Multielement Analysis of Animal Feed, Animal Wastes, and Sewage Sludge Stephen G. Capar, James T. Tanner, Melvin H. Friedman, and Kenneth W. Boyer” Division of Chemistry and Physics, US. Food and Drug Administration, Washington, D.C. 20204
Animal excreta and sewage sludge are currently being used as animal feed ingredients on an experimental basis. The levels of 30 elements are reported for a typical cattle feedlot diet, two dried cattle manures, a commercial cattle waste product, two dried poultry manures, and a metropolitan sewage sludge. The analyses are conducted using neutron activation analysis, induction coupled plasma spectroscopy, atomic absorption spectroscopy, and anodic stripping voltammetry. The levels of most inorganic elements are considerably higher in animal wastes and sewage sludge than in traditional animal feeds. For most elements the levels determined by several techniques are in good agreement. Problems of loss of lead with precipitate formation, accurate quantitation of elements present a t high levels, and obtaining homogeneous samples for analysis are discussed. Solid waste from farm animals has traditionally been used as cropland fertilizer. In recent years municipal sewage sludge is also being increasingly used as crop fertilizer. The safety with respect to animal and human health of cropland application of sewage sludge is currently the subject of considerable research activity (1-5). The broader aspects concerning management of sludge use on land to assure safety and wholesomeness of the food supply have also been discussed (6).
A large amount of undigested protein and high levels of several nutrient elements are found in solid wastes from animal feedlot operations and municipal waste treatment plants. These wastes are currently being considered as animal feed ingredients to be mixed with the more traditional grain and forage animal feeds. The practice of feeding “wastelage” (derived from ensiling ground grass hay and manure), or mixtures of dried animal waste and feedlot diets appears to be commonplace (7-9). Although several states have approved the use of animal waste as a feed ingredient for meat producing animals sold in intrastate commerce, the U S . Food and Drug Administration recently decided to delay issuing proposed guidelines for the safe use of animal waste in feed because legitimate questions about safety remain unanswered (10). While sewage sludge has not been formally approved for use in feed by any governmental agency, several federal and state-supported projects are studying the effects of feeding sewage sludge to animals on an experimental basis (11). Numerous papers have appeared in the literature reporting the elemental content of animal wastes and sewage sludges for various numbers of elements (1-5, 12-1 7). Most feeding studies in which animal wastes or sewage sludge are fed to animals have been conducted with the objective of showing the nutritive value of the material being fed, and not for the
purpose of determining possible effects on human or animal health (12). Bhattacharya and Taylor’s (12) review indicates that little work has been done to determine typical trace element content of animal wastes for a number of important elements. Nor has much work been done to determine whether trace elements accumulate in edible tissue, or edible products, from animals fed diets containing animal wastes. The purpose of our study was to determine the levels of as many elements as possible in typical dried animal wastes having a documented history and to compare these levels with those of the same elements in animal feeds and sewage sludges. These data should be useful in future studies or to other groups in arriving at conclusions about the desirability or relative safety of using animal wastes and sewage sludges as animal feed ingredients. Such data should also be considered in choosing among the various possible alternatives for beneficial use of animal wastes and sewage sludge, including cropland application, conversion to methane, and animal feed ingredients.
