Within Arkansas, Louisiana, Mississippi, Missouri, and Texas, rice acreage is rotated with soybean due to both crops’ adaptability to the clay soils of the midsouthern USA. Two row patterns, two maturity groups, and six seeding rates were examined at Stoneville, MS, in 2009-2010, with respect to soybean growth and yield produced on silt loam soil. Optimal yield for MG IV was 333,000 seed ha-1 (297,000 plants ha-1). Twin-row soybean increased seed yield 7 to 10% more than single-row due to greater LAI, NDVI, and node and pod production. Rice field experiments quantified N loss via ammonia volatilization and determined grain yield for various N sources and preflood application timing. Cumulative ammonia volatilization loss on Tunica clay was minimal (10% of applied N). Grain yields were 6% less when fertilizer was applied 10 days before flood (dbf) as compared to 1 dbf; N sources are available to minimize ammonia volatilization loss.

New low volatile formulations of dicamba and 2,4-D are being developed to reduce injury to non-tolerant crops. Experiments were conducted comparing dicamba and 2,4-D volatility as affected by formulation, soil moisture content and soil texture. All formulations were applied to soil contained in greenhouse flats. After application greenhouse flats were then placed between a bio-indicator row of cotton and soybean. An open ended plastic dome covered the treated flats and crops for 48 hours prior to removal. Plant heights and yield of bio-indicator crops were unaffected by 2,4-D and dicamba volatility. Soil texture and moisture content did not significantly affect volatility. The greatest soybean crop injury was observed following application of the dimethylamine salt of dicamba and cotton injury was greatest following application of the ethylhexylester salt of 2,4-D. Volatility was generally reduced with new formulations of dicamba or 2,4-D; however, volatility was not completely eliminated.

Nitrogen (N) fertilizer application constitutes a major portion of farmers' cost of production since N is the most applied nutrient in U.S. cotton production. Despite this, N uptake and use efficiency (NUE) in cotton remains below 50%, which presents challenges of environmental quality. Studies were conducted across 4 states in the US Cotton Belt with the overall objective of evaluating strategies to reduce loss of N to the environment, increase N uptake and NUE. The first study had two objectives: 1) compare NH3 volatilization from surface versus subsurface application/placement of granular (urea) and fluid N source (urea ammonium nitrate; UAN32); and 2) compare NH3 volatilization from urea and UAN treated with enhanced-efficiency fertilizer products. For this study, four A horizon soils of different types were collected from four sites in Virginia (VA), Georgia (GA), Tennessee (TN), and Texas (TX). The EEF products were N-(n-butyl) thiophosphoric triamide (NBPT), nitrapyrin, and ESN. In the first set of experiments (N placement experiments), it was found that across soil types, subsurface placement of granular N source reduced NH3 volatilization by 58 – 81% and subsurface placement of UAN reduced NH3 volatilization by 56 – 98%. In the second set of experiments (EEF experiments), it was found that NBPT reduced NH3 volatilization by 5 – 77% across soil types, and the highest reduction in losses by NBPT was observed on sandier and low CEC soils. Treating urea with both nitrapyrin and NBPT was more effective at reducing NH3 volatilization compared to treating urea with nitrapyrin alone. Based on our findings, subsurface application of granular and fluid N sources is recommended as strategy to reduce NH3 volatilization. Where subsurface placement is not possible, EEF products should be considered. The objectives of the second study were: 1) determine the effects of small grain and legume cover crops on N cycling; 2) evaluate the effects of cover crops and N fertilization on N uptake; and 3) evaluate the effects of cover crops on lint yield. Cover crops were winter fallow (winter weeds), small grain [cereal rye (Secale cereale)], legume mix [(50% crimson clover (Trifolium incarnatum): 50% hairy vetch (Vicia villosa)], and legume mix + rye [(67% legume mix:33% hairy vetch)]. Fertilizer N application rates were 0, 45, 90, and 135 kg ha-1. Soil inorganic N in the top 30 cm depths of the legume mix and legume mix + rye plots was consistently higher than in the rye lone or fallow plots. Cotton lint yield following legume mix with 45 kg ha-1 fertilizer N application was comparable to following fallow plots with 135 kg N ha-1. Thus, fertilizer N rate could considerably be reduced when cotton follows legume cover crops. The objectives of the third study were: 1) evaluate urea and UAN placement (broadcast, dribbling, and injection) on lint yield and fiber quality of three cotton maturity groups (early-, mid-, and full-maturity); (2) assess N use and agronomic efficiencies as influenced by N source, rate, and placement; (3) evaluate the impact of N source and placement on fiber quality. A study including 9 