Can biochar covers reduce emissions from manure lagoons while capturing nutrients?

Abstract

The unique physical and chemical properties of biochars make them promising materials for odor, gas, and nutrient sorption. Floating covers made from organic materials (biocovers) are one option for reducing odor and gas emissions from livestock manure lagoons. This study evaluated the potential of floating biochar covers to reduce odor and gas emissions while simultaneously sorbing nutrients from liquid dairy manure. This new approach has the potential to mitigate multiple environmental problems. Two biochars were tested: one made via gasification of Douglas-fir chips at 650°C (FC650), and the other made from a mixture of Douglas-fir bark and center wood pyrolyzed at 600°C (HF600). The HF600 biocover reduced mean headspace ammonia concentration by 72 to 80%. No significant reduction was found with the FC650 biocover. Nutrient uptake ranged from 0.21 to 4.88 mg N g⁻¹ biochar and 0.64 to 2.70 mg P g⁻¹ biochar for the HF600 and FC650 biochars, respectively. Potassium ranged from a loss of 4.52 to a gain of 2.65 mg g⁻¹ biochar for the FC650 and HF600 biochars, respectively. The biochars also sorbed Ca, Mg, Na, Fe, Al, and Si. In a separate sensory evaluation, judges assessed odor offensiveness and odor threshold of five biocover treatments including four biochars applied over dairy manure. Reductions in mean odor offensiveness and mean odor threshold were observed in three treatments compared to the control. These results show that biochar covers hold promise as an effective practice for reducing odor and gas emissions while sorbing nutrients from liquid dairy manure.

Introduction

The increasing prevalence of confined livestock production has led to an increase in the use of liquid manure (slurry) storage structures. These structures enable manure to be stored and subsequently spread when conditions are favorable or nutrients are needed for crop growth. However, lagoons can be sources of gaseous emissions during manure storage that are malodorous (Blanes-Vidal et al., 2009) and radiatively active, including greenhouse gases such as methane, carbon dioxide, and nitrous oxide (Leytem et al., 2011). Strong odors from livestock operations are commonly associated with volatile compounds including ammonia (NH₃), sulfides, fatty acids, phenols and indoles (Hobbs et al., 2000). Livestock manure is a particularly significant source of odorous NH₃ emissions (Vaddella et al., 2011). Ammonia and sulfide-containing compounds from livestock manure can contribute to ground-level air pollution (National Research Council (U.S.), 2003). Emissions of NH₃ can also contribute to eutrophication of water via deposition and have other detrimental impacts on ecosystems (Camargo and Alonso, 2006). Additionally, land application of the slurry can be difficult to synchronize with weather patterns and can lead to inadvertent nutrient runoff and leaching when manure application exceeds the needs of crops (Carpenter et al., 1998). Regulations regarding odor emissions and field application of manure are likely to make it increasingly difficult to manage livestock manure in this manner in the future (NRC, 2003).

One practice that has proven effective in reducing gas emissions is the application of a floating cover on the surface of the manure slurry. Various biologically-based materials (biocovers) have been tested for this purpose (Clanton et al., 1999; Guarino et al., 2006; Regmi et al., 2007; Hudson et al., 2008). Straw, vegetable oil, corn stover, and wood chip biocovers have shown variable effectiveness for reducing odor, ammonia (NH₃) and hydrogen sulfide (H₂S) emissions (VanderZaag et al., 2008). While somewhat effective for reducing odor and gas emissions, biocovers have not been widely used. In many cases their use is not justified due to the expense and difficulty of application and maintenance of the cover. More efficient, multi-purpose and multi-benefit biocover materials are needed to address the problems associated with slurry storage.

One material that has the potential to offer multiple benefits is biochar. Biochars are the result of thermal alteration of biomass in the absence of oxygen in a process known as pyrolysis. The process results in the loss of volatile organic compounds, leaving behind a carbon-rich material similar to charcoal. Thermally altered carbon compounds are not a preferred substrate for decomposer organisms and may hence persist for long periods of time, meaning that a portion of the carbon in biochars can be considered sequestered carbon if returned to the soil (Lehmann et al., 2015). In addition to sequestering carbon, biochars can improve soil productivity, particularly for degraded and lower quality soils (Lehmann and Joseph, 2009). Biochars have unique chemical and physical properties including a highly porous structure at scales ranging from nanometers to millimeters, leading to large surface areas that may exceed 100 m² g^-1. This makes them suitable for a range of environmental applications including odor, gas, and nutrient sorption (Xie et al., 2015).

