Investigations on pool boiling critical heat flux,transient characteristics and bonding strength of heater wire with aqua based reduced graphene oxide nano fluids

R.Kamatchi*,G.Kumaresan

1 School of Mechanical Engineering,VIT University,Vellore Institute of Technology 632014,Tamil Nadu,India

2 Department of Mechanical Engineering,Bannari Amman Institute of Technology,Sathyamangalam 638401,Tamil Nadu,India

1.Introduction

System miniaturization and high heat removal rate are the primary concern of any heat transfer devices,and hence,a lot of investigations are carried out in this field for the last two decades.Pool boiling is considered as one of the efficient ways of dissipating large amounts of heat flux at low wall superheat.Critical heat flux(CHF)is also known as peak heat flux that explains the thermal limit during pool boiling beyond which the heater material gets burn out.Considerable research works have been reported to enhance the CHF since Nukiyama[1].However,most of the investigations are performed in a scaled model system due to the limitations of technical feasibility and financial constraints.

The modeling techniques generally used in pool boiling are geometric and fluid modeling.In comparison with geometric modeling, fluid modeling(nano fluid is used instead of base fluid)plays a major role due to its enhanced thermal properties since 2003.As sesen from the literature,the use of nanoparticles with different base fluids enhances the CHF to a greater extent.Till date,different types of nanomaterials like alumina,diamond,magnetite,reduced graphene oxide(rGO),CNT,CuO,SiO2,TiO2,and ZnO are used to enhance the CHF[2–8].Recently,Rainho Neto et al.[5]found the CHF of alumina,maghemite and CNT nano fluids vary from 26%to 37%for the tested concentrations.Further,they reported that the wetting behavior of deposited heater surface is the reason forthe enhancement.Kamatchi et al.[7]reported a poolboiling CHF of about 245%for 0.3 g·L?1concentrations of aqua based reduced graphene oxide nano fluids.Most of the studies explained that the enhancement is mainly because of the changes in surface characteristics of the heater material by the deposition of nanomaterials[2–6].However,the growth of deposited layer increases with concentrations,and hence,the CHF enhancement may be related to the thickness of deposited layer[7].Thus,the physical mechanism behind the enhancement is not clear.Different research groups correlated their results with different boiling heat transfer models[2].For instance,Kamatchi et al.[7]used macrolayer dryout model[9]while Rainho Neto et al.[5]correlated their results with Kandlikar model[10].

As observed from the literature,a very few studies are reported on transient heat transfer characteristics and nanocoated surface with base fluid.For instance,Kwark et al.[11]investigated the transient characteristics of alumina,CuO,and diamond nano fluids.However,the CHF value is unaltered during the study on transient heat transfer characteristics.Kim and Kim[12]used nanocoated wires to study the enhancement of CHF with base fluids.They found that the CHF enhancement of nanocoated wires with base fluid has higher value than bare surface using nano fluids.However,this requires a number of test runs to establish the CHF enhancement.On the other hand,Kwark et al.[11]reported nearly an identical pool boiling CHF value up to 16 tests,indicating that the nanoparticles deposits have good bonding with the heater surface.

Graphene(an atomically thin sheet of SP2hybridized carbon in 2-D form and honeycomb lattice structure)has excellent thermal properties than any other nanomaterial.Interestingly,the enhancement in CHF with graphene nano fluids and graphene coated surfaces is higher than the metal/metal oxide nanoparticles.This is because of the formation of uniform porous structure on the heater material[13,14].However,they are notwidely used in poolboiling due to theirpoor dispersion stability and hydrophobic nature.Kamatchi et al.[7]used aqua based rGO nano fluids in their experimental work.They reported the rGO flakes ensure better dispersion stability than graphene.This powerful merit can be utilized in preparing rGO-waternano fluids and tested in poolboiling.

Therefore,this work mainly investigates the transient characteristics during boiling of rGO nano fluids and bonding strength of rGO deposits on the Ni-Cr wire.As a part of the study,the rGO is synthesized in our lab,and various characterization studies are done including the stability analysis to identify the maximum limit in concentrations.

2.Synthesis and Characterization of rGO Flakes

2.1.Synthesis of rGO

The method of synthesis of rGO is represented in Fig.1.Itis obtained first by converting graphite to graphite oxide and later by reducing graphite oxide to rGO.

