Influence of Tool Shoulder Diameter on Mechanical ...

2 downloads 0 Views 16MB Size Report
[19] Li, X., Tabil, L.G., Panigrahi, S. and Crerar, W.J. (2009),. “The Influence of Fiber Content on ...... John Wiley and Sons, p. 275. [8] Gupta, V. and Murthy, P.N. ...
Contents 1. Influence of Tool Shoulder Diameter on Mechanical Properties of Friction Stir Welded Dissimilar Aluminium Alloys 2014 and 6082 Ashwani Kumar, Pardeep Kumar and Balwinder S. Sidhu

5

2. Improving Wear Behaviour of GIHC Grade of Grey Cast Iron by Using WC-12CO and Stellite-6 Coatings Baljit Singh

11

3. Mechanical Characterization of Polypropylene (PP) and Polyethylene (PE) Based Natural Fiber Reinforced Composites Deepak Varshney, Kishore Debnath and Inderdeep Singh

16

4. Optimization of Process Parameters of Gas Metal ARC Welding by Taguchi’s Experimental Design Method Deepak Kumar and Sandeep Jindal

24

5. Nano Science and Technology through Friction Stir Processing and Bulk Metallic Glass Routes Part II: Bulk Metallic Glass Brij K. Dhindaw, Harpreet Arora, J. Eckert, Nilam Barekar and Sundeep Mukherjee

28

6. Jet Impingement Heat Transfer: Stationary Disc Gus Nasif, Ron Barron and Ram Balachandar

34

7. Development of High Temperature Coatings for Wear and Oxidation Resistance Using Cold Spray and HVOF Coatings A.S. Khanna and W.S. Rathore

39

8. Surface Engineering Analysis of HVOF Sprayed Cr3C2-NiCr Coating under High-Temperature Oxidation V.N. Shukla, R. Jayaganthan and V.K. Tewari

44

9. Investigation on Multi-Response Parameter Optimization of Cold Spray Coating Process Tarun Goyal, R.S. Walia and T.S. Sidhu

50

AUTHOR INDEX

56

Influence of Tool Shoulder Diameter on Mechanical Properties of Friction Stir Welded Dissimilar Aluminium Alloys 2014 and 6082 Ashwani Kumar1*, Pardeep Kumar2 and Balwinder S. Sidhu3 1,3

Mechanical Engineering, Giani Zail Singh Punjab Technical University Campus, Bathinda, Punjab–151001, India 2 Mechanical Engineering, Yadavindra College of Engineering, P.B.I. Univ. G.K. Campus, Talwandi Sabo, Bathinda, Punjab–151302, India E-mail: *[email protected]

Abstract—Friction Stir Welding (FSW), a new solid state welding technology has created a worldwide attraction especially in automobile and aerospace industries as conventional fusion welding techniques are susceptible to various welding defects like cracks, voids, porosity etc. Welding of dissimilar aluminium alloys exhibits poor weldability by fusion welding processes. It’s a matter of concern, because of thin oxide layer formation on the surface of aluminium alloys as this oxide layer tends to thicken at higher temperatures. As compared to many other fusion welding techniques those are used for joining various alloys in industries, FSW is a new technique in which the material to be welded does not actually melt. FSW is used for applications, where the original metal characteristics must remain unchanged as far as possible. 2000 series and 6000 series of aluminium alloys in combination are highly used in aerospace and automobile industry because of their mechanical properties, corrosion resistant properties and good-strength to weight ratio. A lot of work had been done to improve the quality and strength of friction stir welded joints by changing welding process parameters. Tool shoulder diameter also has a great impact on the strength and quality of friction stir welded joints. The present research investigates the influence of tool shoulders of different diameters on the mechanical properties of friction stir welded dissimilar aluminium alloys 2014 and 6082. Aluminium alloys 2014 and 6082 were welded successfully by using three different tool shoulder diameters. From the results, it was clear that the mechanical properties of welded joints are influenced by tool shoulder diameter. The joint fabricated by using 18 mm tool shoulder diameter given superior tensile properties and micro-hardness among all tested. Keywords: Friction Stir Welding (FSW), Mechanical Properties, Dissimilar Aluminium Alloys, Tool Shoulder Diameter

