Use of nano-metal particles as catalyst under electromagnetic heatingfor in-situ heavy oil recovery

John Greff, Tayfun Babadagli n

Department of Civil and Environmental Engineering, School of Mining and Petroleum Engineering, University of Alberta, 3-112 Markin CNRL-NREF,Edmonton, AB, Canada T6G 2W2

a r t i c l e i n f o

Article history:Received 2 May 2013Accepted 2 November 2013Available online 11 November 2013

Keywords:heavy-oil recoveryelectromagnetic wavesnano-metal particlescatalysisin-situ upgrading

a b s t r a c t

In order for heavy oil and bitumen recovery to be efficient, all components present within the oil must beproduced. To achieve a highly efficient production process it is essential that we are able to produceasphaltenic components and limit their precipitation. Solvent and conventional thermal techniques arelargely limited in their ability to crack asphaltenic components; thus, new techniques and catalysts areneeded to more efficiently recover heavy oil.

When nano-size metal particles are present they catalyze the breaking of carbon–sulfur bonds withinasphaltenic components. The result of this process is an increase in saturates and aromatics, whilesimultaneously reducing the asphaltene content. This process dramatically lowers the viscosity of heavyoil and bitumen by significantly reducing the average molecular weight. This effect can be dramaticallyincreased by having a strong hydrogen donor present and can be completely inhibited by the removal ofall hydrogen donors. When conducting these types of reactions in-situ, it is very difficult and expensiveto introduce strong hydrogen donors. Therefore, it is imperative that hydrogen donors be created withinthe oil rather than be introduced from an external source.

In this paper, we investigated the effects of microwave radiation, using a 2.45 GHz emitter, on therecovery of heavy oil from a sand pack. Experiments were conducted with and without nano-size nickelcatalyst being present. Heavy oil samples were heated at differing power levels until recovery of heavy oilleveled out. In all cases, the nano-nickel catalysts performed better than their microwave-only counter-parts due to the increased cracking and vaporization demonstrated by Greff and Babadagli (2011) to takeplace in the presence of nano-size metal catalysts and microwaves.

& 2013 Elsevier B.V. All rights reserved.

1. Introduction

Due to the worldwide depletion of easily accessible light crudereservoirs, other more difficult and economically prohibitivereservoirs are gaining greater interest. Since light crude produc-tion is peaking at the same time as global demand for oil isincreasing, this is making medium, heavy, and extra heavy oilreservoirs more crucial than ever to the sustained availability ofcheap oil. Heavy oil and bitumen comprise nearly 70% of remain-ing oil reserves (Liu, 2005), which makes it the most prominentalternative oil source to light oil reservoirs. However, the highviscosity of these oil sources makes them technically challengingand economically prohibitive to produce. The high average mole-cular weight of these substances and the interesting interactionof aspaltenic components give these oils a very high viscosityand density. These two properties cause problems not only for

production, but also for transportation. Currently these propertiesare primarily negated through the use of thermal heating dueto the reduction of viscosity which accompanies the increase intemperature.

Steam is the most commonly used heat carrier for thermalmethods. Steam induced heating relies upon the temporaryreduction of oil viscosity in order to increase the flow rate of theoil through the reservoir and likewise increase the production ofoil. Cyclic Steam Stimulation and Steam Assisted Gravity Drainageare the two most commonly utilized methods by operators. Thesetwo methods rely on the large heat capacity of steam and theavailability of a sufficient water supply. However, sufficient watersupplies are not available globally and likewise these methods arenot universally suitable. In addition, due to the high steam–oilratios that are required for these methods, there are additionalcapital expenditures that are needed to not only create the steambut also to treat the waste water byproduct. These two issuesrepresent major liabilities in the form of lost profitability.

Additionally, beneficial chemical reactions also take placeduring steam stimulation techniques lowering the viscosity of

Contents lists available at ScienceDirect

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

Journal of Petroleum Science and Engineering

0920-4105/$ - see front matter & 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.petrol.2013.11.012

n Corresponding author.E-mail address: tayfun@ualberta.ca (T. Babadagli).