Fig.1.Method of preparation of rGO.

2.1.1.Conversion of graphite into graphite oxide

The preparation of graphite oxide(GO)is followed by modified Hummer method[15].Concentrated sulfuric acid(350 ml)is added to a mixture of graphite powder(8 g)and sodium nitrate(8 g)and is surrounded by an ice bath in order to avoid the evaluation of heat generation during the process.Next,Potassium permanganate(45 g)is added into the mixture by keeping the reaction temperature below 18°C.With the aid of heater cum magnetic stirrer,the solution is then heated to 40°C and stirred for about 18 h at which hydrogen peroxide(1250 ml)is added slowly.It is then stirred for about 30 min at which DI water is added and the color of the solution is turned to brilliant yellow.In order to remove the ions,the mixture is washed and filtered with 3%Hydrochloric acid,ethanol and repeatedly,DI water.Finally,the end product is kept in vacuum to obtain the dry graphite oxide(GO).

2.1.2.Reduction of GO into reduced graphene oxide

The thus obtained GO(4 g)is dispersed in water and sonicated for about 3 h to obtain a homogeneous dispersion of GO solution.During magnetic stirring,sodium borohydride(16 g)is added,and the reaction is held for half a day period.It is necessary to maintain the pH of the solution(9–10)during this process.The mixture is then filtered and repeatedly washed with water.The final product is dried in vacuum for a period of one day to obtain the rGO.

2.2.Characterization

Fig.2 depicts the detailed characterization techniques followed on the prepared rGO.The XRD measurements are performed by Rigaku Ultima III(Make:Rigaku,US)to identify the crystallite nature of the as-synthesized rGO.As seen from Fig.2(a),the 2θ value of rGO is about25.03°which indicates the formation of mono or multilayer structure with smaller crystallite size[7].Fig.2(b)shows the Raman spectrum of rGO obtained over the range of 500–3200 cm?1by Raman Microscopy(Make:Renishaw,UK).The peak intensity values of G and 2D band are used to identify whether the rGO is mono or multilayer(2–6 layers)structure[7].In the present study,the G band is at 1579 cm?1which is lower than the peak value of monolayer(1585 cm?1),and hence,the synthesized rGOhas a chance of multilayer structure.Chakrabarti et al.[16]stated a single sharp peak in 2D band exhibits monolayer graphene while splitting in the 2D band indicates multilayer graphene.Fig.2(b)illustrates a splitting in the 2D band,and hence,the as-synthesized rGO has multilayer structure.Also,the rGO has 7–10 layers when the 2D band value exceeds 2700 cm?1[16].In Fig.2(b),the 2D band shows a value of 2640 cm?1,and hence,our product has 2–6 layer structure.

The presence of functional groups is identified by FT-IR spectroscopy(Make:Perkin Elmer,US).Fig.2(c)represents the FT-IR spectrum of rGO,obtained over the range between 4000 and 500 cm?1.The wavelength at 1035,1190,1448,1636,and 1743 cm?1indicates the absorption band,C--OH groups,O--H groups,phenolic O--H deformation,C--C stretching,and carbonyl stretching respectively.Kamatchi et al.[7]observed the presence of hydrogen bonding,C--OH and O--H groups favors to acquire stable rGO in water medium.In this work,the existence of carbonyl and carboxyl groups is evident from the FT-IR and hence,our product has a chance of high dispersion in water.

The SEM image of rGO flakes are examined by TESCAN VEGA3(Make:TESCAN,Czech Republic)and are shown in Fig.2(d)and(e).The rGO is found to be folding,irregular and layer structured.Also,the thickness of flakes is examined by AFM(Make:Agilent Technologies,US)and found as～4 nm(Fig.2(f)).The obtained thickness value is 5 times higher than the thickness(～0.7 nm)of monolayer graphene[17].Therefore,the as-synthesized rGO has 2–6 layers which is in accordance with Raman's spectroscopy.

Fig.2.Characterization of reduced graphene oxide(a)XRD(b)Raman's spectrum(c)FT-IR spectrum of rGO(d)and(e)SEM image of rGO flakes(f)AFM image of rGO flakes.

