Cleanup of leaking underground storage tank (LUST) sites has been a priority for the United States of America (USA) for more than 20 years due to the large number of sites, the potential harmful health affects associated with gasoline components and the fact that single owners may not have the ability to pay for cleanup of these sites. In June 2006, the US Environmental Protection Agency (EPA) reported that of the 459,637 confirmed releases from USTs that had occurred previously, 342,688 had been remediated, which leaves 116,949 sites yet to be completed across the USA. Petroleum cleanup programs tend to be managed at the State level; however, there are wide variations among State programs in terms of information access, risk perception and funding availability. While each of the Federal and State UST programs has evolved to meet specific requirements, there has not been a comprehensive comparison of the individual State programs.In this thesis, State petroleum cleanup programs across the USA are evaluated to determine similarities and differences in an effort to identify factors that affect petroleum cleanup progress. Many parameters enter the equation in determining petroleum cleanup effectiveness. Not only are the parameters of the State program operation important, but also the characteristics of each State, including drinking water source and perceived risk associated with petroleum contamination, factor into the determination.A representative group of States and State petroleum cleanup programs were evaluated and the characteristics of States were compared to cleanup progress to determine factors affecting efficiency. Based on trend analysis the cleanup levels for toluene, ethylbenzene and total xylenes correlate directly to the cost of LUST site cleanup. For States with less perceived risk from petroleum contamination, the cleanup goals are less stringent; therefore, fewer resources and less time are required to complete site cleanup. Consequently, petroleum cleanup in States with less-stringent goals is achieved more efficiently. The knowledge of these drivers of efficient petroleum cleanup can be used to expeditiously pursue completion of the thousands of sites remaining across the USA.

The first chapter of this dissertation examines welfare impacts of a combination of subsidies to alternative fuels (AFs) and alternative fuel vehicles (AFVs), and how they compare to gasoline taxes. The particular AF examined here is ethanol that is produced from agricultural products in a small open economy. The model in this paper characterizes a country or state where gasoline is the major source of fuel for automobiles, but that also produces and consumes ethanol as an AF. Gasoline combustion is polluting and its use equals the total amount of emissions produced. Thus, a gasoline tax here is the same as an emissions tax and is the most direct environmental instrument. However, increasing gasoline taxes for pollution purposes is often politically not feasible. Thus, this paper studies how closely subsidies to alternative fuels (AFs) and alternative fuel vehicles (AFVs) emulate abatement behavior from a unit gasoline tax in a simple three sector general equilibrium model, and in the presence of pre-existing labor taxes. The model can also be used to track the effects of each policy on outputs, exports, and fuel use. The analytical results of the model are then calibrated to data from the largest ethanol producing state in the U.S., namely Illinois. The paper finds that subsidies can achieve up to 64 percent of the welfare gains from the gasoline tax, if the uncompensated wage elasticity is low enough or the elasticities of substitution between the transportation goods is high enough. The second chapter estimates behavior of households who jointly make discrete decisions about vehicle ownership and continuous decisions about miles driven. The paper uses seven years of data from 1995-2001 for the 35 states and union territories of India. The estimated parameters will be used to calculate elasticities of each different type of vehicle for percentage changes in petrol price per unit distance travelled and in vehicle taxes. The paper also computes income and price elasticities for petrol consumption. Two types of vehicles predominant in India are cars and two-wheelers such as motorcycles, mopeds, and scooters. The latter type of vehicle is more fuel efficient than the former. However, patterns of vehicle ownership across the country reflect a growing number of cars relative to motorcycles. This paper investigates the impact alternative policies such as taxes on petrol or on cars have on efficient methods of vehicle emission abatement in India. In particular, the chapter estimates the effect of each such policy on vehicle choice and driving behavior, and how they in turn affect emissions. The main results are summarized as follows: First, continuous choice own-price elasticities are higher for 4w relative to 2w, given age, and for older vehicles relative to newer ones, within each category. Second, discrete choice own-price elasticities with respect to capital cost are higher for 2w relative to 4w. Moreover, older vehicles of each type are more sensitive to higher vehicle prices relative to their newer counterparts. Third, income elasticities for discrete vehicle choices are all positive and greater than unity. Thus, higher income encourages purchase of newer vehicles of each type. Moreover, usage of vehicles rises with income, conditional on the particular vehicle choice. Finally, the paper conducts