Reducing greenhouse gas emissions from pig slurry by acidification with organic and inorganic acids.

Methane (CH4) emission from pig slurry is a large contributor to the climate footprint of livestock production. Acidification of excreta from livestock animals with sulfuric acid, reduce CH4 emission and is practiced at many Danish farms. Possible interaction effects with other acidic agents or management practices (e.g. frequent slurry removal and residual slurry acidification) have not been fully investigated. Here we assessed the effect of pig slurry acidification with a range of organic and inorganic acids with respect to their CH4 inhibitor potential in several batch experiments (BS). After careful selection of promising CH4 inhibitors, three continuous headspace experiments (CHS) were carried out to simulate management of manure in pig houses. In BS experiments, more than <99% CH4 reduction was observed with HNO3 treatment to pH 5.5. Treatments with HNO3, H2SO4, and H3PO4 reduced CH4 production more than acetic acid and other organic acids when acidified to the same initial pH of 5.5. Synergistic effects were not observed when mixing inorganic and organic acids as otherwise proposed in the literature, which was attributed to the high amount of acetic acid in the slurry to start with. In the CHS experiments, HNO3 treatment reduced CH4 more than H2SO4, but increased nitrous oxide (N2O) emission, particularly when the acidification target pH was above 6, suggesting considerable denitrification activity. Due to increased N2O emission from HNO3 treatments, HNO3 reduced total CO2-eq by 67%, whereas H2SO4 reduced CO2-eq by 91.5% compared to untreated slurry. In experiments with daily slurry addition, weekly slurry removal, and residual acidification, HNO3 and H2SO4 treatments reduced CO2-eq by 27% and 48%, respectively (not significant). More cycles of residual acidification are recommended in future research. The study provides solid evidence that HNO3 treatment is not suitable for reducing CO2-eq and H2SO4 should be the preferred acidic agent for slurry acidification.

Introduction

Methane (CH4) emission from management of livestock manure contributes approximately 6% of global anthropogenic CH4 emission [1] and is a considerable source of global climate change [2]. Methane emission factors are higher for pig slurry than cattle slurry [3] and the potential for mitigation is greater in pig slurry management systems. In many finisher pig production facilities, the slurry is stored underneath the slatted floor in pits or channels and kept there throughout the production cycle [4]. In northern regions, temperatures are lower in outside storage tanks and the rate of CH4 production per ton slurry will therefore decrease significantly if slurry from pig houses is pumped to external storages. Frequent slurry removal from the pig houses by gravimetric emptying, scraping, or flushing, therefore holds a CH4 reduction potential, but such practices are carried out in <1% of pig production facilities [4]. Another promising mitigation technology is slurry treatment with acidic agents. This approach was implemented to reduce ammonia emission as it shifts the acid-base equilibrium towards ammonium ions (NH4+) that are nonvolatile [5]. Lowering the slurry pH also reduces CH4 emission, but the mode of inhibition is not completely understood. Common practice is to lower the pH to 5.5 with H2SO4 with reported CH4 reductions of 63–99% [6]. Proposed inhibition mechanisms include H2S-mediated inhibition [7], which is mainly derived from sulfate reduction [8], uncoupling of the cell membrane by protonated fermentative products [9, 10], or competitive inhibition of methanogens by microorganisms able to respire more energetic electron acceptors [11]. The latter may be relevant if acidification is carried out using H2SO4 or HNO3, which dissociates to the more preferable electron acceptors, sulfate and nitrate, than CO2 during methanogenesis [11]. Nitrate is a highly preferable electron acceptor in the absence of oxygen and it may be partially reduced by microorganisms to N2O or fully reduced to N2 in a cascade of reactions termed denitrification [12, 13]. Nitrous oxide is a greenhouse gas 273 times more potent than CO2 over 100 years [14]. It may also be produced as a side product of NH3-oxidation under semi-aerobic conditions when NH4+ is oxidized to NO2-, but this process is considered nearly absent in slurry channels and storages without crusts [12].

As NO3- is a preferable electron acceptor, CH4 potential may be significantly reduced due to competition for substrates by denitrifying bacteria if treated with HNO3. Large CH4 reductions were previously observed from HNO3 acidified cattle slurry, but simultaneous increases in N2O emission attributed to denitrification activity were also detected [15, 16]. It appears difficult to avoid N2O emission from sole HNO3 treatment, but combining acids might generate synergistic effects. Lately, CH4 emission from slurry was reduced by combining H2SO4 with acetic acid as compared to pure H2SO4 treatment to pH 5.5 [6]. This effect was possibly related to protonated acetic acid uncoupling the cell membrane [9, 10] and draws attention to the idea of combining organic and inorganic acids to achieve synergistic mitigation effects. Other organic acids have been examined as replacements for H2SO4, but with a focus on NH3 emission and manure characteristics [17, 18].

Acidification strategies must be optimized and integrated with other management practices such as frequent slurry removal from slurry channels inside animal houses to external storages. Frequent slurry removal leaves a residual slurry fraction that may function as an inoculum that speeds up the anaerobic processes in newly produced slurry resulting in accelerated CH4 production [19]. However, complete removal of residual slurry is impractical at large farms [20] or in farms with typical designs of pits and slurry channels. Coupling acidification of the residual slurry with frequent slurry removal may enhance the total mitigation potential and reduce acidification expenses as compared to bulk acidification. Laboratory tests of residual slurry acidification have shown very promising results a with reduction of CH4 emission of up to 99% over 116 days [21]. In pilot-scale storage tanks (10 m3), 2.12 m3 residual slurry which was acidified before addition of 8.48 m3 fresh slurry reduced CH4 emission by 77% over 155 days compared to a control case where the residual slurry was not acidified before the addition of fresh slurry [20]. Neither of the studies examined residual slurry acidification in a setup with continuous or daily addition of slurry, which could potentially fuel CH4 emissions. Further, the studies focused on cattle slurry in outside storages, which has a lesser potential of CH4 emission than pig slurry stored inside a pig house [3].

In this study, the aim was to determine the optimal acid treatment for reducing greenhouse gas emissions from pig slurry and to understand how acid treatment efficiency depends on dosage and frequent removal of slurry. This was accomplished in two steps; 1) Screening of CH4 production from pig slurry incubated with organic acids, inorganic acids, and combinations thereof in a simple assay using closed batch vials. 2) Following initial screening, we selected the most promising greenhouse gas mitigating acids for further investigations in a continuous gas monitoring setup, designed to simulate frequent slurry removal and residual slurry acidification in a pig pen. This novel experimental approach gives a realistic indication of the mitigation technology potential in pig houses. Gas emissions were measured using state-of-the-art cavity ring-down spectrometry to capture temporal evolution of gas emission and its response to treatment and slurry management strategy. The work conducted here clearly indicated that H2SO4 should be the preferred acidic agent for slurry acidification, independently of the management practice.

Materials and methods

Two types of incubation experiments were conducted in this study. These are referred to as batch screening experiments (BS experiment A, B, and C) and continuous headspace experiments (CHS experiment A, B, and C). A simplified overview of the experimental setups is shown in Fig 1. The batch screening experiments were done first to choose the most promising acidic agents for the more complex continuous headspace experiments. In the batch screening experiments the target pH was 5.5 since this is the target pH of commercial in-house acidification systems [5]. The methodology and procedures are detailed in the following sections.

Fig 1

Experimental overview of batch screening experiments and continuous headspace experiments.

GC-TCD is gas chromatography with a thermal conductivity detector, MFC is mass flow controller, CRDS is cavity ring-down spectrometry, GHG is greenhouse gas.

Slurry characterization

Slurry dry matter was measured as the slurry mass remaining after heating at 105°C for 24 h in a B180 Oven (Nabertherm). The pH was measured with a Portamess module (Knick) and a DJ114 pH electrode. For NH4+concentration, the slurry was pre-diluted 10–50 times with ultrapure water and measured with an ammonium test kit 1.00683 (Merck) on a Spectroquant NOVA 60 (Merck).

Site description and slurry collection

Experimental pig facility

Pig slurries for all incubation experiments were obtained from an experimental pig facility at Aarhus University (Foulum, Denmark). There were two pens in each section and 15 finishing pigs (30–110 kg) per pen. Batch time for the pigs was 77 days and batch days were scheduled in parallel in the two sections. In each pen, there was a 60 cm deep slurry pit below the slatted floor. The slurry pits were emptied with vacuum flushing leaving approximately 4 cm of residual slurry in the pit after vacuum flushing. The slurry management differed in the two sections: In the experimental section, the slurry was flushed every 7th day whereas, in the control section, the slurry was flushed on day 40 and at the end of the batch on day 77.