The application of biochar as a biocover on liquid manure storage structures has the potential to provide multiple environmental benefits. These include reduced odor and gas emissions along with sorption of valuable nutrients. When applied to soil, the nutrient-enriched biochar may serve as a fertilizer source for plants while improving soils and sequestering carbon. The high porosity, low bulk density, and nutrient retention capabilities of biochars make them promising candidates for use as a biocover. Biochar has been shown to be effective in adsorbing and retaining nutrients from dairy lagoon effluent (Streubel et al., 2012; Sarkhot et al., 2013). Research has further demonstrated that a biochar enriched with dairy manure effluent reduced greenhouse gas emissions from the soil surface and increased C and N storage when applied to soil (Sarkhot et al., 2012). It has also been proposed that biochars can be used to capture excess nutrients from dairy wastewater to create a sustainable nutrient recycling loop (Ghezzehei et al., 2014).

In order to justify the use of biochar as a lagoon cover, it must offer advantages over traditional cover materials. Demonstrating the efficacy of floating biochar covers is necessary for proof-of-concept of this technology. Consequently, the objectives of this study were to determine the effectiveness of floating biochar covers for (i) reducing NH₃ and H₂S emissions, (ii) adsorbing nutrients, and (iii) mitigating odor from dairy manure slurry.

Conceptual approach

To achieve objectives (i) and (ii), concentrations of NH₃ and H₂S emitted from floating biochar and straw covers were measured over an eight week period. The covers were then removed and analyzed to determine nutrient uptake. We will refer to this as the “adsorption experiment”. To achieve objective (iii), a panel of judges evaluated odors emanating from barrels of manure covered with a variety of floating biochar and straw covers over a 12 week period. This will be referred to as the “odor study”.

Materials and Methods

Manure collection and properties

Two different dairy manure slurries were used for the adsorption study. Manure 1 (M1) was a mixture of fresh manure and barn flush water retrieved from a collection tank at the Oregon State University Dairy Research Center near Corvallis, OR. Manure 2 (M2) was fresh manure scraped to the end of a freestall barn alley at a dairy near Gervais, OR. The manure was scooped into twelve 26.5 L polyethylene pails at each site. For the odor study, Manure 3 (M3) was pumped into six 210 L plastic barrels from the same location as M1. Subsamples of manure were drawn from each pail for M1 and M2, and each barrel for M3 and combined into single samples for analysis. The composited samples were frozen and express mailed to Dairyland Laboratories, Arcadia, WI for analysis. Nitrogen was measured via combustion and detection of thermal conductivity. All other elements were measured via microwave digestion with 50% HNO₃ solution and subsequent analysis with Inductively-Coupled Plasma (ICP) mass spectrometry. Results are shown in Table 1.

Table 1.

Elemental analysis of manures used in the adsorption experiment (Manures 1 and 2) and the odor study (Manure 3)

	Manure 1 (M1)†	Manure 2 (M2)†	Manure 3 (M3)†
Solids (% mass)	0.64	5.39	0.64
Element	_____________________ mg L−1 of manure ___________________
N	300	1200	600
P	200	500	100
K	900	1200	1300
Ca	500	3700	600
Mg	200	600	200
S	100	300	100
Na	180	336	210
Fe	17	237	33
Al	7	183	24
Zn	4	18	22
Mn	5	16	5
Cu	3	14	1
B	1	3	1

^†n = 1 for each manure

Biocover Source and Analysis

All biochars were acquired from Bio-Logical Carbon, LLC in Philomath, OR. Two biochars were used in the adsorption study. The first was a Douglas-fir chip biochar produced in a downdraft gasifier with a peak temperature of approximately 650°C (FC650). The second was made via slow pyrolysis in a retort at approximately 600°C (HF600) from an unprocessed mix of coarse chips of Douglas-fir bark and wood fiber, commonly referred to as “hog fuel”. Four biochars were used for the odor study. Two were Douglas-fir hog fuel biochars made via slow pyrolysis at 550°C (HF550) and 350°C (HF350) respectively. The third was a pine chip biochar made via slow pyrolysis at 500°C (PC500). The fourth was a Douglas-fir hog fuel biochar produced via rapid (seconds to a few minutes) combustion in an industrial cogeneration facility at 1200°C (HF1200). The wheat straw (Straw) for both experiments was acquired from the Oregon State University Dairy Research Center in Corvallis, OR. The cover materials were selected based on local feedstock availability.

