Simultaneous Application of Heat and Electron Particles to Effectively

Article pubs.acs.org/EF

Simultaneous Application of Heat and Electron Particles to Effectively Reduce the Viscosity of Heavy Deasphalted Petroleum Fluids Masoud Alfi,*,† Maria A. Barrufet,† Paulo F. Da Silva,‡ and Rosana G. Moreira‡ †

Petroleum Engineering Department, and ‡Biological and Agricultural Engineering Department, Texas A&M University, College Station, Texas, United States ABSTRACT: Application of ionizing incidents, as an innovative way of combining engineering physics and chemistry to efficiently deliver energy to the electronic structure of the molecules, has introduced great opportunities to the developing oil and gas industry. Although heavy petroleum resources have been considered as a potential alternative to fulfill the growing energy inquiries, the lack of cost-effective technologies for extraction, transportation, or refinery upgrading hinders the development of heavy oil reserves. Nevertheless, electron irradiation technology can economically overcome principal problems of heavy oil processing arising from the heavy oil’s unfavorable physical and chemical properties. This technology promises to increase the yields of valuable and environmentally satisfying products in thermal cracking. Electron particles were observed to reduce the viscosity of heavy deasphalted petroleum fluid and to provide a higher concentration of light hydrocarbons in the final product. This behavior is attributed to the intensified cracking in the presence of ionizing electrons. Molecular distribution of the hydrocarbons in liquid and gas products shows that although simultaneous application of heat and electron particles accelerates the cracking process, the reaction mechanism does not differ from that of thermal cracking. The results also demonstrate the substantial influence of reaction temperature and absorbed dose upon the radiolysis throughput. Moreover, aging analyses of the post-treatment samples proves the time-stable nature of the irradiation products.

1. INTRODUCTION Recent studies on unconventional oil and gas reserves have introduced these resources as potential alternatives for conventional hydrocarbon reservoirs. The question is whether the industry can increase the production of unconventional resources to a level that compensates the declined production of depleting conventional reservoirs. Although these resources offer a long-life production and a good upside potential to boost recoveries through new technologies, transformation of such unfamiliar resources into supplies has posed awkward challenges to the industry. Extreme drilling conditions; complicated reservoir rock characteristics; upgrading, processing, and refining capacities; and environmental concerns are part of the current challenges. Among unconventional resources, bitumen and extra-heavy oil reservoirs form a considerable portion of the reserves. Although bitumen and heavy oil are encountered worldwide, Canada and Venezuela contain the largest portions of these reserves.1,2 Difficulties arise while dealing with such a heavy hydrocarbon fluid from the time it is extracted, using heavy oil recovery methods, through the time it is transported to the refinery units and finally when it is upgraded into more utilizable hydrocarbons.3−5 Olsen and Ramzel6 discussed some of these limitations and difficulties, encountered in domestic heavy oil production from the U.S. reservoirs. Although viscosity reduction is known as the minimum objective of the upgrading processes, higher quality of the hydrocarbon molecules can only be achieved through a full upgrading approach. Rana et al.7 have discussed noncatalytic processes (including thermal processes and solvent deasphaltening), catalytic processes, and hydroprocessing methods as different technologies for upgrading of heavy oils © 2013 American Chemical Society

and residua. Even though they are effective in throughput, these processes are not technically efficient, as they require a substantial energy investment and, in some cases, expensive chemical catalysts. Additionally, unavoidable environmental pollution problems, such as sulfur dioxide emission, are the byeffects of such processes.8 As a potential alternative, electron irradiation provides an efficient way of delivering energy to the target molecules. The excited species generate highly reactive free radicals that are capable of initiating chain reactions. Using ionizing electron particles, we can minimize any probable energy loss, which is unavoidable in other energy delivery methods. Zhuravlev et al.9 investigated thermal and radiation thermal cracking of vacuum gas oil fraction of Western Siberian at different temperatures and absorbed doses. The rate of cracking was 1.5−2 times higher when irradiation was applied. Moreover, the amount of gasoline and diesel fraction increased in the irradiation products. Ionizing irradiation has been also used as a way to treat heavy petroleum samples. Zaykin and Zaykina10 compared radiation thermal cracking of petroleum bitumen to the other processing methods such as thermal cracking, thermocatalytic cracking, and ozone thermal cracking. The irradiated samples provided a higher concentration of the light synthetic oil through a well controlled process. From an economical perspective, some authors reported radiation methods to be more favorable compared to catalytic processes.11,12 More specifically, Yang et al.13 showed that Received: May 13, 2013 Revised: July 30, 2013 Published: August 5, 2013 5116