INTRODUCTION FSW was invented in UK in December, 1991 at The Welding Institute (TWI) [1]. FSW is a new solid state welding technology that has created a worldwide attraction especially in automobile and aerospace industries as compared to conventional fusion welding techniques for the welding of light weight alloys. For reducing the weight of automobiles the use of lightweight material is most effective [2]. A lot of research has been going on to make a possible use of aluminium alloys for body work and structural components as compared to steel. In past some year’s, due to their wide range of properties aluminium alloys are growing as a main element that is used for engineering applications. Due to their mechanical properties, goodstrength to weight ratio and anti-corrosion properties aluminiumalloys are now highly used in aerospace and automobile industry. It is big challenge to weld aluminium alloys with the help of conventional fusion welding processes. Conventional fusion welding techniques are susceptible to various welding defects like cracks, voids, porosity etc. It’s a matter of concern, because of thin oxide layer formation, high thermal conductivity, high coefficient of thermal expansion, solidification shrinkage and high solubility of hydrogen and other gases in molten state [3]. The fusion welding of aluminium alloys leads to the melting and re-solidification of the fusion zone which results in the formation of brittle inter-dendritic structure and eutectic phases. The formation of brittle structure in

the weld zone leads to the drastic decrease in the mechanical properties like lower hardness, strength and ductility [4, 5]. A solid state joining technique FSW is greatly recommended to solve these problems. FSW process overcomes these defects as in this process joining of materials take place in solid state. Also with FSW process, dissimilar aluminium alloys give superior tensile properties [6]. In fact, the undesirable low mechanical properties resulting from melting and re-solidification are absent [7]. Friction stir welded joints lead to improved mechanical properties, such as ductility and strength in some alloys [3, 8, 9]. FSW is used for applications where the original metal characteristics must remain unchanged as far as possible. FSW is an emerging solid state joining process in which the material that is being welded does not melt and recast [10]. The dissimilar aluminium alloys 2xxx series and 6xxx series exhibits poor weld ability by fusion welding process but were successfully welded with FSW [11]. The process and the terminology of FSW are schematically explained in Fig. 1. In FSW thermo-mechanical joining of the two materials takes place as the materials are joined due mechanical deformation caused by the heat produced due to friction. A non-consumable rotating shouldered pin tool is plunged into interface between two plates being welded, until the shoulder touches the surface of the base material and is the transferred along the weld line [13]. The shoulder is pressed against the surface of the materials being welded, while the probe or pin is forced between the two components by a downward force. The rotation of the tool

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

5

Kumar, Kumar and Sidhu

Influence of Tool Shoulder Diameter on Mechanical Properties of Friction Stir Welded

changing the welding process parameters. Tool shoulder diameter also has a great impact on the strength and quality of friction stir welded joints. The present research paper investigates the influence of different tool shoulder diameters on the mechanical properties of friction stir welded dissimilar aluminium alloys 2014 and 6082. FSW tool parameters are shown in Fig. 2.

EXPERIMENTAL WORK

Fig. 1: FSW Process [12]

under this force generates a frictional heat that decreases the resistance to plastic deformation of the material [14]. Upon reaching the end of the weld, the tool was withdrawn, while it was still being rotated [15]. Benefits of FSW as compared to fusion processes are low distortion, excellent mechanical properties in the weld zone, execution without a shielding gas and suitability to weld all aluminium alloys [16]. The microstructure evolution and the resulting mechanical properties depend strongly on the variation of the processing parameters leading to a wide range of possible performances [17]. FSW parameters such as tool transverse speed, tool rotation speed, tool tilt angle and tool geometry (tool pin profile, tool shoulder diameter, tool pin length and tool pin diameter) plays a major role in deciding the quality of the weld. Lot of work had already been done to improve the strength and quality of friction stir welded joints by