Journal of Petroleum Science and Engineering 112 (2013) 258–265

oil. Hyne et al. (1982) used the term “aqua-thermolysis” in order todescribe the chemical reaction that occurs at high pressure andhigh temperature between steam and certain components ofheavy oil/bitumen. The dominating hypothesis used to explainthis phenomenon is that high pressure and temperature steam iscapable of breaking the carbon–sulfur bonds present within theasphaltenic components of heavy oil/bitumen (Clark et al., 1990;Liu et al., 2002). This effect permanently lowers the viscosity of theheavy oil/bitumen by reducing the average molecular weight ofthe oil.

The “aqua-thermolysis” reaction can be more efficiently per-formed by the addition of catalytic aqueous metal species (Clarkand Hyne, 1990; Clark et al., 1990). When the steam–oil reaction isenhanced by the addition of these metal species, the viscosityreduction of the heavy oil/bitumen is increased significantly. Thiscatalytic property has been found to be associated with all of thefirst row transition metal species, which makes all of these metalscapable of increasing the efficiency of the “aqua-thermolysis”reaction. Additionally, more recent experiments performed byZhao et al. (2002) have demonstrated that these catalytic proper-ties are maintained in-situ. Typically, in the “aqua-thermolysis,”water acts as both the hydrogen and heat donor. However,stronger hydrogen donors such as Tetralin have been shown tobe more effective in combination with a catalyst at reducing theviscosity of heavy oil compared to a weak hydrogen donor such aswater (Zhong et al., 2003). However, strong hydrogen donors suchas Tetralin are very expensive and would therefore be very costprohibitive for field scale implementation.

On the hand, there are many circumstances that steam injec-tion may not be feasible such as deep, and highly heterogeneousand shaly formations. Electromagnetic heating could be an alter-native to steam injection for this type of reservoirs. Microwavesare very photoreactive and therefore it may be possible to usethem in order to produce extremely reactive hydrogen-radicals in-situ. These hydrogen radicals would in turn act as an idealhydrogen donor, both increasing the rate of the “aqua-thermo-lysis” reaction while spontaneously being replenished in-situ. Inaddition, by having catalytic nano-metal catalysts present, it maybe possible to catalyze the breaking of the carbon–sulfur bonds.This would enable us to not only temporarily lower the viscosity ofheavy oil/bitumen through thermal mechanisms, but also toinduce a permanent upgrading of the oil through photochemicalalteration. This permanent change in the oil's properties wouldenable us to increase our overall recovery while producing the oilat lower temperatures. Therefore, we would effectively be able toproduce highly upgraded crude more efficiently, without thepotential environmental liabilities.

2. Basic microwave theory

Microwaves have an extremely high frequency with a rangebetween 300 MHz and 300 GHz. Microwaves are very effectivestimulators of dielectric reactions and are able to cause atomicpolarization, interfacial polarization, and dipolar turning to polar-ization. When we polarize dielectric materials it causes an innerpower dissipation which results in an increase in temperature ofthe material. Microwave heating does not rely solely on convectionor conduction and can likewise heat objects internally regardlessof whether or not physical contact is achieved between themicrowave source and the sample. This enables us to remotelyheat samples very quickly and efficiently, provided that thesubstance absorbs the particular frequency of radiation that isbeing applied (Li et al., 2003). In addition to the heating propertiesof microwaves, there are many specific documented and undocu-mented photochemical reactions that they participate in.

3. Experimental set-up and materials

3.1. Microwave generator

The microwave set-up used for these experiments consisted ofa microwave generator with a set frequency of 2450 MHz. Thismicrowave generator's available power range was 100–1000 W.Power settings were adjustable in 100 W increments. The micro-wave generator and reactor were equipped with an output powerreadout as well as ports to collect produced gases/condensates andliquids, which enabled us to collect gases/condensates and liquidson a continual basis for analysis. A diagram of the experimentalsetup can be found in Fig. 1.

