Farm-Scale Testing of Soybean Peroxidase and Calcium Peroxide

Agricultural and Biosystems Engineering Agricultural and Biosystems Engineering Publications

10-2017 Farm-scale testing of soybean peroxidase and calcium peroxide for surficial swine manure treatment and mitigation of odorous VOCs, ammonia and hydrogen sulfide mie ssions Devin L. Maurer Iowa State University, [email protected]

Jacek A. Koziel Iowa State University, [email protected]

Kelsey Bruning Iowa State University, [email protected]

David B. Parker United States Department of Agriculture

Follow this and additional works at: http://lib.dr.iastate.edu/abe_eng_pubs Part of the Agriculture Commons, and the Bioresource and Agricultural Engineering Commons The ompc lete bibliographic information for this item can be found at http://lib.dr.iastate.edu/ abe_eng_pubs/813. For information on how to cite this item, please visit http://lib.dr.iastate.edu/ howtocite.html.

This Article is brought to you for free and open access by the Agricultural and Biosystems Engineering at Iowa State University Digital Repository. It has been accepted for inclusion in Agricultural and Biosystems Engineering Publications by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. Farm-scale testing of soybean peroxidase and calcium peroxide for surficial swine manure treatment and mitigation of odorous VOCs, ammonia and hydrogen sulfide mie ssions

Abstract The swine industry, regulatory agencies, and the public are interested in farm-tested methods for controlling gaseous emissions from swine barns. In earlier lab- and pilot-scale studies, a renewable catalyst consisting of soybean peroxidase (SBP) mixed with calcium peroxide (CaO2) was found to be effective in mitigating gaseous emissions from swine manure. Thus, a farm-scale experiment was conducted at the university's 178-pig, shallow-pit, mechanically-ventilated swine barn to evaluate SBP/CaO2 as a surficial manure pit additive under field conditions. The BPS was applied once at the beginning of the 42-day experiment at an −2 application rate of 2.28 kg m with 4.2% CaO2 added by weight. Gas samples were collected from the primary barn exhaust fans. As compared to the control, significant reductions in gaseous emissions were observed for ammonia (NH3, 21.7%), hydrogen sulfide H( 2S, 79.7%), n-butyric acid (37.2%), valeric acid (47.7%), isovaleric acid (39.3%), indole (31.2%), and skatole (43.5%). Emissions of dimethyl disulfide/ methanethiol (DMDS/MT) increased by 30.6%. Emissions of p-cresol were reduced by 14.4% but were not statistically significant. There were no significant changes to the greenhouse gas (GHG) emissions of methane (CH4), carbon dioxide (CO2) and nitrous oxide (N2O). The ott al (material + labor) treatment cost was $2.62 per marketed pig, equivalent to 1.5% of the pig market price. The osc t of CaO2 catalyst was ∼60% of materials cost. The osc t of soybean hulls (SBP source) was $0.60 per marketed pig, i.e., only 40% of materials cost.

Keywords Swine manure, Emissions control, Mitigation, Soybean peroxidase, Odor

Disciplines Agriculture | Bioresource and Agricultural Engineering

Comments This article is from Atmospheric Environment 166 (2017), 467-478, doi:10.1016/j.atmosenv.2017.07.048.

Rights Works produced by employees of the U.S. Government as part of their official duties are not copyrighted within the U.S. The onc tent of this document is not copyrighted.

This article is available at Iowa State University Digital Repository: http://lib.dr.iastate.edu/abe_eng_pubs/813 Atmospheric Environment 166 (2017) 467e478

Contents lists available at ScienceDirect

Atmospheric Environment

journal homepage: www.elsevier.com/locate/atmosenv

Farm-scale testing of soybean peroxidase and calcium peroxide for surﬁcial swine manure treatment and mitigation of odorous VOCs, ammonia and hydrogen sulﬁde emissions

* Devin L. Maurer a, Jacek A. Koziel a, , Kelsey Bruning a, David B. Parker b a Dept. of Agricultural and Biosystems Engineering Iowa State University, Ames, IA 50011, United States b USDA-ARS, Conservation and Production Research Laboratory, Bushland, TX 79012, United States highlights graphical abstract

Post application of SBP/CaO2 to manure at a swine farm was tested for 42 d. Surﬁcially-applied renewable biocatalyst was used to mitigate gaseous emissions. 2.28 kg m 2 reduced non-sulfur VOC

(36%), NH3 (22%), and H2S (80%) emissions.

No signiﬁcant change to CO2,CH4 or N2O emissions were observed. Est. material cost ¼ $1.45 pig 1, equivalent of ~0.8% of the pig market price. article info abstract

Article history: The swine industry, regulatory agencies, and the public are interested in farm-tested methods for con- Received 7 February 2017 trolling gaseous emissions from swine barns. In earlier lab- and pilot-scale studies, a renewable catalyst Received in revised form consisting of soybean peroxidase (SBP) mixed with calcium peroxide (CaO2) was found to be effective in 10 July 2017 mitigating gaseous emissions from swine manure. Thus, a farm-scale experiment was conducted at the Accepted 25 July 2017 university's 178-pig, shallow-pit, mechanically-ventilated swine barn to evaluate SBP/CaO as a surficial Available online 27 July 2017 2 manure pit additive under field conditions. The SBP was applied once at the beginning of the 42-day 2 experiment at an application rate of 2.28 kg m with 4.2% CaO2 added by weight. Gas samples were Keywords: fi Swine manure collected from the primary barn exhaust fans. As compared to the control, signi cant reductions in fi Emissions control gaseous emissions were observed for ammonia (NH3, 21.7%), hydrogen sul de (H2S, 79.7%), n-butyric Mitigation acid (37.2%), valeric acid (47.7%), isovaleric acid (39.3%), indole (31.2%), and skatole (43.5%). Emissions of Soybean peroxidase dimethyl disulfide/methanethiol (DMDS/MT) increased by 30.6%. Emissions of p-cresol were reduced by Odor 14.4% but were not statistically significant. There were no significant changes to the greenhouse gas (GHG) emissions of methane (CH4), carbon dioxide (CO2) and nitrous oxide (N2O). The total (material þ labor) treatment cost was $2.62 per marketed pig, equivalent to 1.5% of the pig market price. The cost of CaO2 catalyst was ~60% of materials cost. The cost of soybean hulls (SBP source) was $0.60 per marketed pig, i.e., only 40% of materials cost. © 2017 Elsevier Ltd. All rights reserved.

* Corresponding author. E-mail address: [email protected] (J.A. Koziel). http://dx.doi.org/10.1016/j.atmosenv.2017.07.048 1352-2310/© 2017 Elsevier Ltd. All rights reserved. 468 D.L. Maurer et al. / Atmospheric Environment 166 (2017) 467e478