Proximate analysis was performed on three replicates of each biochar to determine ash, volatile matter, and fixed carbon content on a dry basis per the American Society of Testing and Materials (ASTM) D1762–84 method (ASTM, 2007). Electrical conductivity (EC) and pH were determined by preparing suspensions of biochar particles (<250μm) in water (1% w/w). The suspensions were heated to 90°C and stirred for 20 minutes, then cooled to room temperature. Electrical conductivity was measured with an EC4083 meter (Amber Science Inc., Eugene, OR) and pH was measured with a PT-15 meter (Sartorius Corp., Bohemia, NY). The results are shown in Table 2.

Table 2.

Characteristics of biochars used in the adsorption experiment (FC650 and HF600) and the odor study (HF1200, HF550, HF350, and PC500)

	__Adsorption experiment__		______________ Odor study _______________
Property	(FC650)	(HF600)	(HF1200)	(HF550)	(HF350)	(PC500)
Moisture, mg g−1	145.8 ± 7.38	81.9 ± 0.63	128.2 ± 1.42	91.7 ± 3.57	138.6 ± 3.85	163.7 ± 11.66
Volatile matter, mg g−1	72.4 ± 1.45	113.3 ± 4.39	103.3 ± 10.77	86.9 ± 7.76	162.2 ± 7.95	112.6 ± 3.84
Ash, mg g−1	86.9 ± 19.30	6.8 ± 4.01	71.5 ± 5.31	71.4 ± 5.91	119.7 ± 4.45	17.8 ± 1.90
Fixed C, mg g−1	840.6 ± 20.74	879.9 ± 2.99	825.2 ± 7.33	841.8 ± 12.48	718.1 ± 11.35	869.6 ± 4.49
pH†	9.32 ± 0.04	7.28 ± 0.03	8.53 ± 0.02	8.65 ± 0.04	8.45 ± 0.02	8.19 ± 0.02
EC, μS cm−1†	260.9 ±0.52	42.4 ±0.12	790 ± 10	111.2 ±0.50	107.5 ±0.86	58.7 ±0.09

All values are means (n=3) ± 1 SE. Volatile matter, ash, and fixed carbon values are given on a dry basis.

^†Reported values are for 1% w/w biochar/water solutions

Adsorption Experiment

The adsorption experiment was conducted in a temperature-regulated greenhouse on the Oregon State University campus in Corvallis, OR. Three cover treatments (FC650, HF600, and Straw) were tested on two manures (M1 and M2). Each treatment was replicated three times. For each replicate, 17 L of slurry was placed in a 26.5 L polyethylene pail model S-16969 (Uline, Pleasant Prairie, WI). Biocovers were applied on the manure surface to a thickness of 5 cm one day prior to initiation of sampling. These were compared to a control (Control) with no cover. A gas sampling apparatus was constructed from a threadable airtight lid model S-17945W (Uline, Pleasant Prairie, WI). Holes were drilled on opposite sides of the lid and fitted with 4.75 mm I.D. PVC tubing, which was joined with a tee to create a single gas sampling port.

The containers were sealed with the lid once every 14 days to enable sampling of gases in the headspace above the slurry. A 70 mm dia. battery powered fan mounted on the underside of the lid was run for 2 minutes to homogenize the headspace air prior to drawing gas samples. Concentrations of NH₃ and H₂S were measured with Colorimetric gas analysis tubes (RAE Systems Inc., San Jose, CA. A single gas sample (50 mL volume drawn over 45 s) was drawn from each replicate on days 0, 14, 28, 42, and 56, with air temperatures during sampling of 27, 19, 23, 20, and 26°C, respectively. Additional aged slurry was added weekly as needed to compensate for evaporative losses and maintain a consistent headspace volume of 8 L. The slurry was slowly poured down the sidewall of the pails to minimize cover disturbance.