Slurry sampling

An overview of applied slurry types is presented in Table 1. Vacuum flushed slurry for the batch experiments and CHS experiment A was sampled from the vacuum flushing system in the control section at day ~ 40 in a batch of pigs. Due to decreasing methanogenic activity in the vacuum flushed slurry a residual slurry with higher methanogenic activity was used in CHS experiments B and C to ensure that potential differences in treatments were still detectable. Residual slurry was sampled from the bottom of the pits in the control section on the day after the pigs were taken out of the section and after the pits were emptied by vacuum flushing. The residual slurry was sampled by scraping the 3–5 cm residual slurry layer on the pit bottom with the beaker and the slurry was stored in PFTE containers at 5°C until further experimental treatment. For diurnal slurry addition in CHS experiment C, weekly vacuum flushed slurry was sampled from the experimental section and stored at 5°C between daily additions to the incubation flasks. New vacuum flushed slurry was collected from the experimental section weekly to ensure the slurry was no more than 7th days old when applied in the experiments.

Table 1

Description and characteristics of slurries.

Description	Applied in experiments	pH	Dry matter (%)	TANa (g/L)
Vacuum flushed slurry from control section, sampled 2-7-2020	BSb experiments and CHSc experiment A	6.96	4.85	2.65
residual slurry from control section, sampled 6-11-2020	CHS experiments B and C	7.21	9.35	3.11
vacuum flushed slurry from experimental section, sampled 26-11-2020	CHS experiment C	6.66	8.93	2.07
vacuum flushed slurry from experimental section, sampled 03-12-2020	CHS experiment C	6.77	7.76	2.17

a Total ammonia nitrogen.

b Batch screening experiments.

c Continuous headspace experiments.

Batch screening experiments

Vacuum flushed slurry was agitated with a magnet stirrer (Heidolph, MR 3001 K) in a 5 L bucket while pH was measured with a pH sensor (Knick) and samples were collected for later slurry characterization. Eight grams of the slurry was transferred to 20 mL serum vials while still agitating the slurry using a 10 mL pipette with cut tips to avoid clogging. The headspaces of the serum flasks were flushed with nitrogen then crimped with aluminum caps and incubated for 24 h at room temperature (~23°C). After 24 h the vials were opened and treated with inorganic or organic acids alone (BS experiment A), inorganic acids combined with acetic acid (BS experiment B), and organic acids (BS experiment C). Organic acids included acetic acid, lactic acid, propanoic acid, citric acid, and formic acid whereas inorganic acids used were hydrochloric acid, sulfuric acid, nitric acid, and phosphoric acid. Details of treatment dose and BS experiment design are presented in Table 2. After acid treatment, the vials were flushed with nitrogen, crimped with aluminum caps, and incubated at room temperature. On a weekly or biweekly basis, the pressure and CH4 gas concentration in the serum vials headspaces were measured using a pressure sensor fused with a needle and a gas chromatograph with a thermal conductivity detector (GC-TCD). After pressure measurement or gas sampling for GC-TCD analysis, the vials were flushed with nitrogen and crimped with aluminum caps. The BS experiments were terminated after 1–3 weeks depending on the gas production rate.

Table 2

Treatment details of batch screening experiments.

Treatment	n	Dose (pH or millimolar (mM))
BS experiment A, length: 6 or 18 days, Fig 5A
Untreated	2	–
Sulfuric acid	3	To pH 5.5
Phosphoric acid	3	To pH 5.5
Acetic acid	3	To pH 5.5
Hydrochloric acid	3	To pH 5.5
Nitric acid	3	To pH 5.5
Lactic acid	3	To pH 5.5
BS experiment B, length: 9 days, Fig 5B
Untreated	3	–
Acetic acid + nitric acid	3	42 mM + to pH 5.5
Acetic acid + sulfuric acid	3	42 mM + to pH 5.5
Acetic acid + phosphoric acid	3	42 mM + to pH 5.5
Acetic acid	3	42 mM
Nitric acid	3	To pH 5.5
Sulfuric acid	3	To pH 5.5
Phosphoric acid	3	To pH 5.5
BS experiment C, length: 7 days, Fig 5C
Untreated	2	–
Acetic acid + sulfuric acid	1	21 mM + to pH 5.5
Acetic acid + sulfuric acid	1	60 mM + to pH 5.5
Acetic acid + sulfuric acid	1	89 mM + to pH 5.5
Propionic acid + sulfuric acid	1	23 mM + to pH 5.5
Propionic acid + sulfuric acid	1	53 mM + to pH 5.5
Propionic acid + sulfuric acid	1	82 mM + to pH 5.5
Citric acid + sulfuric acid	1	13 mM + to pH 5.5
Citric acid + sulfuric acid	1	24 mM + to pH 5.5
Citric acid + sulfuric acid	1	36 mM + to pH 5.5
Lactic acid + sulfuric acid	1	28 mM + to pH 5.5
Lactic acid + sulfuric acid	1	70 mM + to pH 5.5
Lactic acid + sulfuric acid	1	112 mM + to pH 5.5
Formic acid + sulfuric acid	1	30 mM + to pH 5.5
Formic acid + sulfuric acid	1	46 mM + to pH 5.5
Formic acid + sulfuric acid	1	60 mM + to pH 5.5
Sulfuric acid	3	To pH 5.5

Continuous headspace experiments

Vacuum flushed or residual slurry was thoroughly stirred in a 5 L bucket with a large spatula while measuring the pH with a Portamess pH sensor (Knick) and sampling for later characterization of the slurry. Then 300–800 g of slurry was amended to 1 L DURAN flasks and immediately treated with different acids alone or in combinations. The slurry-treated flasks were then placed into a continuous gas monitoring setup. In the setup wet air was continuously dosed with a mass flow controller (Bronkhorst EL-FLOW, Ruurlo, Netherlands) and distributed evenly to the headspace of the 15 x 1 L flasks containing treated pig slurry. An empty flask was used for background measurements. The continuous air flow was adjusted to 1.3 L/min per slurry flask. The outlet flow from the flasks was directed to a 16-port manifold (Picarro, Santa Clara, CA, USA), which changed position between the 16 flasks every 15th minute. An exhaust tube was placed before the manifold to avoid pressure buildup when the manifold was closed for a particular flask. The concentration of CH4, N2O, and CO2 from the flasks was measured with a G2508 cavity ring-down spectrometer (Picarro) with extended CH4 concentration range. Due to spectral interference at high concentration levels, NH3 was removed by applying an H2SO4 scrubber and a Nafion tube before the inlet of the G2508 as previously described [22].

In total three CHS experiments each lasting 2 weeks were carried out. A presentation of the treatments is detailed in Table 3. In CHS experiment A, 800 g of pig slurry (> two months old) with considerable methanogenesis activity (observed from in-situ measurements before sampling), was added to 1 L incubation flasks. Then 10–20 g HNO3 (69% purity) (kg slurry)-1 was added to the slurry flasks to reach pH values between 5.5 and 6.25. Acidification in all CHS experiments was carried out by alternately pipetting acid into the slurry under agitation with a magnet stirrer followed by pH measurement. This procedure was repeated until the pH reached and remained stable at the target pH over 5 minutes. In CHS experiment A, nitrous oxide was measured continuously over two weeks, but CH4 and CO2 were only measured periodically when a Nafion filter was applied to remove NH3 and possibly other volatile organic compounds, which caused spectral interference with CH4 and CO2. On the contrary, NH3 was measured in periods where the Nafion filter was not applied. In CHS experiment B, 500 g of residual pig slurry was added to 1 L incubation flasks. Compared to CHS experiment A, the slurry mass was reduced (from 800 to 500 g) to avoid excessive foam formation during acidification. The slurries were acidified with either HNO3 (69%), H2SO4 (96%), or mixtures of both acids to final pH values of 5.5 for all treatments. Greenhouse gases were monitored continuously the following two weeks with the Nafion filter permanently applied. For CHS experiment C, the intention was to simulate weekly slurry flushing combined with residual slurry acidification. To start with, 300 g residual slurry was added to 1 L incubation flasks and the slurries were acidified with HNO3 (69%), H2SO4 (96%), or mixtures of both to final pH values of 5.5 for all treatments. On the day after treatment (day 1) and on a diurnal basis hereafter, 78 g of vacuum flushed slurry (<7 days old) was added to all incubation flasks to simulate daily excretion of slurry from the animals, resulting in a total of 768 g slurry in each flask on day 6. On day 7, the slurries were mixed by gently shaking and slurry was removed until 300 g was left in each flask to simulate frequent slurry removal practice. The remaining slurries were then reacidified (residual acidification) with HNO3 (69%), H2SO4 (96%), or mixtures of both, to final pH values of 5.5. Later on day 7, 78 g vacuum flushed slurry (<7 days old) from a new slurry aliquot was amended and diurnal slurry addition was repeated until and including day 13. Greenhouse gases were measured continuously and with the Nafion filter permanently applied. For all CHS experiments, the pH was measured every 3–4th day by opening the incubation flasks completely and carefully submerging the pH sensor to 1 cm above the bottom of the incubation flask without stirring the slurry. All treatments in continuous headspace experiments A, B, and C were carried out in triplicates.

