Simultaneous effect of initial moisture content and airflow rate on biodrying of sewage sludge.

w a t e r r e s e a r c h x x x ( 2 0 1 5 ) 1 e1 1

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/watres

Simultaneous effect of initial moisture content and airflow rate on biodrying of sewage sludge ~ ir*, Manuel Villegas Cesar Huilin Departamento de Ingenierı´a Quı´mica, Universidad de Santiago de Chile, Casilla 442, Correo 2, Santiago, Chile

article info

abstract

Article history:

The simultaneous effect of initial moisture content (initial Mc) and air-flow rate (AFR) on

Received 3 November 2014

biodrying performance was evaluated. For the study, a 32 factorial design, whose factors

Received in revised form

were AFR (1, 2 and 3 L/min kgTS) and initial Mc (59, 68 and 78% w.b.), was used. Using energy

18 March 2015

and water mass balance the main routes of water removal, energy use and efficiencies

Accepted 13 April 2015

were determined. The results show that initial Mc has a stronger effect on the biodrying

Available online xxx

than the AFR, affecting the air outlet temperature and improving the water removal, with higher maximum temperatures obtained around 68% and the lowest maximum matrix

Keywords:

temperature obtained at initial Mc ¼ 78%.Through the water mass balance it was found

Biodrying

that the main mechanism for water removal was the aeration, with higher water removal

Water mass balance

at intermediate initial Mc (68%) and high AFR (3 L/min kgTS). The energy balance indicated

Energy balance

that bioreaction is the main energy source for water evaporation, with higher energy

Initial moisture content

produced at intermediate initial Mc (68%). Finally, it was found that low values of initial Mc

Air flow rates

(59%) improve biodrying efficiency. © 2015 Elsevier Ltd. All rights reserved.

1.

Introduction

Biological drying or simply ‘‘biodrying’’, an alternative pretreatment method intended for combustion, has been developed in recent years. Biodrying, which is based on a process similar to composting, aims at removing water from biowastes with high water content using the heat generated during the aerobic degradation of organic substances, in addition to forced aeration (Frei et al., 2004; Navaee-Ardeh et al., 2011; Velis et al., 2009; Zhao et al., 2010). Among the parameters that affect the biodrying process, initial moisture content (initial Mc) and air-flow rate (AFR) have proven to be the most important ones. Although both have been studied independently, the simultaneous effect of initial Mc and AFR in the process has not been addressed in the studies. A

study of the simultaneous effect of these parameters may shed light on which of them has more influence on biodrying performance, information that so far has not been published in the literature. According to the literature, AFR is the main operational variable used for process control in biodrying, both in laboratory (Adani et al., 2002; Navaee-Ardeh et al., 2006; Sugni et al., 2005) and commercial applications. The inlet airflow rate can be manipulated to control matrix temperature, in turn, affecting the air dew point and biodegradation kinetics (Velis et al., 2009). The effect of AFR on biodrying has been studied recently by several researchers. Zhao et al. (2010) studied the effect of air-flow rate and turning frequency on bio-drying of dewatered sludge, showing that the higher air-flow rate, heat consumed by sensible heat of inlet air and heat utilization efficiency for evaporation was higher than the lower one. Cai et al. (2013) showed that forced aeration controlled the pile

* Corresponding author. Tel.: þ56 02 7181814. ~ ir). E-mail address: [email protected] (C. Huilin http://dx.doi.org/10.1016/j.watres.2015.04.046 0043-1354/© 2015 Elsevier Ltd. All rights reserved.

~ ir, C., Villegas, M., Simultaneous effect of initial moisture content and airflow rate on Please cite this article in press as: Huilin biodrying of sewage sludge, Water Research (2015), http://dx.doi.org/10.1016/j.watres.2015.04.046

2


temperature and improved evaporation, making it the key factor influencing water loss during the process of sewage sludge biodrying. Colomer-Mendoza et al. (2013) studied the effect of AFR on the biodrying of gardening wastes, showing that high airflow affects the biodrying process, because the thermophilic phase is avoided, so that the waste is dried only by physical phenomena and not by biodrying. Finally, Sharara et al. (2012) showed that high aeration level was superior in terms of both drying energy and time requirements than the other considered rates. None of these studies worked with different levels of initial Mc, therefore, the effect of both parameters on the biodrying performance should be evaluated. Moisture is a critical parameter involved in biodrying technology that influences the complex biochemical reactions associated with microbial growth and the biodegradation of organic matter that occurs during the process (Cai et al., 2012; Ryckeboer et al., 2003). The initial Mc is important because an excessively high initial Mc limits oxygen transport and microbial activity is hindered, invalidating the biodrying (Navaee-Ardeh et al., 2011). On the contrary, if initial Mc is too low, microbial activity will be slowed by lack of moisture, which results in reduced drying performance. Recently, Yang et al. (2014) studied the effect of initial Mc in the biodrying process. They showed that 50e70 wt% was the optimal initial Mc range for the sludge biodrying process, with VS reduction between 12.3 and 21.2%; however, they worked at only one constant air flow rate. Other studies in composting process (Komilis et al., 2011; Liang et al., 2003; Petric et al., 2009; Tremier et al., 2009) also showed the critical importance of initial Mc in the biodegradation process. One of the variables that should be taken into account in biodrying is the outlet air temperature. The increased outlet air temperature (as a consequence of the microbial activity) allowed the air to hold greater quantities of moisture, thus increasing the drying rate (Frei et al., 2004; Navaee-Ardeh et al., 2006). Several researchers have developed mass and energy balance by supposing that the outlet air temperature is equal to that of the matrix temperature (He et al., 2013; Yang et al., 2014; Zhao et al., 2010); however, the outlet air temperature is not necessarily equal to the matrix temperatures, depending mainly on the AFR value (Navaee-Ardeh et al., 2011). To date, only Frei et al. (2004) present values of outlet air temperatures, working with these values using only one AFR. Therefore, the effect of AFR and initial Mc on the outlet air temperature is one of the goals of this work. In light of the above, the study of the simultaneous effect of both initial Mc and AFR, on the biodrying performance has not been presented so far. Thus, this work focuses on investigating the interactive influence of these two parameters on outlet air temperature, water removal and energy utilization of dewatered sludge as a biomass resource. As a result of the mass and heat balance calculation, the water holding capacities of air-flow and heat uses were clarified.