The floating covers were skimmed from the surface of each pail on day 56. The cover material was spread on a 1.5 mm mesh screen and allowed to drain for 30 minutes, then frozen. For analysis, three samples from each replicate were thawed and oven dried at 105°C for 24 hours. For each sample, 3.0 g of material was combined with 100 mL of deionized water in an Erlenmeyer flask and placed in an orbital shaker at 120 rpm for 60 minutes to leach the easily extractable nutrients. The resulting mixture was poured through a 630 μm screen. The solid portion remaining on the screen was oven dried at 105°C for 24 hours, then ground and sieved through a 250 μm screen. The N content of the dried cover material (three replicates) was determined with a CNS-2000 Macro Analyzer (Leco Corp., St. Joseph, MI). Phosphorus, K, Ca, Mg, Na, Fe, Al, Mn, and Si were determined via ICP mass spectrometry using an Optima 2100DV (PerkinElmer, Waltham, MA). To release elements for subsequent ICP analysis, the dried cover material was digested using a mix of 3.0 mL each of HNO₃ (70%), HCl (37%), and HF (48%) added to 0.30 g of <250 μm dried sample. This mixture was heated for 25 min in a MDS2000 microwave digester (CEM Corp., Matthews, NC). Deionized water was used to dilute 2.0 mL of the digestate to 40 mL prior to analysis. The same procedure was used to analyze unused cover materials for comparison.

Odor study

The odor study was conducted at the Oregon State University Dairy Research Center near Corvallis, OR. Six 210 L plastic barrels (Uline model s-9945, Pleasant Prairie, WI) were filled with 70 L of dairy manure slurry (M3) each on Oct. 14, 2015. Five barrels received 10 cm thick biocovers consisting of HF1200, HF550, HF350, PC500, and Straw respectively. The sixth barrel received no biocover and served as the control (Control). A seventh barrel was left empty (Empty) to serve as a standard for measuring background odor. Odor was judged by a panel of 5 to 10 judges once every two weeks over a 12-week period. To help prevent odor fatigue (Riskowski et al., 1991; Sheffield and Ndegwa, 2008) the treatments were not replicated. Judges were recruited and selected per Oregon State University Institutional Review Board rules and procedures. Judges used a variation of the compost odor wheel presented by Suffet et al. to evaluate the character of the odor (Suffet et al., 2009). Methods described by Sheffield and Ndegwa (2008) were used to standardize odor evaluation. The offensiveness of the odor was ranked from −10 (extremely unpleasant) to +10 (extremely pleasant). Odor threshold was ranked from 1 to 10, with 10 being the worst smell the judges had ever experienced. Judges donned an odor blocking mask upon entering the site to prevent odor desensitization. The mask was removed only during evaluation of each barrel. The barrels were covered with muslin cloth one hour prior to each judging session. This eliminated bias from the judges based upon the visible appearance of the biocovers. Manure slurry was removed by inserting a siphon pump down the sidewall of the barrel as needed to compensate for rainfall and maintain a consistent slurry depth.

Statistical Analyses

All experiments were evaluated using OriginPro software version 9.3. Mean gas concentrations from the Control were compared to individual biocover-manure combinations using unpaired two-sample t-tests (unequal variance, p = 0.05). Odor offensiveness and odor threshold mean judge response and mean nutrient concentrations in the biocovers pre vs. post experiment for all biocover-manure combinations were compared using ANOVA. The Tukey test was used to adjust for multiple comparisons.

Results

Adsorption Experiment

The FC650, HF600, and Straw biocovers all maintained surface coverage for the eight-week duration of the study. Differences in evaporative losses between treatments were not significant at p = 0.05. Concentrations of H₂S in the headspace were below the detectable level of 0.2 μL L⁻¹ for most replicates on every sampling date. Due to lack of measurable H₂S concentrations no results are reported here. Figure 1 shows the pattern of NH₃ concentrations measured in the headspace above M1 (Fig 1a) and M2 (Fig 1b). The concentration of NH₃ in the pail headspace declined over time with the control, straw, and FC650 covers. The HF600 cover showed an increase at day 14. The same trend was observed for all replicates on both manures. All replicates had higher NH₃ concentrations through day 28 on M2 compared to the more dilute M1. Ammonia concentrations above the FC650 cover were higher than the Control on day 0 for both manures.

