Fuel 105 (2013) 258–265

Contents lists available at SciVerse ScienceDirect

Fuel

journal homepage: www.elsevier .com/locate / fuel

Surface coal mine land reclamation using a dry flue gas desulfurization product:Long-term biological response

Liming Chen a, Yongqiang Tian a,e, Richard Stehouwer b, Dave Kost a, Xiaolu Guo a,f, Jerry M. Bigham c,Joel Beeghly d, Warren A. Dick a,⇑a School of Environment and Natural Resources, The Ohio State University/The Ohio Agricultural Research and Development Center, Wooster, OH, USAb Department of Crop and Soil Sciences, Pennsylvania State University, University Park, PA, USAc School of Environment and Natural Resources, The Ohio State University, Columbus, OH, USAd Bessemer, PA, USAe Department of Vegetable Science, College of Agronomy and Biotechnology, China Agricultural University, Beijing, Chinaf Key Laboratory of Advanced Civil Engineering Materials, Tongji University, Shanghai, China

h i g h l i g h t s

" Flue gas desulfurization (FGD) product effectively remediated acidic coal mined-land." Plant biomass was higher by FGD than by soil treatment 16 years after application." Heavy metals measured in plant tissues were not significantly increased." Bacterial populations and microbial biomass C were greatly increased by treatments.

a r t i c l e i n f o

Article history:Received 7 February 2012Received in revised form 18 June 2012Accepted 19 June 2012Available online 3 July 2012

Keywords:Coal mine land reclamationPlant biomassTrace elementMicrobial biomass carbonBacteria diversity

0016-2361/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.fuel.2012.06.081

⇑ Corresponding author. Tel.: +1 3302633877; fax:E-mail address: dick.5@osu.edu (W.A. Dick).

a b s t r a c t

Abandoned surface coal mined lands are a worldwide environmental concern due to their low productiv-ity and potential negative impact on water and soil quality. A field study was conducted to investigate theuse of a dry flue gas desulfurization (FGD) product, i.e. a fluidized bed combustion (FBC) product, for rec-lamation of an abandoned surface coal mined land in Ohio, USA. The FGD product was applied to the minesite at a rate of 280 Mg ha�1 alone or with 112 Mg ha�1 yard waste compost, and these treatments werecompared to a conventional reclamation treatment that included 20 cm of resoil material plus45 Mg ha�1 of agricultural limestone. A grass-legume sward was planted, and plant biomass yields andelements in plant tissues were determined as long as 16 years after treatments. Bacterial populationsand diversity and microbial biomass C in the reclaimed surface coal mined land were analyzed in the16th year after treatments. Compared with the conventional soil treatment, plant biomass on plots trea-ted with FGD product was lower in the first and third years, not different in the 14th year, and higher inthe 16th year after application. Magnesium, S, Mo and B concentrations in plant tissues were increased bythe treatments with FGD product in the first three years but not in the 14th year after application, and theheavy metals measured were not significantly increased. Bacterial populations and diversity and micro-bial biomass C in the reclaimed coal mine plots were significantly increased compared to adjacentuntreated area and were generally similar among reclamation treatments. These results suggest thatuse of FGD product, used alone or in combination with compost, for reclamation of acidic surface coalmined lands can provide effective, long-term remediation.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Coal mining and coal preparation practices in the years prior tothe 1977 enactment of stringent environmental protection and

ll rights reserved.

+1 330263788.

reclamation laws (Surface Mining Control and Reclamation Act)frequently resulted in dumping of coal cleaning refuse into largepiles. Many surface coal mine sites were simply abandoned with-out adequate reclamation of iron sulfide-containing materials,the source of much of the acid mine drainage. Reclamation of aban-doned coal mine lands is a worldwide environmental concern be-cause these lands surface water and groundwater quality,revegetation, and aesthetics [1]. In Ohio, USA, more than 14,000

Fig. 1. Location of the Fleming abandoned surface coal mined land site in Ohio, USA.

Table 1Selected chemical characteristics of Fleming abandoned mine land (AML) spoil andamendments.

Parameters Spoil FGD product Compost Borrow topsoil

pH (1:1 water) 3.1 12.4 7.4 4.3

Macronutrients (g kg�1)N NDa ND 8.4 NDP 0.7 0.3 1.5 0.4K 23.6 3.6 5.6 23.5Ca 0.4 261 16.9 0.7Mg 5.0 36.5 3.5 5.3S 10.2 123 ND 0.6

Micronutrients (mg kg�1)B ND 418 39.1 NDCu 26.8 49.5 69.0 62.8Fe 55700 59000 17700 39600

L. Chen et al. / Fuel 105 (2013) 258–265 259

hectares of highly degraded abandoned coal mine lands have notyet been reclaimed [2].

When compared to native soils, new mine soils often have largequantities of coarse fragments, lowered nutrient status, poor waterholding capacity [3], decreased organic matter content [4,5], lowpH, and increased Fe oxides [6], as well as low microbial activity[7,8] and microbial biomass C [9,10,5,11]. The properties of dis-turbed mined soils make them a poor medium for plant growth,and natural recolonization by plants on these soils is slow. Somerefuse or abandoned mine sites can produce copious amounts ofacid that eventually drain into many streams and rivers. Acid pro-duction in soils can make them unsuitable for production of cropsor for use as pasture lands or woodlands.