simulations that alter the price per kilometer by adding either an additional gas tax, a distance tax or an emissions tax. Results show that a distance tax reduces vehicle kilometers traveled the most, followed by an emissions tax and lastly by the gas tax. However, local emissions are reduced the most by an emissions tax, followed by a distance tax and then by a gasoline tax. Even though it would be ideal to compare the results obtained in this paper to results generated using a micro-level data set, the estimates presented here are indicative of whether a distance tax or a gasoline tax is more effective for emissions abatement in India. The third chapter of this dissertation evaluates how information asymmetry in private automobile markets affects programs to accelerate vehicle retirement, also known as scrappage programs. We use a dynamic framework where agents have heterogenous preference for car quality. Cars can either be new, or used. While all new cars have the same quality, used cars can be of high- or low-quality. The quality of a car is perfectly correlated with emissions. The goal of a scrappage program is to induce car owners to voluntarily scrap low-quality used cars. One key result is that in the presence of adverse selection a subsidy that maintains an active resale market unambiguously makes all types of consumers better off. However, if this option of implementing the subsidy does not exist, then the only other way to induce effective scrappage in our framework is to shut down the used car market. Welfare implications suggest that it might be better not to do anything rather than have a scrappage program such as the latter. / text

Petroleum hydrocarbons (PHCs) such as gasoline are ubiquitous organic compounds present at contaminated sites throughout the world. Accidental spills and leakage from underground storage tanks results in the formation of PHC source zones that release hundreds of organic compounds, including the high impact, acutely toxic and highly persistent aromatics (e.g., benzene, toluene, ethylbenzene, xylenes, trimethylbenzenes and naphthalene) into groundwater. Contamination by these compounds continues to persist until the PHC source zone is treated in place or removed. In situ chemical oxidation (ISCO) employing persulfate was identified as a potentially viable technology for the treatment of PHC source zones. The effectiveness and efficiency and, therefore, the overall economic feasibility of a persulfate-based ISCO treatment system depend upon the reactivity of the target organic compounds and the interaction of persulfate with aquifer media. The objective of this research was to investigate the persistence of unactivated and activated persulfate in the presence of aquifer materials, and to examine persulfate oxidation of PHC compounds at both the bench- and pilot-scales. A series of bench-scale studies were performed to estimate persulfate degradation kinetic parameters in the presence of seven well-characterized, uncontaminated aquifer materials and to quantify the changes in specific properties of these materials. Batch experiments were conducted in an experimental system containing 100 g of solids and 100 mL of persulfate solution at 1 or 20 g/L. Column experiments were designed to mimic in situ conditions with respect to oxidant to solids mass ratio and were performed in a stop-flow mode using a 1 g/L persulfate solution. The degradation of persulfate followed a first-order rate law for all aquifer materials investigated. An order of magnitude decrease in reaction rate coefficients was observed for systems that used a persulfate concentration of 20 g/L as compared to those that used 1 g/L due to ionic strength effects. As expected, the column experiments yielded higher reaction rate coefficients than batch experiments for the same persulfate concentration due to the lower oxidant to solids mass ratio. Bench-scale data was used to develop a kinetic model to estimate the kinetic response of persulfate degradation during these tests. The push-pull tests involved the injection of persulfate (1 or 20 g/L) and a conservative tracer into a hydraulically isolated portion of the sandy aquifer at CFB Borden, Canada. The kinetic model developed from the bench-scale data was able to reproduce the observed persulfate temporal profiles from these push-pull tests. This implies that persulfate degradation kinetics is scalable from bench-scale to in situ scale, and bench tests can be employed to anticipate in situ degradation. The estimated reaction rate coefficients indicate that persulfate is a persistent oxidant for the range of aquifer materials explored with half lives ranging from 2 to 600 days, and therefore in situ longevity of persulfate will permit advective and diffusive transport in the subsurface. This is critical for successful delivery of oxidant to dispersed residuals in the subsurface. Activation of persulfate is generally recommended to enhance its oxidation potential and reactivity towards organic compounds. This approach may influence the stability of persulfate-activator system in the presence of aquifer materials. A series of batch tests were performed to investigate persistence of persulfate at two concentrations (1 or 20 g/L) using three contemporary activation strategies (citric acid chelated-ferrous, peroxide and high pH ) in the presence of 4 well-characterized, uncontaminated aquifer materials. Chelation by citric acid was ineffective in controlling the interaction between persulfate and Fe(II) and a rapid loss in persulfate