2.

Materials and methods

2.1.

Characteristics of the raw material

Dewatered secondary sludge was obtained from a slaughterhouse wastewater treatment plant in Puente Alto, Santiago,

Chile. Sludge was dewatered by screw filter with the addition of organic flocculating agents. Wood shavings of 2.5 mm in average diameter were used as bulking agent. The characteristics of the raw materials are presented in Table 1.

2.2.

Experimental equipment and process operation

The biodrying experiments were performed in three 64 L cubic reactors (40 cm in height, 40 cm in width, and 40 cm in length) made of acrylic plastic connected to data acquisition system OPTO22. Heat losses were reduced by wall insulation provided by a 5 cm layer of polyurethane foam. A perforated baffle with a 2 mm mesh was fixed above the bottom to support the material and facilitate aeration. Temperature measuring ports were set in the middle of the reactor. The temperature and relative humidity of the inlet and outlet air were measured each 15 min using a series HMP110 humidity and temperature transmitter (Vaisala, Vantaa, Finland) with ±0.1 C and ±1.5% accuracy, respectively. A constant and uninterrupted air-flow rate was used in all the assays using a mini-compressor (150 L/ min, ACO-012, China) connected to the bottom of the column, while a rotameter (1e10 lpm, Veto, Santiago, Chile; 4e40 lpm, Veto, Santiago, Chile) was used to measure the air flow rate. In the reactors, matrix temperatures were measured each 15 min through a heat-resistant temperature electrode (Pt100, Veto, Chile), placed in the middle of substrate and in the inlet of bioreactor. All the sensors were connected through the acquisition module SNAP PAC OPTO22 (California, USA) on a PC. The schematic representation of the laboratory bioreactor is shown in Fig. 1. The biodrying process was studied through a 32 factorial design, whose factors were air flow rate AFR (1, 2 or 3 L/min kgTS) and initial Mc (59, 68 and 78% w.b.). The AFR levels were chosen based on the work of Huilinir and Villegas (2014), while the initial Mc values were chosen according to values reported by Yang et al. (2014) and Petric et al. (2009). Table 2 shows the experimental matrix. When the matrix temperature was close to the environmental temperature, the reactors were weighed. The matrix was removed from the reactors and loaded again after mixing for 30e60 min, as recommended by Zhao et al. (2010). Weight loss, which is directly related to the loss of moisture and VS from the sample, was measured using a portable scale. Before the material was loaded again into the bioreactor, three 10 g samples of mixed waste were collected to measure moisture content and volatile solids (VS) of the matrix.

2.3.

Analytical and statistical methods

The dry matter content was determined by drying the sample at 105 C for 24 h in an OV-180 Blue M (Stabil-Therm, USA) oven. The difference between the initial and final weights was the water content or the moisture content of the solids. The volatile solids content (VS) was analyzed by heating the sample at 550 C for 5 h in a muffle furnace. Nitrogen content was determined by the total Kjeldahl nitrogen method, using the modified. Nessler method (method 8075), with 0.25 g of homogenized material. Digestion was first done using digester Digesdahl Hach (HACH, USA), based on a single total digestion of organic


3


Table 1 e Characteristics of the raw materials. Material

Moisture content (w.b.,%)

Secondary sludge Wood shavings for Mc ¼ 59% Wood shavings for Mc ¼ 68% Wood shavings for Mc ¼ 78%

88.25 9.43 19.53 11.92

%C

± 8.36 ± 0.78 ± 2.25 ± 1.45

47.08 54.12 51.34 53.60

%N

± 1.25 ± 0.66 ± 1.23 ± 0.56

6.02 0.26 0.23 0.26

C/N ratio, g C g1 N

± 0.45 ± 0.02 ± 0.02 ± 0.03

7.82 208.15 223.22 206.15

Fig. 1 e Schematic diagram of lab-scale system for biodrying.

matter with sulfuric acid and hydrogen peroxide. The digested solution was used for determination of nitrogen using Hach's DR-3900 spectrophotometer (HACH, USA). To determine the differences between temperatures (matrix and outlet air) of each treatment, a one-factor variance analysis (ANOVA) was applied, using Excel 2010.

2.4.

Mass and energy balances

Biodrying process is a batch process in a non-steady state; the final water content is the result of the initial water content, water produced by metabolism, evaporated water and water loss by turning. The basic equation of the integral water

balance in the matrix, for any time between sampling, can be written as follows: Wmatrix;i Wmatrix;f ¼ Wevap þ Wturning Wmetabolic

(1.1)

where Wmatrix,i is the initial mass water in the matrix; Wmatrix,f is the final mass water in the matrix; Wevap is the mass water lost by evaporation; Wturning is the mass water lost by turning and Wmetabolic is the generated mass water by microbial metabolism. Initial and final mass water in the matrix was calculated with the mass change in the bioreactor and its moisture content, measured as percentage. Mass water generated by microbial metabolism was calculated assuming that water is

Table 2 e Experimental matrix, in real (and coded) levels for studying the biodrying process. Assay 1 2 3 4 5 6 7 8 9

Initial Mc, % w.b. 58.98 58.98 58.98 68.23 68.23 68.23 78.00 78.00 78.00

± 1.23 ± 1.23 ± 1.23 ± 1.45 ± 1.45 ± 1.45 ± 1.65 ± 1.65 ± 1.65

(1) (1) (1) (0) (0) (0) (þ1) (þ1) (þ1)

