Restoration of Soil Incineration Ash at the Drake Chemical Superfund Site

This applied research project successfully demonstrated in-situ topsoil manufacture by incorporating large amounts of yard trimmings compost into the soil incineration ash.

Incineration of contaminated soil at the Drake Chemical Superfund site located in Lock Haven, PA generated approximately 300,000 tons of alkaline, saline soil ash. The soil ash was placed at maximum density in a pile covering approximately 3.5 acres. Final closure of the site called for establishment of a permanent vegetative cover on the soil ash by first covering the pile with 18 inches of clean soil fill and then 6 inches of topsoil. This would have required excavation, transport, and placement of approximately 14,000 tons of soil material. Members of the site cleanup team (US EPA, PA Department of Environmental Protection, US Army Corp of Engineers, OHM Remediation Services) were interested in developing an in-situ manufactured soil alternative that would allow direct vegetation of the soil ash and eliminate the need for natural soil cover. A coordinated effort was initiated between Dr. Richard Stehouwer of Penn State University and the Drake Team to develop an alternative strategy that would allow establishment of a permanent vegetative cover directly on the soil ash. Given the chemical and physical constraints of the soil ash, the decision was made to focus our efforts on soil ash amendment with yard trimmings compost.

Numerous organic chemicals were present in soil at the Drake site, with beta-naphthylamine (a bladder carcinogen) the primary contaminant of concern. These contaminants were removed by thermally treating the soil in a rotary kiln operated within a narrow temperature range around 1,700 degrees F, which also combusted all soil organic matter and sterilized the soil. Prior to incineration, soil was screened to remove large debris, and quick lime (CaO) was added to drive off excessive moisture. Chlorides in combustion gasses from the incineration process were scrubbed by reacting with NaOH forming a spent scrubber liquor solution of NaCl (common table salt). This spent scrubber liquor was sprayed on hot soil ash to cool it as it came out of the incinerator. These aspects of the incineration process produced soil ash that was alkaline (pH 8.4 to 8.8), very strongly saline and contained no organic matter. The bulk of the soil ash collected was as bottom ash, although much of silt and clay sized particles in the original soil were swept from the incinerator in the combustion gasses and collected in a bag house. This depletion of fine particulates made soil ash a relatively coarse-textured (sandy-loam) material. Following testing to insure complete removal of contaminants, the soil ash was placed in a pile at maximum density.

The ash was very infertile with essentially no nitrogen and very little phosphorus (potash levels were moderate). All of these factors combined to make the soil ash very inhospitable to plant growth. The relatively steep slopes on the soil ash pile (1:4) combined with loss of structure and a high degree of compaction promoted surface runoff of rainwater rather than infiltration, percolation and leaching of salts from the surface. Salinity levels were high enough that only the most salt tolerant plant species could survive. The coarse texture of soil ash further exacerbated the soluble salt problems due to its limited capacity to store plant available water. The soil ash also had a low cation exchange capacity due to its coarse texture and loss of organic matter. Thus, plant growth was further restricted by the limited capacity of the soil ash to retain and supply essential plant nutrients. Nutrient cycling was also very limited in the soil ash because all soil biotic activity was eliminated during incineration, and recolonization would be very slow due to the lack of organic carbon energy sources for soil microbes.

We thought that large amounts of yard trimmings compost, supplemented with inorganic fertilizer, could ameliorate many of these constraints to plant growth. Addition of the compost was expected to:
- Dilute the soluble salts in the ash and might decrease the availability of sodium (thereby decreasing its toxicity),
- Help to reestablish a soil microbial community by inoculating the soil ash with microorganisms and providing organic carbon to stimulate soil microbial activity,
- Create a friable rooting medium,
- Increase water infiltration, soil permeability, water holding capacity, and leaching of salts, and
- Increase cation exchange capacity, nutrient retention and supply, and nutrient cycling.