concentration was observed. Higher Fe(II) concentration (600 mg/L) led to greater destabilization of persulfate than lower Fe(II) concentration (150 mg/L) and the persulfate loss was stoichiometrically equivalent to the Fe(II) concentration employed. Subsequent to this rapid loss of persulfate, first-order degradation rate coefficients (kobs) were estimated which were up to 4 times higher than the unactivated case due to the interaction with Fe(III) and CA. Total oxidation strength (TOS) was measured for peroxide activation experiments and was observed to decrease rapidly at early time due peroxide degradation. This was followed by slow degradation kinetics similar to that of unactivated persulfate implying that the initial TOS degradation was peroxide dominated and the long-term kinetics were dominated by persulfate degradation. The kobs used to capture TOS degradation for later time were shown to depend upon unactivated persulfate and peroxide degradation rate coefficients, and peroxide concentration. Either a slow peroxide degradation rate and/or higher peroxide concentration allow a longer time for peroxide and persulfate to interact which led to kobs ~1 to 100 times higher than kobs for unactivated persulfate. For alkaline activation, kobs were only 1 to 4 times higher than unactivated persulfate and therefore alkaline conditions demonstrated the least impact on persulfate degradation among the various activation strategies used. For all activation trials, lower stability of persulfate was observed at 1 g/L as compared to 20 g/L due to insufficient persulfate and/or ionic strength effects. A series of batch reactor trials were designed to observe the behavior of the nine high impact gasoline compounds and the bulk PHC fraction measures subjected to various persulfate activation strategies over a 28-day period. This bench-scale treatability used unactivated persulfate (1 or 20 g/L) and activated persulfate (20 g/L). Activation employed chelated-Fe(II), peroxide, high pH or two aquifer materials as activators. No significant oxidation of the monitored compounds was observed for unactivated persulfate at 1 g/L, but 20 g/L persulfate concentration resulted in their near-complete oxidation. Oxidation rates were enhanced by 2 to 18 times by activation with peroxide or chelated-Fe(II). For alkaline activation, pH 11 trials demonstrated ~2 times higher oxidation rates than the unactivated results. For pH 13 activation the oxidation rates of benzene, toluene and ethylbenzene were reduced by 50% while for the remaining monitored compounds they were enhanced by 5 to 100%. Natural activation by both aquifer materials produced oxidation rates similar to the unactivated results, implying that either activation by minerals associated with aquifer material was not significant or that any potential activation was offset by radical scavenging from aquifer material constituents. Acid-catalyzation at pH <3 may enhance oxidation rates in weakly buffered systems. Oxidation of the monitored compounds followed first-order reaction kinetics and rate coefficients were estimated for all the trials. Overall, activated and unactivated persulfate appear to be suitable for in situ treatment of gasoline. Persulfate under unactivated or naturally activated conditions demonstrated significant destruction of gasoline compounds and showed higher persulfate persistence when in contact with aquifer solids as compared to chelated-Fe(II) or peroxide-activated persulfate systems. This observation was used as the basis for selecting unactivated sodium persulfate for a pilot-scale treatment of gasoline-contaminated source zone at CFB Borden, Canada where a ~2000 L solution of persulfate (20 g/L) was injected into a PHC source zone. Concentration of organics and inorganics were frequently monitored over a 4 month period across a 90 point monitoring fence line installed down-gradient. Treatment performance was measured by estimating organic and inorganic mass loading across the monitoring fence. Increased mass loading for sodium was observed over time as the treatment volume moved across the fence-line indicating transport of the inorganic slug created upon oxidant injection. The mass loading also increased for sulfate which is a by-product generated either due to persulfate degradation during oxidation of organic compounds or during its interaction with aquifer materials. Oxidation of organic compounds was evident from the enhanced mass loading of dissolved carbon dioxide. More importantly, a significant (45 to 86%) decrease in mass loading of monitored compounds was observed due to oxidation by injected persulfate. The cumulative mass crossing the monitoring fence-line was 20 to 50% lower than that expected without persulfate treatment. As the inorganic slug was flushed through the source zone and beyond the monitoring fence, the mass loading rate of sodium, sulfate and carbon dioxide decreased and approached background condition. Mass loading of the monitored compounds increased to within 40 to 80% of the pre-treatment conditions, suggesting partial rebound. These investigations assessed the impact of activation on persulfate persistence and treatability of gasoline and served to establish guidelines for anticipating field-scale persulfate behavior under similar conditions. In summary, unactivated persulfate is a stable oxidant in the presence of aquifer materials and its persistence depends upon TOC and Fe(Am) content of the materials, ionic strength, and aquifer to solids mass ratio. Persulfate exhibits significant