site-years studies was conducted in VA, GA, and TX. It was found that placement had no effect on yield in VA, had effect in all 3 years in TX, and had effect in 1 year in GA. Yield responded to N application in 8 out of 9 site-years in this study. Nitrogen use efficiency was highest among the early- and mid-season varieties. Overall, N rate and variety, rather than application/placement strategy, had the most pronounced effects on lint yield. / Doctor of Philosophy / Nitrogen (N) is usually the major limiting nutrient in cotton production and represents a significant cost of production. On average, the current proportion of applied N recovered in the aboveground crop biomass, (i.e., nitrogen use efficiency, or NUE) ranges from 33 – 50%, meaning that up to two-thirds of applied N is not recovered by crops. This unrecovered N not only represents economic loss to growers, but acts as a potential pollutant in the environment. There is a need for practices which increase N uptake in cotton production, agronomic efficiency, and environmental sustainability. Previous studies conducted outside the U.S. Cotton Belt reported that NUE is influenced by N source and rate of application. Data on NUE of contemporary cotton varieties utilized in the humid and semi-arid regions of the U.S. Cotton Belt would prove useful in efficient N management in the region. First study evaluated gaseous N loss from fertilizer application. It was found that subsurface placement of granular urea reduced NH3 loss as much as 58 – 81% compared to surface broadcast granular urea and subsurface placement of fluid N source reduced NH3 loss by 56 – 98%. In a second study, N rate and method of application/placement were evaluated. Here, it was found that N rate and cotton variety, rather than application/placement strategy had a more pronounced effect on cotton yield. Lastly, the potential of cover crops as alternate N source in cotton production was investigated. It was found that cotton yield following legume mix and fertilized with 45 kg of N per hectare was comparable with cotton yield following no cover crop and fertilized with 135 kg of N per hectare. The results of these studies indicate that subsurface placement of granular and fluid N sources can reduce NH3 loss. In addition, all other things being equal, choosing the right cotton variety as well as applying the right N rate are critical for yield. Furthermore, by growing cotton after legume cover crops, N fertilizer application rates can be significantly reduced.

Phytoremediation is an attractive remediation technology due to its relative low cost and maintenance requirement. Acceptance of phytoremediation requires that the contaminant removal mechanisms are characterized and demonstrated in the field. Quantification of contributions from each mechanism to the overall remediation rate is crucial for optimization of phytoremediation systems, risk management and prediction of the total remediation time. The objective of this research was to characterize and quantify removal mechanisms for naphthalene at a creosote-contaminated site with poplar trees in Oneida, Tennessee. Groundwater monitoring for seven years in the surficial aquifer at this site demonstrated a reduction in polycyclic aromatic hydrocarbons (PAHs) with selective removal of naphthalene and three-ring compounds. Naphthalene mass loss mechanisms investigated at this site are biodegradation in the saturated zone, volatilization and biodegradation in the vadose zone and phytovolatilization. This is probably the most comprehensive field study of PAH phytoremediation mechanisms conducted to date. The significance of this research is to contribute to predictions of remediation time and end result for phytoremediation of PAHs. The understanding of in situ factors controlling each mechanism can facilitate future optimization of phytoremediation systems as well as improve risk assessment and monitoring strategies. Biodegradation rates were determined for different conditions at this site with in situ respiration tests, laboratory soil microcosms and laboratory soil columns. The combined remediation mechanisms of volatilization and biodegradation in the vadose zone were investigated in the field and in laboratory columns. Field measurements show that lower groundwater elevations in the summer and early fall lead to elevated groundwater concentrations of naphthalene and increased volatilization. The increase in the fraction of the porespace occupied by gas (gas saturation) in the unsaturated zone during the summer and fall further enhances the volatilization by increasing effective diffusion rates. Water consumption and interception by the phytoremediation system are believed to enhance mass transfer to the vadose zone. Column experiments and field measurements show that more than 90% of the naphthalene vapors are biodegraded within 5-10 cm above the groundwater table. The data indicate that biodegradation increases the overall volatilization flux out of the source by 10-300 times, when the source is exposed directly to the gas phase. In