3.2. Crude oil and viscosity measurements

We used a crude oil sample obtained from a heavy oil reservoirin Northern Alberta, Canada. The stock tank crude oil viscositycurve is shown in Fig. 2a. We also measured the viscosity of theproduced liquids and condensates using a Brookfield DV-IIþProviscometer, which enabled us to measure the viscosities of theproduced substances at varying temperatures to establish viscositycurves. A detailed analysis of the effects of the nano- and micro-particle catalyst on the viscosity of heavy-oil was provided in ourearlier study (Greff and Babadagli, 2011). An example of theseanalyses is given in Fig. 2b and c. After treating the oil sampleunder the microwave at 300 W for 5 h, with and without nano-metal particles, the viscosity of the oil samples was measured (thisis called the control case). For comparison, the original oil sampleviscosity was also measured at different temperatures but withoutexposing to any microwave heating. Obviously, the control case(heated in microwave without any metal particles) viscosity inFig. 2b and c (solid lines) is higher than the original oil (Fig. 2a)due to the removal of lighter ends under microwave heating.

In both metal particle addition cases (iron and iron (III) oxide),increasing particle concentration yielded an increase in viscosity.However, below a certain concentration, the catalyst yielded alower viscosity than the control case (sample without catalyst)indicating the existence of an optimal value of concentration. Thisvalue was observed to be around (or less than) 0.1% as can beinferred from Fig. 2b and c. A similar trend was also observed byHamedi and Babadagli (2010, 2011) reporting the optimal

Fig. 1. Microwave design and experimental setup.

J. Greff, T. Babadagli / Journal of Petroleum Science and Engineering 112 (2013) 258–265 259

concentrations for a wide variety of nano- and micro-metal particlesyielding minimal viscosity.

3.3. Catalyst

In the present study, based on observations by Hamedi andBabadagli (2010, 2011), nickel was selected as the metal particledue to its superior performance as a catalyst compared to otherparticles in terms viscosity reduction and aquathermolysis processfor heavy-oil recovery. The optimal concentration of nickel wassuggested to be around 0.1% and this value was used as the nickelconcentration throughout the experiments. A comprehensive listof nickel properties can be found in Table 1.

The effects of nano-metal particles on heavy oil recovery underelectromagnetic heating were investigated using nickel particles.Then, the recovery performances with nickel were compared withother metal particles of nano-size such as iron and iron (III) oxide.The properties of iron and iron (III) oxide particles can be found inGreff and Babadagli (2011).

3.4. Buchner filter funnel

The Buchner Filter Funnel is used to contain the model made ofglass beads and heavy oil. A diagram of the funnel filter iscontained in Fig. 3 and a comprehensive list of properties is givenin Table 2.

4. Sample preparation

Samples were prepared by measuring 40 g of glass beads into60 mL Buchner filter funnels. All Buchner filter funnels wereidentical and had the same dielectric properties. If a catalyst wasto be added to the sample, it was added to the glass beads anddispersed evenly throughout the sample. This sand pack/catalyst

0

20

40

60

80

100

120

140

160

180

200

50 60 70 80 90 100

Visc

osity

(Cp)

Temperature (°C)

Stock Tank

Fig. 2. (a) Viscosity of stock tank oil vs. temperature. (b) Viscosity change withtemperature for different concentrations of nano-iron particles (data from Greffand Babadagli, 2011). (c) Viscosity change with temperature for different concen-trations of nano-iron (III) oxide particles (data from Greff and Babadagli, 2011).

Table 1Nickel nano-catalyst properties.

Assay Z99% Trace metals basisForm NanopowderResistivity 6.97 μΩ-cm, 20 1CParticle size o100 nmbp 2732 1C (l)mp 1453 1C (l)Density 8.9 g/mL at 25 1C (l)

Fig. 3. Buchner filter funnel.

Table 2Buchner funnel filter properties.

Product type Filter funnels

Material GlassDiameter mm �40Joint size number 24/40Porosity 4–8 m

J. Greff, T. Babadagli / Journal of Petroleum Science and Engineering 112 (2013) 258–265260

mixture was then fully saturated with heavy oil through theutilization of partial vacuum. The saturation process was kept inlow pressure vacuum and at a slow rate to ensure that thehomogeneous distribution of particles was not disturbed. Eightexperiments were run and the sample setup and catalyst concen-trations for the experiments are presented in Table 3.

