An induction reactor for studying crude-oil oxidation relevant to in situ combustion

Mohammad Bazargan, Alexandre Lapene, Bo Chen, Louis M. Castanier, and Anthony R. Kovscek Citation: Review of Scientific Instruments 84, 075115 (2013); doi: 10.1063/1.4815827 View online: http://dx.doi.org/10.1063/1.4815827 View Table of Contents: http://scitation.aip.org/content/aip/journal/rsi/84/7?ver=pdfcov Published by the AIP Publishing

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 177.20.130.9 On: Tue, 10 Dec 2013 18:34:33

REVIEW OF SCIENTIFIC INSTRUMENTS 84, 075115 (2013)

An induction reactor for studying crude-oil oxidation relevant to in situ combustion

Mohammad Bazargan, Alexandre Lapene,a) Bo Chen, Louis M. Castanier, and Anthony R. Kovscekb)
Energy Resources Engineering Department, Stanford University, Stanford, California 94305-4007, USA

(Received 21 February 2013; accepted 1 July 2013; published online 24 July 2013) In a conventional ramped temperature oxidation kinetics cell experiment, an electrical furnace is used to ramp temperature at a prescribed rate. Thus, the heating rate of a kinetics cell experiment is limited by furnace performance to heating rates of about 0.53 C/min. A new reactor has been designed to overcome this limit. It uses an induction heating method to ramp temperature. Induction heating is fast and easily controlled. The new reactor covers heating rates from 1 to 30 C/min. This is the rst time that the oxidation proles of a crude oil are available over such a wide range of heating rate. The results from an induction reactor and a conventional kinetics cell at roughly 2 C/min are compared to illustrate consistency between the two reactors. The results at low heating rate are the same as the conventional kinetics cell. As presented in the paper, the new reactor couples well with the isoconversional method for interpretation of reaction kinetics. 2013 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4815827]
I. INTRODUCTION

In situ combustion (ISC) is a thermal enhanced oil recovery (EOR) method that uses air injection to generate heat inside the reservoir. It has been successfully applied in numerous oil elds.14 Ignition is usually initiated using gas burners or electrical heaters.5 The air burns a small portion of the oil but gives, in return, signicant amounts of heat and produces combustion gases. Heat signicantly reduces the viscosity of the heavy oil and combustion gases push the oil toward the producer thereby increasing recovery.6 In situ combustion is implemented at lab scale in an experiment known as a combustion tube.7 The combustion tube experiment is a one-dimensional process. This experiment is also valuable for screening eld candidates. Even though it provides useful information, however, it is difcult to distinguish and quantify various phenomena such as transport and chemical reactions. Dedicated experiments are required especially for kinetics study.8 Various techniques have been applied to study the kinetics of combustion. Tadema9 was the rst to use a differential thermal analysis (DTA) method to study the combustion kinetics of the crude oil. In this experiment, the temperature differences between an oil/sand sample and a blank sand sample are measured while they are heated under the same conditions of heating rate and air ow. Thermo-gravimetric analysis (TGA) is another experiment to measure oxidation kinetics. It uses the amount and rate of the change in the weight of the oil/sand sample as a function of the temperature. The advantage of TGA is that it requires less time. But TGA uses a small sample size and it is sensitive to mass transfer effects.10

Bae11 used both DTA and TGA to analyze the combustion behavior of different crude oil samples. The derivative of thermo-gravimetric curves (DTG) is used by Vossoughi and Shoubary12 together with TGA to investigate the effect of different parameters such as specic surface area and oxygen partial pressure on coke combustion. Differential scanning calorimetry (DSC) is yet another technique to study the kinetics of combustion. DSC monitors differential heat ow of the sample as a function of temperature. Kok13 used DSC together with TG/DTG to characterize the combustion behavior of two crude oil samples. Yannimaras et al.14 introduced accelerating rate calorimetry (ARC) to probe oxidation of crude oils. The ARC method consists of a small spherical sample holder subjected to accurate temperature and heat supply control. Yannimaras and Tifn15 used the modied accelerating rate calorimeter to characterize the combustion behavior of different heavy and medium crude oil samples based on the continuity of the ARC trace ties. Finally, ramped temperature oxidation (RTO)16 is another technique that aims to study the combustion kinetics of crude oil. Among kinetics experiments, RTO has been mostly used by the authors and others to propose reaction models to simulate in situ combustion processes.1720

A. Kinetics cell experiment

a) Present address: TOTAL - EP/DEV/GIS/SIM, Avenue Larribau, 64018 b) Electronic mail: kovscek@stanford.edu

Pau, France.

