Steel Industry Slags Compared with Calcium Carbonate in Neutralizing Acid Mine Soil

Ohio has substantial lands impacted by surface mining for coal and an active steel industry. Steel industry slags have been used as liming compounds for agriculture and acid mine soil reclamation. This 3-year study evaluates slags from Ohio steel mills in greenhouse trials where these materials are compared to reagent grade CaCO 3 in their ability to improve plant growth on acid mine soil. The objectives of this study were to evaluate the effectiveness of these materials at two rates of application in raising acid mine soil pH and to address concerns about metals in such slags. Three slags and reagent grade CaCO 3 were applied at rates equivalent to 12.5 and 25 g CaCO 3 kg soil on acid mine soil (pH = 3.5). Five consecutive crops of oats (Avena sativa L.), wheat (Triticum aestivum L.), corn (Zea mays L.), wheat and soybean (Glycine max (L.) Merr.) were grown and harvested at the seedling stage. The slags and CaCO 3 increased yields (P <0.01 level) compared to unlimed control pots. Soil and plant Ca were increased and plant Al and Mn decreased by application of all four materials. The slags increased soil and plant Mg. Particle size of the slags was somewhat coarse which decreased their effectiveness, but overall these slags proved to be satisfactory liming materials. The fineness efficiency developed for carbonate forms of lime may not adequately characterize slag effectiveness. Micronutrient metals including iron were not found to be in excess in plant tissue treated with slags despite the steel slags’ high Fe content. OHIO J SCI 105 (4):79–87, 2005 Manuscript received 7 February 2004 and in revised form 22 June 2005 (#04-05). INTRODUCTION Ohio has thousands of hectares of lands disturbed by resource extraction, principally the surface mining of coal, sand and gravel, limestone, and other building stone. Sutton and Dick (1987) reported in a review of reclamation of lands disturbed by mining that Ohio had 138,000 ha of surface mined lands of which 111,000 ha (80%) were coal mine sites located in the southeast region of the state’s Appalachian Plateau. Many of these coal mine sites contain FeS 2 (mascarite and pryrite) that can become sources of extreme soil acidity and acid mine drainage. Acid conditions solubilize metals and increase concentration of Al, Mn, and Fe to toxic levels, severely restricting plant growth and leading to increased soil erosion sources of sediments and acid mine drainage. In their review paper, Pichtel and others (1994) discussed reclamation strategies that made use of some of society’s by-products, such as sewage sludge (biosolids), papermill sludge, and power plant fly ash, in attempting to create a soil environment that would support vegetative cover, such as trees and forage plants, on acid mine soil. They reported more success with biosolids and papermill sludge than with power plant fly ash on extremely acid mine soil sites. Another such industrial by-product is steel industry slag. Ohio, like other Midwest and Northeast states, has significant steel making capacity. Iron and steel slag sales in the United States during 1996, 1997, and 1998 indicate that over half of all sales are generated in the North Central States of Illinois, Indiana, Michigan, and Ohio (Kalyoncu and Kaiser 1998; Kalyoncu 1999). Slags are nonmetallic byproducts of iron and steel making, and they are unique to the limestone sources used and the specialty steel being manufactured. They are comprised of Ca and Mg silicates and are able to supply Ca and/or Mg and also neutralize acidity. Typically a blast furnace with an ore feed with 60 to 66% iron would generate 220 to 370 kg of slag for each metric ton of pig iron produced, and steel slag outputs are typically 20% weight/weight (w/w) of steel output (Kalyoncu and Kaiser 1998). While blast furnace and steel furnace slags have many uses, one historic and widespread use has been as a liming material. The use of steel industry slags as a substitute for traditional limestone has a substantial history in Ohio (Williams 1946; Volk and others 1952; Jones 1968). Volk and others (1952) evaluated granulated slag (water quenched), air-cooled slag, and dolomitic limestone reconstituted by combinations of various sieve sizes into “ag screenings,” “ag meal,” and “ag ground” size grades as defined by Ohio’s limestone law. They compared the three materials in both greenhouse and field studies at Wooster, Canfield, and St. Clairsville, OH. The granulated and air-cooled slags were much higher in SiO 2 and Al 2 O 3 , somewhat higher in Ca, and much lower in Mg than the dolomitic limestone at all 3 field study locations. Greenhouse pot tests for the correction of soil acidity were conducted with Wooster silt loam (Fine-loamy, mixed, mesic Oxyaqic Fragiudalfs) and plant growth performance in the greenhouse at several levels of each liming material were performed with Trumbull silty clay loam (Fine, illitic, mesic Typic Epiaqualfs). In field studies the average hay yields were higher for granulated slag than for air-cooled slag or limestone at each size gradation (screenings, meal, and ag ground). Air-cooled slag and dolomitic limestone were equally effective when they had the same particle size distribution. In greenhouse studies granulated slag 80 VOL. 105 SLAGS AND CACO 3 TREAT ACID MINE SOIL and dolomitic limestone were more effective at correcting soil acidity than air-cooled slag. Since air-cooled slags produced crop yields quite comparable to limestone and yet did not raise pH as effectively, the studies indicated that crop yield response to the slags involved more than simply neutralizing soil acidity. Jones (1968) reevaluated the effectiveness of granulated steel industry slag in field studies on acid Canfield silt loam (Fine-loamy, mixed, mesic active Aquic Fragiudalfs) of pH 