situ the naphthalene is generally dissolved from the source into the groundwater and then volatilezed from the groundwater to the gas phase. Under these conditions biodegradation in the vadose zone will still indirectly have an enhancing effect on the flux out of the source. This is the result of removal naphthalene from the soil gas by biodegrdation driving removal from the groundwater by volatilization, which in turn drives dissolution form the source into the groundwater. Phytovolatilization was quantified in flux chambers mounted on trees and calculated from transpiration rates. A laboratory uptake study and analysis of tree cores from the site provided supplementary evidence for naphthalene uptake by poplar trees. Phytovolatilization was detected throughout the year and was highest in the summer and fall when the groundwater concentrations were highest and transpiration was active. The role of biodegradation relative to physical removal mechanisms was compared for a year, for winter and summer conditions and with and without the impact of phytoremediation. Biodegradation of naphthalene in the saturated zone dominates by orders of magnitude over the removal by volatilization and phytovolatilization of naphthalene at this site. The removal of the total residual naphthalene mass was estimated to require up to 100 years with phytoremediation, but more than twice as long without phytoremediation. The estimated removal of naphthalene was three times larger in the summer than in the winter due to slower biodegradation in the saturated zone and smaller rates of volatilization to the vadose zone in the winter. The research shows that phytoremediation enhances the overall naphthalene removal, mainly by stimulating faster biodegradation in the rhizosphere and promoting mass transfer of naphthalene to the vadose zone followed by rapid vadose zone biodegradation. In the future, phytoremediation research focusing on the capillary zone is desirable. / Ph. D.

Switchgrass (Panicum virgatum L.) has potential as a biofuels feedstock. Major management questions include cultivar selection, cutting management, and optimizing N fertilization. Four cultivars of switchgrass were evaluated under two cutting regimes at eight locations within KY, NC, TN, VA, and WV in 2000 and 2001. Harvests were made once (in early November) or twice (midsummer and early November). Biomass yields averaged 15 Mg ha-1 and ranged from 10 to 22 Mg ha-1 across locations and years. There was no yield advantage to taking two harvests of the lowland cultivars ("Alamo" and "Kanlow"). If harvested twice, the upland cultivars ("Cave-in-Rock" and "Shelter") provided yields equivalent to the lowland ecotypes. A closer look at Alamo revealed much higher N removal in the midsummer harvests, late-season N translocation out of tillers, and fewer tillers developing under one-cut management. Switchgrass appears to be capable of truly perennial productivity in the upper Southeast USA with 50 kg N ha-1 yr-1 and a single harvest. A second field study was conducted on "Cave-in-Rock" switchgrass at Orange and Blacksburg, VA to examine N dynamics. For the 3-yr study, N fertilizer was applied once in May 2001 at 0, 90, 180, or 270 kg N ha-1. Switchgrass was harvested once (early November) in 2001 and twice (early July and early November) in 2002 and 2003. Tissue, root, and soil samples were collected in May, July, September, and November each year. Nitrogen fertilization had no effect on yield in 2001 and small residual effects in 2002 and 2003. Higher N removal was observed with two-cut management, where a high-yielding July cut had high shoot N concentrations. The amount of N removed as biomass from the 0 N treatments over 3 yr was 227 kg N ha-1; obviously significant amounts of N can be made available by these soils without any fertilizer applied. During the growing season, higher mineral N in soil was observed in July and September, when warmer temperatures increase microbial activity and N mineralization. Nitrogen use efficiency declined with increasing N rates. The low N response could be due to "native" N, to microbiological interactions, and/or to the ability of the plants to create internal N reserves. Proper N management of switchgrass must take into account the dynamics of several N pools. Greenhouse studies were conducted to establish switchgrass' responses to N and P under well-defined, soilless conditions and to examine two N sources. Shoot biomass increased with N fertilization with an observed inflection point at 210 kg N ha-1. In these pot studies, root biomass increased with N only to 115 kg N ha-1. No significant effect of P above 30 kg ha-1 was observed in shoot or root biomass. Biomass yield and tiller number were highly correlated. Biomass production was two times greater with ammonium sulfate than with urea when each was applied at equivalent N rates. Taken together, these findings suggest soils in the upper Southeast USA can supply significant amounts of the N needs of switchgrass, especially when harvested once at the end of the season. / Ph. D.