1. Introduction supply of fresh manure into the deep storage pit below the slatted floor; and variations of ventilation rate, inlet and barn set- Emissions of odor, volatile organic compounds (VOCs), ammonia point air temperatures and relative humidity (RH), phased animal (NH3), hydrogen sulfide (H2S) and greenhouse gases (GHGs) (CH4, diet, and waste management practices. The pork industry typically N2O and CO2) are a side effect of pork production. Gaseous emis- requires proven farm-scale mitigation technologies and their eco- sions originate from animal housing, manure storage, treatment, nomic evaluations prior to adoption. and handling, and from land application of manure. Mitigation of these emissions is of importance due to their effect on local and 2. Materials and methods regional air quality and their association with climate change. There has been considerable research in the past two decades to 2.1. Experimental design quantify and mitigate gaseous emissions from swine farms (Van Huffel et al., 2016; Philippe and Nicks, 2015; Cai et al., 2015; Chen Farm-scale testing was conducted at the Iowa State University et al., 2009; Sun et al., 2008b; Hoff et al., 2006), poultry farms Allen E. Christian Swine Teaching Farm. The research was con- (Cai et al., 2007; Heber et al., 2006; Rockafellow et al., 2012); dairy ducted in two adjacent rooms within a single confinement barn. farms (Sun et al., 2008a); and beef cattle feedyards (Parker et al., Each room housed 89 pigs (Fig. 1 and Fig. S1, Supplementary Ma- 2005, 2016b). Maurer et al. (2016) recently summarized the state terial). The shallow manure pit (11 m ✕ 11 m ✕ 0.61 m depth) in one of emission mitigation measures for livestock and poultry in a room was used as the control, and an identical manure pit in the scientific literature database on the Iowa State University (ISU) other room was treated with surficial application of SBP at a 2 Extension and Outreach website (Air Management Practices treatment (TRT) of 2.28 kg m with 4.2% CaO2/SBP by weight Assessment Tool; AMPAT). AMPAT is a user-friendly website that (Figs. S2 and S3). The application resulted in a 6 mm thick layer of aims to provide an objective overview of best management prac- SPB/CaO2 on the surface of the pit (Fig. S4). The SBP was produced tices to address odor, emissions and dust at livestock operations from ground soybean hulls (Bio-Research Products, Inc. North (Iowa State University Extension and Outreach, 2016a). Manure Liberty, IA, USA). The manure pit was treated once at the start of the treatment is one of 12 technologies that has been researched for the experiment. The SBP/CaO2 was premixed and manually applied control of odors from manure storage and handling. However, most through 2 cm gaps in the fully-slatted floor using a special funnel of the previous manure treatment research has been limited to lab- device. or pilot-scale testing. Field/farm-scale trials were conducted in less Gas samples were collected directly from the primary exhaust than ~25% of the research. Unfortunately, none of the reported fan in each room (Fig. 1 and S5). Initially, emissions data was research projects provided a comprehensive data for all gases of collected for 14 d prior to SBP/CaO2 application to establish baseline interest. This is important, because emissions reduction of one gas emissions from TRT and control rooms. Then, emissions data target pollutant can sometimes result in increased emissions of was collected for 42 d after SBP/CaO2 application, for a total other important target pollutants (e.g., NH3 vs. N2O). Thus, it is experiment duration of 56 d (VOCs were only monitored for a total important to perform comprehensive testing of mitigation tech- of 50 d). The experiment was conducted in the winter, when nologies using farm-scale trials. outside temperatures ranged from 11 to 18 C. Inside tempera- Researchers have shown that peroxidase-based treatment can tures were maintained at 14 to 22 C. Each room was 100% me- reduce some odorous phenolic and indolic compounds in lab- or chanically ventilated. Only a primary ventilation fan (the only pilot-scale studies (Morawski et al., 2001; Tonegawa et al., 2003; Ye emissions exhaust point during the trial period) in each room was et al., 2009; Parker et al., 2012, 2016a; Steevensz et al., 2014). operating resulting in ~5.4 air exchanges per h during sampling. Soybean peroxidase is a bio-based, value-added product that is Barn ventilation airflow rate was determined using airflow cali- produced by grinding soybean hulls, a low value and low utility bration for the primary ventilation fan in each room (Fig. S6). byproduct. The swine industry uses soybeans as a key ingredient of Measurement of NH3 and H2S concentrations, temperature and animal feed, thus minimizing any concerns about adding a RH were conducted in real-time. Gas samples for GHGs and soybean-based product to the manure. odorous VOCs were collected and subsequently analyzed in the lab. Most recently, as a follow-up to lab-scale testing, Maurer et al. Emission rates were calculated as a product of measured gas con- (2017a) investigated the efficacy of surficial application of SBP/ centrations and the total airflow rate through each room, adjusted CaO2 to swine manure on gaseous emissions of odorous volatile for standard conditions and dry air. Environmental data was VOCs, NH3,H2S and GHG in pilot-scale experiments. Effects of dose collected along with manure for quality evaluation (manure was and time were studied over the 137 day trial. Maurer et al. (2017a) collected before SBP/CaO2 application, immediately after SBP/CaO2 reported significant reductions in emissions for DMDS/MT (36.2%e application, and again at the end of the study). Ammonia, H2S and 84.7%), p-cresol (53.1%e89.5%), skatole (63.2%e92.5%) and NH3 RH were measured every other day after SBP/CaO2 application. (14.6%e67.6%). Significant increases in CH4 (32.7%e232%) and CO2 Percent RH was monitored via an 850071 Environmental Quality (20.8%e124%) emissions were observed. The key finding from the meter (Sper Scientific, Scottsdale, AZ, USA). Methane, CO2 and N2O pilot-scale experiment (Maurer et al., 2017a) was that an SBP/CaO2 were measured every other day after SBP/CaO2 application. VOC dose of 2.28 kg m 2 performed as well as higher doses. Thus, this measurements were collected on n ¼ 19 d over the 42 day moni- 2 2.28 kg m SBP/CaO2 dose was selected for this farm-scale trial toring after the SBP/CaO2 application. Pig weights were monitored because of the economical and practical application aspects. throughout the trial. This study aimed to evaluate the farm-scale efficacy of SBP/CaO2 surficial treatment on finisher pig manure emissions over a 42-day 2.2. Volatile organic compounds (VOCs) evaluation period. This study follows the lab-pilot-farm-scales progression of testing for a promising emissions mitigation tech- Air samples for VOC measurement were collected using 65 mg nology. The farm-scale experiment addresses an important defi- Tenax TA sorbent tubes (4 mm O.D. 0.10 m long) constructed of ciency in controlled lab-scale and pilot scale studies, i.e. the effects 304-grade stainless steel that had been double passivated with a of growing animals. Some of these effects include the presence of proprietary surface-coating process. Field air samples were taken other emitting sources (e.g. breathing and excreting animals, using a portable vacuum sampling pump with a set flow rate of manure on slatted floor, feed); the continuous and increasing re- 50 mL min 1 for 15 min, and analyzed within two days. The D.L. Maurer et al. / Atmospheric Environment 166 (2017) 467e478 469

Fig. 1. Schematic of one of the two identical rooms used in the farm-scale experiment. Each room had eight pens and housed 89 pigs. Fans with gray lines through them were not used during the experimental period.

sampling flow rates were verified with a NIST-traceable digital flow determining the extent of that conversion with field air samples, meter (Bios International, Butler, NJ, USA). Chemical analyses of we conducted additional laboratory testing to determine the po- swine odorants were completed using a thermal desorption gas tential extent of MT to DMDS/DMTS conversion (Cai et al., 2015). chromatography - mass spectrometer (TD-GCeMS) system (Agilent We determined that complete conversion of MT to DMDS 6890 GC; Microanalytics, Round Rock, TX, USA) using procedures (97.5e99.5%) and DMTS (0.5e2.5%) was possible for moist standard previously described (Zhang et al., 2012; Cai et al., 2015). Work by gas stored on sorbent tubes from 1 to 3 d. Therefore, these com- Andersen et al. (2012) has shown that thermal desorption of sor- pounds are reported as DMDS/MT and DMTS/MT in this bent tubes can convert methanethiol (MT) to dimethyl disulfide manuscript. (DMDS) and dimethyl trisulfide (DMTS). Because of the difficulty in 470 D.L. Maurer et al. / Atmospheric Environment 166 (2017) 467e478

2.3. Ammonia and hydrogen sulfide testing while “After SBP/CaO2 application” represent the first 42 d post TRT application: Ammonia and H2S concentrations were measured using a E E Drager X-am 5600 portable gas analyzer (Luebeck, Germany) with %R ¼ Con Treat*100 (3) NH3 and low range H2S XS sensors. The analyzer was calibrated ECon using Drager calibration software, an Environics 4040 gas dilution fl system (Tolland, CT, USA) and standard gases (Praxair, Ames, IA, where: %R is the % of reduction, ECon is the average ux estimate of the desired time interval (day, week, biweek or overall) of the USA) (NH3: 102 ppm and H2S: 15.6 ppm). control, and ETreat is the average flux estimate of the desired time interval (day, week, biweek or overall) of the treated. 2.4. Greenhouse gases

2.7. Statistical analyses Gas samples were collected via syringe and 5.9 mL Exetainer vials (Labco Limited, UK) and were analyzed for GHG concentra- An autoregressive like model using a spatial power option in tions on a GHG GC equipped with FID and ECD detectors (SRI In- PROC MIXED, in SAS System (version 9.4, SAS Institute, Inc. Cary, struments, Torrance, CA, USA). Gas method detection limits were NC, USA) accounting for the repeated measures on each room 1.99 ppm, 170 ppb, and 20.7 ppb for CO2,CH4 and N2O, respectively. across time was used to analyze the data by comparing mean flux Standard calibrations were constructed daily using 2 ppm and values to determine the p values, p < 0.05 was used as the signif- 10.3 ppm CH4; 510 ppm, 1010 ppm and 2010 ppm CO2; and icance level. 0.101 ppm, 1.02 ppm and 10.1 ppm N2O (Air Liquide America, Plumsteadville, PA, USA). Standards used for calibrations were done 3. Results in duplicate for CH4 and CO2 while N2O standards were done in triplicate. 3.1. Temporal changes in SBP/CaO2 surficial coverage 2.5. Swine manure analysis The 6 mm surface layer of SBP/CaO2 floated on the surface after application but was incorporated into the manure over time as new Swine manure analyses were completed as described in Maurer feces, urine, water, and spilled feed entered the pit. The incorpo- et al. (2017a,b) using standard methods described in AOAC (2000) ration was not consistent across the entire pit surface. The SBP/CaO and APHA (1998). Total C, H, N, and S were determined using a PE 2 was incorporated sooner near the waterers and feeders, as well as 2100 Series II combustion analyzer (Perkin Elmer Inc. Waltham, in the localized areas where the pigs defecated. In other areas of the MA) with a cysteine calibration standard and an expected precision room where pigs were cleaner, including the alley way which and accuracy of ± 0.3% for each element. The combustion and received little pig traffic, the SBP/CaO was incorporated slowly reduction temperatures were both 975 C. All standards and re- 2 over several weeks, similar to what was observed in the former agents were from Perkin Elmer and Elementar America's Inc. pilot-scale experiment (Maurer et al., 2017a).

