Synthesis of highly stable γ-Fe2O3 ferrofluid dispersed in liquid paraffin, motor oil and sunflower oil for heat transfer applications

This article aims at the synthesis of highly stable γ-Fe2O3 magnetic nanoparticles and their ferrofluids using different base liquids such as liquid paraffin, motor oil and sunflower oil for heat transfer applications. Phase and morphology of the synthesized nanoparticles were probed using XRD, SEM and FTIR spectroscopy. The average nanoparticle size of γ-Fe2O3 magnetic nanoparticles was found to be 13 nm. Stability of the ferrofluids was monitored by visually observing the aggregation nature of the nanoparticles for 180 days. The ferrofluid prepared using motor oil as a base fluid exhibited high stability (for more than 1 year) and a mean enhancement of 77% in thermal conductivity at 1.5 vol% nanoparticles was observed as compared to base fluid. The viscosity of the ferrofluids was also measured and found to be 18, 38 and 8 cP at 27 °C for the liquid paraffin based, motor oil based and sunflower oil based ferrofluid, respectively.


Introduction
Ferrouids are colloidal dispersions of magnetic particles of size 10 nm in a liquid carrier. These uids have drawn great interest in material research due to their distinct chemical, physical, mechanical and magnetic properties. 1 The magnetic property of uids helps in tracking the location and movement of ferrouids and for precise control by applying and manipulating an external magnetic eld. 2,3 Ferrouids nd applications in technological, materials research and biomedical domains. 4,5 Technological applications include dynamic sealing, heat dissipation, damping, doping of technological materials etc. 6 In Biomedical eld, ferrouids are being used in cell separation, drug delivery in cancer therapy, magnetic induced hyperthermia, as a MRI contrast agent, immunomagnetic separation etc. 7,8 Due to immense applications of ferrouids in wide novel domains, researchers are focusing on synthesis of stable ferro-uids. Ferrouids can be synthesized using nanoparticles of ferromagnetic metals as well as magnetic compounds. The most frequently used ferromagnetic metal is the iron. In specic, iron oxide especially Fe 2 O 3 is the most common oxide of iron. This oxide of iron has been extensively focused in the current research community due to its magnetic and polymorphic properties. Among the polymorphs of Fe 2 O 3 i.e., alpha, beta, gamma and epsilon, 9 the stable structure and properties of alpha (hematite) and gamma (maghemite) are creating interest among researchers. Maghemite nanoparticles have been synthesizing using a variety of techniques which includes co-precipitation, 10 sol-gel, 11 microemulsion, 12 ball-milling, 13 and sonochemistry. 14 Despite of synthesis of ferrouids, stabilization of ferrouids is a major challenging task. Stability of the ferrouids depends on various parameters such as type of solvent, concentration of particles, pH etc. Bateer et al. (2014) synthesized Fe 3 O 4 nano-uid using paraffin as a solvent and claim that nanouid is highly stable in paraffin for 90 days. 15 Jain et al. (2011) prepared ionic liquid ferrouids containing bare and sterically stabilized superparamagnetic iron oxide nanoparticles dispersed in protic ethylammonium and aprotic imidazolium ionic liquids. 16 Kim et al. (2005) synthesized oleic acid stabilized superparamagnetic iron oxide nanouids using chitosan as a solvent. 14 They observe that these nanouids are stable for at least 30 days without showing any sign of aggregation. From the overview of the literature, it has been observed that the ferrouids were mainly dispersed in ionic liquids, paraffin, chitosan solvents etc. to suit best for magnetic applications. However, the synthesis and application of stable ferrouids using motor oil and sunower oil as base solvents for heat transfer is very sparse in the literature other than a work recently published by our group on motor oil based ferrouid (prepared using Fe 3 O 4 nanoparticles) which shows enhancement of 51% in terms of thermal conductivity. 17 This paper aims at resolving this problem.
Moreover, ferrouids also shows good enhancement in thermal conductivity as compared to base solvents. The enhancement of thermal properties depends on various factors such as temperature, pH, volume fraction of nanoparticles etc. 18,19 Yu et al. (2010) synthesized kerosene based Fe 3 O 4 nanouid using phase transfer method and reported that the thermal conductivity increases linearly with the volume fraction of Fe 3 O 4 nanoparticles. 20 Guo et al. (2010) prepared Fe 2 O 3 nanouid using mixture of ethylene glycol and water as base solvent and reported that these uids are exhibiting greater enhancement in thermal conductivity and viscosity as compared to base solvents. 21 Colla et al. (2012) synthesized water based Fe 2 O 3 nanouid and claims that the thermal conductivity and viscosity increases with increase in temperature and particle concentration. 22 From the foregoing discussion, it has been observed that most of the work has been conducted on preparing moderately stable water based Fe 2 O 3 nanouids for heat transfer applications. No work has been conducted on preparing highly stable paraffin oil, motor oil and sunower oil based Fe 2 O 3 nanouid for heat transfer applications. This paper focuses on the synthesis of highly stable Fe 2 O 3 nanouids (ferrouids) using three different oils of varying viscosity such as paraffin oil, motor oil and sunower oil as base uids for heat transfer applications. Thorough analysis of stability of ferrouids has been conducted by observing the aggregation behaviour of fer-rouids with respect to time. Thermal conductivity and viscosity measurements were also conducted by varying various parameters to enhance the thermal and rheological properties of base uids for their application in heat transfer as coolants.