destruction of gasoline compounds and can be employed for the remediation of gasoline-contaminated sites. Peroxide and chelated-Fe(II) enhance oxidation rates of these compounds, but reduce stability of the persulfate-activator system. Persulfate activation using high pH conditions does not significantly impact persulfate persistence but reduces the overall destruction of gasoline compounds. Therefore, activation imposes a trade-off between enhanced oxidation rates and reduced persulfate persistence. Kinetic model is representative of persulfate degradation at bench- and pilot-scales and can be used for estimation of in situ degradation. The quantification of oxidation rates for gasoline compounds under activated and unactivated persulfate conditions will assist decision-making for identification of appropriate remediation options when targeting contamination by gasoline or by specific high impact gasoline compounds. While persulfate oxidation resulted in partial treatment of a small gasoline source zone, aggressive persulfate load will be required during injection for a complete clean-up. Overall, persulfate-based in situ chemical oxidation was demonstrated to be an effective and a viable technology for the remediation of contaminated soil and groundwater.

This thesis develops a new methodological approach to test for threshold cointegration. It determines the threshold locations, the number of thresholds, and tests the null hypothesis of a unit root against the alternative of a stationary threshold process using p-values based on a residual-based block bootstrap for the nonlinear threshold autoregressive specification (TAR). Chapter 2 describes the methodological approach which combines Gonzalo and Pitarakis (2002) and Seo (2008). Chapter 3 employs Monte Carlo analysis to investigate the properties of the new approach. The results indicate that the methodology performs well and is suited for application to real world time series. Chapter 4 applies the new approach in combination with a threshold error correction model (ECM) to determine the spatial relationships among three crude oil prices: WTI, Brent, and Oman, from 2008 through 2011. The results indicate that the crude oil benchmarks are tied together by a long run relationship; however, the recent reversal in price premium between the two main crude oil benchmarks, WTI and Brent, is an anomaly that has resulted in a time period in which the series do not have a tendency to move back toward their long run relationship. Chapter 5 applies the new approach, in combination with threshold ECMs, with regime switches being triggered by the upstream markup margin to determine the vertical relationships between the crude oil, rack, and retail gasoline prices for six cities across North America. The results using both daily and weekly data between 2008 and 2011 suggest that upstream and downstream prices are cointegrated. There is evidence of band-TAR in which the crude, rack, and retail prices are free to diverge until the markup margin is squeezed or stretched beyond a lower or upper threshold. This suggests that abnormally high margins cannot be sustained indefinitely. The threshold ECMs indicate that there is no systematic relationship between the speed of adjustment and the markup margin; however, the residuals exhibit a leverage effect in which volatility and price changes are negatively correlated. Chapter 6 concludes with a summary of Chapters 2 through 5 and makes suggestions for future research. / Thesis (Ph.D, Management) -- Queen's University, 2012-06-17 22:53:24.922

In the past few decades spatial econometric models have become a standard tool in empirical research. Nevertheless applications in binary-choice models remain scarce. This paper makes use of Bayesian Spatial Probit Models to model and estimate spatial interactions in location decisions. For this purpose, we focus on the Austrian retail gasoline market, which is going through a process of remarkable structural changes. A short analysis shows that, during the last decade 10.9% of the stations had left the market and a percentage of 29.6% had either left the market or had changed the brand. This paper aims at investigating this process. A special characteristic of this market is the local competition structure which is characterized by spatial dependencies along local competitors. To capture these spatial dependencies and since the dependent variable is binary in nature (an exit had taken place or not), we apply a Bayesian spatial probit model using MCMC estimation on station level data for the whole Austrian retail gasoline market. Our results suggest, that the decision to leave the market, does not only depend on own characteristics, but also on competitors. In particular, we find the exit decisions to exhibit a negative spatial correlation. Moreover, our model allows to quantify spatial spillover effects of this market. (authors' abstract) / Series: Department of Economics Working Paper Series