2.6. Gas emissions 3.2. Environmental parameters

Measured gas concentrations were used for estimation of gas 1 Over the course of the farm-scale trial the measured tempera- emissions from each room (mass time ) and also for estimation of ture in the control room and the SBP/CaO treated room ranged fl 1 2 gas emissions expressed as a ux (mass time per surface area of from 15.1 to 21.1 C and 14.1 to 21.8 C respectively. The trial started fl manure in the pit under slatted oor). Gas emissions then were mid-October and ended mid-December resulting in lower room calculated using Eq. (1): temperatures later in the trial (Fig. S7). There was no significant difference in the temperature between the control and SBP/CaO Estd ¼ Qair *C (1) 2 std std treated rooms (p ¼ 0.4643). The RH measured in each room also was not significantly different after (p ¼ 0.1995) the SBP/CaO2 TRT where: Estd is standardized emissions in ng/h. Qairstd is the stan- 1 was applied. The RH of the control and SBP/CaO2 treated rooms dardized air flow rate (mL h ) from Eq. S(4), Cstd is the standardized e concentration (mgmL 1) from Eq. S(9). ranged from 59 to 100% and 50 100%, respectively (Fig. S8). The Gas emissions of all measured gases were normalized to account atmospheric pressure was recorded over the course of the trial for the differences observed during the baseline measurements (Fig. S9). The total weight of growing pigs (expressed in animal unit (A.U.) equivalents defined here as total pigs weight in kg divided by completed prior to SBP/CaO2 application. The normalization was made by adding the average daily difference between the rooms of 500 kg) ranged from 9.41 A.U. (treated room) and 9.31 A.U. (control room) on day nine to 20.44 A.U. for both rooms on day 58 before the the two weeks prior to SBP/CaO2 TRT to the emissions from the control room: hogs began to be sold for market (Fig. S10).

¼ð Þþ Enorm Etb Ecb Eca (2) 3.3. Volatile organic compounds where: Enorm is the normalized flux estimate of the control room, Etb Reductions in the emissions of odorous volatile fatty acids were is the average daily flux estimate for the treated room before SBP/ statistically significant and were 37.2% (p ¼ 0.0012), 47.7% CaO2 application (first 14 d), Ecb is the average daily flux estimate (p < 0.0001) and 39.3% (p ¼ 0.0004) for n-butyric acid, valeric acid for the control room before SBP/CaO2 application (first 14 d), and and isovaleric acid, respectively (Fig. 2A, B, C). However, all VFAs Eca is each daily flux estimate for the control room over the entire fluxes were not reduced below a calculated odor detection trial. threshold with exception of one day for valeric acid and isovaleric Overall mean % reduction for each measured gas was estimated acid. using all measured flux for either “Before” or “After” period. “Before Reductions of odorous indole and skatole were both statistically SBP/CaO2 application” represent the 2-week period of baseline significant with reductions of 31.2% (p ¼ 0.0017) and 43.5% D.L. Maurer et al. / Atmospheric Environment 166 (2017) 467e478 471

2 Fig. 2. Effects of SBP/CaO2 on measured flux of n-butyric acid (A), valeric acid (B), isovaleric acid (C), indole (D), skatole (E), p-cresol(F), as a function of time. >: 2.28 kg m SBP/ CaO2 dose, -: Control, ——— day of SBP/CaO2 application, $$$$$ odor detection threshold. Odor detection threshold flux values were calculated based on Devos et al. (1990) and NOAA (1999) and average temperature, atmospheric pressure, relative humidity and exhaust fan air flow of the treated and control rooms over the trial period.

(p < 0.0001) respectively (Fig. 2D and E). However, the SBP/CaO2 room. The SBP/CaO2 TRT resulted in an overall statistically signifi- TRT did not reduce indole concentrations below that of the calcu- cant reduction of 79.7% (p < 0.0001) in H2S and a significant lated odor detection threshold of for any of the observed days. The reduction (p 0.0001) for every sampling period over the 42 d after SBP/CaO2 TRT reduced the concentration of skatole below its application ranging from 42.0% to 99.5% (Fig. 4B). The H2S emis- calculated odor detection threshold for two days. Emissions of p- sions were mitigated to below odor threshold in the treated room cresol were also reduced by 14.4% but were not statistically sig- on 56% of the observed days. The reduction of H2S is important nificant (p ¼ 0.34) (Table 1 and Fig. 2F). There were no observed because of toxicity risks associated with the inhalation of H2Sby days of p-cresol flux below the calculated odor detection threshold workers and pigs in swine barns. in either the control or treated rooms. Similarly, no statistical difference was observed for DMTS/MT emissions (p ¼ 0.98) (Fig. 3A). DMDS/MT flux from the SBP/CaO2 3.5. Greenhouse gases treated room were also not significantly different (p ¼ 0.94) compared to that of the control room (Fig. 3B). The DMDS/MT flux There were no statistically significant differences in estimated fl ¼ ¼ ¼ estimates for the control and SBP/CaO2 treated rooms were both ux for CH4 (p 0.23), CO2 (p 0.30) and N2O(p 0.07) among the fl ± 1 2 below the calculated odor detection threshold over the entire trial TRT (Fig. 5). The overall mean CH4 uxes were 304 59 mg h m ± 1 2 period. Mean measured gas concentrations of SBP/CaCO2 treated and 285 97 mg h m for the control and treated room, ± room after TRT are summarized in Table S1. respectively. The mean CH4 concentrations were 23.5 7.5 ppm and 23.9 ± 4.7 ppm for the control and treated room, respectively. 1 2 The overall mean CO2 fluxes were 83,790 ± 12,746 mg h m and 3.4. Ammonia and hydrogen sulfide 81,284 ± 17,492 mg h 1 m 2 for the control and treated room, respectively. The mean CO2 concentrations were 2346 ± 461 ppm There was an overall statistically significant reduction of 21.7% and 2533 ± 394 ppm for the control and treated room, respectively. 1 2 (p ¼ 0.0172) in NH3 flux after the SBP/CaO2 TRT (Fig. 4A), with two The overall mean N2O fluxes were 12.8 ± 2.1 mg h m and days below the calculated odor reduction threshold in the treated 11.6 ± 2.5 mg h 1 m 2 for the control and treated room, 472 D.L. Maurer et al. / Atmospheric Environment 166 (2017) 467e478

Table 1 Mean ﬂux and percent reductions for the treatment and control, as compared to previous lab- and pilot-scale research.

This Study Maurer et al., 2017a Parker et al., Parker et al., (Farm-scale) (Pilot-scale) 2016a 2012 (Lab-scale) (Lab-scale)

¡ ¡ Mean Flux (mg h 1 m 2) % Reduction

Control 0 Control 15 TRT 0 TRT 15 Total 42 Day 14 Daya Total 136 42 Dayb 14 Daya 14 day 2 day e14d e56d e14d e56d Day

¡2 2 2c 2d SBP/CaO2 2.28 kg m 2.28 kg m 2.50 kg m 0.585 kg m Dose Application Surface Surface Surface Mixed VOCs n-butyric 4.65 5.74 4.65 3.61 37.2 (0.0012) 34.6 8.50 (0.944) 17.7 (0.922) 19.6 (0.933) “VFAs” 29.3 90.6 acid (0.0114) valeric acid 1.24 1.42 1.24 0.743 47.7 47.2 87.5 (0.331) 5.57 (0.781) 18.5 75.8 (<0.0001) (0.0190) (0.396) isovaleric 0.860 1.22 0.860 0.739 39.3 (0.0004) 41.9 42.7 (0.474) 46.9 (0.485) 57.5 (0.238) 87.7 acid (0.0009) indole 0.0489 0.0813 0.0489 0.0559 31.2 (0.0017) 41.3 3.18 82.9 (0.267) 73.0 (0.785) 13.0 22.1 (0.0002) (0.811) skatole 0.144 0.182 0.144 0.103 43.5 49.1 72.6 81.4 87.4 83.3 32.4 (<0.0001) (<0.0001) (<0.0001) (<0.0001) (<0.0001) p-cresol 1.33 2.64 1.33 2.26 14.4 (0.3417) 26.3 64.9 (0.02) 58.3 (0.03) 67.8 (0.131) 86.4 92.5 (0.0328) DMTS/MTe NA NA NA NA NA NA 16.9 (0.495) 19.5 (0.458) 4.77 (0.887) “Sulﬁdes” 122 NA DMDS/MT 0.0537 0.0949 0.0537 0.124 ¡30.6 0.114 36.2 (0.212) 65.1 (0.205) 0.0 (1.0) NA (0.9400) (0.4961)

