the daily average between
Tmin and Tmax. The Fisher LSD at a = 0.05 is shown in plot for the mean separation.

33% in CPFP, CPFP+WW, and CPWW, respectively. CPFP tted the expo-
CPFP > CPFP+WW > CPWW pattern throughout the process; at the end nential (rst order) model which showed a k value of 0.03 d1.
of composting TS was 63%, 61% and 59% in the above-mentioned Conversely both CPFP+WW and CPWW tted the logistic (sigmoidal)
respective mixtures (Table 3). No signicant variation of TS was model in which the doubling time (b) was 12.4 and 20.4 d
detectable in the mixtures in the 1530 d and 6090 d intervals; (Fig. 2a). The maximum mineralization indicated by the models
the main variation occurred in the 015 and 3060 d intervals. was: 45% in CPFP; 36% in both CPFP+WW and CPWW. t0 corresponded
The pH was unaffected by the mixtures or time, while the EC to 36 and 43 d in CPFP+WW, and CPWW, respectively (Fig. 2a).
was inuenced only by composting time with a common decrease Having established the signicant differences in the initial val-
signicant at 10% (P = 0.080). ues of the OUR of different mixtures (79, 57 and 41 mmol O2
kg1 VS h1 respectively in CPFP, CPFP+WW, and CPWW), the samples
showed a similar pattern in the decrease of the respiration rate
3.3. OM course during composting during the composting process (Fig. 2b). CPFP showed a steep
reduction in the rst 15 d (52%), followed by the other mixtures
Following the process of mineralization, the TOC progressively which showed a reduction of around 3035%. After 60 d of com-
decreased during composting depending on time and on the inter- posting, all mixtures reached similar values around 12 mmol
action mixture  time (Table 2). CPFP was the only mixture which O2 kg1 VS h1. At the end of the process, all samples showed a to-
showed a signicant mineralization in the rst 2 weeks (19%; tal OUR reduction in the order of 80% of the initial value. A deeper
P 6 0.001). The central phase of composting (1560 d) was charac- insight into the decay of OUR by means of the exponential model
terized by a modest decrease of TOC; only CPFP+WW showed a sig- revealed signicant differences among the three mixtures (Table
nicant decrease in this phase (9.3%; P 6 0.05). Later, in the last 3): CPFP showed a constant k rate of 0.06 (d1), compared to
month of composting, a further signicant decrease was recorded 0.03 (d1) of CPFP+WW and to the intermediate 0.05 (d1) of CPWW.
for all mixtures; on the whole, during the entire process, CPFP, These values corresponded to a half-life of the OM of 12, 22 and
CPFP+WW and CPWW showed a respective mineralization of 42%, 14 d in CPFP, CPFP+WW, and CPWW, respectively.
29% and 20%. The VS content was dependent on mixture and time; The Norg content was affected by mixture, time and by their
the latter showed no signicant variation between 15 and 30 d and interaction (Table 2). At the beginning of the process the Norg con-
in the last month. The OM mineralization reached: 43%, 35% and centration was lowest in CPFP, and CPWW (14.5 mg g1) while
M. Grigatti et al. / Chemosphere 83 (2011) 4148 45

Table 3
CPFP (a) Equation parameters, regression coefcient and signicance of the OUR measured

CPFP+WW * during the composting process in the three different mixtures.

Compost Equation parameters
y (mmol O2 y0 (mmol O2 a k (d1) R2
kg1 VS h1) kg1 VS h1)
CPFP 17 16* 62** 0.0601* 0.984**
N (mg kg-1)

CPFP+WW 13 8 50 ns 0.0315 ns 0.985**

CPWW 11 9* 32** 0.0503* 0.991***
According to the exponential model: ym y0 aekt , in which: ym represent the
* * OUR over time (mmol O2 kg1 VS h1); y0 is the potentially OUR (mmol O2 kg1
VS h1); a is the intercept; k is the respiration rate constant (d1); t is the com-
200 * posting time (d).
Signicant at P < 0.05.
Signicant at P < 0.01.
Signicant at P < 0.001.

and the only signicant variation was detectable after the rst
LSD0.05 * (b) month. The subsequent samplings showed lower values than the
initial ones but there was no difference between them. The total
phenols showed a similar pattern, being inuenced only by time;
they showed a rst reduction step in the 1530 and in the 30
60 d intervals.
N (mg kg-1)