3.Preparation and Stability Analysis of Aqua Based rGO Nano fluids

The thus prepared rGO flakes are quantified using a digital weighing scale(accuracy:0.1 mg).It is of interest to know the maximum limit in concentrations,and hence,the different concentrations such as 0.01,0.05,0.1,0.2,0.3,and 0.4 g·L?1of aqua based rGO nano fluids are prepared initially.The fluids are kept in an ultrasonic homogenizer(Ultraturrax T25,IKA,Germany)and probe sonicator(Vibra-cell,Sonics,US)to get uniform dispersion of aqua based rGO nano fluids.As observed from the literature,agglomeration is the major problem and this affects the performance of heat transfer equipment.Hence,sedimentation and Zetapotential study are carried out in the present work.

3.1.Sedimentation

The sedimentation study is conducted by keeping the different concentrations of rGO nano fluids in an isolated place.Fig.3 illustrates the photographic image of rGO nano fluids.It is observed that there is no sedimentation for all the tested concentrations even after 7 days.This is mainly due to the presence of C--OH and--OH groups at the edge of the flakes which helps to obtain stable nano fluids.However,the agglomeration starts to appear on 10th day and completely settles at the bottom of the vial on 15th day for 0.1,0.2,and 0.3 g·L?1concentrations.Also,some agglomeration is seen in 0.01 and 0.05 g·L?1of rGO nano fluids during 15th day of preparation.

Fig.3.Sedimentation analysis of rGO-water nano fluids.

Fig.4 shows the photographic image 0.4 g·L?1concentration of aqua based rGOnano fluids.As seen from the Fig.4,some agglomeration and sedimentation is observed during the third day itself.After 5 days,the accumulated rGO flakes settled at the bottom of the vial.Itis because of higher amount of rGO flakes in DI water in 0.4 g·L?1which further leads to faster rate of agglomeration.Therefore,it is concluded that the maximum limit in concentration of the present study is 0.3 g·L?1.

Fig.4.Sedimentation study of 0.4 g·L?1 concentration(a)Day 1(b)Day 3(c)Day 5.

3.2.Zetasizer

The size of the flakes and stability of nano fluids are evaluated by Zetasizer ver.6.20(Make:Malvern Instruments Limited,UK).As observed from the literature,most of the studies measured the Zetapotential value immediately after the preparation.In the present study,Zetapotential is measured after 5th day of preparation since there is no sedimentation even after seven days as evident from the sedimentation study.Also,the maximum concentration(0.3 g·L?1)is chosen to evaluate the dispersion stability rather than conducting experiments for all the concentrations.Fig.5 depicts the Zetapotential curve of rGO nano fluid and shows a value of?39.1 mV.This refers that the prepared nano fluids have better dispersion stability[7].Dynamic light scattering technique is followed to examine the size of the flakes and is represented in Fig.6.The size of flakes is found to be 0.322 μm which is slightly higher than the results obtained from the AFM study(Fig.2(f)).This might be due to the agglomeration of flakes on 5th day after preparation.

Hence,it is recommended to use the rGO nano fluids in the pool boiling test facility for a period of 7 days after the preparation.

4.Pool Boiling Experimental Setup

Fig.7 shows the photographic image ofpoolboiling CHF experimental setup.The boiling vessel is made of borosilicate vessel whose diameter and thickness are 120 mm and 4 mm respectively.The required quantity of working fluid is poured into the vessel and is allowed to boil until it attains its saturation temperature.The top of the boiling vessel is enclosed by a te flon cover whereby the re flux condenser and pressure gauge are fixed.In order to condensate the steam and to maintain the constant level of working fluid during the boiling process,cold water is circulated into the re flux condenser with the help of peristaltic pump.Also,a pressure gauge is used to ensure whether the boiling experiments are conducted at atmospheric conditions or not.The thermocouple(T-type)is inserted into the pool to measure the temperature of the working fluid.A thin Ni-Cr wire heater(diameter:0.42 mm;length:80 mm)is soldered to the copper electrodes and the other end of the electrodes is connected to a DC power supply(32 V/60 A)as shown in Fig.7.

