This is the accepted version of the journal article: González Alé, Daniel; Guerra-Gorostegi, Nagore; Colón Jordà, Joan; [et al.]. «Filling in sewage sludge biodrying gaps : Greenhouse gases, volatile organic compounds and odour emissions». Bioresource Technology, Vol. 291 (November 2019), art. 121857. DOI 10.1016/j.biortech.2019.121857 This version is available at https://ddd.uab.cat/record/281350 under the terms of the license 1 Filling in sewage sludge biodrying gaps: Greenhouse gases, volatile organic 2 compounds and odour emissions 3 a, b 4 Daniel González , Nagore Guerra c, Joan Colón c, David Gabriel b, Sergio Ponsá c, 5 Antoni Sánchez a, * 6 a 7 Composting Research Group (GICOM) Dept. of Chemical, Biological and 8 Environmental Engineering, Universitat Autònoma de Barcelona, 08193-Bellaterra 9 (Barcelona), Spain. b 10 Group of biological treatment of liquid and gaseous effluents (GENOCOV) Dept. of 11 Chemical, Biological and Environmental Engineering, Universitat Autònoma de 12 Barcelona, 08193-Bellaterra (Barcelona), Spain 13 c BETA Technology Centre: “U Science Tech”, University of Vic-Central University of 14 Catalonia, 08500 Vic, Barcelona, Spain 15 16 * Corresponding author: 17 Antoni Sánchez 18 Composting Research Group (GICOM), Dept. of Chemical, Biological and 19 Environmental Engineering 20 Universitat Autònoma de Barcelona 21 Bellaterra, 08193 (Spain) 22 Email address: antoni.sanchez@uab.cat 1 23 Abstract 24 In the present work, a complete study of the sewage sludge (SS) biodrying 25 technology was conducted at bench-scale, aiming at assessing its performance and 26 providing a valuable insight into the different gaseous emission patterns found for 27 greenhouse gases (GHG) and odorant pollutants. As process key parameters, 28 temperature, specific airflow, dynamic respiration index, final moisture content and 29 Lower Calorific Value (LCV) were evaluated. At the end of the biodrying, a product 30 with a 35.9% moisture content and a LCV of 7.1 MJ·kg-1product was obtained. GHGs 31 emission factor was 28.22 kgCO2eq per Mg of initial mass of dry matter in the SS (DM0- 32 SS). During the biodrying process, maximum odour concentration measured was 3043 33 ou·m-3 and the estimated odour emission factor of the biological treatment was 34 3.10E+07 ou per Mg DM0-SS. Finally, VOCs were completely identified and 35 quantified. The most abundant VOCs found in the biodrying gaseous emissions were 36 terpenes, sulphur-compounds and ketones. 37 Keywords: Sewage sludge, biodrying, gaseous emission, greenhouse gases (GHG), 38 odour. 2 39 1. Introduction 40 The large amount of sewage sludge (SS) production is becoming a worldwide 41 environmental problem. Nowadays, different treatment processes are used to manage SS 42 produced, such as land application, thermal processing, composting or anaerobic 43 digestion, aiming at reducing its volume and quantity as well as stabilizing it. In fact, 44 composting, which is one of the most applied SS stabilization approaches, is considered 45 as an environmental friendly technology. Nevertheless, there are some limitations when 46 applying SS or the compost obtained from SS on soil due to the possible presence of 47 heavy metals or other emergent contaminants (European Commission, 2014). Regarding 48 that, and taking into account the limited land resources, thermal processes have risen as 49 an attractive sludge treatment option in many countries. However, due to the high water 50 content present in SS, a pre-drying step is required prior to incineration. Between the 51 existing alternatives for SS drying, biodrying is catching attention of researchers in the 52 last years (Huiliñir and Villegas, 2015; Villegas and Huiliñir, 2014; Zhao et al., 2011). 53 The objective of SS biodrying process, which is based on the mechanisms of a 54 composting process, is to remove water from biowaste with high moisture content and 55 to reduce the volume and the biological activity of the solid matrix to obtain a product 56 with a considerable Lower Calorific Value (LCV) and low moisture content that can be 57 used as a biomass fuel (Adani et al., 2002). To this aim, both the metabolic heat 58 generated during the aerobic degradation of the easily biodegradable organic matter as 59 well as forced aeration are used (Frei et al., 2004). In this sense, and compared with 60 composting, higher aeration rates are used during biodrying to enhance water removal 61 (Cai et al., 2013; Yuan et al., 2016). 62 Different studies have been conducted to find operating strategies to optimize the SS 63 biodrying process in terms of aeration and initial moisture content, as they are the key 3 64 parameters to achieve an optimum SS biodrying performance. For example, Cai et al. 65 (2013) studied the influence of aeration on water removal reporting higher evaporation 66 rates at higher aeration rates during the thermophilic phase of the SS biodrying process. 67 Moreover, Huiliñir and Villegas (2015) worked on the simultaneous effect of the 68 aeration and the initial moisture of the material on the water evaporation. They reported 69 that higher material temperature is achieved when initial moisture content is around 70 68% and that, under these conditions, higher aeration rates enhance water evaporation. 71 However, critical aspects such as gaseous emissions and its associated environmental 72 impacts have not been systematically studied yet as it has been done with SS 73 composting. Greenhouse gases (GHG) and odorant compounds characterization have 74 been widely reported for SS composting at bench-scale (Maulini-Duran et al., 2013; 75 Rincón et al., 2019), whereas few studies have been conducted in full-scale facilities 76 (González et al., 2019a; Shen et al., 2012). Nonetheless, few information about gaseous 77 emissions from bench-scale SS biodrying can be found in literature (Kim et al., 2018; 78 Wei et al., 2016), but with no concluding or profitable information in terms of 79 environmental impact of the technology. In this sense, gaseous emissions assessment 80 studies must be carried out to fill existing knowledge gaps and to help improving the 81 performance of such biological treatment technology. Emission factors referred to 82 specific pollutants are a useful tool for environmental impact assessment and facilitate 83 the estimation of the overall emission rate of a treatment process based on a specific 84 activity index, which should represent the process evaluated (Capelli et al., 2009). The 85 US Environmental Protection Agency defined the emission factor of a specific pollutant 86 as the relation between the quantity of the pollutant emitted to the atmosphere with an 87 activity associated with the release of the pollutant, such as the mass of waste to be 88 treated, the emitting surface or time units (United States Environmental Protection 4 89 Agency (USEPA), 1995). Analogously, in the last years this tool has been referenced 90 for the assessment of odour impact from different treatment processes and facilities, 91 namely Odour Emission Factor (OEF) (Capelli et al., 2014; Sironi et al., 2006). 