* 3.4. Mineral nitrogen

* The mineral nitrogen content varied in dependence of mixture,

100 time and their interaction; Fig. 2 shows that at the beginning of the
process the NH 4 N content was the lowest in CPFP and CPFP+WW
(28 mg kg1 on average), while CPWW showed the highest value
50 (196 mg kg1). All mixtures showed a rapid increase of the ammo-
nium content at day 15: up to 200 mg kg1 in CPFP and to 600 on
average for both CPFP+WW and CPWW. After this stage, all blends
showed a rapid decrease, down to 11, 81 and 133 mg kg1 in CPFP,
CPFP+WW, and CPWW, respectively (Fig. 2). A new increase was sub-
(c) sequently detectable only in CPFP and CPFP+WW (700 and
180 mg kg1), while CPWW shifted toward a background level.
The NO 3 N content was similar in the different composts from
the start until day 15 (46 mg kg1 on average). NO 3 N then
showed an accumulation up to 130 mg kg1 in both CPFP+WW and
CPWW at day 30, and up to 250 mg kg1 in CPFP at day 60. All mix-

tures showed a subsequent nitrate reduction to values similar to

the initial ones. NO 2 was not detectable in any of the analyzed

samples. The NH 4 /NO3 ratio which was very low at the beginning
10 of composting (1.6 on average), after 15 d showed a steep increase
up to 6, 25 and 12 in CPFP, CPFP+WW, and CPWW, respectively (Fig. 2).
After that, it decreased down to 0 in CPWW, 3 in CPFP+WW while it
rose up to 9 in CPFP.

4. Discussion
0 15 30 45 60 75 90
Days of composting 4.1. Main physicalchemical characteristics

Fig. 3. Mineral nitrogen course and NH 4 /NO3 ratio during composting of different It was shown that during composting the temperature was af-
blends; NH
4 N (a), NO3 N (b), NH4 /NO3 ratio (c). Data are expressed on TS. The
Fisher LSD at a = 0.05 is shown in plot for the mean separation.
fected only by time following the typical pattern of windrow com-
posting (Rashad et al., 2010); all mixtures showed a rapid
temperature increase and the same peak at the rst sampling date;
CPFP+WW showed the highest concentration (15.9 mg g1). During the temperature attained and its duration were adequate to the
composting, CPWW showed a regular increase tendency, while CPFP mass sanitization (Mupondi et al., 2010). After 1 month, the tem-
and CPFP+WW showed a decrease followed by an increase. This on peratures were still higher than 50 C, and after 2 months these
the whole resulted in a moderate increase with respect to the ini- were 20 C higher than the external temperature, indicating the
tial values only in CPWW (11.6%; P 6 0.01), while CPFP and CPFP+WW presence of a high quantity of degradable OM. Temperature de-
showed no difference compared to the initial values. As shown in creases were less steep than reported in literature (Rashad et al.,
Table 2, the DOC content was dependent only on time (P 6 0.05); 2010), and probably due to the less frequent pile turning (monthly
no appreciable difference was therefore found between mixtures, instead of weekly) and the related oxygenation level, anyway, at
46 M. Grigatti et al. / Chemosphere 83 (2011) 4148

the end of composting piles temperatures were only a few degrees 4.3. Mineral nitrogen course
above external temperature. The O2 saturation levels, similar in the
different composting piles, showed the opposite pattern to temper- Based on the literature, it is generally recognized that the com-
ature; they increased after the temperature reached the peak at posting process is characterized by the disappearance of readily
15 d, indicating a general reduction of the microbial activity re- degradable N (normally NH 4 ) present in the rst process stage,
lated to the disappearance of degradable OM, as commonly re- in favour of the accumulation of NO 3 in the latter stages (Kalamd-
ported during composting (Kalamdhad and Kazmi, 2009). had and Kazmi, 2009; Gao et al., 2010). Many studies (Veijalainen,
2007; Tontti et al., 2010) showed that nitriers are a low compet-
4.2. OM evolution during composting itive community which generally becomes active when the tem-
perature of the composting mass decreases and the readily