As a first step,the preheater is made ON and the working fluid is allowed to boil to uproot the non-condensable gases.After that,the power input is given to the Ni-Cr wire with the aid of DC power supply.A step increase in voltage of～0.5 V is given initially.Once the expected CHF is attained,the power input is given in small increments(～0.25 V).The appearance of red hot conditions on the wire implies the occurrence of CHF and is calculated by,

As proposed by Holman[18],the uncertainty in CHF is determined using Eq.(2).The accuracy of ammeter and voltmeter is 0.25%and 0.01%respectively.The T-type thermocouple has an error of±0.5 °C in the measurement of temperature.The error in measuring diameter and length are±0.0001 and±0.1 mm respectively.The uncertainty in CHF is found as±2.55%.

Using the same test facility,transient characteristics during pool boiling with Ni-Cr wire and CHF Studies on rGO deposited wires with DI water are also investigated.Results of this study are discussed in Sections 5.2 and 5.3.

Fig.5.Zeta potential curve for 0.3 g·L?1 concentration.

Fig.6.Size of rGO flakes.

Fig.7.Photographic view of pool boiling heat transfer facility.

5.Results and Discussion

5.1.CHF investigations with water and various concentrations of nano fluids

Initially,three trials of CHF experiments are carried out with water at different timings to ensure repeatability,reproducibility,and also to get deep understanding of boiling phenomenon.A mean CHF value of 0.985 MW·m?2is observed for DI water and compared with Zuber's correlations[19],as given in Eq.(3).The difference in CHF value between the experimental data and estimated result is less than 4%.

After that,the aqua based rGO nano fluids(0.01–0.3 g·L?1)are poured in the test facility to investigate the enhancement in CHF.It is found that the CHF occurs on the cathode side of the wire for the tested samples while it occurs randomly over the wire for base fluid.The experimental investigation of Ahn et al.[13]explained that this is because of the presence of negatively charged carboxyl groups(--COOH)on the rGO flakes and are attracted by the Coulomb force towards the anode side.Hence,the formation of rGO structure on the anode side of the wire.However,a thin layer of rGO deposits occurs on the cathode side which determines the enhancement in CHF.Also,Ahn et al.[13]concluded that the coating time is not the in fluencing factor for the determination of CHF when the tests are performed with one side coating.Therefore,the power inputis given to the heater wire after every 10 min in the present study.The average value of three trials is used to determine the CHF and the values are 1.42,1.62,1.79,2.16 and 2.40 MW·m?2respectively for the tested samples.

Fig.8 shows the CHF enhancement of aqua based rGOnano fluids.The CHF is found to increase with concentrations and attains a maximum of 245%.Results of this study are in contradiction with Kwark et al.[11]who reported the CHF remains the same beyond 0.025 g·L?1concentration for alumina,CuO,and diamond nano fluids.Therefore,the obtained CHF may be due to the uniformity of rGO deposits on the Ni-Cr wire and this uniformity increases with nano fluid concentrations.This is also in coincidence with the experimental investigations of Seo et al.[20]who found the rGO deposits on the heater surface forms uniform porous layer and each pores may act as a nucleation sites which further helps in enhancing the CHF.

Fig.8.Enhancement in CHF for various concentrations of rGO nano fluids.

In order to identify the uniformity of the rGO deposits,the SEM image of heater wire after boiling with low(0.01 g·L?1),medium(0.1 g·L?1),and high(0.3 g·L?1)concentrations are captured at different magnification levels and are shown in Fig.9.As seen from the Fig.9,the heating surfaces are covered by the deposits according to its concentrations.At low concentration,the flakes are distributed evenly on the Ni-Cr wire surface whereas the heater material is entirely covered by the rGO deposits at medium and high concentrations.

The mechanism for the enhancement of CHF with thin Ni-Cr wire is analyzed in detail using various boiling heat transfer models in our earlier study[7].Finally,we found the gain in heat flux using macrolayer dry out model[9]:

Fig.9.SEM image of Ni-Cr wire after boiling with nano fluids.

Seo et al.[20]found the porosity(ε)of porous graphene as 0.131,and the same porosity value is chosen in this study.The boiling process of all the tested concentrations are recorded and the bubble growth time(τd)is then found,which varies from 10 to 15 ms.Further,the thickness(δp)of rGO deposit layer of each nano fluid boiled thin Ni-Cr wire is measured using laser scan micrometer(Make:Keyence,Japan).The average of three trials is used to calculate the thickness of rGO layer and the values are 26,30,34,40,and 47 μm respectively.It is observed that the thickness increases with concentrations.