92 To the authors’ knowledge, this is the first bench-scale work where a systematic 93 study of the gaseous emissions generated during the biodrying process of conventional 94 SS is conducted. Due to the lack of this scientific information and its importance for 95 further development of the biodrying technology, the aim of this work was to study the 96 SS biodrying process performance and, specially, to acquire knowledge about the GHGs 97 and odorant gaseous emission patterns related to this treatment. Emission factors for 98 specific pollutants are provided, as well as a screening on the diversity of VOC families 99 found in the gaseous emissions from the biodrying of SS, which can be valuable 100 information in further environmental assessment studies. 101 2. Materials and methods 102 2.1. Characteristics of feedstock 103 SS was obtained from the wastewater treatment plant of Navàs, Barcelona, which 104 treats 1500 m3·d-1 of domestic wastewater. Diatomaceous earth (DE), which was used 105 as a high-energy content co-substrate to enhance metabolic activity and to adjust 106 moisture content (Johnson, 1997), was obtained from the organic fraction of Municipal 107 Solid Waste (MSW) biomethanization plant of Can Barba in Terrassa, Barcelona. 108 Finally, pruning waste (PW), used as bulking agent, was obtained from the composting 109 plant of Manresa, Barcelona. The mixture was prepared manually with an industrial 110 mixer, and the used mixing ratio of the three feedstocks (SS:DE:PW) was 1:0.25:3 on a 111 volume basis and 1:0.15:0.48 on weight basis, which is a typical mixing ratio used 112 during SS composting (González et al., 2019a; Maulini-Duran et al., 2013). As 5 113 aforementioned, the addition of a 10-15% of DE (weight basis) has been reported to 114 enhance the metabolic activity of the mixture in different treatment processes (Pognani 115 et al., 2012). Table 1 shows the physical-chemical characteristics of each material and 116 mixture. 117 2.2. Experimental set-up and process operation 118 Figure 1 shows a scheme of the biodrying pilot plant. A cylindrical bioreactor with 119 an operative volume of 100 L was used to carry out the SS biodrying process. The 120 reactor was filled with 41.4 kg of the mixture with a density of 414 kg·m-3. The 121 dimensions of reactor were 0.85 m high and had a diameter of 0.50 m. The internal and 122 external walls of the reactor were made of stainless steel. A thermal isolation layer of 123 polyurethane foam (2 cm) was used to avoid heat losses and to maintain near-to- 124 adiabatic conditions. A perforated plate was fixed at the bottom of the reactor to support 125 the material and to facilitate aeration, working as a diffusor. The biodrying reactor was 126 opened but covered with a straw cover to prevent heat loss and vapour condensation on 127 the mixture. Temperature of the material was measured with a temperature probe (PT- 128 100, Iserntech S.A., Spain) introduced into the reactor, with sensors located at the top, 129 middle and bottom of each one. An air compressor (Dixair DNX 2050, Worthington 130 Creyssensac, Spain) and digital mass flow controllers (D-6311-DR, Bronkhorst High- 131 Tech B. V., Ruurlo, Netherlands) were used for continuous aeration. During biodrying, 132 the aeration flowrate was regulated based on the temperature of the material using a 133 feedback control loop. The control software adapted the aeration level depending on 134 different temperature ranges referred to the temperature in the middle of the reactor 135 (<35 ºC; 35-45 ºC; 45-55 ºC; 55-70 ºC; >70 ºC), increasing the airflow rate when the 136 material temperature increased (2, 6, 10, 14 and 20 L·min-1, corresponding to 0.12, 0.36, 137 0.60, 0.84 and 1.20 L·min-1·kg-1 of initial total VS of mixture). In this case, the middle 6 138 temperature was used to regulate airflow because both the bottom and the top 139 temperatures of the reactor can be influenced by the temperature of the inlet airflow or 140 due to the loss of volume of the material along the biodrying process, respectively. To 141 follow the evolution of the process in terms of biodegradation, the oxygen content of the 142 outlet gas of the reactor was monitored using an O2 sensor (O2 A1, Alphasense, UK) 143 located after a condensation trap to prevent sensors damage. The biodrying process 144 lasted 13 days and the material was mixed manually every day during the thermophilic 145 phase and every two days during the mesophilic stage. Water and VS loss during the 146 biodrying process was monitored through weight loss using a scale (Gram Precision k3- 147 k3i, Gram Group, Spain). Arduino UNO was used for data acquisition and LabView 148 2017® was used for data analysis and airflow control. 149 2.3. Gas sampling and analysis methodologies 150 A semi-spherical stainless steel flux chamber (0.443 m of base diameter, 0.154 m2 of 151 base area and 0.045 m3 of volume) provided by Scentroid (IDES Canada Inc., 152 Whitchurch-Stouffville, ON, Canada) was used daily to perform emissions sampling 153 (González et al., 2019b). Nalophan® bags were used to store gas samples, which were 154 flushed twice before obtaining the final gaseous sample for analysis. All gas samples 155 were always obtained before mixing of the material. 156 CH4 and N2O analysis were performed using an Agilent 6890N Gas Chromatograph 157 (GC) (Agilent Technologies, Inc., Santa Clara, CA, USA), equipped with a flame 158 ionization detector (FID) and an electron capture detector (ECD) for CH4 and N2O 159 detection, respectively. The column used for the analysis was a HP-PLOT Q semi- 160 capillary column (30 m x 0.53 mm x 40.0 µm, Agilent Technologies, Inc.) with N2 as 161 carrier gas at 2 psi pressure, and a post-column particle trap (2 m, nº 5181-3352, Agilent 162 Technologies, Inc.). For CH4 analysis, the injector temperature was 240 ºC, the detector 7 163 temperature was 250 ºC and the oven, which worked isothermally, was set at 60 ºC. For 164 N2O analysis, the injector temperature was 120 ºC, the detector temperature was 345 ºC 165 and the oven, which worked isothermally, was set at 60 ºC. The injection volume used 166 for each sample was 500 µL and the total time of analysis was 4 and 6 minutes for CH4 167 and N2O, respectively. 