The utilization of different types of sludge strongly affected the degradable substances are low. The NH 4 /NO3 ratio has thus been
OM content and mineralization over time; the latter was more in- proposed as a stability parameter (Meunchang et al., 2005). In
tense in CPFP especially in the rst composting stage. This was in the current experiment we did not nd a direct relationship be-
agreement with the utilization of an undigested sludge obtained tween NH
4 /NO3 ratio and composting time; the NH4 /NO3 ratio
by simple mechanical separation which had a lower stability in showed a saw-tooth pattern and in CPFP it was higher at the end
comparison with digested sludge, as proved by the OUR; the pres- than at the beginning. As previously discussed, we found that all
ence of a more easily biodegradable OM was described as the main mixtures exhibited an increase of the NH 4 content at the rst sam-
cause of carbon losses during composting (Barrington et al., 2002), pling at 15 d. This was more pronounced in the mixtures with sew-
by measuring the BOD. From the TOC mineralization, it appears age sludge. There was a considerable reduction afterwards, which
that CPFP and CPFP+WW had 53% and 32% more mineralizable carbon was very probably related to the full utilization of the NH 4 by
compared to CPWW. This is consistent with the OUR test: in fact at means of microorganisms, as reported by Rashad et al. (2010). In
the beginning of composting CPFP and CPFP+WW showed 97% and this experiment we found a further step of NH 4 accumulation
40% more oxygen demand compared to CPWW. Despite different which occurred only in the food waste processing based blends
TOC initial contents, the DOC were similar in mixtures, making a (CPFP and CPFP+WW). This kind of uncommon NH 4 accumulation
real separation of starting material based on this factor unfeasible. was described for denitrication beds which utilized composted
The DOC course was in agreement with what found by Zmora- materials (Mahimairaja et al., 1995; Gilbert et al., 2008; Cameron
Nahum et al. (2005) in a compost formed with similar raw and Schipper, 2010). Besides, a very uncommon nitrication pat-
materials; at the same time, only CPFP satisfactorily tted the decay tern was detected during composting: mixtures containing acti-
exponential model in our experiment, which allowed a useful vated sludges showed an earlier nitrication compared to CPFP,
evaluation of stabilization during composting (data not shown). probably due to the microbial community of the sludge itself. At
The OM mineralization showed a constant trend to the plateau the same time, these mixtures showed an earlier denitrication
in CPFP. Conversely both CPFP+WW and CPWW showed a doubling of and a lower NH 4 accumulation compared to CPFP. It has now been
the OM mineralization until the point of inection (weeks 5 and 6); recognized that NO 3 diffuses to zones which are made anaerobic
afterwards, OM mineralization in these two mixtures rapidly de- by high biological demand (BOD), resulting in denitrication (Zhou
creased. In other words the WW based blends showed a steeper et al., 2002). The differences in the rate of denitrication between
decrease of the OM compared to CPFP until the beginning of the samples have been related to the amount of soluble C and the pop-
mesophilic phase. This high carbon availability strongly stimulated ulation size of denitriers. In the current experiment we did not
the microbial growth, in turn promoting signicant Norg losses, nd any direct relationship between the OUR and the NO 3 courses,
most likely due to the respiration of proteins (Rashad et al., 2010). or any evidence of a relationship with the DOC. On the contrary,
It is widely recognized that carbon and nitrogen mineralization the highest NO 3 losses were detectable in the last month of com-
courses are strictly related; in this way different mixtures had also posting when temperature, DOC and OUR were low and the O2 sat-
a strong effect on the Norg course over time. Food waste based uration level had returned to the initial values. This behavior led us
blends showed an initial Norg reduction followed by a new in- to think that the denitrication process occurred thanks to the fun-
crease. This reected on the C/N values, which showed an uncom- gal activity as reported by Kobayashi et al. (1996), who found den-
mon saw-tooth pattern. Conversely, the compost made from itrifying systems in the mitochondria of different fungi. This fungi
sewage sludge showed a Norg increase tendency during compost- utilized the nitrite and nitrate reduction for ATP synthesis.
ing. This was related to the different mineralization rate of the
OM and reected a more regular C/N decrease over time, similar 4.4. OUR and OM mineralization relationship
to what generally reported in literature (Rashad et al., 2010; Mu-
pondi et al., 2010). It is recognized that carbon availability plays In the literature there are few examples in which the compost-
a major role in N immobilization (Barrington et al., 2002): while ing process has been followed step by step by means of the OUR in
degrading organic compounds, microbes waste 6070% of the C a nutrient non-limiting environment for microorganisms, and
as CO2 while immobilizing the rest as cellular components these mainly refer to the work of Lasaridi and Stentiford (1998).
(Kalamdhad and Kazmi, 2009). Since the yield factor for N by Kalamdhad and Kazmi (2009), in the evaluation of the stability of
microbial community is 0.6 as function of available C, it results that composting mixtures by the measurements of the OUR, found sim-
the remaining 0.4 can be lost through volatilization (Barrington ilar values at the beginning of the composting process; they also
et al., 2002). Although we did not measure the NH3 emissions, found the same trend in the OUR which decreases over time. Anal-
we found an intense NH 4 N accumulation up to 650 mg kg
in ogous was the OUR trend found by Lasaridi and Stentiford (1998).
correspondence with the temperature peak. In this phase we mea- Having established the signicant differences between the mix-
sured only the O2 saturation level in the interspaces. At the lowest tures at the beginning of the process the OUR was the only param-
level of the composting process, it was about 80% of the atmo- eter which successfully tted a mathematical exponential model as
spheric value. In addition, the compost aeration was limited by a function of time for all mixtures. The adopted model is a rst or-
the absence of active ventilation and the infrequent turning. It is der model in which the kinetics is inuenced only by the substrate
thus possible that the increase of the CO2 concentration preserved concentration in time (Mason, 2006). Since during the respiration
the pH increase which was around 7.5, and the NH 4 stripping, as test the only energy source is the carbon from compost, its refrac-
reported by Liang et al. (2004). tivity to the microbial degradation is recognized to be the only rate
M. Grigatti et al. / Chemosphere 83 (2011) 4148 47