dx.doi.org/10.1021/ef400883z | Energy Fuels 2013, 27, 5116−5127

Energy & Fuels

Article

density is undertaken by the integrated density measuring cell, which works on the proven principle of the oscillation U-tube. The density measurement in SVM 3000 complies with the ASTM D7042 standard. 2.1.2. Gas Chromatography (GC). To analyze the evolved gas during the experiment, we used a refinery gas analyzer (RGA-7890A Agilent). The machine is able to analyze light liquid and gas samples and has two different detectors, working at the temperature of 200 °C. Thermal conductivity detectors (TCD) analyzes permanent gases, and the flame ionization detector (FID) evaluates hydrocarbon molecules. A capillary column of 60 m length and 0.32 mm ID, which contains silica as the stationary phase, was used to separate hydrocarbon molecules in the mixture (the column is well suited for hydrocarbon and sulfur gases). Helium with a flow rate of 1.5 mL/min was employed as the carrier gas. The separation process started at the temperature of 40 °C, and the oven held this temperature for 10 min. Afterward, the temperature increased to 100 °C with a ramp of 3 °C/ min and was kept at 100 °C for 40 min. The RGA machine provides very promising information about the gas samples with a solid resolution. 2.1.3. Simulated Distillation (SIMDIS). ASTM D7169 was used to determine the boiling point distribution of the cut point intervals of crude oil and residues, using high temperature gas chromatography. This test is designed to determine the boiling point distribution of the hydrocarbons up to n-C100 with corresponding elution temperature of 720 °C. When analyzing the samples, the GC oven’s initial temperature is set at −20 °C with the initial hold time of 0 min. The oven is then heated at the rate of 15 °C/min to the final temperature of 425 °C and held for 10 min. The column has a length of 5 m, an inner diameter of 0.53 mm, and a stationary phase thickness of 0.15 μm where a carrier (mobile) phase flow of 25 mL/min passes through it. 2.2. Experimental Setup. 2.2.1. Electron Generation Facility. The oil samples were irradiated with a 1.35 MeV Van de Graaff electrostatic accelerator (High Voltage Engineering Corp., Cambridge, MA) located at the Department of Biological and Agricultural Engineering, Texas A&M University. A Farmer ionization chamber (Marcus Chamber, type 23343) with a volume of 0.055 cm3 was used for the expected dose calibration procedure. A dose distribution mapping was performed at the target area where the sample was supposed to be located at during the irradiation treatment. The chamber exposure (Roentgen-R) was related to the sample’s expected absorbed dose based on its composition. A parallel plate transmission ion chamber, attached to the exit electron beam window, was used to monitor the beam flow during the irradiation time, therefore providing the expected dose at that particular target area for that particular time duration. In order to verify proper beam penetration, a radiochromic film (FWT-60 Series from Far West Technologies, Goleta, CA) was placed inside a decoy aluminum chamber, and its readings were verified with a previously developed calibration curve. The aluminum chamber did not contain any sample on it, and it was not heated during this procedure. Dose distribution was assumed to be uniform due to the sample convection during heating process. Also, it was assumed that at any time all the sample volume was in contact with the irradiation beam due to the same reason (strong convection during heating). 2.2.2. Reaction Chamber and Condenser Design. RTC and TC experiments took place in thin-wall aluminum reaction chambers. Note that the thin wall of the reaction chamber (0.2 mm) reduces energy attenuation of electron particles, delivering more energy to the sample. According to Yang,13 an accelerated electron loses about 4% of its energy while passing through the aluminum can reactor walls. The chamber was connected to a reflux condenser unit that is designed to cool down the condensable evolved gas and turn it into the liquid to be exposed further to ionizing particles (as we can see in Figure 1, the condensed gas returns to the reactor). On the other hand, noncondensable gas was collected in a specially designed gas sampling bag for further analysis. To provide the required heat for thermal and radiation thermal cracking, a temperature control hot plate was used. In addition to that, a custom-made copper base, which is designed based on the shape of the aluminum chamber, was used to heat the

electron beam irradiation, when performed in an industrial scale, could decrease the upgrading cost by 31%. They also suggested a practical design to use electron irradiation in dynamic flow processing units to upgrade heavy fluids in pipelines.14 In another work, Yang et al.15 coupled heat transfer and radiation Monte Carlo simulation to model electron beam upgrading process of multiphase and single-phase hydrocarbons. This work is known as a pioneer in this field as no previous effort has been done to simulate electron beam processing of petroleum. According to Skripchenko et al.16 radiation is capable of promoting destructive reactions in heavy petroleum fluids, coal samples, and their mixtures. To better understand the radiolytic behavior of hydrocarbons and collect practical information about the mechanism and kinetics of the reactions, simple hydrocarbon molecules were studied as models to simulate complicated situations. The experiments demonstrated the vital role of C−C bond rupture in the formation of saturated and unsaturated species.17−19 Although application of ionizing particles in petroleum industry has had a rapid growth during the recent decades, quite a few works have investigated the rheological property changes that are brought about because of radiation-induced reactions. Alfi et al.20−22 showed that electron-induced thermal cracking of heavy petroleum fluids results in products with lower viscosities than the thermal cracking cases. In this study, we have analyzed radiation thermal cracking (RTC) and thermal cracking (TC) of heavy deasphalted samples. In fact, this research is meant to see if electron irradiation of heavy petroleum fluids can be an effective means of cracking larger, and less expensive hydrocarbon molecules into smaller, and probably more expensive species. This way, transportation of those fluids becomes easier as well. The physical and chemical properties of the fluids were analyzed before and after the treatment. The influence of the reaction temperature and deposited radiation energy upon the radiation thermal cracking throughput was also investigated to find the optimum operating condition. The effect of ionizing radiation is always evaluated as a function of the amount of absorbed energy per unit mass of the material which is measured in joules (J) per kilogram (kg) or gray (Gy).