168 Odour concentration analysis were performed using a SM-100 portable field 169 olfactometer provided by Scentroid (IDES Canada, Inc., Whitchurch-Stouffville, ON, 170 Canada), as explained in González et al. (2019b). All gaseous samples were analysed 171 three times by the same panellist in a closed, well-ventilated room the same day of 172 sampling. Total volatile organic compounds (tVOCs), NH3 and H2S concentration of the 173 outlet gases of the reactor were measured daily in situ with a MultiRAE Lite portable 174 analyser (RAE Systems, San José, CA, USA), equipped with a 10.6 eV PID lamp for 175 tVOCs measurement and two electrochemical sensors for NH3 and H2S measurement, 176 respectively. tVOCs detection ranged from 0 to 1000 ppmveq isobutylene with 1 ppmveq 177 isobutylene increments, NH3 detection ranged from 0 to 100 ppmv with 1 ppmv 178 increments and H2S detection ranged from 0 to 100 ppmv with 1 ppmv increments. 179 During each analysis, the portable analyser was placed inside a closed recipient with 180 inlet and outlet ports and the outlet gases of the reactor were passed through it until a 181 steady value was read on the analyser. Since the biodrying process was held in an open 182 reactor, a suction pump was used to redirect part of the outlet gases from the surface of 183 the material to the hermetic recipient where the portable analyser was placed. The gases 184 were measured just before the water trap installed to protect the rest of the measurement 185 devices from the gaseous flow moisture. 186 To perform the VOCs characterization, two different gaseous samples of 1 L each 187 were pumped through sorbent tubes packed with two different hydrophobic sorbents 8 188 (Tenax® TA and Carbograph™ 1TD, Markes International, Inc., Gold River, CA, 189 USA) by means of a manual sampling pump (Markes International, Inc.). The gaseous 190 samples were obtained the 3rd and the 12th day of running, coincident with the 191 thermophilic and the mesophilic phases of the biodrying process. First, a UNITY-2 192 Thermal Desorber (TD) (Markes International, Inc.) was used to desorb the gaseous 193 samples retained in the sorbent tubes. Then, an Agilent 7820 GC coupled to an Agilent 194 5975 Mass Spectrometer (MS) (Agilent Technologies, Inc.) was used to analyse and 195 characterize the different VOCs present in the gas samples (González et al., 2019b). 196 Both samples were analysed the same day that they were obtained in order to preserve 197 the stability of VOCs in sorbent tubes (Ribes et al., 2007). 198 2.4. Estimation of the gaseous emission factors 199 The emission factor was calculated based on the measured concentration of a specific 200 pollutant, the aeration flow through the reactor at time of each sampling event and 201 normalised by the initial mass of dry matter in the SS (DM0-SS) introduced in the 202 reactor. Firstly, daily emission rates of each pollutant were calculated (Eq. 1). 203 𝐸𝑅𝑖 = 𝐶𝑖 · 𝐹 Eq. (1) 204 where ERi is the emission rate of the target pollutant (mg pollutant·d-1); Ci is the 205 concentration of the target pollutant (mg pollutant·m-3); F is the aeration flow through 206 the reactor during sampling (m3·d-1). 207 Then, the emission rates of a target pollutant obtained for each sampling day were 208 represented versus process time, thus the area below the curve obtained corresponds to 209 the emitted total mass of the pollutant throughout the process studied. Finally, the 210 emitted total mass of a target pollutant was divided by the initial mass of DM0-SS 211 introduced in the reactor to obtain the emission factor (Eq. 2). 9 𝑚𝑖 212 𝐸𝐹𝑖 = Eq. (2) 𝐼 213 where EFi is the emission factor of the target pollutant (mg pollutant·kg-1 DM0-SS); 214 mi is the total emitted mass of the target pollutant during the process (mg pollutant); I is 215 the initial mass of DM0-SS treated as specific activity index (kg DM0-SS). 216 2.5. Analytical methods for solid samples 217 Moisture content, dry matter, organic matter content, N-Kjeldhal, C/N ratio, pH and 218 electrical conductivity were determined according to the standard procedures (US Dept. 219 of Agriculture and US Composting Council, 2001; Water Environment Federation, 220 1995). To evaluate the final biological stability of the treated material, the Dynamic 221 Respiration Index in the 24 hours of maximum biological activity (DRI24h) and the 222 cumulative oxygen consumption in 4 days (AT4) were used. These indices provide 223 information about the biological stability of an organic solid sample, and are expressed 224 in g O2·h-1·kg-1 VS and g O2·kg-1 VS, respectively. The determination of the DRI24h and 225 the AT4 was performed using a dynamic respirometer (Ponsá et al., 2010) as explained 226 in González et al. (2019a). 227 The determination of the Higher Calorific Value (HCV) was done by means of a 228 bomb calorimeter (1341 Plain Jacket Calorimeter – 1108 Oxygen Combustion Vessel, 229 Parr, IL, USA), according to manufacturer’s instructions and following the 230 methodology proposed at the DIN 51900-3:2005 (German Institute for Standarisation, 231 2003). The Lower Calorific Value (LCV) was calculated from HCV according to Eq. 3, 232 taking into account the moisture content of the final product and its estimated hydrogen 233 content (Koppejan and van Loo, 2008). 𝑀𝐶 𝑀𝐶 𝐻 𝑀𝐶 234 𝐿𝐶𝑉 = 𝐻𝐶𝑉 · (1 − 100) − (2.444 · 100) − 2.444 · 100 · 8.936 · (1 − 100) Eq. (3) 10 235 where LCV is the Lower Calorific Value of the biomass fuel (MJ·kg-1 product); 236 HCV is the Higher Calorific Value of the biomass fuel (MJ·kg-1 DM); MC is the 237 moisture content of the biomass fuel (%); H is the estimated hydrogen content of the 238 biomass fuel (%), which was considered to be a 5 % (Choi et al., 2014); 2.444 is the 239 enthalpy difference between liquid water at 25 ºC and gaseous phase; 8.936 is the 240 molecular weight ratio between both molecules (mH2O·m-1H2). 241 3. Results and discussion 242 3.1. Biodrying performance 243 Figure 2a shows the performance of the biodrying reactor while a summary of the 244 process parameters evaluated during the treatment is presented in Table 2. In Figure 2a, 245 the evolution of the temperature in the middle of the solid matrix, the specific airflow, 246 the decrease of moisture content and the mixing pattern are represented for the whole 247 process time. As expected, two temperature stages were encountered, a thermophilic 248 stage (> 41 ºC) and a mesophilic stage (≤ 41 ºC). Temperature increased rapidly during 249 the first 48 hours to the maximum values registered (51.0 ºC). 250 The key parameters to explain the biodrying process are temperature, specific airflow 251 and moisture content of the solid material, which are shown in Figure 2a for the whole 252 process time. Similar to a typical SS composting process, material temperature reached 253 its peak on day 2. Then, temperature was maintained around 40 ºC until day 7. 254 Thereafter, the temperature progressively decreased to a final value of 30 ºC, as 255 previously observed in this kind of biological treatment (Huiliñir and Villegas, 2015). 