Applying topsoil and limestone to restore coal mined land is awell-established technology with many sites in the USA remainingin sound plant cover condition for 20 years or more after treatment[4,12,9]. Current reclamation laws require the spreading of stock-piled topsoil on mine spoil to facilitate revegetation. However, top-soil was generally not conserved when sites were surface minedfor coal prior to the reclamation laws. Thus, soil must be borrowedfrom adjacent land thereby creating another disturbed area. Thecost of reclamation becomes prohibitive if a sufficient amount ofborrowed soil is not available adjacent to the mine site. Moreover,in mine land soil 20 years after reclamation using topsoil, totalmicrobial biomass, microbial biomass carbon, and soil organicmatter were only 20%, 44% and 36% of values found in adjacentundisturbed soils [9].

The Clean Air Act, as amended in 1992, confirmed the need todevelop and implement processes to remove SO2 from flue gasesproduced by burning. Some flue gas desulfurization (FGD) pro-cesses generate by-product materials consisting of variousamounts of excess sorbent, reaction products containing SO2�

4 /SO2�

3 and fly ash. Because of the unspent sorbent component, theseFGD products are usually alkaline and have significant neutraliza-tion potential. Several studies have shown that this property en-ables FGD products to be used as alkaline amendments foragricultural and mine land soils [13–16].

Composts are rich in organic C and plant macronutrients andalso contain a balanced level of micronutrients [17], which oftendoes not exist in a highly degraded coal mine land situation. Incor-poration of organic matter can improve the physical and chemicalproperties of mine wastes and is known to improve fertility andbind trace elements, thereby facilitating reclamation efforts [3].Microorganisms in soils play important roles in organic matterdecomposition and nutrient cycling. Soil microbial communitiesimpact plant reestablishment and the development of ecosystemsin the mine soil. Microbial recovery is important to sustainable coalmine land reclamation. Application of organic wastes may increasemicrobial number and activity in mine soils [18,7].

The objectives of this study were to determine the effects ofFGD product, compost and topsoil on (1) plant growth, (2) concen-trations of plant essential and environmental concern elements inplant tissues, and (3) soil bacterial populations and diversity, andmicrobial biomass C to assess their potential capacities to achievelong-term reclamation success.

Mo 14.0 22.4 27.8 <0.2Ni 28.5 78.8 383 44.8Zn <0.3 112 108 138

RCRAb-regulated elements (mg kg�1)As 46.3 71.5 11.5 5.5Ba 701 204 ND 503Cd 0.8 1.5 <0.2 3.3Cr 94.4 42.2 284 95.6Pb 78.0 17.4 26.0 15.9Se 4.5 8.6 0.3 <0.7

a Not determined.b Resource Conservation and Recovery Act.

2. Materials and methods

2.1. Site description

The study site (Fig. 1), the Fleming abandoned mine land (AML)site, is located in Franklin Township, Tuscarawas County, Ohio, USA(40�3301900 north latitude and 81�3101300 west longitude). Prior toreclamation, the Fleming AML site consisted of approximately10 ha of exposed, highly erodible underclay (a stratum of clay lying

beneath a coal bed) bordered on two sides by 18 ha of spoil and2 ha of coal refuse. Acid mine drainage was a significant problemwith surface water pH ranging from 2.4 to 3.9 and electrical con-ductivity ranging from 0.7 to 3.0 dS m�1. Oxidation of pyrite(FeS2) associated with the Middle and Upper Kittanning coal bedswas a major cause of the acidity. The spoil was derived from Penn-sylvanian age rocks of the Allegheny Formation, which consist ofsandstones and shales interbedded with coal, clay, and limestone[19]. The spoil at the site were extremely acidic (Table 1), andthe ability of these materials to support plant growth was severelylimited (Fig. 2). The entire Fleming AML site was classified as

Fig. 2. Comparison of vegetation on reclaimed surface coal mined area (right) and an untreated area (left) in 2010.

260 L. Chen et al. / Fuel 105 (2013) 258–265

Bethesda soil series (loamy-skeletal, mixed, acid, mesic TypicUdorthent) prior to reclamation.

2.2. Reclamation

In autumn of 1994, six 0.4 ha plots were constructed by regrad-ing underclay and mine spoil at the Fleming AML site. First, ex-posed underclay was graded, using earthmoving equipment, to a4% slope and recompacted as an aquitard. Thickness of the aquitardranged from 3 m to greater than 10 m. A 1.5 m wide by 30 cm highberm was constructed to hydrologically separate each plot. Next,1.2 m of acid mine spoil was placed over the underclay aquitardand graded to 4% slope. Berms were formed 3 m wide by 60 cmhigh above the aquitard. Berms were constructed in a similarway at the bottom of the plots to direct flow into water samplingdevices. The three treatments applied to the six plots shown inFig. 1 and described in Table 2.

The FGD product came from an atmospheric fluidized bed com-bustion (AFBC) burner located at a General Motors plant in Pontiac,MI, which burned eastern Ohio coal and used dolomitic limestone(from Findlay, Ohio) for desulfurization. The material was trans-ported as a backhaul from the General Motors plant back to themine site and was not stored. It was applied within a few days toweeks to the spoil after being transported to the mine site. Thecompost was obtained from Earth-N-Wood Inc., Canton, OH. It con-sisted of composted yardwaste (leaves, twigs, grass clippings). Theborrowed soil was obtained from a designated area to the north ofthe study site. Predisturbance classification of the borrowed soil isunknown. Selected properties of the various amendment materialsare shown in Table 1.

The application rates for the Fleming AML site were selectedbased on the lime test index (SMP buffer pH � 10) of the spoil[20]. The lime test index, as determined by The Ohio State Univer-sity, was 44 for the Fleming AML site spoil. In order to adjust thematerial to pH 7, the Ohio Agronomy Guide [21] recommends add-ing 112 Mg ha�1 of limestone with 100% calcium carbonate equiv-alent (CCE). Since the FGD products had a CCE of approximately40%, its application rate had to be adjusted accordingly to provideneutralization potential equivalent to that of pure limestone.