Table 3

Treatment details of continuous headspace experiments.

Treatment	Dose at experiment start (g (kg slurry)-1)	Slurry start + daily addition (g)
CHS experiment A, datetime; 21-10-2020–05-11-2020, Fig 4
Untreated	–	800
Nitric acid to pH 6.25	10.11 ± 0.46
Nitric acid to pH 6.00	12.29 ± 0.04
Nitric acid to pH 5.75	15.47 ± 0.10
Nitric acid to pH 5.5	17.51 ± 0.04
CHS experiment B, datetime; 11-11-2020–25-11-2020, Fig 3
Untreated	–	500
Sulfuric acid to pH 5.5	11.29 ± 0.15
Nitric acid to pH 5.5	20.23 ± 0.04
Nitric acid (L)a and sulfuric acid to pH 5.5	5.28 ± 0.05 and 7.94 ± 0.06
Nitric acid (H)a and sulfuric acid to pH 5.5	10.62 ± 0.13 and 6.29 ± 0.28
CHS experiment C, datetime: 26-11-2020–17-12-2020, Fig 4
Untreated	–	300 + 78
Sulfuric acid to pH 5.5	11.68 ± 0.017
Nitric acid to pH 5.5	19.15 ± 0.28
Nitric acid (L)a and sulfuric acid to pH 5.5	5.51 ± 0.07 and 8.63 ± 0.11
Nitric acid (H)a and sulfuric acid to pH 5.5	10.44 ± 0.04 and 5.35 ± 0.02

(L) and (H) refers to “low” and “high” doses of nitric acid.

Gas analysis and emission calculations

For batch screening (BS) experiments CH4 concentration in the headspace of the vials was measured on a GC-TCD (Agilent Technologies 7890A, USA) equipped with a Porapak Q column (Agilent) and argon carrier gas. The pressure in the headspace of the vials was measured with an MPX4250AP absolute pressure sensor. CH4 production was calculated as the product of vial headspace pressure, vial headspace volume, and vial headspace CH4 concentration.

In continuous headspace experiments (CHS), a G2508 analyzer (Picarro, Santa Clara, CA, USA) was used to measure CH, N2O, CO2, NH3, and H2O concentrations. For NH3, humidity correction was done according to [23], using Eq 1.

Where NH3,raw is the output from the G2508 analyzer. Gas emission rate was calculated using Eq 2.

Where r is the emission rate (g day-1) of gas j, C is the gas concentration (atmosphere) of gas j measured by the CRDS instrument, Q is the airflow rate (L day-1) in the headspace of each incubation flask, and ρ is the density of gas j at 25°C. The cumulated gas emission was calculated by the trapezoidal method of integration in Eq 3.

Where E is the cumulative gas emission (g) of gas j after i rate measurements, r is the emission rate (g day-1) of gas j at i rate measurement, and t is the time (days) at rate measurement i.

Emission in CO2 equivalents (CO2-eq) was calculated using 100-year global warming potentials without inclusion of climate-carbon feedbacks of 273 (N2O) and 27.2 (CH4) [14]. Carbon dioxide emission was not included in the calculation of CO2-eq as the CO2 flux to and from the atmosphere is considered to be net-zero from livestock animals [3].

The non-dissociated form of acetate (acetic acid or HAc) was calculated from Eq 4, using a pKa value for HAc of 4.76

Statistics

Error bars in figures and tables represent sample standard deviations. For batch experiments with many different treatments, statistical significance and reported p-values were obtained from one-way ANOVA tests using a level of significance (α) of 0.05. The ANOVA test was done in MatLab using the “anova1” function with default settings. The statistics from “anova1” were used as input for a pairwise comparison between means of treatments to indicate which treatments differed from each other. The pairwise comparison was done using the “multcompare” function in MatLab with the default Tukey Kramer post hoc method. For continuous headspace experiments, two-sample t-test with assumed unequal variance was conducted to test for the difference of individual treatments compared to the untreated slurry. The level of significance (α) was 0.05 for the t-tests, which was conducted in MS Excel 2016. The abbreviation “ns” is applied when presented results that are not statistically significant in the text. Error propagation was used for calculating errors on CO2-eq and where otherwise relevant according to [24]. For rate plots (Figs 2A, 3A, and 4A) data is presented as rolling means over 12 h using the “zoo” R-package.

Fig 2

Gas emission rates (A) and cumulative gas emissions (B) from continuous headspace experiment A. In (A) colors indicate different treatments according to legend. In (B) black dots represent average of triplicates and the blue error bars indicate sample standard deviation. Treatments that were statistically different from the untreated slurry concerning cumulative emission are indicated with * (p<0.05), ** (p<0.01), *** (p<0.001).

Fig 3

Gas emission rates (A) and cumulative gas emissions (B) from continuous headspace experiment B. In (A) colors indicate different treatments according to legend. In (B) black dots represent average of triplicates and the blue error bars indicate sample standard deviation. Treatments that were statistically different from the untreated slurry concerning cumulative emission are indicated with * (p<0.05), ** (p<0.01), *** (p<0.001).

Fig 4

Gas emission rates (A) and cumulative gas emissions (B) from continuous headspace experiment C. In (A) colors indicate different treatments according to legend. In (B) black dots represent average of triplicates and the blue error bars indicate sample standard deviation. Treatments that were statistically different from the untreated slurry concerning cumulative emission are indicated with * (p<0.05), ** (p<0.01), *** (p<0.001).

Results

The batch experiments consisted of three separate experiments, where the approximate effect of organic, inorganic, and mixtures from both groups was assessed with respect to their inhibiting effect on CH4 production. Fig 5 shows inhibition potential by presenting the cumulative CH4 production in vials incubated with acid-treated pig slurries (experimental details in Table 2). Both organic and inorganic acids all caused a significant inhibition of CH4 production compared to untreated slurry (Fig 5A and 5B). Nitric acid was the most effective inhibitor with > 99% CH4 reduction at pH 5.5 (Fig 5A and 5B). The inhibition effect of acid treatments lasted throughout the experimental period with only untreated slurry producing CH4 from day 6 to day 18. The untreated slurry had an initial acetate (HAc + Ac) concentration of 7.26 ± 0.95 g/L and there was no effect of combining acetic acid with the inorganic acids (Fig 5B) as compared to inorganic acids alone when acidifying to a final pH of 5.5. Different organic acids and concentrations thereof were tested to examine potential synergistic mitigation effects with H2SO4 to hereby reduce the total volume of acids needed to inhibit CH4 production. However, no synergistic effects were observed (Fig 5C). Most of the acid treatments resulted in a pH increase from 5.5 to pH 5.6–6.15 by the end of the experiments, and the pH increase was positively correlated with the amount of CH4 produced (S1 Appendix). Formic acid + H2SO4 increased pH to 6.48–6.79 and stimulated CH4 production compared to untreated slurry (Fig 5C). Similarly, lactic acid alone increased pH to 7, but CH4 was still reduced compared to the untreated slurry in Fig 5A. A smaller dosage of lactic acid in combination with H2SO4 produced equal amounts of CH4 compared to the untreated slurry in Fig 5C.

Fig 5

Inhibition potential of different acids on pig slurry in batch screening experiments A (A), B (B), and C (C). Uncertainty bars indicate standard deviation. In (C), (L), (M), and (H) indicate low, medium, and high dose treatments as described in Table 2. p-values are for one-way ANOVA test on differences between CH4 production and bars with letters in common are not statistically different with a pairwise comparison of means (Tukey Kramer).

The batch screening experiments suggested that HNO3 was an efficient CH4 inhibitor and in the subsequent continuous headspace experiments, the effect of HNO3 was studied in more detail.