256 Specific airflow, referred to the total VS in the initial mixture (VS0) or to the initial VS 257 in the SS (VSSS), depended on the temperature in the middle of the biodrying reactor. Its 258 average value during the whole process was 0.46 L·min-1·kg-1 VS0 (2.54 L·min-1·kg-1 11 259 VSSS), which is a typical specific airflow for SS biodrying (Cai et al., 2013; Huiliñir and 260 Villegas, 2015) and higher than the specific airflows reported during conventional SS 261 composting (Maulini-Duran et al., 2013; Yuan et al., 2016). These high aeration rates 262 used in biodrying are necessary to supply oxygen to microorganisms but also to enhance 263 water evaporation. The maximum specific airflow registered during the biodrying 264 process (0.84 L·min-1·kg-1 VS0; 4.63 L·min-1·kg-1 VSSS), coincides with the peak of 265 temperature found in order to maximise water evaporation during the period of 266 maximum biological activity. In contrast, the minimum specific airflow registered (0.12 267 L·min-1·kg-1 VS0; 0.66 L·min-1·kg-1 VSSS) was found during the first day of biodrying, 268 which helped to rapidly reach thermophilic conditions inside the reactor. The biodrying 269 process was stopped on day 13, when a 34.3% reduction of the moisture content of the 270 material was achieved, obtaining a final material with a 35.9% of moisture content, even 271 lower to what is reported in other studies (Cai et al., 2013; Hao et al., 2018). The 272 biodrying process time used is between typical reported values (Huiliñir and Villegas, 273 2015; Tambone et al., 2011) and, even though moisture content could have been 274 reduced even more, its extension will imply higher aeration costs, especially at full- 275 scale. The final DRI24h of the biodried material was 3.5 ± 0.1 g O2·h-1·kg-1 VS, 276 corresponding to a 44.4% reduction of the initial DRI24h. Despite the DRI24h reduction 277 achieved, it is worth mentioning that the main goal of SS biodrying is not to stabilize 278 the organic waste, but to reduce its moisture content in a short-term process to obtain a 279 product with a LCV enough to be used as a biomass fuel. Moreover, no related literature 280 presenting the biological stability of the final biodried material has been found for SS 281 biodrying processes, as it happens with other wastes such as MSW (Evangelou et al., 282 2016). In contrast, conventional SS composting seeks reducing the biological activity of 283 the material but applying longer process times, obtaining final DRIs24h lower than 1 g 12 284 O2·h-1·kg-1 VS (Maulini-Duran et al., 2013). Finally, the LCV of the final material was 285 analysed and was found to be 7.1 MJ·kg-1 product –increasing a 65.1% compared to the 286 initial LCV–, which is known to be sufficient for self-sustaining combustion (Hao et al., 287 2018). 288 3.2. Greenhouse gases emissions 289 GHGs emissions from SS composting processes have been widely reported in the 290 past years due to their potential impact to global warming (Pan et al., 2018; Yuan et al., 291 2016). However, an information gap exists about GHGs emission generated during the 292 biodrying of SS. It has to be mentioned that during this study, only GHGs emissions 293 referred to the biological process have been accounted, and CO2 has not been taken into 294 account due to its biogenic origin. Figure 2b shows the daily emission rates of CH4 and 295 N2O from the biodrying processes, altogether with the temperature profile of the 296 material and the specific airflow supplied to the reactor. At first sight, two different 297 emission patterns were found for CH4 and N2O. In general, the CH4 emission pattern 298 followed the temperature profile of the material, presenting its emission peak when 299 maximum temperature and maximum aeration occurred. On the other hand, N2O 300 showed a high emission peak the first sampling day due to the stripping effect of the 301 remaining N2O present in the SS, which rapidly decrease to lower values. However, an 302 increase of the N2O emission was observed when the temperature of the material was 303 below 40 ºC, suggesting that an increase in the SS biodrying process time would have a 304 negative effect in terms of GHG emissions. 305 Generally, CH4 is formed by methanogens under anaerobic conditions. Anaerobic 306 spots can be created around the reactor by the oxygen demand at this temperature range 307 due to the rapid degradation of organic matter during thermophilic phase (Yuan et al., 308 2016). As observed in Figure 2b, the peak of CH4 emission coincided with that of 13 309 temperature, and then CH4 emission decreased as temperature decreased into the 310 mesophilic range. Maximum and minimum CH4 concentration measured during the SS 311 biodrying process were 8.16 ppmv and 1.07 ppmv, respectively. 312 The formation of N2O is complex, which can be produced by denitrification when 313 NO3- is converted to N2O and nitrogen gas, and by incomplete nitrification during the 314 conversion of NH3 to NO2- (Moënne-Loccoz and Fee, 2010). The N2O emission rates 315 for the SS biodrying process are presented in Figure 2b. During the first sampling day, a 316 peak of N2O emission appeared, mainly because of the stripping effect of remaining 317 N2O present in the SS used for the experiment. Once this first N2O was stripped out of 318 the SS, there was no sign of N2O production until the temperature of the material 319 decreased to a mesophilic range. At that point, an increase of the N2O emission 320 appeared until the end of the process. Results suggest that the temperature of the 321 material below the high-end of the mesophilic range (40ºC), together with the low 322 carbon availability at the latter stages of the process, affected the microbial activity and, 323 concomitantly, the formation and stripping of N2O. Under these conditions some 324 denitrifying bacteria can promote the N2O formation by NO3- denitrification at the end 325 of the process (Ahn et al., 2011; Fukumoto et al., 2003). This fact should be confirmed 326 by characterizing the activity and the diversity of the bacterial communities in future 327 works. Moreover, and despite the forced aeration and the fact that the material was 328 mixed every day, different anoxic spots could had been formed inside the reactor and 329 promote the formation of N2O by denitrification (Awasthi et al., 2018). Maximum and 330 minimum N2O concentrations measured during the SS biodrying process were 18.75 331 ppmv and 0.37 ppmv, respectively. 332 Table 3 shows the emission factor of each GHG in the SS biodrying process as kg of 333 CH4 or N2O per Mg DM0-SS and per Mg DM0 of mixture and in kg of CO2eq per Mg 14 334 DM0-SS and per Mg DM0 of mixture, respectively. These emission factors are valuable 335 information to help comparing different treatment strategies and evaluating its 336 environmental impact. Besides, it is important to highlight the much higher 337 environmental impact in terms of Global Warming Potential of N2O (265 times higher 338 than that of CO2) compared with CH4 (28 times higher than that of CO2) (IPCC, 2014). 