The base materials used in this study were aluminium alloys 2014 and 6082. 6 mm thick rolled plates of both the aluminium alloys were cut in the required sizes at slow speed in order to avoid any temperature rise, so as to avoid any grain structure changes in the materials. The size of rectangular pieces to be welded was 100 mm × 70 mm × 6 mm. Square butt joint configuration (100 mm × 140 mm) was prepared by welding the samples. The chemical compositions of both the alloys are presented in Table 1. FSW of the samples was done by using single pass welding process. Direction of welding was perpendicular to the rolling direction of the samples. Table 1: Chemical Composition of Work Material (wt.%). Al Alloy 2014 6082

Mn 0.44 0.58

Cu 2.87 0.028

Mg 0.37 0.47

Si 0.74 1.23

Fe 0.084 0.13

Zn 0.048 0.030

Al 95.5 97.5

Table 2: Chemical Composition of Tool Material (wt.%) Tool C Cr S P Mn Si Fe HCHCr 2.2 11.5 0.04 0.02 0.45 0.51 Balance

Non-consumable tools of High carbon high chromium steel were used for this FSW process. Table 2 shows the chemical composition of the tool material. Tools used in this study have pentagonal pin profile with three different tool shoulder diameters. Detailed tool specifications are presented in Table 3. As reported elsewhere [12], the pentagon pin profiled tool gives superior mechanical properties as compared to square and hexagon pin profiled tools, when the same dissimilar materials, aluminium alloys 2014 and 6082, welded with FSW. The tool rotation speed and welding speed (tool transverse speed) were kept constant throughout the process. Axial load and tool tilt angle were also kept constant. Welding Speed

= 120 mm/ min

Tool Rotation Speed = 1600 RPM Tool Tilt Angle

= 0

Table 3: FSW Tool Tool Length Tool Shoulder Diameter (D) Tool Pin Diameter (d) Tool Pin Length (L) Tool Pin Profile Fig. 2: FSW Tool Dimensions [12]

6

Tool Tilt Angle

Specifications 80 mm 15 mm, 18 mm, 21 mm 6 mm 5.7 mm Pentagon 0

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Influence of Tool Shoulder Diameter on Mechanical Properties of Friction Stir Welded

Kumar, Kumar and Sidhu

RESULTS Visual Inspection

Vertical milling machine with CNC was used for fabricating the friction stir welded joints. The samples to be welded were held in position on the table of CNC vertical milling machine by using a specially designed fixture. The fixture had dimensions 200 mm  200 mm and was made up of mild steel plate of 20 mm thickness. A pocket of depth 3 mm was cut at the center of the fixture plate for holding samples. After FSW, the samples were sliced at slow speed to the required dimensions in order to prepare tensile specimens as shown in Fig. 3. ASTM standard E8 / E8M [18] guidelines were followed for preparing the tensile test specimens. Tensile test was performed to evaluate the mechanical properties of the joints. By using universal testing machine having maximum load capacity of 25 KN, ultimate tensile strength of the specimens was recorded. To reveal the internal defects like porosity, voids and cracks etc. in the FSW zone, X-ray radiography of the samples was done after welding. The specimens for micro-structure analysis were cut to the required sizes at slow speed. The metallographic samples consisting FSW zone, Thermo Mechanically Affected Zone (TMAZ), Heat Affected Zone (HAZ) and Base Metal region were cut. Samples were first polished by using emery papers of different grades (240, 320, 600, 800, 1000, and 1500). Final polishing of the samples was done by using a double disc polishing machine with the use of diamond paste on velvet cloth. Specimens were then etched with Hydrofluoric (HF) acid to reveal the micro structures. Metallographic specimens were prepared by following ASTM standard E3-11 [19] guidelines. Micro-structural analysis was performed by using an inverted optical microscope (Radical, India make) at a magnification of 200X. Vicker’s micro-hardness test was done to record the micro-hardness in the weld zone (stirred zone). Vickers micro-hardness tester VHS 1000 A (Banbros, India make) was used for this test. The micro-hardness testing was performed at a load of 100gf for a dwell time of 10s.ASTM standard E384 [20] guidelines were followed for micro-hardness testing of the specimens.