Grading, construction, and application of the reclamation treat-ments were completed in October 1994. The plots were seeded inNovember of the same year using a seed mixture consisting of

Table 2Treatments applied to the plots at the Fleming abandoned surface coal mine land.

Treatmentcodes

Treatments

SOIL 112 Mg ha�1 of agricultural limestone was incorporated into gradedborrowed soil treated with additional 45 Mg ha�1 of agricultural lim

FGD 280 Mg ha�1 of FGD product was incorporated into graded spoil witFGD/C 280 Mg ha�1 of FGD product and 112 Mg ha�1 of yard-waste compo

orchard grass (Dactylis glomerata), timothy (Phleum pratense), an-nual ryegrass (Lolium multiflorum), ladino clover (Trifolium repenseLadino), birdsfoot trefoil (Lotus sp.) and winter wheat (Triticum sp.).

2.3. Plant biomass and element determinations

Plant biomass production in 1995, 1996, 1997, and 1998 wasdetermined by cutting three randomly selected 46 cm wide by30 m long strips across each plot. The biomass from each stripwas dried and weighed. Plant biomass production in 2008 and2010 was determined by cutting three randomly selected 1-m2

areas. The biomass was dried in a greenhouse for 2 weeks andweighed. Plant tissues containing a mixture of grasses and legumesfrom 1995, 1996, 1997, and 2008 were ground to pass a 1-mmsieve and stored in the lab at room temperature until analysis.They were analyzed in the years when they were harvested. Thesame laboratory and procedures were use regardless of the yearsthe samples were collected and analyzed. The concentrations ofelements P, K, Ca, Mg, S, B, Cu, Fe, Mn, Mo, Ni, Zn, Al, Na, Co, Li,Sb, Si, Sr, V, As, Ba, Cd, Cr, Pb, and Se in the plant tissues were deter-mined using inductively coupled plasma atomic emission spec-trometry (ICP-AES) after acidic digestion [22]. Concentration ofHg in plant tissues from 2008 was also determined by cold vaporatomic fluorescence spectroscopy (CVAFS). Concentrations of ele-ments in the plant tissues from 1998 and 2010 were not analyzed.

2.4. Soil bacterial populations and diversity and microbial biomasscarbon determinations

2.4.1. Soil samplingSix soil cores (0–20 cm depth) from each plot and the adjacent

untreated area were randomly collected with a soil core probe andcombined to one sample in the 16th year after reclamation in 2010.The soil samples were stored at �20 �C until bacterial populationsand diversity and microbial biomass carbon were determined.

2.4.2. Bacterial population determinationPotential viable counts were determined to enumerate the bac-

teria in different treatments. One gram fresh soil was homogenizedin 9 ml sterilized water. Then 0.3 ml of the soil solution was trans-ferred to a culture tube containing 2.7 ml Nutrient Broth (5 g Bactopeptone and 3 g Bacto beef extract in 1 L water, DIFCO Laboratories,

spoil with a chisel plow to a depth of 20 cm and then covered with 20 cm ofestone.h a chisel plow to a depth of 20 cm.st were incorporated into graded spoil with a chisel plow to a depth of 20 cm.

L. Chen et al. / Fuel 105 (2013) 258–265 261

Detroit, MI). Log serial dilutions were made until a dilution oc-curred where bacterial growth was no longer evident. The cultureswere incubated by shaking at 180 rpm at 28 �C for 3 days, andmicrobial number was determined by noting the highest level ofdilution in which positive bacteria growth was evident. Bacteriapopulation was calculated based on dry weight of soil.

2.4.3. Microbial biomass C determinationMicrobial biomass C was prepared using microwave irradiation

and K2SO4 extraction and determined by the rapid oxidation spec-trophotometric method [23]. A 10 g oven-dried equivalent (ODE)of field-moisture soil was placed into a 50-ml centrifuge tubeand adjusted to 80% water-filled porosity by adding deionizedwater as needed. Microwave energy was applied at a total of800 J g�1 ODE soil. To minimize heat pockets within the moist soil,samples received two times of 400 J g�1 ODE soil with stirring inbetween for uniform mixing. Control samples remained at roomtemperature and were not exposed to microwave irradiation.Added 25 ml 0.5 M K2SO4 (pH 7.0) to a centrifuge tube and ex-tracted by horizontal shaking at 250 rpm for 60 min, then filteredto obtain soil-free filtrate. Concentrations of C in the extracts weredetermined by the rapid oxidation spectrophotometric method asdescribed by Islam and Weil [23].

2.4.4. Bacterial diversity determination2.4.4.1. DNA extraction. DNA was extracted and purified from250 mg of fresh soil from each plot and untreated area by usingthe Power Soil DNA Kit (MoBio Laboratories, CA) following manu-facturer’s recommendations. DNA integrity was checked by elec-trophoresis on a 1% agarose gel. Purity of extracted DNA wasfurther measured and quantified using a Nanodrop 1000 Spectro-photometer (Thermo Scientific, Wilmington, DE). DNA yields ran-ged between 7.5 and 12.5 lg g�1 dried soils.

2.4.4.2. Polymerase chain reaction (PCR) amplification of the 16 S RNAgene. A PCR was conducted using a set of universal bacterial prim-ers, PRBA 338 and PRUN518R, that amplify the 338–518 region ofthe 16s rRNA gene of bacteria. For PCR reactions, 100 ll of finalmixture volume containing 1 lM of each primer, 50 ll of GoTaqGreen Master Mix, 2X (Promega, Madison, WI) and 1 ll DNA(10–20 ng) template were used. The PCR reactions were performedusing an automated thermal cycler (PTC-100, MJ Research, Wal-tham, MA). The temperature program for the PCR reaction startedwith a 94 �C denaturation step for 9 min. Then 30 cycles were con-ducted in which each cycle included a denaturing step of 94 �C for30 s, an annealing step of 55 �C for 30 s and an extension step of72 �C for 30 s. The last step in the PCR program was a final exten-sion at 72 �C for 7 min. The samples were then held at 4 �C beforebeing stored in a freezer at �20 �C.