339 To contextualize the results obtained, the GHG emission factors have been contrasted 340 with GHG emission factors obtained from SS composting, because, to the authors’ 341 knowledge, no previous GHG emission factors from SS biodrying have been reported. 342 For example, Yuan et al. (2016) reported GHG emission factors depending on the 343 aeration rate in the range of 170.43 – 272.52 kg CO2eq·Mg-1 DM0-SS during the 344 composting of dewatered non-digested sewage sludge with cornstalks as bulking agent 345 in a 60 L aerated bench-scale reactor. On the other hand, Maulini-Duran et al. (2013) 346 found lower GHG emission factors referred to the SS composting in a 50 L aerated 347 bench-scale reactor, using wood chips as bulking agent, ranging from 4.98 to 9.90 kg 348 CO2eq·Mg-1 DM0-SS. Pan et al. (2018) observed GHG emission factors ranging from 349 36.05 to 134.56 kg CO2eq·Mg-1 DM0 of mixture while composting dewatered SS with 350 wheat straw and different acidic additives in a 150 L bench-scale composting reactor. 351 Moreover, the GHG emission factor related to SS biodrying process was lower than 352 those reported for conventional SS drying systems, which are in a higher range between 353 367 to 21900 kg CO2eq·Mg-1 DM0-SS due to the use of fossil fuels and chemicals (Chen 354 and Kuo, 2016). In comparison with the experimental emission factors reported for SS 355 composting processes, the SS biodrying treatment seems a feasible technology in terms 356 of GHG emission impact. However, future studies are warranted to minimise N2O 357 emissions by understanding the biological pathways for N2O generation during the SS 358 degradation and the effect of the aeration rate on its emission in order to decrease the 15 359 total GHGs emissions related to the SS biodrying process. Ensuring aerobic conditions 360 in the SS biodrying reactor is a key factor to minimize GHG emissions. As reported by 361 different authors studying SS composting processes, low aeration and inadequate initial 362 moisture content have a great impact on increasing GHG emission (Villegas and 363 Huiliñir, 2014; Yuan et al., 2016). Results reported herein are a good starting point for 364 future environmental impact assessment of the SS biodrying technology, which can be 365 used for Life Cycle Assessment studies. 366 3.3. Odorant compounds emissions 367 3.3.1. NH3, H2S, tVOCs and odour emissions 368 Emissions of NH3, H2S, tVOCs and odour from the SS biodrying process were 369 monitored during the whole process to fill the existing knowledge gap about gaseous 370 emissions associated to this SS biological treatment. Table 3 presents the emission 371 factors obtained as kg of pollutant per Mg DM0-SS and per Mg DM0 of mixture, 372 respectively, and in ou per Mg DM0-SS and per Mg DM0 of mixture, respectively, when 373 monitoring the biodrying processes. NH3, H2S, tVOCs are the typical precursors of 374 odour nuisance generated during the biological treatment of organic wastes such as SS. 375 NH3 emissions during aerobic biodegradation of organic wastes have been widely 376 reported (Colón et al., 2012; He et al., 2017), and it is known to occur at their maximum 377 rate during the thermophilic phase of the biological process, since high temperatures 378 favour NH3 volatilization by displacing the NH4+/NH3 equilibrium to NH3. H2S during 379 SS biological treatment also occurs during the first stage of the process and it is 380 enhanced by high temperatures. Its formation results from bacterial reduction of 381 sulphate and decomposition of sulphur-containing compounds present in the SS under 382 anaerobic conditions (Chen et al., 2010). VOCs have been also reported to be released 383 at their highest production rate during the initial forced aeration of the biological 16 384 treatment of SS. In this case, NH3, H2S and tVOCs maximum concentrations measured 385 during SS biodrying were 330 ppmv of NH3, 1 ppmv of H2S and 18 ppmveq isobutylene, 386 respectively, which were all observed during the peak of maximum temperature of the 387 material in the reactor. In SS composting processes, different ranges of target pollutants 388 concentration can be found in literature depending on the characteristics of the 389 feedstocks or the composting process itself. Hort et al. (2009) reported NH3 390 concentrations above 70 ppmv during the bench-scale SS and yard waste composting, 391 but no H2S was reported due to a proper aeration of the material during the first steps of 392 the biological process. Besides, Pagans et al. (2006) reported maximum concentrations 393 of 132 ppmv of NH3 and 260 ppmv of C-tVOCs while composting raw sludge in a 394 forced-aerated 30 L lab-scale reactor. On the other hand, González et al. (2019a) 395 measured NH3 and tVOCs of 250 ppmv and 18 ppmv, respectively, during the SS 396 composting in a full-scale composting plant. 397 To put the obtained results into context, the emission factors of the SS biodrying 398 process have been compared with other reported emission factors from different 399 composting processes, since no information has been found about other biodrying 400 processes. Maulini-Duran et al. (2013) reported NH3 and tVOCs emission factors in the 401 range of 1.33 to 4.12 kg NH3·Mg-1 DM0-SS and 0.66 to 1.00 kg C-VOC·Mg-1 DM0-SS, 402 respectively, during composting of raw sludge in a 50 L aerated bench-scale reactor. 403 Rincón et al. (2019) studied the gaseous emissions from the SS composting in a 300 L 404 pilot-scale reactor, reporting an NH3 emission factor of 2.16 kg NH3·Mg-1 DM0 of SS 405 and a tVOCs emission factor of 1.21 kg C-VOC·Mg-1 DM0-SS. González et al. (2019a) 406 studied the emission of NH3 and tVOCs in a full-scale sewage sludge composting plant, 407 reporting emission factors of 8.53 kg NH3·Mg-1 DM0 of mixture and 2.09 kg C- 408 VOC·Mg-1 DM0 of mixture. In addition, Gomez-Rico et al. (2008) and Deviatkin et al. 17 409 (2019) reported higher tVOCs and NH3 emission factors (0.59 kg C-VOC·Mg-1 DM0- 410 SS and 1.03 to 6.00 kg NH3·Mg-1 DM0-SS, respectively) during different conventional 411 drying of SS at 100-120ºC at lab-scale. The variability between the different studies 412 shows that, even by treating the same type of organic waste, the substrates nature and 413 process conditions (aeration level, initial moisture, scale, etc.) affect the emission of the 414 odorant pollutants. However, results obtained in the present work represent a valuable 415 insight that will allow a better understanding of the environmental impact related to this 416 new technology. 