5. Experimental procedure

In order to determine the magnitude of the effects of bothmicrowave intensity and catalytic effects on oil recovery, thefollowing steps were performed: (1) prepare samples as outlinedin the previous section and Table 3; (2) measure the mass of eachsand pack before being saturated with oil; (3) measure the mass ofeach sample after being saturated with oil and before microwaveirradiation; (4) place each sample into the center of the microwaveoven; (5) heat sample at appropriate microwave intensity until oilrecovery levels off; (6) measure the mass of each sample at 5 minintervals; and (7) collect oil produced and condensate producedfor future analysis.

6. Results and discussion

A comparison of experiments with and without a nano-nickelcatalyst is given in Figs. 4, 5, 6 and 7 with four different energies:200, 300, 400, and 500 W, respectively.

The most significant improvements in both ultimate recoveryand recovery rate between catalyst and non-catalyst containingsamples are experienced at the 200 W power setting (Fig. 4). Atthis power setting, the microwave-only sample's production rate isvery slow with a comparatively low final recovery of roughly 30%.However, the nano-nickel catalyst sample perform quite wellreaching a very high peak production rate quickly before leveling

off to a total recovery of roughly 87%. Due to the profoundenhancement of recovery experienced by the nano-nickel contain-ing sample, multiple runs were performed to show reproducibilityof the results. During each trial run, results were completelyreproducible with only minor variations in production markers.The results of two such trials are summarized in Fig. 4.

While all other samples did not experience as profound resultsas the 200 W power range, they were all very significant. All other

Table 3Sample setup and catalyst concentrations.

Sample # Particle Particle Microwavepower setting

(wt%) (W)

1 None 0 2002 None 0 3003 None 0 4004 None 0 5005 Nano-nickel 0.1 2006 Nano-nickel 0.1 3007 Nano-nickel 0.1 4008 Nano-nickel 0.1 500

0102030405060708090

100

0 100 200 300 400 500 600 700 800 900

% R

ecov

ery

Minutes

Recovery vs. Time

200 Watt trial 1 200 Watt Catalys Trial 1

200 Watt Trial 2 200 Watt Catalyst Trial 2

Fig. 4. Recovery vs. time for 200 W with and without catalyst.

0102030405060708090

100

0 20 40 60 80 100

% R

ecov

ery

Minutes

Recovery vs. Time

400 Watt400 Watt Catalyst

Fig. 6. Recovery vs. time for 400 W with and without catalyst.

0102030405060708090

100

0 20 40 60 80 100 120 140

% R

ecov

ery

Minutes

Recovery vs. Time

300 Watt300 Watt Catalyst

Fig. 5. Recovery vs. time for 300 W with and without catalyst.

0102030405060708090

100

0 20 40 60 80 100

% R

ecov

ery

Minutes

Recovery vs. Time

500 Watt500 Watt Catalyst

Fig. 7. Recovery vs. time for 500 W with and without catalyst.

J. Greff, T. Babadagli / Journal of Petroleum Science and Engineering 112 (2013) 258–265 261

samples demonstrated a general trend of producing comparablelevels of oil at roughly half of the time required by their non-catalyst containing counterparts. Additionally, in all cases, totalrecovery was slightly higher for the catalyst-containing samples.All experiments performed showed that nano-nickel catalystgreatly enhanced the recovery of oil achieved through microwaveirradiation. Comparisons of the efficiency of all power settings andconfigurations are summarized in Fig. 8. As seen, the effect ofnano-particles as catalyst is critically important at low powers ofmicrowave. 200 W can be considered threshold value as powersabove it did not show any important change in the recovery whenmetal particles are added. Fig. 8 presents the recovery with respectto energy input, which directly controls the applicability and theeconomics of the process. Another important factor that impactsthe economics is the duration of the application. Fig. 4 indicatesthat time to reach reasonably high recoveries at 200 W withadditives (less than 35 min to reach 75% recovery) is comparableto the time values reached at much higher powers (�30 min at300 W and �15 min at 400 W with additives as can be inferredfrom Figs. 5 and 6). Hence, one may reach a conclusion that,considering the energy input and time, 200–300 W range wouldyield an optimal recovery at laboratory scale conditions from timeand energy input points of view.