Materials
Ferrous sulfate heptahydrate (FeSO 4 $7H 2 O) and NaOH were purchased from Sigma-Aldrich Corporation, USA. Ferric chloride hexahydrate (FeCl 3 $6H 2 O) was procured from Scientic Limited, UK and extra pure oleic acid was obtained from Gem-Chem, India and used without further purication. The engine oil of 20W-50 grade and vegetable oil (sunower) were purchased from SASO, Saudi Arabia. Paraffin oil was obtained from Loba Chemie, India. HCl was purchased from Chem-Lab, Belgium and absolute alcohol of analytical grade was purchased from Scharlau, Spain. De-ionized (D.I.) water used in this work was puried with a Puris-Expe water system.

Synthesis of g-Fe 2 O 3 nanoparticles
Iron oxide (g-Fe 2 O 3 ) nanoparticles were synthesized by a controlled co-precipitation technique. Initially, ferrous sulfate heptahydrate and ferric chloride hexahydrate were dissolved in deionised water. Aer that, the pH of the above solution was adjusted to 11 by adding alkaline solution drop by drop. Aer adding, the solution turned to black precipitate immediately and the resulting black precipitate was collected with a magnet and the supernatant was removed from the precipitate by decantation. 2 Later, washing of the precipitate was conducted by treating the precipitate with deoxygenated ultrapure water and alcohol mixture. The procedure was repeated for 5 times. Aer washing the precipitate for 5 times, 0.01 M HCl solution was added to the precipitate to neutralize the anionic charges on the surface of nanoparticles. The resulting black powder was isolated using an external magnetic eld. Finally, the magnetic nanoparticles were calcined in an oven at 450 C for 3-4 hours.
2.3 Synthesis of iron oxide ferrouids using paraffin oil, motor oil and sunower oil as base uids Briey, the synthesized magnetic nanoparticles (g-Fe 2 O 3 ) were initially coated with oleic acid by mixing 5% (w/v) Fe 2 O 3 nanoparticles with 10% (v/v) oleic acid and the resultant viscous solution was stirred vigorously for 1 hour at 40 C. Aer stirring, the viscous solution was transferred into a beaker and diluted to 100 ml by adding liquid paraffin oil to obtain colloidal solution. This colloidal solution was further ultra-sonicated for 1 h at 70 C to obtain stable nanouid. Aer sonication, the prepared nanouid was kept in a bottle and the stability of the ferrouid was observed with respect to time. The above procedure has been repeated for the preparation of ferrouids using motor oil and sunower oil as base solvents.