417 Figure 2c shows the daily odour emission rates (OER) obtained during the SS 418 biodrying process. Similarly to other works where odour emission from SS composting 419 were studied (González et al., 2019b; Rincón et al., 2019) the odour emission profile 420 reached its maximum during the thermophilic phase, when biological activity was 421 higher. Temperature is one of the key factors affecting the emission of odorant 422 compounds during SS biodrying. However, reaching thermophilic conditions during the 423 process, as well as in SS composting, is inevitable and needed to obtain a proper drying 424 of the material. Maximum odour concentration measured during SS biodrying was 3043 425 ou·m-3. As the process proceeded, odour emissions decreased progressively down to 426 low odour concentrations between 275 and 290 ou·m-3. In this sense, few works dealing 427 with the OEFs of SS biological treatment have been found. Rincón et al. (2019) 428 evaluated the OEFs of different substrates in a 300 L aerated pilot-scale reactor, 429 reporting an OEF of 9.35E+08 ou·Mg-1 DM0-SS and González et al. (2019b) obtained 430 an OEF of 9.45E+07 ou·Mg-1 DM0-SS in a full-scale sewage sludge composting plant. 431 Nevertheless, some works have been conducted over different treatment facilities to 432 estimate their OEFs. Sironi et al. (2006) and Zarra et al. (2016) reported OEFs of 433 1.01E+08 ou·Mg-1 of MSW and 4.15E+06 ou·Mg-1 of MSW, respectively, from 18 434 different MSW mechanical-biological treatment plants. The obtained results in terms of 435 DM0 of biodryied mixture show that OEF referred to the SS biodrying process are 436 between one and two orders of magnitude below the OEFs found in the few literature 437 mentioned, what represents lower odour impact from this kind of technology. Even 438 though these results are valuable information that can help to estimate odour emissions 439 from this biological process and the odour impact derived from similar process, they 440 must be carefully used when extrapolating into larger SS biodrying treatment systems 441 due to its dependence on different aspects such as feedstock characteristics, operational 442 parameters or plant capacity. 443 In summary, the main advantages that can be highlighted from SS biodrying over SS 444 composting are that SS biodrying is a faster biological treatment that requires less space 445 to treat the same flux of material, that energy recovery is produced and that lower 446 gaseous emissions are generated. However, it can be a more expensive alternative than 447 SS composting due to the high aeration requirements. 448 3.3.2. VOCs characterization 449 A chemical characterization was made by comparing the VOC families found in the 450 different samples as well as by quantifying the concentration of 17 out of the 35 451 detected compounds. Figure 3 presents the distribution of the different VOC families 452 found in the samples of two different days obtained during the SS biodrying process. 453 Terpenes –with α-pinene as predominant– are the main VOC family found in the 454 gaseous samples analysed, with relative abundances between 70-80%. Furthermore, 455 other odorant species such as sulphur compounds –dimethyl sulphide (DMS), dimethyl 456 disulphide (DMDS) and dimethyl trisulphide (DMTS)–, ketones –mainly 2-butanone 457 and 2-nonanone– and some alkanes and alkenes such as pentane or pentene appeared in 458 different percentages. Generally, terpenes such as α-pinene, β-pinene or limonene may 19 459 originate as intermediate products of aerobic biodegradation of organic matter. Besides, 460 terpenes are also related to the PW used as bulking agent for the biodrying process 461 (González et al., 2019b; Schiavon et al., 2017). In the present work, its abundance was 462 that high probably due to the high SS:PW v/v ratio used. On the other hand, sulphur 463 compounds are typical chemicals present in the gaseous emissions generated during the 464 SS treatment, which are normally formed and emitted during the first stages of the SS 465 biological treatment (He et al., 2017; Schiavon et al., 2017). In the work presented 466 herein, a 13% abundance of organosulphur compounds was found in the first sample, 467 which coincide with the thermophilic phase of the biological process. Then, when the 468 process fell into mesophilic temperature, the organosulphur compounds abundance 469 decreased to 3%. Finally, some other odorant compounds typically found in the gaseous 470 emissions generated during the biological treatment of SS were found, such as ketones, 471 alkanes or alkenes. Ketones were found at maximum abundance during the first stage of 472 the SS biodrying (6.6% abundance), which was then reduced down to a 0.5% 473 abundance during the mesophilic phase. 474 4. Conclusions 475 Results indicated a proper performance of the SS biodrying process. The 476 corresponding GHG and odorant emissions were accounted to be lower than that 477 reported for conventional SS composting. GHG emission factor and OEF referred to the 478 bench-scale SS biodrying process were 28.22 kg CO2eq·Mg-1 DM0-SS and 3.10E+07 479 ou·Mg-1 DM0-SS, respectively. In conclusion, since there is no available scientific 480 information dealing with the gaseous emissions from SS biodrying to compare with, this 481 study shows a new insight into the gaseous emissions generated during the SS biodrying 482 process at bench-scale, which will serve in further development of this treatment. 483 E-supplementary data for this work can be found in e-version of this paper online. 20 484 References 485 1. Adani, F., Baido, D., Calcaterra, E., Genevini, P., 2002. The influence of biomass 486 temperature on biostabilization-biodrying of municipal solid waste. Bioresour. 487 Technol. 83, 173–179. doi:10.1016/S0960-8524(01)00231-0 488 2. Ahn, H.K., Mulbry, W., White, J.W., Kondrad, S.L., 2011. Pile mixing increases 489 greenhouse gas emissions during composting of dairy manure. Bioresour. Technol. 490 102, 2904–2909. 491 3. Awasthi, M.K., Wang, Q., Chen, H., Wang, M., Awasthi, S.K., Ren, X., Cai, H., Li, 492 R., Zhang, Z., 2018. In-vessel co-composting of biosolid: Focusing on mitigation 493 of greenhouse gases emissions and nutrients conservation. Renew. Energy 129, 494 814–823. 495 4. Cai, L., Chen, T., Gao, D., Zheng, G., Liu, H., 2013. Influence of forced air volume 496 on water evaporation during sewage sludge bio-drying. Water Res. 47, 4767–4773. 497 5. Capelli, L., Sironi, S., Del Rosso, R., Céntola, P., 2009. Predicting odour emissions 498 from wastewater treatment plants by means of odour emission factors. Water Res. 499 43, 1977–1985. 500 6. Capelli, L., Sironi, S., Rosso, R. Del, 2014. Odour Emission Factors: Fundamental 501 Tools for Air Quality Management. Chem. Eng. Trans. 40. 502 7. Chen, J., Liu, H.-T., Lei, M., Chen, T.-B., Zheng, G.-D., Gao, D., Guo, S.-L., Cai, L., 503 2010. Reducing H2S production by O2 feedback control during large-scale sewage 504 sludge composting. Waste Manag. 31, 65–70. 505 8. Chen, Y.C., Kuo, J., 2016. Potential of greenhouse gas emissions from sewage sludge 506 management: A case study of Taiwan. J. Clean. Prod. 129, 196–201. 