NH3 & H2S NH3 145 276 145 216 21.7 (0.0172) 29.9 14.6 (0.120) 15.3 (0.110) 13.9 (0.193) NA NA (0.0287)

H2S 21.5 29.5 21.5 5.99 79.7 83.2 10.9 (0.688) 14.3 3.63 NA NA (<0.0001) (<0.0001) (0.680) (0.989) GHGS

CH4 286 304 286 285 6.15 (0.2253) 2.56 32.7 32.2 17.1 NA NA (0.5627) (0.077) (0.161) (0.730)

CO2 69700 83800 69700 81300 2.99 (0.3026) 8.23 20.8 24.6 16.3 NA NA (0.1979) (0.0138) (0.023) (0.294)

N2O 12.8 12.8 12.8 11.6 9.76 (0.0672) 4.61 2.63 2.91 3.06 NA NA (0.3951) (0.919) (0.944) (0.818)

Note: Values in parentheses are p values. a Values at Day 14 for comparison with Parker et al., 2016a. b Values at Day 42 for comparison with this study. c Corresponds to Parker et al., 2016a 25 g L 1 recalculated based on 3.8 L manure and a manure surface of 0.038 m2. d Corresponds to Parker et al., 2012 experiment 5, 50 g L 1 recalculated based on 0.175 L manure and a manure surface of 0.015 m2. e Gas not detected consistently above detection limits to be considered as reliable data. Parker et al., 2016a,b reduction values calculated from average means over the trial of Parker et al., (2016a), Table 2. Parker et al., 2012 reduction values calculated from average means over the trial of Parker et al., (2012), Table 4.

respectively. The mean N2O concentrations were 0.4 ± 0.1 ppm and application were compared (Table 1). The pilot-scale and the farm- 2 0.3 ± 0.1 ppm for the control and treated room, respectively. scale flux estimates for the 2.28 kg m SBP/CaO2 dose resulted in reductions of n-butyric acid, valeric acid, isovaleric acid, indole and 3.6. Manure analysis NH3, respectively, all statistically significant at the farm-scale but not at the pilot-scale (Maurer et al., 2017a) at the same SBP/CaO2 The manure from the treated room accumulated TKN at a faster dose and time (Table 1). Reductions in VFAs were also observed in rate (34.8 mg L 1 d 1) than the control room (26.4 mg L 1 d 1) shorter (i.e., 14 day and 2 day) lab-scale trials with the use of SBP/ (Fig. S11). Total NH3-N losses were 30.4 kg from the control room CaO2, while increases in indole were observed (Parker et al., 2012, and 27.6 kg from the treated room. Total N2O-N losses were 2016a). Ammonia was not measured at the lab-scale reported in 0.843 kg from the control room and 1.17 kg from the treated room Parker et al. (2012, 2016a). Skatole emissions were significantly 2 (Table 2). reduced by the 2.28 kg m SBP/CaO2 in both the farm-scale and The pH of the manure in both rooms increased during the 14- pilot-scale (Maurer et al., 2017a) trials over similar time periods. day baseline measurements, before the SBP/CaO2 TRT was applied. Reductions were also observed as a result of SBP/CaO2 TRT in Parker 2 After the SBP/CaO2 application, the pH of the control room manure et al. (2012, 2016a) for skatole. The 2.28 kg m SBP/CaO2 treatment continued to increase over the next 42 d while the pH of the treated did not result in significant reduction in p-cresol emissions at the room decreased (Fig. S12). farm-scale as was observed at pilot- (Maurer et al., 2017a) and lab- scales (Parker et al., 2012, 2016a). Closer inspection of data shows 4. Discussion that the mitigating effect for p-cresol does not last as long as for the other compounds, as there was a significant 26% (p ¼ 0.03) fi 4.1. Comparison with pilot and lab-scale experiments reduction in the rst 14 d but only 14% for the entire 42 d period. Similar, lower % reduction with time was also observed for skatole and indole. Maurer et al. (2017a) reported the same type of In general, mitigating effect trends of SBP/CaO2 were similar at diminishing mitigation effect for phenolics with time (Table 1) the farm- and pilot-scales when the first 42 d following SBP/CaO2 D.L. Maurer et al. / Atmospheric Environment 166 (2017) 467e478 473

Fig. 3. Effects of SBP/CaO2 on measured flux of DMTS/MT (A), DMDS/MT (B) as a 2 function of time. >: 2.28 kg m SBP/CaO2 dose, -: Control, ——— day of SBP/CaO2 application, $$$$$ odor detection threshold. Odor detection threshold flux values were calculated based on Devos et al. (1990) and average temperature, atmospheric pressure, relative humidity and exhaust fan air flow of the treated and control rooms over the trial period.

Fig. 5. Effects of SBP/CaO2 on CH4 (A), CO2 (B) and N2O (C) flux as a function of time. 2 >: 2.28 kg m SBP/CaO2 dose, -: Control, ——— day of SBP/CaO2 application. Fig. 4. Effects of SBP/CaO2 on measured flux of NH3 (A), H2S (B), function of time. >: 2 2.28 kg m SBP/CaO2 dose, -: Control, ——— day of SBP/CaO2 application, $$$$$ odor detection threshold. Odor detection threshold flux values were calculated based on farm-scale. Flux estimates at the pilot-scale (Maurer et al., 2017a) Devos et al. (1990) and Smeets et al. (2007) and average temperature, atmospheric pressure, relative humidity and exhaust fan air flow of the treated and control rooms showed an increase in H2S and a reduction in DMDS/MT, while the over the trial period. farm-scale resulted in reduction of H2S and no significant effect on DMDS/MT. Parker et al. (2016a) reported an overall increase in sulfides resulting from SBP/CaO2 application. The true nature of S- while previous lab-scale studies (Parker et al., 2012, 2016a) did not containing gas generation is obviously complicated and should be test the effect of time beyond 14 d. researched further. These observations illustrate the delicate bal- Flux estimates for sulfur containing compounds were drastically ance of chemical and microbial processes that are at work in a different between the pilot-scale (Maurer et al., 2017a) and the complex system such as a manure pit. 474 D.L. Maurer et al. / Atmospheric Environment 166 (2017) 467e478

Table 2 Physical/chemical properties of treated and untreated swine manure.

Analysis Control SBP/CaO2 Treated Day 2 Day 14* Day 56 Day 2 Day 14* Day 56

TS% 4.65 ± 0.37 8.07 ± 0.05 8.78 ± 0.02 3.01 ± 0.05 8.27 ± 0.05 8.49 ± 0.23 VS% 3.60 ± 0.36 6.59 ± 0.04 7.23 ± 0.10 2.21 ± 0.05 6.66 ± 0.15 6.69 ± 0.26 Volatility% 77.3 ± 1.70 81.6 ± 0.1 82.3 ± 0.9 73.5 ± 0.4 80.5 ± 1.3 78.8 ± 0.9 pH 7.64 ± 0.01 7.80 ± 0.03 7.92 ± 0.02 7.96 ± 0.03 8.16 ± 0.01 7.75 ± 0.01 COD mg L 1 19200 ± 132 22300 ± 1630 24400 ± 2470 19400 ± 2180 25400 ± 4700 23700 ± 3850 NH3-N mg L 1 4430 ± 74 5320 ± 177 6130 ± 117 4030 ± 64 5520 ± 168 6590 ± 141 TKN mg L 1 5120 ± 170 6380 ± 209 7490 ± 198 4360 ± 261 7390 ± 64 8850 ± 356 PO4-P mg L 1 161 ± 11 130 ± 21 142 ± 24 98 ± 10 96 ± 4 107 ± 16 TP mg L 1 376 ± 82 332 ± 37 461 ± 14 416 ± 82 607 ± 4 477 ± 22

1 *Sample taken immediately after SBP/CaO2 addition. TP ¼ total phosphorus. * volatility ¼ VS TS 100%. Results presented as value ± standard deviation.