Pin-hole defect was produced when welding tool was removed from the welded zone. Tool leaves a hole in the weld when removed, which is called pin hole. This pin hole was produced at the end of all the joints, fabricated at various combinations of parameters. This pin hole can be filled later on with the help of TIG welding.

Tensile Properties After FSW all the specimens were tested to find out their ultimate tensile strength. Three specimens were tested for each combination of parameter and their average values were considered. All the results were plotted in the form of a bar chart ultimate tensile strength v/s tool shoulder diameter as shown in Fig. 4. From the results of tensile test it can be studied that the ultimate tensile strength of the ##### 66

Tensile Strength in MPa

Fig. 3: Drawing of Tensile Test Specimen (all Dimensions in mm)

The friction stir welded samples were first visually inspected. Almost, all the joints welded with FSW were produced with a smooth surface finish. Defects like porosity, slag, cracks etc. are formed in most of the fusion welded joints as a results of melting of the material which affects the quality of the weld. Friction stir welded joints were free from these types of defects as the solidification takes place in solid state. Sample welded with 15 mm shoulder diameter tool produced avisual defect may be infilling of material on the top surface of the joint. It was also clear from the X-ray radiography of the welded samples that there is a little bit porosity in the sample welded with 15 mm shoulder diameter tool. The defect produced was on the advancing side of the weld. The material could not flow properly from retarding side to the advancing side of the weld when unusual welding parameters were used. From visual inspection and X-ray radiography it was clear that the formation of welding defect is a function of welding tool shoulder diameter.

66 64 62 60 58 56 54 52 50

58.9 56.04

15 mm

18 mm

21 mm

Tool Shoulder Diameter

Fig. 4: Effect of Tool Pin Profile on Ultimate Tensile Strength of FSW Joints

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

7

Kumar, Kumar and Sidhu

Influence of Tool Shoulder Diameter on Mechanical Properties of Friction Stir Welded

friction stir welded joint has been influenced by the tool shoulder diameter of the welding tool. The joint produced by using 18 mmshoulder diameter tool has given superior tensile properties as compared to 15 mm and 21 mm shoulder diameter tool.

of 149Hv. The hardness of the stirred zone for all the friction stir welded samples was considerably higher than both of the base metals irrespective of the tool shoulder diameters used.

The factors which determine the tensile strength of dissimilar aluminium alloy joints are: (i) presence of macroscopic defects in weld zone (ii) degree of plastic flow and amount of mixing of both the materials [21].

DISCUSSION

Micro-Hardness From the tensile test, it was clear that all the specimens were failing in the FSW region. The accurate location of the joint failure was either on the advancing side or on the retarding side of the weld region. Hence, micro-hardness analysis of the joints was carried out in the FS weld zone. The values of micro-hardness were plotted in the form of a bar graph and are presented in Fig. 5. 166

170 160

Microhardness (Hv)

150

154 149

140 130 120 110 100 15 mm

18 mm

21 mm

Tool Shoulder Diameter Fig. 5: Effect of Tool Pin Profile on the Micro-Hardness of Stir Zone

Two main reasons are responsible for the hardness improvement in the stirred zone [22]. 1.

2.

The size of the grains present in the welding zone, if the grains in the welding zone are finer then the base metal that plays an important role to provide strength in the material. According to the Hall-Petch equation, hardness increases as the grain size decreases. The small particles of inter-metallic compounds are also a benefit to hardness improvement according to the Orowan hardening mechanism [23].