21 507 9. Choi, H.L., Sudiarto, S.I.A., Renggaman, A., 2014. Prediction of livestock manure 508 and mixture higher heating value based on fundamental analysis. Fuel 116, 772– 509 780. 510 10. Colón, J., Cadena, E., Pognani, M., Barrena, R., Sánchez, A., Font, X., Artola, A., 511 2012. Determination of the energy and environmental burdens associated with the 512 biological treatment of source-separated Municipal Solid Wastes. Energy Environ. 513 Sci. 5, 5731–5741. 514 11. Cometto-Muñiz, J.E., Cain, W., Abraham, M., Kumarsingh, R., 1998. Sensory 515 properties of selected terpenes. Thresholds for odor, nasal pungency, nasal 516 localization, and eye irritation. Ann. N. Y. Acad. Sci. 855, 648–651. 517 12. Deviatkin, I., Lyu, L., Chen, S., Havukainen, J., Wang, F., Horttanainen, M., 518 Mänttäri, M., 2019. Technical implications and global warming potential of 519 recovering nitrogen released during continuous thermal drying of sewage sludge. 520 Waste Manag. 90, 132–140. 521 13. European Commission, 2014. http://ec.europa.eu/environment/waste/sludge/ 522 [WWW Document]. 523 14. Evangelou, A., Gerassimidou, S., Mavrakis, N., Komilis, D., 2016. Monitoring the 524 performances of a real scale municipal solid waste composting and a biodrying 525 facility using respiration activity indices. Environ. Monit. Assess. 188. 526 15. Frei, K.M., Cameron, D., Stuart, P.R., 2004. Novel drying process using forced 527 aeration through a porous biomass matrix. Dry. Technol. 22, 1191–1215. 528 16. Fukumoto, Y., Osada, T., Hanajima, D., Haga, K., 2003. Patterns and quantities of 529 NH3, N2O and CH4 emissions during swine manure composting without forced 22 530 aeration - Effect of compost pile scale. Bioresour. Technol. 89, 109–114. 531 17. German Institute for Standarisation, 2003. DIN 51900-3:2005. Testing of solid and 532 liquid fuels - Determination of gross calorific value by the bomb calorimeter and 533 calculation of net calorific value - Part 3: Method using the adiabatic jacket. 534 18. Gomez-Rico, M.F., Fullana, A., Font, R., 2008. Volatile organic compounds 535 released from thermal drying of sewage sludge. WIT Trans. Ecol. Environ. 111, 536 425–433. 537 19. González, D., Colón, J., Gabriel, D., Sánchez, A., 2019a. The effect of the 538 composting time on the gaseous emissions and the compost stability in a full-scale 539 sewage sludge composting plant. Sci. Total Environ. 654, 311–323. 540 20. González, D., Colón, J., Sánchez, A., Gabriel, D., 2019b. A systematic study on the 541 VOCs characterization and odour emissions in a full-scale sewage sludge 542 composting plant. J. Hazard. Mater. 373, 733–740. 543 21. Hao, Z., Yang, B., Jahng, D., 2018. Combustion characteristics of biodried sewage 544 sludge. Waste Manag. 72, 296–305. 545 22. He, P., Wei, S., Shao, L., Lü, F., 2017. Emission potential of volatile sulfur 546 compounds (VSCs) and ammonia from sludge compost with different bio-stability 547 under various oxygen levels. Waste Manag. 73, 113–122. 548 23. Hort, C., Gracy, S., Platel, V., Moynault, L., 2009. Evaluation of sewage sludge and 549 yard waste compost as a biofilter media for the removal of ammonia and volatile 550 organic sulfur compounds (VOSCs). Chem. Eng. J. 152, 44–53. 551 24. Huiliñir, C., Villegas, M., 2015. Simultaneous effect of initial moisture content and 552 airflow rate on biodrying of sewage sludge. Water Res. 82, 118–128. 23 553 25. IPCC, 2014. Climate Change 2014: Synthesis Report. Contribution of Working 554 Groups I, II and III to the Fifht Assessment Report of the Intergovernmental Panel 555 on Climate Change. IPCC, Geneva, Switzerland. 556 26. Johnson, M., 1997. Management of spent diatomaceous earth from brewing 557 industry. 558 27. Kim, H., Hao, Z., Yang, B., Kim, S., Kwon, G., Jahng, D., 2018. Characterization 559 and Treatment of Emission Gas Condensate Generated from Biodrying of Sewage 560 Sludge. Environ. Eng. Sci. 35, ees.2017.0362. 561 28. Koppejan, J., van Loo, S., 2008. The Handbook of Biomass Combustion and Co- 562 firing. Earthscan, London. 563 29. Leffingwell, 2018. http://www.leffingwell.com/odorthre.htm, Accessed on 564 December 2018. [WWW Document]. 565 30. Maulini-Duran, C., Artola, A., Font, X., Sánchez, A., 2013. A systematic study of 566 the gaseous emissions from biosolids composting: Raw sludge versus 567 anaerobically digested sludge. Bioresour. Technol. 147, 43–51. 568 31. Moënne-Loccoz, P., Fee, J.A., 2010. Catalyzing NO to N2O in the nitrogen cycle. 569 Science (80-. ). 330, 10–11. 570 32. Nagata, Y., 2003. Odor measurement review, Measurement of odor threshold by 571 triangle odor bag method. Minist. Enviornmental Gov. Japan 122–123. 572 33. Pagans, E., Font, X., Sánchez, A., 2006. Emission of volatile organic compounds 573 from composting of different solid wastes: Abatement by biofiltration. J. Hazard. 574 Mater. 131, 179–186. 575 34. Pan, J., Cai, H., Zhang, Z., Liu, H., Li, R., Mao, H., Awasthi, M.K., Wang, Q., Zhai, 24 576 L., 2018. Comparative evaluation of the use of acidic additives on sewage sludge 577 composting quality improvement, nitrogen conservation, and greenhouse gas 578 reduction. Bioresour. Technol. 270, 467–475. 579 35. Pognani, M., Barrena, R., Font, X., Sánchez, A., 2012. A complete mass balance of 580 a complex combined anaerobic/aerobic municipal source-separated waste 581 treatment plant. Waste Manag. 32, 799–805. 582 36. Ponsá, S., Gea, T., Sánchez, A., 2010. Different Indices to Express Biodegradability 583 in Organic Solid Wastes. J. Environ. Qual. 39, 706–712. 584 37. Ribes, A., Carrera, G., Gallego, E., Roca, X., Berenguer, M.J., Guardino, X., 2007. 585 Development and validation of a method for air-quality and nuisance odors 586 monitoring of volatile organic compounds using multi-sorbent adsorption and gas 587 chromatography/mass spectrometry thermal desorption system. J. Chromatogr. A 588 1140, 44–55. 589 38. Rincón, C.A., De Guardia, A., Couvert, A., Le Roux, S., Soutrel, I., Daumoin, M., 590 Benoist, J.C., 2019. Chemical and odor characterization of gas emissions released 591 during composting of solid wastes and digestates. J. Environ. Manage. 233, 39–53. 592 39. Schiavon, M., Martini, L.M., Corrà, C., Scapinello, M., Coller, G., Tosi, P., 593 Ragazzi, M., 2017. Characterisation of volatile organic compounds (VOCs) 594 released by the composting of different waste matrices. Environ. Pollut. 231, 845– 595 853. 596 40. Shen, Y., Chen, T. Bin, Gao, D., Zheng, G., Liu, H., Yang, Q., 2012. Online 597 monitoring of volatile organic compound production and emission during sewage 598 sludge composting. Bioresour. Technol. 123, 463–470. 25 599 41. Sironi, S., Capelli, L., Céntola, P., Del Rosso, R., Il Grande, M., 2006. Odour 600 emission factors for the prediction of odour emissions from plants for the 601 mechanical and biological treatment of MSW. Atmos. Environ. 40, 7632–7643. 