Greenhouse experiments

These concepts were tested in a greenhouse pot experiment using three compost and fertilizer addition rates as well as various depths of mixing. Rates of compost addition were based on the objective of attaining a soil material with 2 - 3% stable organic matter. Assuming a compost with 50% organic material was used, and that approximately half of the organic material in the compost would decompose, an addition of 100 dry tons of compost per 1000 tons of soil ash (acre plow depth) would be needed to achieve a 2.5% stable organic matter content. We used this as our base rate and also tested levels of 0, 50, and 150 tons of dry compost per acre. Using a typical mature yard trimmings compost, 150 dry tons per acre would equate to 400 - 425 cubic yards per acre, or a little over 3 inch depth per acre. This amount of compost would be difficult to mix in one application, so the two higher rates (100 and 150 tons/acre) were mixed into the pots to simulate compost spreading and incorporation in two equal applications. The first application was mixed to a depth of 12 inches to simulate deeper incorporation and to provide increased rooting depth. The second application was mixed to a depth of six inches. For the 50 ton/acre rate only a single application and incorporation to six inches was made. Each level of compost addition was tested with four rates of fertilizer addition: a base rate of 100 lb N, 200 lb P2O5, and 100 lb K2O per acre, and 0, 0.5, and 1.5 times this amount. Fertilizers were mixed in the upper 6 inches of soil ash.

Pots were planted with a mix of tall fescue and birdsfoot trefoil. After planting the pots were watered with only enough water to keep the surface moist and, following seedling emergence, to meet plant needs for water. This was done to maintain maximum salt concentrations in the soil surface and thus to simulate a worst-case situation. One month after planting water equivalent to a 1.3 inch rainfall was added over a 12 hour period. This was sufficient to produce leachates from all the pots and presumably moved some salts away from the soil surface. A second fertilizer application, equivalent to the first application was made 19 weeks after planting by spreading fertilizer on the pot surfaces. Plants were harvested 13, 19, and 24 weeks after planting by cutting at 1 inch height, drying and weighing. Following each harvest, another 1.3 inch simulated rainfall was applied.

During the first six weeks we monitored germination and establishment of tall fescue and birdsfoot trefoil under conditions of minimum moisture and no leaching of salts. Because of the limited watering, soluble salts were wicked to the soil surface and deposited there by evaporation. With increased compost addition the amount of evaporite salts on the surface was reduced. Four weeks after planting only 0.5% of birdsfoot trefoil seeds and fewer than 30% of tall fescue seeds in pots with no compost had produced viable seedlings. However, all levels of added compost improved establishment of birdsfoot trefoil to 15 to 20% and tall fescue to 50 to 60%. The first simulated rainfall event did not further increase birdsfoot trefoil germination and establishment, but did cause additional fescue seeds to germinate and increased establishment to the 80 to 90% range. Following the simulated rainfall event there was a marked increase in plant growth rates in all of the compost amended pots. Seedlings that survived in the unamended soil ash showed clear signs of salt injury, and growth was greatly reduced compared to seedlings growing with compost amendment. These results clearly showed that compost had a beneficial effect on plant establishment and early growth even under conditions with no salt leaching. It appeared that compost had diluted soluble salt concentrations and reduced salt toxicity. The germination and plant growth response to the first leaching also indicated that any steps taken to increase water infiltration and percolation would greatly increase the chances for successful establishment of vegetation on the soil ash.

Plant growth at each harvest, and the total yield of all harvests, was increased by each level of compost addition. Within each compost level, plant growth was further increased by each level of fertilizer addition. No yield (0 g/pot) was obtained with no compost or fertilizer and the largest total yield (27 g/pot) was obtained with the highest levels of compost and fertilizer. Growth response to added fertilizer was also improved by compost addition. With no compost, fertilizer addition increased yield by only 0.5 g/pot, but with 150 tons/acre of compost addition of fertilizer increased total yield by 16 g/pot (from 11 to 27 g/pot). Further evidence that compost improved retention and utilization of added plant nutrients was seen in the nitrate content of pot leachates. In pots with the highest fertilizer addition, final leachate nitrate was 166 mg/L with no compost and only 8 mg/L with 150 tons/acre of compost. Compost had decreased soil ash pH. At the end of the experiment pH was 8.6 with no compost and 7.6 with 150 tons/acre of compost.