2. EXPERIMENTAL METHODOLOGY AND MATERIALS This section describes the electron irradiation facility, reaction chamber design, petroleum samples characteristics, and analytical tools to monitor the changes in physical and chemical properties of heavy petroleum fluids. 2.1. Analytical Tools. Radiation-induced physical and rheological changes were analyzed using viscometer and densitometer instruments. To control the radiolytic reactions in favor of a better process throughput, we should develop an in-depth knowledge of the radiolysis reaction mechanism, chemical changes, and dominating factors. The gas chromatography instruments used in this research provided us comprehensive information about the chemical distribution of the products after different treatment scenarios. The following sections discuss the analytical methods used to evaluate radiation products. 2.1.1. Viscosity and Density. The viscosity of the liquid products was measured using Brookfield HBDV-IIIUCP and LVDV-IIIUCP viscometers. These machines enable us to measure viscosities in the range of 15 cP to 8 × 106 cP. All the samples showed Newtonian fluid behavior before and after irradiation, as determined from the linear behavior of shear stress versus shear rate, and the viscosity of the samples was calculated using linear regression. The density of the samples was measured using the Anton Paar SVM 3000 machine, where the required determination of the sample 5117


Energy & Fuels

Article

the beam exit window, and Figure 2 provides a schematic diagram of the experimental and analytical setup). Note that to fully confirm the integrity of the experiments, we have performed mass balance analysis for each experiment considering the weight of the liquid sample before the experiment and that of the remaining liquid and evolved gas after irradiation. 2.3. Petroleum Fluid Characteristics. In this study, we have used deasphalted oil (DAO) as the heavy petroleum sample. The DAO fluid is a stream of highly asphaltic atmospheric residuum that is produced through a pilot scale solvent deasphaltening unit. Note that the samples contain less than 0.06 wt % asphaltene, so this work is not intended to study any possible interference of heavy asphaltic structures with the radiation process and that will be investigated in a separate work. Simulated distillation analysis shows that 10% of the liquid boils at temperatures above 720 °C, which is an indication of the heavy nature of the samples (Figure 3a). This temperature corresponds to the boiling point of C100 (this value is just an indicator of this group, and the true boiling point of C100 isomers depends on their structure). Figure 3b provides the viscosity of the untreated fluid at different temperatures. Very high viscosity of the sample (11000 cP at 20 °C) reflects its heavy nature. More information about the sample characteristics can be found in Table 1.

3. RESULTS AND DISCUSSION This section discusses the results of different treatment scenarios on the DAO fluid. Viscosity and density measurements, SIMDIS of the liquid products, light liquid fraction analysis, and evolved gas analysis were used to evaluate the changes caused by electron particles. We have used the provided information to investigate the mechanism of hightemperature radiation-induced cracking of complex hydrocarbon molecules. Moreover, time stability of the irradiated

Figure 1. Reaction chamber and the condenser along with the heat source and thermocouple assembly in front of the beam window. chamber uniformly. Three thermocouples were used at different places to measure the temperature of the evolved gas, hydrocarbon liquid, and heater (Figure 1 provides a picture of the reaction setup in front of

Figure 2. A schematic picture of the reaction setup and the related analytical tools. 5118


Energy & Fuels

Article

Figure 3. (a) Simulated distillation of the untreated deasphalted oil demonstrates the heavy nature of the fluid (initial boiling point is 330 °C). (b) Viscosity of the untreated deasphalted oil measured at different temperatures.

petroleum fluids, we should make sure that both TC and RTC experiments will be performed at identical temperatures. This way, we can claim that the differences in the results can be solely attributed to the presence of high energy electron particles and the reactions initiated by those species. This was achieved with the assistance of the temperature control setup and thermocouple assembly we used during the experiment. The duration of experiments was 2 h, and the liquid temperature was kept at 385 °C throughout the run time for both thermal and radiation thermal cracking experiments (for the TC case, we had just heating, but for the RTC case heating and irradiation took place simultaneously and an amount of 20 KGy energy was uniformly deposited into the sample). It should be mentioned that the amount of liquid sample in the reactor was 30 g, which means that total additional energy, deposited in the RTC sample as a result of irradiation, is 0.6 KJ (this is approximately 2% of the thermal energy required to raise temperature of our fluid to 385 °C). Repeatability of the experiments was examined with three replications for each experiment set, and the results are denoted by error bars.

Table 1. Characteristics of the DAO Sample parameter °API (60 °F/60 °F) ash, wt % heptane insolubles, wt % pentane insolubles, wt % pour point, °F CHNS (wt %) carbon (wt %) hydrogen (wt %) nitrogen (wt %) sulfur (wt %)

DAO fluid 14.5 0.002 0.06 0.16 45 99.75 85.55 11.09 0.25 2.86

samples, as well as the effect of reaction temperature and absorbed dose on radiation throughput, is examined. 3.1. Irradiation Experiments. To investigate the effect of electron particles on heavy petroleum fluids, we conducted two types of experiments: thermal cracking (TC) and radiation thermal cracking (RTC). It should be noted that to accurately analyze the effect of ionizing electron particles on heavy 5119


Energy & Fuels

Article

Figure 4. Viscosity of the RTC and TC samples (measured at 20 and 30 °C) shows that further viscosity reduction is achieved when DAO samples are exposed to electron irradiation.

Figure 5. The difference in boiling point distribution of RTC and TC samples with that of the untreated DAO shows an intensified cracking for DAO samples as a result of electron irradiation.