limiting step. Sobratee et al. (2009), adopted a similar approach in variation during the composting process. The adoption of this test
the study of the microbial growth in the study of composting sub- in a non-limiting nutrient environment that stimulates microbial
strate extracts; the authors found that the microbial community growth, gives the real extent of potential mineralizable OM follow-
tted a logistic model in which the maximum growth decreased ing the addition to a nutrient rich environment as soils can be.
as function of composting time. The study of the I order equation
parameters from the respiration test showed that CPFP+WW had References
the lowest rate constant k. This was in agreement with the OM
mineralization course which was slower at the beginning of the Barrena, G., Vzquez Lima, F., Gordillo Bolasell, M.A., Gea, T., Snchez Ferrer, A.,
process, probably due to the adaptation of the microbial commu- 2005. Respirometric assay at xed and process temperatures to monitor
composting process. Bioresource Technol. 96, 11531159.
nity to the new environment produced by mixing the two types
Barrington, S., Choinire, D., Trigui, M., Knight, W., 2002. Effect of carbon source on
of sludge (food processing and waste water). On the whole, CPWW compost nitrogen and carbon losses. Bioresource Technol. 83, 189194.
was the most stable mixture at the beginning of the process and Bernal, M.P., Paredes, C., Snchez-Monedero, M.A., Cegarra, J., 1998. Maturity and
stability parameters of compost prepared with a wide range of organic wastes.
the OUR was the lowest; this was in agreement with the digestion
Bioresource Technol. 63, 9199.
process which this sludge underwent prior to composting, which Bremner, J.M., 1996. Nitrogen-total. In: Sparks, D.L. (Ed.), Methods of Soil Analysis.
mineralized most of the easily degradable OM (Tomei et al., Part 3 Chemical Methods. Soil Science Society of America, Madison, WI, USA.
2008). This mixture showed a mineralization rate very similar to Bustamante, M.A., Paredes, C., Marhuenda-Egea, F.C., Prez-Espinosa, A., Bernal,
M.P., Moral, R., 2008. Co-composting of distillery wastes with animal manures:
that of CPFP (Table 3). In other words, despite a lower quantity of carbon and nitrogen transformations in the evaluation of compost stability.
mineralizable OM, this was oxidized faster in the more stable Chemosphere 72, 551557.
mixture in a nutrient non-limiting environment. It is therefore per- Cameron, S.G., Schipper, L.A., 2010. Nitrate removal and hydraulic performance of
organic carbon for use in denitrication beds. Ecol. Eng. 36, 15881595.
ceived a lack of available nitrogen in CPWW during composting. Chefetz, B., Hatcher, P.G., Hadar, Y., Chen, Y., 1996. Chemical and biological
It is recognized that each biological redox reaction involves at characterization of organic matter during composting of municipal solid waste.
least two substrates: an oxidizing and a reducing agent, such as J. Environ. Qual. 25, 776785.
Ciavatta, C., Govi, M., Pasotti, L., Sequi, P., 1993. Changes in organic matter during
oxygen and organic substrate in the case of OM oxidation, oxygen stabilization of compost from municipal solid wastes. Bioresource Technol. 43,
and ammonium in the case of nitrication, nitrate and OM in the 141145.
case of denitrication. The maximum specic growth is a constant Eghball, B., Power, J.F., 1999. Phosphorus- and nitrogen-based manure and compost
applications. Corn production and soil phosphorus. Soil Sci. Soc. Am. J. 63, 895
typical for each microbial species, substrate and temperature; 901.
since composting mixtures are heterogeneous, this constant can Gao, M., Liang, F., Yu, A., Li, B., Yang, L., 2010. Evaluation of stability and maturity
be considered as the average of many microbial species and sub- during forced-aeration composting of chicken manure and sawdust at different
C/N ratios. Chemosphere 78, 614619.
strates. Theoretically a zero order kinetic can be adopted for each
Gilbert, Y., Le Bihan, Y., Aubry, G., Veillette, M., Duchaine, C., Lessard, P., 2008.
of all involved substrate; as a rule of thumb, each one of the in- Microbiological and molecular characterization of denitrication in biolters
volved substrate can be limiting to microbial growth and thus pro- treating pig manure. Bioresource Technol. 99, 44954502.
duce different kinetic patterns. The comparison between the OM Govi, M., Ciavatta, C., Gessa, C., 1993. Evolution of organic matter in sewage sludge:
a study based on the use of humication parameters and analytical
mineralization and the OUR course highlights that the limiting fac- electrofocusing. Bioresource Technol. 44, 175180.