Air flow rate (AFR), L min1 kg1TS 1 2 3 1 2 3 1 2 3

(1) (0) (þ1) (1) (0) (þ1) (1) (0) (þ1)

Air flow rate (AFR), L min1 kg1VS 1.05 2.09 3.14 1.09 2.18 3.27 1.18 2.36 3.54

Initial VS content, % d.b 95.52 95.52 95.52 91.81 91.81 91.81 84.65 84.65 84.65

± 1.52 ± 1.52 ± 1.52 ± 2.34 ± 2.34 ± 2.34 ± 1.80 ± 1.80 ± 1.80

Initial mixture mass, kg

Sewage sludge: bulking agent ratio

4.50 4.50 4.50 6.77 9.00 8.00 14.71 14.48 15.74

2.00 2.00 2.00 3.33 3.33 3.33 10.00 10.00 10.00


4


generated only in the process of VS oxidation, where its value is:

(Qrad) and by turning (Qturning). The following equations were used:

Wmetabolic ¼ VSconsumed $YH2 O=VS

1. Biologically generated heat:

(1.2)

where YH2 O=BVS is a constant yield coefficient. This coefficient was assumed as the average value between values obtained from literature (Zhang et al., 2012). Mass water loss by evaporation was calculated using the inlet and outlet temperatures and relatively humidity of the air. This way, the evaporated water was: Ztf Wevap ¼

Fda ðtÞ$ðYi Yo Þdt

Qbio ¼ DVS$Hc

Where DVS is the VS consumption and Hc is the biodegradation enthalpy. 2. Heat consumed by aeration (sensible heat)

(1.3) Ztf

ti

where Fda is the inlet mass flow rate of dry air in the bioreactor and Yi andYo are the absolute humidity of inlet and outlet air, respectively. Dry air mass flow rate can be calculated as (Treybal, 1986): Qair

Fda ðtÞ ¼ 1 MWair

Qdryair ¼ cp;dryair $

MWwater Pv;w $ MWair Pabs Pv;w

Fda ðtÞ$ Tair;outlet Tair;inlet dt

(1.12)

ti

Ztf Qwatevap ¼ cp;watvap $

Fda ðtÞ$Yo ðtÞ$Tair;outlet ðtÞ

ti

Fda ðtÞ$Yi ðtÞ$Tair;inlet ðtÞ dt

(1.4)

þ MW1water $R$Tair;inlet ðtÞ

where Qair is the measured inlet air flow rate, MWair is the molecular mass of air, MWwater is the molecular mass of water, R is the universal constant of gases and T is the temperature of the humid inlet air. Specific humidity of air was calculated as: Y¼

(1.11)

(1.13)

3. Heat consumed by water evaporation (latent heat)

Ztf (1.5)

Qevapo ¼

Fda ðtÞ$levap ðtÞdt

(1.14)

ti

where Pv,w is the actual water vapor pressure in air at a defined temperature and Pabs is the absolute pressure of gas (supposed 1 atm). Actual vapor pressure can be calculated as:

where levap is the latent heat of water evaporation and can be calculated as (Zhao et al., 2010):

Pv;w ¼ HR$Pvs;w

levap ðtÞ ¼

(1.6)

where Pvs,w is the saturation water vapor pressure. This variable is dependent on temperature and can be mathematically modeled by the Antoine equation: log10 Pvs;w

2238 ¼ 8:896 Tð KÞ

(1.7)

Therefore, measuring the temperature and relative humidity of inlet and outlet air, the Wevap was calculated. The mass water loss by turning was calculated as: Wturning ¼ Wb;i Wb;f Wevap þ Wmetabolic

(1.8)

The energy balance was calculated based in the methodology presented by Zhao et al. (2010) and He et al. (2013). The energy balance equation is as follows: Qbio ¼ Qconsu þ Qloss

(1.9)

where Qbio is the energy generated by bioreaction; Qconsu is the energy consumed by the process and Qloss is the energy lost in the process. The consumed energy was calculated as:

Tmatrix ðtÞ þ 32 1055 $9 $ 1093:7 0:5683$ 5 454

(1.15)

4. Heat consumed by matrix temperature increase (sensible heat): Qwater ¼ Wmatrix $cp;water $DTmatrix

(1.16)

Qsolid ¼ Msolid $cp;solid $DTmatrix

(1.17)

In this case, DTmatrix was the change of matrix temperature in a time element, i.e., between matrix mass water measurements. 5. Heat consumed by radiation: Qradi ¼ s$Atop $Fa $Fe $ T4air;0 T4a

(1.18)

6. Heat consumed by turning:

Qconsum ¼ Qdryair þ Qwatvap þ Qwater þ Qsolid þ Qevapo þ Qrad þ Qturning (1.10) According to the last equations, the energy is consumed by aeration (Qdryair and Qwatvap), by evaporated water (Qevapo), by matrix temperature increase (Qwater and Qsolid), by radiation

Qturning ¼ Wmatrix $cp;water $ðTmatrix Ta Þ þ Msolid $cp;solid $ðTmatrix Ta Þ (1.19) The parameters used in the balances are given in Table 3.



3.

Results

3.1.