Figure 1.

Change in ammonia (NH₃) concentration (μL/L) in the headspace above Manure 1 (a) and Manure 2 (b) over 56 days for the FC650, Control, Straw, and HF600. Error bars show ± 1 SE.

Figure 2 shows the mean headspace NH₃ concentration over the 56-day trial. Mean concentrations were higher above all covers on M2 compared to M1. Mean concentrations above M1 were 9.4, 9.3, 2.8, and 1.9 μL L⁻¹ above the Control, FC650, Straw, and HF600 replicates respectively. Mean concentrations above M2 were 18.9, 16.7, 8.9, and 5.3 μL L⁻¹ above the Control, FC650, Straw, and HF600 replicates respectively. A majority of the variation in cover performance was observed on days 0 and 14. The Straw and HF600 covers resulted in statistically significant reductions in mean NH₃ concentration above both manures. The Straw cover reduced mean NH₃ concentration by 70% and 53% on M1 and M2 respectively. The HF600 cover reduced mean NH₃ concentration by 80% and 72% on M1 and M2 respectively. The FC650 cover did not significantly reduce NH₃ concentrations compared to the Control due to the increase in NH₃ observed at the onset of the study. The difference between HF600 and straw covers was not significant at p = 0.05. Results indicate that the HF600 and Straw covers were effective for reducing headspace NH₃ concentrations over the 56 day experiment, whereas the FC650 cover was not.

Figure 2.

Comparison of mean headspace ammonia (NH₃) concentration above the Control, FC650, Straw, and HF600 treatments. Asterisk denotes significant reduction compared to the control at p=0.05. Error bars show ± 1 SE.

Elemental analysis indicates that the FC650, HF600, and Straw biocovers all sorbed nutrients to some degree while floating on the manure surface. To determine nutrient sorption, the elemental concentrations of the covers before vs. after the experiment were compared. It should be noted that some volatilization of N may have occurred when the samples were dried at 105°C prior to analysis. Table 3 shows the concentrations pre-experiment.

Table 3.

Elemental analysis of fir chip (FC650), hog fuel (HF600), and straw (S) biocovers prior to application on the manure surface.

Element	FC650	HF600	Straw
	_____________ mg g−1 of biocover ____________
N	2.16	3.30	2.36
P	0.57 ± 0.03	0.05 ± 0.04	0.32 ± 0.03
K	9.74 ± 0.30	0.02 ± 0.02	1.64 ± 0.30
Ca	4.04 ± 0.08	0.87 ± 0.25	2.33 ± 0.14
Mg	0.56 ± 0.03	0.03 ± 0.02	0.39 ± 0.03
Na	0.41 ± 0.01	0.33 ± 0.01	0.36 ± 0.01
Fe	1.00 ± 0.03	0.60 ± 0.06	‡
Al	0.10 ± 0.01	0.04 ± 0.01	‡
Mn	0.15 ± 0.01	‡	‡
Si	3.95 ± 0.07	3.01 ± 0.37	54.01 ± 3.06

Values are means (n=6) ± SE, except for N (n=1)

^‡Concentration is below the detection limit

The values in Table 4 reflect the change in mean nutrient concentrations at the end of the adsorption experiment compared to pre-experiment means. The initial concentration of nutrients in the biochar may affect the ability of the covers to sorb nutrients. The FC650 covers had a high initial K content (9.74 mg g⁻¹) and were not effective at adsorbing K. Pre-experiment Si levels were high (>50 mg g⁻¹) in the Straw replicates and did not change significantly. The initial concentration of nutrients in the manure also appears to be a factor. At the beginning of the study, M2 had a greater concentration of all nutrients than the more diluted M1. Subsequently, all covers floating on M2 sorbed more of every nutrient than those same covers floating on M1.

Table 4.

Elemental uptake by biocovers, expressed as post minus pre-adsorption experiment concentration. Cases of elemental release are underlined for greater clarity.