tor for the composting rate can be the available nitrogen in Grigatti, M., Ciavatta, C., Gessa, C., 2004. Evolution of organic matter from sewage
CPFP+WW and CPWW. While the OM mineralization in CPFP tted sludge and garden trimming during composting. Bioresource Technol. 91, 163
an exponential model (rst order), both CPFP+WW and CPWW tted Grigatti, M., Cavani, L., Ciavatta, C., 2007a. A multivariate approach to the study of
a logistic model: it is well described in literature that logistic mod- the composting process by means of analytical electrofocusing. Waste Manage.
els reect the inuence of a limiting factor on the mineralization 27, 10721082.
Grigatti, M., Dios Prez, M., Blok, J.W., Ciavatta, C., Veeken, A., 2007b. A standardized
kinetic (Sobratee et al., 2009) which is very probably the amount method for the determination of the intrinsic carbon and nitrogen
of available nitrogen. CPWW mixture in fact was the most conserva- mineralization capacity of natural organic matter sources. Soil Biol. Biochem.
tive for N which increased during the process, this almost com- 39, 14931503.
Iannotti, D.A., Pang, T., Toth, B.L., Elwell, D.L., Keener, H.M., Hoitink, H.A.J., 1993. A
pletely depleted the mineral nitrogen forms. Similar trends were quantitative respirometric method for monitoring compost stability. Compost
found by Bustamante et al. (2008) with nitrate loss corresponding Sci. Util. 1, 5265.
to the slowdown of OM mineralization. Iannotti, D.A., Grebus, M.E., Toth, B.L., Madden, L.V., Hoitink, H.A.J., 1994. Oxygen
respirometry to assess stability and maturity of composted municipal solid
waste. J. Environ. Qual. 23, 11771183.
IPLA Istituto per le Piante da Legno e per lAmbiente (Institute for Wood Plants
5. Conclusions and Environment), 1998. Analytical Methods for Compost. Chemical, Physical,
Biological and Microbiological Determinations, Torino, Italy, p. 186.
Kalamdhad, A.S., Kazmi, A.A., 2009. Effects of turning frequency on compost
The adoption of different sludges in the mixtures strongly af- stability and some chemical characteristics in a rotary drum composter.
fected the OM mineralization course during the composting pro- Chemosphere 74, 13271334.
cess. The addition of food processing waste strongly enhanced Kobayashi, M., Matsuo, Y., Takimoto, A., Suzuki, S., Maruo, F., Shoun, H., 1996.
Denitrication, a novel type of respiratory metabolism in fungal mitochondrion.
the OM mineralization and the Norg losses, compared to the addi- J. Biol. Chem. 271, 1626316267.
tion of a digested sewage sludge. The parameters which are often Lasaridi, K.E., Stentiford, E.I., 1998. A simple respirometric technique for assessing
adopted in the description of the compost stability like the C/N, compost stability. Water Res. 32, 37173723.
 Liang, Y., Leonard, J.J., Feddes, J.J., McGill, W.B., 2004. A simulation model of
the DOC content, and the NH 4 N/NO3 N ratio, did not well de- ammonia volatilization in composting. In: Trans. ASABE 47, 16671680.
scribe the relationship between composting time and stabilization Mahimairaja, S., Bolan, N.S., Hedley, M.J., 1995. Denitrication losses of N from fresh
level reached by the mass. The study showed that the kind of mix- and composted manures. Soil Biol. Biochem. 27, 12231225.
Mason, I.G., 2006. Mathematical modelling of the composting process: a review.
ture and the time of composting strongly affected these parame- Waste Manage. 26, 321.
ters. This was particularly evident regarding the mineral nitrogen Meunchang, S., Panichsakpatana, S., Weaver, R.W., 2005. Co-composting of lter
course: nitrate losses were important, and were probably due to cake and bagasse; by-products from a sugar mill. Bioresource Technol. 96, 437
the denitrication operated by fungi especially in the last part of
MiPAF Ministero delle Politiche Agricole e Forestali (Italian Ministry of
the composting process. Mineral nitrogen appeared to be the rate Agricultural and Forestry Politics), 2001. Ofcial Analytical Methods for
limiting step in the composting process of more stabilized raw Fertilizers. Decreto Ministeriale 21 dicembre 2001. Supplemento Ordinario N.
materials as digested sewage sludge. The regression analysis 6, Gazzetta Ufciale n. 21 del 26 gennaio 2001, Roma, Italy.
Mupondi, L.T., Mnkeni, N.S., Muchaonyerwa, P., 2010. Effects of a precomposting
showed that the measurement of the OUR by means of a standard- step on the vermicomposting of dairy manurewaste paper mixtures. Waste
ized test proved to be a valid tool to describe the stabilization level Manag. Res.. doi:10.1177/0734242X10363142.
48 M. Grigatti et al. / Chemosphere 83 (2011) 4148

