Investigating corrosion effects and heat transfer enhancementin smaller size radiators using CNT-nanofluids

W. Rashmi A. F. Ismail M. Khalid

A. Anuar T. Yusaf

Received: 13 November 2013 / Accepted: 8 March 2014 / Published online: 2 April 2014

Springer Science+Business Media New York 2014

Abstract Nanofluids have been extensively studied in the

past to enhance the heat transfer performance and effi-

ciency of systems. However, corrosion effects have been

paid very little attention and thus this work presents an

experimental study on the effect of carbon nanotubes

(CNT) on corrosion of three different metals under study

such as aluminium alloy, stainless steel and copper,

respectively. The work was further extended to study the

heat transfer performance in a car radiator of two different

sizes. Both the studies were performed using four different

fluids such as water, ethylene glycol, 0.02 % CNT-nano-

fluid and 0.1 % CNT-nanofluid, respectively. It was

observed that among the three metals, the highest rate of

corrosion occurs to aluminium, followed by stainless steel

and copper, irrespective of the fluid used. The rate of

corrosion increased with the increase in temperature

(2790 C) in all cases. The experimental results showedthat the stable CNT-nanofluids prepared in this work

showed better heat transfer performance in both engines.

Moreover, the smaller radiator using the CNT-nanofluids

depicted enhanced heat transfer rates compared to the

standard radiator using water and ethylene glycol.

Introduction

Nanofluids have attracted a large number of scientific com-

munity since their discovery by Choi [1] in 1995, due to their

enhanced heat transfer properties and large applications

ranging from chemical and process industries, electronic

cooling, cooling of heavy vehicles, solar energy harvesting

and heat transfer applications as suitable coolants [2]. Many

review papers are available in this field with regards to the

effect of various parameters on thermal conductivity

enhancement, stability and heat transfer mechanisms and its

applications [37]. Carbon nanotubes (CNT) are widely being

used due to its high thermal conductivity and unique

mechanical, electrical and optical properties, respectively [8].

Previous researchers have measured the thermal conductivity

of CNT-nanofluids dispersed in various base fluids and further

reported that the CNT-nanofluids displayed significantly

higher thermal conductivities than their base fluids [914].

In the design of automotive systems, engineers are

exploring the idea of using nanofluids as coolants to reduce the

weight and cost of the radiator. This will result in better fuel

consumption and lower car prices in near future. Nevertheless,

not many studies have been reported to check the compati-

bility of nanofluids to be used as a coolant in the engine

radiator. A good quality coolant must possess a high thermal

capacity, low viscosity, and it should be non-toxic, low cost,

chemically inert, low electrical conductivity and resists oxi-

dation. Water is a basic coolant that is commonly used as a

heat transfer fluid, due to its high thermal conductivity, cheap

and it is readily available. However, water freezing point at

W. Rashmi (&)Energy Research Group, School of Engineering, Taylors

University, Subang Jaya, Malaysia

e-mail: rashmi.walvekar@gmail.com

A. F. Ismail A. AnuarNanoscience and Nanotechnology Research Group (NANORG),

Kulliyyah of Engineering, International Islamic University,

Gombak, Kuala Lumpur, Malaysia

M. Khalid

Division of Manufacturing and Industrial Processes, Faculty of

Engineering, University of Nottingham, Semenyih,

Kuala Lumpur, Malaysia

T. Yusaf

Department of Mechanical & Mechatronics Engineering,

University of Southern Queensland, Toowoomba, Australia

123

J Mater Sci (2014) 49:45444551

DOI 10.1007/s10853-014-8154-y

0 C limits its capability to act as a pure coolant without anyaddition of ethylene glycol. In addition, most of coolants are

blended with inhibitors to prevent corrosion in the engine

radiator. In the case of a highly corroded cooling system, the

performance of the vehicle will be severely affected. The

damage occurs due to increase in engine temperature and as

the heat that cannot be transferred directly from the engine to

the coolant effectively. Furthermore, corrosion may cause the

cavitation in the radiator that which can further restrict the

flow of coolant in the cooling system.