The GHGs, CO2 and CH4, also showed differences between pilot- 4.3. SBP treatment cost and effectiveness compared to other and farm-scale. The pilot-scale (Maurer et al., 2017a) resulted in treatments in the literature significant increases in CO2 and non-significant increases in CH4, most likely due to the oxidation byproducts of VOCs as a result of Overall, the cost estimate of the SBP/CaO2 TRT ($1.45 per mar- the SBP/CaO2 application, while at the farm-scale significant in- keted pig, 0.8% price of a marketed pig; material only) was at the creases were not observed. This observation of CO2 and CH4 not lower range of prices reported for comparable products which had increasing on the farm-scale was most likely due to the contribu- a mean cost of $4.28 ± $5.80 (ranging from $0.01 to $18.2 per tion of other sources (e.g., pig respiration, flatulence, eructation, marketed pig for materials only) (Table 3). The cost of CaO2 catalyst feed and barn heating) that were present on the farm-scale which was ~60% of materials cost. The cost of soybean hulls was $0.60 per might have ‘overwhelmed’ any possible increases in GHG emissions marketed pig, i.e., only 40% of materials cost. Labor cost is highly from the manure. variable and in U.S. Midwest conditions the total cost estimate with labor included is $2.62 per marketed pig (1.5% price of marketed pig). 4.2. SBP modes of action The cost of SBP/CaO2 treatment was based on the following: $191 metric ton 1 for SBP based on bulk purchase of soybean hulls There are several modes of action (physical, chemical and bio- from a Midwest-based co-op market (February-2014 availability logical) in which SBP/CaO2 application could be contributing to pricing, mean of 3 estimates from Iowa, Missouri and Minnesota; 1 reduce emissions from swine manure. The physical mode of action University of Missouri Extension, 2014); $6.48 kg for CaO2 based of the SBP/CaCO2 was the act of covering the manure surface, on a bulk purchase of 900 kg (American International Chemical, Inc. creating a permeable cover or crust interfering with the mass Framingham, MA); and $6.00 1000 kg 1 for grinding of bulk soy- transfer of gases from the manure to the headspace. This is similar beans. The summed material cost was $1.45 per marketed pig, ® to the use of Leca balls as demonstrated by Balsari et al. (2006). equivalent to 0.8% price of a marketed pig (2010e2015 average) Once the SBP/CaO2 was saturated and incorporated into the (Iowa State University Extension and Outreach, 2016b) at the SBP manure, it possibly facilitated the formation of a crust on the dose of 2.28 kg m 2 (Maurer et al., 2017a). manure in less-disturbed areas of the pit. The crust can be also a Labor cost involved in mixing and application could vary greatly source of biological activity where bacterial populations utilizing depending on method of application and local labor costs. Based on gases emitted from manure develop. the application approach of mixing and adding the TRT through the Govere et al. (2007) and Ye et al. (2009) discuss an increase in pH slats by hand, the labor cost (2014 price benchmarking) was $1.17 following the application of horseradish peroxidase and CaO2 to per marketed pig, 0.7% price of a marketed pig (8 man-hours and 1 swine slurry due to CaO2. A rise in pH of the manure was not $13 h ). Thus, the estimated total cost (material plus labor) was observed in this study due to the CaO2 in the SBP/CaO2 TRT, in fact $2.62 per marketed pig, which is about 1.5% of the price of a mar- the SBP/CaO2 application reduced the pH of the manure. This keted pig. Larger commercial facilities could consider using reduction of the pH of the manure is likely due to the oxidation of external-to-barn pit access wells for a premixed SPB/CaO2 appli- organic compounds (Dec et al., 2007) resulting from the SBP/CaO2 cation using a powered spreader. Eliminating the need to apply TRT outweighing the initial pH increase of the CaO2 reacting with additives through the slats inside the barn could possibly lower the H2O. The reduction in pH resulting from SBP/CaO2 also was application cost, and would lower biosecurity risks as well, an observed at the pilot-scale (Maurer et al., 2017a) with the trend of important consideration for the swine industry. Using the same 1 1 higher SBP/CaO2 doses resulting lower pH of the manure. This pH assumptions, the estimated cost was $2.19 pig space year of the 1 1 reduction is a possible mode of action in reducing NH3 emissions (material only) and $3.95 pig space year when the cost of labor (Ottosen et al., 2009; Dai and Blanes-Vidal, 2013; Kai et al., 2008). was included. Similarly, the estimated cost per barn area was 2 1 2 1 The last possible (chemical) mode of action of the SBP/CaO2 $2.96 m year (material only) and $5.34 m year when the application is enzymatic oxidation. Enzymatic oxidation has been cost of labor was included (2014 price benchmarking). shown to polymerize phenolic and indolic compounds (Tonegawa The SBP/CaO2 TRT resulted in a more comprehensive mitigation et al., 2003) and to a lesser extent VFAs (Govere et al., 2005). The of greater number of gases of concern for swine industry. Heber enzymatic oxidation mode renders VOCs less volatile and/or less et al. (2000) studied the effects of Alliance treatment developed offensively odorous. The CaO2 may also react and oxidize com- by Monsanto during a 6 month study. Alliance treatment, a pro- pounds in the manure slurry. In summary, there is still need to prietary mix of surfactants, neodol, glyoxal, copper sulfate, and conduct in-depth studies aiming to understand the actual com- water was sprayed onto manure pit surface every 4 h for 4 min at a plementary/synergistic effects of aforementioned modes of SBP/ time, resulting in 20% dilution of fresh manure. Heber et al. (2000) CaO2 TRT. Table 3 Farm-Scale e comparison of estimated % emissions reduction and material cost of treatment (labor not figured in).

Reference Scale Additive Gaseous Emissions Reduction (%) Additive Cost ($) 2 Mode of NH3 H2SCH4 CO2 N2O DMDS/ n-Butyric Valeric Isovaleric p- Phenol Indole Skatole Per per m per per pig space a Action DMTS/MT Acid Acid Acid Cresol marketed year per year pig

Manure Storage and Handling Dai and Blanes- Lab Sulfuric Acid C 77 NS NM NS NM NM NM NM NM NM NM NM NM NA NA NA Vidal, 2013 Whitehead et al., Lab Tannins B NM 90 95 NM NM NM NM NM NM NM NM NM NM NA NA NA 2013 ® b Shah and Kolar, Pilot ManureMax B NM NM 34 NS NS NM 100 NM NM NM NM NA NA NA

2012 c Parker et al., 2012 Lab SBP/CaO2 C NMNMNMNMNMNM NS 92 NA 69 51 NA NA NA © Rahman et al., 2011 Farm Digest3þ3 BNSNSNMNMNMNMNMNMNMNMNMNMNMNANANA Nykanen et al., 2010 Farm Carbohydrate and Bacterial B NM S NM NM NM S NM NM NM NM NM NM NM NA NA NA 467 (2017) 166 Environment Atmospheric / al. et Maurer D.L. d d d d Ye et al., 2009 Pilot HRP/CaO2 C NMNMNMNMNMNM 38 93 46 100 NA NA NA NA NA NA Banhazi et al., 2009 Farm WonderTreat™ BNSNSNMNMNMNMNMNMNMNMNMNMNMNANANA Ottosen et al., 2009 Farm Sulfuric Acid C 100 NM >50 NM NM NM NM NM NM NM NM NM NM NA NA NA Kai et al., 2008 Farm Sulfuric Acid C >90 NM NM NM NM NM NM NM NM NM NM NM NM NA NA NA Predicala et al., Pilot Na-nitrile and Na- B/C NS >92 NM NM NM NM NM NM NM NM NM NM NM NA NA NA 2008 molybdate e f Govere et al., 2007 Pilot HRP/CaO2 C NMNMNMNMNMNM 22 100 NM NA NA NA Lee et al., 2007 Lab Aqueous Foams B/C 88 70 NM NM NM NM NM NM NM NM NM NM NM NA NA NA ® Balsari et al., 2006 Pilot LecaBalls P 80 NM NM NM NM NM NM NM NM NM NM NM NM 0.01 g 0.06h 0.05i Huang et al., 2006 j Lab L. plantarum and B9249 NM NM NM NM NS 47 24 NM NM NM NM NA NA NA carbohydrates Portejoie et al., 2003 Pilot Oil P 93 NM NM NM NM NM NM NM NM NM NM NM NM NA NA NA Portejoie et al., 2003 Pilot Peat P 92 NM NM NM NM NM NM NM NM NM NM NM NM NA NA NA ® ® ® Martinez et al., 2003 Pilot NX 23 , Staloson , Biosuper CNMNM4721 NM NM NM NM NM NM NM NM NM NA NA NA e64 e28 Martinez et al., 2003 Pilot Water P NM NM 36 22 NM NM NM NM NM NM NM NM NM NA NA NA e57 e37 Zhu et al., 1997 Pilot MPC, Biosafe, Shac, X-Stink B/C NS NS NM NM NM NM NM NM NM NM NM NM NM NA NA NA and CPPD e