Rashad, F.M., Saleh, W.D., Mohamed, A., Moselhy, M.A., 2010. Bioconversion of rice Tremier, A., de Guardia, A., Massiani, C., Paul, E., Martel, J.L., 2005. A respirometric
straw and certain agro-industrial wastes to amendments for organic farming method for characterising the organic composition and biodegradation kinetics
systems: 1. Composting, quality, stability and maturity indices. Bioresource and the temperature inuence on the biodegradation kinetics, for a mixture of
Technol. 101, 59525960. sludge and bulking agent to be co-composted. Bioresource Technol. 96, 169
Slinkard, K., Singleton, V.L., 1977. Total phenols analysis: automation and 180.
comparison with manual methods. Am. J. Enol. Viticult. 28, 4955. Veijalainen, A.M., 2007. Sustainable Organic Waste Management in Tree-seedling
Sobratee, N., Mohee, R., Driver, M.F.B., 2009. Variation of broth composition by Production, vol. 217. Natural and Environmental Sciences. Doctoral
addition of broiler litter composting substrate extracts: inuence on faecal Dissertation. Kuopio University Publications C, p. 114.
bacterial growth. J. Appl. Microbiol. 107, 12871297. Zhou, Z., Takaya, N., Nakamura, A., Yamaguchi, M., Takeo, K., Shoun, H., 2002.
Tomei, M.C., Braguglia, C.M., Mininnia, G., 2008. Anaerobic degradation kinetics of Ammonia fermentation, a novel anoxic metabolism of nitrate by fungi. J. Biol.
particulate organic matter in untreated and sonicated sewage sludge: role of Chem. 277, 18921896.
the inoculum. Bioresource Technol. 99, 61196126. Zmora-Nahum, S., Markovitch, O., Tarchitzky, J., Chen, Y., 2005. Dissolved organic
Tontti, T., Heinonen-Tanski, H., Karinen, P., Reinikainen, O., Halinen, A., 2010. carbon (DOC) as a parameter of compost maturity. Soil Biol. Biochem. 37, 2109
Maturity and hygiene quality of composts and hygiene indicators in agricultural 2116.
soil fertilised with municipal waste or manure compost. Waste Manage. Res..

View publication stats