Effect of initial Mc and AFR on temperature profiles

Temperatures are important for the biodrying process. The increasing of matrix temperatures increases the air temperature and improves its capacity to remove water. The effect of Mc and AFR on the temperatures are shown in Figs. 2 and 3. The final moisture content in the matrix at initial Mc ¼ 59%, was 52.5% for initial AFR of 1 and 2 L/min kgTS and 51% for AFR ¼ 3 L/min kgTS. At initial Mc ¼ 68%, the final moisture content in the matrix varied between 64% (1 and 2 L/min kgTS) and 60% (3 L/min kgTS); finally, at initial Mc ¼ 78% the moisture content reduction was limited, varying between 1 and 2 % in all the AFR studied. The analysis of moisture content was re~ ir (2014). ported in Villegas and Huilin As shown in Fig. 2, after the initial filling of the reactors, a rapid temperature increase occurred in all of them, indicating a high microbial activity in the first 24 h. The increase was more pronounced at low AFR for all Mc studied. In all the experiments, the matrix temperature was higher than both inlet and outlet temperatures. In Fig. 2A, it is shown that the temperature rose up to 48 C at AFR ¼ 1 L/min kgTS, while at AFR ¼ 3 L/min kgTS the temperature reached a maximum of only 34 C. However, at the three AFR studied, the matrix temperature was higher than environmental temperature at least the first 3 days. It is also observed that the temperatures of the assays were affected by the dayenight cycle, showing peaks every 24 h. Statistical analysis of daily average temperatures throughout the duration of the study showed a significant difference (P < 0.05) between different aeration levels. At initial Mc ¼ 68% (Fig. 2B) the maximum temperature (60 C) was obtained in all the assays, occurring at the lowest studied AFR (1 L/min kgTS). At the highest AFR (3 L/min kgTS), there was also a high temperature in comparison to the temperatures obtained at initial Mc of 59% and 78%, reaching values over 37 C. As the assay with initial Mc ¼ 59%, the temperatures were affected by the dayenight cycle, and the statistical analysis of daily average temperatures showed a significant difference (P < 0.05) between different aeration levels. Finally, Fig. 2C shows that at initial Mc ¼ 78% the temperature's profile was different from the other conditions studied. The maximum temperature for all the cases was

Table 3 e Parameters used in mass and energy balances. Parameter YH2 O=VS , g H2O/g VSconsumed MWair, g/mol MWwater, g/mol R, atm L/K mol Hc, kJ/kg VS cp,dryair, kJ/kg C cp,watvap, kJ/kg C s, kJ/s m2 K4 Atop, m2 Fa, dimensionless Fe, dimensionless cp,water, kJ/kg C cp,solid, kJ/kg C

Value

Reference

0.41 28.97 18.02 0.083 21,000 1.004 1.841 5.67 1011 0.5867 0.7911 0.85 4.184 1.046

Zhang et al., 2012 Treybal, 1986 Treybal, 1986 Treybal, 1986 He et al., 2013 Zhao et al., 2010 Zhao et al., 2010 Ahn et al., 2007 Calculated Calculated Ahn et al., 2007 Zhao et al., 2010 Zhao et al., 2010

5

obtained after three days, decreasing slowly and reaching levels close to environmental temperature after 8 days. The maximum temperatures obtained for these conditions were around 40 C at AFR ¼ 1 L/min kgTS. As in the other assays, the temperatures were affected by the dayenight cycle, but the statistical analysis of daily average temperatures does not show a significant difference (P > 0.05) between different aeration levels. This result clearly shows that the AFR range studied does not affect the average temperature profiles at high initial Mc. Fig. 3 shows the outlet air temperature profiles for all the conditions studied. As the matrix temperature, the outlet air temperatures at AFR ¼ 1 L/min kgTS were always higher than outlet air temperatures at higher AFR, although the difference was lower than that found in the matrix. In addition, outlet air temperatures at AFR ¼ 2 and 3 L/min kgTS were similar, being lower than 35 C for all the assays. At initial Mc ¼ 59% (Fig. 3A), the outlet air temperature ranged between 30 and 35 C and the profiles were affected by dayenight cycle. At higher AFR, the increasing of outlet air temperature was lower and it occurred only the first day of operation, being it very close to the environmental temperature after the first day. Even though the values obtained for the outlet air temperature under this condition were similar, the ANOVA analysis of the average temperatures showed a significant difference (P < 0.05) between different aeration levels. At initial Mc ¼ 68% (Fig. 3B), the outlet air temperatures were higher than those obtained at initial Mc ¼ 59%, especially on the first two days. This behavior coincides with the higher temperatures obtained in the matrix (Fig. 2B). At AFR ¼ 1 L/min kgTS the highest outlet air temperatures were obtained, reaching values up to 43 C, situation that improved the water removal (see Table 4). At AFR ¼ 2 and 3 L/min kgTS, outlet air temperature increased only up to 36 C and decreased rapidly the second day. After day 2, the outlet air temperature presented values close to environmental temperature. Contrary to the temperatures obtained at initial Mc ¼ 59%, average temperatures did not show a significant difference (P > 0.05) between different aeration levels. At initial Mc ¼ 78% (Fig. 3C), the profile of outlet air temperature follows the matrix temperature profile, with a low increasing of temperature. As the rest of the assays, AFR ¼ 1 L/min kgTS obtained the highest air temperature, while AFR ¼ 2 and 3 L/min kgTS showed similar temperature values, however, average daily temperatures showed a significant difference (P < 0.05) between different aeration levels. From Figs. 2 and 3 it is possible to compare the matrix temperatures with the outlet air temperatures. As indicated above, the difference between matrix temperature and outlet air temperature (DTma) increases at lower AFR in all the assays. For instance, at initial Mc ¼ 68%, the higher DTma at AFR ¼ 1 L/min kgTS was 19 C, while at AFR ¼ 2 L/min kgTS and AFR ¼ 3 L/min kgTS was only 10 C and 2 C, respectively. This behavior shows the effect of AFR on the energy transfer between the matrix and the air. At higher AFR there is a lower transfer energy resistance because of the decrease of the boundary layer on the solid matrix, allowing higher energy transfer and obtaining closer values between matrix and outlet air temperature values. This was also reported by Du and Wang (2001), who indicated that by increasing fluid velocity within the pores of the porous media, the heat transfer