Experimental Samples Analyzed. The samples analyzed were all received from the same source (Colorado State University) and included: (I) a typical feedlot diet without animal waste products added (the dry weight composition of this ration was 70% corn, 3% hay, 5% beet pulp, 20% corn silage, and 2% mineral supplement); (11) oven-dried cattle manure from feedlot heifers fed a low fiber diet containing 59% corn, 2% alfalfa hay, 3% molasses, 33% corn silage and 3% mineral supplement; (111) oven dried cattle manure from feedlot heifers fed a high fiber diet containing 24% corn, 29% alfalfa hay, 3% molasses, 41% corn silage, and 3% mineral supplement; (IV) a commercial high-protein feedlot animal waste product similar to (111) in which a portion of the fiber and ash was removed and the product was pelletized; (V) dried poultry waste with litter consisting of wood shavings mixed with excreta from hens fed a layer ration; (VI) dried poultry waste from caged laying hens without any litter; (VII) dried sewage sludge from Metropolitan Denver Sewage Disposal District No. 1 (Metro Denver). These samples correspond to columns I-VII, respectively, in Table I. Samples (V), (VI), and (VII) are being used as feedlot diet ingredients at Colorado State University. Sample Homogenization. The feedlot diet was analyzed as received without further homogenization. Portions of the animal waste and sludge samples (100 g) were initially blended with a Waring Blendor (hereafter referred to as “blended” samples),and subsamples were analyzed by neutron activation analysis (NAA), flame atomic absorption spectrometry (AAS), and anodic stripping voltammetry (ASV) as described below. However, because most of the waste samples were still visibly inhomogeneous, additional portions of the as-received unhomogenized and unblended waste and sludge samples
This article not subject to US. Copyright. Published 1978 American Chemical Society
Volume 12, Number 7, July 1978
785
(70-105 g each) were mixed in the ratio of 4 parts distilled deionized water to 1 part dry material and homogenized at high speed with a Model PCU-2 Polytron homogenizer (hereafter referred to as “homogenized” samples) equipped with a PT-35 K sonic probe-type generator until a homogeneous slurry was obtained (usually about 5 min). The slurries were then freeze dried in a VirTis Model 10-MRTR freeze drier. The resulting homogenized dry samples were analyzed by using AAS, ASV, NAA, and induction coupled plasma optical emission spectroscopy (ICP) in conjunction with the appropriate procedures described below. Sample Mineralization. Dry samples for AAS, ASV, and ICP analyses were mineralized by acid digestion and analyzed following the procedures detailed below. Sample sizes for analysis were 1.00 g “blended” animal wastes and sewage sludge, 2.00 g for “homogenized” animal wastes, 0.20 g for “homogenized” sewage sludge, 2.00 g for unhomogenized feedlot diet, and 1.00g for NBS Standard Reference Materials. Samples for AAS, ASV, and ICP analyses were weighed into 100-mL micro-Kjeldahl flasks and digested with 25 mL H N 0 3 and 5 mL HC104. Sample digests were transferred to 50-mL volumetric flasks and diluted to volume with deionized water. As routine practice at least one 1.00-g sample of an NBS Standard Reference Material (Orchard Leaves, Bovine Liver, or Spinach) and one reagent blank were also carried through the entire analytical procedure for each set of 10 samples. “Blended” animal wastes and sewage sludge samples were digested and analyzed in duplicate, while “homogenized” samples and the unhomogenized, unblended feedlot diet were digested and analyzed in triplicate. AAS Determination of Pb, Cd, Cu, Zn, Fe, Mg, a n d Mn. A Perkin-Elmer Model 403 or Model 503 AAS instrument with an air-acetylene flame was used for determining these elements. The following analytical lines were used: P b 2833.1, Cd 2288.0, Fe 2483.3, Zn 2138.6, Cu 3247.5, Mg 2852.1, and Mn 2794.8, all in angstroms. Sample solutions were diluted when necessary to achieve elemental concentrations within the following ranges: 0.2-15 pg Cu/mL, 0.5-2.0 pg Zn/mL, 2.0-10.0 pg Fe/mL, 0.1-2.0 pg Mn/mL, and 0.1-0.5 kg Mg/mL. A 10% (v/v) HC104 concentration was maintained in diluted solutions. Sample solutions for Mg determinations contained 0.6% (v/v) La to prevent refracting oxide formation. Sewage sludge was analyzed for P b and Cd, and NBS Orchard Leaves were analyzed for P b by AAS. These were the only samples with concentrations of P b and/or Cd high enough for AAS analysis for the sample sizes analyzed. ASV Determination of Pb a n d Cd. Those sample solutions prepared for AAS analyses which could not be quantitated for P b and/or Cd by AAS were analyzed for these elements by using ASV. This included all sample solutions except those of sewage sludge and NBS orchard leaves. One-milliliter aliquots of the sample solution were buffered at pH 4.3 with a sodium acetate/acetic acid buffer and analyzed with a Princeton Applied Research (PAR) Model 174 polarographic analyzer equipped with a PAR Model 315 automatic electroanalysis controller, a hanging mercury drop electrode, SCE reference electrode, and platinum counter electrode. The technique of standard additions and the instrumental parameters previously reported (20) were used for quantitating the P b and Cd present. Hydride Generation-AAS Determination of As, Sb, and Se. All samples were analyzed for As, Se, and Sb by using the hydride generation-AAS procedure and apparatus reported by Fiorino et al. (21). ICP Determination of 17 Elements. A Jarrell-Ash Model 975 Plasma Atomcomp was used to simultaneously analyze for 25 elements (Al, As, B, Be, Ca, Cd, Co, Cr, Cu, Fe, Mg, Mn, 786