From the results, it can be clearly studied that 18 mm shoulder diameter tool provides the highest hardness value of 166 Hv. 21 mm shoulder diameter tool provides the second highest hardness value of 154Hv. 15 mm shoulder diameter tool provides the lowest hardness value 8

The tool shoulder diameter has directly proportional relationship with the heat generation due to friction [24]. Lima et al. [25] and Oosterkamp et al. [26] had reported that at the top surface of the FSP region, a material transport occurred due to the action of the rotating tool shoulder. Material near the top of the FSP region, approximately the upper one-third, moves under the influence of the shoulder rather than the profiles on the pin. The tool pin generates the heat and stirs the material being welded but the tool shoulder also plays an important role by providing additional frictional treatment as well as preventing escapement of the plasticized material from the weld region. The friction between the shoulder and work piece results as the biggest component of heating. From the heating aspect, the relative size of pin and shoulder is important [22]. Hence, tool shoulder diameter plays a crucial role in heat generation and material flow in the FSW process. If the tool shoulder diameter is small, it will produce less amount of heat due to small contact area (less amount of frictional heat generation) between tool shoulder and work material and vice versa. In the present study, from the results of tensile testing it had been found that the tools with a large shoulder diameter of 21 mm leads to have a wider contact area and produces large amount of frictional heat, subsequently wider TMAZ and HAZ region and then resulted in the deterioration of tensile strength. The akin results are also reported by Rajakumar et al. [22]. It had also been observed that the tools with small shoulder diameter of 15 mm leads to have a small contact area and produces less amount of frictional heat and hence the proper plasticized flow of metal was not happened. That’s why the consolidation of the metal in the weld joint was not good and hence resulted in poor tensile properties, also reported by [27]. On the other hand, the tools with 18 mm shoulder diameter produced sufficient amount of frictional heat due to large contact area, that’s why proper consolidation of metal was there in the weld joint and as a result superior tensile properties are produced. Hence, the tool shoulder diameter should be optimum neither too large nor too small. As the effect of tool shoulder diameter on FSW joints is concerned, the joints produced by using tools with shoulder diameter 18 mm (D/d = 3) have shown superior ultimate tensile strength as compared to other joints [27]. It was also clear from the X-ray radiography which shows a little bit porosity in the joint fabricated with 15 mm shoulder diameter tool.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Influence of Tool Shoulder Diameter on Mechanical Properties of Friction Stir Welded

For comparison purpose, micro-hardness values and microstructure images of all the joints welded with different shoulder diameter tools from the FSW region (stirred zone) are presented in Fig. 5 and Fig. 6, respectively.

Kumar, Kumar and Sidhu

CONCLUSION In the above investigation an attempt has been made tostudy the influence of tool shoulder diameter on the mechanical properties of friction stir welded dissimilar aluminium alloys 2014 and 6082. It has been found that the tool shoulder diameter has a great impact on the strength and quality of friction stir welded joints. 1.

Among all tested joints, the joint fabricated by using 18 mm shoulder diameter tool (D/d = 3) provided superior mechanical properties.

2.

Sufficient amount of frictional heat generation in the stirred zone and proper plasticized flow of the material produced defect free welds.

50 m (a)

REFERENCES [1]

50 m (b)

50 m

(c) Fig. 6: Micrographs of Stir Zone of FSW Samples Produced by Using Different Shoulder Diameter Tools (a) 15 mm (b) 18 mm (c) 21 mm

It was observed from the results that the 18 mm shoulder diameter tool provided higher value of micro-hardness as compared to other joints produced by 15 mm and 21 mm shoulder diameter tools. Joints fabricated using smaller shoulder diameter tool resulted in small contact area, which caused defects in FSW zone and clustering of precipitates in the weld zone, which resulted in lower values of micro-hardness. On the other hand joints fabricated by using larger shoulder diameter tool resulted in large contact area, which consisted of coarse grains but the strengthening precipitates were uniformly distributed throughout the matrix. The value of micro-hardness in the nugget zone or stirred zone were coherent with the grain size, a higher value of micro-hardness was due to a finer structure [28]. The grain structure produced by 18 mm shoulder diameter tool contained very fine equiaxed grains as observed in Fig. 6 (b). This may be the reason for the higher values of micro-hardness at these parameters.