6


Fig. 2 e Variation of matrix temperature at different AFR. A) At initial Mc ¼ 59%; B) At initial Mc ¼ 68%; C) At initial Mc ¼ 78%.

coefficient between the solid phase and the fluid phase would increase. As a result, the temperature difference between the solid phase and the fluid phase would decrease, as was observed in this work. Initial Mc affects the matrix temperature profiles and the outlet air temperature profiles. According to the results, at initial Mc ¼ 68% the maximum matrix and outlet air temperatures were always higher than the maximum temperatures obtained at initial Mc ¼ 59% and 78%, for all the AFR studied. This behavior can be explained because values of initial Mc close to 68% generated a better microbial environment in the matrix. A similar situation was obtained at initial Mc of 59%, where at AFR ¼ 1 L/min kgTS the temperature rose up to 48 C in the first 24 h. On the contrary, at high initial Mc (78%), the system had a delay in the temperature increase, with slow microbial activity and lower maximum temperatures of the matrix. This also was observed by Yang et al. (2014), who showed that the optimal initial Mc of sludge for biodrying seemed to be in the range of 50e70 %, whereas microbial activity was lower when the Mc was either too high or too low. A similar situation was reported for the composting process (Haug, 1993; Petric et al., 2009). However, the outlet air temperature at AFR ¼ 1 L/min kgTS and initial Mc ¼ 78% reached values close to the values observed at AFR ¼ 1 L/min kgTS and

initial Mc ¼ 59%, showing that the delay in the increasing of temperature does not affect the energy transfer. Regarding the AFR effect, at low AFR the temperature was always higher than high AFR values for both matrix and outlet temperatures. The higher temperatures at lower AFR has been reported by Zhao et al. (2010) working with sewage sludge, Sharara et al. (2012) working with dairy manure and Huilinir and Villegas (2014) working with secondary paper and pulp sludge. In all the cases, the low AFR allows to keep the energy inside the reactor, avoiding energy loss by aeration. Thus, it is possible to conclude that the initial Mc has a remarkable effect on the matrix and outlet air temperature, with higher maximum temperatures obtained around 68% and the lowest maximum matrix temperature obtained at initial Mc ¼ 78%. Lower values of AFR obtained higher matrix and outlet temperatures at all the initial Mc studied; however, the difference between matrix and outlet air temperatures increases under these conditions.

3.2. Effect of initial Mc and AFR on water removal, vs removal, energy balance and biodrying efficiency In order to study the effect of initial Mc and AFR in the water removal, a water mass balance was done, and whose results


7


Fig. 3 e Variation of outlet air temperature at different AFR and initial Mc. A) Initial Mc ¼ 59%; B) Initial Mc ¼ 68%; C) Initial Mc ¼ 78%.

are shown in Table 4. It is observed that initial Mc influences the water removal, where the initial Mc of 68% showed the highest moisture removal conditions. This behavior can be attributed to the higher outlet air temperatures obtained in this condition. The outlet air temperature affects the air capacity of water removal, with an increase of the saturation capacity of the air at higher temperatures. In addition, water

removed by evaporation during the time with higher outlet air temperatures (first two days) represented between 38 and 20% of water removed in all the assays. The effect of higher outlet air temperatures was more remarkable at initial Mc ¼ 68%, where the water evaporated during the first two days was higher than 32%. In fact, under this initial Mc, the water evaporated during the first two days was higher at lower AFR

Table 4 e Mass balance of biodrying assays. Initial Mc, %

59%

68%

78%

1

2

3

1

2

3

1

2

3

Total water removal, kg Evaporated water, kg Generated water, kg Water loss by turning, kg Water removal by evaporation, % Water removal by evaporation first 2 days, %