2.4.4.3. Electrophoresis of the 16 S RNA gene. A BioRad DCode appa-ratus (BioRad, Hercules, CA) was used to conduct the denaturing gra-dient gel electrophoresis (DGGE) analysis. An 8% (w/v)polyacrylamide gel, with denaturing gradients ranging from 35% to65% was used for separation of PCR products obtained as describedabove. Urea and formamide were used as denaturants to facilitatethe separation of DNA fragments. DGGE was performed using theDcode Universal Mutation Detection System (BioRad Laboratories)and a 16 cm/16 cm gel apparatus. The gel was loaded and run in 1X TAE (20 mM tris–Cl, 10 mM acetate, 0.5 mM Na2EDTA) buffer at60� C for a total of 780 V h (constant voltageof 130 V for 6 h). Gels werethen stained with ethidium bromide and visualized on a UV transillu-minator and photographed (Gel Logic Unit, Kodak, California, USA).

2.4.4.4. Calculation of bacterial richness and Shannon–Weaverindexes. A diversity richness index calculated using the DGGE

banding pattern was used to quantify the different soils numeri-cally. The mean band number for each soil was used to calculatethe richness index. For our analyses, bands that could be clearlydiscerned as being distinct and separated from other bands, evenif faint, were marked. The existence of the bands was further con-firmed by comparing the normal gel pictures with an invertedimage.

A maximum value of 1.00 was assigned to the mine soil withFGD treatment due to the maximum number of bands for this soil.By using the richness index it was possible to differentiate the soilsbased on the bacterial diversity observed in each soil. The phylo-type richness (S, number of bands) was calculated for each minesoil and was normalized in comparison to the mine soil withFGD treatment that was assigned an index value of 1.00. In thisevaluation of richness, the higher the value, the more diverse interms of the number of dominant species that were in the soilsample.

The Shannon–Weaver index (H) was used to measure the DNAsequence diversity of soil microbial community. The parameterwas estimated using the equation below:

H ¼ �Xs

i¼1

pi ln pi ¼ �Xs

i¼1

ðNi=NÞ lnðNi=NÞ

where Pi is the percentage of the ith DGGE fragment gray degree toeach DNA sample, Ni is the net gray degree quantity (subtracted bythe background gray degree quantity of a gel) of the ith DGGE frag-ment to each DNA sample. N is the total net gray degree quantity ofall DGGE fragments examined in each DNA sample, and s is thenumber of RAPD fragments to each DNA sample.

2.5. Statistical analysis

The results in this study were analyzed statistically using amodel that included treatment as independent variables. Datawere subjected to analysis of variance (ANOVA) using the PROCGLM statement of the SAS statistics program [24]. When the anal-ysis generated a significant F value (P 6 0.05) for treatments, themeans were compared by the least significant difference (LSD) testusing the appropriate error term to calculate the LSD value.

3. Results and discussion

3.1. Effects of FGD product on plant biomass production

Vegetative cover is absolutely necessary to attenuate soil ero-sion of surface coal mined lands. Fourteen and 16 years after recla-mation, visual observation indicated that grasses and legumesgrew well in all treatment plots, but no plants grew in an adjacentuntreated area (Fig. 2). Generally, natural soil material is regardedas superior to amended spoil for plant growth. As expected, forshort-term treatments, in 1995 and 1997 the plant biomass pro-duction was higher for the SOIL treatment than the FGD andFGD/C treatments. Treatment differences were not significant in1996, 1998, 2008, but, plant biomass production was significantlyhigher for the FGD and FGD/C treatments than for the SOIL treat-ment in 2010 (Fig. 3).

Also, in 1995 and 1997 plant biomass production for the SOILtreatment was significantly greater than in 1996 and 1998. Thisdifference may be attributed to variation in climatic conditions.The annual total precipitation and total rainfall of the plant grow-ing season were below normal in 1996 and 1998, and the plant bio-mass data were not considered representative of years with normalrainfall. Long-term reclamation treatment effects in 2008 and 2010showed that plant biomass production was greater than in 1996and 1998 and representative of years with normal rainfall. There

0

1

2

3

4

5SOILFGD FGD/C

1995 1996 1997 1998 2008 2010

A

A

A A

B

B

B

B

B

Pla

nt B

iom

ass

(Mg

ha-1

)

Fig. 3. Vegatative biomass production in 1995, 1996, 1997, 1998, 2008, and 2010 atthe Fleming surface coal mine land reclaimed with SOIL, FGD and FGD/C.Treatments are described in Table 2. Different letters over each bar in the sameyear represent a significant difference at P 6 0.05.

0

2000

4000

6000

8000

10000

12000

14000

SOIL FGDFGD/C

0

1000

2000

3000

4000

5000

3000

4000

5000

Ca

Mg

S

C

A

B

B

A

A

A

A

BB

A

A

A A AA

BB

Con

cent

rati

ons

of C

a, M

g, a

nd S

(m

g/kg

)

262 L. Chen et al. / Fuel 105 (2013) 258–265

was no significant difference in the biomass production betweenthe FGD and FGD/C treatments in any year during the study periodwhen plant biomass was analyzed. In comparison to conventionalreclamation, plant biomass production for the plots with FGD andFGD/C treatments demonstrates the potential of FGD product, withor without addition of an organic amendment, to provide long-term sustainability of vegetative cover. However, in the early stageof the experiment (1–3 years), decreased plant biomass in thetreatments with FGD product was probably due to the saltinessof the FGD material, which impacts the germination of seeds andplant growth. Visual observation indicated germination rate waslower for the FGD and FGD/C treatments than the SOIL treatmentin 1995 and the compost seemed to help alleviate the effect ofthe salinity of the FGD material.