There are also many studies reported to investigate the

effect of coolant especially ethylene glycol (EG) towards

different materials used in the cooling system. Some of the

common metals used to manufacture the cooling system are

aluminium, steel, cast iron and copper. The corrosive effects

of EG on steel and copper were investigated by Khomomi

et al. [15] and May et al. [16], whereas Liu and Cheng [17

19] and Niu and Cheng [20] studied the corrosion effects on

aluminium. All of the tests were mostly restricted to EG with

different inhibitors and concentrations. Celata et al. [21]

conducted erosion tests on aluminium, stainless steel and

copper using TiO2, Al2O3, SiC and ZrO2 nanofluids. The

results showed that the mechanical effects strongly depended

on the target material and type of nanoparticle material.

Among all, Al2O3 has proved to give a largest damage to the

gears followed by SiC, ZrO2 and TiO2 being the last. In

addition, it is also reported that the use nanofluids in radiators

can lead to a reduction in the frontal area of the radiator up to

10 % and the fuel saving up to 5 % due to the reduction in

aerodynamic drag [22].

Thus, this study aims to study the corrosion effects of

CNT-nanofluids using water and EG as base fluids. Two

different CNT concentrations have been studied, 0.02 and

0.1 wt%, respectively. Further, these nanofluids are applied

in two different size radiators to test their heat transfer

performance and efficiency.

Experimental procedure

Materials

Multiwalled CNTs were used in the present work with size

of 2030 nm OD and 510 nm ID, a length of approxi-

mately 30 lm and the purity of the CNT was greater than95 %. CNT was purchased from Sab Bayan Enterprise,

Klang Malaysia. Ethylene glycol used in this work was

purchased from Chemolab, Malaysia.

Preparation of CNT-nanofluid

Stable CNT-nanofluids of 0.02 and 0.1 wt% were prepared

by measuring the correct amount of CNT nanoparticles and

adding them to the respective measured amount of base

fluid. Optimum amount of gum Arabic (GA) dispersant

was added to stabilize the CNT in base fluid. The solution

was further homogenized using high shear homogenizer

(Fluko, Germany) at 28000 rpm for 10 min, respectively.

This was done in order to break the agglomerates and

homogenize the suspension. The solution was immediately

sonicated in a water bath sonicator at 25 C for 4 h. Duringthe sonication process, the CNTs which are initially

entangled gets separated due to the effect of ultrasonic

vibration and a thin layer of GA is coated on the CNT

surface which will prevent the agglomeration and sedi-

mentation. From, our previous studies, it is reported that

the optimum amount of GA for 0.02 wt% CNT is 1.0 wt%

and for 0.1 wt% CNT is 2.5 wt%, respectively. Further, 4 h

sonication was found to be optimum as increasing the

sonication time would damage the CNT structure and

morphology. More details on stability of CNT-nanofluids

can be found in our previously published paper [13].

Corrosion test method

Specimen preparation

Metal specimens were prepared before being used as the

working electrode (Fig. 1). Prior to testing, cutting,

grinding, polishing, washing and drying of the metal

specimens were carried out. The specimen of copper,

stainless steel and aluminium alloy plate were cut with

dimension of 1 9 1 cm2 with metal scissors. The metals

are then ground with different grades of emery paper from

100 degrees of fineness and increasing up to 800 degrees of

fineness. Since the samples are small in size, they are first

mounted together before grinding as shown in Fig. 2.

The metal pieces were mounted with copper wire con-

nected to the metal. The copper wire was used as a wire

connection of the working electrode to the circuit. The

mounted metal mould makes the grounding process easier.

Fig. 1 Copper, aluminium alloy and stainless steel in 1 9 1 cm2 size

J Mater Sci (2014) 49:45444551 4545

123

Then, the specimens were polished with aluminium oxide

on the horizontal polishing machine manually. This was

done to remove all the scratch marks on the metal surfaces

thus making the surface smooth. These mounted metals

were polished until they looked like a mirror image. The

shinier the metal surfaces, the more easy reading can be

taken. Next, the metals were washed with distilled water

twice to remove any metal residue. Finally, the specimens

were dried with a hair dryer to remove the moisture.

Polarization test

Before carrying out the polarization test, all of the equip-

ment and samples were prepared in advance and test setup

is done as shown in Fig. 3. A 50 ml solution of water is

placed in a 100-ml beaker which acts as the electrolyte.