Fig 6 presents pH profiles from the three CHS experiments. In CHS experiment A, pH values were relatively stable, except for the HNO3 to pH 6.25 treated slurry, which increased from pH 6.25 to 7.25 during the first week. The reason that the variant with HNO3 to pH 6.25 had a higher end pH than the untreated slurry is unclear, but it is likely a combination of denitrification producing strong base [11] and initial loss of acidic CO2 during acidification. Treatments with HNO3 to pH 6 or below did not change considerably in pH. In CHS experiment B (Fig 6 middle), all treatments to pH 5.5 increased only to pH ~5.75 and stabilized after 2–3 days, independent of the acidic agent used. However, during CHS experiment C, new slurry was added daily, resulting in a significant pH increase of more than 1 pH unit during week one and again during week two.

Fig 6

Mean pH values measured in continuous headspace experiments A, B, and C. Black arrows in CHS experiment C indicate addition of vacuum flushed slurry and the red arrow indicates addition of slurry + slurry removal + acidification. Uncertainty bars indicate sample standard deviation of triplicates.

Fig 2 shows gas emission rates (a) and cumulative gas emission (b) from CHS experiment A, where slurry was acidified to different pH values using HNO3. Spectral interference from high NH3 levels in periods without a Nafion filter applied resulted in erroneous CH4 measurements and has been omitted. Nitric acid reduced CH4 emission severely, but treatment with HNO3 to pH 6.25, which was the lowest dosage of HNO3, showed a trend of increasing CH4 emission by the end of the experiment (Fig 2A). Slurry treatment with HNO3 to pH 6.25 produced a considerable N2O emission peak around day 5. The simultaneous pH increase (Fig 6), suggests that N2O was produced from denitrification of NO3- from HNO3. Ammonia emission from slurry increased with increasing pH due to a shift from nonvolatile NH4+ to NH3 (pka = 9.25) (S2 Appendix). The significantly higher N2O emission from slurry treated with HNO3 to pH 6.25 (p = 0.042) resulted in significantly higher emission of CO2-eq (p = 0.044) than the untreated slurry. However, with HNO3 treatment to pH 5.75 and pH 5.5, CO2-eq emission was significantly reduced by 52.9% (p = 0.021) and 59.5% (p = 0.013) compared to untreated slurry.

As N2O emission was very pronounced at pH 6.25, another round of experiments was conducted in which all slurries were acidified to pH 5.5 and the N-pool available for microbial production of N2O was minimized by replacing different amounts of HNO3 with H2SO4 (Table 3). Fig 3 shows gas emission rates from CHS experiment B, with slurry acidified with mixtures of HNO3 and H2SO4 to pH 5.5. All treatments with HNO3 reduced CH4 emission rates by approximately three orders of magnitude (> 99.9%) compared to the untreated pig slurry and more than two orders of magnitude (> 99%) compared to H2SO4 treatment (Fig 3). These reductions are difficult to perceive from Fig 3, as all treatments showed close to zero CH4 emission and we refer to S3 Appendix for log10 transformed y-axis. The N2O emission from treatments with HNO3 was 14.8–18.4 fold higher than from untreated slurry (p = 0.001–0.1), indicating that denitrification still proceeded at pH 5.5, however at very limited rates compared to CHS experiment A at pH 6.25. Carbon dioxide emission was lower for all acidification treatments compared to untreated slurry (ns, p = 0.2) suggesting that microbial oxidation of substrates and intermediates, during either hydrolysis, fermentation, or methanogenesis, was severely slowed down at pH 5.5. In terms of CO2-eq, untreated slurry had higher emission than acidification treatments, primarily due to the higher CH4 emission. The most effective inhibitor was H2SO4 with 91.5% CO2-eq emission reduction (p = 0.034). Reduction of CO2-eq emission with HNO3 treatments was 67%–59% (ns, p = 0.059–0.071).

To simulate frequent slurry removal and daily excretion of pig feces and urine, treatments from CHS experiment B were repeated, but with daily addition of slurry and weekly slurry removal to a threshold level, which was equivalent to a situation with weekly vacuum flushing in a pig pen. In addition, the residual slurry was acidified after one week to reduce acid consumption and target the methanogenic inoculum. During the first week in CHS experiment C, CH4 emission was lower for all HNO3 treatments, but during week two, CH4 emissions accelerated to levels nearly identical to those from untreated slurry (Fig 4). This resulted in total CH4 reductions ranging from 48–72% compared to the untreated slurry, with HNO3 (L) + H2SO4 being the most effective treatment based on cumulative emission and the only treatment that was significantly different from the untreated slurry (p = 0.036, others p = 0.054–0.081) (Fig 4B). Nitrous oxide emission showed a less dramatic increase during week two than CH4 and the increase was more pronounced for slurries with HNO3 addition, which is consistent with results from CHS experiment B. The CO2-eq emission was lowest for H2SO4, with a reduction of 49% compared to untreated slurry (ns, p = 0.085) and 27.5% reduction for HNO3 treatment (ns, p = 0.225). Based on these observations pure H2SO4 acidification of residual slurry would be preferable from a 100-year climate perspective. This was mainly due to the N2O production outweighing the extra reduction of CH4 gained by using HNO3 instead of H2SO4.

Discussion

The batch screening of combined acidic agents was initiated with expectations that synergistic effects could occur due to the uncoupling effect previously described [9]. This theory has not been properly tested in stored animal slurry until recently, where Fuchs et al (2021) reported that CH4 emission from acidified cattle slurry showed a larger reduction with acetic acid treatment than H2SO4 treatment (both to pH 5.5). Fuchs et al. (2021) reported 97% CH4 inhibition at 1631 mg HAc/L in cattle slurry acidified to pH 5.5 at room temperature. In comparison, H2SO4 and acetate reduced CH4 by 93% and 85%, respectively over 18 days of incubation (Fig 5A) in the experiments reported here. Zhang et al. 2018 also studied methanogen inhibition in the pH range of 5–7 and acetate concentration ranging from 0–15 g/L and found that methanogen activity could not be recovered when exceeding 810 mg HAc/L [25]. In this study, we did not observe additional CH4 reduction with acetate addition. The acetate concentration in the untreated slurry was 7.26 ± 0.95 g/L, which equals 1119 ± 147 mg HAc/L at pH 5.5 (calculated from Eq 4). This being well above the recovery limit of 810 mg HAc/L reported by Zhang et al. (2018), we deem that further addition of acetate would have no additional inhibitory effect on methanogenesis. Importantly, this finding is not conflicting with previous literature findings, but highlights that utilization of acetate or other organic acids (Fig 5C) as combination inhibitors with H2SO4, will probably only be beneficial in a relatively VFA depleted slurry similar to that used by Fuchs et al. (2021), which contained 104 ± 12 mg HAc/L at pH 5.5 (H2SO4 acidified). Acetate treatment alone resulted in a final pH of 6.05 and a larger CH4 production (Fig 5B) than other treatments (all pH 5.5), although it still reduced production by 55% compared to the control. The lesser inhibition was likely due to a smaller proportion of the acetate (and other organic acids) being protonated at pH 6.05 than at pH 5.5. But the absence of sulfate or reduced sulfur species from H2SO4, which have earlier been reported to inhibit methanogenesis [7], could also explain lesser inhibition. The latter theory of sulfur species hampering methanogenesis is consistent with results in Fig 5A since H2SO4 acidification (pH 5.5) was more efficient than pure acetate acidification (pH 5.5) and HCl acidification (pH 5.5).

CH4 production was observed for acidified pig slurry with formic acid addition (Fig 5C) and it is well known that formate is an electron donor of CH4 production [26]. Still, at an initial pH of 5.5, formic acid fueled CH4 production and exceeded CH4 production in untreated slurry. This suggests that both pH and acid type are important predictors of CH4 inhibition potential in acidified slurry. This finding disqualifies formic acid as an inhibitory agent even under acidic conditions. The high CH4 reduction potential reported by Berg et al. (2006) for lactic acid was not confirmed in our screenings experiments. Instead, an increase in pH (S1 Appendix) and limited effect on CH4 production was observed with lactic acid treatments. This discrepancy could be related to differences in acidification target pH (pH 3.8–4.8 versus 5.5) and slurry type (cattle versus pig), which would affect buffer capacity and microbial community. Importantly buffer systems in slurry can result in pH increases after acidification. Hence, the CH4 reductions reported from the batch screening experiments are not comparable to commercial in-house acidification systems with continuous pH adjustment but instead hint to the relative efficacy of the different acids in such a system.

