NUTRITIONAL MANIPULATION OF SWINE DIETS TO REDUCE HYDROGEN SULFIDE EMISSIONS

INTRODUCTION

Hydrogen sulfide gas is one of hundreds of known odor producing compounds emitted from livestock facilities. The smell of hydrogen sulfide is best described as a "rotten egg" smell. Hydrogen sulfide gas is heavier than air and tends to be in highest concentrations at the surface of manure storage containments. However, unlike other gases produced in livestock operations, it can be deadly if high concentrations (over 100,000 ppb) are present. Generally, this only occurs when anaerobic deep pits beneath swine confinement buildings are agitated during manure removal and adequate ventilation is not provided. Low levels of hydrogen sulfide can cause headaches, dizziness, nausea, or insomnia. Because of these potential adverse health effects, it has become the latest environmental component being studied and monitored near swine facilities in 23 states including Minnesota.

The Minnesota Department of Health considers a level of less than 50 ppb hydrogen sulfide of ambient air exposure in one hour to be safe for humans to breathe. This level is not an established standard yet in Minnesota, but is considered a maximum safe value for monitoring purposes. However, OSHA has established a level of 10,000 ppb hydrogen sulfide exposure per eight-hour day to be the maximum tolerable level in the work place for worker safety.

Currently, MPCA is monitoring hydrogen sulfide levels on livestock farms. Hydrogen sulfide levels on most farms fall below the 50 ppb maximum, but some farms occasionally have a brief increase in emissions before dropping back to less than 50 ppb. Furthermore, we do not understand why hydrogen sulfide is a problem on some farms and not on others. The concern over hydrogen sulfide emissions from confinement swine operations has made it more difficult for producers to obtain building permits because permits cannot be issued if there is knowledge that a state environmental standard is going to be violated.

The purpose of this paper is to review our current understanding of hydrogen sulfide production and odor, as well as the influence of diet formulation and feeding programs for reducing hydrogen sulfide gas in swine facilities.

SULFUR NUTRITION AND METABOLISM IN THE PIG

Sulfur is an essential element for the pig, and the sulfur requirement appears to be adequately met by providing adequate levels of the sulfur containing amino acids methionine, cystine, and cysteine, and the water soluble viatmins biotin and thiamin. Certain mucopolysaccharides, including chondroitin sulfate and the mucoitin sulfates, heparin, glutathione, taurine, and coenzyme A are also important sulfur containing metabolic compounds. Because proteins are present in every cell of the body, and S-containing amino acids are components of almost all proteins (usually 0.6 to 0.8% of the protein), sulfur is widely distributed throughout the body and is found in every cell to make up about 0.15% of body weight. Sulfur functions mainly through its presence in organic metabolites. Inorganic sulfate from exogenous dietary sources and from endogenous release from S-containing amino acids is used in synthesizing the chondroitin matrix of cartilage, in biosynthesis of taurine, heparin, cystine, and other organic constituents of the animal body (Baker, 1977).

Inorganic sulfur is considered to be essential for animals for normal maintenance or productive functions. Furthermore, because intestinal absorption of inorganic sulfur compounds is low, sulfur toxicity is not a practical problem. Absorption of inorganic sulfate from the gastrointestinal tract is inefficient. Active transport of SO2-4 takes place from the upper small intestine. Both inorganic and organic forms of sulfur are used for sulfation of cartilage mucopolysaccharides. Organic forms of S are absorbed readily relative to all sulfur compounds. Inorganic sulfur is excreted via feces and urine. Unabsorbed sulfur is likely reduced in the lower GI tract and excreted as sulfate. Urinary sulfur is present mainly as inorganic SO2-4, but also as a component of thiosulfate, taurine, cystine, and other organic compounds. Because the bulk of body sulfur is present in amino acids, it is not surprising that urinary S excretion tends to parallel urinary N excretion. High protein diets are associated with large amounts of urinary sulfur and nitrogen.

Banwart and Bremner (1975b) were able to detect only one (dimethyl sulfide) of six sulfur containing gases in fresh swine manure using gas chromatography methodologies. Reduction of inorganic sulfate to sulfide occurs to a limited extent in nonruminants (Kline et al., 1971). Therefore, it appears that most of the production of hydrogen sulfide and other volatile sulfur containing gases occurs as a result of microbial fermentation during manure storage.

PRODUCTION OF GASES AND ODOR IN STORED MANURE

Once urine and feces enter an anaerobic pit, numerous chemical transformations occur. For example, urea is hydrolyzed to ammonia and CO2, and sulfate is reduced to hydrogen sulfide. Plant fiber and protein are also anaerobically degraded to low molecular weight compounds during anaerobic digestion. Over 150 volatile odorous compounds have been identified in swine manure, and most are presumed to be products of anaerobic microbial degradation of waste (Spoelstra, 1980). Of these 150 volatile compounds, the following compounds are considered to be the main components responsible for offensive odors in swine waste (Spoelstra, 1980).

Air Components	Waste Components
ethanoic acid	ethanoic acid
propanoic acid	propanoic acid
butanoic acid	butanoic acid
	2-methyl propanoic acid
	pentanoic acid
	2-methylbutanoic acid
	3-methylbutanoic acid
	2-methylpentanoic acid
phenol	phenol
	3-methylphenol
4-methylphenol	4-methylphenol
4-ethylphenol	4-ethylphenol
	benzoic acid
indole
skatole	skatole
	phenylacetic acid
diethyldisulfide*
propanethiol*
butanethiol*
dipropyldisulfide*
2-methylthiophene*
propylprop-1-enyldisulfide *
2,4 dimethylthiophene*
2-methylfuran

* Indicates sulfur containing compounds.

Notice that approximately 50% of the compounds considered to be primary contributors to swine odors in air of confinement swine facilities contain sulfur. The reason that some compounds are undetectable in air but are found in swine waste is because of chemical reactions between these compounds and the atmosphere. The reduced sulfur compounds are very reactive in air.