Synthetic nitrogen (N) fertilizers play a key role in human nutrition and crop production. The most widely used N source globally is urea; however, N loss via ammonia volatilization can be great in agricultural systems where urea is surface-applied. The objectives of the experiments reported in this dissertation were: 1) evaluate the performance of a new laboratory ammonia volatilization measurement system for measuring ammonia volatilization from coated granular urea; 2) determine if urease can be extracted from corn and soybean residues; 3) determine if differences in urease activity are present in corn and soybean residues; and 4) evaluate N content and yield of corn treated with surface-applied coated urea fertilizers. The laboratory ammonia volatilization system had a system recovery efficiency (SRE) of 97% of the applied N and the lowest variation in mg N captured in the acid traps when the air flow rate was 1.00 L min⁻¹, at 26°C, and an acid trap volume of 100 ml 0.02M phosphoric acid. Ammonia volatilization was greatest from 12-24 h after N application with a total of 17% of the applied N being lost during that period. The urease inhibitor N-(n-butyl) thiophosphoic triamide (NBPT) was the most effective ammonia volatilization control treatment and reduced ammonia losses 30-40% compared to urea in the laboratory trials. Urease was extracted from soybean residue and retained activity during extraction; however, urease from corn residue could not be identified in extracts. The agronomic field trials indicated that NBPT increased N concentration in corn ear leaves; however the effect on corn grain yield was masked by environmental conditions. The data from this study suggests that ammonia volatilization from granular urea can be effectively controlled using NBPT, and corn tissue N content in the field indicates that NBPT allows for more N to be utilized by the plant. The urease extraction showed that there may be differences in urease activity in different crop residues. Further research is needed to determine if varying levels of volatilization control are needed for urea applied to different crop residues in no-till systems. / Ph. D.

Nitrogen deficiency is characteristic of many mid-rotation loblolly pine (Pinus taeda L.) plantations in the Piedmont region of the southeastern USA. Fertilization with urea is the most common method used to correct this deficiency. Previous studies show that urea fertilization produces a rapid pulse of available nitrogen (N) with only a portion being utilized by plantation trees. Controlled release fertilizers release available N more slowly over a longer period of time and therefore may result in greater uptake efficiency. The objective of this study was to compare Nitroform®, a urea-formaldehyde controlled release N fertilizer versus urea and a control by measuring the effects of the two fertilizer treatments on N availability and loss as: total KCl extractable-N, total ion exchange membrane-N (IEM-N), N mineralization, and N volatilization, in a mid-rotation loblolly pine plantation in the Piedmont of Virginia. In addition, mid-summer and mid-winter fertilizations were compared to assess fertilizer uptake as a function of season. After the summer fertilization, Nitroform® significantly increased total KCl-extractable N, IEM-N, and N mineralization for two to three months over urea and the control. Three hundred times more N volatilized from urea than from controlled release Nitroform®. Interestingly, seven months after the summer application, the controlled release Nitroform® showed marked immobilization for three months while urea demonstrated greater N mineralization. After the winter application, fertilization with urea demonstrated greater soil inorganic N concentrations for two to three months over Nitroform®, very little N was immobilized, and volatilization was only 10 times that of Nitroform®. After summer and winter fertilizations, both fertilizer treatments significantly increased soil inorganic N concentrations and N volatilization over controls, however did not significantly increase N mineralization over controls when average response was tested over the entire sampling period. In addition to the fertilizer effects measured, a thinning only treatment was also incorporated into this study with soil N-availability indices compared to a control with no thinning or fertilization. The results from the thinning only treatment demonstrated no significant increases over the control in total KCl extractable-N, IEM-N, N-mineralization, or N volatilization when average responses were tested over the entire sampling period. / Master of Science