0

1000

2000 A A

B

1995 1996 1997 2008

Year

Fig. 4. Concentrations of Ca, Mg, and S in plant tissues in 1995, 1996, 1997, and2008 at the Fleming surface coal mine land reclaimed with SOIL, FGD and FGD/C.Treatments are described in Table 2. Different letters over each bar in the same yearrepresent a significant difference at P 6 0.05.

3.2. Effects of FGD product on concentrations of essential elements inplant tissues

Calcium, Mg and S are three secondary macronutrients for high-er plants and the major elements in the FGD product. Concentra-tion of Ca in the plant tissues was not significantly different forthe FGD or FGD/C treatment as compared to the SOIL treatmentin 1995, 1996 and 2008, and was increased only in 1997 (Fig. 4).Furthermore, Ca concentration in the plant tissues in 1997 wasgreater by the FGD/C treatment than by the FGD treatment. Con-centrations of Mg and S in the plant tissues were significantlygreater in 1995, 1996, and 1997 when mine soil was reclaimedwith FGD and FGD/C treatments compared with mine soil treatedwith SOIL. There were no differences in the concentrations of Mgand S between the FGD and FGD/C treatments. The large input ofMg and S to mine land by FGD product, as compared with soil, isan additional benefit of the FGD product as plant macronutrients.However, there were no significant differences for the concentra-tions of Mg and S in the plant tissues among treatments 14 yearsafter reclamation in 2008. Chen et al. [25] reported that S concen-tration in alfalfa tissues was increased by fluidized bed combustionproduct addition to acid soil but Mg concentration in the plant tis-sues was decreased.

Nitrogen, P, and K are three primary macronutrients for higherplants. Nitrogen concentration in the plant tissues was increased in1997 by the FGD/C treatment compared with the SOIL treatmentand was not affected in 2008 by the treatments (Table 3). Nitrogenconcentration was not determined in 1995 and 1996. Phosphorusconcentration was decreased in 1996 and 2008 by the FGD treat-ment compared with the SOIL treatment and was not affected in1995 and 1997 by the treatments. Potassium concentration was in-creased in 1996 and decreased in 2008 by the FGD treatment com-pared to the SOIL treatment and was not affected in 1995 and 1997

by the FGD and FGD/C treatments. Overall, the results of N, P, and Kconcentrations in the plant tissues were inconclusive. Becauseplant biomass contained a mixture of grasses and legumes, andthe treatments caused different proportion of grasses to legumesas was observed in 1995–1998, the differences in the concentra-tions of some elements were partly due to different proportionsof grasses to legumes.

Copper and Mo are essential micronutrients for higher plants.However, high concentration of Mo in forage tissues may causeMo toxicity in animals and induce a decrease of Cu concentrationin forage tissues, which may cause Cu deficiency in animals. Con-centrations of Mo in the plant tissues were increased in 1995,1996, and 1997 and were not changed in 2008 by the FGD andFGD/C treatments compared to the SOIL treatment (Fig. 5). Thetemporary increase of Mo in the plant tissues may cause Mo toxic-ity in beef cattle if Mo concentration in forages is more than 10 mg/kg [26]. Compared with plants by the SOIL treatment, vegetationgrown on mine soil treated with the FGD/C had greater concentra-tions of Cu in plant tissues in 1995 and 1997 but not in 1996 and2008. There was no statistical difference for Cu concentration be-tween the treatments of FGD and FGD/C. However, the Cu/Mo ratioin the plant tissues was generally less than 2:1, which might cause

Table 3Concentrations of selected plant essential elements in vegetation grown on the mineland reclaimed with SOIL, FGD, and FGD/C.

Elements SOIL FGD FGD/C LSD0.05

mg kg�1

1995

Fe 64.8 ba 170 a 168 a 69.3K 8490 6560 8070 2230Mn 257 b 312 b 465 a 78N NDb ND NDNi 0.74 b 3.47 a 4.08 a 1.14P 1350 1530 1580 244Zn 26.3 b 30.9 a 35.2 a 4.5

1996Fe 95.8 b 90.2 b 163 a 64.0K 15500 b 18700 a 14800 b 1640Mn 177 b 308 a 255 ab 82N ND ND NDNi 1.69 b 5.65 a 4.72 a 1.46P 1860 a 1620 b 1670 ab 221Zn 33.8 29.2 30.3 10.3

1997Fe 55.6 225 112 230K 21900 23400 22100 2780Mn 123 196 186 101N 15400 b 19300 ab 20900 a 4040Ni 1.42 b 3.99 a 3.98 a 1.69P 2120 2220 2250 263Zn 25.8 28.5 27.4 6.8

2008Fe 173 289 340 728K 10300 a 8250 b 7080 b 1180Mn 98.8 220 101 442N 12500 13200 11900 6410Ni 1.70 7.66 1.97 20.7P 2510 a 1900 b 2210 ab 598Zn 16.1 17.4 26.4 34.0

a Different letters in the same row are significantly different at the P 6 0.05 level.b Not determined.