In addition to the sulfur compounds listed above, other sulfur containing compounds are also present in air and waste of swine facilities. During manure storage, sulfate is reduced to hydrogen sulfide (Riviere, 1974). The following are additional sulfur containing compounds that have been detected in air and pig manure:

 

Air Components	Waste Components
carbonyl sulfide
hydrogen sulfide	hydrogen sulfide
methanethiol	methanethiol
dimethylsulfide	dimethylsulfide
	diethylsulfide
dimethyldisulfide	dimethyldisulfide
dimethyltrisulfide	dimethyltrisulfide
	ethanethiol
diethyldisulfide	diethyldisulfide
propanethiol	propanethiol
butanethiol	butanethiol
dipropyldisulfide
2-methylthiophene
propylprop-1-enyldisulfide
2,4-dimethylthiphene
2-methylfuran

Most of these compounds are present only in trace amounts (Banwart and Bremner, 1975a). Hydrogen sulfide and methyl mercaptan are most frequently reported as constituents of pig waste and are quantitatively the most important S-containing volatile constituents (Spoelstra, 1980). In ventilation air, only traces of these compounds have been reported (Schaefer et al, 1974; Avery et al, 1975). This is probably due to oxidation of mercaptans to the less volatile disulfides by air (Kadota and Ishida, 1972) and possibly by adsorption. Hydrogen sulfide is likely to originate mainly from microbial reduction of sulfate. Urine contains about 1100 mg/l of sulfur, mainly as sulfate, which originates from animal metabolism (Loehr, 1974). Sulfate reducing organisms have been found to be present in pig wastes in amounts up to 103-104 per ml (Riviere et al., 1974). Sulfate reducing bacteria have been shown to produce trace amounts of carbon disulfide, carbonyl sulfide, and methyl, ethyl and propyl mercaptans (Hatchikan et al., 1976). In addition, hydrogen sulfide can be produced by microbial degradation of cysteine and cystine (Freney, 1967; Riviere et al, 1974). Carbon disulfide and diethyl sulfide have been reported as products from cysteine. Methionine is decomposed mainly to methyl mercaptan and dimethyl sulfide (Freney, 1967; Kadota and Ishnada, 1972). Most of the other identified S-containing volatiles seem to be derived from more seldomly occurring amino acids like substituted cysteine, which occur in plants (Meister, 1965; Freney, 1967).

CHEMISTRY OF HYDROGEN SULFIDE PRODUCTION AND OTHER VOLATILE SULFUR COMPOUNDS DURING MANURE STORAGE

Sulfur is found in a variety of chemical forms and is interconverted between forms depending on chemical conditions in manure storage systems. Figure 1 shows interconversions that occur in the sulfur cycle (Sawyer and McCarty, 1978).

Sulfates are indirectly responsible for odor and corrosion of waste handling systems resulting from reduction of sulfates to hydrogen sulfide under anaerobic conditions (Sawyer and McCarty, 1978).

Sulfates serve as a source of electron acceptors for biochemical reactions produced by anaerobic bacteria in the absence of dissolved oxygen and nitrate. Sulfate ions are reduced to sulfide ions, which establish an equilibrium with hydrogen ions to form hydrogen sulfide based on its primary ionization constant. Thus, depending on pH of the slurry, the chemical form of sulfur compounds can be very different. When pH of slurry is 8 or more (basic), most reduced sulfur exists in solution as HS- and S2- ions, and the amount of free H2S is so small that odor problems do not occur. At a pH below 8, equilibrium shifts rapidly toward formation of un-ionized H2S and is about 80% complete at pH 7. Under these conditions, the partial pressure of hydrogen sulfide becomes great enough to cause significant odor problems whenever sulfate reduction produces significant quantities of sulfide ion. Figure 2 shows the relationship of pH on hydrogen sulfide equilibrium.

 Hydrogen sulfide production in swine confinement finishing units has been shown to be highly correlated with average outside air temperature, ratio of pit area to building volume, air exchange rate for the building and daily dietary sulfur intake (Avery et al., 1975). Anaerobic fermentation is essential for production of hydrogen sulfide gas, as well as most other volatile sulfur gases, except dimethyl sulfide which can be produced under aerobic fermentation (Banwart and Bremner, 1975b). Methyl mercaptan and hydrogen sulfide have been shown to be the predominant (80%) volatile S gases produced under simulated anaerobic fermentation conditions (Banwart and Bremner, 1975b). However, during a 30-day incubation period, only 0.03% of total S present in swine manure was volatilized to sulfur gases (Banwart and Bremner, 1975b).

The major proportion of sulfur in the diet is provided by the sulfur containing amino acids methionine, cystine, and cysteine. Hydrogen sulfide, methyl mercaptan, dimethyl sulfide, and dimethyl disulfide are produced by microbial decomposition of methionine and other sulfur compounds present in manure (Freney, 1967; Kadota and Ishida, 1972). For example, DL-methionine is decomposed by bacteria using the following pathway (Kadota and Ishida, 1972):

Carbon disulfide is produced by microbial decomposition of cystine and cysteine in manure (Banwart and Bremner, 1975b). However, it is not known how carbonyl sulfide is produced because there is no evidence in the literature that suggests that this compound is produced by microbial decomposition of sulfur compounds.

PROBLEMS ASSOCIATED WITH MEASURING SULFUR COMPOUNDS IN SWINE MANURE

Reduced sulfur compounds are very reactive in air (Spoelstra, 1980). Therefore, there is rapid interconversion of various sulfur forms. In addition, hydrogen sulfide is one of the two most volatile sulfur containing compounds, making it difficult to measure accurately. Most analytical methodologies for measuring sulfur, sulfate, and sulfide compounds have been developed for soil tests. There are two standard methods that have been used to measure sulfates in waste water: gravimetric and turbidimetric procedures (Sawyer and McCarty, 1978). The gravimetric method is considered to be the most accurate method (Sawyer and McCarty, 1978) and the method that we have used in our research. However, comparing the fecal and urine sulfur and sulfate excretion values obtained in our studies, it is a concern that levels of total fecal and urine sulfur excretion expressed as mg/pig/day are lower than absolute values for fecal and urine sulfate excretion (Table 6 and 7). Therefore, we question the accuracy and application of current standard analytical methods for sulfate measurements in swine feces and urine.