While making sure the working electrode is fully immersed

in the electrolyte, the other two electrodes that are counter

electrode and reference electrode were also connected to

the circuit. The reference electrode that is used here is

saturated calomel electrode (SCE) and the counter elec-

trode is the platinum plate (Fig. 4). After all of the elec-

trodes are connected to the circuit system, the computer to

run the test was turned on. The constant maximum and

minimum voltage of ?2 and -2 mV were applied to the

circuit system. This is the potential limit for the circuit.

The polarization test was carried out by varying the

temperature of the nanofluid keeping the solution static

(magnetic stirrer was set off). The nanofluid was heated

using the hot plate. After it reached the desired tempera-

ture, the polarization test was conducted on the metal and

solution. During the experiment, the temperature was var-

ied from 27 to 90 C by 10 C increment. Readings weretabulated and plotted for every measurement.

Radiator test

Engine specifications and dimensions

Two different sizes of radiators were tested with a 1.5 L

Proton Wira engine at idle conditions. Tables 1 and 2 list

the dimensions of both the standard radiator used for the

engine and the smaller radiator used as its replacement.

Where L, H and W denotes the length, height and width of

the radiators, tubes and fins, and N is the number of tubes

and fins used in both radiators. While the width of both

Fig. 2 a Metal pieces connected to copper wire to be mounted together. b Mounted specimen with copper wire

Fig. 3 Polarization test setup

4546 J Mater Sci (2014) 49:45444551

123

radiators is the same, the length and the height of the

smaller radiator are approximately 47 and 9 % less than

that of standard radiator, respectively. The engine specifi-

cation is listed in Table 3.

Test procedure

Half cut car which is Proton Wira using the Magma engine

12 valves is used in this experiment as shown in Fig. 5 to

determine the heat transfer of the cooling system. Custom

made connectors or radiator hoses are connected from the

engine to the radiator. Thermocouples were attached to the

connectors by using epoxy resin to gain a strong bond that

can withstand high temperature and pressure of coolant.

Later, the thermocouple probes were connected to the

thermocouple to measure and record the coolant tempera-

ture. Initially, water as a coolant was poured into the

radiator and the engine was started. After 10 min, the

temperature readings for the coolant inlet and outlet of the

radiator were recorded with every 5 s increment until 60 s

for every 1000, 2000 and 3000 rpm. Tables 4 and 5 give

the summary of the experimental values through smaller

and standard radiators, respectively.

Strobotest was used to indicate rpm reading for every

throttle position. Finally, the coolant was flushed out from

the radiator. This procedure was repeated for other types of

coolant, which are the ethyleneglycol, CNT-nanofluids

0.1 wt% and CNT-nanofluids 0.02 wt%, respectively.

Later, similar procedure is repeated for different radiator

sizes.

Results and discussion

Corrosion rates

The corrosion rates of aluminium alloy, stainless steel and

copper, using 0.1 wt% CNT-nanofluid at different tem-

peratures are presented in Fig. 5. Tables 4, 5, and 6 present

the corrosion rates of four different coolants measured at

Fig. 4 Images of metals. a Counter electrode; platinum and nickel. b Aluminium alloy, copper and stainless steel. c Saturated calomel electrode(SCE) as the reference electrode

Table 1 Standard radiatordimensions for the 1.5 L Proton

Wira

Lradiator(mm)

Hradiator(mm)

Wradiator(mm)

Wtube(mm)

Htube(mm)

Lfin(mm)

Wfin(mm)

Hfin(mm)

Ntube Nfin

678 375 40 25.4 2.11 3.96 58.42 0.0762 45 188

Table 2 Smaller radiatordimensions used as the

replacement

Lradiator(mm)

Hradiator(mm)

Wradiator(mm)

Wtube(mm)

Htube(mm)

Lfin(mm)

Wfin(mm)

Hfin(mm)

Ntube Nfin

318 350 40 25.4 2.11 3.96 58.42 0.0762 28 168

J Mater Sci (2014) 49:45444551 4547

123

different temperatures and different metals used, respec-

tively. As observed from Fig. 6, copper shows the lowest

corrosion rate value compared with the other two metals.

The highest corrosion rate at a fixed temperature is

obtained by aluminium alloy in all of the coolants as

depicted by the results in Tables 4, 5, and 6. This can be

explained by the galvanic series for metal and alloy, where

aluminium alloy position is the nearest to the active side of

metals. For all coolants, the stainless steel corrosion rates

are between copper and aluminium alloy. The results also

demonstrate that the rate of corrosion increases with the

increase in temperature in all cases.