Abstract Electric Vehicles (EVs) are considered a more sustainable alternative vehicle because of their efficient electric motor when compared to internal combustion engines (ICE), and thus help to mitigate environmental problems and reduce fossil fuel dependency. In or-der to support drivers of plug-in hybrid electrical vehicles (PEVs), the installation and adequate distribution of Electric Vehicle Service Equipment (EVSE) is a major factor. The availability of EVSE is a vital requirement in order to charge the vehicle’s battery pack through connection to the electricity grid. This thesis evaluates the likely distribu-tion of a sufficient number of charging stations, measured as the demand of EVSE, for initial deployment in the E14 highway. This highway is also known as the Green High-way region, where a plan has been outlined with the aim to create a fleet of 15% EVs in the area by 2020.In order to model EVSE distribution, the first step was to complete a survey in 2012 on the population density and location of cities, along with the location of already estab-lished charging station locations on the Green Highway. The survey was done with ge-ography information survey (GIS) software. The second step was to create a map of the region. Based on the map, the initial estimate of EVSE locations on the Green Highway project plan was analyzed, as the third step. This was used as an initial analysis. The forth step was to use the location of current gasoline stations to provide as alternative pattern for the EVSE sites.It was observed that the network of gasoline stations correlates positively with population density. Through using these stations, the optimal location of the EVSEs was proposed. However, the model results do not provide for sufficient placement of EVSE sites where the population density is very low. In order to assess the different potential options, it was necessary to create analytical models in Arc-GIS, in which buffer zones were created with a variable size of 10, 15, 20 and 31 miles. This permitted allocation of a geographical area to estimate the optimum sites for charging stations. The resultsiiishowed that for a buffer zone of 10 miles, 28 charging stations were calculated, using buffer zone of 15 miles gives 18 stations, and a buffer zone of 20 miles results in 13 charging station sites. Notably, the estimate of the 20-mile buffer zone gives the same results as for the 50 km (31 miles) buffer zone for residential areas along E14. Therefore, the results show that the optimal design is to deploy 14 fast charging stations with three-phase DC, or 14 fast charging stations with three-phase AC, installed adjacent to the E14 road.

The contamination of aquifer systems by petroleum hydrocarbons is a global problem. Underground storage tanks used for storing these hydrocarbons often leak, resulting in subsurface contamination. The hazards associated with petroleum hydrocarbon contamination are mainly attributable to the BTEX compounds, namely benzene, toluene, ethylbenzene and xylenes together with trimethylbenzenes (TMBs) and naphthalene due to their potential to impact human health and the ease with which they can enter the groundwater system. In situ chemical oxidation (ISCO) is the delivery of strong chemical oxidants to the subsurface for the purpose of treating organic contaminants. ISCO can be an effective way to remediate organic contaminants from the soil and groundwater. Sodium persulphate is one of the newer oxidants to gain widespread use in treating petroleum hydrocarbon contamination, though without being fully understood. This investigation tested the ability of unactivated sodium persulphate in treating dissolved phase and residual BTEX contamination through bench-scale laboratory tests and a pilot-scale field study. A degradation potential batch reactor test was carried out to assess the efficacy of unactivated sodium persulphate in oxidizing petroleum hydrocarbons present in contaminated groundwater as well as its effect on aquifer material from a field site. This test was carried out at a sodium persulphate concentration of 20 g/L. Results from this test did not follow the expected first-order degradation, and so subsequent experiments were carried out using a sodium persulphate concentration of 100 g/L. A test to determine the degree of interaction between the oxidant and aquifer material was also conducted. It was found that the degree of natural oxidant interaction for the field site in question was very low. 1000 kg of sodium persulphate was dissolved in nearly 10,000 L of water and injected into the subsurface. Electrical conductivity (EC), pH, sodium, persulphate, sulphate and BTEX were all monitored during the subsequent 152-day post-injection monitoring period. An empirical relationship was determined between EC and the concentration of sodium in groundwater. This enabled the use of EC as a real-time tracer to track the progress of the injectate. Field results supported predictions based on a simulation model that density-driven flow would play an important role in the delivery of the injectate. A portion of the injectate was believed to have been missed by the monitoring network. Areas that did show elevated tracer results in some cases showed a decrease in BTEX concentrations. Results were categorized in four ways. The first category had wells that showed strong evidence of injectate presence but little to no change in BTEX levels. The second category was comprised of wells that showed a reduction in BTEX levels along with the presence of injectate. BTEX levels in some wells rebounded towards the end of the study period. The third category consisted of wells that showed the presence of dilute injectate but did not show any reduction in BTEX concentrations. The fourth and final category was of wells that showed no evidence of having been affected by the injectate in any way. BTEX levels were the same as background. The oxidation of BTEX by unactivated sodium persulphate was found to be successful, though the vagaries of oxidant delivery and field sampling made difficult the accurate determination of the degree of success.