0.633 0.602 0.017 0.047 95.187 28.835

0.462 0.448 0.013 0.027 97.120 35.463

0.660 0.638 0.017 0.039 96.649 37.547

0.671 0.657 0.020 0.034 97.926 38.296

1.247 1.195 0.103 0.155 95.882 32.202

1.423 1.392 0.082 0.114 97.762 35.799

0.875 0.840 0.052 0.087 96.015 21.049

1.288 1.238 0.104 0.154 96.071 22.540

1.502 1.471 0.073 0.104 97.948 29.254

Total water removal, %

15.329

14.205

15.748

15.840

20.500

26.367

7.606

11.376

12.198

AFR, L/min kgTS


8


due to the higher outlet air temperature reached (Fig. 3B). However, when the outlet air temperatures were similar, the water evaporation was higher under conditions with higher AFR. This is clear at initial Mc ¼ 59%, where outlet air temperatures are similar and the water removal depends mainly on dry air volume. Regarding the AFR, it also affects the water removal, with higher water removal at higher AFR. This behavior is stronger at initial Mc of 68% and 78%, while at 59% the effect of AFR was slightly superior to lower AFR. The higher removal at higher AFR can be attributed to the different volume of dry air used. At higher AFR, a higher dry air volume was injected to the system allowing an increase of the rate of water removal. According to the Table 4, the main mechanism of water removal was evaporation, with values higher than 95% in all the assays. These values are higher that reported in the literature (Zawadzka et al., 2010; Zhao et al., 2010) and can be attributed to the continuous aeration used in this work. Nevertheless, the water removal obtained in this work was lower than values reported in literature. Zhao et al. (2010) reported water removal between 58% and 66%, while ColomerMendoza et al. (2013) reported water removal between 20% and 69%. This difference can be attributed to the length of our assays. Our experiments lasted 8 days, 60% lower than the reported by other researchers (20 days). Thus, it is possible to say that the main mechanism for water removal in all the assays was the aeration, with higher water evaporation at higher dry air flow rates and higher outlet air temperatures. Fig. 4 shows the water removal rate (kg d1) during all the assays. The tendency was a decreasing drying rate with time, except for trials at initial Mc 59%, where the drying rate had no clear tendency (Fig. 4A). At initial Mc 68% (Fig. 4B), drying rate decreases continuously with time, due to the higher temperatures obtained the first three days. On days 6 and 8 there are negative drying rate values, indicating that on these two days the matrix was wetted. At initial Mc ¼ 78%, the drying rate decrease was slower, due to the lower and slower temperature increase under these conditions. Regarding the effect of AFR on drying rate, it was clear that lower drying rates were obtained at lower AFR in all the assays. In Fig. 4A, even though the trend is not clear, the drying rates values at 1 L/min kgTS are lower than the rest of AFR. At initial Mc of 68% and 78% (Fig. 4B and C) the drying rate was also always faster at higher AFR, except on the two first days of the experiment, where the temperature at 2 L/min kgTS was higher than the rest of the conditions. This behavior was observed also by Zhao et al. (2010), who indicated that the air-flow rate had more impact on water removal than on VS loss. Thus, both parameters (AFR and initial Mc) have an important effect on the drying rate. Fig. 5 shows the VS removal in all the assays. In this figure it is possible to observe that initial Mc has a remarkable effect on the VS consumption. The highest VS consumption obtained was at initial Mc ¼ 68%, where in all the AFR studied the consumption was higher than 15%. This situation, together with the temperature rise, shows that the initial Mc close to 70% increases the microbial activity. At initial Mc ¼ 59%, the VS consumption is lower than 6% in all the AFR studied, being the lowest VS consumption obtained in all experiments. At

initial Mc ¼ 78% the VS consumption is also low, with maximum VS consumption of 11% at AFR ¼ 2 L/min kgTS. According to Fig. 5, there is no effect of AFR on the VS consumption. At initial moisture content of 59% and 68%, the VS consumption was similar at different values of AFR, while at initial Mc ¼ 78%, the VS consumption at 1 L/min kgTS was the lowest. This situation can be attributed to low Free Air ~ ir, 2014) under this Space (FAS) values (Villegas and Huilin condition, which increases the resistance to the oxygen diffusion in the matrix. Therefore, under these conditions the AFR does not have an effect on the VS consumption. The VS consumptions are according to the levels founded in other reports. Yang et al. (2014) reported consumption between 12% and 21% of initial VS at different initial Mc. Petric et al. (2009), working in poultry manure composting, reported values between 12.75 and 33.57 at different initial Mc. The results found in this work show that VS consumption depends on initial Mc values. Thus, it can be said that the range of initial Mc studied affect the VS consumption, with higher VS consumption at initial Mc close to 70%. The AFR range studied does not have effect on the VS consumption. Researchers have developed Biodrying Index (I) as performance indicator for Biodrying process expressed as (Zhang et al., 2008): It ¼

WLt OLt

(1.20)

where It is I at time t; OLt (kg) the VS loss at time t; and WLt(kg) the water loss at time t. Fig. 6 shows the biodrying index for all the assays. It is shown that I values are higher at lower initial Mc and higher AFR, however, the influences of each one are different. Initial Mc has a strong effect on I, being its value up to three times higher at initial Mc ¼ 59% than the values obtained at initial Mc of 68% and 78%. On the other hand, I has its lowest values at initial Mc ¼ 68%, even though the water removal under this conditions was the highest. This can be explained by the higher VS consumption under this condition (see Fig. 5), in which the I value decreases. The values of I obtained in this work are higher than those reported in the literature. Zhang et al. (2008) obtained values that varied between 3.27 and 6, while Zhao et al. (2010) obtained values around 6. Only Yang et al. (2014) obtained values as high as 19.8, values slightly higher than the values obtained in this work. Without taking the values of I obtained at initial Mc ¼ 59% into account, the values of I ranged between 5 and 8.4, also being higher to the values obtained by other reports. According to the results, initial Mc has a higher effect on the Biodrying Index (I) than the AFR, with higher values of I at initial Mc close to 60%. Table 5 shows the energy balance results for all the assays. It is shows that the energy produce by bioreaction varied depending on the initial Mc conditions, with lower energy produce at lower initial Mc. Regarding the effect of AFR on the energy produce by bioreaction, its effects is not clear. At high initial Mc (68% and 78%), the higher energy was obtained at AFR ¼ 2 L/min kgTS, while the lowest energy was obtained at AFR ¼ 1 L/min kgTS. However, at initial Mc ¼ 59%, the higher energy was obtained at AFR ¼ 3 L/min kgTS, while the lower energy was obtained at AFR ¼ 2 L/min kgTS. This behavior can



9

Fig. 4 e Variation of drying rate at different AFR and initial Mc. A) Initial Mc ¼ 59%; B) Initial Mc ¼ 68%; C) Initial Mc ¼ 78%.

Fig. 5 e Variation of VS removal (%) at different initial Mc and AFR.

Fig. 6 e Variation of biodrying efficiency at different initial Mc and AFR.