The recovery under high microwave energy was obtained in twophases: liquid and vapor. The amount of the recovery in the form ofgas was determined by subtracting the liquid produced from thechange in mass of the model. The results for the recovery achievedin vaporized form are summarized in Fig. 9. As seen, the majority ofoil is produced from the nano-nickel catalyst system through thecracking and consequent vaporization of the heavy oil sample. Thiscracking and vaporization is significant even at low power settings

with only incremental amounts of extra vaporization occurring athigher power levels. This incremental vaporization is likely due to aslight increase in the rate of cracking, while predominantly depen-dent upon the increased temperature of the sample.

The gas produced was subject to condensation and the averageviscosity of it was measured. As seen in Fig. 10, the viscosity of theproduced vaporized gas is significantly lower than that of the stocktank oil (oil collected during the experiments as liquid). Due torelatively high temperature of the experimental system, all vapor-ized components are expected to have a fairly low molecularweight. This validates the results of our previous experimentsconducted earlier to characterize the catalytic effects of nano-sizemetal catalysts (Greff and Babadagli, 2011). In these experiments,it was noted that when heavy oil is exposed to microwaves in thepresence of nano-metal catalysts, significant portions of the heavyoil are cracked and vaporized out of the system. Since the averageviscosity of the produced condensate is in the medium-light range,it confirms that the majority of this vaporization is due to thecracking and consequent vaporization of the heavy oil sample.

As can be inferred from Figs. 11 to 13, the sand which is leftafter nano-nickel assisted production is extremely clean and hasvery little residual oil present. Furthermore, for any given durationof microwave exposure, the nano-nickel catalyst containing sam-ple can visually be confirmed as having a more favorable produc-tion profile. This is due to the sheer volume of vaporization thattakes place within these reactions. Since heavier components arebeing cracked in large quantities, the light components areproduced very easily out of the sample in their vaporized form.This leaves the remaining sand with very little residual oil present.

To give insight into the selection of catalyst for the heavy-oilrecovery process and the oil recovery performances of microwave,

0

10

20

30

40

50

60

70

80

90

100

0 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000

% R

ecov

ery

Joules

Recovery vs. Energy Input

500 Watt 500 Watt Catalyst 400 Watt 400 Watt Catalyst

300 Watt 300 Watt Catalyst 200 Watt 200 Watt Catalyst

Fig. 8. Comparison of recoveries with and without nickel particles at for differenttotal energy input.

0

20

40

60

80

100

0 100 200 300 400 500 600

% o

f Tot

al R

ecov

ery

Watts

Recovery as Gas Phase vs. Power SettingMicrowaveNano Catalyst

Fig. 9. Recovery of oil achieved in the gas phase.

020406080

100120140160180200

0 20 40 60 80 100

Visc

osity

cP

Temperature (°C)

Viscosity of Condensate Stock Tank OilAverage Viscosity of Condensate

Fig. 10. Average viscosities of the recovered condensate and liquid oil produced.

4 cm

Fig. 11. Sample before microwave exposure.

J. Greff, T. Babadagli / Journal of Petroleum Science and Engineering 112 (2013) 258–265262

the experiment with different nano-nickel particle concentrationswere compared with those of nano-iron particles. As seen inFig. 14, nickel acts as a better catalyst compared to iron, especiallyin lower energy cases. The recovery is the same at 400 W casesfor both catalysts due to the high temperature factor; however,at lower energies (200 and 300 W), the catalytic effect of metal

particles are more critical and the nickel case showed a betterrecovery performance than iron.