	N	P	K	Ca	Mg	Na	Fe	Al	Mn	Si
	________________________________ mg g−1 of biocover ______________________________
Manure 1
FC650§	3.66	2.53	−4.52	1.54	1.57	0.61	33.24	0.54	0.34	17.02
		(0.10)	(0.33)	(0.53)	(022)	(0.07)	(4.10)	(0.11)	(0.05)	(1.83)
HF600§	0.21	0.64	2.57	1.48	0.96	0.39	1.21	0.03	‡	2.54
		(0.06)	(0.14)	(0.20)	(0.05)	(0.02)	(0.15)	(0.01)		(0.36)
Straw#	5.45	2.40	0.19	3.03	1.53	0.23	‡	0.01	‡	−3.55
		(0.14)	(0.30)	(0.39)	(0.24)	(0.04)		(0.01)		(3.03)
Manure 2
FC650#	4.88	2.70	−0.81	3.34	2.43	1.40	74.46	1.46	0.62	32.96
		(0.22)	(0.71)	(0.71)	(0.44)	(0.07)	(12.80)	(0.33)	(0.07)	(3.14)
HF600#	1.40	0.82	2.65	2.23	1.10	0.69	5.20	0.15	‡	3.54
		(0.05)	(0.16)	(0.30)	(0.07)	(0.04)	(1.34)	(0.03)		(0.61)
Straw#	11.29	4.39	1.82	3.32	2.24	0.91	0.56	0.53	‡	3.19
		(0.24)	(0.32)	(0.17)	(0.08)	(0.04)	(0.07)	(0.02)		(2.44)

Upper values are means, lower values (in parentheses) are pooled SE

^§For N n = 3 post-experiment, n = 1 pre-experiment. n = 8 for all other elements

^#For N n = 3 post-experiment, n = 1 pre-experiment. n = 9 for all other elements

^‡Concentration is below the detection limit

Statistically significant differences pre vs post adsorption experiment are in italics (t-test with p < 0.05)

The Straw cover was most effective at sorbing N, P, and Ca. The FC650 cover was most effective at sorbing Mg, Na, Fe, Al, Mn, and Si. The HF600 cover was most effective at sorbing K. Levels of K in M1 Straw and M2 FC650 replicates did not change significantly. No significant change was measured in Ca levels in M1 FC650 or Al levels in M1 HF600 replicates. All other replicates had statistically significant increases in mean nutrient concentration at p = 0.05.

Odor Study

Figure 3 shows mean odor offensiveness from the 12 week study. The HF350, HF550, and PC500 biocovers had significant reductions in mean odor offensiveness and mean odor threshold over 12 weeks compared to the control, whereas the HF1200 and Straw biocovers did not. The HF1200 cover sank completely by week 6. The PC500 biochar started sinking at week 6 but maintained some surface coverage through the duration of the trial. Data collection continued after cover submergence. The HF350 and HF550 covers cracked and broke into chunks, leading to inconsistent surface coverage. Straw maintained consistent surface coverage throughout the study.

Figure 3.

Mean odor offensiveness from each treatment over 12 weeks. Odor offensiveness is ranked on a scale of −10 (extremely unpleasant) to +10 (extremely pleasant). Asterisk denotes significant reduction compared to the control at p=0.05. Error bars show ± 1 SE.

A significant portion of the HF550 and HF350 biochars consisted of fine particles, which tended to clump together and adhere to the sides of the barrels. Bulging in the center of the covers became evident as rainwater infiltrated through the surface. This caused the HF550 and HF350 covers to crack and break up over time due to fluctuating effluent levels during rainfall and effluent removal events. Rainfall readily penetrated the HF1200, PC500, and Straw biocovers. They were able to fluctuate up and down with no adhesion to the barrel sides. Little adhesion of manure particles to the biochar itself was noted. The biocovers were removed and examined at the end of the experiment. The HF350 biochar was visibly drier in the interior portion of the cover compared to the HF550 biochar.