In CHS experiment A, a clear tendency of increasing pH was observed for HNO3 acidification to pH 6.25 (Fig 6). This pH increase could be related to release of acidic CO2 (Fig 2A) produced from organic matter transformation by microbes. For example, dissimilatory nitrate reduction to ammonium or denitrification of NO3- with acetate as electron donor produce strong base thereby increasing pH [11]. The N2O peak in Fig 2A suggests that denitrification of NO3- indeed occurred during the period where the pH increased for the HNO3 treatment to pH 6.25. Earlier experimental work has shown that denitrification produced high N2O fluxes from stored cattle slurry acidified with HNO3 to pH 6.0 but at pH < 5 N2O flux was reduced by several orders of magnitude [16]. That trend was demonstrated again in the current study, where N2O emission was completely inhibited at pH < 6 (reduced > 99% compared to HNO3 to pH 6.25). The optimum pH for denitrification in soils is neutral to slightly alkaline [27], which is in line with an interpolated pH value around ~7 at day 5–6 when the N2O peak is at its highest (Figs 2A and 6). A mass balance revealed that N2O emission corresponded to ~ 12% of the added NO3—N from HNO3, suggesting that the remaining NO3- was either not reduced or reduced through complete denitrification to N2 or dissimilatory NO3- reduction to NH4+. For slurries with pH ≤ 6 (Fig 6) there was only a small pH increment in the beginning and hereon after a pH stabilization 0.1–0.25 pH units above the initial pH. The small pH increase could be due to release of retained CO2, volatile fatty acids, or hydrogen sulfide produced during the acidification process. It has been shown that enzymatic N2O reductase activity is significantly reduced at low pH [28], thereby decreasing the enzymatic reduction of N2O to N2. However, the fact that CO2 production was reduced below pH 6, suggests that general inhibition of heterotrophic activity (including heterotrophic denitrifiers) is a more plausible explanation for the reduced N2O emission below pH 6. Given these considerations, a critical threshold for microbial activity likely exists somewhere between pH 6 and pH 6.25. This is consistent with the batch experiment, where acetate alone to pH 6.05 produced CH4 to a larger extent than treatments to pH 5.5, but still not as much as in untreated slurry (Fig 5B).

Emission of N2O was diminished substantially in CHS experiment B for HNO3 treatments, but still, they were significantly higher than the untreated slurry, probably due to a limited but persistent denitrification activity (Fig 3). Conversely, CH4 was reduced more in HNO3 treatments, which we attribute to increased competition for substrates by denitrifiers in the presence of NO3-. To assess the most efficient acidic agent a reasonable approach is a comparison of the total climate impact in terms of CO2-eq. Here we did not find evidence that suggests HNO3 is a superior greenhouse gas inhibitor to H2SO4. The CHS experiment A and B were conducted without daily slurry addition and without optimizing the timing of acidification. This justified carrying out the final CHS experiment C, where new slurry was added daily. Here the initial low pH of acidified slurries was expected to increase (Fig 6) as the vacuum flushed slurry had a pH of 6.7–6.8 (see Table 1). Addition of new slurry would dilute and thereby neutralize the acidic environment and add new substrates and microbial biomass to fuel organic matter transformation. This pH increase resulted in increasing CH4 production rates, which accelerated quickly at day 9 (Fig 4). This indicates that residual acidification by day 7, was overcome more quickly than at day 0, most likely due to a more adapted microbial community in week two. Considering the development in emission patterns during the two weeks of monitoring, the CH4 reduction potential is likely to decrease with time and this should be assessed in long-term experiments with more cycles of flushing and acidification. In comparison to the 48–72% CH4 reduction in CHS experiment C, Petersen et al. (2012) reported a 67–87% reduction in CH4 emission from H2SO4 bulk acidified cattle slurry in pilot storage tanks, without addition of new slurry. Sokolov et al. (2020a; b), achieved high CH4 reductions (>99% in lab; 77% in pilot storages) with residual acidification, but similarly to other studies, new slurry was not amended. Therefore, the abovementioned studies are perhaps more comparable to CHS experiment B without slurry addition and slurry removal where we observed > 99% CH4 reduction. Only a few studies have studied the effect of acidification (bulk or residual) on greenhouse gas emission from pig barns, where slurry is naturally and continuously added. Petersen et al. 2016 measured 50% CH4 reduction (not significant) from a pig barn in a spring campaign, but low CH4 emissions from the slurry compared to enteric emission reduced the confidence of the evaluation [29]. A report by the Danish knowledge institution, SEGES, reports a 60% methane reduction from bulk slurry acidification including enteric CH4 emission [30]. The mentioned studies both utilized photo-acoustic spectroscopy for CH4 monitoring, which suffers from spectral interference of volatile organic compounds and H2O [31, 32]. Based on results from the available literature Olesen et al. 2018 assessed that a CH4 reduction of 60% for bulk acidification in barns, including enteric CH4 emission, is realistic [33]. Assuming that 30% of CH4 comes from enteric fermentation, CH4 reductions from the slurry is probably > 90%, which as expected is higher than in CHS experiment C where only residual acidification was tested. Notably, there was not a significant effect of any acidification treatment on the CO2-eq emission in CHS experiment C, which is a consequence of the chosen t-test with unequal variances, which is more conservative. Using the regular student’s t-test with assumed equal variance would increase statistical power and result in a statistical difference for H2SO4 treatment compared to the untreated slurry, but the chance of concluding a Type I error would also increase [34]. In summary, CHS experiment C shows that a complete evaluation of residual slurry acidification with weekly slurry removal in pig barns would require more cycles of flushing and should be conducted using H2SO4. The reduced effect of residual slurry acidification compared to bulk acidification suggest that residual slurry acidification should be carried out with a lower target pH than 5.5 to maintain the inhibiting effect once new slurry is added. It is necessary to examine if the residual slurry can form an adapted microbial community, which will reduce the long-term effects of methanogenic inhibition through acidification.

Perspectives of acidification strategy

The described CHS experiments as well as the initial BS experiments in general find no compelling reason to shift to other acidic agents than H2SO4 when taking into account greenhouse gas emission from the slurry alone. From a farm-scale perspective, in certain scenarios, supplementing or acidifying slurry with other organic acids could be beneficial due to increased nutritional value as field fertilizer [17], and the risk associated with handling hazardous H2SO4 would be eliminated [35]. In addition, if the slurry is further processed in anaerobic digesters to produce biogas, acetic acid could hold potential as a methanogen substrate under controlled feeding. On the contrary, H2SO4 acidification, hampers downstream use of the slurry in biogas reactors, unless it is co-digested with untreated slurry at a maximum of 10% acidified slurry [36]. However, the degradation rate of organic acids is rapid compared to H2SO4 [37] and the timing of pure organic acid treatment is therefore critical in this context. In slurries with limited nitrogen, HNO3 may be supplied to increase the nutritional fertilizer value when landspreading the slurry to crops fields, but comes with a considerable risk of increasing N2O emissions [37].

Conclusion

Two types of experiments were conducted; 1) batch vials experiments and 2) continuous headspace experiments. In the batch vial experiments, CH4 production was reduced by >99% with HNO3 treatment to pH 5.5. Sulfuric acid and other inorganic and organic acids also reduced CH4 production significantly, but no synergistic effects between mixed acids were observed. It is likely that high concentrations of VFAs present before acidification reduce the efficacy of organic acid treatment. In the continuous headspace experiments, three rounds of experiments were conducted. Nitric acid treatment to pH 5.5, 5.75, 6.0, and 6.25, revealed that the degree of inhibition increased with lower pH for both CH4 and N2O emissions. Nitric acid treatment to pH 6.25 produced significant amounts of N2O and a concurrent pH increase suggested that this was due to denitrification activity. For bulk acidification with H2SO4 to pH 5.5 a 91.5% reduction in CO2-eq was found, whereas HNO3 and mixtures of HNO3 with H2SO4 reduced CO2-eq by only 59–67% due to increases in N2O emission. When simulating weekly removal of slurry combined with acidification of the residual slurry and daily addition of fresh slurry (CHS exp C), a considerable increase in greenhouse gas emissions was observed for all treatments during week two, resulting in a CO2-eq reduction of 48% for H2SO4 (ns) and 27% for HNO3 (ns). Further in CHS exp C, CH4 was reduced significantly only for HNO3 mixed with H2SO4 by 72%, but N2O emission largely outweighed this reduction. This suggests that when slurry is added daily and the microbial community adapts, acidification of the residual slurry becomes less efficient, but further studies are necessary to determine more clearly the long-term effect of adaptation mechanisms related to residual slurry acidification.

Correlation between CH4 and pH.