10


Table 5 e Heat generation and utilization of biodrying system (kJ). Initial Mc, % AFR, L/min kgTS Qbio Qevapo Qwater Qrad Qdryair Qwatvap Qsolid Qturning Qconsuming Qevap/Qbio

59%

69%

78%

1

2

3

1

2

3

1

2

3

867.075 1437.372 100.056 4.405 127.096 33.831 17.278 11.570 1731.608 1.66

682.556 1074.719 128.528 1.366 59.406 22.643 17.529 15.395 1319.585 1.57

880.179 1532.917 66.041 0.251 31.820 30.208 8.377 28.278 1641.337 1.74

1997.654 1558.130 145.965 6.650 155.967 39.355 17.172 13.234 1936.473 0.78

5299.754 2853.490 204.040 2.656 188.821 63.350 23.727 197.156 3533.238 0.54

4208.740 3347.399 13.098 1.785 168.947 71.740 1.121 6.212 3610.302 0.80

2651.576 2002.084 29.022 13.469 397.332 56.552 30.018 6.313 2522.164 0.76

5312.186 2965.830 449.115 2.751 300.064 62.780 115.019 5.506 3890.053 0.56

3753.407 3520.958 416.048 2.404 406.496 74.546 28.710 2.037 4447.125 0.94

be explained with the VS consumption tendency. As the energy produce by bioreaction was associated to the VS consumption and this parameter was not affected by AFR, the bioreaction energy was not affected by AFR in an important way. From Table 5, it is clear that the main consumption of energy was through water evaporation. At higher initial Mc (68 and 78%), the Qevapo varied between 54% (initial Mc ¼ 68% and AFR ¼ 2 L/min kgTS) and 94% (initial Mc ¼ 78% and AFR ¼ 3 L/ min kgTS), being the energy of biochemical reaction enough to evaporate the water. In comparison to the literature (He et al., 2013; Zhao et al., 2010), these values are out of the reported range and may be attributed to the different conditions studied in this work. At initial Mc ¼ 59%, the energy used for water evaporation was higher than the energy coming from the biochemical reaction. A similar situation, but at initial Mc superior to 75%, was reported by Yang et al. (2014), who indicated that if the energy coming from the bioreaction was inferior to the energy used by evaporation, biodrying was not active. However, Navaee-Ardeh et al. (2010) proposed that the energetic biodrying efficiency index (h ¼ Q_ evap =Q_ bio ) should vary between 1.5 and 3.5 in order to increases its efficiency, showing that the energy for evaporation can be higher than the energy coming from the bioreaction and, therefore, biodrying can be active under conditions with low VS consumption. At initial Mc ¼ 59%, the percentage of energy produce by biochemical reaction was a 60, 64 and 57% (h of 1.66, 1.57 and 1.74, respectively) of the energy required by evaporation at AFR ¼ 1, 2 and 3 L/min kgTS, respectively. This shows that, even though the system could use other energy sources (inlet air energy) for water evaporation, the biochemical reaction is the main mechanism for obtaining energy in the biodrying process. Table 5 also shows that the AFR has effects on the Qevap/Qbio ratio, obtaining higher ratios at higher AFR under all initial studied Mc. Furthermore, the lowest Qevap/Qbio ratio always was obtained at AFR ¼ 2 L/min kgTS. At higher initial Mc (68% and 78%), this situation can be explained by the higher VS consumption under these conditions, a situation that generated more energy. At lower initial Mc (59%), there is lower energy produce and also lower energy by water evaporation. This also shows that the main source of energy was the bioreaction. Thus, the energy coming from the bioreaction is the main energy source for water evaporation. Initial Mc content has a strong effect on the energy produce by bioreaction and the

Qevap/Qbio ratio, with lower energy produce by bioreaction and higher Qevap/Qbio ratios at lower initial Mc. AFR also affects the energy production, with the lowest Qevap/Qbio ratio at AFR ¼ 2 L/min kgTS.

4.

Conclusions

Initial Mc has a stronger effect on biodrying than the AFR, affecting the air outlet temperature and improving the water removal, with higher maximum temperatures obtained around 68% and the lowest maximum matrix temperature obtained at initial Mc ¼ 78%. The water mass balance indicated that the main mechanism for water removal was the aeration, with higher water removal at intermediated initial Mc (68%) and high AFR (3 L/min kgTS). The energy balance indicated that bioreaction is the main energy source for water evaporation, with higher energy produce at intermediated initial Mc (68%). At low values of initial Mc (59%) the biodrying efficiency increases, with higher values of Biodrying Index and Qevap/Qbio ratios.

Acknowledgments This work was supported by FONDECYT Project No. 11121160.

references

Adani, F., Baido, D., Calcaterra, E., Genevini, P., 2002. The influence of biomass temperature on biostabilization-biodrying of municipal solid waste. Bioresour. Technol. 83 (3), 173e179. Ahn, H.K., Richard, T.L., Choi, H.L., 2007. Mass and thermal balance during composting of a poultry manuredwood shavings mixture at different aeration rates. Proc. Biochem. 42 (2), 215e223. Cai, L., Chen, T.-B., Gao, D., Zheng, G.-D., Liu, H.-T., Pan, T.-H., 2013. Influence of forced air volume on water evaporation during sewage sludge bio-drying. Water Res. 47 (13), 4767e4773. Cai, L., Gao, D., Chen, T.-B., Liu, H.-T., Zheng, G.-D., Yang, Q.-W., 2012. Moisture variation associated with water input and evaporation during sewage sludge bio-drying. Bioresour. Technol. 117 (0), 13e19.