O manejo eficiente de fertilizações nitrogenadas em sistemas de exploração a pasto é peça chave para que melhores índices produtivos sejam alcançados. O conhecimento das perdas intrínsecas a esta prática e de fontes nitrogenadas alterntivas às normalmente usadas proporciona a determinação de melhores eficiências e flexibilidade na escolha da forma do nitrogênio aplicado. Assim o presente estudo teve o objetivo de avaliar as perdas por lixiviação de nitrato e volatilização de amônia, juntamente com a recuperação aparente do nitrogênio, em pastagens de capim elefante manejadas sob elevada fertilidade do solo. O experimento foi conduzido em área do depzrtamento de Zootecnia da ESALQ/USP, em Piracicaba. Adotou-se o delineamento experimental blocos ao acaso, com quatro blocos e seis tratamentos, que constaram da aplicação de 100 kg de N.ha-1 por ciclo, sob a forma de uréia, nitrato de amônio, etserco de curral, cama de frango e ajifer. Ao todo foram realizados três ciclos de amostragem, de novembro de 2006 a fevereiro de 2007, totalizando a aplicação de 300 kg de N.ha-1. Todos os resultados foram submetidos ao teste t à 5% de significância para a realização da análise estatística. As médias de produção de foram maiores para os tratamentos adubados com cama de frango e esterco, sendo de 3878,89 kg MSV.ha-1 e 3873,67 kg MSV.ha-1, respectivamente. Não foram observadas diferenças entre as recuperações aparentes do nitrogênio aplicado. As perdas por lixiviação de nitrato foram diferentes somente entre os ciclos, como consequência dos maiores índices de precipitação observados no terceiro ciclo amostrado, que apresentou média de -0,2.10-4 kg de NO3-.ha-1dia-1. Os valores de amônia volatilizada foram diferentes entre os ciclos e entre os tratamentos, sendo que no primeiro ciclo as maiores perdas foram atribuídas ao esterco, com 20,58% do nitrogênio aplicado. A cama de frango apresentou as maiores perdas de volatilização durante o segundo ciclo, com perda de 10,71% do nitrogênio aplicado, enquanto no terceiro ciclo a uréia apresentou a maior perda de nitrogênio, com 22,23% do nitrogênio volatilizado. As perdas por volatilização mais expressivas foram registradas até 60 horas após a aplicação das fontes. / The efficiency of nitrogen fertilization is one of the key elements to grass production. The knowledge of nitrogen losses within this practice associated with the application of alternative nitrogen sources provide efficient use and flexibility in choosing between nitrogen fertilizers. The objective of this study was to evaluate nitrate leaching, ammonium volatilization losses and apparent recovery of the applied nitrogen in elephant-grass pasture managed under high soil fertility. It was conducted at the Animal Science Department of ESALQ/USP, in Piraicaba - S.P. The statistical design was in complete randomized blocks, with four replicates and six treatments, which were the use of 100 kg N.ha-1 as urea, ammonium nitrate, dairy manure, chicken litter and ajifer. There were three cycles of evaluations, from November 2006 to February 2007, resulting in 300 kg N.ha-1 applied during the whole period. The results were statistically analysed by the t test with 5% significance. Chicken litter and dairy manure treatments provided the higher dry matter yields of 3878,89 DM.ha-1 and 3873,67 kg DM.ha-1, for each source respectively. Nitrate leaching losses were significant in the third cycle when -0,2.10-4 kg NO3-.ha-1.day-1 was lost. Ammonia volatilization was different between cycles and sources. In the first cycle the dairy manure resulted in higher losses of 26,42 kg N.ha-1. Chicken litter and urea lost 23,97 and 22,27 kg N.ha-1, respectively. In the second cycle urea presented the higher losses of 22,50, while dairy manure and chicken litter lost 6,01 and 15,88 kg N.ha-1. During the third cycle urea presented higher losses than , with 40,12 kg N.ha-1, dairy manure and chicken , which were 2,71 and 11,93 kg N.ha-1. Higher volatilization losses were observed until 60 hours after fertilization.