In a RTO kinetics cell experiment, a mixture of crude oil, water, sand (reservoir rock), and clay is placed in the cell that is subjected to a controlled temperature program. The air is injected at a constant ow rate from the bottom of the cell. The furnace is programmed to ramp temperature and the temperature of the cell is measured continuously. The composition of the produced gas is measured using a gas analyzer. Figure 1 shows a schematic of a furnace kinetics cell and the result for a 13 API 21 oil sample at a temperature ramp of 2.77 C/min.
2013 AIP Publishing LLC

0034-6748/2013/84(7)/075115/7/$30.00

84, 075115-1

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 177.20.130.9 On: Tue, 10 Dec 2013 18:34:33

075115-2

Bazargan et al.

Rev. Sci. Instrum. 84, 075115 (2013)

FIG. 1. Schematic of a furnace kinetics cell design (left) and a result for a 13 API oil sample at a temperature ramp of 2.77 C/min (right).

As shown in Fig. 1, the oxygen consumption curve (in mole percent) shows two peaks. The rst peak is referred to as low temperature oxidation (LTO) and the second peak as high temperature oxidation (HTO).16 These two peaks are associated with two deviations in the linear temperature prole of the kinetics cell. In a kinetics cell experiment (from now on, referred to as a furnace kinetics cell), the temperature of the furnace, and not the temperature of the kinetics cell, is commonly controlled (Fig. 1). The system possesses signicant thermal inertia due to the thick (order cm) walls of the kinetics cell itself and the furnace elements. Large heating rates that are accurately controlled are virtually precluded for exothermic reactions. Furnace kinetics cell experiments are usually operated in the maximum heating range of 3 C/min.

tube. As suggested in Fig. 3, the heating rates in the reaction zone are between 15 and 30 C/min. The range of heating at the combustion front (Fig. 3) is much larger than the working limit of the heating rate in a furnace kinetics cell. This does not necessarily mean that the kinetics of combustion change with the heating rate. But the predictability of the pseudo reaction model that has been proposed based on the furnace study is questionable. The induction kinetics cell has been designed to work at wider ranges of heating rate including the heating rate at the combustion front, as described next. It also aims to enable us to quench the reaction at different stages and analyze the reactant remaining in the sample. This capability should be useful for examining the reaction intermediates that are important to understanding ISC.22
II. INDUCTION REACTOR A. Induction heating

B. The importance of heating rate

Furnace kinetics cell experimental results are mostly used to screen ISC candidates and propose a lumped reaction model. The reaction model is only a part of the full eld model. The combustion tube experiment, which is used as an intermediate scale experiment between the kinetics cell and the eld, can be availed for validating a given reaction model by comparing combustion tube simulations and experimental results. Figure 2 shows a temperature prole along the combustion tube in a successful in situ combustion process. Each curve in Fig. 2 corresponds to a thermocouple that has been placed at a xed position along the tube. Each thermocouple is separated by roughly 4 cm. The temperature peak is called the combustion front (or re front). As the front moves, the temperature at each position increases smoothly and then the rate of increase slows down. This is the time period where the water vaporizes. After the vaporization of water, the combustion front arrives, the temperature increases quickly, and the temperature peaks. After the front passes each position, the temperature of that position starts decreasing. Figure 3 shows the rates of temperature increase (heating rates) versus time at several positions along the combustion

Induction heating has found many industrial applications of annealing, brazing, bonding, forging, and so on. It has

FIG. 2. Temperature proles for a combustion tube test of a 13 API oil sample. The legend gives the location of each thermocouple.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 177.20.130.9 On: Tue, 10 Dec 2013 18:34:33

075115-3

Bazargan et al.

Rev. Sci. Instrum. 84, 075115 (2013)

FIG. 4. Induction reactor design.