The net gain in heat flux for the tested concentrations is found as 0.493,0.582,0.693,1.12,and 1.33 MW·m?2respectively.The macrolayer dryoutmodel[9]predicts the CHF of about235%for 0.3 g·L?1concentrations whereas the experimental result shows 245%.Therefore,the rGO deposited layer delays the occurrence of local dry out and a subsequent rise of temperature,which is the mechanism behind the enhancement in CHF of rGO nano fluids with thin Ni-Cr wire.

Fig.10 represents the flow visualization of base fluid and rGO nano fluids boiled with thin Ni-Cr wire.The images are taken at different heat flux conditions,and the lastpicture in each column indicates the moment just before attaining the CHF.It is seen that the Ni-Cr wire in water burns out at～1 MW·m?2,while the wires in the 0.05 and 0.3 g·L?1of rGOnano fluid have an extended life.During early nucleate boiling region,the existence of small vapor bubbles and an increased bubble density are noticed with increasing concentrations.Further,this porous layer on the heater surface acts as secondary nucleation sites and hence,the enhancement in CHF of rGO nano fluids.This is in coincidence with the earlier study of Chang and You[21]who showed a direct relation between porous layer and nucleation site density.

Interestingly,the appearance of red hot conditions on the thin wire differs for each fluid,which is shown in the last picture of each sequence.When DI water is used as working fluid,the entire length of the wire turns into red hot conditions after attaining the CHF and immediately breaks.On the other hand,the red hot condition appears on the cathode side of the wire for rGO-water nano fluids.Also,the glowing length differs for each concentration(i.e.,for 0.05 g·L?1,it appears only a few millimeters whereas it is more for0.3 g·L?1).At0.05 g·L?1concentrations,the local dry out at the surface leads to sudden jump in temperature for the applied heat flux and the wire burns out at the local hot spot whereas the dry out is global instead of local at higher concentration,as evidenced by increasing glowing length[21].

5.2.Investigations on transient characteristics during pool boiling of rGO nano fluids

The experimental study of Kwark et al.[11]reported the deposits modify the heater surface and alters the boiling phenomenon.As seen from the Fig.9,the thickness of rGO deposits increases with concentrations which results surface modification on the heater material.Therefore,the rGO-water nano fluids may exhibit transient characteristics and are investigated in this work.As a first step,a bare Ni-Cr wire is used in the pool boiling experimental setup.Three trials of CHF investigations are carried out for each concentration(0.001,0.005,0.1,0.2,and 0.3 g·L?1)with out taking the Ni-Crwire heater assembly from the pool.Fig.11 represents the CHF value obtained for every boiling run.As seen from the Fig.11,the CHF of all the tested concentrations remains constant for the consecutive experiments.However,the boiling performance mainly depends on the incremental in heat input and boiling duration[11].Hence,the transient characteristics studies are conducted by considering these parameters.5.2.1.Effect of heat input in steps to the thin wire

Fig.10.Flow visualization in Ni-Cr wire surface.

Fig.11.CHF value at different boiling run.

The effect of heat input in steps to the thin wire is investigated for low(0.01 g·L?1)and higher(0.3 g·L?1)concentrations,rather than conducting CHF experiments for all the concentrations.Two different methods,denoted by M–1 and M–2 are followed for giving the heat input to the Ni-Cr wire during pool boiling investigations.In M–1,the increment in heatinput is:0.01 MW·m?2until the onset of nucleate boiling(0.05 MW·m?2);0.05 MW·m?2until 0.9 MW·m?2;and 0.01 MW·m?2thereafter until the CHF is attained.In M–2,the incremental in heat input is kept constant(0.01 MW·m?2)from the beginning to the attainment of CHF.Three trials of experiments are conducted for M–1 and M–2 and are represented in Fig.12.

Fig.12.Effect of incremental in heat input to the Ni-Cr wire.

At low concentration,a negligible change in CHF is seen for both methods and this is mainly due to the slower rate of flakes deposition on the heater material.Also,the obtained CHF value is same as that of the consecutive experiments(Fig.11).At 0.3 g·L?1concentration,a slight increase in CHF is obtained in M–2 in comparison with M–1.This is because of the incremental in heat input is very low which leads to more uniformity of the surface and thus the CHF.