O presente experimento de laboratório foi realizado no Departamento de Ciência do Solo da ESALQ/USP, em Piracicaba, SP, com o objetivo de avaliar o efeito da presença de diferentes resíduos vegetais na superfície do solo sobre as perdas de amônia por volatilização decorrentes da aplicação de uréia em três doses. Foram utilizadas amostras da camada 0-0,20m de um Nitossolo Vermelho Eutroférrico do município de Piracicaba, SP. Os tratamentos constaram de arranjo fatorial com três repetições envolvendo quatro doses de uréia (0, 60, 120 e 180 mg dm-3 de N), aplicadas na superfície da terra, e quatro formas de cobertura da terra: sem cobertura e com cobertura constituída de fragmentos de folhas de plantas de milho, cana-de-açúcar e amendoim forrageiro, igualmente aplicados na superfície na quantidade de 29g de matéria seca por frasco (equivalente a 5 Mg ha-1). A volatilização de amônia foi avaliada em frascos de plástico de 1,6 L de capacidade, fechados, contendo 0,4 kg de terra umedecida e, para captar a amônia volatilizada, um disco de papel de filtro tratado com 1 mL de H2SO4 de concentração variando de 0,05 a 0,5 mol L-1. Esse disco era substituído diariamente, sendo o ácido remanescente no filtro exposto à amônia, titulado com NaOH 0,02 mol L-1 na presença de indicador bromocresol verde. Ao final do período experimental de 40 dias, amostras da camada superficial (0-0,05m) de terra dos frascos foram coletadas com amostrador especial e submetidas à determinação do pH em CaCl2 0,01 mol L-1. Os picos de máxima perda de amônia por volatilização ocorreram no 3º e no 12º dia após a aplicação da uréia, não havendo efeito dos resíduos culturais sobre a distribuição das perdas ao longo do tempo. As perdas de amônia aumentaram com o aumento da dose de uréia aplicada. Nas terras sem cobertura e naquelas contendo resíduos de milho e de amendoim forrageiro os aumentos foram mais expressivos para as doses menores (60 e 120 mg dm-3 de N-uréia), enquanto nas terras com restos de cana-de-açúcar os maiores aumentos ocorreram para a dose mais alta (180 mg dm-3 de N-uréia). A presença de restos de cultura de amendoim forrageiro na superfície do solo aumentou a quantidade de amônia volatilizada das terras tratadas ou não com uréia, enquanto a de restos de cultura de cana-de-açúcar reduziu a quantidade volatilizada. Os restos de cultura de milho exerceram pouca influência nas perdas de NH3 por volatilização. O pH da camada superficial das terras aumentou em mais de duas unidades com a aplicação de uréia e dos resíduos vegetais. / A laboratory experiment was carried out at the University of São Paulo, Piracicaba campus, Brazil, in order to evaluate how the presence of different plant residues on soil surface will affect ammonia volatilization losses from urea fertilizer applied on soil surface in three rates. Samples from 0-0.2m layer of a ferric Red Latosol located in Piracicaba, State of São Paulo, Brazil, were used in the study. Treatments consisted of a factorial arrangement with three replicates involving four rates of urea (0, 60, 120 and 180 mg dm-1 of N), applied at soil surface and four forms of soil cover: with no cover and with cover consisted of leaf fragments of corn, sugar-cane and peanut plants also applied on the surface at the rate of 29 g of dryed material per flask (equivalent to 5 Mg per hectare). Ammonia volatilization was measured in closed 1.6 L flasks containing 0.4 kg of moistened soil sample and, to capture NH3, a filter paper disk treated with 1 mL of H2SO4 of concentration varying from 0.05 to 0.5 mol L-1. The paper filter was substituted every day and the remaining acid in the exposed filter was titrated with NaOH 0.02 mol L-1 in the presence of bromocresol green. At the end of the experimental period of 40 days, samples of the surface layer (0-0.05 m) of the soil in the flasks were collected using a special sampler and subjected to determination of the pH in CaCl2 0.01 mol L-1. Maximum volatilization occurred at the 3rd and the 12th day of urea application and no influence was observed of the plant residues on the distribution of the losses during the experimental period. Ammonia losses increased with the rate of urea application. For the bare soil samples and for samples covered with corn and peanut residues, these increases were more evident when urea was applied at the smaller rates (60 and 120 mg dm-1 of urea-N), whereas for samples covered with sugar-cane residues, increases were higher when the higher rate (180 mg dm-1 of urea-N) was used. Covering the soil with peanut residues increased the amount of volatilized ammonia in the samples treated and not treated with urea, while covering with sugar-cane residues reduced the volatilization losses. Corn residues showed to have only small influence on ammonia loss. The pH at the surface of the soil samples increased more than two units as the result of urea and plant residues application.