(DOCX)

Click here for additional data file.

NH3 emission rate.

(DOCX)

Click here for additional data file.

log10 transformed emission rates.

(DOCX)

Click here for additional data file.

Data for figures.

(XLSX)

Click here for additional data file.

9 Feb 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Xueming Chen

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

3. Thank you for stating the following in your Competing Interests section:

" No"

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state "The authors have declared that no competing interests exist.", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf.

4. Thank you for stating the following in the Acknowledgments Section of your manuscript:

"The research was funding by the Danish Agricultural Agency under the Ministry of Environment and Food (Miljø- og Fødevareministeriet) (grant no. 33010-NIFA-19-725) and would like to thank for the opportunity to conduct this research. "

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

"NO"

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

Comments to the Author

Reviewer #1: OVERVIEW AND GENERAL RECOMMENDATION

1. The study gives a comprehensive overview of the use of different acids for acidification on greenhouse gas emissions from slurry. The continuous addition of slurry, as is also common in barns, was also investigated. This leads to a special character of the manuscript. However, the studies are very extensive and are divided into several sub-studies. It is therefore difficult for the reader to follow the intentions of the authors.

2. In my opinion, the title is not appropriate because it mentions organic acids. However, most of the study deals with the inorganic acids H2SO4 and HNO3. Even in the abstract, acetic acid is only briefly mentioned once. Additionally, the authors should point out to the reader which of the acids mentioned are inorganic or organic acids.

3. Why was the pH value of 5.5 in BS chosen?

4. There are buffer systems in the slurry which lead to an increase in the pH value after acidification. This was not taken into account or mentioned. How practically relevant are the studies from BS in which acidification was only carried out once?

5. The figures look blurred. Please improve the resolution.

6. The actual data from which the mean values were formed were not provided.

7. It is recommended that the manuscript is read by a native speaker before submission, as there are some linguistic and grammatical errors in the manuscript.

For example

Page 21, l. 407, have instead of has

Page 23, l. 455, was instead of were

Page 23, l. 464 ...and I would like...

DETAILED COMMENTS

1. Page 1, l. 3, Why does it say "Introduction" after the title?

2. Page 3, ll.47-56, This information does not add much to the understanding of the study.

3. Page 4, l. 78, There is a missing space before the source [20].

4. Page 5, l. 95, With state-of-the-art, one space is too many.

5. Page 5, ll.97-98, It is unusual for results to already be presented in the aim in the introduction.

6. Page 6, ll. 120-124, Why were different types of slurry used (especially for CHS experiment 1 and 2)?

7. Page. 7 l. 130, Table 1 is not mentioned in the text.

8. Page 7, l. 130, Why are the abbreviations for the slurry relevant? Are they even mentioned again in the manuscript?

9. Page 7, l. 140, Table 2 instead of Table 1, please check all cross-references again!

10. Page 8, l. 147, BS experiment 1: length 6 or 18 days; write out abbreviations such as H2SO4 or HCl in the text; explain mM in the table heading; BS experiment 3: H2SO4 better mentioned in 2nd place

11. Page 9; ll. 161-163, Show H2SO4 scrubber and Nafion tube in Figure 1 for better understanding.

12. Page 10, l. 165, Table 3 instead of Table 2

13. Page 10, l. 167, The correct spelling is 20 g HNO3 (69% purity) (kg slurry)-1 (please check in the whole manuscript).

14. Page 10, l. 168, Methane is also a greenhouse gas.

15. Page 10, l. 185-187, How was the pH measurement carried out? Was the vessel opened so that air could enter? Was the sample stirred, which also led to oxygen entering and thus influenced the pH development (stronger pH increase as mentioned in other studies)?

16. Page 11, l. 188, the letters "L", "H" and "B" were not explained

17. Page 11, l. 188, Why were different amounts of slurry used in experiment 1 and 2?

18. Page 11, Chapter 2.4 it is better to mention this chapter earlier in the material and methods section, as results have already been presented in Table 1. Alternatively, the results of the slurry characterisation could also be presented in the results chapter.

19. Page 12, l. 207, The abbreviation ATM was not explained

20. Page 14. l. 251, Why not experiment A, B and C instead of 1, 2 and 3. A better naming of the individual sub-experiments would help the reader to understand the manuscript.

21. Page 15, ll. 259-260, It is better to mention this in the material and methods section

22. Page 15, l. 261-262, It is better to mention this in the material and methods section

23. Page 15, l. 265, Why is the pH value of the variant acidified with HNO3 higher than that of the control?

24. Page, 16, l. 293, one point too many at the end of the sentence

25. Page 23, ll. 458-461, Although the authors correctly point out that a daily addition of fresh slurry makes acidification less efficient, this cannot be avoided in practice in the barn. Or do the authors have any ideas?

26. Fig. 2C, Are there any statistics for this illustration?

27. Fig. 3, The pH value normally increases more quickly after acidification of slurry. Presumably, the slower increase in experiment 1 can be explained by the old slurry > 2 months. This slurry is therefore not representative, especially if you think of an in-house slurry acidification.

28. Fig. 4A, Why do the CH4 emissions decrease on day 7/8 and increase again on day 12/13 after the interruption of the measurement period? The CO2 concentration is also lower between days 6-8? Could the H2SO4 scrubber have had an influence?

Reviewer #2: This study studied GHG emissions from acidified pig slurry with different organic and inorganic acids. Results showed HNO3 could result in N2O emissions and therefore not suitable. This study design followed a straightforward while rigorous manner by applying firstly a large amount of screening tests, followed by continuous headspace experiments. I find the results interesting. I have only some minor comments.

1. It was believed the N2O was due to denitrification of NO3-, was this N2O production/accumulation due to acidic condition or inherent partial denitrification?

2. There is no synergy between inorganic and organic acid. Is there synergy between H+ and NO3-/SO4- (as electron acceptors)?

3. You can also discuss the economic implications of these investigated acids.

Reviewer #3: This study investigated the effect of pig slurry acidification with a range of organic and inorganic acids with respect to their CH4 inhibitor potential in a number of batch experiments. There are some issues with this manuscript which need to be addressed before it can be accepted to publish, and they are listed below.

1. Line 97, the subscript should be corrected. Please check similar problems in the text.

2. In Table 1, could the difference in original TAN between the slurries affect the batch experiments?

3. Line 293, a redundant full stop.

4. Line 339, 344 and 348. I don’t know if the references format meet the requirements of PLOS ONE?

5. Line 386-387, have the denitrification activity been measured? If not, it is not appropriate to state like that.

6. Line 446-447, “It is likely that high concentrations of VFAs present prior to acidification reduce the efficacy of organic acid treatment.” The evidence to support the conclusion should be clarified in the text.

10 Apr 2022

In the text below we have addressed all comments from the editor and the reviewers.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

This has been checked and corrected

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

We don’t see any section called “Financial disclosure” in the online resubmission form. If this is the same as the “Financial statement” we would like it to be:

“This work was funded by the Danish Agricultural Agency under the Ministry of Environment and Food, Denmark (grant no. 33010-NIFA-19-725). The funder provided support in the form of salaries for authors Frederik R. Dalby, Lise B. Guldberg, Anders Feilberg, and Michael V.W. Kofoed but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

3. Thank you for stating the following in your Competing Interests section:

" No"

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state "The authors have declared that no competing interests exist.", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf.

We have included the statement in the cover letter

4. Thank you for stating the following in the Acknowledgments Section of your manuscript:

"The research was funding by the Danish Agricultural Agency under the Ministry of Environment and Food (Miljø- og Fødevareministeriet) (grant no. 33010-NIFA-19-725) and would like to thank for the opportunity to conduct this research. "

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

"NO"

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

We have removed the funding information from the acknowledgments section and the manuscript in general. We don’t see any section called “Funding Statement” in the online submission form.

We would like the following statement to be used, which is also included in the cover letter.

“This work was funded by the Danish Agricultural Agency under the Ministry of Environment and Food, Denmark (grant no. 33010-NIFA-19-725). The funder provided support in the form of salaries for authors Frederik R. Dalby, Lise B. Guldberg, Anders Feilberg, and Michael V.W. Kofoed but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

Comments to the Author

Reviewer #1: OVERVIEW AND GENERAL RECOMMENDATION

1. The study gives a comprehensive overview of the use of different acids for acidification on greenhouse gas emissions from slurry. The continuous addition of slurry, as is also common in barns, was also investigated. This leads to a special character of the manuscript. However, the studies are very extensive and are divided into several sub-studies. It is therefore difficult for the reader to follow the intentions of the authors.