Measurement of thermal conductivity
Heat transfer studies were conducted in WL-373 heat conduction instrument which is particularly calibrated for the estimation of thermal conductivity of liquids and gases. Generally, this unit comprises a double walled cylinder with an integrated heater acting as a heat source, and the surrounding cylinder as a heat sink. The medium to be investigated is placed in between a measurement slot. Aer that, the temperature of heat source and sink were measured using thermocouples and transmitted to a measurement control unit. The heat transfer in the medium is completely due to thermal conduction. Due to the constant width of the measurement slot, it has been assumed that thermal conduction occurs in a plane wall. Hence, the rate of heat ow is calculated using Fourier law:

Characterization
The morphology of nanoparticles was analyzed using Scanning Electron Microscope (JEOL JSM-7600F). The phase purity of nanoparticles were investigated using X-ray diffraction technique (XRD) with D8 AaS Advance X-ray diffractometer using Cu Ka radiation (l ¼ 1.54156 A). The surface bonding was measured using Fourier Transform Infrared (FTIR) spectroscopy (ATR-FT-IR model ''Nicolet IS 10'') equipped with the specular reectance accessory. The viscosity of ferrouid was investigated using a viscometer (Brookeld DV-II + Pro). The thermal conductivity of ferrouids was estimated using WL-373 heat conduction instrument (GUNT, Germany).

Phase and surface morphology
Three different types of ferrouids have been prepared using different base uids. Ferrouid-1 (FF-1) was prepared by dispersing g-Fe 2 O 3 nanoparticles in paraffin oil. Similarly, ferrouid-2 (FF-2) and ferrouid-3 (FF-3) were prepared using motor oil and sunower oil as base uids. Fig. 1 represents the XRD of g-Fe 2 O 3 nanoparticles synthesized using controlled coprecipitation technique. The peaks situated at several diffraction angles such as 31, 36, 43.5, 57 and 63.5 suggests 220, 311, 400, 511 and 440 crystalline phases of g-Fe 2 O 3 nanoparticles (PDF no.-01-089-5892). Moreover, the increased resolution of the planes with peak broadening suggests that g-Fe 2 O 3 crystal was showing a mean size of 12.6 nm. Fig. 2 shows the SEM of g-Fe 2 O 3 nanoparticles. The particles are showing slight agglomerating nature for high concentration of particles even though the sample was sonicated for 15 minutes in motor oil before drop casting the sol on the aluminium foil. From the gure, it has been observed that the particles are spherical in nature and showing a mean size of 13 nm (around 100 particles were taken from Fig. 2 for calculating mean size using ImageJ soware) which is in consistent with the crystallite size calculated from the XRD analysis. The aggregation behavior of the particle were possibly due to the enhanced magnetic and inter particle interaction between the particles. 2 The nature of produced ferrouid was observed using FTIR analysis. Fig. 3 shows the FTIR spectra of g-Fe 2 O 3 nanoparticles synthesized by co-precipitation method. The sample for FTIR analysis was prepared as a pellet by mixing g-Fe 2 O 3 nanoparticles with KBr powder using hydraulic press. The characteristic bands at 562 cm À1 and 630 cm À1 were attributed to Fe-O bonding in g-Fe 2 O 3 nanoparticles. Another characteristic peak has been noticed at 3416 cm À1 corresponds to -OH group of water. A small dip in the FTIR spectra has been observed at 2362 cm À1 due to the presence of atmospheric carbon dioxide. 23