Discussion

Adsorption experiment

The reductions in NH₃ concentrations associated with biocovers observed in the adsorption study are in line with results from other studies (VanderZaag et al., 2008). Although the study was not set up to determine precise gas fluxes, the reduction in headspace concentrations suggests that gas emissions would also be reduced. The gas sampling method likely modified the conditions in the headspace above the biocovers relative to ambient air. The reported concentrations may not provide an accurate estimate of actual emission rates. These concentrations are only for comparison of differences between different biocovers under the same conditions. However, it is reasonable to assume that the biocovers would function similarly when not being sampled, thus periodic measurements of headspace concentration would show emission trends.

The initial spike in NH₃ concentration with the FC650 covers was unexpected. The mechanism leading to this initial increase is unknown. A possibility is that the high pH of the FC650 biochar (9.32 in a 1% w/w solution of biochar in water) caused a conversion of ammonium (NH₄⁺) in the liquid to NH₃ gas near or on the surface of the biochar. The HF600 biochar had a lower pH (7.28 in 1% w/w solution) and appeared to be more hydrophobic than the FC650 biochar. The FC650 covers saturated completely within four weeks, whereas the HF600 covers still had dry material on top. This may explain the delayed spike in NH₃ concentrations seen in the HF600 covers. Delayed wetting of the biochar may have caused a delay in the conversion of NH₄⁺ to NH₃. One potential explanation is that the high pH of the FC650 biochar caused a temporary rise in pH of the surrounding manure, thus creating favorable conditions for NH₃ formation. The presence of oxygen or acidic functional groups in biochar affects biochar pH (Li et al., 2013). Functional groups on the surface of biochar particles will interact with H⁺ ions, thus affecting solution pH. Streubel et al. found that adding biochar (pH 9.3) to anaerobic digester effluent caused an increase in solution pH (Streubel et al., 2012). It may be necessary to avoid high pH biochars for biocover applications where reducing NH₃ emissions is a primary goal.

There was no clear advantage to using biochar rather than straw in this experiment. However, the short duration of the study may have advantaged the straw covers. Biologically-based cover materials tend to decay and lose effectiveness over time (Meyer and Converse, 1982; Clanton et al., 1999; Bicudo et al., 2004). Biochars, being more inert materials, should be less susceptible to decay and could provide cover function for a longer time period. There also appeared to be edge effects from the small diameter pails used in the study. The biochar covers tended to adhere to the sides and bridging was noticeable in comparison to the straw covers.

The mechanisms behind control of odor and gas emissions with biocovers have been discussed in other studies (VanderZaag et al., 2008). A reduction in mass transport of gases from the lagoon surface is generally considered to be the dominant mechanism. Microbial consumption of gases may also be a factor in cases where covers provide a substrate for aerobic bacterial growth at the surface. Further research is needed to determine if the porous structure of biochars can provide for enhanced microbial growth and reduced emissions from the lagoon surface.

The biochar covers (FC650 and HF600) exhibited differing sorption behavior in this study. This agrees with other research showing that biochars with differing properties exhibit different sorption behaviors in the environment (Lehmann and Joseph, 2009; Mukherjee and Zimmerman, 2013; Zhang et al., 2013). Biochars have been shown to be capable of sorbing and releasing many elements and compounds from manures (Streubel et al., 2012; Sarkhot et al., 2013) and soils, (Uchimiya et al., 2010; Sun et al., 2011). Previous research has shown that the surface of biochars is typically negatively charged due the presence of oxygenated surface functional groups on biochar surfaces. This can enhance sorption of positively charged ions such as ammonium, but may inhibit sorption of negatively charged ions including nitrate and phosphate (Yao et al., 2011; Hollister et al., 2013; Gai et al., 2014). Biochar pH also affects sorption characteristics because much of biochar cation exchange capacity is pH dependent and can become protonated at lower pH and thus ineffective for cation sorption (Wang et al., 2015). The results reported here suggest that sorption and release of nutrients can also occur when biochar is used as a biocover. When biochar covers are blended with manure and soil-applied this dynamic nutrient exchange may continue. Other work suggests that biochar may reduce nutrient leaching from soils when combined with manure (Laird et al., 2010; Troy et al., 2014; Bradley et al., 2015). To date we are unaware of any studies where biochar has been used as a lagoon cover and then field-applied. This concept needs further research to determine the potential benefits and drawbacks of the practice.