For water, Table 6 indicates that the corrosion rate for

aluminium alloy varies from 3.01 9 10-2 mm/year at 27 Cto 9.75 9 10-2 mm/year at 90 C. Meanwhile, the corrosionrates for stainless steel (Table 7) and copper (Table 8) vary

from 1.59 9 10-2 mm/year to 8.60 9 10-2 mm/year and

1.06 9 10-2 mm/year to 7.97 9 10-2 mm/year, respec-

tively. It is further observed that aluminium alloy has the

highest corrosion rate compared to stainless steel and copper

in both CNT-nanofluids with the value from 3.95 9

10-5 mm/year (at 27 C) to 2.54 9 10-4 mm/year (at 90 C)for 0.1 wt% CNT-nanofluid and from 9.23 9 10-5 mm/year

(at 27 C) to 6.66 9 10-3 mm/year (at 90 C) for 0.02 wt%CNT-nanofluid.

On the other hand, lower corrosion rates are produced

when EG and the CNT-nanofluids are being used as the

solution. The reason for this is the existent of inhibitor in

EG that prevents corrosion from happening. Interestingly,

the corrosion rates of metals using the CNT-nanofluid are

the lowest compared to the corrosion rates using pure water

and ethylene glycol. This could be explained on the basis

that nanofluids show lower and more stable friction coef-

ficients and they also have self-healing lubricating effects

[23]. This could be the main reason for low corrosion rate

of metals in nanofluids. It has been reported that the stable

Table 3 Engine specification

Code 4G15

Manufacturer Mitsubishi

Type S-4

SOHC

12 valves

3 valves/cylinder

Bore 9 stroke 75.5 9 82 mm2

Bore/stroke ratio 0.92

Displacement 1468 cc

Unitary capacity 367 cc/cylinder

Density ratio 9.20:1

Fuel system MPFi

Aspiration Normal

Intercooler No

Catalytic converter Yes

Maximum output 89.2 PS (88.0 bhp) (65.6 kW) @ 6000 rpm

Maximum torque 126.0 Nm (93 lbft) (12.8 kgm) @ 3000 rpm

Specific output 59.9 bhp/l

Specific torque 85.83 m/l

Fig. 5 Experimental setup with Magma engine 12 value and thermocouple probe placement

Table 4 Summary of heat transfer test for smaller radiator

Type of

coolant

rpm Mass

flow

rate, _m(kg/s)

Specific

heat, Cp(J/kg C)

Temp

difference,

DT (C)

Heat

transfer

rate,_Q (kW)

Water 1000 0.066 4191 16.217 4.485

2000 0.066 4198 16.267 4.507

3000 0.066 4203 16.308 4.524

Ethylene

glycol

1000 0.071 3583 19 4.833

2000 0.071 3583 19.025 4.840

3000 0.071 3598 22.108 5.648

CNT-

nanofluids

0.02

1000 0.142 4129 21.842 12.595

2000 0.142 4134 23.15 13.590

3000 0.142 4139 24.533 14.419

CNT-

nanofluids

0.01

1000 0.142 3851 23.017 12.586

2000 0.142 3856 24.517 13.424

3000 0.142 3860 25.5 13.977

4548 J Mater Sci (2014) 49:45444551

123

and homogeneously dispersed nanoparticles in mineral oils

are effective in reducing wear and increasing load carrying

capacity. Furthermore, the friction can be reduced between

the moving mechanical parts. The corrosion rates produced

by the 0.1 wt% CNT-nanofluid and 0.02 wt% CNT-nano-

fluid are very similar.

Another reason for the low corrosion rates of both CNT-

nanofluids under study is the presence of additive (GA) in

the solution. It has been widely known that GA is com-

monly used as an inhibitor and its function is to prevent or

slow down the corrosion process [2426]. GA is a natural

biopolymer which contains hydroxyls, aldehydes, ketones,

carboxyls, double bonds, ester, ether and other functional

groups. These functional groups impart good adhesion and

corrosion resistance performance to the substrate. As these

CNT-nanofluids uses GA as dispersant, indirectly the

inhibitor characteristics of the nanofluids are enhanced. As

a result, lower corrosion rates of metals are recorded in the

experiment using the CNT-nanofluids.