7. Effect of clay particles on the process

Finally, the experiments were extended to investigate the effectof clay concentrations on the recovery process for differentcatalysts. High permeability unconsolidated oilsands and heavyoil reservoirs in Alberta contain significant amount of clay in thereservoir rock. As known, clay minerals may act as catalyst inheavy-oil up-grading reactions. On the other hand, clay particlesmay result in permeability reduction, i.e., barrier to oil recoverythat eventually yields lower recovery factor during in-situ thermalapplications. To clarify these effects, different clay concentrationswere used during sample preparation. Then, the tests wereperformed for different metal particles at the concentration of0.1 wt%. The clay used was standard bentonite, which composedpredominantly of montmorillonite. The experimental details andresults are given in Table 4. Graphical representation of theseresults is provided in Fig. 15.

One may notice that the addition of clays significantly reducedthe recovery when the recoveries from Exps. 1, 2, and 3 arecompared to the ones without any catalyst in Fig. 8. However,catalyst inclusion significantly contributed to recovery eventhough they never reached the ultimate recoveries given inFigs. 8 and 14. When the recoveries for the same energy inputare compared for different catalysts (Exps. 1, 4, 7, and 9 for 300 Wor Exps., 2, 5, 8, 11 for 400 W, or Exps., 3, 6, 9, 12, for 500 W), onemay observe a significant increase in recovery due to the additionof nano-metal particles for the same clay concentration (10%). Inshort, the importance of catalyst addition in the oilsands environ-ment with clays is more critical than the systems with no clayduring electromagnetic heating applications.

Increasing clay concentration yielded a slight decrease in theultimate recovery. The reduction is around 2–4% when Exps. 7 and9 (10% clay concentration) were compared with their 30% claycontent equivalents (Exps. 13 and 14) for iron oxide. For the nickelcases, this reduction is more significant; 4% for the 300 W cases(Exps. 10 and 15) and 7.5% for the 500 W cases (Exps. 12 and 16).

Also note that most of the recovery is in the form of liquid ifthere is no catalyst in the system (Exps. 1, 2, and 3). Oil recovery inthe form of vapor remarkably increases when nano-iron (Exps. 4,5, and 6) and nano-nickel (Exps. 10, 11, and 12) were used ascatalyst. This indicates that oxidation capability of iron generated asimilar oil recovery in the form of vapor but nickel is still the bestin oil recovery in the form of liquid due to its capability of breakingasphaltenic bonds as also observed by Hamedi-Shokrlu andBabadagli (2012) in their steam injection tests with differentcatalysts.

It should also be emphasized that viscosity reduction is relatedto the clay content. Slightly higher viscosity was obtained for thehigher concentration clay cases (Exps. 13–16) than the other with10% clay (Exps. 4–11). At first sight, this can be contributed to thecatalytic effect of clay on viscosity reduction at lower concentrations(10%) of clay; however more research is needed to clarify this.

8. Conclusions and remarks

Nano-nickel catalysts increase the efficiency of microwave heatingfor heavy oil production. They allow for more oil to be recovered fasterand at a lower total energy input. This is significant because it directlytranslates into lower operating and extraction costs while simulta-neously increasing the overall recovery factor. This relationship is very

4 cm

Fig. 12. Picture of sample after exposure to microwaves at 400 W for 20 min.

4 cm.

Fig. 13. Sample after exposure to microwaves and nano-nickel catalyst for 20 min.

0

10

20

30

40

50

60

70

80

90

100

0 10000 20000 30000 40000 50000 60000

Rec

over

y, %

OO

IP

Energy Input, Joules

400 Watt Nano-Iron

300 Watt Nano-Iron

200 Watt Nano-Iron

400 Watt Nano Nickel

300 Watt Nano Nickel

200 Watt Nano Nickel

Fig. 14. Recovery performances of iron and nickel particles against energy input.

J. Greff, T. Babadagli / Journal of Petroleum Science and Engineering 112 (2013) 258–265 263

apparent at lower energy levels where very little oil can be recoveredthrough the utilization of microwaves alone.