A contaminação ambiental provocada pelo despejo de resíduos industriais e pela aplicação de agrotóxicos pode resultar no desequilíbrio dos ecossistemas, além de causar danos diretos à saúde humana. No Brasil, existem várias áreas contaminadas com resíduos industriais, tais como o hexaclorobenzeno (HCB), um composto organoclorado usado no passado como fungicida, mas que atualmente tem seu uso proibido devido a sua alta persistência no ambiente e alta capacidade de bioacumulação. Este estudo verificou a possibilidade de remoção de resíduos de HCB do solo contaminado além da possibilidade de contaminação do ar e da água por este composto. Verificou-se: a) o efeito de adição de matéria orgânica, alteração de pH e alagamento da terra sobre a comunidade microbiana e a degradação do HCB; b) a volatilização do HCB e, c) a lixiviação deste composto. A contaminação inicial da terra foi determinada através de extração das amostras de terra contaminada e análise dos extratos por cromatografia gasosa (CG). Alterações das características da terra foram promovidas pela adição de matéria orgânica (bagaço de cana de açúcar ou vermicomposto), adição de Cal ou alagamento das amostras de terra contaminada, além de combinações destes tratamentos. Cada amostra de terra, colocada em frascos de vidro, recebeu uma solução de HCB radiomarcado (14C-HCB). A mineralização do 14C-HCB foi analisada através da captura do CO2 proveniente das amostras de terra por uma solução de hidróxido de potássio e quantificação do 14CO2 por Espectrometria de Cintilação Líquida (ECL). A volatilização do 14C-HCB foi analisada através da captura dos compostos voláteis por lâmina de poliuretano, extração desta lâmina e análise do extrato por ECL, para quantificação dos 14C-compostos voláteis, e por CG, para quantificação dos compostos voláteis (HCB e metabólitos). A formação de 14C-resíduos ligados e de metabólitos do HCB nas amostras de terra foi analisada através da extração destas amostras e posterior análise da terra extraída por ECL e do extrato por CG. Cada amostra de terra também foi analisada quanto à atividade microbiana, medida através da respiração dos microrganismos, e quanto a densidade microbiana, através da metodologia de contagem de unidades formadoras de colônias (UFC) de bactérias e fungos. A mobilidade e a lixiviação do HCB foram estudadas através de percolação de água em tubos de PVC contendo amostras de terra contaminada tratadas com bagaço de cana de açúcar e/ou cal e solução de 14C-HCB, dispostas sobre amostras de terra não contaminada; os tubos foram seccionados e a terra e a água lixiviada foram analisadas por ECL. O período dos estudos foi de 270 dias. A determinação da contaminação inicial da terra indicou a presença do HCB (3400 mg g-1 terra) e dos metabólitos 1,2,4,5 TCB (24 mg g-1 terra), 1,2,3,4 TCB (6 mg g-1 terra) e PCB (267 mg g-1 terra). A volatilização de 14C-compostos ocorreu em todas as amostras, principalmente nas amostras alagadas e com adição de matéria orgânica (29% a 40 %, após 270 dias). O alagamento favoreceu a volatilização provavelmente devido a baixa hidrossolubilidade do HCB e, a presença de matéria orgânica pode ter favorecido a formação de metabólitos mais voláteis através do estímulo à descloração redutiva, que ocorre sob condições de anaerobiose. Entretanto, a descloração redutiva não foi comprovada pois não houve formação nem aumento na concentração de metabólitos do HCB entre os compostos volatilizados. Houve uma diminuição na quantidade de 14C-compostos extraíveis no decorrer do tempo em todas as amostras, mas a concentração de HCB ou de seus metabólitos permaneceu constante independente do tratamento. A atividade e a densidade microbiana foram maiores nas amostras com bagaço de cana de açúcar mas não tiveram efeito sobre a volatilização ou a degradação do HCB. Não ocorreu mineralização ou mobilidade do 14C-HCB na terra contaminada e nem a formação de 14C-resíduos ligados. Portanto, a adição de matéria orgânica, a alteração de pH e o alagamento da terra não se mostraram eficientes para a remoção do HCB de terra contaminada. Além disso, a ocorrência