FIG. 3. Heating rate at several positions for the combustion tube test of a 13 API oil sample.

a high heating speed and good efciency. Quick response makes the induction heating process precisely controllable.23 A current (dc or ac) passing through a solenoid coil creates a magnetic eld around the coil. When a magnetic solid bar is placed inside the coil, the induced magnetic eld becomes stronger. Relative permeability (r ) indicates the ability of the material (that is placed inside the coil) to conduct magnetic ux better than a vacuum or air.24 According to Faradays law, alternating voltage applied to a solenoid coil produces a time variable magnetic eld that has the same frequency as the coil current.25 This magnetic eld induces eddy currents (Foucault currents) in the work-piece placed inside the coil.26 These eddy currents produce heat by Joules law.27 Larger relative permeability and resistivity of the work-piece make induction heating more efcient.28 Table I shows the electrical resistivity, relative permeability, and melting point of several metals. The list helps us to choose the best metal to be used in the induction reactor.
B. Reactor design

Figure 4 shows the elements of the induction reactor. A 1.2 kW-100 kHz induction generator (The Roy-1200, FLUXEON) has been used to heat a carbon-steel tube. Carbon-steel
TABLE I. Electrical point31 of several metals. Metal resistivity,29 relative permeability,30 and melting

has very good features for use as the induction heating workpiece. It has high melting temperature, relative permeability, and resistivity (Table I). The oxidation of carbon-steel is reduced by coating the tube with multilayers of silicon (SilcolloyTM 1000). The tube is 12 cm long, has an OD of 1.91 cm, and wall thickness of 2.11 mm. The mixture that contains the oil sample is placed in the middle of the tube. It lls 3 cm of the tube in the middle. This corresponds to about 7 g of the mixture (depending on its composition). The amount of sample in the tube is chosen accurately as a compromise to minimize temperature gradient along the sample, as shown later, and maximize the oxygen consumption. The rest of the tube on both sides is lled with sand. The sand on the inlet side helps to preheat the injected gas. The temperature of the sample is measured with a J type thermocouple. To control the temperature, we use a temperature controller that is connected to the induction generator through a double pole double throw (DPDT) ice cube relay (LY series, OMEGA). The induction generator is attached to the coil. The coil is a copper tube with OD of 0.64 cm and has 11 loops with a diameter of 5.08 cm. Cooling water passes through the coil. The tube has been placed exactly in the middle of the coil using a side arm. The reactor is wrapped with insulation. Insulation does not affect the induced magnetic eld. An efcient heating mechanism and small thermal mass enable us to achieve all of our desired heating rates while consuming less than 300 W.
C. Process design

Resistivity Relative permeability Melting temperature ( C) ( m) (r ) 7.0 107 1.60 108 1.70 108 2.30 108 2.80 108 1.60 107 1.0 107 6.20 108 6.80 108 1.02 0.99 0.99 0.90 1.00 3800 4500 255 600 1510 761 1083 1063 660 1540 1536 1495 1495

Stainless steel 316 Silver Copper Gold Aluminum Carbon-steel Iron Cobalt Nickel

Figure 5 demonstrates the different parts of the system. Two inlets are considered for injection: One for air and one for other gases (e.g., N2 ). The ow rate of each stream is controlled by mass ow controllers (SLA5800 series, Brooks Instrument). The injected gas passes through the reactor and enters the liquid trap. Any particular matter is separated from the ue gas using lters. A back pressure regulator controls the system pressure. Eventually, a ow of 0.2 SLPM is sent to the compact gas analyzer (4200 series, SERVOMEX).
III. RESULT

The mixture for testing consists of 1.5 g of the crude-oil sample (optimal mass depends on the oil), 3.76 g of water,

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 177.20.130.9 On: Tue, 10 Dec 2013 18:34:33

075115-4

Bazargan et al.

Rev. Sci. Instrum. 84, 075115 (2013)

FIG. 5. Process design.

2.00 g of clay, and 50.00 g of the cooked sand. The sand is rst mixed with clay and then water is added. The oil sample is then added and mixed thoroughly. The induction reactor needs to be fed with about 10.00 g (corresponds to 3.00 cm in the middle of the tube) of the prepared mixture for each experiment. Sand is packed adjacent to the sample. The temperature controller applies the heating program. The oxygen consumption associated with the sample, at the given heating rate, must be enough to be detected accurately by the gas analyzer. This means that if less of the sample is used, the injection rate must be accordingly decreased. Also, the sample amount should not be excessive, as it causes deviation from linearity in the temperature history. The amount of sample is kept constant for all tests at all heating rates to compare oxygen consumption. The amount of sample needed, however, decreases as the heating rate increases. For 30 C/min, 2.00 g of sample (0.6 cm in the middle of the tube) is enough. Figure 6 shows the temperature prole and associated oxygen consumption of a 9 API oil sample at a heating rate of 2.60 C/min. The pressure is kept constant at 100 psi while