Sulfides have typically been measured using colorimetric or volumetric methods. Both methods have significant limitations resulting in determining the sulfide content by difference after separating inorganic from organic sulfur, and elemental sulfur from sulfate to determine percentage of sulfide. We are currently experimenting with a new technique using a sulfide electrode manufactured by Cole-Parmer in an attempt to be able to quickly, simply, accurately and economically measure sulfide ions in aqueous solutions rather than using laborious, expensive, and questionable chemical analysis procedures to separate various sulfur forms in feces and urine. Hopefully, this technique will provide more rapid, precise measurements of sulfide concentrations in manure samples than current chemical analysis methodologies.

CORRELATION OF HYDROGEN SULFIDE WITH ODOR AND FEED AND WATER SOURCES ON COMMERCIAL SWINE FARMS

 We recently completed a study (Jacobson et al., 1998, unpublished) to determine if sulfur in swine feed and drinking water on commercial swine farms could be correlated with hydrogen sulfide emmissions from manure storage systems. Six commercial swine farms representing three different manure storage types (earthen basin, outdoor pit or tank, and indoor deep pit), which were part of an ongoing odor monitoring project, were used in this study. Three farms were located in southwestern Minnesota and the other three farms were located in central Minnesota. Each pair of farms with common manure storage types were compared. Air samples were collected at the surface of the manure storage unit and analyzed for odor threshold using an olfactometer, and a hydrogen sulfide was measured using a Jerome^TMmeter or sensidyne indicator tubes. Similar odor levels were obtained for each pair of the three manure storage types compared. However, one farm within each manure storage type had a much higher concentration of hydrogen sulfide than the other paired site. To determine if sulfur content of water and feed samples were related to odor and hydrogen sulfide levels on these farms, samples of feed and water were collected and analyzed for total sulfur content. Estimated sulfur consumption and excretion were calculated and weighted based upon number of pigs in various phases of production and their expected consumption and manure excretion levels.

Hydrogen sulfide levels for all six farms during spring, summer, and fall seasons are shown in Figure 3. Note that each pair of farms had either a high or low level of hydrogen sulfide gas within manure storage type. The fall season had the highest hydrogen sulfide levels among the three seasons for air sampled directly above the manure storage surface. No single manure storage type seems to have higher hydrogen sulfide levels than another. These two observations are consistent with those observed with a larger, ongoing odor monitoring project being conducted at the University of Minnesota.

Variation in odor threshold level for these three paired manure storage units during spring, summer, and fall are shown in Figure 4. There were only small differences in odor levels between paired farms, and odor levels were much higher during spring (April and May) compared to those measured during summer and fall. Odor levels are generally higher in spring because increasing temperatures accelerate microbial activity for decomposing "poorly fermented" organic compounds that have accumulated during colder months, and subsequently produces large quantities of odorous gases. These data, along with those obtained from a larger U of M odor monitoring study, suggest that there does not appear to be a particular manure storage system that consistently produces high odor levels. Furthermore, correlation between odor levels and hydrogen sulfide levels is poor as shown in Figure 5 and Table 1. Thus, these preliminary data suggest that if high sulfate drinking water and high dietary levels of sulfur are found in on-farm production conditions, hydrogen sulfide levels may be increased on some farms compared to farms with low sulfur feed and water.

Table 1. Comparison of manure storage type, average hydrogen sulfide level, and total sulfur in water and feed fed to swine on commercial swine farms in Minnesota

 

Farm	Average Hydrogen Sulfide Level, ppm	Water Sulfur, ppm	Feed Sulfur x 103, ppm
Earthen basin 1	22	331	244
Earthen basin 2	3	6	278
Outdoor tank 1	37	223	412
Outdoor tank 2	4	195	344
Indoor pit 1	25	4	225
Indoor pit 2	5	7	186

DIETARY SOURCES AND LEVELS OF SULFUR

Sulfur content of common swine feed ingredients ranges from 0.02% in dried bakery product to 1.59% in low lactose dried whey (NRC, 1998). However, since most sulfur values published in NRC (1998) were determined several years ago, and many are based on only a few observations, we chose to analyze all ingredients used in complex starter diets for total sulfur content and compare these values with those published in NRC (1988 and 1998). The results of this comparison are shown in Table 2. We used a LECO procedure to determine sulfur values of each ingredient shown in the analyzed column. Note that our analyzed values for sulfur were very similar to NRC (1988, 1998) values except for a much lower sulfur value for spray dried whey (edible grade) and a somewhat higher value for spray dried blood meal. The discrepancy for dried whey is likely due to potential differences in nutrient levels of various sources and grades coupled with natural analytical variability. The discrepancy for spray dried blood meal is likely due to variability in product quality among sources. NRC does not list sulfur values for spray dried porcine plasma, IPC 790 fish meal, tricalcium phosphate, D,L methionine, or copper sulfate. Therefore, sulfur values were either calculated based on sulfur amino acid content listed on ingredient specification sheets for spray dried porcine plasma; or in the case of IPC 790 fish meal and tricalcium phosphate, taken directly from product specification sheets; or calculated using the chemical formula (copper sulfate) assuming 100% purity. Note that calculated and product specification values did not match our analyzed values very closely except for tricalcium phosphate. These results suggest that NRC sulfur values are accurate and can be used for most commonly used ingredients in starter diets. However, analysis of spray dried plasma, spray dried blood meal, fish meal, and spray dried whey should be conducted to establish reliable values of sulfur content when selecting ingredient sources to minimize sulfur content of starter diets.