0

10

20

30

40

50

60

SOIL FGD FGD/C

0

2

4

6

8

0

2

4

6

8

10

12

B

A

A

C

A

B

B

AA

AB

BA

B

A

AB

B

A A C

A

B

B

A

A

B

Cu

Mo

1995 1996 1997 2008

Con

cent

rati

ons

of B

, Cu,

and

Mo

(mg/

kg)

Year

Fig. 5. Concentrations of B, Cu, and Mo in plant tissues in 1995, 1996, 1997, and2008 at the Fleming surface coal mine land reclaimed with SOIL, FGD and FGD/C.Treatments are described in Table 2. Different letters over each bar in the same yearrepresent a significant difference at P 6 0.05.

L. Chen et al. / Fuel 105 (2013) 258–265 263

Cu deficiency in beef cattle [26]. If the reclaimed mined land isimmediately used for grazing, animals should be supplied somefoods rich in Cu.

Boron is an essential micronutrient for higher plants. Borondeficiency is a widespread nutritional disorder, and B deficiency-induced reduction or even failure of seed and fruit set is wellknown. The FGD product used in this study contained high concen-tration of B (Table 1). The concentration of B in the plant tissue wasgreatly increased by the FGD and FGD/C treatments in 1995, 1996,and 1997 compared to the SOIL treatment and was no differentamong treatments 14 years after reclamation in 2008 (Fig. 5). Itwas 7.1–8.1 times greater in 1995 by the FGD or FGD/C treatmentcompared with the SOIL treatment and was gradually decreasedduring the study period. This indicated that B in the FGD productwas easily taken up by the reclamation plants. Critical deficiencyrange is about 5–10 mg kg�1 in monocotyledons (e.g. orchardgrass, timothy, annual ryegrass, and wheat) and 25–60 mg kg�1

in red clover [27]. This was probably why legumes dominatedthe plots treated with the FGD and FGD/C, while grasses dominatedthe plots treated with SOIL determined by visual observation from1995 to 1998. Phytotoxic B concentrations vary from 100 to1000 mg kg�1 depending on plant species and cultivars. For exam-ple, critical toxicity levels in leaves are 100 mg kg�1 for corn,400 mg kg�1 for cucumber, and 1000 mg kg�1 for squash [27].The concentrations of B in the plant tissues reclaimed with theFGD product were far below the toxic levels (Fig. 5). Boron concen-tration in plant tissues often reaches to a toxic level when largeamounts of fly ash alone are applied to soil. Relatively low B con-centrations in this study were probably due to the FGD resulting

in higher soil (spoil) pH that makes the B less available for plantuptake.

Iron, Mn, Ni, and Zn are also essential micronutrients for higherplants. Compared to the SOIL treatment, Fe concentration in planttissues was increased in 1995 by the FGD and FGD/C treatmentsand in 1996 by the FGD/C treatment (Table 3). Concentration ofMn was increased in 1995 by the FGD/C treatment and in 1996 bythe FGD treatment. Concentration of Ni in plant tissues was in-creased in 1995, 1996, and 1997 by the FGD and FGD/C treatmentsbut not in 2008. Concentration of Zn in plant tissues was increasedby the FGD and FGD/C treatments in 1995 but not in later years.There was no difference for Ni and Zn concentrations between thetreatments of FGD and FGD/C. Overall, concentrations of plantmicronutrients B, Cu, Fe, Mn, Mo, Ni, and Zn in the plant tissues wereincreased by the FGD and FGD/C treatments short-term (1–3 years)after reclamation but not long-term (14 years) after reclamation.

3.3. Effects of FGD product on concentrations of environment regulatedelements in plant tissues

Table 4 shows concentrations of Resource Conservation andRecovery Act (RCRA)-regulated elements of As, Ba, Cd, Cr, Pb, Hg

264 L. Chen et al. / Fuel 105 (2013) 258–265

and Se in the plant tissues as affected by the treatments. Comparedwith the SOIL treatment, only the concentration of Pb was in-creased in 1996 by the FGD/C treatment and the concentration ofBa was decreased in 1995 and 1996 by the FGD or FGD/C treatmentdue to Ba in soil being precipitated by SO2�

4 in the FGD product.Chen et al. [25] reported that concentration of Ba was decreasedin alfalfa growing on acid soil treated fluidized bed combustionproduct compared to the untreated control. We do not have anexplanation for the concentration increase in Pb. There were nosignificant differences for the concentrations of As, Cd, Cr, Pb, Hgand Se among the treatments of SOIL, FGD, and FGD/C in all theyears of determination. The concentrations of some regulated traceelements in the plant tissues varied in different years. One reasonmay be simply due to laboratory variation from one year to an-other. Hopefully, this was a minor reason as we followed strictQA/QC protocols. Another reason was probably the differences inthe sampling time because in different stages of plant growth, con-centrations of elements in plant tissues are different. The third rea-son was probably different proportions of plant species in differentyears as mentioned previously.

3.4. Soil bacterial populations, microbial biomass carbon, and bacterialdiversity

All treatments were effective in raising surface (0–20-cm depth)soil bacterial populations 16 years after applications in 2010 com-pared to the adjacent untreated control (Fig. 6). It should be notedthat bacteria populations in the adjacent untreated area were onlyin the range of 103–104 cells g�1 soil, while in the reclaimed area

Table 4Concentrations of selected Resource Conservation and Recovery Act (RCRA)-regulatedelements in vegetation grown on the mine land reclaimed with SOIL, FGD, and FGD/C.