This thesis describes an experimental investigation of lean blowoff for H2/CO/CH4 mixtures in a swirling combustor. This investigation consisted of three thrusts. The first thrust focused on correlations of the lean blowoff limits of H2/CO/CH4 mixtures under different test conditions. It was found that a classical Damköhler number approach with a diffusion correction could correlate blowoff sensitivities to fuel composition over a range of conditions. The second part of this thesis describes the qualitative flame dynamics near blowoff by systematically characterizing the blowoff phenomenology as a function of hydrogen level in the fuel. These near blowoff dynamics are very complex, and are influenced by both fluid mechanics and chemical kinetics; in particular, the role of thermal expansion across the flame and extinction strain rate were suggested to be critical in describing these influences. The third part of this thesis quantitatively analyzed strain characteristics in the vicinity of the attachment point of stable and near blowoff flames. Surprisingly, it was found that in this shear layer stabilized flame, flow deceleration is the key contributor to flame strain, with flow shear playing a relatively negligible role. Near the premixer exit, due to strong flow deceleration, the flame is negatively strained i.e., compressed. Moving downstream, the strain rate increases towards zero and then becomes positive, where flames are stretched. As the flame moves toward blowoff, holes begin to form in the flame sheet, with a progressively higher probability of occurrence as one moves downstream. It is suggested that new holes form with a more uniform probability, but that this behavior reflects the convection of flame holes downstream by the flow. It has been shown in prior studies, and affirmed in this work, that flames approach blowoff by first passing through a transient phase manifested by local extinction events and the appearance of holes on the flame. A key conclusion of this work is that the onset of this boundary occurs at a nearly constant extinction strain rate. As such, it is suggested that Damköhler number scalings do not describe blowoff itself, but rather the occurrence of this first stage of blowoff. Given the correspondence between this first stage and the actual blowoff event, this explains the success of classical Damköhler number scalings in describing blowoff, such as shown in the first thrust of this thesis. The physics process associated with the actual blowoff event is still unclear and remains a key task for future work.

Thesis (Ph. D.)--University of Texas at Austin, 2007. / Vita. Includes bibliographical references.