3.1.1. Viscosity and Density. Figure 4 provides the viscosity of the DAO fluid, treated in different ways, measured at two temperatures. Assuming that heavy oil transportation in the pipeline occurs at temperatures around 25 °C, we have selected 20 and 30 °C as our viscosity measurement temperatures. As it is apparent from the graph, irradiation has substantially decreased the viscosity of the fluids. A 55% viscosity reduction (viscosity decreased from 2750 to 1200 cP) in irradiated samples is evidence for intensified cracking as a consequence of ionizing irradiation. The effectiveness of irradiation, as an efficient means of delivering energy to the electronic structure of the molecules, becomes more pronounced when we notice that thermal energy is coupled tightly to translational, rotational, and vibrational modes, and only a small portion of the energy goes into the electronic structure of the absorber. In fact, irradiation will impact initiation as one of the most energyintensive steps in chain reactions. This intensified cracking results in lighter molecules in the final product and causes the viscosity of the irradiated DAO fluid to decrease considerably. The following sections provide more details about the similarities and differences of thermal and radiation-induced

cracking. Although different in viscosity, RTC and TC products have similar API gravities (14.70 and 14.44 for RTC and TC, respectively). 3.1.2. Boiling Point Distribution. Figure 5 provides detailed information about the boiling point distribution of hydrocarbon components in heavy oil samples (the horizontal axis represents the boiling temperature, and the vertical axis specifies weight percent of the components with that boiling point). For the boiling point range of 36−460 °C, we have considered intervals of 10 °C to group components together, and intervals of 20 °C were used for temperatures above 460 °C. To better see the changes that are brought about to our DAO samples as a result of different treatment scenarios (radiation thermal cracking and thermal cracking) and also to compare the outcome of different scenarios with each other, we have subtracted boiling point distribution of the original untreated DAO from that of RTC and TC fluids (Figure 5). The positive value of Δwt % lines at temperatures below 430 °C means that we have a higher concentration of components with (Tb < 430 °C) in RTC and TC than the untreated DAO. The negative value of Δwt % lines for T > 430 °C is, however, analyzed as a higher concentration 5120


Energy & Fuels

Article

Figure 6. RTC and TC gas products have similar components distribution.

take place when either thermal or radiation thermal cracking is employed as a means of reducing viscosity of heavy petroleum fluids. However, looking more precisely, we can see that there is a traceable amount of hydrogen in RTC gas. The amount of evolved hydrogen in TC experiments was lower than the GC instrument’s threshold. Radiation-induced excited molecules can be the potential source of H2 generation. The evolved hydrogen has two origins: it may be formed through a molecular elimination or a hydrogen atom mechanism.23−26 For the case of molecular elimination, the excited hydrocarbon (A*) loses a hydrogen molecule, resulting in an unsaturated molecule (Aunsaturated). This process is also called unimolecular hydrogen formation.

of components in this boiling point range for untreated DAO fluids compared to RTC or TC. Keeping in mind that 430 °C corresponds to the boiling point of C28, we can claim that the + net effect of both treatment scenarios is to crack C 28 components into lighter species. However, this is not the most important conclusion that can be drawn from the provided graph. The upward shift of RTC line for Tb < 430 °C, compared to that of the TC one, means that we have a higher concentration of lighter components in the irradiated sample, which reduces the viscosity from 2750 cP for TC to 1200 cP for RTC (at 20 °C). Higher concentration of light molecules in RTC samples is because irradiation reinforces the cracking process. Considering the trend lines more precisely, it is apparent that both treatments follow a similar pattern, an increase starting at 150 °C that reaches a maximum at 320 °C followed by a decrease to 650 °C. Additionally, the relative ratio of the concentration of different hydrocarbon molecules in the TC sample is the same as that of the RTC sample. It can be concluded from the discussion that although irradiation improves the cracking process, it does not change the reaction mechanism in favor of molecules with a specific boiling point or molecular weight. 3.1.3. Evolved Gas Analysis. Investigation of the gas products helps us to better understand the reaction mechanism. In this study, gas samples were analyzed from two perspectives: quantitative and qualitative. A quantitative point of view gives us information about the amount of gas evolved during the experiments (as mentioned in the previous section, evolved gases from TC and RTC were collected in specifically designed gas sampling bags). The results showed that RTC generates 2− 3 times more gaseous products compared to TC, as manifested by the volume of the evolved gas. Gas molecules are, in fact, the product of the upgrading process, when heavy complex molecules break into smaller compounds; hence, more amounts of liberated gas in RTC are analyzed as reinforced cracking, which is a result of electron irradiation. On the other hand, the qualitative analysis discusses the chemical composition of the evolved gas to inspect the differences and probable enhancements. Figure 6 shows the distribution of different molecules in the gas products (note that nitrogen is excluded from the analysis). According to the graph, both gas samples have similar composition; this leads to the conclusion that similar reactions

A* → A unsaturated + H 2

(1)

On the other hand, let us consider a model for the production of hydrogen in which a C−H bond is initially broken in an excited molecule or ion to produce a hydrocarbon radical and a hydrogen atom. The hydrogen atoms possess a range of kinetic energies, but most are sufficiently excited to abstract another hydrogen atom on their first collision (defined as hot hydrogen atoms). If the abstraction takes place from another molecule, the process is called bimolecular hydrogen formation. If the abstraction occurs from the same carbon atom or an adjacent one on the same molecule, from which the hot hydrogen atom has been released, the process will be indistinguishable from the molecular elimination (both processes are unimolecular). Hydrogen atoms, that do not react on their first collision, are defined as epithermal or thermal, depending on their kinetic energy. With a few exceptions, epithermal and thermal hydrogens will react similar to each other. Compared to the hot atoms, the contribution of thermal atoms to the hydrogen yield is not that significant. A* → Ȧ + Ḣ

(2)

Ḣ + A → H 2 + Ȧ

(3)

Here, A is a hydrocarbon molecule, and A* represents the excited state of that molecule. 3.1.4. Light Liquid Fraction Analysis. This section provides detailed information about the composition of the liquid products with boiling points less than 250 °C (called light liquid 5121


Energy & Fuels

Article

Figure 7. Distribution of different hydrocarbon groups in the light liquid components (Tb < 250 °C) shows similar composition for RTC and TC.