Thanks for the comments in general. Our responses refer to line numbers in the revised manuscript with tracked changes and “all markup”.

We have tried to be more clear in stating the purpose of the study in the last section of the introduction. L103-111.

We do believe that an initial “simple” screening to select acids with best performance, followed by a more complex measurement to increase understanding, accuracy, and comparability to real pig house conditions is a logic and scientifically correct methodology to asses, the better acid agent for reducing greenhouse gases from pig slurry.

2. In my opinion, the title is not appropriate because it mentions organic acids. However, most of the study deals with the inorganic acids H2SO4 and HNO3. Even in the abstract, acetic acid is only briefly mentioned once. Additionally, the authors should point out to the reader which of the acids mentioned are inorganic or organic acids.

We do believe that the part of the manuscript where the organic acids were tested are an essential part of the work, since we here rule out the possibility that combining organic acids with e.g. sulfuric acid would add significant value. Acetate was specifically mentioned in the abstract as this was most comprehensively tested in the batch screenings and is also the organic acid being most frequently mentioned in the literature. In our opinion it is redundant to mention all organic acids in the abstract. We have instead specified in materials and methods which acids are organic acids and inorganic acids. See L160-162

“Organic acids included acetic acid, lactic acid, propanoic acid, citric acid, and formic acid whereas inorganic acids used were hydrochloric acid, sulfuric acid, nitric acid, and phosphoric acid.”

For the reasons above we do think the current title is a good description of the work presented and have therefore kept it in the revised version.

3. Why was the pH value of 5.5 in BS chosen?

We have added a phrase about this in L113-114

“In the batch screening experiments the target pH was 5.5, since this is the target pH of commercial in-house acidification systems [5].”

Other pH values could have been chosen and has also been tested in the literature. However, in commercial acidification systems in Denmark the target pH is normally 5.5 (NH4+ infarm technology, and JH acidification technology). For consistency and comparison purposes we chose this as the pH target also.

4. There are buffer systems in the slurry which lead to an increase in the pH value after acidification. This was not taken into account or mentioned. How practically relevant are the studies from BS in which acidification was only carried out once?

We agree with the reviewer that this is important and we have added a few lines in the results sections (L310-L312) where we comment on pH increase during the batches. In addition, we included Appendix A, where the pH by the end of the experiments is plotted versus the methane production. We included some discussion about the effects of pH and how batch assays can be used and compared to commercial continuous acidification systems (L412-417)

“This discrepancy could be related to differences in acidification target pH (pH 3.8-4.8 vs 5.5) and slurry type (cattle VS pig), which would affect buffer capacity and microbial community. Importantly buffer systems in slurry can result in pH increases subsequent to acidification. Hence, the CH4 reductions reported from the batch screening experiments are not comparable to commercial-in house acidification systems with continuous pH adjustment, but instead hints to the relative efficacy of the different acids in such a system.”

5. The figures look blurred. Please improve the resolution.

The PDF builder automatically reduces resolution of figures. Therefore to see the figures clearly the reviewer should download the “high resolution” version of the figures, by clicking the download tab in the PDF.

6. The actual data from which the mean values were formed were not provided.

We have included S4 Appendix with all the data used for making the figures.

7. It is recommended that the manuscript is read by a native speaker before submission, as there are some linguistic and grammatical errors in the manuscript.

For example

Page 21, l. 407, have instead of has

Page 23, l. 455, was instead of were

Page 23, l. 464 ...and I would like...

We have corrected these mistakes and other grammatical errors throughout the manuscript.

DETAILED COMMENTS

1. Page 1, l. 3, Why does it say "Introduction" after the title?

Corrected

2. Page 3, ll.47-56, This information does not add much to the understanding of the study.

We have reconsidered this part and agree that it is redundant. Therefore we have removed this paragraph and replaced it in L62-64

“Large CH4 reductions were previously observed from HNO3 acidified cattle slurry, but simultaneous increases in N2O emission attributed to denitrification activity was also detected [15,16]. “

In addition we shortened the paragraph about denitrification in L52-53

3. Page 4, l. 78, There is a missing space before the source [20].

Corrected

4. Page 5, l. 95, With state-of-the-art, one space is too many.

Corrected

5. Page 5, ll.97-98, It is unusual for results to already be presented in the aim in the introduction.

We included this to comply with the PLOS author guidelines for the introduction:

“Conclude with a brief statement of the overall aim of the work and a comment about whether that aim was achieved”

6. Page 6, ll. 120-124, Why were different types of slurry used (especially for CHS experiment 1 and 2)?

The reviewer is correct in that CHS1 (CHS A) could have been conducted with the same slurry as in CHS2 (CHS B).

However, as the slurry by the end of the experiment showed reduced methanogenic activity we chose a more potent “residual slurry” for the next CHS experiments. More activity in the slurry is necessary to see clear differences between treatments. We don’t believe this affects our conclusions since comparisons are only made within each experiment

We have added and explanation of the different in slurries in L137-139

“Due to decreasing methanogenic activity in the vacuum flushed slurry a residual slurry with higher methanogenic activity was used in CHS experiment B and C to ensure that potential differences in treatments were still detectable.”

7. Page. 7 l. 130, Table 1 is not mentioned in the text.

This is a mistake. It has been added in L135

8. Page 7, l. 130, Why are the abbreviations for the slurry relevant? Are they even mentioned again in the manuscript?

Thanks for catching this redundancy. Abbreviations has now been removed from the table.

9. Page 7, l. 140, Table 2 instead of Table 1, please check all cross-references again!

Corrected and checked

10. Page 8, l. 147, BS experiment 1: length 6 or 18 days; write out abbreviations such as H2SO4 or HCl in the text; explain mM in the table heading; BS experiment 3: H2SO4 better mentioned in 2nd place

Corrected: mM explained in table subheading. “6 or 18 days” added

We understand this as the reviewer prefers to replace chemical formulas with names (e.g. H2SO4 with sulfuric acid) only in the Table (not in the main text). This has been done in Table 2.

We have moved sulfuric acid to 2nd place in the table.

We have still kept the chemical formulas in the main text of the manuscript. We were unsure if the reviewer also thought these should be changed. We could not find any requirement about this in the guidelines.

11. Page 9; ll. 161-163, Show H2SO4 scrubber and Nafion tube in Figure 1 for better understanding.

This has now been included in Fig 1.

12. Page 10, l. 165, Table 3 instead of Table 2

Corrected

13. Page 10, l. 167, The correct spelling is 20 g HNO3 (69% purity) (kg slurry)-1 (please check in the whole manuscript).

Corrected

14. Page 10, l. 168, Methane is also a greenhouse gas.

Rephrased to avoid confusion about this. See L195-199.

15. Page 10, l. 185-187, How was the pH measurement carried out? Was the vessel opened so that air could enter? Was the sample stirred, which also led to oxygen entering and thus influenced the pH development (stronger pH increase as mentioned in other studies)?

We have added details about acidification procedure in L193-195 and general pH measurement in L216-218.

“Acidification in all CHS experiments were carried out by alternately pipetting acid into the slurry under agitation with a magnet stirrer followed by a pH measurement. This procedure was repeated until the pH reached and remained stable at the target pH over 5 minutes.”

“For all CHS experiments the pH was measured every 3–4th day by opening the incubation flasks completely and carefully submerging the pH sensor to 1 cm above the bottom of the incubation flask without stirring of the slurry.”

When only measuring the pH (no pH adjustment) the flasks were opened completely by taking of the lid and the bulk pH was measured by submerging the pH sensor 1 cm above the bottom. The slurry was not stirred during this process. The continuous headspace gas was air, so the slurry surface was exposed to oxygen at all times during the experiment, independently of the pH measurement (described in 2.4). Therefore pH measurements was not considered to influence much on the biochemical processes in the slurry, except for breaking the pH gradient at the place of submerging. This, however, is no different from barn conditions where fresh manure is excreted to the pit, disturbing the bulk slurry.

In the case of acidification and residual slurry acidification it was necessary to stir the manure intensely to ensure the target pH was correctly reached.

16. Page 11, l. 188, the letters "L", "H" and "B" were not explained

This information has been added in a Table footnote. B was referring to the manure type, earlier described in Table 1. But we have now removed “B” (and other letters referring to slurry types) to avoid redundancy.

17. Page 11, l. 188, Why were different amounts of slurry used in experiment 1 and 2?

The lower volume used was for practical reasons when acidifying the slurry, which created a lot of foam. Since the pH was decrease to 5.5 for all treatments in CHS experiment 2, more foam was expected.