5.2.2.Effect of maintaining constant heat flux for a specified duration

The transient nature of nano fluid boiling and its dependence on the heat flux is also demonstrated by maintaining heat flux for a specific duration.The higher concentration(0.3 g·L?1)of the nano fluid is chosen in this study.The experiments are conducted by giving the heat input in steps at 10-min interval until it attains 1 MW·m?2.The heat input is held constant for about 30,45,and 60 min for the first,second and third experiments respectively.After this wait time,experiments are performed by incrementing the heat input until it reaches the CHF.It is noted that three trials of investigations are carried out for 30-,45-,and 60-min duration.It is found that the CHF do not vary much with time,which is shown in Fig.13.A maximum difference of 0.08 MW·m?2is observed between 30-and 60-min duration.The results obtained are in agreement with Ahn et al.[13]who suggested that the coating time does not play a role when the tests are conducted in one side coating.

Fig.13.The influence of coating time on the CHF.

It is of interest to know the reason for the no change in CHF during transient heat transfer studies.The thickness of rGO deposits on the wire surface is measured using a laser scan micrometer.Surprisingly,the thickness of rGO layer does not vary much for all the experiments,and hence,the CHF is unaltered.Therefore,it is concluded that transient characteristics during the boiling of rGO-water nano fluid lead to an interesting feature that the CHF remains same for the respective concentrations irrespective of heat input in steps and coating time.Also,the CHF depends on the thickness of flakes deposits and its uniformity.

5.3.CHF experiments with nanocoated heater

Similar to transient characteristics,low(0.01 g·L?1)and high(0.3 g·L?1)concentrations are chosen to investigate the effect of nanocoating on the enhancement of CHF.Two sets of experiments are conducted in this study and each set of tests consists of(i)a bare Ni-Cr wire heater in rGO-water nano fluids and(ii)nanoparticles deposited Ni-Cr wire in DI water.Hereafter,the nanocoating produced during boiling with 0.01 g·L?1of rGO nano fluid is defined as “0.01 g·L?1nanocoated heater”(0.01 g·L?1NCH)and the nanocoating produced during the 0.3 g·L?1concentration is designated as “0.3 g·L?1nanocoated heater”(0.3 g·L?1NCH).

In most of the research work,the heater material is removed from the pool after the CHF experiments and allowed to dry in air.Due to this,the presence of rGO nano fluid droplets on Ni-Cr wire evaporates,causing the additional coating.These additional nanoparticles deposition may alter the boiling performance and does not represent the actual surface condition of nano fluid boiling.In the present study,the heater which is removed from the nano fluid bath after CHF experiments is immediately immersed in DI water in order to avoid the dry out in air.Hence,this method ensures that the nanoparticles having good bonding with the heater surface has an influence on the boiling.

Fig.14 depicts the CHF of rGO nano fluids and nanocoated heaters.The CHF obtained with 0.01 g·L?1NCH is～15%lower than the bare Ni-Cr wire in 0.01 g·L?1concentration.The reason for lower CHF value in the case of 0.1 g·L?1NCH is due to the detachment of nanocoating from the surface when tested in DI water.On the other hand,the CHF value of 0.3 g·L?1NCH tested in DI water is～20%higher than the bare heater tested in 0.3 g·L?1concentration.This may be due to the higher thickness of nanocoating formed at0.3 g·L?1NCHand any removal of this coating during boiling does not affect the CHF[11].

Fig.14.CHF experiments with bare and nanocoated heater.

In order to understand the CHF of nanocoated wires with base fluid,the SEM image of Ni-Cr wire after boiling with nano fluid and nanocoated wire after boiling with base fluid are taken(Fig.15).The 0.1 g·L?1NCH before boiling and after boiling with base fluid is shown in Fig.15(a).It is seen that some of the coating is peeled off from the wire after boiling with base fluid,and hence,the deterioration in CHF.Fig.15(b)depicts the SEM image of 0.3 g·L?1NCH before and after boiling with DI water.Although the surfaces are modified after boiling,an enhancement of CHF with 0.3 g·L?1NCH is obtained.This is in accordance with Kwark et al.[11]who stated that any removal of coating athigher concentrations does not affect CHF.Moreover,the rGO layer acts as porous structure which may be the factor for enhancing the CHF when tested in DI water[20].Hence,these experiments show the requirement of minimum(critical)coating thickness for the enhancement of CHF.Any coating thickness which is less than the critical value leads to the deterioration in CHF,as observed in 0.01 g·L?1NCH.