Figure 8. Carbon number distribution of the radiolyzed fluid looks similar to that of the thermally cracked fluid.

f raction), which can be used as a fingerprint for investigation of RTC and TC treatment mechanisms. According to Figure 7, the distribution of different hydrocarbon species in both treatment scenarios is quite similar to each other. Aromatic molecules are the most abundant group and form 20 wt % of the light liquid components. Mono-aromatics, i-paraffins, and nolefins stand after aromatics with wt % of 14, 13, and 10, respectively, for both TC and RTC products. This similarity is also observed for less abundant species (di-olefins, naphthenoolefins, indanes, indenes, and naphthalenes). Figure 8 represents the carbon number distribution of the light liquid products. According to the graph, irradiated and unirradiated samples share similar patterns. C8−C12 have the highest concentration, while the concentration of C13 decreases steeply. Moreover, we have analyzed mass distribution of different hydrocarbon groups in the light liquid products to gain a better idea of probable changes that may happen to a specific group of components as a result of irradiation (Figure 9). Except for the paraffins (the composition of light and heavy hydrocarbons differs slightly in RTC and TC cases) and n-olefins (C4 concentration is higher in TC), the other groups (isoparaffins, monoaromatics, mononaphthenes, iso-olefins, and naphthenoolefins) exhibit a similar composition distribution. The analyses

performed on the light liquid products interestingly confirm the results of SIMDIS and gas analyses, supporting the claimed theory about the role of irradiation on chain reactions and the fact that the reaction path will not be altered by radiolytic methods. 3.2. Effective Experimental Factors. This section discusses the effective experimental conditions in radiolytic reactions. Although there are a number of important factors that may influence the radiolysis process (i.e., reaction temperature, absorbed dose, dose rate, energy level of individual electron particles, heat losses, etc.), based upon our experimental capabilities and the factors we have control over, reaction temperature and irradiated dose were selected as two of the dominant ones. 3.2.1. Reaction Temperature. Finding out the optimum reaction temperature, we are able to lower the operating cost of radiation-induced cracking while not obstructing the radiolysis process. We performed RTC and TC experiments on the DAO sample at four different liquid temperatures (this is the temperature of the liquid inside the reactor). The liquid temperature started at 230 °C to simulate low-temperature radiation thermal cracking. Medium-temperature cracking experiments were performed at 270 and 320 °C, while the high-temperature test was done at the liquid temperature of 5122


Energy & Fuels

Article

Figure 9. (a−g) Similar carbon number distribution of different hydrocarbon groups in the light liquid product (Tb < 250 °C) is considered as evidence for the presented theory about the reaction mechanism in thermal and radiation thermal cracking.

380 °C. Like the previous tests, the liquid temperature was the

took place simultaneously and the amount of 20 KGy energy

same for both TC and RTC experiments, and the only difference between the runs referred to the presence of

was absorbed by the fluids. Table 2 provides the viscosity of the RTC and TC products

irradiation in RTC cases. In the TC case, the fluid was heated

(measured at 20 and 30 °C) at different reaction temperatures.

for 2 h, while in the RTC experiment heating and irradiation

It is apparent that at lower reaction temperatures, irradiation 5123


Energy & Fuels

Article

Table 2. Viscosity Values of the DAO Fluid at Different Reaction Temperatures μ at 20 °C (cP) temperature (°C)

RTC

230 270 320 380

12891 12606 11070 1190

± ± ± ±

μ at 30 °C (cP) TC

460 450 350 120

12584 12425 11362 3001

± ± ± ±

RTC 500 450 330 150

4062 3947 3548 502

± ± ± ±

340 340 270 80

TC 3961 3904 3617 1133

± ± ± ±

350 345 260 86

Figure 10. Low temperature RTC and TC increase the viscosity, while at higher temperatures viscosity reduction is observed for both the cases. The solid black line represents the viscosity of the original untreated DAO (all viscosities measured at 20 °C).

lower reaction temperatures are close to each other, within the range of the error bars. In fact, higher values of absorbed dose with different dose rates might be required to see more pronounced differences in RTC and TC products of low temperature reactions. On the other hand, irradiation improves the upgrading process at higher temperatures because of the more efficient energy delivery process. The following discussion explains the theory behind the observed behavior. Electron particles give out their energy to the surrounding molecules while passing through a media, causing ionization or excitation to the target molecules. Depending on the situation, electrons may have elastic or inelastic scattering. Inelastic scattering results in energy transfer to the molecules, producing excited molecules and ions, secondary and Auger electrons, photons, and X-ray, while elastic scattering causes deflection in the electron track without any energy loss. Appearance of electric charges is one of the most obvious consequences of exposing materials to ionizing irradiation. Ionizing incidents result in abstraction of electrons from the molecules and creation of positive ions in a so-called “ionization” process. The electrons abstracted from the irradiated molecules will be strongly pulled by ions of positive charge, resulting in “charge recombination”. The recovered ionization potential generates highly excited molecules with energy levels much higher than bond strength. The remaining energy that is deposited by ionizing irradiation causes “excitation” for the molecules exposed to irradiation (eqs 5 and 6).30−32 Ionization:

does not cause any improvement in the viscosity of heavy petroleum fluids. To analyze the results and see the differences, we plotted the viscosity values of Table 2 (measured at 20 °C) in a graph (Figure 10, viscosities vs reaction temperature). At relatively low reaction temperatures (T < 320 °C), both treatments increase the viscosity of the samples by 10%. This can be attributed to the formation of heavier components as a result of high-energy species recombination (high-energy species stabilize through establishing chemical bonds and formation of larger molecules). Note that, as indicated in Figure 3a, vaporization of light components is ruled out as a factor that may cause viscosity to increase due to the very heavy nature of the DAO fluid. Similar reaction temperature dependency has been also reported by other authors.27−29 Somewhere between 320 to 380 °C, cracking reactions become activated, causing the viscosity to decrease in both experiments. However, this reduction is more pronounced for the case of irradiated samples. To better evaluate the impact of temperature on hydrocarbon irradiation, we have defined relative viscosity reduction (RVR) as the relative reduction in the fluid viscosity, achieved as a result of radiation-induced reactions (eq 4). RVR =

μTC − μRTC μTC

× 100 (4)

Figure 11 shows the values of RVR at different reaction temperatures. The negative values at lower temperatures mean that the viscosity of the RTC product is higher than that of TC and can be interpreted as reinforced free radical or charge recombination in irradiated samples. However, it should be noted that the viscosity values for RTC and TC samples at

AB ⇝ AB+ + e ̇

(5)

Excitation: 5124


Energy & Fuels

Article

Figure 11. Radiation-induced viscosity reduction becomes more evident at 380 °C, which means that cracking reactions become activated after a certain threshold energy is provided.

AB ⇝ AB*

(6)

amount of heat that is required to produce the same concentration of active radicals, one can conclude that the consumed energy in the form of heat is excessively higher. The feedstock should be heated to high temperatures in order to deliver a relatively small amount of energy into the molecules. Therefore, consuming almost the same amount of energy, the concentration of active radicals, generated in the chain initiation step, is much higher in RTC than in TC. •Propagation (chain process)

These primary reactions result in development of secondary reactions where the ions, secondary electrons, and excited species exchange energy and charge with nearby neighbors, resulting in generation of short-living intermediate components which may finally evolve into new stable products (Figure 12).

hydrogen exchange : Ci̇ + Cn → CiH + Cṅ

(8)

β scission: Cṅ → Ċk + 1‐alkene

(9)

where k can be any value in the range 1 ≤ k < n. radical addition: Cṅ + 1‐alkene → Cṅ +

Figure 12. Schematic representation of primary and secondary radiolysis events (regenerated after Cleland32).

(10)

The free radicals generated in the initiation step (with either thermal or radiation origins) have the ability to propagate through a series of chain development reactions. Each active radical can serve as the starting point for hundreds or thousands of consecutive reactions. In other words, the ultimate product of a series of chain reactions is substantially dominated by the products of the chain propagation step. The major reactions in the propagation step of hydrocarbon cracking chain reactions exhibit an endothermic nature. It means that they require energy to be activated and develop further in favor of generating lighter species. •Termination

Having all the prerequisites available, generated free hydrocarbon radicals may start a series of chain reactions causing hydrocarbon molecules to upgrade into lighter species.18,33 Chain reactions are comprised of three main stages known as chain initiation, propagation (reaction process), and termination: •Initiation Cn → Ci̇ + Cj̇ (7) Any reaction proceeding by a free radical mechanism must include some radical-producing reactions, which are generally referred to as initiation reactions, and these initiator derived radicals (species resulting from hydrocarbon radiolysis along with the ones generated by thermal hydrocarbon cracking) react with the reactants, producing reactant-derived radicals. Considering the chain reaction of hydrocarbons, the energy consumed for the cracking initiation in the form of heat or ionizing irradiation will not change the product enthalpy. In fact, this initiation energy is consumed to create a generous concentration of active radicals necessary for chain initiation. Suppose the initial concentration of radicals to be a specific value. Using electron irradiation as a way to generate that concentration of active radicals, the energy is directly transferred to the molecules. Comparing this energy to the

(11) Ẋ + Ẏ → XY Here, Ẋ and Ẏ can be any of the radical species formed during the initiation or propagation steps. Now, let us consider four different scenarios, low and high temperature thermal and radiation thermal cracking. When working at lower temperatures, as the required activation energy for chain development is not supplied, the chain reactions stay inert. In fact, the free radicals, generated either through thermal or radiolytic processes, will not take part in cracking reactions. However, high energy radicals can stabilize through the formation of heavier species in recombination reactions. The higher the concentration of the reactive free radical, the more probable the formation of high energy 5125