We added this in L201-202

“Compared to CHS experiment A, the slurry mass was reduced (from 800 to 500 g) to avoid excessive foam formation during acidification.”

18. Page 11, Chapter 2.4 it is better to mention this chapter earlier in the material and methods section, as results have already been presented in Table 1. Alternatively, the results of the slurry characterisation could also be presented in the results chapter.

We moved up “Slurry characterization” as the first section in materials and methods.

19. Page 12, l. 207, The abbreviation ATM was not explained

Has been written out as “atmosphere”

20. Page 14. l. 251, Why not experiment A, B and C instead of 1, 2 and 3. A better naming of the individual sub-experiments would help the reader to understand the manuscript.

We have changed 1,2,3 to A,B, and C for batch experiments and CHS experiments throughout the manuscript

21. Page 15, ll. 259-260, It is better to mention this in the material and methods section

We removed this and included it in the M&M

22. Page 15, l. 261-262, It is better to mention this in the material and methods section

We removed this and included it in the M&M

23. Page 15, l. 265, Why is the pH value of the variant acidified with HNO3 higher than that of the control?

It could be due to denitrification, but the many processes that could occur (e.g. CO2 loss in the beginning) are blurring the picture of what caused this increase. We have written this into the manuscript L 310-312.

24. Page, 16, l. 293, one point too many at the end of the sentence

Corrected

25. Page 23, ll. 458-461, Although the authors correctly point out that a daily addition of fresh slurry makes acidification less efficient, this cannot be avoided in practice in the barn. Or do the authors have any ideas?

We are aware this is unavoidable. In the authors opinion it could mean that residual slurry acidification should have a lower target pH – perhaps even 4.5.

We added a phrase in L489-492

“The reduced effect of residual slurry acidification compared to bulk acidification suggest that residual slurry acidification should be carried out with a lower target pH than 5.5 to maintain the inhibiting effect once new slurry is added.”

26. Fig. 2C, Are there any statistics for this illustration?

No statistics were done here since all treatments were different, even the ones with the same organic acid used different concentrations of it. Therefore we chose to show all variants in Fig 2C.

27. Fig. 3, The pH value normally increases more quickly after acidification of slurry. Presumably, the slower increase in experiment 1 can be explained by the old slurry > 2 months. This slurry is therefore not representative, especially if you think of an in-house slurry acidification.

It is difficult to say how the pH developed in the start of CHS experiment 1 (now called CHS A), since it was measured for the first time after 1 week (the line is an interpolation). We are not convinced that general statements of how fast pH increases can be made in this case where 1) HNO3 was used and at different pH values. However, the reviewer might have a point that the slurry is not completely representative of barn conditions, but within CHS A, the different treatments are still comparable.

28. Fig. 4A, Why do the CH4 emissions decrease on day 7/8 and increase again on day 12/13 after the interruption of the measurement period? The CO2 concentration is also lower between days 6-8? Could the H2SO4 scrubber have had an influence?

This is a mistake in the data treatment and we are glad the reviewer caught this mistake.

The CH4 increase and decrease is due to the scrubber being applied and removed. Since values are averaged over 12 h intervals to make plots more easily interpretable (now specified in the statistics section). The concentration on Fig 4A was influenced also outside the intervals of the scrubber. We have corrected this in the figure now to avoid confusion of whether the CH4 is increasing/decreasing or if it is due to the scrubber being used or not. The decrease in CH4 emission by the end of the experiment (in the untreated sample) is probably due to substrate depletion or depletion of easily degradable substrate.

For CO2, we have since found that some VOCs may have interfered with CO2 concentrations and therefore only periods with the scrubber + nafion tube applied is now shown. Our background measurements of CO2 and air shows expected CO2 values. We have cleared this out in the new Fig 4 and explained the interference in L196-199

“In CHS experiment A Nitrous oxide was measured continuously over two weeks, but CH4 and CO2 were only measured periodically when a Nafion filter was applied to remove NH3 and possibly other volatile organic compounds, that caused spectral interference with with CH4 and CO2.”

Reviewer #2: This study studied GHG emissions from acidified pig slurry with different organic and inorganic acids. Results showed HNO3 could result in N2O emissions and therefore not suitable. This study design followed a straightforward while rigorous manner by applying firstly a large amount of screening tests, followed by continuous headspace experiments. I find the results interesting. I have only some minor comments.

Thanks for the comments in general. Our responses refer to line numbers in the revised manuscript with tracked changes and “all markup”.

1. It was believed the N2O was due to denitrification of NO3-, was this N2O production/accumulation due to acidic condition or inherent partial denitrification?

This is a good question. Since low pH inhibit N2O-reductase (N2O to N2) we would expect low N2O production with higher dose of HNO3 (lower pH). However, as seen from Fig 4 the lower pH also reduced CO2 production, which indicates that microbial activity was inhibited as well. We believe therefore that the inhibition of microbes able to carry out partial denitrification was inhibited at acidic conditions.

In L442-446 we added the following

“It has been shown that enzymatic N2O reductase activity is significantly reduced at low pH [29], hereby decreasing the enzymatic reduction of N2O to N2. However, that fact that CO2 production was reduced below pH 6, suggests that general inhibition of heterotrophic activity (including heterotrophic denitrifiers) is a more plausible explanation for the reduced N2O emission below pH 6.”

2. There is no synergy between inorganic and organic acid. Is there synergy between H+ and NO3-/SO4- (as electron acceptors)?

This is a good question and we don’t believe that our data can support any conclusion about this. It has previously been shown that sulfate inhibits methanogenesis it self – possibly due to SO4 being preferable electron acceptor compared to CO2 for methanogenesis.

In order to answer the question we should have conducted experiments with pure SO4 and NO3 addition and compared to that of H2SO4 and HNO3. However, this was out of the scope, but certainly an interesting point.

3. You can also discuss the economic implications of these investigated acids.

Good point. We have, however, chosen not to go into the economics of acidification as a good cost benefit analysis is beyond the scope.

Reviewer #3: This study investigated the effect of pig slurry acidification with a range of organic and inorganic acids with respect to their CH4 inhibitor potential in a number of batch experiments. There are some issues with this manuscript which need to be addressed before it can be accepted to publish, and they are listed below.

Thanks for the comments in general. Our responses refer to line numbers in the revised manuscript with tracked changes and “all markup”.

1. Line 97, the subscript should be corrected. Please check similar problems in the text.

Corrected

2. In Table 1, could the difference in original TAN between the slurries affect the batch experiments?

For the batch experiments the same slurry was used, so there is not any difference in the TAN concentration here (Table 1 in first data row). If the reviewer refers to the continuous headspace experiments, there was a difference between CHS1 vs CHS2 + CHS3. However, in these experiments the methanogenic potential as well as dry matter and pH was also different. We believe it is more likely that other factors than TAN affected the experiments.

In “Astals et al. 2018. Characterising and modelling free ammonia and ammonium inhibition in anaerobic systems” TAN inhibition was studied in pig slurry and using their equations it is unlikely that TAN was inhibiting much on methanogenesis in our study.

In any case CHS experiments were compared only within and not among experiments, and therefore any possible TAN inhibition would be equally effecting the experiments.

3. Line 293, a redundant full stop.

We combined the sentences into one

4. Line 339, 344 and 348. I don’t know if the references format meet the requirements of PLOS ONE?

This has been corrected.

5. Line 386-387, have the denitrification activity been measured? If not, it is not appropriate to state like that.

We measured it indirectly by measuring N2O production. Previous studies with HNO3 addition to slurry found that denitrification was the main source of N2O. See Oenema, O. and Velthof, G.L. 1993 (cited in manuscript).

However we softened the sentence by replacing “accounted for“ to “corresponded to” in L437

6. Line 446-447, “It is likely that high concentrations of VFAs present prior to acidification reduce the efficacy of organic acid treatment.” The evidence to support the conclusion should be clarified in the text.

This conclusion, which is still soft (using the word “likely”), is based on L389-401 under the discussion of the batch experiments.

Submitted filename: response to reviewers.docx

Click here for additional data file.

14 Apr 2022

Reducing greenhouse gas emissions from pig slurry by acidification with organic and inorganic acids

PONE-D-22-02024R1

Dear Dr. Dalby,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Xueming Chen

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

26 Apr 2022

PONE-D-22-02024R1

Reducing greenhouse gas emissions from pig slurry by acidification with organic and inorganic acids

Dear Dr. Dalby:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Xueming Chen

Academic Editor

PLOS ONE