Radiator performance

Figure 7 indicates the temperature difference between the

inlet and outlet as a function of the engine speed for the

standard size radiator using the four different types of

coolants used in this work. It is observed that the temper-

ature difference increases with the increase in rpm. For all

rpms, the highest temperature difference is produced by the

0.1 wt% CNT-nanofluid followed by 0.02 % CNT-nano-

fluid, EG and water. These results prove that the CNT-

Table 5 Summary of heat transfer test for bigger radiator

Type of

coolant

rpm Mass

flow

rate, _m(kg/s)

Specific

heat, Cp(J/kg C)

Temp

difference,

DT (C)

Heat

transfer

rate, _Q(kW)

Water 1000 0.041 4198 20.442 3.518

2000 0.041 4198 21.85 3.761

3000 0.041 4198 22.708 3.908

Ethylene

glycol

1000 0.044 3583 22.417 3.534

2000 0.044 3583 23.317 3.676

3000 0.044 3622 23.708 3.778

CNT-

nanofluids

0.02

1000 0.089 4129 25.075 9.215

2000 0.089 4139 25.017 9.215

3000 0.089 4139 31.5 11.604

CNT-

nanofluids

0.01

1000 0.089 3851 26.942 9.234

2000 0.089 3856 27.092 9.307

3000 0.089 3860 32.492 11.162

Table 6 Corrosion rate of aluminium alloy (mm/year)

Temperature

(C)Water Ethylene

glycol

Nanofluid

0.1 wt%

Nanofluid

0.02 wt%

27 3.01E-02 1.85E-02 3.95E-05 9.23E-05

40 3.64E-02 2.01E-02 4.49E-05 1.36E-04

50 3.94E-02 2.39E-02 6.29E-05 1.64E-03

60 4.15E-02 3.02E-02 7.47E-05 1.74E-03

70 6.99E-02 3.72E-02 9.29E-05 4.82E-03

80 7.88E-02 3.84E-02 2.24E-04 6.09E-03

90 9.75E-02 4.15E-02 2.54E-04 6.66E-03

0

0.00005

0.0001

0.00015

0.0002

0.00025

0.0003

0 20 40 60 80 100

Corr

osio

n Ra

te (m

m/ye

ar)

Temperature (C)

CopperStainless StealAluminium Alloy

Fig. 6 Corrosion rates of metals using 0.1 wt% CNT-nanofluid

Table 7 Corrosion rate of stainless steel (mm/year)

Temperature

(C)Water Ethylene

glycol

Nanofluid

0.1 wt%

Nanofluid

0.02 wt%

27 1.59E-02 1.42E-02 2.48E-05 3.32E-05

40 2.01E-02 1.76E-02 3.53E-05 3.59E-05

50 2.03E-02 1.94E-02 3.66E-05 4.85E-05

60 3.68E-02 2.47E-02 5.14E-05 7.11E-05

70 3.94E-02 2.91E-02 7.81E-05 7.66E-05

80 8.31E-02 3.90E-02 1.03E-04 9.73E-05

90 8.60E-02 4.03E-02 2.46E-04 1.41E-04

Table 8 Corrosion rate of copper (mm/year)

Temperature

(C)Water Ethylene

glycol

Nanofluid

0.1 wt%

Nanofluid

0.02 wt%

27 1.06E-02 1.46E-03 8.54E-06 4.56E-05

40 1.30E-02 1.57E-03 1.03E-05 5.64E-05

50 1.57E-02 1.59E-03 1.07E-05 5.94E-05

60 2.17E-02 4.18E-03 1.13E-05 7.10E-05

70 2.89E-02 2.01E-02 1.16E-05 7.10E-05

80 3.72E-02 3.17E-02 1.35E-05 9.16E-05

90 7.97E-02 3.81E-02 1.43E-05 1.01E-04

J Mater Sci (2014) 49:45444551 4549

123

nanofluids have greater thermal conductivity that can

absorb and transfer more heat to the surroundings.

The temperature difference between the inlet and outlet

temperatures as a function of the engine speed for the

smaller radiator is shown in Fig. 8. The plots reveal similar

trends depicted in Fig. 7. Comparing the values of the

results in Figs. 7 and 8, it can be seen that the temperature

difference (between 22 and 25 C) obtained using CNT-nanofluids in the smaller radiator match the temperature

difference obtained using EG or water in the standard size

radiator.