Ingredient	%S (NRC, 1988)	% S (NRC, 1998)	% S Analyzed
Corn, yellow dent	0.11	0.13	0.11
Dried whey, edible	1.04	0.72	0.21
Soybean meal, 44%	0.42	0.43	0.39
Lactose	0.00	-	0.01
Oat groats	0.20	0.20	0.19
Spray dired porcine plasma	0.75*	-	1.05
IPC 790 fish meal	0.45**	-	0.93
Skim milk, dried	0.32	0.32	0.38
Choice white grease	0.00	-	0.00
Spray dried blood meal	0.34	0.47	0.51
Dicalcium phosphate	0.80	0.80	0.84
Tricalcium phosphate	0.00**		0.02
Limestone	0.00	0.04	0.02
Salt	0.00	0.00	0.00
Mecadox-10	-	-	0.05
U of M vitamin premix	0.00**	-	0.20
D, L methionine	21.49**	-	8.46
High sulfur TM premix	6.20*	-	8.21
Low sulfur TM premix	0.00*	-	0.05
L-lysine HCL	0.00**	-	0.00
Choline chloride	0.00*	-	0.05
Copper sulfate	12.84*	-	9.39
Zinc oxide	0.00*	-	0.03

* Values were calculated.
** Values were obtained from supplier specification sheets.

RELATIVE CONTRIBUTION OF FEED INGREDIENTS TO THE TOTAL SULFUR CONTENT OF PHASE I, PHASE II AND PHASE III DIETS 

We formulated experimental high and low sulfur Phase I, II, and III diets using book values from NRC (1988) and obtained remaining sulfur values from product specification sheets if not provided in NRC (1988). Using published values shown in Table 2, we calculated that our high sulfur, conventional diet should contain 0.45% sulfur and our modified low sulfur diet should contain 0.24% sulfur. In other words, by adjusting inclusion rates of some ingredients that contribute major amounts of sulfur, we theoretically should achieve a 47% reduction in sulfur content while satisfying all other required nutrient levels. However, due to the wide disparity between NRC (1988) and our analyzed sulfur values for dried whey and IPC 790 fish meal, the actual reduction of total sulfur in our low sulfur Phase I diet was only 13%. This points out the importance of analyzing each ingredient for sulfur content before formulating low sulfur diets.

As shown in Table 3, the primary sulfur contributors in the high sulfur Phase I diet were:

1. Spray dried plasma (20.5%)
2. Fish meal (15.3%)
3. Spray dried whey (13.7%)
4. Soybean meal (13.0%)
5. Corn (9.8%)
6. Oat groats (6.2%)
7. Skim milk (6.2%)

These ingredients collectively account for nearly 85% of total sulfur in this diet formulation. Dicalcium phosphate, D,L methionine, TM premix and copper sulfate contribute the remaining 15% of sulfur to the diet. By eliminating dried whey from the formulation, replacing dicalcium phosphate with tricalcium phosphate, and copper sulfate with zinc oxide, and using a no sulfate TM premix, small improvements were achieved in reducing total diet sulfur content.

A 39% reduction in total sulfur content of the low sulfur Phase II diet was achieved by minimizing the use of high sulfur ingredients, and replacing them with lower sulfur alternatives (Table 4). However, only a 19% reduction in total sulfur was achieved for the low sulfur Phase III diet compared to the high sulfur Phase III diet due to the use of fewer ingredients and less flexibility for diet manipulation (Table 5).

Table 3. Diet composition and percentage contribution of each ingredient to the total sulfur content of a high and low sulfur Phase I starter diet

Ingredient	High Sulfur Diet, %	% of Total Dietary S	Low Sulfur Diet, %	% of Total Dietary S
Corn, yellow dent	27.334	9.8	28.990	12.2
Dried whey, edible	20.000	13.7	0.000	0.0
Soybean meal, 44%	10.250	13.0	12.250	17.9
Lactose	10.000	0.3	21.500	0.8
Oat groats	10.000	6.2	10.000	7.3
Spray dried porcine plasma	6.000	20.5	6.000	24.0
IPC 790 fish meal	5.000	15.3	5.000	17.9
Skim milk, dried	5.000	6.2	10.000	14.5
Choice white grease	3.000	0.0	2.250	0.0
Spray dried blood meal	0.000	0.0	0.000	0.0
Dicalcium phosphate	1.400	3.9	0.000	0.0
Tricalcium phosphate	0.000	0.0	1.900	0.0
Limestone	0.650	0.0	0.150	0.0
Salt	0.000	0.0	0.000	0.0
Mecadox-10	0.400	0.0	0.400	0.0
U of M vitamin premix	0.300	0.3	0.300	0.4
D, L methionine	0.150	4.2	0.160	5.3
High sulfur TM premix	0.150	3.9	0.000	0.0
Low sulfur TM premix	0.000	0.0	0.056	0.0
L-lysine HCL	0.150	0.0	0.150	0.0
Choline Chloride	0.116	0.0	0.116	0.0
Copper sulfate	0.100	2.9	0.000	0.0
Zinc oxide	0.000	0.0	0.278	0.0
Nutrient level
Calculated sulfur, %	0.45	100.00	0.24	100.00
Analyzed sulfur, %	0.32	100.00	0.28	100.00

EFFECT OF DIET MANIPULATION ON SULFUR EXCRETION, GROWTH PERFORMANCE, ODOR, AND HYDROGEN SULFIDE EMISSIONS

After an extensive literature review, we were unable to find any published studies related to diet manipulation on sulfur retention and excretion, growth performance, or odor and hydrogen sulfide emissions. Therefore, we conducted a series of trials with the following objectives:

1. Quantify the amount of sulfur consumed, retained, and excreted from pigs weaned at 18 days of age when fed either a typical or low sulfur Phase I, Phase II, and Phase III diet sequence.

2. Quantify the amount of elemental sulfur, sulfate, and sulfide consumed in feed and excreted in feces and urine when feeding a typical and low sulfur Phase I, Phase II and Phase III diet.

3. Determine the effect of feeding a typical and low sulfur diet on energy and nitrogen retention during Phase I, Phase II, and Phase III.

4. Determine the effects of feeding low sulfur diets on pig performance.

5. Determine the effects of feeding low sulfur diets on odor, hydrogen sulfide and other gas levels in confinement nursery rooms.