air is injected at the ow rate of 2 SLPM. Figure 6 shows that oxygen consumption is as expected displaying LTO and HTO regions. Figure 6 also represents the capability of the device to maintain an accurate heating rate. Fast response of the induction heating system minimizes the difference between the programmed temperature and kinetics cell temperature. So not only a RTO function, but any other temperature history may also be achieved by the device. This provides a unique opportunity for kinetics reaction models to test their ability to predict the reaction at combustion tube conditions.
A. Repeatability

In order to demonstrate the repeatability achieved by the induction reactor, two experiments at a low heating rate of 2.84 C/min were conducted. A total amount of 10.13 g of mixture was used in each experiment. Figure 7 shows the oxygen consumption of these two experiments. The two curves in Fig. 7 almost perfectly overlap, and demonstrate repeatability of the induction kinetics cell.

FIG. 6. Temperature prole (left axis) and associated oxygen consumption (right axis) of a 9 API oil sample at heating rate of 2.60 C/min and air ow of 1 SLPM.

FIG. 7. Repeatability of the induction kinetics cell: Two experiments have been conducted at heating rate of 2.84 C/min and air ow of 1 SLPM on an identical 9 API oil sample.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 177.20.130.9 On: Tue, 10 Dec 2013 18:34:33

075115-5

Bazargan et al.

Rev. Sci. Instrum. 84, 075115 (2013)

FIG. 8. Temperature is measured along the sample through thermal well and across the sample by thermocouples with ceramic insulator.

FIG. 10. Oxygen consumption proles at ve different heating rates versus time for the induction reactor and a 13 API oil sample and air ow of 0.6 SLPM.

B. Homogeneous temperature condition

The accuracy of a RTO kinetics study depends on the reactor achieving a homogenous temperature distribution.32 Fast heating and heat transfer in the induction reactor allow it to work in homogenous conditions close to a constant temperature prole along the reactor. The temperature of the sample and of the outside surface of the reactor wall are each measured by thermocouples. Because temperature measurement on the outside surface of the reactor wall is disturbed by eddy currents that are formed during induction heating, thermocouples are covered in OMEGATITE 450 ceramic insulator. The accuracy of the temperature measurement by a thermocouple encased in ceramic insulator was veried. This was achieved by removing the insulation and using a Raytek MI3 infrared temperature sensor (reported accuracy of 1 C) together with the thermocouple with ceramic insulation. Hence, the temperature along and across the sample (Fig. 8) is measured. Figure 9 shows the maximum temperature difference inside the sample at different heating rates. The x-axis in Fig. 9 is the temperature of the sample and the y-axis is the maximum temperature difference inside the reactor.

C. Comparison with furnace kinetics cell

Before increasing the heating rates, we show the consistency of the induction reactor with results from a conventional furnace kinetics cell. We use the isoconversional method18 for comparison. To perform isoconversional analysis, several experiments at different heating rates are needed. The isoconversional method provides a model-free apparent activation energy versus conversion.18 If two reactors are consistent, they should yield a similar apparent activation energy curve versus conversion. Figure 10 shows oxygen consumption proles at ve different heating rates (Fig. 11) for a 13 API oil sample and air ow of 0.6 SLPM. Figure 12 shows the apparent activation energy (isoconversional ngerprint) obtained by the induction reactor and the conventional furnace kinetics cell. The shaded region in Fig. 12 represents the condence region of 95% of the result after taking into account measurement uncertainties (primarily temperature) coming from induction reactor. Figure 12 demonstrates that the two devices give consistent isoconversional ngerprints.

FIG. 9. Temperature difference inside the sample versus temperature of the sample at different heating rates for the induction reactor using a 13 API oil sample and an air ow of 1 SLPM.

FIG. 11. Temperature proles at ve different heating rates versus time in the induction reactor and a 13 API oil sample and air ow of 0.6 SLPM.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 177.20.130.9 On: Tue, 10 Dec 2013 18:34:33

075115-6

Bazargan et al.