Fig.15.Nanocoated wires before and after boiling with base fluid(a)0.1 g·L?1 NCH(b)0.3 g·L?1 NCH.

Experiments are also conducted with 0.3 g·L?1NCH to determine the bonding strength(reliability)of the nanocoating.Bonding strength means the ability of the nanoparticles remain stick to the heater material and to delay the occurrence of CHF.We have performed 16 pool boiling CHF experiments with 0.3 g·L?1NCH in DI water.Although the surfaces are modified during every boiling run,the thickness of rGO layer is still above the critical thickness value and hence,almost a constant value of CHF up to 10 tests.However,the deterioration in CHF is observed from 11th test onwards,and a maximum reduction in CHF of about8%is occurred between the 1stand 16th experiments.Similar to transient heat transfer studies,the thickness of rGO layer is measured after every boiling run.It is found to be ～46 μm after the 10th experiment.This is the reason for the no change in CHF until the 10th boiling run.However,the thickness shows～44μm after the 11th experiment which indicates the detachment of flakes from the Ni-Crwire,and hence,the deterioration in CHF occurs.Therefore,a value of～46 μm is the critical coating thickness which is required for the enhancement.Hence,it is suggested that the thickness of coating layer plays an important role for the enhancement of CHF of rGO nano fluids.

6.Conclusions

The pool boiling CHF investigations show a maximum enhancement of about 245%for 0.3 g·L?1concentrations.This is due to the fact that the porous structure acts as secondary cavities thereby decreasing wall superheat of Ni-Cr wire.Also,the increase in thickness of the rGO layer is proportional to the nano fluid concentrations.The experimental results well match with macrolayer dry out model.The listed below are the main findings of the present investigations:

i.)As seen from the flow visualization,there is increase in bubble density and the presence small bubbles with increasing concentrations,resulting to an enhanced CHF.Interestingly,the red hot conditions on the wire differ for each nano fluid concentration.

ii.)Transient characteristics during boiling are investigated in two methods:the first experiment is by incrementing heat input in steps and the later by increasing coating time.However,both the studies led to interesting phenomena that there is no change in CHF of rGO nano fluids.

iii.)The CHF of 0.01 g·L?1NCH is about ～15%less than bare heater tested in 0.01 g·L?1concentration while 0.3 g·L?1NCH in DI water exhibits～20%higher than the CHF of bare heater wire with 0.3 g·L?1concentration.When 0.01 g·L?1NCH is tested in DI water,some of the rGO flakes might detach and hence the deterioration in CHF is occurred.But,0.3 g·L?1NCH has thick layer,and even if some of the flakes might detach the CHF is not affected.Instead,the CHF increases due to the presence of secondary cavities formed by the rGO flakes and its uniformity.

iv.)The bonding strength of 0.3 g·L?1NCH is investigated by conducting repetitive experiments.It is found that the CHF remains fairly constant up to 10 tests and deteriorates beyond this.

Nomencleture

d diameter of the Ni-Cr wire,m

g acceleration due to gravity,m·s?2

hfglatent heat of vaporization,J·kg?1

I current,A

l length of the Ni-Cr wire,m

q heat flux,W·m?2

V voltage,V

Δ increment

δ thickness of the layer,m

ε porosity

ρ density,kg·m?3

σ surface tension,mN·m?1

τ time,s

Subscripts

CHF critical heat flux

d bubble growth

l liquid

p porous

v vapor

Z Zuber

[1]S.Nukiyama,The maximum and minimum values of the heat transmitted from metal to boiling water under atmospheric pressure,J.Jpn Soc.Mech.Eng.37(1934)367–374.

[2]R.Kamatchi,S.Venkatachalapathy,Parametric study of pool boiling heat transfer with nano fluids for the enhancement of critical heat flux:a review,Int.J.Therm.Sci.87(2015)228–240.

[3]A.Suriyawong,S.Wong wises,Nucleate pool boiling heat transfer characteristics of TiO2–water nano fluids at very low concentrations,Exp.Thermal Fluid Sci.34(2010)992–999.