 Procedures for Experiment I

A three-phase nursery diet feeding sequence of "typical," high sulfur diets and modified, low sulfur diets shown in Tables 4 and 5 was used. These diets provided equivalent concentrations of nutrients except sulfur. Based on our initial calculations using NRC (1988) sulfur values for feed ingredients, our modified, low sulfur diet should have reduced total sulfur consumption by 30 %. The greatest percentage of this reduction in sulfur consumption should have been obtained in Phase I (53.5%) and Phase II (55%), followed by only a modest reduction in sulfur consumption during Phase III (12.5%). Therefore, we hypothesized that if sulfur digestibility/bioavailability is similar between "typical," high sulfur diets and modified, low sulfur diets, and if sulfur values of feed ingredients published in NRC (1988) were accurate, these dietary modifications would reduce sulfur excretion by 30%. 

A total of 20 PIC barrows (10 pigs/treatment) were weaned at 18 days of age at the St. Paul Swine Research unit. Pigs were weighed, blocked by weight and litter, and assigned within block to one of two dietary treatment sequences. Pigs were placed in individual stainless steel collection cages and were fed either the high sulfur or low sulfur Phase I diet for seven days. Pigs were fed an amount of feed from their respective experimental diet, equivalent to 2% of their initial body weight twice daily. Total fecal and urine excretion was collected for three days (day 5 to 7) and stored for later laboratory analysis. 

On day 8, pigs fed the high sulfur Phase I diet were weighed and switched to the high sulfur Phase II diet for a 14-day feeding period. Similarly, pigs fed the low sulfur Phase I diet were weighed and switched to the low sulfur Phase II diet. Pigs were again fed an amount of experimental diet equivalent to 2% of their body weight on day 8 for one week, and urine and feces were collected from day 12 to 14. This same procedure was used for each of the five weekly collection periods for pigs fed both experimental diets in all phases.

Table 4. Diet composition and nutrient values of high sulfur and low sulfur Phase I and Phase II experimental diets

Ingredient	Phase I High Sulfur Diet, %	Phase II High Sulfur Diet, %	Phase I Low Sulfur Diet, %	Phase II % of Total Dietary S
Corn, yellow dent	27.334	49.484	28.990	48.350
Dried whey, edible	20.000	20.000	-	-
Soybean meal, 44%	10.250	17.000	12.250	23.500
Lactose	10.000	-	21.500	14.000
Oat groats	10.000	6.2	10.000	-
Spray dried porcine plasma	6.000	-	6.000	-
IPC 790 fish meal	5.000	5.000	5.000	5.000
Skim milk, dried	5.000	-	10.000	-
Choice white grease	3.000	3.000	2.250	2.800
Spray dried blood meal	-	2.500	-	2.500
Dicalcium phosphate	1.400	1.250	-	-
Tricalcium phosphate	-	-	1.900	1.950
Limestone	0.650	0.550	0.150	0.100
Salt	-	-	-	0.400
Mecadox-10	0.400	0.400	0.400	0.400
U of M vitamin premix	0.300	0.300	0.300	0.300
D, L methionine	0.150	-	0.160	-
High sulfur TM premix	0.150	0.150	-	-
Low sulfur TM premix	-	-	0.056	0.156
L-lysine HCL	0.150	0.150	0.150	0.150
Choline Chloride	0.116	0.116	0.116	0.166
Copper sulfate	0.100	0.100	-	-
Zinc oxide	-	-	0.278	0.278
Nutrient level
Crude protein, %	-	-	-	-
Lysine, %	-	-	-	-
Methionine + cystine, %	-	-	-	-
Calculated Sulfur, %	0.45	0.40	0.24	0.18
Analyzed Sulfur, %	0.32	0.31	0.28	0.19

Table 5. Diet Composition and Nutrient Values of High Sulfur and Low Sulfur Phase III Diets

Ingredient	Phase III High Sulfur Diet, %	Phase III Low Sulfur Diet, %
Corn, yellow dent	57.834	58.100
Soybean meal, 44%	34.000	34.000
Choice white grease	4.00	4.000
Dicalcium phosphate	1.700	-
Tricalcium phosphate	-	1.750
Limestone	0.850	0.350
Salt	0.400	0.400
Mecadox-10	0.400	0.400
U of M vitamin premix	0.300	0.300
High sulfur TM premix	0.150	-
Low sulfur TM premix	-	0.056
L-lysice HCL	0.150	0.150
Choline chloride	0.116	0.116
Copper sulfate	0.100	-
Zinc oxide	-	0.278
Nutrient Levels
Crude protein, %	-	-
Lysine, %	-	-
Methionine + cystine, %	-	-
Calculated Sulfur, %	0.24	0.21
Analyzed Sulfur, %	0.26	0.21

Feed samples of each diet, as well as fecal and urine samples from each collection period, were analyzed for nitrogen, sulfur, sulfate, sulfide content. Gross energy was determined for only feed and feces. Sulfur content of feed and feces was determined using a Total Sulfur Analyzer manufactured by LECO Corporation, St. Joseph, MI. Sulfur content of urine was determined using Inductively Coupled Plasma Atomic Emission Spectrophotometry (Perkin-Elmer, Norwalk, CT). Sulfate was analyzed using a Gravimetric method with ignition of residue, which is a standard method for evaluation of sulfate in water and wastewater. We devoted considerable effort toward finding acceptable methods for measuring sulfide in feed, feces and urine. Although we developed a colorimetric assay for sulfide determination in urine, it could not be successfully applied toward measuring sulfide in feces because it is difficult to determine color change in cloudy, high organic matter solutions. Thus, precision of this measurement was poor and sulfide data were not included in this study.