In this study the straw covers provided greater capture of N and P per gram than either of the biochars. From an agronomic standpoint, sorption of N, P, and K would be desirable characteristics of a biocover material intended for soil application. The possibility remains that biochars would retain sorbed N and P for longer periods in the environment, but this was not evaluated. Whether nutrients are contained within manure particles adhered to the surface or entrained in pore spaces vs. nutrients sorbed to the internal structure of the cover materials will affect persistence in the soil. When the pails were emptied at the end of the adsorption study, visual observation indicated that many of the small (< 2mm) biochar particles had accumulated at the bottom of the pails. These particles would have considerable surface area for nutrient sorption, but only the particles that remained floating were included in the elemental analysis. The FC650 biochar had noticeably more organic material stuck to the surface at the end of the adsorption experiment than did the HF600 biochar. The reason for this remains unclear, but is likely related to the differing sorption characteristics of the biochars. The more hydrophobic nature of the HF600 biochar may have prevented attraction of slurry material to its surface. This may partially explain why the FC650 biochar appeared to be considerably more effective than HF600 at sorbing nutrients. Conversely, the HF600 biochar was more effective at reducing NH₃ concentrations.

Odor Study

Weather events during the odor study likely affected odor perception. The covers were frozen during odor judging on week 6. It rained during judging on weeks 2, 8, and 10. Rainfall totaled 460 mm over the 12-week period, causing the manure to become more dilute over time. This may have affected odor intensity. Odor is known to be very subjective and this was reflected in the significant variability in the judges’ responses. Odor perception from the empty barrel could be background odor, odor from the barrel itself, or a placebo effect where the judges perceived odor because they thought the barrel contained manure. The muslin cloth used to prevent the judges from seeing the contents of the barrel appeared to make odor detection difficult, especially during rain events. At the conclusion of the study on week 12, the judges were asked to re-evaluate the odor without the cloth covers in place. The change in odor offensiveness without the covers in place was not statistically significant. There was a noticeable hesitation amongst the judges to sniff the barrel when the cloth was removed and they could see the contents.

Durability of the biochar biocovers is likely related to their density and hydrophobicity. Biochars produced at lower temperatures tend to be more hydrophobic (Gray et al., 2014). This may explain the lower moisture uptake by the HF350 cover. Biochars that repel water and maintain air-filled voids would be expected to float for longer periods compared to hydrophilic biochars. Most biochars have a skeletal density greater than 1g cm⁻³ and would be expected to sink if the pore spaces become fully saturated. Brewer et al. reported skeletal density values ranging from 1.34 to 1.96 g cm⁻³ for biochars (Brewer et al., 2014). Adhesion of manure particles to the biochar surface could affect the buoyancy of biochar covers. In addition to low skeletal density, hydrophobicity would be desirable for flotation over long periods of time. The tradeoff with greater hydrophobicity of a biocover is likely to be lesser nutrient sorption.

Conclusions

The adsorption experiment suggests that a biochar cover can reduce NH₃ emissions while sorbing nutrients from livestock manure. The odor study showed that biochar covers can reduce odor. The variation in performance seen in these experiments underscores the importance of selecting a biochar with properties suited to nutrient sorption while providing effective lagoon surface coverage. Despite the challenges, using biochar as a biocover material holds promise. Blending biochar with other materials may be a strategy to improve both efficacy and economics of biocovers. Floating biochar covers could provide a mechanism for passive capture of nutrients with no need for pumps, filters, or additional manure holding tanks. The covers could be blended with the manure and field-applied, or collected and sold as a means for exporting excess nutrients off the farm. This preliminary investigation of biochar biocovers provides a basis for future research. We have shown that some biochar covers can sorb nutrients and provide reductions in odor and gas emissions and were able to identify pitfalls to be avoided in future investigations. Subsequent research activities should be directed at determining the optimal hydrophobicity, pH, and particle size distribution for improved cover durability and nutrient sorption. Methods for applying and removing biochar covers should also be explored.

Abbreviations

M1

manure 1

M2

manure 2

M3

manure 3

HF 350, 550, 600, and 1200

hog fuel biochars pyrolized at 350°C, 550°C, 600°C, and 1200°C respectively

FC650

fir chip biochar gasified at 650°C

PC500

pine chip biochar pyrolized at 500°C