These effects are due to the asphaltene-reducing reactions thatthese catalysts take place in. This enables the oil to be extracted andproduced much more easily from the porous medium as most of therecovery takes place in the gas phase for the catalyst-microwaverecovery. These reactions were characterized in our earlier experi-ments (Greff and Babadagli, 2011) and the results were confirmedonce again through the observations given in the present paper.When heavy oil is exposed to microwaves in the presence of nano-metal catalysts, significant portions of the heavy oil are cracked andvaporized out of the system. The viscosity of the recovered con-densate is significantly lower, measuring roughly 20 cP at roomtemperature. This is extremely significant because not only are werecovering more oil with a higher efficiency with a nano-nickelcatalyst, but we are also producing an extremely upgraded crude oilwhich can be classified as medium-light crude instead of heavy oil.

In the presences of clay, which is very common in heavy oilcontaining sand reservoirs, the effect of catalysts is more crucial.Nickel yielded a better recovery than the other two catalysts (ironand iron oxide), especially in higher energy cases.

Obviously, the technical success at the laboratory scale leads tothe next – and critical – question as to how this can be applied inthe field practically. Although it is beyond the scope of this paper,we can provide some highlights in this regard on the basis ofrecent studies performed in our research group. Davletbaev et al.provided a mathematical analysis of the penetration of radiofrequency waves from a single well (Davletbaev et al., 2011) withsome field scale applications (Davletbaev et al., 2010) emphasizingthe applicability of electromagnetic heating for well stimulations.Moreover, Kovaleva et al. (2011) experimented the injection ofsolvent under radio frequency heating for heavy-oil recovery. Thenext question after the possibility of the application of electro-magnetic waves for single well stimulation is how to inject the

Table 4Experiments with different catalysts at different clay concentrations.

Exp. # Clay content Catalyst Catalyst Power level Viscosity Maximumtemperature

Total oil recovery Vapor recovery Liquid recovery

(%) (wt%) (W) (1C) (% OOIP) (% OOIP) (% OOIP)

1 10 None 0.1 300 5.32 161 11.7 1.4 10.32 10 None 0.1 400 3.98 164 21.5 2.5 18.83 10 None 0.1 500 2.17 172 31.7 4.5 27.24 10 Nano-iron 0.1 300 1.37 141 37.7 8.7 29.05 10 Nano-iron 0.1 400 0.73 173 45.7 11.5 34.26 10 Nano-iron 0.1 500 0.66 213 49.6 35.0 14.17 10 Nano-iron oxide 0.1 300 0.75 195 45.1 – –

8 10 Nano-iron oxide 0.1 400 0.74 236 46.4 – –

9 10 Nano-iron oxide 0.1 500 0.6 284 56.5 – –

10 10 Nano-nickel 0.1 300 0.84 138 42.5 7.8 34.711 10 Nano-nickel 0.1 400 0.58 169 56.6 11.1 45.512 10 Nano-nickel 0.1 500 0.23 196 68.9 14.4 54.513 30 Nano-iron oxide 0.1 300 1.53 228 43.9 – –

14 30 Nano-iron oxide 0.1 500 1.13 337 52.0 – –

15 30 Nano-nickel 0.1 300 1.65 162 38.5 – –

16 30 Nano-nickel 0.1 500 0.86 248 60.4 – –

Fig. 15. Recovery performances of iron, iron oxide, and nickel particles at different clay concentrations.

J. Greff, T. Babadagli / Journal of Petroleum Science and Engineering 112 (2013) 258–265264

nano-particles. Hamedi and Babadagli (2011) studied the injectiv-ity and recoverability of nano-particles in high permeabilityoilsands environment using different carriers. A similar injectionstrategies could be applicable at different stages of electromag-netic heating, practically after a certain amount of heat has beeninjected for initial viscosity reduction of oil to improve theinjectivity of nano-particles carrying fluids.

Acknowledgments

This research was conducted under the second author's NSERCIndustrial Research Chair in Unconventional Oil Recovery (indus-trial partners are Schlumberger, CNRL, SUNCOR, Petrobank,Sherritt Oil, APEX Eng., and PEMEX) and an NSERC DiscoveryGrant (No. G121210595). The funds for the equipment used in theexperiments were obtained from the Canadian Foundation forInnovation (CFI) (Project # 7566) and the University of Alberta. Wegratefully acknowledge these supports. We are also thankful toNSERC and Alberta Innovates Technology Future providing the firstauthor with scholarships through their graduate student supportprograms. This paper is a modified and improved version of IPTC14720, which was presented at the International Petroleum Technol-ogy Conference held in Bangkok, Thailand, 15–17 November 2011.