Figures 9 and 10 demonstrate the heat transfer rates as a

function of the engine speed rpm for the standard radiator

and the smaller radiator, respectively. For both radiators,

the heat transfer rate for water and ethyleneglycol dem-

onstrates almost similar values for each rpm. The two

different CNT-nanofluids used in this study show similar

heat transfer rates which are higher than those of water and

ethylene glycol. This is expected due to the higher thermal

properties of the CNT-nanofluids and the results suggest

that 0.02 wt% CNT concentration is adequate to enhance

the heat transfer rates. Overall, the heat transfer rates

increase with the increase of the engine speeds, since more

heat needs to be dissipated at higher rpm.

Comparing the results in Figs. 9 and 10, the heat transfer

rates produced by the two CNT-nanofluids in the smaller

radiator are much higher that the rates obtained by either

water or EG in the standard radiator.

Conclusion

Corrosion rates using three different metals and heat

transfer studies in two different size radiators have been

performed using water, EG and two different CNT-water

nanofluids. The results indicated that the corrosion rates of

10

15

20

25

30

35

0 500 1000 1500 2000 2500 3000 3500

Tem

pera

ture

Diff

eren

ce (

C)

RPM

WaterEthylene Glycol0.02% CNT-Nanofluid0.1%CNT-Nanofluid

Fig. 7 Temperature difference between inlet and outlet temperaturesfor the standard radiator

10

15

20

25

30

35

0 500 1000 1500 2000 2500 3000 3500

Tem

pera

ture

Diff

eren

ce (

C)

RPM

WaterEthylene Glycol0.02% CNT-nanofluid0.1% CNT-nanofluid

Fig. 8 Temperature difference between inlet and outlet temperaturesfor the smaller radiator

0

2

4

6

8

10

12

14

0 500 1000 1500 2000 2500 3000 3500

Hea

t Tra

nsfe

r Rat

e (kW

)

RPM

WaterEthylene Glycol0.02%CNT Nanofluid0.1%CNT Nanofluid

Fig. 9 Heat transfer rates for the standard radiator

0

2

4

6

8

10

12

14

16

0 500 1000 1500 2000 2500 3000 3500

Hea

t Tra

nsfe

r Rat

e (kW

)

RPM

WaterEthylene Glycol0.02%CNT Nanofluid0.1%CNT Nanofluid

Fig. 10 Heat transfer rates for the smaller radiator

4550 J Mater Sci (2014) 49:45444551

123

the three metals are found to be lower using CNT-nanofl-

uids under study. Aluminium alloy showed the highest

corrosion rate compared to stainless steel and copper in

both sets of CNT-nanofluids demonstrating their self-

healing lubricating properties. Moreover, the addition of

GA not only enhanced the stability of the CNT-nanofluid

but also improved the corrosion resistance. The results on

heat transfer rates and the temperature difference using

CNT-nanofluids in the smaller radiator were found to be

similar or higher than those using water or EG in the

standard radiator. Thus, proving that the smaller radiator

can be deployed using the nanofluids in the car cooling

system. This will further lead to improved system that will

provide better engine performance as well as lower fuel

consumption in the automotive industry. The results of this

study should promote further investigations on the opti-

mum operating process conditions using the CNT-

nanofluids.

References

1. Choi SUS (1995) Enhancing thermal conductivity of fluids with

nanoparticles. In: Siginer DA, Wang HP (eds) Developments and

applications of non-newtonian flows. ASME, New York,

pp 99105

2. Taylor R, Coulombe S, Otanicar T et al (2013) Small particles,

big impacts: a review of the diverse applications of nanofluids.

J Appl Phys 113:011301011319

3. Fan J, Wang L (2011) Review of heat conduction in nanofluids.

J Heat Transfer 133:040801-1040801-14

4. Yu W, Xie H (2012) A review on nanofluids: preparation, sta-

bility mechanisms, and applications. J Nano Mater. doi:10.1155/

2012/435873

5. Wang XQ, Mujumdar AS (2008) A review on nanofluidspart I:

theoretical and numerical investigations. Braz J Chem Eng

25:613630

6. Kakac S, Pramuanjaroenkij A (2009) Review of convective heat

transfer enhancement with nanofluids. Int J Heat Mass Transfer

52:31873196

7. Lee JH, Lee SH, Choi J, Jang SP, Choi SUS (2010) A review of

thermal conductivity data, mechanics and models for nanofluids.