de formação de 14C-compostos voláteis e a ausência de lixiviação do 14C-HCB indicaram que a presença de resíduos de HCB no solo pode levar à contaminação do ar mas não de água subterrânea. / The environmental contamination by industrial wastes and pesticides can cause harmful effects to ecosystem besides direct human health damages. In Brasil, there are several areas contaminated with industrial wastes like the hexachlorobenzene (HCB), a chlorinated organic compound used in the past as fungicide but that has no longer been produced due to its high persistence in the environment and bioaccumulation potential. This study verified the possibility of HCB residues removal from contaminated soil besides the air and water contamination risk. It was verified: a) the effect of organic matter addition, pH change and flooding of the contaminated soil on the microbial community and HCB degradation; b) HCB volatilization and, c) HCB leaching. The initial soil contamination was determined by extraction of contaminated soil samples and analysis of the extract by Gaseous Chromatography (GC). Soil properties changes were promoted by organic material addition (sugar cane bagasse or earthworm humus), lime addition and flooding of the soil samples besides a combination of these treatments. Each soil sample, placed in glass flasks, was treated with radiolabelled HCB solution (14C-HCB). The 14C-HCB mineralization was analyzed by the entrapment of CO2 originated from the soil samples by potassium hydroxide solution and quantification of the 14CO2 by Liquid Scintillation Spectrometry (LSS). The 14C-HCB volatilization was analyzed by the entrapment of the volatiles compounds by a polyurethane lamina, extraction of this lamina and analysis of the extract by LSS, for 14C-volatile compounds quantification, and by GC, for volatile compounds (HCB and metabolites) quantification. The 14C-bound residues and the HCB metabolites of the soil samples were analyzed by the extraction of these samples and further analysis of the extracted soil by LSS and analysis of the extract by GC. Each soil sample was analyzed with regard to its microbial activity by microbial respiration measure and microbial density by the plate count technique. The mobility and leaching of the HCB were studied by water percolation through PVC tubes containing contaminated soil treated with sugar cane bagasse and/or lime and 14C-HCB placed over no contaminated soil; the tubes were cut in sections and the soil and the leached water were analyzed by LSS. All these studies lasted 270 days. The initial soil contamination determination indicated the presence of the HCB (3400 mg g-1 soil) and the metabolites 1,2,4,5 TCB (24 mg g-1 soil), 1,2,3,4 TCB (6 mg g-1 soil) and PCB (267 mg g-1 soil). The volatilization of the 14C-compounds occurred in all the samples, mainly in the flooded samples with organic matter addition (29% to 40% after 270 days). The flooding enhanced the volatilization due the low HCB water-solubility and, the presence of organic material can have favoured the formation of metabolites more volatile than the HCB due to reductive dechlorination that occurs under anaerobic conditions. But the reductive dechlorination occurrence was not proved as there was no formation or increase in metabolite concentration. There was a decrease in 14C-extractable compounds quantity through the time in all the samples but the HCB and metabolites concentrations keep constant independently of the treatment. The microbial activity and density were higher in the samples with sugar cane bagasse but it did not affected the HCB volatilization or degradation. The 14C-HCB mineralization and the 14C-bound residues formation did not occur. The mobility of the 14C-HCB was not observed too. Therefore, the organic material addition, the pH change and the soil flooding were not efficient to HCB removal from the contaminated soil. And, the occurrence of the volatilization of 14C-compounds and the absence of 14C-HCB leaching indicated that the presence of HCB residues in the soil can result in contamination of air but not the ground water.