Given the positive results obtained in the greenhouse study, the Drake team decided not to use imported fill and topsoil cover opting instead to amend soil ash with compost and establish vegetation directly on the soil ash. Penn State University Scientists, Rick Stehouwer and Pete Landscoot, in conjunction with the Drake team, used the results of the greenhouse study to develop an alternative plan which was implemented at the Drake site in July and August, 1999.

Field implementation

We opted to use a compost application rate of 150 tons/acre, since this had given best results in the greenhouse study. Yard trimmings compost was added to the pile in two equal lifts of 75 tons/acre (dry weight basis). Each lift was spread on the surface with a bulldozer and then incorporated into the soil ash. The first lift was to be incorporated to a depth of at 12 inches and deeper if possible. The second compost lift also included inorganic fertilizer (150 lb N, 300 lb P2O5, and 150 lb K2O per acre) and 10 tons/acre (wet weight) of Bionsoil®, a processed manure product. The second application was to be incorporated to a depth of 12 inches.

Compost application and mixing began on July 29 and was completed on Aug 11, 1999. We attempted to incorporate the first compost lift with a chisel plow using several passes in perpendicular directions. However, we were not able to achieve the desired depth of mixing with the chisel plow due to slippage on the relatively steep slope, loss of traction from the compost, and the highly compacted soil ash,. We switched to a subsoiler plow with fewer shanks and were able to mix compost to 12 inches. Mixing was not uniform with depth but was enriched near the soil surface and decreased with depth. The second application was mixed using the chisel plow and accomplished more uniform mixing to a depth of at least 6 inches. The supplied compost was not fully mature and had a lower than expected bulk density (0.17 dry tons/yd3). This increased the volume of material to be incorporated which also contributed to the mixing difficulties. Consequently, compost concentrations near the pile surface were significantly higher than those used in the greenhouse study. Following mixing the pile surface was very loose and fluffy. To prepare a more suitable seedbed the surface was firmed by driving up and down the pile sides with a bulldozer. This left a firm surface with small ridges on the contour formed by the bulldozer tracks.

Two days after compost was mixed and the surface firmed (but before seeding and mulching) an intense storm dropped 1.75 inches of rain on the pile. There was no erosion from the pile indicating the compost had achieved the objective of increased water infiltration and percolation. This rainfall was also fortuitous in that it moistened the rooting zone and undoubtedly leached some salts from the surface.

The pile was hydroseeded on August 16, 1999 with a mixture of tall fescue, red fescue, perennial ryegrass, birdsfoot trefoil, and red clover. The hydroseed tank mix included additional fertilizer (40 lb N, 80 lb P2O5, 40 lb K2O/acre), legume inoculant and hydromulch. Following hydroseeding straw mulch was spread over the surface of the pile. Finally, a spray irrigation system was installed that enabled the entire soil ash pile to be watered. One inch of water was applied during the next 24 hours to leach additional salts from the surface, and for one month thereafter the pile was irrigated as needed to maintain adequate surface moisture for plant growth.

Initial germination and plant establishment was rapid and uniform over the entire pile. One month after planting, however, seedlings were showing clear signs of nitrogen deficiency and growth rates were lagging behind what would normally be expected of new seedings. Analysis of the compost used on the site showed a C:N ratio of 29:1, a further indication that the compost was not fully mature. The high C:N ratio, combined with higher than intended concentrations of compost in the soil surface (due to the problems in mixing), indicated that much of the added N fertilizer had been immobilized. To overcome this problem 40 lb N/acre was applied to the pile on September 20, 1999. The added N produced a rapid improvement in plant growth, but did not fully overcome evidence of N deficiency. To insure N was not limiting, two additional 40 lb N/acre applications were made during the next three months. The pile will be overseeded with legumes early in the spring of 2000, at which time a final fertilizer application will be made.

Amendment of soil incineration ash with yard trimmings compost proved to be highly successful and allowed the establishment of vigorous vegetative cover on previously phytotoxic and infertile material. Use of compost was less costly than importing soil cover - both economically and environmentally. Although the compost was more expensive per ton than soil, very large savings were garnered by purchasing, transporting, and handling approximately 25 times less material. Composted waste materials were recycled to a beneficial use, and prime farmland was not stripped of its topsoil.