Colomer-Mendoza, E.J., Herrera-Prats, L., Robes-Martinez, F., Gallardo-Izquierdo, A., Pina-Guzman, A.B., 2013. Effect of airflow on biodrying of gardening wastes in reactors. J. Environ. Sci. China 25 (5), 865e872. Du, H., Wang, X., 2001. Forced convective heat transfer for fluid flowing through a porous medium with internal heat generation. Heat Transf. Asian Res. 30 (3), 213e221. Frei, K.M., Cameron, D., Stuart, P.R., 2004. Novel drying process using forced aeration through a porous biomass matrix. Dry. Technol. 22 (5), 1191e1215. Haug, R.T., 1993. The Practical Handbook of Compost Engineering, second ed. CRC Press LLC, Florida. He, P.J., Zhao, L., Zheng, W., Wu, D., Shao, L.M., 2013. Energy balance of a biodrying process for organic wastes of high moisture content: a review. Dry. Technol. 31 (2), 132e145. Huilinir, C., Villegas, M., 2014. Biodrying of pulp and paper secondary sludge: kinetics of volatile solids biodegradation. Bioresour. Technol. 157, 206e213. Komilis, D., Evangelou, A., Voudrias, E., 2011. Monitoring and optimizing the co-composting of dewatered sludge: a mixture experimental design approach. J. Environ. Manag. 92 (9), 2241e2249. Liang, C., 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 (2), 131e137. Navaee-Ardeh, S., Bertrand, F., Stuart, P.R., 2011. Development and experimental evaluation of a 1D distributed model of transport phenomena in a continuous biodrying process for pulp and paper mixed sludge. Dry. Technol. 29 (2), 135e152. Navaee-Ardeh, S., Bertrand, F., Stuart, P.R., 2006. Emerging biodrying technology for the drying of pulp and paper mixed sludges. Dry. Technol. 24 (7), 863e878. Navaee-Ardeh, S., Bertrand, F., Stuart, P.R., 2010. Key variables analysis of a novel continuous biodrying process for drying mixed sludge. Bioresour. Technol. 101 (10), 3379e3387. Petric, I., Sestan, A., Sestan, I., 2009. Influence of initial moisture content on the composting of poultry manure with wheat straw. Biosyst. Eng. 104 (1), 125e134. Ryckeboer, J., Mergaert, J., Vaes, K., Klammer, S., De Clercq, D., Coosemans, J., Insam, H., Swings, J., 2003. A survey of bacteria

11

and fungi occurring during composting and self-heating processes. Ann. Microbiol. 53 (4), 349e410. Sharara, M.A., Sadaka, S., Costello, T.A., VanDevender, K., 2012. Influence of aeration rate on the physio-chemical characteristics of biodried dairy manure e wheat straw mixture. Appl. Eng. Agric. 28 (3), 407e415. Sugni, M., Calcaterra, E., Adani, F., 2005. Biostabilizationbiodrying of municipal solid waste by inverting air-flow. Bioresour. Technol. 96 (12), 1331e1337. Tremier, A., Teglia, C., Barrington, S., 2009. Effect of initial physical characteristics on sludge compost performance. Bioresour. Technol. 100 (15), 3751e3758. Treybal, R., 1986. Mass Transfer Operations, third ed. McGrawHill, Mexico. Velis, C.A., Longhurst, P.J., Drew, G.H., Smith, R., Pollard, S.J.T., 2009. Biodrying for mechanical-biological treatment of wastes: a review of process science and engineering. Bioresour. Technol. 100 (11), 2747e2761. ~ ir, C., 2014. Biodrying of sewage sludge: Villegas, M., Huilin kinetics of volatile solids degradation under different initial moisture contents and air-flow rates. Bioresour. Technol. 174 (0), 33e41. Yang, B., Zhang, L., Jahng, D., 2014. Importance of initial moisture content and bulking agent for biodrying sewage sludge. Dry. Technol. 32 (2), 135e144. Zawadzka, A., Krzystek, L., Stolarek, P., Ledakowicz, S., 2010. Biodrying of organic fraction of municipal solid wastes. Dry. Technol. 28 (10), 1220e1226. Zhang, D.-Q., He, P.-J., Jin, T.-F., Shao, L.-M., 2008. Bio-drying of municipal solid waste with high water content by aeration procedures regulation and inoculation. Bioresour. Technol. 99 (18), 8796e8802. Zhang, Y., Lashermes, G., Houot, S., Doublet, J., Steyer, J.P., Zhu, Y.G., Barriuso, E., Garnier, P., 2012. Modelling of organic matter dynamics during the composting process. Waste Manag. 32 (1), 19e30. Zhao, L., Gu, W.-M., He, P.-J., Shao, L.-M., 2010. Effect of air-flow rate and turning frequency on bio-drying of dewatered sludge. Water Res. 44 (20), 6144e6152.


Biodrying of sewage sludge: kinetics of volatile solids degradation under different initial moisture contents and air-flow rates.

Effects of cationic polyacrylamide characteristics on sewage sludge dewatering and moisture evaporation.

Influence of Initial Moisture Content on Heat and Moisture Transfer in Firefighters' Protective Clothing.

Decreasing effect and mechanism of moisture content of sludge biomass by granulation process.

Thermophilic anaerobic digestion of sewage sludge with high solids content.

Effect of soil moisture content on the splash phenomenon reproducibility.

Effect of grain moisture content on the degradation rate of methyl phoxim in corn, sorghum, and wheat.

The effects of sewage sludge and sewage sludge biochar on PAHs and potentially toxic element bioaccumulation in Cucumis sativa L.

Hydrothermal carbonisation of sewage sludge: effect of process conditions on product characteristics and methane production.

Effect of whey storage on biogas produced by co-digestion of sewage sludge and whey.

Simultaneous Determination of Fluoroquinolones and Sulfonamides Originating from Sewage Sludge Compost.

EU policy on sewage sludge utilization and perspectives on new approaches of sludge management.

Effective management of sewage sludge.

Mycoflora of activated sewage sludge.

PAHs content of sewage sludge in Europe and its use as soil fertilizer.

Effect of pyrolysis temperature on chemical and physical properties of sewage sludge biochar.

Use of sewage and sludge in agriculture.

Anaerobic codigestion of sewage sludge and glycerol, focusing on process kinetics, microbial dynamics and sludge dewaterability.

Mineral content of selected tissues and feces of rats fed turnip greens grown on soil treated with sewage sludge.

Effect of solarization on the removal of indicator microorganisms from municipal sewage sludge.

The effect of moisture content on physicochemical properties of extruded waxy and non-waxy rice flour.

Effect of grain moisture content during milling on pasting profile and functional properties of amaranth fractions.

Stabilization of sewage sludge by different biochars towards reducing freely dissolved polycyclic aromatic hydrocarbons (PAHs) content.

Microwave oxidation treatment of sewage sludge.