Rev. Sci. Instrum. 84, 075115 (2013)

FIG. 12. Apparent activation energy curves of a 13 API oil sample obtained by induction reactor and a conventional furnace kinetics cell.

FIG. 14. Oxygen consumption versus time at heating rates of 8, 12, 16, 20, 25, 30 C/min for a 13 API oil sample and air ow of 0.6 SLPM.

D. Large heating rate

As mentioned earlier, the induction reactor is able to operate at wide ranges of heating rate. Figure 13 shows the oxygen consumption (right axis) and temperature proles (left axis Figure 13) at heating rates of 2, 5, 8, 12, 16, 20, 25, 30 C/min. This is the rst time that such data are available over such a broad range of heating rates for crude oil oxidation.

There are several interesting observations suggested by these gures. First, for heating rates greater than 5 C/min, oxygen consumption is not complete even at 750 C. This means that the heating rate is greater than the reaction rate. Consequently, oxidation reactions continue even after the peak temperature. Second, LTO and HTO peaks merge at larger heating rates. Figure 14 depicts oxygen consumption versus time for heating rates of 8, 12, 16, 20, 25, 30 C/min.

FIG. 13. (a)(h) Temperature prole (left axis) and associated oxygen consumption (right axis) versus time for the induction reactor and a 13 API oil sample at an air ow of 0.6 SLPM.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 177.20.130.9 On: Tue, 10 Dec 2013 18:34:33

075115-7

Bazargan et al.
6 M.

Rev. Sci. Instrum. 84, 075115 (2013) Prats, Thermal recovery, Society of Petroleum Engineers Monograph No. 7 (Society of Petroleum Engineers, 1982). 7 W. L. Penberthy and H. J. Ramey, Design and operation of laboratory combustion tubes, SPE J. 6(2), 183198 (1966). 8 C. Y. Lin, W. H. Chen, S. T. Lee, and W. E. Culham, Numerical simulation of combustion tube experiments and the associated kinetics of in situ combustion processes, SPE J. 24(6), 657666 (1984). 9 H. J. Tadema, Mechanism of oil production of underground combustion, in Proceedings of the World Petroleum Conference, 1959. 10 A. Jess and A. K. Andresen, Inuence of mass transfer on thermogravimetric analysis of combustion and gasication reactivity of coke, Fuel 89, 1541 (2010). 11 J. H. Bae, Characterization of crude-oil for reooding using thermal analysis method, SPE J. 17(3), 211218 (1977). 12 S. Vossoughi and Y. E. Shoubary, Kinetics of crude-oil coke combustion, SPE J. 4(2), 201206 (1989). 13 M. V. Kok, Use of thermal equipment to evaluate crude oils, Thermochim. Acta 214(2), 315324 (1993). 14 D. V. Yannimaras, A. H. Su, and M. R. Fassihi, The case for air injection into deep light oil reservoir, in Proceedings of the 6th European Symposium on Improved Oil Recovery, 1991. 15 D. V. Yannimaras and D. L. Tifn, Screening of oil for in situ combustion at reservoir conditions by accelerating-rate calorimetry, SPE J. 10(1), 36 39 (1995). 16 B. C. Burger and J. G. Sahuquet, Chemical aspects of in situ combustionheat of combustion and kinetics, SPE J. 12(5), 410422 (1972). 17 A. Lapene, G. Debenest, M. Quintard, L. M. Castanier, M. G. Gerritsen, and A. R. Kovscek, Kinetics oxidation of heavy oil. 1. Compositional and full equation of state model, Energy Fuels 25(11), pp. 48864895 (2011). 18 M. Cinar, L. M. Castanier, and A. R. Kovscek, Isoconversional kinetic analysis of the combustion of heavy hydrocarbons, Energy Fuels 23, 4003 (2009). 19 D. D. Mamora, New ndings in low-temperature oxidation of crude-oil, in Proceedings of the SPE Asia Pacic Oil and Gas Conference, 1995. 20 M. R. Fassihi, W. E. Brigham, and H. J. Ramey, Reaction kinetics of in situ combustion: Part 2Modeling, SPE J. 24(4), 408416 (1984). 21 API is a measure how heavy or light a petroleum uid is compared to water. API gravity of a petroleum uid is related to its specic gravity( ) at 60 F .5 by: AP I = 141 131.5.
22 R.