References

Clark, P.D., Clarke, R.A., Hyne, J.B., et al., 1990. Studies on the effect of metal specieson oil sands undergoing steam treatments. AOSTRA J. Res. 6 (1), 53–64.

Clark, P.D., Hyne, J.B., 1990. Studies on the chemical reactions of heavy oils understeam stimulation. AOSTRA J. Res. 6 (1), 29–39.

Davletbaev, A., Kovaleva, L., Babadagli, T., Minnigalimov, R. 2010. Heavy Oil andBitumen Recovery Using Radio-Frequency Electromagnetic Irradiation andElectrical Heating: Theoretical Analysis and Field Scale Observations. PaperSPE 136611 Presented at the SPE Canadian Unconventional Resources andInternational Petroleum Conference, Calgary, AB, Canada, 19–21 October.

Davletbaev, A., Kovaleva, L., Babadagli, T., Minnigalimov, R., 2011. Mathematicalmodeling and field application of heavy oil recovery by radio-frequencyelectromagnetic stimulation. J. Petr. Sci. Eng. 78 (3–4), 646–653.

Greff, J.G., Babadagli, T., 2011. Catalytic Effects of Nano-Size Metal Ions in BreakingAsphaltene Molecules during Thermal Recovery of Heavy-Oil. Paper SPE146604 Presented at the SPE Annual Technical Conference and Exhibition,Denver, Colorado, 29 October–4 November.

Hamedi, Y., Babadagli, T., 2010. Effects of Nano-Size Metals on Viscosity Reductionof Heavy Oil/Bitumen during Thermal Applications. Paper SPE 137540 Pre-sented at the SPE Canadian Unconventional Resources and InternationalPetroleum Conference, Calgary, AB, Canada, 19–21 October.

Hamedi, Y., Babadagli, T., 2011. Transportation and Interaction of Nano- and Micro-Size Metal Particles during Thermal Injection Applications for Heavy-OilRecovery. Paper SPE 146661 presented at the 2011 SPE Annual TechnicalConference and Exhibition, Denver, Colorado, 30 October–2 November.

Hyne, J.B., Clark, P.D., Clarke, R.A., et al., 1982. Aqua-Thermolysis of Heavy Oils. In:Proceedings of the 2nd International Conference on Heavy Crude and TarSands, Caracas, Venezuela.

Kovaleva, L., Davletbaev, A., Babadagli, T., Stepanova, Z, 2011. Effects of electricaland radio-frequency electromagnetic heating on the mass transfer processduring miscible injection for heavy-oil recovery. Energy Fuels 25 (2), 482–486.

Li, D., et al., 2003. The study and application of gas hydrate decomposition undermicrowave action. Chem. Ind. Dev. 22 (3), 19–22.

Liu, Y., Zhong, L., Fan, H., et al., 2002. Study on the aqua-thermolysis and viscosityreduced mechanism of heavy oil. J. Daqing Pet. Inst. 26 (3), 95–99.

Liu, Z.W., 2005. Technical Challenges of China's Heavy Oil Development. In:Proceedings of the China–Canada Symposium on Heavy Oil Development,Process and Utilization, Beijing, China: China University of Petroleum.

Zhao, X.F., Liu, Y.J., Fan, H.F., et al., 2002. Study on feasibility of heavy oil aqua-thermolysis. J. Fuel Chem. Technol. 30 (4), 381–384.

Zhong, L.G., Liu, Y.J., Fan, H.F., 2003. Liaohe Extra-Heavy Crude Oil UndergroundAquathermolytic Treatments using Catalyst and Hydrogen Donors under SteamInjection Conditions. Paper SPE 84863 Presented at the SPE InternationalImproved Oil Recovery Conference, Asia Specific, Kuala Lumpur, 20–21 October.

J. Greff, T. Babadagli / Journal of Petroleum Science and Engineering 112 (2013) 258–265 265