Environmental Science & Technology
Mo, Ni, P, Pb, Sb, Se, Sn, Te, Ti, T1, V, and Zn) in the same solutions prepared for AAS analysis. Operating parameters for ICP were 1100 W forward power; less than 5 W reverse power; plasma observation height 17 mm above the load coil; sample solution aspiration rate 1.2 mL/min. The high and variable salt content of the sample solutions caused slight aspiration problems. Occasionally, the solution would clog the nebulizer and stop aspiration. Because the undiluted sample solutions contained levels of Fe and Ca beyond the linear range of the instrument, they had to be diluted 1-5. NAA Determination of 22 Elements. The dry powdered samples were analyzed in three parts: a long irradiation-decay for long-lived radionuclides, a short irradiation-decay for short-lived radionuclides, and a chemical separation for Hg analysis. All irradiations were carried out a t the FDA NAA facility, which uses the National Bureau of Standards (NBS) reactor in suburban Washington, D.C. Long-Lived Radionuclides. Approximately 300 mg of the dry material was weighed into clean, high-purity quartz vials, sealed, and irradiated together with standards in the high flux position (6 X 1013 neutrons/cm2/s) of the NBS reactor for 6 h. These were allowed to undergo radioactive decay for 2-4 weeks and then counted by using the automated data acquisition system described elsewhere (22).Briefly, this consists of an Ortec Ge(Li) detector (2.2 keV resolution, 15% efficiency) in an automatic sample changer (modified Nuclear Chicago Unit) connected to a Nuclear Data 4410 analyzer (16K memory). The data are collected and stored on magnetic tape for later reduction using the program MLTELMT (for Multielement) developed at FDA. Under these conditions 65Zn,szBr, 51Cr, 59Fe, 140La,1 5 2 E ~46Sc, , 86Rb, 131Ba, 6oCo, 75Se,and 124Sbwere determined. Short-Lived Radionuclides. Approximately 400 mg of dry sample was weighed into cleaned polyvials and heat sealed. These samples and standards were irradiated in the low flux position (1X 1013neutrons/cm2/s) of the NBS Reactor for 15 s and then counted after a 5-min decay. The Ge(Li) detector was the same as described earlier, but was connected to a Nuclear Data ND-100 analyzer with 2K memory. Again, the data were collected, stored on magnetic tape, and reduced later. Under these conditions 38Cl, 28A1, 56Mn,42K,52V,49Ca, 27Mg, 66Cu, and 24Nawere determined. Hg Analyses. Mercury was determined by irradiating the samples, which were sealed in high-purity quartz, in the high flux position (6 X 1013neutrons/cm2/s) of the NBS reactor for 1h. The samples and standards were then allowed to undergo radioactive decay for about 3 days before separating the Hg by the Rook et al. (23) volatilization procedure. The samples were counted with an Ortec low-energy photon detector connected to the Nuclear Data ND-4410 analyzer. This counting procedure was previously described by Friedman et al. (24).Concentrations were calculated on all three peaks (67, 68.8, and 77.3 keV) and then averaged for the final concentration.
Results The analytical results from determination of 30 elements in the animal feedlot diet, animal wastes, sewage sludge, and the NBS Standard Reference Materials are reported in Table I. Values are reported only for those elements with concentrations within the reliable quantitation range for the method indicated. For the highly toxic elements Be and Hg, “lessthan” values are reported when quantitation was not possible. The mean levels of the same elements in sewage sludges from the 16-city study reported by Furr et al. (15)are also tabulated in Table I for comparison with the data obtained in this study. These mean values also compare well with the results of Sommers ( 17).
Table 1. Elemental Content of Animal Feed, Animal Waste, Processed Waste Pellets, Sewage Sludge, and Reference Materialsa,'
AnalyEletlcal ment method
As
AAS
I Feed lot diet
0.10
...
Concentrations, ppm In drled samples
II
111
IV
V
Cattle manure low flber diet
Cattle manure hlgh flber diet
Processed cattle waste pellets
PouItry waste with lltter
0.88
2.2
0.60
0.57
V.. I
P0ulVI1 try Metro waste Denver wlthout sewage litter sludge
16 Cities average sewage sludge
0.66
14.3
8.1
Ba
NAA
18.
105.
305.
70.
54.
57.
1066.
621.
Be
ICP