Int J Micro-Nano Scale Transp 1:269322

8. Dresselhaus MS, Dresselhaus G, Charlier JC, Ernandez EH

(2004) Thermal and mechanical properties of carbon nanotubes.

Philos Trans R Soc Lond A 362:20652098

9. Ding Y, Alias H, Wen D, Williams RA (2006) Heat transfer of

aqueous suspensions of carbon nanotubes (CNT nanofluids). Int J

Heat Mass Transfer 49:240250

10. Rashmi W, Ismail AF, Khalid M (2012) Thermal conductivity of

carbon nanotube nanofluid-Experimental and theoretical study.

Heat Tran Asian Res 41:145163

11. Xie H, Lee H, Youn W, Choi M (2003) Nanofluids containing

multiwalled carbon nanotubes and their enhanced thermal con-

ductivities. J Appl Phys 94:49674971

12. Rashmi W, Khalid M, Ismail AF, Saidur R, Rasheed AK (2013)

Experimental and numerical investigation of heat transfer in CNT

nanofluids. J Exp Nanosci. doi:10.1080/17458080.2013.848296

13. Rashmi W, Ismail AF, Sopyan I, Jameel AT, Yusof F, Khalid M

(2011) Stability and thermal conductivity enhancement of carbon

nanotube nanofluid using gum Arabic. J Exp Nanosci 6:567579

14. Liu MS, Lin MCC, Huang IT, Wang CC (2005) Enhancement of

thermal conductivity with carbon nanotube for nanofluids. Int

Commun Heat Mass 32:12021210

15. Khomami MN, Danaee I, Attar AA, Peykari M (2012) Effects of

NO2- and NO3

- ions on corrosion of AISI 4130 steel in ethylene

glycol?water electrolyte. Trans Indian Inst Met 65:303311

16. May PM, Ritchie IM, Tan ET (1991) The corrosion of copper in

ethylene glycolwater mixtures containing chloride ions. J Appl

Electrochem 21:358364

17. Liu Y, Cheng YF (2009) Cathodic reaction kinetics and its

implication on flow-assisted corrosion of aluminum alloy in

aqueous ethylene glycol solution. J Appl Electrochem

39:12671272

18. Liu Y, Cheng YF (2011) Characterization of passivity and pitting

corrosion of 3003 aluminum alloy in ethylene glycolwater

solutions. J Appl Electrochem 41:151159

19. Liu Y, Cheng YF (2011) Inhibition of corrosion of 3003 alumi-

num alloy in ethylene glycolwater solution. J Mater Eng Per-

form 20:271275

20. Niu L, Cheng YF (2007) Electrochemical characterization of

metastable pitting of 3003 aluminum alloy in ethylene glycol

water solution. J Mater Sci 42:86138617. doi:10.1007/s10853-

007-1841-1

21. Celata GP, DAnnibale F, Mariani A (2011) Nanofluid flow

effects on metal surfaces. Energ Ambiente e Innovazione

45:9498

22. Singh D, Toutbort J, Chen G, (2006) Heavy vehicle systems

optimization merit review and peer evaluation. Annual Report,

Argonne National Laboratory, USA

23. Sahoo RR, Bhattacharjee S, Das T (2013) Development of

nanofluids as lubricant to study friction and wear behavior of

stainless steels. Int J Mod Phys Conf Ser. doi:10.1142/

S2010194513010829

24. Arthur DE, Jonathan A, Ameh PO, Anya C (2013) A review on

the assessment of polymeric materials used as corrosion inhibitor

of metals and alloys. Int J Ind Chem. doi:10.1186/2228-5547-4-2

25. Kesavan D, Gopiraman M, Sulochana N (2012) Green inhibitors

for corrosion of metals: a review. Chem Sci Rev Lett 1:18

26. Rani BEA, Basu BBJ (2012) Green inhibitors for corrosion

protection of metals and alloys: an overview. Int J Corros. doi:10.

1155/2012/380217

J Mater Sci (2014) 49:45444551 4551

123

Investigating corrosion effects and heat transfer enhancement in smaller size radiators using CNT-nanofluidsAbstractIntroductionExperimental procedureMaterialsPreparation of CNT-nanofluidCorrosion test methodSpecimen preparationPolarization test

Radiator testEngine specifications and dimensionsTest procedure

Results and discussionCorrosion ratesRadiator performance

ConclusionReferences