5.5. He compared the performance of slag to comparable rates of calcitic and dolomitic limestone and concluded that granulated slag was as effective at the two rates used (3.36 and 6.72 Mg ha) as either type of limestone. Crop yields were equal or slightly higher for the slag treatment compared to the calcitic and dolomitic lime treatments. When soil and plant Ca and Mg were measured, the granulated slag and dolomitic limestone both boosted the soil and plant Mg to a significantly greater degree than did the calcitic limestone. This was because of the high Mg content of the granulated slag (12% MgO equivalent) and the dolomitic lime. Rodriguez and others (1994) evaluated the performance of Basic Linz-Donowitz slag as a liming material for pastures in Northern Spain. In European steel manufacturing, approximately 150 kg of such slag are generated for every 1000 kg of steel produced. The slag used was 290 g kg Ca and 50 g kg Mg (w/w). In addition, it contained substantial quantities of Fe, Se, and Mn, and traces of Pb, Zn, Cu, K, Na, and Cd. The slag application resulted in a decrease in Al saturation and increases in Ca and Mg saturation of the soil cation exchange capacity. Slag applications resulted in higher Ca, Mg, and P concentrations and higher yields than the untreated plots. During the 3-year study, slag application reduced plant K and Mn concentrations. Oguntoyinbo and others (1996) compared four liming materials: basic slag, cement flue dust, indigenous limestone, and imported slaked lime Ca(OH) 2 on two acid soils in greenhouse trials in Nigeria. The relative effectiveness at neutralizing soil acidity was Ca (OH) 2 > basic slag > cement flue dust > limestone. Lime requirements were related to soil exchangeable Al on both soils. There were no significant differences in plant yield among lime sources. The soils of the North Central/Northeast US need periodic treatment (every 3-6 yrs) with liming materials to correct soil acidity, enhance nutrient availability and microbial activity, and to restore the supplies of exchangeable Ca and Mg (McLean and Brown 1984). Ohio monitoring networks for acid precipitation show rain and snow to have a pH of 4.0 to 4.4, lower than the normal pH 5.6 of water in equilibrium with ambient atmosphere CO 2 (National Atmospheric Deposition Program, National Trends Network, 1988, 1989; data for Wooster, OH). The objectives of this study were to evaluate the effectiveness of these slag materials with the size characteristics provided at raising soil pH for acid mine soil remediation, and to note the uptake of metals when slags were used compared to CaCO 3 . METHODS AND MATERIALS Three steel industry slags were supplied by Stein Inc., Broadview Heights, OH, from LTV steel in Cleveland, OH, or USS Kobe in Lorain, OH. These slags were identified as blast furnace slag (BFS), non-metallic steel slag (NMSS) poured off from slag far from the molten steel, and metallic steel slag (MSS) poured from the slag very close to the molten steel. Table 1 gives particle size and chemical properties of the slags and the reagent grade CaCO 3 . The particle size data was obtained by sieving the slags. The chemical analyses were performed by The Ohio State University Research and Extension Analytical Laboratory (REAL), Wooster, OH, and the Penn State Ag Analytical Laboratory, University Park, PA, on samples digested in concentrated perchloric or nitric acid. Penn State’s soil test lab procedures used Mehlich III for soil P and extractable cations (Wolf and Beegle 1995) and pH in water (1:1) and SMP buffer to assess the soil active and exchangeable acidity respectively (Eckert and Sims 1995). The change in supporting laboratories was made necessary by Ohio State’s closure of the REAL Lab in December 1998. The greenhouse studies used an acid mine soil from Ohio Agricultural Research and Development Center (OARDC) Unit 2 in Caldwell, OH. Hall (1977) proposed classification of the mine soils at this location as Barkcamp (Loamy-skeletal siliceous, acid, mesic Typic Udorthents) and Enoch (Loamy-skeletal, siliceous, acid, mesic Typic Udorthents) series based on pH and the potential for oxidation of FeS 2 minerals. The mine soil pH collected at Caldwell Unit 2 averaged pH 3.5. Air dried and ground mine soil at 1.5 kg per pot was amended with four materials: blast furnace slag (BFS), steel furnace slag (NMSS), metallic steel slag (MSS), and reagent grade CaCO 3 at two rates: 12.5 g kg and 25 g kg of 100% CaCO 3 equivalent material (equivalent to 16.7 and 33.3 Mg ha, respectively). The experiment was designed as a 4 × 2 factorial (4 treatments at 2 rates). Each treatment was replicated four times and placed on the available greenhouse bench with treatments completely randomized on the bench space. The statistical analyses utilized the General Linear Model of SAS for analysis of variance and construction of appropriate orthogonal comparisons. The pots were watered with deionized water throughout the course of these experiments with the opportunity for free drainage at the bottom of the pots. On 30 April 1998, 20 seeds per pot of oat cultivar (cv) ‘Armor’ were planted 1.0 cm deep in each pot. After germination and emergence, all pots were treated three times with 100 ml of 0.01 M KH 2 PO 4 and 0.01 M KNO 3 . The plants were harvested by cutting at the soil surface on 3 June 1998, at about the 4-leaf stage (~20 cm tall), oven dried, weighed, and saved for tissue analysis. After the oat crop was harvested, all soils were crushed in the top 5.0 cm of each pot, a small (15-20 g) portion was removed for soil pH measurement, and the pots were reseeded with 20 seeds of wheat (cv ‘Freedom’) on 2 October 1998. The wheat seedlings (Feekes Stage 6) were harvested by cutting at the soil surface on 9 December 1998, oven dried, weighed, and saved for subsequent analysis. After the OHIO JOURNAL OF SCIENCE 81 DAVID A. MUNN