Livestock Housing 478 2 k i This Study Farm 2.28 kg/m SBP/CaO 2 C/P 22 80 NS NS NS NS 37 48 39 NS NM 31 44 1.45 2.96 2.19 2 k i Maurer et al., 2017a Pilot 2.28 kg/m SBP/CaO 2 C/P NS NS NS 21 NS NS NS NS NS 65 NM NS 73 1.45 2.96 2.19 2 k i Maurer et al., 2017a Pilot 4.57 kg/m SBP/CaO 2 C/P 25 NS 52 27 NS 68 NS NS NS 53 NM NS 63 2.90 5.93 4.39 2 k i Maurer et al., 2017a Pilot 22.8 kg/m SBP/CaO 2 C/P 58 48 107 84 NS 85 NS NS NS 90 NM NS 93 14.53 29.62 21.95 2 k i Maurer et al., 2017a Pilot 45.7 kg/m SBP/CaO 2 C/P 68 144 232 124 NS NS 235 NS 355 78 NM NS 77 29.06 59.33 43.90 Maurer et al., 2017b Pilot Non-activated Biochar P 23 NS 25 NS NS NS NS NS NS NS NM NS NS 8.72 32.41 24.09 Moreno et al., 2010 Pilot Sodium Molybdate B/C NS 89 NM NS NM NM NM NM NM NM NM NM NM 0.45 l 1.98i 1.46m Kai et al., 2008 Farm Sulfuric Acid C 70 NM NM NM NM NM NM NM NM NM NM NM NM NA NA NA Varel and Wells, Farm Thymol and Urease B NM NM 93 NM NM NM NM NM NM NM NM NM NM 17.98 36.58 26.64 2007 Inhibitor Schneegurt et al., Farm Bio-Kat B 75 NM NM NM NM NM NM NM NM NM NM NM NM NA NA NA 2005 Smith et al., 2004 Farm aluminum chloride C 52 NM NM NM NM NM NM NM NM NM NM NM NM NA NA NA j Heber et al., 2001 Pilot AgriKlenz Plus B 6 34 NM NM NM NM NS NS NS NS NS NS NS 0.57 2.14 1.59 Heber et al., 2001j Pilot Alken Clear-Flo B 4 47 NM NM NM NM NS NS NS NS NS NS NS 18.18 67.51 50.18 Heber et al., 2001j Pilot AWL-80 B 10 NS NM NM NM NM NS NS NS NS NS NS NS 0.41 1.52 1.13 (continued on next page) 475 476 D.L. Maurer et al. / Atmospheric Environment 166 (2017) 467e478 . 3 2 reported 24% reduction in NH emissions and no signiﬁcant dif-

US 3 ference in H2S emissions. Moreno et al. (2010) studied the effect of molybdate treatment on H2S, NH3 and CO2 emissions from swine manure due to microbial inhibition. These tests were conducted in two chamber ation using the 1.89 per pig space per year

fl rooms housing 8 pigs each. The sodium molybdate was sprayed and raked over to mix with manure. Gas concentrations were measured per 2 1 $CAD) in 2009 (year of nishing cycles per year. h: during manure agitation on day 28 and 48 post application at the i fi ¼ pit, animal and human breathing zone level. Moreno et al. reported per m year approximate 79e97% reduction in H2S in pit headspace on days 28, and 48 respectively. Mean percent reduction, 89% is reported in Table 3. No statistically significant difference in NH3 and CO2 concentrations was measured. marketed pig Additive Cost ($) Balsari et al. (2006) reported 73e87% reduction of NH3 emissions from open swine slurry storage tank (300 m3, ~80,000 gal capacity) with the use of 0.1 m layer of Leca (extruded clay) floating balls. Four measurement trials, each lasting 6 d per season, were made over 1 year period. The mean of the lowest, wintertime (73%) to highest summertime (87%) NH3 emissions reduction, i.e., 80% is reported in Table 3. Phenol Indole Skatole Per nisher building with a half full 8 ft pit and a stocking density of 0.74 m

ﬁ Heber et al. (2001) reported concentrations of NH3 and H2Sin Methanethiol; DMDS, DMTS and MT cannot be individually distinguished 129 NS NS NS 2.95 10.94 8.13 - ¼

-ethylphenol. g: calculated based on 2.76 swine headspace of 35 manure additives. The tests were done on pilot- p Cresol p

cid. c: VFAs measured in the study were; acetic acid, propionic acid and butyric scale over 42 d with three replications of each, with samples taken over 4 h period on a weekly bases. All additives were added with no pH adjustments or agitation. Concentrations of acetic acid, Isovaleric Acid -cresol and propionic acid, isobutyric acid, butyric acid, isovaleric acid, valeric p acid, phenol, p-cresol, indole and skatole in manure were also Not available; MT -cresol), indolics (indole and skatole) and values are also determined in the manure not the ) per pig space. j: Concentrations were taken from the manure not the headspace for all but NH p ¼ 2 26 NS NS NS NS NS 7.96monitored. 29.56 21.97 Statistically signiﬁcant reductions in headspace con- Valeric Acid

(8 ft centrations were observed for 7 and 8 of the additives for H2S and 2 NH respectively. (8 ft2/pig). l: calculated based on the average exchange rate (0.8704 USD 3 1

Maurer et al. (2017b) reported a reduction in NH3 emissions of -Butyric nishing cycles per year). All previously published results on cost have been adjusted for in n Acid

pig e

ﬁ 13 23% resulting from the use of non-activated biochar at pilot- 2 -valeric acid. f: phenol, Not measured; NA scale over a one-month period. Concentrations of H2S, DMDS/MT, n ¼ DMTS/MT, n-butyric acid, isovaleric acid, valeric acid, p-cresol, indole, skatole, CH4,CO2 and N2O in manure were also monitored.

DMTS/MT Methane emissions were increased as a result of the treatment, O DMDS/ 2 while the remaining monitored gases were not signiﬁcantly be- N

cance level; NM tween the control and treated groups. 2 ﬁ There are several publications reporting swine manure treat- CO ments discussed below however, no cost of treatment was re- 4 ported. Ye et al. (2009) reported 32e54% and 28e41% reduction in indolic compounds and volatile fatty acids (VFAs), respectively and SCH 58 NM70 NM NM NM NM NM NM NM47 NM27 NM NM NS NM NS NM NM NM NM NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS 14.87 NS 8.27 NS 55.24 NS 30.72 41.05 0.16 22.84 0.59 0.44 2 H nearly 100% reduction in p-cresol from 21 L bucket trials on swine 3

2NNNN NS923NMNMNMNM NS NS NSNS NS NS NA NA NA manure treated with surﬁcially applied horseradish peroxidase and NH peroxides. Reductions in volatiles were determined by concentration in the manure over a 72 h period. Portejoie et al. (2003) re- cant difference at 95% signi a ﬁ ported a 93% and 92% reduction in NH3 emissions from 5 kg scale

/pig) was lower than industry standards, i.e., 0.74 m ﬁ 2

1 Euro) in 2006 (year of publication). i: calculated based on 0.74 m trails over 15 d with sur cial application of kitchen oil and peat, C 24 NM NM NM NM NM NM NM NM NM NM NM NM 0.69 2.55 Mode of Action ¼ No signi respectively on swine manure. There were other additives reported ¼

were based on supplier recommended application rates and frequencies and on a modern 2500 head in the literature but the effectiveness and general practicality of (14.6 ft nishing cycles per year (from publication). 1 ﬁ employing these additives was not to the point of real world use, cant; NS pig and so were not included for comparison. ﬁ 2 Signi

¼ 5. Conclusions ® Heber et al., 2001 This study concludes the lab-pilot-farm-scales progression of -caproic acid, isobutyric acid, isocaproic acid, isovaleric acid, propionic acid and n testing for a promising gas emissions mitigation technology, i.e., . Cost for optimizing the effects of time and TRT dose, broadening the array of Pilot Biocharge DryPilot BMTPilot EM WastePilot Treatment GT-2000OC BC-2000AFPilot InhibodorPilot Krystal B B AirPilot B Manure ManagementPilot Plus Peroxy Odor BPilot Control ROEPilot SMOC 15 Pilot NS 7 B ZymPlex 34Farm C C Alliance 37 6 NM C NM NM NM 5 NM NM NM 3 NS NM 36 7 C NM B C NS NS NM NM NS NM NM NM NM NS NM NS NS NS 37 27 NS NS NM NM NS NM NM NS NM NM NM NM NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS 12.15 NS NS 0.94 NS NS NS 45.12 NS NS 3.48 NS NS 33.53 1.78 NS 1.11 2.59 NS 6.61 NS 4.13 4.91 0.82 3.07 3.05 2.27 key gases to be measured, while keeping in mind the practical j j j j j j j j j j j ) application constraints and cost. In this farm-scale study, we -butyric acid,

n observed signiﬁcant reductions in NH3 (21.7%), H2S (79.7%), n- Horseradish peroxidase; S