[4]Ahmad Azari,Mansour Kalbasi,Masoud Derakhshandeh,Masoud Rahimi,An experimental study on nano fluids convective heat transfer through a straight tube under constant heat flux,Chin.J.Chem.Eng.21(10)(2013)1082–1088.

[5]A.Rainho Neto,J.L.G.Oliveira,J.C.Passos,Heat transfer coefficient and critical heat flux during nucleate pool boiling of water in the presence of nanoparticles of alumina,maghemite and CNTs,Appl.Therm.Eng.111(2017)1493–1506.

[6]M.Khoshvaght-Aliabadi,M.H.Akbari,F.Hormozi,An empirical study on vortexgenerator insert fitted in tubular heat exchangers with dilute Cu–water nano fluid lf ow,Chin.J.Chem.Eng.24(2016)728–736.

[7]R.Kamatchi,S.Venkatachalapathy,C.Nithya,Experimental investigation and mechanism of critical heat flux enhancement in pool boiling heat transfer with nano fluids,Heat Mass Transf.52(2016)2357–2366.

[8]M.M.Sarafraz,F.Hormozi,S.M.Peyghambarzadeh,Poolboiling heattransferto aqueous alumina nano- fluids on the plain and concentric circular micro-structured(CCM)surfaces,Exp.Thermal Fluid Sci.72(2016)125–139.

[9]Y.Haramura,Y.Katto,A new hydrodynamic model of critical heat flux,applicable widely to both pool and forced convection boiling on submerged bodies in saturated liquids,Int.J.Heat Mass Transf.26(1983)389–399.

[10]S.G.Khandlikar,A theoretical model to predict poolboiling CHF incorporating effects of contact angle and orientation,ASME J.Heat Transf.123(2001)1071–1079.

[11]S.M.Kwark,Ratan Kumar,Gilberto Moreno,Jaisuk Yoo,Seung M.You,Pool boiling characteristics of low concentration nano fluids,Int.J.Heat Mass Transf.53(2010)972–981.

[12]H.Kim,M.Kim,Experimental study of the characteristics and mechanism of pool boiling CHF enhancement using nano fluids,Heat Mass Transf.45(2009)991–998.

[13]H.S.Ahn,J.M.Kim,M.Kaviany,M.H.Kim,Pool boiling experiments in reduced graphene oxide colloids,part I–boiling characteristics,Int.J.Heat Mass Transf.74(2014)501–512.

[14]S.D.Park,S.Wonlee,S.Kang,I.C.Bang,J.H.Kim,H.S.Shin,D.W.Lee,D.Won Lee,Effects of nano fluids containing graphene/graphene-oxide nanosheets on critical heat flux,Appl.Phys.Lett.97(2010)(023103–023103-3).

[15]W.S.Hummers,R.E.Offeman,Preparation of graphite oxide,J.Am.Chem.Soc.80(1958)1339.

[16]Amartya Chakrabarti,Jun Lu,Jennifer C.Skrabutenas,Tao Xu,Zhili Xiao,John A.Maguireb,Narayan S.Hosmane,Conversion of carbon dioxide to few-layer graphene,J.Mater.Chem.21(2011)9491–9493.

[17]Yifan Dong,Shuo Li,Hongmei Xu,Mengyu Yan,Xiaoming Xu,Xiaocong Tian,Qing Liu,Liqiang Mai,Wrinkled-graphene enriched MoO3 nanobelts with increased conductivity and reduced stress for enhanced electrochemical performance,Phys.Chem.Chem.Phys.15(2013)17165–17170.

[18]N.Zuber,Hydrodynamics Aspect of Heat Transfer PhD,Thesis UCLA,1959.

[19]Han Seo,Jae Hwan Chu,Soon-Yong Kwon,In Cheol Bang,Pool boiling CHF of reduced graphene oxide,graphene,and SiC-coated surfaces under highly wettable FC-72,Int.J.Heat Mass Transf.82(2015)490–502.

[20]J.Y.Chang,S.M.You,Enhanced boiling heat transfer from microporous surfaces:effects of a coating composition and method,Int.J.Heat Mass Transf.40(1997)4449–4460.

[21]Diane Marie Vazquez,Experimental Studies of the Heat Transfer Characteristics of Silica Nanoparticle Water-based Dispersion in Pool Boiling Using Nichrome Flat Ribbon and Wires,PhD Thesis University of Central Florida,2008.