Stability of ferrouids
Since stability is the most important characteristics of ferro-uid, the detailed stability analysis of prepared ferrouids was conducted by observing the aggregation behavior of ferrouids with respect to time. In this work, the stability analysis of 1.5 vol% nanoparticle based three ferrouids (FF-1, FF-2 and FF-3) was conducted by storing the ferrouids in a beaker in an open ambience for 180 days. Digital images of the ferrouids were taken in regular intervals of time to observe the aggregation behavior of nanoparticles in uid. The digital images of ferrouids with respect to time were shown in Fig. 4 (FF-1), Fig. 5 (FF-2) and Fig. 6 (FF-3) respectively. Fig. 4 represents the stability analysis of FF-1 synthesized using paraffin oil as a base uid. From the gure, it has been observed that ferrouid is highly stable for 30 days and partially stable for 120 days. Aer 120 days, slight formation of ocs was noticed in the beaker due to detachment of oleic acid from the surface of nanoparticles. Complete aggregation of nanoparticle was not observed even aer 180 days. These formed ocs were broken down by mild sonication for 5 minutes. Aer sonication, ferrouid is showing good stability i.e. stable for more than 1 month. Fig. 5 show the stability of FF-2 with respect to time. FF-2 is showing highly stable nature even aer storing for 180 days in open atmosphere.   No sign of aggregation has been noticed in the beaker even aer 180 days. From the gure, it is clear that even ocs are not formed. This high stability of ferrouid in motor oil may be due to the refractive index or dielectric constant of the solvent. [24][25][26] Formation of ocs has been observed aer 9 months and these ocs can be broken down by mild sonication. Aer sonication, ferrouid is highly stable for more than one year.      6 represents the visual observation of aggregation behaviour in FF-3. From the gure, it can be seen that the fer-rouid is highly stable for 90 days and partially stable for 120 days. Aer 120 days, a slight formation of ocs was observed which can be broken by mild sonication for 5 minutes.

Thermal properties of ferrouids
The variation of coefficient of thermal conduction for FF1 with various volume fraction of g-Fe 2 O 3 nanoparticles (0.5, 1.0 and 1.5 vol%) is shown in Fig. 7(a). From gure, it can be seen that the thermal conductivity of FF-1 increases with increase in the concentration of g-Fe 2 O 3 nanoparticles. The values of thermal conductivity at 0.5, 1.0 and 1.5 vol% are 0.145, 0.165 and 0.210 W mK À1 respectively. FF-1 prepared by using 1.5 vol% of g-Fe 2 O 3 nanoparticles shows 45.2% enhancement (mean) of thermal conductivity as compared to paraffin oil. Similar trends were observed for FF-2 and FF-3 as shown in Fig. 7(b) and (c). The mean enhancement of thermal conductivity in FF-2 and FF-3 are 77% and 20.6% as compared to mineral oil and sunower oil as shown in Fig. 7(d). A remarkable enhancement of thermal conductivity was observed in ferrouids prepared using mineral oil as base uid due to highly stable nature of ferrouid, small size of nanoparticle and interaction between g-Fe 2 O 3 nanoparticles and the baseuid. 27

Rheological properties of ferrouids
Rheological behavior of ferrouids signicantly impacts their application in various elds such as in valves, pneumatic servo controller, and shock absorber etc. 28 Viscosity is one of the important properties associated with ferrouid, which depends on the nature of carrier liquid as well as the concentration and size of magnetic nanoparticles. 29 The viscosity of ferrouids for 1.5 vol% nanoparticle concentration were measured initially at room temperature (i.e., 27 C) and found to be 20 cP, 42 cP and 8 cP for FF-1, FF-2 and FF-3 respectively. These viscosities of fer-rouids were compared with the base uids as shown in the Fig. 8. Viscosities of all ferrouids and base uids were found to be decreasing with temperature and for all ferrouids it was found to be around 1 cP at 55 C. Moreover, viscosities of base uids are comparable with their ferrouid counterpart till 33 C for both paraffin and motor oil, whereas 28 C for sunower oil. This conrms that the synthesized ferrouids is inuenced by the presence of nanoparticles at higher temperatures as reported by Ahammed et al. (2016). 30

Conclusions
g-Fe 2 O 3 nanoparticles was successfully synthesized by coprecipitation method and its ferrouids were prepared by dispersing nanoparticle in paraffin oil, motor oil and sunower oil. The average crystal size of g-Fe 2 O 3 nanoparticles was calculated (12.6 nm) using Scherrer equation from XRD analysis. SEM analysis conrms the spherical nature of the particles and average size was also comparable with the XRD analysis result. A remarkable enhancement (mean) in thermal conductivity was observed for motor oil based ferrouid (FF-2) i.e., 77%. Viscosity behavior of ferrouids was found to be similar for all ferrouids with increase in temperature. This conrms the high heat transfer rate and ow ability of ferrouids in heat transfer applications such as in coolant pipes and other channels.

Conflicts of interest
There are no conicts of interest with the authors.