Elements SOIL FGD FGD/C LSD0.05

mg kg�1

1995As 0.076 <0.001 0.356 0.476Ba 10.4 aa 2.51 b 3.60 b 1.86Cd 0.101 0.108 0.134 0.082Cr 0.191 0.251 0.704 0.838Pb 1.07 0.91 1.14 0.44Hg NDb ND NDSe 2.37 2.84 2.73 2.29

1996As 1.20 1.17 0.16 1.54Ba 5.58 a 0.96 c 1.93 b 0.88Cd 0.298 0.276 0.254 0.205Cr 0.542 0.317 0.577 0.607Pb 0.47 b 0.83 b 1.51 a 0.56Hg ND ND ND NDSe 18.6 11.2 9.8 19.5

1997As <0.01 <0.01 <0.01Ba 3.77 3.06 1.84 2.37Cd 0.052 0.068 0.045 0.038Cr 0.147 0.309 0.194 0.192Pb 0.61 0.54 1.17 0.72Hg ND ND NDSe 0.09 0.55 <0.01 0.84

2008As <1.28 <1.28 <1.28Ba 12.1 12.0 7.1 21.7Cd 0.057 0.117 0.050 0.230Cr 0.74 2.83 1.13 7.04Pb 0.77 1.04 0.77 0.93Hg 0.020 0.026 0.015 0.020Se <2.32 <2.32 <2.32

a Different letters in the same row are significantly different at the P 6 0.05 level.b Not determined.

were in the range of 106–108 cells g�1 soil. Furthermore, bacteriapopulations were greater by the SOIL and FGD/C treatments thanby the FGD treatment. Emmerling et al. [7] reported that applica-tion of sewage sludge at 10–25 Mg ha�1 increased microbial activ-ities in mine soils 2 years after reclamation compared withapplication of mineral fertilizers. Lindemann et al. [18] reportedthat amendment of mine spoil with sewage sludge at 11.2 Mg ha�1

increased the number of microorganisms in the mine spoil oneyear after application.

Microbial biomass C is defined as the living portion of the soilorganic matter excluding plant roots and soil animals. It is an indi-cation of the overall microbial population size in the soil withoutregard to the types of microorganisms. Sixteen years after reclama-tion in 2010, there was no significant difference in microbial bio-mass C among the treatments of SOIL, FGD, and FGD/C (Fig. 6).However, soil microbial biomass C in the treatment plots was sig-nificantly higher than in the adjacent untreated area.

Amount of microbial biomass C in soil are dependent on parentmaterial. Many researches indicated that microbial biomass C inunclaimed mine spoils was very low [8,11,10]. Chodak et al. [11]reported that in Upper Silesia, Poland microbial biomass C in minesoils gradually increased with the post-mining time of 6, 20, and27 years under spontaneously developing pine forest stands. Mac-hulla et al. [10] reported in Halle, East Germany microbial biomassC in mine spoils at 0–1 cm depth was 9–39 mg kg�1 soil before rec-lamation, increased to 148–497 mg kg�1 within one year after ahay mulch-seeding treatment, and sustained these levels in thesecond and third years. Mummey et al. [9] reported that 20 yearsafter mine reclamation in Wyoming, USA microbial biomass C inthe mine soil was only 44% of that found in adjacent undisturbedsoils. Anderson and Stahl [12] reported that microbial biomass Cin the mine soils from 2 to 32 years after reclamation in Wyoming,USA was only approximately half of that found in adjacent undis-turbed soils. Jedidi et al. [28] applied different types of organicwastes at 40 ton ha�1 to a loamy–clayey soil and incubated for8 weeks at 25 �C. Microbial biomass C in the soil was increased

0

2

4

6

8

10

Untreated SOIL FGD FGD/C

C

A

B

A

Treatments

0

50

100

150

200

250

AA A

B

Mic

robi

al b

iom

ass

C (

mg

kg-1

)M

icro

bial

num

ber

10 c

ells

g-1

soi

l) (

Log

Fig. 6. Bacterial number and microbial biomass C in soil 16 years after treatment atthe Fleming surface coal mine land reclaimed with SOIL, FGD and FGD/C.Treatments are described in Table 2. Different letters over each bar represent asignificant difference at P 6 0.05.

Table 5Diversity indexes of microbial community in the mine land soils with differenttreatments calculated from DGGE.

Treatments Band richness index Shannon–Weaver index

Untreated soil 0.03 da 0.52 cSOIL 0.44 c 7.64 bFGD 1.00 a 14.9 aFGD/C 0.92 b 14.2 aLSD0.05 0.05 1.22

a Different letters in the same column are significantly different at the P 6 0.05level.

L. Chen et al. / Fuel 105 (2013) 258–265 265

by the organic wastes and positively correlated with the organic Ccontent but decreased with the period of incubation. However,microbial biomass C in this study was not increased by the FGD/C treatment compared to the FGD treatment (Fig. 6).

The total DNA patterns of the SOIL, FGD, and FGD/C treatmentswere characterized by higher diversity than that of the adjacentuntreated area. The mine soil treated with FGD had the highestnumber of distinguishable bands in the gel, while the adjacent un-treated control mine soil had only 3 and 0 discernible bands for thetwo replicate samples. The DGGE profile of DNA showed bandsattributable to specific bacterial species that were only dominantin the soil from the FGD treatment.

The phylotype richness (S, number of bands) was calculated foreach soil and was normalized with the soil of FGD treatment whichwas assigned an index value of 1.00 (Table 5). In this evaluation ofrichness, the higher the value, the more diverse in terms of numberof dominant species that were in the soil sample. Thus, the minesoil with FGD treatment had the greatest diversity of bacterial spe-cies and the untreated control mine soil, the least. However, thediversity from Shannon–Weaver index showed no significant dif-ference between the soils with FGD and FGD/C treatments(Table 5).

4. Conclusions

Reclamation of acid mine soil using FGD product can be suc-cessfully accomplished with a grass-legume sward. Long-termmonitoring of plants and soil microbial properties do not indicateany potential long-term negative impacts associated with the uti-lization of FGD product.