 Results from Experiment 1

As expected, feeding the low sulfur diets tended to reduce total sulfur consumption during Phase I, and did reduce sulfur consumption during Phase II, Phase III and overall average sulfur consumption for the entire 5-week experiment (Table 6). The reason for lack of a significant reduction of total sulfur intake during Phase I was a result of using ingredient values for sulfur from NRC (1988) instead of actual analyzed values of sulfur from ingredients used in the formulation. Surprisingly, total fecal sulfur excretion of pigs fed the low sulfur diet was higher compared to pigs fed the high sulfur diet in each phase and for the overall 5-week feeding period, while total urinary sulfur excretion and total sulfur excretion (fecal + urine) were reduced in all feeding phases by feeding the low sulfur diet sequence (Table 6). Although the reason for higher fecal sulfur excretion for pigs fed the low sulfur diet is unclear, the net effect of feeding the low sulfur diet was a 30% reduction in total fecal excretion. Another interesting, but difficult to explain, finding was reduction in total sulfur retention for pigs fed the low sulfur Phase II and Phase III.

Since sulfate appears to be the predominant chemical form of sulfur excreted in swine manure, we wanted to measure effect of dietary sulfur level on sulfate excretion. As expected, Table 7 shows that sulfate intake was less for pigs fed the low sulfur diet compared to pigs fed the high sulfur diet. However, amount of fecal and urine sulfate excretion was greater than total amount of fecal and urine sulfate excretion shown in Table 6, which is theoretically impossible. This result confirms the difficulty and challenge of accurately quantifying various sulfur compounds in feces and urine. Although fecal sulfate excretion tended to be higher for pigs fed the low sulfur diet, increased variability in sulfate values due to the analytical procedure used prevented these apparent differences from being significantly different. Amount of sulfate excreted in urine was dramatically higher than in feces, and was higher for pigs fed the high sulfur diet compared to pigs receiveing low sulfur intake in Phase II, Phase III, and for the average of the 5-week experiement (Table 7). Net sulfate retention was negative for pigs fed both high and low sulfur diets in all phases of the experimental period. This means that more sulfate was excreted in feces and urine than the total sulfate consumed by the pig, suggesting significant conversion of various chemical forms of non-sulfate, sulfur forms consumed by the pig into primarily the sulfate form for excretion. Pigs fed the high sulfur diet had higher loss of sulfate (less retention) than pigs fed the low sulfur diet during Phase III. This finding suggests that excess sulfur consumed by pigs is primarily converted to sulfate for excretion. Pigs fed the low sulfur diet had less sulfate intake and excretion resulting in less sulfate loss.

 Table 6. Effects of feeding a 3-phase, high sulfur diet sequence compared to a 3-phase, low sulfur diet sequence on total sulfur intake, fecal sulfur excretion, urinary sulfur excretion, and sulfur retention

Measure	High Sulfur	Low Sulfur	Standard Error	Significance
Sulfur intake, mg/pig/day
Week 1 (Phase I)	541	469	38	NS
Weeks 2 & 3 (Phase II)	650	404	38	P<.0003
Weeks 4 & 5 (Phase III)	948	732	57	P<.015
Weeks 1-5 (Average)	747	548	45	P<.006
Fecal S excretion, mg/pig/day
Week 1 (Phase I)	77	111	9	P<.01
Weeks 2 & 3 (Phase II)	87	115	10	P<.06
Weeks 3 & 4 (Phase III)	158	193	16	P=.15
Weeks 1-5 (Average)	114	145	10	P<.05
Urine S excretion, mg/pig/day
Week 1 (Phase I)	137	42	15	P<.0005
Weeks 2 & 3 (Phase II)	110	30	7	P<.0001
Weeks 4 & 5 (Phase III)	212	53	16	P<.0001
Weeks 1-5 (Average)	156	42	10	P<.0001
Sulfur excretion (Fecal S + Urine S), mg/pig/day
Week 1 (Phase I)	214	153	17	P<.02
Weeks 2 & 3 (Phase II)	197	145	13	P<.01
Weeks 4 & 5 (Phase III)	371	245	27	P<.005
Weeks 1-5 (Average)	270	187	17	P<.003
Sulfur retention (S intake - Fecal S - Urine S), mg/pig/day
Week 1 (Phase I)	327	315	33	NS
Weeks 2 & 3 (Phase II)	453	258	30	P<.0002
Weeks 4 & 5 (Phase III)	577	486	33	P<.065
Weeks 1-5 (Average)	477	361	30	P<.01

Table 7. Effects of feeding a 3-phase, high sulfur diet sequence compared to a 3-phase, low sulfur diet sequence on total sulfate intake, fecal sulfate excretion, urinary sulfate excretion, and sulfate retention

Measure	High Sulfur	Low Sulfur	Standard Error	Significance
Sulfate intake, mg/pig/day
Week 1 (Phase I)	384	275	25	P<.007
Weeks 2 & 3 (Phase II)	469	315	28	P<.001
Weeks 4 & 5 (Phase III)	889	576	50	P<.0003
Weeks 1-5 (Average)	620	411	36	P<.0007
Fecal sulfate excretion, mg/pig/day
Week 1 (Phase I)	99	124	13	NS
Weeks 2 & 3 (Phase II)	119	253	84	NS
Weeks 3 & 4 (Phase III)	211	255	34	NS
Weeks 1-5 (Average)	152	216	41	NS
Urine sulfate excretion, mg/pig/day
Week 1 (Phase I)	374	300	60	NS
Weeks 2 & 3 (Phase II)	686	419	112	P=.11
Weeks 4 & 5 (Phase III)	1537	620	123	P<.0001
Weeks 1-5 (Average)	964	476	81	P<.0005
Sulfate excretion (Fecal + Urine), mg/pig/day
Week 1 (Phase I)	473	423	62	NS
Weeks 2 & 3 (Phase II)	804	673	133	NS
Weeks 4 & 5 (Phase III)	1748	845	142	P<.0003
Weeks 1-5 (Average)	1116	692	93	P<.005
Sulfate retention (Intake - Fecal - Urine), mg/pig/day
Week 1 (Phase I)	-89	-149	57	NS
Weeks 2 & 3 (Phase II)	-335	-357	129	NS
Weeks 4 & 5 (Phase III)	-859	-269	104	P<.0008
Weeks 1-5 (Average)	-496	-280	73	P<.05

 

Our goal in formulating low sulfur diets was to reduce sulfur excretion without reducing energy and nitrogen digestibility, retention and pig performance. Tables 8, 9, and 10 show results of feeding the low sulfur diet on energy digestibility, nitrogen retention and excretion, and growth performance, respectively. There were no differences in digestible energy and nitrogen retention between pigs fed either high or low sulfur diets. Therefore, a reduction in sulfur and sulfate excretion can be achieved without negatively affecting energy or nitrogen digestibility and retention (Tables 8 and 9, respectively). These results are consistent with the growth performance comparison shown in Table 10, showing similar gain, feed intake and feed conversion by feeding either the high or low sulfur diet sequence.