602 42. Tambone, F., Scaglia, B., Scotti, S., Adani, F., 2011. Effects of biodrying process on 603 municipal solid waste properties. Bioresour. Technol. 102, 7443–7450. 604 43. United States Environmental Protection Agency (USEPA), 1995. Compilation of air 605 pollutant emission factors, AP-42, vol. 1, 5th ed., Stationary point and area 606 sources. Research Triangle Park, NC, USA. 607 44. US Dept. of Agriculture, US Composting Council, 2001. Test methods for the 608 examination of composting and compost purpose, in: Edaphos International, 609 Houston. 610 45. Villegas, M., Huiliñir, C., 2014. Biodrying of sewage sludge: Kinetics of volatile 611 solids degradation under different initial moisture contents and air-flow rates. 612 Bioresour. Technol. 174, 33–41. 613 46. Water Environment Federation, 1995. Standard Methods for the Examination of 614 Water and Wastewater. 615 47. Wei, Y., Tong, J., Wang, Y., ChuLu, B., Yu, D., Sui, Q., Zhang, J., Chen, M., 2016. 616 Who contributes more to N2O emission during sludge bio-drying with two 617 different aeration strategies, nitrifiers or denitrifiers? Appl. Microbiol. Biotechnol. 618 101, 3393–3404. 619 48. Yuan, J., Chadwick, D., Zhang, D., Li, G., Chen, S., Luo, W., Du, L., He, S., Peng, 620 S., 2016. Effects of aeration rate on maturity and gaseous emissions during sewage 621 sludge composting. Waste Manag. 56, 403–410. 26 622 49. Zarra, T., Naddeo, V., Oliva, G., Belgiorno, V., 2016. Odour emissions 623 characterization for impact prediction in anaerobic-aerobic integrated treatment 624 plants of municipal solid waste. Chem. Eng. Trans. 54, 91–96. 625 50. Zhao, L., Gu, W.M., He, P.J., Shao, L.M., 2011. Biodegradation potential of bulking 626 agents used in sludge bio-drying and their contribution to bio-generated heat. 627 Water Res. 45, 2322–2330. 628 629 27 630 Legend to Figures 631 Figure 1. Scheme of the biodrying system and the external industrial mixer used during the 632 experiment (TT represents the temperature measure; FC represents the mass flow controller; 633 O2T represents the O2 measure). 634 Figure 2. (a) Performance of the biodrying process. In black dots, the evolution of the 635 temperature of the material in the middle of the reactor (ºC); in white dots, the evolution of the 636 moisture content of the solid mixture (%); solid line, the specific airflow introduced in the 637 reactor (L·min-1·kg-1 VS0), black arrows represent the mixing pattern. (b) CH4 and N2O 638 emission rate profile during the biodrying process. In black dots, the temperature of the material 639 in the middle of the reactor (ºC); in white squares, the inlet specific airflow rate (L·min-1·kg-1 640 VS0); in blue squares, the evolution of the CH4 emission rate (mg CH4·d-1); in green triangles, 641 the evolution of the N2O emission rate (mg N2O·d-1). (c) Odour emission rate (OER) profile 642 during the biodrying process. In black dots, the temperature of the material in the middle of the 643 reactor (ºC); in white squares, the inlet specific airflow rate (L·min-1·kg-1 VS0); in cyan squares, 644 the evolution of the odour emission rate (OER, ou·d-1). 645 Figure 3. Distribution of the different VOC families found in each sample, expressed in 646 percentage of abundance respect the whole sample, from the outlet gases of the biodrying 647 reactor. 28 648 Tables 649 Table 1. Physical-chemical properties of the feedstocks and mixture used during the biodrying 650 process. Physicochemical Biodrying SS PW DE properties mixture Moisture (%) 83.5 ± 0.1 18.0 ± 0.0 3.9 ± 0.0 54.6 ± 0.1 Organic matter (%) 68.9 ± 0.2 88.7 ± 0.0 35.2 ± 0.1 85.2 ± 0.0 C/N ratio 9.1 31.6 1321.6 13.6 pH 6.7 8.2 6.2 – Electrical 2.5 1.7 1.2 – conductivity (mS/cm) DRI24h 7.3 ± 0.9 – – 6.3 ± 1.7 (g O2·h-1·kg-1 VS) 651 SS: sewage sludge; PW: pruning waste; DE: diatomaceous earth 652 29 653 Table 2. Summary of the process parameters evaluated during the biodrying process. Parameter Biodrying Units Total air volume 1.4E+05 L Air supplied 0.46 L·min-1·kg-1 VS0 Average air supply 2.54 L·min-1·kg-1 VSSS Total evaporated water 13.3 kg H2O Initial 15.1 ± 0.1 HCV MJ·kg-1 TS Final 13.6 ± 0.2 Calorific value Initial 4.3 ± 0.1 LCV MJ·kg-1 product Final 7.1 ± 0.1 Initial 6.3 ± 1.7 DRI24h g O2·h-1·kg-1 VS Biological Final 3.5 ± 0.1 stability Initial 403.6 ± 43.8 AT4 g O2·kg-1 VS Final 175.1 ± 18.9 654 HCV: higher calorific value; LCV: Lower Calorific Value 30 655 Table 3. CH4, N2O, NH3, tVOCs and odour emission factors for the biodrying process. GHG 656 emission factors are expressed in kg of pollutant and in kg of CO2 equivalent per Mg DM0-SS 657 and Mg DM0 of mixture, respectively. Odorant emission factors are expressed in kg of pollutant 658 per Mg DM0-SS and Mg DM0 of mixture and odour emission factor is expressed in ou per Mg 659 DM0-SS and Mg DM0 of mixture, respectively. kg pollutant·Mg-1 DM0-SS kg pollutant·Mg-1 DM0 CH4 emission factor 4.54E-02 1.02E-02 GHGs N2O emission factor 1.02E-01 2.28E-02 kg CO2eq·Mg-1 DM0-SS kg CO2eq·Mg-1 DM0 CH4 emission factor a 1.27 0.28 GHGs N2O emission factor a 26.95 6.04 GHG emission factor a, b 28.22 6.32 kg pollutant·Mg-1 DM0-SS kg pollutant·Mg-1 DM0 NH3 emission factor 1.23 2.75E-01 Odorants H2S emission factor 1.68E-03 3.77E-04 tVOCs emission factor 1.38E-01 3.10E-02 ou·Mg-1 DM0-SS ou·Mg-1 DM0 Odour Odour emission factor 3.10E+07 6.95E+06 a 660 GHGs emission for CH4 and N2O, on a 100-yr frame, is 28 and 265 times higher than that of 661 CO2, respectively (IPCC, 2014). b 662 GHGs emission factor as the sum of the CH4 and N2O emission factors. 31 663 Figures 664 FIGURE 1 665 666 32 667 FIGURE 2 668 a) 669 670 b) 671 672 673 33 674 c) 675 676 34 677 FIGURE 3 678 679 35 680 Filling in sewage sludge biodrying gaps: Greenhouse gases, volatile organic 681 compounds and odour emissions 682 (E-Supplementary Information) a, b 683 Daniel González , Nagore Guerra c, Joan Colón c, David Gabriel b, Sergio Ponsá c, 684 Antoni Sánchez a, * 685 a 686 Composting Research Group (GICOM) Dept. of Chemical, Biological and 687 Environmental Engineering, Universitat Autònoma de Barcelona, 08193-Bellaterra 688 (Barcelona), Spain. b 689 Group of biological treatment of liquid and gaseous effluents (GENOCOV) Dept. of 690 Chemical, Biological and Environmental Engineering, Universitat Autònoma de 691 Barcelona, 08193-Bellaterra (Barcelona), Spain 692 c BETA Technology Centre: “U Science Tech”, University of Vic-Central University of 693 Catalonia, 08500 Vic, Barcelona, Spain 694 695 * Corresponding author: 696 Antoni Sánchez 697 Composting Research Group (GICOM), Dept. of Chemical, Biological and 698 Environmental Engineering 699 Universitat Autònoma de Barcelona 700 Bellaterra, 08193 (Spain) 701 Email address: antoni.sanchez@uab.cat 36 702 E-Supplementary Information 703 Table 1S. Quantified VOCs in the gaseous samples obtained from the biodrying monitoring, with its odour detection threshold (ODT), expressed in ppbv 704 (“