References

[1] Sutton P, Dick WA. Reclamation of acidic mined lands in humid areas. AdvAgron 1987;41:377–405.

[2] ODNR. Abandoned Mine Land (AML); 2010. <http://www.ohiodnr.com/mineral/federal/tabid/17851/default.aspx>.

[3] Sheoran V, Sheoran AS, Poonia P. Soil reclamation of abandoned mine land byrevegetation: a review. Int J Soil, Sediment Water 3 (2): Article 13; 2010.<http://scholarworks.umass.edu/intljssw/vol3/iss2/13>.

[4] Akala VA, Lal R. Soil organic carbon pools and sequestration rates in reclaimedminesoils in Ohio. J Environ Qual 2001;30:2098–104.

[5] Anderson JD, Stahl PD, Ingram LJ. Influence of reclamation managementpractices on microbial biomass carbon and soil organic carbon accumulation insemiarid mined lands of Wyoming. Appl Soil Ecol 2008;40(2):387–97.

[6] Reuter R. Sewage sludge as an organic amendment for reclaiming surface minewastes; 1997. <http://www.hort.agri.umn.edu/h5015/97papers/reuter.htm>.

[7] Emmerling C, Liebner C, Haubold-Rosar M, Katzur J, Shroder D. Impact ofapplication of organic waste materials on microbial and enzyme activities ofmine soils in the Lusatian coal mine region. Plant Soil 2000;220:129–38.

[8] Renella G, Pietramellara G, Mench M, Nannipieri P, Landi L, Ascher J, et al.Long-term effects of aided phytostabilisation of trace elements on microbialbiomass and activity, enzyme activities, and composition of microbialcommunity in the Jales contaminated mine spoils. Environ Pollut2008;152:702–12.

[9] Mummey DL, Buyer JS, Stahl PD. Soil microbiological properties 20 years aftermine reclamation: spatial analysis of reclaimed and undisturbed sites. Soil BiolBiochem 2002;34:1717–25.

[10] Machulla G, Scow KM, Bruns MA. Microbial properties of mine spoil materialsin the initial stages of soil development. Soil Sci Soc Am J 2005;69:1069–77.

[11] Chodak M, Nikinska M, Pietrzykowski M. Development of microbial propertiesin a chronosequence of sandy mine soils. Appl Soil Ecol 2009;41(3):259–68.

[12] Anderson JD, Stahl P. Soil ecological indicators of surface mine landreclamation success. In: 2002 National meeting of the American Society ofMining and Reclamation, Lexington, KY, June 9–13, 2002. Published by ASMR,3134 Montavesta Rd., Lexington, KY; 2002.

[13] Korcak RF. Fluidized bed material applied at disposal levels: effects on an appleorchard. J Environ Qual 1988;17:469–73.

[14] Stehouwer RC, Sutton P, Fowler RK, Dick WA. Minespoil amendment with dryflue gas desulfurization by-products: element solubility and mobility. JEnviron Qual 1995;24:165–74.

[15] Ritchey KD, Feldhake CM, Korcak RF. Calcium sulfate or coal combustion by-product spread on the soil surface to reduce evaporation, mitigate subsoilacidity and improve plant growth. Plant Soil 1996;182:209–19.

[16] Chen L, Dick WA, Nelson Jr S. Flue gas desulfurization by-products additions tosoil: alfalfa productivity and environmental quality. Environ Pollut2001;114:161–8.

[17] Chen L, Dick WA, Streeter JG, Hoitink HAJ. Ryegrass utilization of nutrientsreleased from composted biosolids and cow manure. Compost Sci Util1996;14(1):73–83.

[18] Lindemann WC, Lindsey DL, Fresquez PR. Amendment of mine spoil to increasethe number and activity of microorganisms. Soil Sci Soc Am J 1984;48:574–8.

[19] Haefner RJ, Rowe GL. Water quality at an abandoned Ohio coal mine reclaimedwith dry flue gas desulfurization by-products. US Geological Survey, FS-051-97. Columbus, OH; 1997.

[20] Sims JT. Lime requirement. In: Sparks DL, editor. Methods of soil analysis: Part3 – chemical methods. Madison, Wisconsin: Soil Science Society of America;1996. p. 491–515.

[21] The Ohio State University Extension Service. Ohio Agronomy Guide. BulletionNo. 472–12th edition. Columbus, Ohio: The Ohio State University ExtensionService; 1998.

[22] Jones JB Sr, Wolf B, Mills HA. Microwave digestion using CEM microwavedigestion system. In: Plant analysis handbook. Athens, GA: Micro-MacroPublishing; 1991.

[23] Islam KR, Weil RR. Microwave irradiation of soil for routine measurement ofmicrobial biomass carbon. Biol Fert Soils 1988;27:408–16.

[24] Institute SAS. SAS/STAT user’s guide. Cary, NC: SAS Institute Inc.; 2004.[25] Chen L, Dick WA, Kost D. Circulating fluidized bed combustion product

addition to acid soil: Alfalfa (Medicago sativa L.) composition andenvironmental quality. J. Agric. Food Chem. 2006;54:4758–65.

[26] Bull RC, Bagley CV, Stenquist NJ. Copper: an essential micronutrient for beefcattle. Cattle producer’s library CL629. Western beef resource committee;2008. <http://www.csubeef.com/dmdocuments/629.pdf>(verified 1.10.11).

[27] Marschner H. Mineral nutrition of higher plants. London: Academic Press;1986.

[28] Jedidi N, Hassen A, Cleemput O, M’Hiri A. Microbial biomass in a soil amendedwith different types of organic wastes. Waste Manage Res 2004;22(2):93–9.