Table 8. Effects of feeding a 3-phase, high sulfur diet sequence compared to a 3-phase, low sulfur diet sequence on gross energy intake, fecal gross energy excretion, and digestible energy

Measure	High Sulfur	Low Sulfur	Standard Error	Significance
Gross energy intake, kcal/pig/day
Week 1 (Phase I)	691	664	52	NS
Weeks 2 & 3 (Phase II)	851	869	60	NS
Weeks 4 & 5 (Phase III)	1456	1422	96	NS
Weeks 1-5 (Average)	1061	1049	71	NS
Gross fecal energy, kcal/pig/day
Week 1 (Phase I)	49	52	4.2	NS
Weeks 2 & 3 (Phase II)	58	77	4.3	P<.01
Weeks 3 & 4 (Phase III)	115	139	12.2	NS
Weeks 1-5 (Average)	79	99	6.4	P<.08
Digestible energy (GE intake - Fecal GE), kcal/pig/day
Week 1 (Phae I)	642	612	48	NS
Weeks 2 & 3 (Phase II)	793	792	57	NS
Weeks 4 & 5 (Phase III)	1340	1284	87	NS
Weeks 1-5 (Average)	982	953	66	NS

 Table 9. Effects of Feeding a 3-Phase, High Sulfur Diet Sequence Compared to a 3-Phase, Low Sulfur Diet Sequence on Total Nitrogen Intake, Fecal Nitrogen Excretion, Urinary Nitrogen Excretion, and Nitrogen Retention

Measure	High Sulfur	Low Sulfur	Standard Error	Significance
Nitrogen intake, g/pig/day
Week 1 (Phase I)	5.54	5.44	0.42	NS
Weeks 2 & 3 (Phase II)	6.93	6.84	0.48	NS
Weeks 4 & 5 (Phase III)	11.18	11.26	0.75	NS
Weeks 1-5 (Average)	8.35	8.33	0.56	NS
Fecal N excreted, g/pig/day
Week 1 (Phase I)	0.53	0.63	0.06	NS
Weeks 2 & 3 (Phase II)	0.62	0.85	0.06	P<.03
Weeks 3 & 4 (Phase III)	1.16	1.36	0.12	NS
Weeks 1-5 (Average)	0.82	1.01	0.08	P<.10
Urine N excreted, g/pig/day
Week 1 (Phae I)	0.68	0.58	0.09	NS
Weeks 2 & 3 (Phase II)	0.82	0.93	0.08	NS
Weeks 4 & 5 (Phase III)	2.50	2.35	0.26	NS
Weeks 1-5 (Average)	1.46	1.43	0.13	NS
Nitrogen retained (N intake - Fecal N - Urine N), g/pig/day
Week 1 (Phase I)	4.33	4.22	0.37	NS
Weeks 2 & 3 (Phase II)	5.49	5.07	0.40	NS
Weeks 4 & 5 (Phase III)	7.53	7.55	0.47	NS
Weeks 1-5 (Average)	6.07	5.89	0.40	NS

Table 10. Effects of feeding a 3-phase, high sulfur diet sequence compared to a 3-phase, low sulfur diet sequence on average daily gain, average daily feed intake, and gain/feed during a 5-week feeding period

Measure	High Sulfur	Low Sulfur	Standard Error	Significance
ADG, g/d
Week 1 (Phase I)	273	229	20	NS
Weeks 2 & 3 (Phase II)	391	332	23	NS
Weeks 4 & 5 (Phase III)	536	568	17	NS
Weeks 1-5 (Average)	422	406	20	NS
ADFI, g/d
Week 1 (Phase I)	274	246	9	NS
Weeks 2 & 3 (Phase II)	505	530	30	NS
Weeks 3 & 4 (Phase III)	911	955	48	NS
Weeks 1-5 (Average)	621	644	28	NS
G/F
Week 1 (Phase I)	0.995	0.932	0.049	NS
Weeks 2 & 3 (Phase II)	0.758	0.622	0.035	P<.07
Weeks 4 & 5 (Phase III)	0.592	0.598	0.023	NS
Weeks 1-5 (Average)	0.683	0.636	0.021	P=.12

Procedures for Experiment 2

A second experiment was conducted to determine if reduction in sulfur excretion from feeding low sulfur diet in the first experiment results in reduced hydrogen sulfide gas and odor as well as supporting growth performance of pigs equivalent to pigs fed the typical high sulfur diet. The same experimental diet sequence and formulations used in experiment 1 were used in this experiment. A total of 128 pigs with average initial body weight of 6.97 kg, and average of 20 days of age were weaned in two groups (64 pigs/group) and randomly allocated to one of four environmental rooms. A total of four replications per dietary treatment sequence was used. A total of 16 pigs were housed in raised deck nursery pens in each of four environmental rooms. Each room was randomly assigned to one of the two experimental diets. Pigs were weighed, and feed consumption was determined weekly to calculate average daily gain, average daily feed intake and gain/feed for a 5-week feeding period. Odor, hydrogen sulfide, and ammonia measurements were recorded weekly to determine the effect of diet on odors and gases.

Results from Experiment 2

Growth performance was not different between pigs fed either the high or low sulfur diet sequences (Table 10). Thus, reducing total sulfur content of the diet has no detrimental effects on pig performance as long as requirements for all other nutrients are met.