Energy & Fuels

Article

This section is aimed at investigating the effect of absorbed dose on radiation throughput. To do so, we have irradiated DAO samples at two different absorbed doses (10 and 20 kGy, both at the reaction temperature of 385 °C). As the electron generation machine operates at a constant energy rate, the duration of experiments varies depending on the absorbed energy values (1 h for 10 kGy and 2 h for 20 kGy) and the amount of thermal energy delivered to each sample differs for the two irradiation tests. Consequently, we are not able to compare directly the RTC products of different absorbed doses. Alternatively, to be able to evaluate the effect of irradiation dose on radiolysis throughput, we have used relative viscosity reduction (RVR) as a way to evaluate the viscosity reduction achieved by ionizing irradiation (eq 4). The results of different irradiation doses are shown in Table 3. Looking at Table 3, it is apparent that increasing irradiation dose from 10 kGy to 20 kGy will substantially improve the viscosity reduction. However, to fully characterize the influence of absorbed dose upon the radiolytic behavior of heavy petroleum samples, a broader range of absorbed energy values should be investigated (it requires a more capable electron accelerator). Viscosity reduction may be intensified (linearly or nonlinearly) with the absorbed dose, or, as reported by some other authors,34 the effect of irradiated dose may come to a saturation after a specific amount of absorbed energy, which means that no improvement in physical and chemical properties of the reactant will be achieved afterward and higher doses will just increase the operating costs. 3.3. Aging Effect. In this study, we have offered electron irradiation as an effective way to reduce the viscosity of heavy hydrocarbons. To come up with an affordable solution for the viscous fluid transportation problems, time stability of the products should be also taken under consideration. Thus, a successful upgrading procedure has two different steps: the first step refers to the moment of upgrading and the throughput of the process; the second step, on the other hand, corresponds to the time after upgrading and probable changes. Here, we have plotted the viscosity of treated fluids with time (Figure 13). Viscosity values were measured at different time intervals until after 120 days from the experiments. The samples were kept at the same environmental conditions. According to the graph, viscosity of the TC and RTC samples does not change with time which means no further

charged or free radical species recombination products would be. Consequently, higher degrees of recombination can be expected for the irradiated fluids as electron irradiation intensifies the formation of reactive free radicals (this pronounced recombination can be reflected in the higher viscosity of RTC products at lower reaction temperatures). Note that, as the chain reactions, which could be quite dominating to significantly influence the final products of the process, are not activated yet, active radicals are not able to carry out consecutive chain reactions, and their capability to develop recombination reactions becomes quite limited. Hence, more radiation initiated active radicals is required (this means more deposited energy) to see a pronounced difference in RTC and TC products as a result of the recombination of radiationinitiated free radicals (for T < 320 °C, RTC and TC lines are close to each other, Figure 10). On the other hand, when the Table 3. Higher Absorbed Dose Provides a More Intensified Cracking irradiated dose, kGy

RVR, %

10 20

30 56

reaction temperature goes above a threshold, it provokes endothermic reactions, activating the chain propagation step. Eventually, hydrocarbon cracking will happen because of a chain process. As mentioned before, chain propagation plays an important role in composition distribution of the final product and is fed by the reactive free radicals created during the initiation step. Higher concentration of free radicals in the RTC case provides the essential requirements for an intensified cracking process when compared to the TC case. 3.2.2. Absorbed Dose. Irradiation dose is one of the most important factors affecting the results of radiation thermal cracking. In fact, its importance should be viewed from two different standpoints. First of all, knowing the relationship between irradiation dose and experiment throughput, we are able to determine the best operating conditions with the highest output. On the other hand, it is obvious that delivering more energyin the form of irradiated dosecauses higher operation costs. Hence, there should be a balance between the process outcome and the energy expense for that outcome.

Figure 13. Aging analysis of the RTC samples shows stability with time for irradiated samples (viscosities measured at 20 °C). 5126


Energy & Fuels

Article

reaction with time. For the radiolyzed fluids, it implies the promising nature of irradiation techniques, when coupled with conventional thermal cracking methods. Any probable viscosity increment could be an indication of recombination reactions, caused by the active post-treatment species. When recombination happens, small molecules join together and form larger molecules with higher molecular weight and different characteristics. This gradual change in molecular weight would be reflected in rheological properties of the fluid as well.