¼ butyric acid (37.2%), valeric acid (47.7%), isovaleric acid (39.3%), continued (

S. k: The stocking density of 1.35 m indole (31.2%), and skatole (43.5%) emissions. The SBP/CaO TRT

2 2

ation Calculator (2014) had no effect on DMDS/MT, CO2,CH4 and N2O emissions. Emissions Heber et al., 2001 Heber et al., 2001 Heber et al., 2001 Heber et al., 2001 Heber et al., 2001 Heber et al., 2000 Heber et al., 2001 ﬂ Reference Scale Additive Gaseous Emissions Reduction (%) Heber et al., 2001 Heber et al., 2001 Heber et al., 2001 Heber et al., 2001 Heber et al., 2001 ﬁ a: mode of action of theacid. additive, d: C: publication chemical, B: reductions biological, wereheadspace. P: e: in physical. categories: b: VFAs VFAs reported (isobutyric in acid, this study isocaproic were; acid, acetic isovaleric acid, propanoic acid), acid, phenolics butanoic (phenol a and by the TD tube collection methodIn used. Cost for this study was based on SBP application once per cycle (2.76 swine Note: HRP Table 3 and H publication). m: calculated based on 3.25 swine calculated based on the average exchange rate (1.26 USD of p-cresol were reduced, however, without signi cant difference. D.L. Maurer et al. / Atmospheric Environment 166 (2017) 467e478 477

The SBP/CaO2 TRT resulted in a more comprehensive mitigation of Removal of odorants from animal waste using Fenton's reaction, 701P0907cd. key gases of concern for the swine industry compared to other In: Proc. International Symposium on Air Quality and Waste Management for Agriculture. Colorado, USA, Broomfield. studies in published literature. The total TRT cost was equivalent to Devos, M., Patte, F., Rouault, J., Laffort, P., Van Gemert, L.J., 1990. Standardized Hu- 1.5% (0.8% materials and 0.7% labor) of the pig market price, i.e., man Olfactory Thresholds. Oxford University Press, New York. $2.62 per marketed pig ($1.45 pig 1 in materials and $1.17 pig 1 in Govere, E.M., Tonegawa, M., Bruns, M.A., Wheeler, E.F., Heinemann, P.H., Kephart, K.B., Dec, J., 2005. Deodorization of swine manure using minced labor). The cost of CaO2 catalyst was ~60% of materials cost. The cost horseradish roots and peroxides. J. Agric. Food Chem. 53, 4880e4889. of soybean hulls (SBP source) was $0.60 per marketed pig, i.e., only Govere, E.M., Tonegawa, M., Bruns, M.A., Wheeler, E.F., Kephart, K.B., Voigt, J.W., 40% of materials cost. Dec, J., 2007. Using minced horseradish roots and peroxides for the deodorization of swine manure: a pilot scale study. Bioresour. Technol. 98, 1191e1198. Heber, A.J., Ni, J.Q., Lim, T.T., Diehl, C.A., Sutton, A.L., Duggirala, R.K., Haymore, B.L., Disclaimer Kelly, D.T., Adamchuk, V.I., 2000. Effect of a manure additive on ammonia emission from swine finishing buildings. Am. Soc. Agric. Eng. 43, 1895e1902. Heber, A.J., Ni, J., Sutton, A.L., Patterson, J.A., Fakhoury, K., Kelly, D., Shao, P., 2001. USDA is an equal opportunity provider and employer. The Laboratory Testing of Commercial Manure Additives for Swine Odor Control. mention of trade names of commercial products in this article is Final report for the USDA-ARS-NSRIC. Ames, IA, January, 2001. solely for the purpose of providing specific information and does Heber, A.J., Ni, J.Q., Lim, T.T., Tao, P.C., Schmidt, A.M., Koziel, J.A., Beasley, D.B., Hoff, S.J., Nicolai, R.E., Jacobson, L.D., Zhang, Y., 2006. Quality assured mea- not imply recommendation or endorsement by the USDA. surements of animal building emissions: gas concentrations. J. Air & Waste Manage. Assoc. 56, 1472e1483. Acknowledgements Hoff, S.J., Bundy, D.S., Nelson, M.A., Zelle, B.C., Jacobson, L.D., Heber, A.J., Ni, J.Q., Zhang, Y., Koziel, J.A., Beasley, D.B., 2006. Emissions of ammonia, hydrogen sulfide, and odor before, during and after slurry removal from a deep-pit swine The PI and all collaborators would like to thank the National finisher. J. Air & Waste Manage. Assoc. 56, 581e590. Pork Board and affiliated swine producers for supporting this study Huang, C., Li, J., Kang, W.L., Tang, X.Y., 2006. Effect of adding Lactobacillus plantarum via the NPB project # 12e108 ‘Testing soybean peroxidase for swine and soluble carbohydrates to swine manure on odorous compounds, chemical composition and indigenous flora. J. Environ. Sci. 18, 201e206. manure treatment and mitigation of odorous VOCs, ammonia, Iowa State University Extension and Outreach, 2016a. Iowa State University hydrogen sulfide and greenhouse gas emissions’. The Iowa State Extension and Outreach Air Management Practices Assessment Tool (2016) and USDA teams would like to thank the Bio-Resource Products, Website. http://www.agronext.iastate.edu/ampat/homepage.html. (Accessed 16 November 2016). Inc. for in-kind contribution of SBP and expertise. Special thanks to Iowa State University Extension and Outreach, 2016b. Ag Decision Maker- Historic Dr. Kent Kroeger, formerly with Department of Statistics, Iowa State Hog and Lamb Prices. Website. http://www.extension.iastate.edu/agdm/ University, for his help with statistical analyses. livestock/pdf/b2e10.pdf. (Accessed 7 November 2016). Kai, P., Pedersen, P., Jensen, J.E., Hansen, M.N., Sommer, S.G., 2008. A whole-farm assessment of the efficacy of slurry acidification in reducing ammonia emis- Author contributions sions. Eur. J. Agron. 28, 148e154. Lee, S.-R., Han, J.K., Choi, Y.J., Nam, K., 2007. Reduction of ammonia and hydrogen sulfide emission from swine manure using aqueous foams amended with mi- The manuscript was written through contributions of all au- croorganisms and chemical additives. Clean 35, 230e234. thors. All authors have given approval to the final version of the Martinez, J., Guiziou, F., Peu, P., Gueutier, V., 2003. Influence of treatment tech- manuscript. All authors contributed equally. niques for pig slurry on methane emissions during subsequent storage. Biosyst. Eng. 85, 347e354. Maurer, D.L., Koziel, J.A., Harmon, J.D., Hoff, S.J., Rieck-Hinz, A.M., Andersen, D.S., Funding sources 2016. Summary of performance data for technologies to control gaseous, odor, and particulate emissions from livestock operations: air management practices assessment tool (AMPAT). Data Brief 7, 1413e1429. Funded by National Pork Board via the NPB project # 12-108. Maurer, D.L., Koziel, J.A., Bruning, K., Parker, D.B., 2017a. Pilot scale testing of renewable biocatalyst for swine manure treatment and mitigation of odorous Appendix A. Supplementary data VOCs, ammonia and hydrogen sulfide emissions. Atmos. Environ. 150, 313e321. http://dx.doi.org/10.1016/j.atmosenv.2016.11.021. Maurer, D.L., Koziel, J.A., Kalus, K., Andersen, D.S., Opalinski, S., 2017b. Pilot-scale Supplementary data related to this article can be found at http:// testing of non-activated biochar for swine manure treatment and mitigation of dx.doi.org/10.1016/j.atmosenv.2017.07.048. ammonia, hydrogen sulfide, odorous volatile organic compounds (VOCs), and greenhouse gas emissions. Sustainability 9 (929). http://dx.doi.org/10.3390/ su9060929. References Morawski, B., Quan, S., Arnold, F.H., 2001. Functional expression and stabilization of horseradish peroxidase by directed evolution in Saccharomyces cerevisiae. Andersen, K.B., Hansen, M.J., Feilberg, A., 2012. Minimisation of artefact formation Biotechnol. Bioeng. 76, 99e107. of dimethyl disulphide during sampling and analysis of methanethiol in air Moreno, L., Predicala, B., Nemati, M., 2010. Laboratory, semi-pilot and room scale using solid sorbent materials. J. Chrom. A 1245, 24e31. study of nitrite and molybdate mediated control of H2S emission from swine AOAC, 2000. Official Methods of Analysis, seventeenth ed. Assoc. Anal. Chem, manure. Bioresour. Technol. 101, 2141e2151. Arlington, VA. National Oceanic and Atmospheric Administration. (1999) Website https:// APHA, 1998. Standard Methods for the Examination of Water and Wastewater, cameochemicals.noaa.gov/chris/BRA.pdf. (Accessed January 11, 2017). twentieth ed. American Public Health Association, Washington, DC. Nykanen, A.M., Hamalainen, N., Kostia, S., Mikola, J., Romantschuk, M., 2010. Balsari, P., Dinuccio, E., Gioelli, F., 2006. A low cost solution for ammonia emission Reduction of odorants in swine manure by carbohydrate and bacterial abatement from slurry storage. Int. Congr. Ser. 1293, 323e326. amendments. J. Environ. Qual. 39, 678e685. Banhazi, T., Hudson, N., Dunlop, M., Dyson, C., Thomas, R., 2009. Development and Ottosen, L.D.M., Poulsen, H.V., Nielsen, D.A., Finster, K., Nielsen, L.P., Revsbech, N.P., testing of an evaluation procedure for commercial manure additive products. 2009. Observations on microbial activity in acidified pig slurry. Biosyst. Eng. Biosyst. Eng. 103, 321e328. 102, 291e297. Cai, L., Koziel, J.A., Nguyen, A.T., Liang, Y., Xin, H., 2007. Evaluation of zeolite for Parker, D.B., Pandrangi, S., Greene, L.W., Almas, L.K., Cole, N.A., Rhoades, M.B., control of odorants emissions from simulated poultry manure storage. Koziel, J.A., 2005. Rate and frequency of urease inhibitor application for mini- J. Environ. Qual. 36, 184e193. mizing ammonia emissions from beef cattle feedyards. Trans. ASABE 48, Cai, L., Koziel, J.A., Zhang, S., Heber, A.J., Cortus, E.L., Parker, D.B., Hoff, G.S.J., Sun, 787e793. Heathcote, K.Y., Jacobson, L.D., Akdeniz, N., Hetchler, B.P., Bereznicki, S.D., Parker, D.B., Cai, L., Kim, K.-H., Hales, K.E., Spiehs, M.J., Woodbury, B.L., Caraway, E.A., Lim, T.T., 2015. Odor and odorous chemical emissions from ani- Nickerson, K.W., Patefield, K.D., 2012. Reducing odorous VOC emissions from mal buildings: Part 3 e chemical emissions. Trans. ASABE 58, 1333e1347. swine manure using soybean peroxidase and peroxides. Bioresour. Technol. 124, Chen, L., Hoff, S., Cai, L., Koziel, J.A., Zelle, B., 2009. Evaluation of wood chip-based 95e104. biofilters to reduce odor, hydrogen sulfide, and ammonia from swine barn Parker, D.B., Hayes, M., Brown-Brandl, T., Woodbury, B.L., Spiehs, M.J., Koziel, J.A., ventilation air. J. Air & Waste Manage Assoc. 59, 520e530. 2016a. Surface application of soybean peroxidase and calcium peroxide for Dai, X.R., Blanes-Vidal, V., 2013. Emissions of ammonia, carbon dioxide, and reducing odorous VOC emissions from swine manure slurry. Appl. Eng. Agric. hydrogen sulfide from swine wastewater during and after acidification treat- 32, 389e398. ment: rffect of pH, mixing and aeration. J. Environ. Mange. 115, 147e154. Parker, D.B., Rhoades, M.B., Baek, B.H., Koziel, J.A., Waldrip, H.M., Todd, R.W., 2016b. Dec, J., Bruns, M.A., Cai, L., Koziel, J.A., Snyder, E.M., Kephart, K.B., Watson, J.E., 2007. Urease inhibitor for reducing ammonia emissions from an open-lot beef cattle 478 D.L. Maurer et al. / Atmospheric Environment 166 (2017) 467e478