TABLE II. Total oxygen consumption at different heating rates. Heating rate ( C/min) 2 5 8 12 16 20 25 30 Oxygen consumption (102 SL) 66.58 67.01 65.89 67.22 66.19 65.96 64.55 66.39

It shows that LTO and HTO peaks merge together as heating rate increases. At 30 C/min, just one peak is observed meaning that LTO and HTO are completely merged. Third, the total amounts of oxygen consumption (Table II) is the same within measurement error for all heating rates. This has been conrmed using different crude-oil samples.33

IV. CONCLUSION

A new RTO kinetics reactor has been designed. It enables us to study the kinetics of crude-oil oxidation at a wider range of heating rates, covering the entire range of the refront zone in a combustion tube experiment. This helps us to evaluate better the predictability of reaction models for simulating the combustion propagation front. By using the isoconversional method, the consistency of the induction reactor and the furnace kinetics cell results has been veried. Finally, the results of ramped temperature oxidation of a crude-oil sample have been shown. The results suggest:

r At low heating rates, less than 3 C/min, the new reactor gives the same result as the conventional kinetics cell experiment. r A wide range of heating rates is achieved by the new reactor. For the rst time, the oxidation proles of crude oil at large heating rates up to 30 C/min are shown. r The new reactor appears to exhibit heating rates sufcient to achieve rate-limiting oxidation of crude oil.
1 C.

P. Gadelle, J. G. Burger, C. P. Bardon, V. Machedon, A. Carcoana, and V. Petcovici, Heavy-oil recovery by in situ combustion: Two eld cases in Romania, J. Pet. Technol. 33, 2057 (1981). 2 G. Moritis, EOR oil production up slightly, Oil Gas J. 96(16), 4956 (1998). 3 T. M. Counihan, A successful in situ combustion pilot in the MidwaySunset eld, California, in Proceedings of the SPE California Regional Meeting, Bakerseld, California, 1977. 4 D. M. Marjerrison and M. R. Fassihi, Performance of Morgan pressure cycling in situ combustion project, in Proceedings of the SPE/DOE Improved Oil Recovery Symposium, Tulsa, Oklahoma 1994. 5 P. S. Sarathi, In situ combustion handbook: Principles and practices, Technical Report No. DOE/PC/91008-0374 (National Petroleum Technology Ofce, U.S. Department of Energy, 1999).

G. Moore, J. D. M. Belgrave, R. Mehta, M. Ursenbach, C. C. J. Laureshen, and K. Xi, Some insights into the low-temperature and hightemperature in situ combustion kinetics, in Proceedings of the SPE/DOE Enhanced Oil Recovery Symposium, 1992. 23 V. Rundev, D. Loveless, R. Cook, and M. Black, Handbook of Induction Heating (Marcel Dekker, 2003). 24 S. Zinn and S. L. Semiatin, Elements of Induction Heating: Design, Control, and Applications (ASM International, 1988). 25 D. C. Giancoli, Physics: Principles with Applications, 6th ed. (Prentice Hall, 2004). 26 H. D. Young, R. A. Freedman, T. R. Sandin, and A. L. Ford, Sear and Zemanskys University Physics: With Modern Physics, 12th ed. (Pearson Addison-Wesley, 2007). 27 W. Watson, A Text-Book of Physics: Including a Collection of Examples and Questions, 7th ed. (Longmans, CO, 2007). 28 E. Rapoport and Y. Pleshivtseva, Optimal Control of Induction Heating Processes (Taylor and Francis Group, 2007). 29 R. A. Serway and J. W. Jewett, Principles of Physics (with PhysicsNow and InfoTrac), 4th ed. (Thomson, 2003). 30 C. Moosbrugger and F. Cverna, ASM Ready Reference: Electrical and Magnetic Properties of Metals (ASM International, 2000). 31 D. J. Young, High Temperature Oxidation and Corrosion of Metals (Elsevier, 2008). 32 J. Hepola, Construction and operation of an isothermal ow reactor system for study of SO2 removal by injected limestone sorbents, Rev. Sci. Instrum. 61, 2682 (1990). 33 M. Bazargan, Measurement of in situ combustion reaction kinetics with high delity and consistent upscaling for reservoir simulation, Ph.D. thesis, Stanford University, Stanford, CA, 94305-4007, USA, 2013.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 177.20.130.9 On: Tue, 10 Dec 2013 18:34:33