Submitted by Dilene Paulo (dilene.fatima@ufcg.edu.br) on 2018-01-23T14:36:45Z No. of bitstreams: 1 CLARICE OLIVEIRA DA ROCHA – TESE PPGEQ 2014.pdf: 2084559 bytes, checksum: 35cd9fd0fad2c37f97e9a1e07b4e3877 (MD5) / Made available in DSpace on 2018-01-23T14:36:45Z (GMT). No. of bitstreams: 1 CLARICE OLIVEIRA DA ROCHA – TESE PPGEQ 2014.pdf: 2084559 bytes, checksum: 35cd9fd0fad2c37f97e9a1e07b4e3877 (MD5) Previous issue date: 2014-10-10 / CNPq / Propostas têm aparecido visando à remoção do enxofre nos combustíveis por intermédio dos processos de adsorção. Com isso, a utilização desse método para a dessulfurização vem se expondo como um bom potencial de aplicação, tanto do ponto de vista econômico como ambiental. As peneiras moleculares, do tipo silicoaluminofosfato (SAPO), vêm se apresentando como um excelente adsorvente, com uma ótima estabilidade térmica e hidrotérmica e com a impregnação de metais, tornando-se seletiva a adsorção. Assim, sintetizou-se SAPO-5 (microporoso) e SAPO-5M (mesoporoso), impregnados com diferentes teores dos metais de transição, níquel e/ou zinco. Esses adsorventes foram utilizados para a remoção do enxofre presente na mistura modelo (Heptano / Tiofeno). Para isso, foram realizadas técnicas de caracterização como TG, DRX, EDX e BET, bem como um acompanhamento cinético a fim de avaliar a eficiência da adsorção e as isotermas de equilíbrio. Foi confirmado a mesoporosidade do SAPO-5M, uma vez que o mesmo apresentou as isotermas características tipo IV (típica de material mesoporoso). O volume e o diâmetro dos poros do SAPO-5 microporoso e mesoporosos não foram totalmente obstruídos após a impregnação. Para os adsorventes mesoporosos houve um aumento de quatro vezes no diâmetro do poro, em relação ao micro. Embora, tenha apresentado redução na intensidade de difração de raios X após a impregnação, dos óxidos de níquel e/ou de zinco, obteve-se um material cristalograficamente ordenado. Ocorreu total remoção do direcionador, mostrando que a calcinação foi eficiente e o material é estável a altas temperaturas. Ademais, conseguiu-se atenuar a quantidade de enxofre de 500 ppm para cerca de 96,3 ppm, uma redução de 80%, para o adsorvente mesoporoso 0,25%Ni 0,75%Zn/SAPO-5M, cuja capacidade de adsorção de enxofre foi de 40 mg S/gadsorvente. O adsorvente com 1%Zn/SAPO-5, apresentou maior capacidade de adsorção (30 mg S/gadsorvente), dentre os materiais microporosos, conseguindo reduzir o teor de enxofre de 500 ppm para 163 ppm. Para os demais suportes e adsorventes ocorreu uma redução de aproximadamente 31 % do teor de enxofre na mistura modelo. Conclui-se que a síntese do mesoporoso foi eficiente e que o material utilizado se mostrou eficaz na remoção desse contaminante na mistura modelo. / Proposals have appeared aiming at the sulfur removal in fuels by means of adsorption processes. Thus, the use of this method for the desulfurization exposing themselves like a good application potential, both economically and environmentally. The molecular sieves silicoaluminiumfosfate type (SAPO), has been presented as an excellent adsorbent with optimum thermal and hydrothermal stability and metals impregnation, making it selective adsorption. Thus was synthesized SAPO-5 (microporous) and SAPO-5M (mesoporous) impregnated with different amounts of transition metals, nickel and/or zinc. These adsorbents were used for removal of sulfur present in the mixture model (Thiophene / Heptane). For this, characterization techniques as TG, XRD, EDX and BET were performed, as well as a kinetic monitoring to assess the efficiency of adsorption and equilibrium isotherms. It was confirmed mesoporosity of SAPO-5M, since the isotherms showed the same type IV characteristics (typical of mesoporous materials). The volume and pore diameter of the microporous and mesoporous SAPO-5 were not totally blocked after impregnation. For mesoporous adsorbents an increase of four times the pore diameter in relation to the microporous. Although, has shown a reduction in the intensity of X-ray diffraction after the impregnation of the oxides of nickel and/or zinc, gave a crystallographically ordered materials. Complete removal of the template occurred, showing that the calcination was efficient and the material is stable at high temperatures. Furthermore, we were able to attenuate the amount of sulfur 500 ppm to about 96.3 ppm, a reduction of 80% for the mesoporous adsorbent Ni 0.25% 0.75% Zn/SAPO-5M, whose adsorption capacity sulfur was 40 mg S/g of adsorbent. The adsorbent of 1% Zn/SAPO-5 showed higher adsorption capacity (30 mg S/g of adsorbent) among microporous materials, achieving reduced sulfur content from 500 ppm to 163 ppm. For other supports and adsorbent was reduced by approximately 31% of the sulfur content in the mixture model. It is concluded that the synthesis of mesoporous was effective and that the material used is effective in the removal of this contaminant in the mixture model.