feedyard in the Texas High Plains. Appl. Engr. Agric. 32, 823e832. http://dx.doi. Sun, H., Trabue, S., Scoggin, K., Jackson, W., Pan, Y., Zhao, Y., Malkina, I.L., Koziel, J.A., 10.13031/aea.32.11897. Mitloehner, F., 2008a. Alcohol, volatile fatty acid, phenol, and methane emis- Philippe, F.-X., Nicks, B., 2015. Review on greenhouse gas emissions from pig sions from dairy cows and fresh waste. J. Environ. Qual. 37, 615e622. houses: production of carbon dioxide, methane and nitrous oxide by animals Sun, G., Guo, H., Peterson, J., Predicala, B., Lague, C., 2008b. Diurnal odor, ammonia, and manure. Agr. Ecosyst. Environ. 199, 10e25. hydrogen sulfide, and carbon dioxide emission profiles of confined swine Portejoie, S., Martinez, J., Guiziou, F., Coste, C.M., 2003. Effect of covering pig slurry grower/finisher rooms. J. Air & Waste Manage. Assoc. 58, 1434e1448. stores on the ammonia emission processes. Bioresour. Technol. 87, 199e207. Tonegawa, M., Dec, J., Bollag, J.-M., 2003. Use of additives to enhance the removal of Predicala, B., Nemati, M., Stade, S., Lague, C., 2008. Control of H2S emissions from phenols from water treated with horseradish and hydrogen peroxide. J. Environ. swine manure using Na-nitrite and Na-molybdate. J. Hazard. Mater. 154, Qual. 32, 1222e1227. 300e309. University of Missouri Extension, 2014. University of Missouri Extension, By- Rahman, S., DeSutter, T., Zhang, Q., 2011. Efficacy of microbial additive in reducing product Feed Price Listing (2014) Website. http://agebb.missouri.edu/dairy/ odor, ammonia, and hydrogen sulfide emissions from farrowing-gestation byprod/bplist.asp. (Accessed 6 February 2014). swine operation. Agric. Eng. Int. CIGR J. 13, 1e9. US Inflation Calculator. 2014. Website http://www.usinflationcalculator.com/ Rockafellow, E.M., Koziel, J.A., Jenks, W.S., 2012. UV treatment of ammonia for (Accessed February 6, 2014). livestock and poultry barn exhaust applications. J. Environ. Qual. 41, 281e288. Van Huffel, K., Hansen, M.J., Feilberg, A., Liu, D., Van Langenhove, H., 2016. Level and Schneegurt, M.A., Weber, D.L., Ewing, S., Schur, H.B., 2005. Evaluating biostimulant distribution of odorous compounds in pig exhaust air from combined room and effects in swine production facility wastewater. In: Proc. State of the Science of pit ventilation. Agric. Ecos. Envir. 218, 209e219. Animal Manure and Waste Management Symposium, 1e6. Tex.: National Center Varel, V.H., Wells, J.E., 2007. Influence of thymol and a urease inhibitor on coliform for Manure and Animal Waste Management, San Antonio. bacteria, odor, urea, and methane from a swine production manure pit. Shah, S.B., Kolar, P., 2012. Evaluation of additive for reducing gaseous emissions J. Environ. Qual. 36, 773e779. from swine waste. Agric. Eng. Int. CIGR J. 14, 10e20. Whitehead, T.R., Spence, C., Cotta, M.A., 2013. Inhibition of hydrogen sulfide, Smeets, M.A.M., Bulsing, P.J., van Rooden, S., Steinmann, R., de Ru, J.A., methane, and total gas production and sulfate-reducing bacteria in in vitro Ogink, N.W.M., van Thriel, C., Dalton, P.H., 2007. Odor and irritation thresholds swine manure by tannins, with focus on condensed quebracho tannins. Appl. for ammonia: a comparison between static and dynamic olfactometry. Chem. Microbiol. Biotechnol. 97, 8403e8409. Senses 32, 11e20. Ye, F.X., Zhu, R.F., Li, Y., 2009. Deodorization of swine manure slurry using horse- Smith, D.R., Moore Jr., P.A., Haggard, B.E., Maxwell, C.V., Daniel, T.C., radish peroxidase and peroxides. J. Hazard. Mater. 167, 148e153. VanDevander, K., Davis, M.E., 2004. Effect of aluminum chloride and dietary Zhang, S., Cai, L., Koziel, J.A., Hoff, S.J., Schmidt, D.R., Clanton, C.J., Jacobson, L.D., phytase on relative ammonia losses from swine manure. J. Anim. Sci. 82, Parker, D.B., Heber, A.J., 2012. Field air sampling and simultaneous chemical and 605e611. sensory analysis of livestock odorants with sorbent tubes and GC-MS/olfac- Steevensz, A., Cordova Villegas, L.G., Feng, W., Taylor, K.E., Bewtra, J.K., Biswas, N., tometry. Sens. Actuators, B 146, 427e432. 2014. Soybean peroxidase for industrial wastewater treatment: a mini review. Zhu, J., Bundy, D.S., Li, X., Rashid, N., 1997. Controlling odor and volatile substances J. Env. Eng. Sci. 9 (3), 181e186. in liquid hog manure by amendment. J. Environ. Qual. 26, 740e743.