(7) Rana, M. S.; Samano, V.; Ancheyta, J.; Diaz, J. Fuel 2007, 86, 1216−1231. (8) Gray, M. Upgrading Petroleum Residues and Heavy Oils; Marcell Dekker: New York, 1986. (9) Zhuravlev, G. I.; Voznesenskaya, S. V.; Borisenko, I. V.; Bilap, L. A. High Energy Chem. 1991, 25, 20−23. (10) Zaykin, Y. A.; Zaykina, R. F. Radiat. Phys. Chem. 2004, 71, 471− 474. (11) Mustafaev, I.; Gulieva, N. Radiat. Phys. Chem. 1995, 46, 1313− 1316. (12) Mirkin, G.; Zaykina, R. F.; Zaykin, Y. A. Radiat. Phys. Chem. 2003, 67, 311−314. (13) Yang, D. Heavy Oil Upgrading from Electron Beam (E-Beam) Irradiation. M.Sc. thesis, Texas A&M University, 2009. (14) Yang, D.; Kim, J.; Silva, P.; Barrufet, M. A.; Moreira, R.; Sosa, J. Laboratory Investigation of E-Beam Heavy Oil Upgrading. Proceedings of Latin American and Caribbean Petroleum Engineering Conference; Cartagena de Indias, Colombia, May 31−June 3, 2009. (15) Yang, D.; Kim, J.; Silva, P.; Barrufet, M.; Moreira, R. Electron Beam (E−Beam) Irradiation Improves Conventional Heavy-Oil Upgrading. Proceedings of SPE Annual Technical Conference and Exhibition; Florence, Italy, 19−22 September 2010. (16) Skripchenko, G. B.; Sekrieru, V. I.; Larina, N. K.; Smutkina, Z. S.; Miesserova, O. K.; Rudoi, V. A. Solid Fuel Chem. 1986, 20, 54−58. (17) Wu, G.; Katsumura, Y.; Matsuura, C.; Ishigure, K.; Kubo, J. Ind. Eng. Chem. Res. 1997, 36, 1973−1978. (18) Wu, G.; Katsumura, Y.; Matsuura, C.; Ishigure, K.; Kubo, J. Ind. Eng. Chem. Res. 1997, 36, 3498−3504. (19) Salovey, R.; Falconer, W. E. J. Phys. Chem. 1965, 69, 2345− 2350. (20) Alfi, M.; Da Silva, P.; Barrufet, M.; Moreira, R. Utilization of Charged Particles as an Efficient Way to Improve Rheological Properties of Heavy Asphaltic Petroleum Fluids. Proceedings of SPE Latin America and the Caribbean Petroleum Engineering Conference; Mexico City, Mexico, April 16−18, 2012. (21) Alfi, M.; Da Silva, P.; Barrufet, M.; Moreira, R. Electron-Induced Chain Reactions of Heavy Petroleum Fluids−Dominant Process Variables. Proceedings of SPE Heavy Oil Conference; Calgary, Canada, 12−14 June 2012. (22) Alfi, M. Ionizing Electron Incidents as an Efficient Way to Reduce Viscosity of Heavy Petroleum Fluids. M.Sc. thesis, Texas A&M University, 2012. (23) Falconer, W. E.; Salovey, R. J. Chem. Phys. 1966, 44. (24) Gaumann, T.; Rappoport, S.; Ruf, A. J. Phys. Chem. 1972, 76, 3851−3855. (25) Foldiak, G.; Wojnarovits, L. Acta Chim. Acad. Sci. Hung. 1971, 70, 23−39. (26) Foldiak, G.; Wojnarovits, L. Acta Chim. Acad. Sci. Hung. 1971, 67, 209−219. (27) Mustafaev, I.; Gulieva, N. Radiat. Phys. Chem. 1995, 46, 1313− 1316. (28) Brodskii, A.; Zvonov, N.; Lavrovskii, K.; Titov, V. Neftekhimiya (U.S.S.R.) 1961, 1, 370−381. (29) Lucchesi, P. J.; Baeder, D. L.; Longwell, J. P. Radiation promoted hydrocarbon reactions. Proceedings of the Fifth World Petroleum Congress, June 1959. (30) Chapiro, A. Encyclopedia of Materials: Science and Technology; Elsevier: Oxford, 2004; pp 1−8. (31) Tolbert, B. M.; Lemmon, R. M. Radiat. Res. 1955, 3, 52−67. (32) Cleland, M. R. Ind. Irradiat. Technol. 1983, 1, 191−218. (33) Hoare, M. F.; Garbett, T. A.; Pegg, R. E. The Effect of Gamma Radiation on the Thermal Decomposition of Light Paraffinic Hydrocarbons. Proceedings of 5th World Petroleum Congress; New York, NY, USA, May 30−July 5, 1959. (34) Zaykina, R. F.; Zaykin, Y. A.; Mirkin, G.; Nadirov, N. K. Radiat. Phys. Chem. 2002, 63, 617−620.

4. CONCLUSIONS The effect of accelerated electron particles on physical and chemical properties of heavy deasphalted petroleum fluids is investigated in this study. The molecular distribution and viscosity of the products with different treatment scenarios were used as indices to evaluate the success of the upgrading process. The gas and liquid product analysis shows that radiation thermal cracking will further decrease the viscosity of the heavy oil samples in comparison to conventional thermal cracking. As the amount of absorbed dose increases, the contribution of the radiolytic reactions to the viscosity reduction process becomes more pronounced. The results also show that hydrocarbon cracking chain reactions (initiated either through a radiation or thermal process) are activated only at temperatures higher than a threshold temperatures. However, those reactions become inactive when the reaction temperature falls below that threshold, and electron irradiation can even be detrimental at these conditions. Even though the lighter nature of the irradiated samples is a clear indication of intensified cracking, as a consequence of electron irradiation, the results show that thermal and radiation thermal cracking processes follow the same chain reaction mechanism. In fact, irradiation facilitates the generation of reactive free radicals, as the chain reaction initiators, while the rest of the process is the same for both treatment scenarios.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: masoud.alfi@pe.tamu.edu. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS The authors thank the Petroleum Engineering Department, Texas A&M University, for support of this project. REFERENCES

(1) Stark, P.; Chew, K.; Fryklund, B. The Role of Unconventional Hydrocarbon Resources in Shaping the Energy Future. Proceedings of International Petroleum Technology Conference,;Dubai, UAE, December 4−6, 2007. (2) Pauchon, C. Oil Gas Sci. Technol. 2004, 453−454. (3) Koppel, P. E.; Mazurek, W. L.; Harji, A. Project Scenarios for Bitumen Upgrading. Proceedings of SPE International Thermal Operations and Heavy Oil Symposium and International Horizontal Well Technology Conference; Calgary, Canada, November 4−7, 2002. (4) Ardali, M.; Barrufet, M.; Mamora, D. A Critical Review of Hybrid Steam/Solvent Processes for the Recovery of Heavy Oil and Bitumen. Proceedings of SPE Annual Technical Conference and Exhibition; San Antonio, Texas, USA, October 8−10, 2012. (5) Rahnema, H.; Barrufet, M.; Mamora, D. J. Pet. Sci. Eng. 2013, 103, 85−96. (6) Olsen, D.; Ramzel, E. B. Fuel 1992, 71, 1391−1401. 5127


Simultaneous Application of Heat and Electron Particles to Effectively

Recommend Documents