Figures 6-9 show the effect of feeding high and low sulfur diet sequences on odor detection, hydrogen sulfide level, and ammonia level during the 5-week feeding period. Odor levels were low but tended to increase linearly during the 5-week feeding period (Figure 6). Odor levels tended to be higher in rooms where pigs were fed high sulfur diets during week 3-5, which is likely due to increased microbial fermentation of manure due to increased manure volume. Hydrogen sulfide levels tended to be similar between dietary treatments during weeks 1 and 2, but tended to be higher for rooms where pigs were fed the high sulfur diet sequence during weeks 3-5 (Figure 7). Ammonia level tended to increase linearly each week (Figure 8), but was not affected by feeding a low sulfur diet (Figure 9). These results show that dietary sulfur level is a significant contributor to odor and hydrogen sulfide levels in confinement nursery facilities.

SUMMARY

 There are numerous odor compounds produced during manure decompostion in various manure storage systems. Several of these compounds contain sulfur. Hydrogen sulfide is a specific sulfur containing gas that contributes to total odor emitted from swine confinement facilities, and is currently being monitored by regulatory agencies on many commercial swine farms in Minnesota and 23 other states. Our studies show that level of hydrogen sulfide gas emitted from manure storage structures is not well correlated with odor levels from these same sources. Sulfate levels in drinking water and feed appear to be significant contributors to hydrogen sulfide levels on commercial swine farms, but other poorly identified factors are also involved in the level of hydrogen sulfide produced.

Sulfur is a required nutrient by the pig, and this requirement is met by providing adequate levels of organic sulfur in the form of the sulfur containing amino acids (methionine, cystine, cysteine). Little is known about interconversion of various sulfur compounds during digestion, absorption, and excretion other than the predominant form of sulfur excreted in feces and urine is sulfate. Depending on pH of the slurry in manure storage structures, level of hydrogen sulfide gas may be low (pH greater than 8) or high (pH less than 8). Sulfates excreted in urine and feces are easiliy converted to hydrogen sulfide under anaerobic microbial fermentation processes in the manure storage structure.

 Our studies have shown that by carefully selecting low sulfur feed ingredients and using them to formulate nutritionally adequate, low sulfur starter diets, total sulfur and sulfate excretion can be reduced by approximately 30%, without compromising energy and nitrogen digestibility or pig performance. Furthermore, our studies show that reduction in total sulfur consumption and excretion will lead to a reduction in hydrogen sulfide gas and odor, but not affect ammonia levels in nursery facilities.

REFERENCES

Avery, G.L., G.E. Merva and J.B. Garrish. 1975. Hydrogen sulfide production in swine confinement units. Trans. Am. Soc. Agric. Engrs. 18:149-151.

Baker, D.H. 1977. Sulfur in Nonruminant Nutrition. West Des Moines, Iowa: National Feed Ingredient Association. 123 pp.

Banwart, W.L., and J.M. Brenmer. 1975a. Formation of volatile sulfur compounds by microbial decomposition of sulfur-containing amino acids in soils. Soil Biol. Biochem. 7:359-364.

Banwart, W.L., and J.M. Brenmer. 1975b. Identification of sulfur gases evolved from animal manures. J. Environ. Qual. 4:363-366.

Banwart, W.L., and J.M. Brenmer. 1976. Evolution of volatile sulfur containing compounds from soils treated with sulfur containing organic materials. Soil Biol. Biochem. 8:439-443.

Day, D.L., E.L. Hansen and S. Anderson. 1965. Gases and odors in confinement swine buildings. Trans. Am. Soc. Agric. Engrs. 8:118-121.

Freney, J.R. 1967. Sulfur-containing organics. In: A.D. McLaren and G.H. Petersen (Editors), Soil Biochemistry. Marcel Dekker, New York, NY, pp. 229-259.

Hatchikian, E.C., M. Chaigneau, and J. Le Gall. 1976. Analysis of gas production by growing cultures of three species of sulfate reducing bacteria. In: H.G. Schlegel, G. Gottschalk and N. Pfennig (Editors), Microbial Production and Utilization of Gases. E. Goltz, K. G. Gottingen, pp.109-118.

Kandota, H., and Y. Ishida. 1972. Production of volatile sulfur compounds by microorganisms. Ann Rev. Microbiol. 26:127-138.

Kline, R.D., V.W. Hays, and G.L. Cromwell. 1971. Effects of copper, molybdenum and sulfate on performance, hematology and copper stores of pigs and lambs. J. Anim. Sci. 33:771.

Loehr, R.C. 1974. Agricultural Waste Management. Academic Press, New York, NY, p. 519.

Meister, A. 1965. Biochemistry of the Amino Acids. Academic Press, New York, NY, London., 1084 pp.

Miller, E.R. 1975. Utilization of inorganic sulfate by growing-finishing swine. Mich. Agric. Exp. Stn. Res. Rep. 289:100-104.

NCR. 1988. Nutrient Requirements of Swine. 9th Revised Edition. National Academy Press, Washington D.C.

NCR. 1998. Nutrient Requirements of Swine. 10th Revised Edition. National Academy Press, Washington D.C.

Riviere, J., J.C. Subtil, and G. Catroux. 1974. Etude de l'evolution physico-chiminique et microbiologique du lisier de porcs pendant le stockage anaerobie. Ann. Agron. 25:383-401.

Sawyer and McCarty. 1978. Sulfates. In: Chemistry for Environmental Engineering, Third Edition. McGraw-Hill Book Company, New York, NY. pp. 476-481.

Spoelstra, S.F. 1980. Origin of objectionable odorous components in piggery wastes and the possibility of applying indicator components for studying odor development. Ag. and Environment 5:241-260.

Last updated July 1, 1998 by David Schmidt
For questions and further information, send email to David Schmidt at: schmi071@maroon.tc.umn.edu
© 1998 Regents of the University of Minnesota. All Rights Reserved.

The University of Minnesota is an equal opportunity educator and employer.

This page is part of the Manure Education and Research Home Page at http://www.bbe.umn.edu/extens/manure/manure.html
Which is part of the Biosystems and Agricultural Engineering Depart ment web at http://www.bbe.umn.edu/