Thomas, W.M., Nicholas, J., Needham, J.C., Murch, M.G., Temple-smith, P. and Dawes, C.J. (1995), Friction Stir Welding, GB Patent Application No. 9125978.8, December 1991, US Patent, No. 5460317. [2] Yan, J., Zhiwu, Xu., Li, Z., Lei, Li. and Yang, S., “Microstructure Characteristics and Performance of Dissimilar Welds between Magnesium Alloy and Aluminium Formed by Friction Stirring”, Scripta Materialia, Vol. 53, pp. 585–589. [3] Matrukanitz, R.P. (1990), “Selection of Weldability of HeatTreatable Aluminium Alloys”, ASM Handbook-Welding, Brazing and Soldering, Vol. 6, pp. 528–536. [4] Su, J.Q., Nelson, T.W., Mishra, R. and Mahoney, M. (2003), “Microstructural Investigation of Friction Stir Welded 7050-T651 Aluminium”, Acta Materialia, Vol. 51, pp. 713–729. [5] Rhodes, C.G., Mahoney, M.W. and Bingel, W.H. (1997), “Effects of Friction Stir Welding on Microstructure of 7075 Aluminium”, Scripta Materialia, Vol. 36, pp. 69–75. [6] Shanmuga Sundaram, N. and Murugan, N. (2010), “Tensile behavior of Dissimilar Friction Stir Welded Joints of Aluminium Alloys”, Materials and Design, Vol. 31, pp. 4184–4193. [7] Da Silva, A.A.M., Arruti, E., Janeiro, G., Aldanondo, E., Alvarez, P. and Echeverria, A. (2011), “Material Flow and Mechanical behaviour of Dissimilar AA2024-T3 and AA7075-T6 Aluminium alloys Friction Stir Welds”, Materials and Design, Vol. 32, pp. 2021–2027. [8] Sato, Y.S., Urata, M., Kokawa, H. and Ikeda, K. (2003), “Hall-/Petch Relationship in Friction Stir Welds of Equal Channel Angular-pressed Aluminium Alloys”, Material Science and Engineering, Vol. 354, pp. 298–305. [9] Berbon, P.B., Bingel, W.H., Mishra, R.S., Bampton, C.C. and Mahoney, M.W. (2001), “Friction Stir Processing: A Tool to Homogenize Nanocomposites Aluminium Alloys”, Scripta Materiala, Vol. 44, pp. 61–66. [10] Thomas (1991), “Friction Stir Welding”, International Patent Application No. PCT/GB92/02203 and GB Patent Application No. 9125978.8”, December 1991, U.S Patent No. 5,460,317. [11] Amancio-Filho, S.T., Sheikhi, S., Dos Santos, J.F. and Bolfarini, C. (2008), “Preliminary Study on the Microstructure and Mechanical Properties of Dissimilar Friction Stir Welds in Aircraft Aluminium Alloys 2024-T351 and 6056-T4”, Journal of Material Processing Technology, Vol. 206, pp. 132–142. [12] Kumar, A., Kumar, P. and Sidhu, B.S. (2013), “Influence of Tool Pin Profile on the Mechanical Properties of Joint of Aluminium Alloys 2014 and 6082, Welded with Friction Stir Welding”, Asian Review of Mechanical Engineering, Vol. 2, No. 2, pp. 66–71.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

9

Kumar, Kumar and Sidhu

Influence of Tool Shoulder Diameter on Mechanical Properties of Friction Stir Welded

[13] Kumar, K. and Kailas Satish, V. (2008), “On the Role of Axial Load and the Effect of Interface Position on the Tensile Strength of a Friction Stir Welded Aluminium Alloy”, Materials and Design, Vol. 29, pp. 791–797. [14] Scialpi, A., De Giorgi, M., De Filippis, L.A.S., Nobile, R. and Panella, F.W. (2008), “Mechanical Analysis of Ultra Thin Friction Stir Welding Joined Sheets with Dissimilar and Similar Material”, Materials and Design, Vol. 29, pp. 928–936. [15] Khaled Terry (2005), “An Outsider Looks at Friction Stir Welding”, Report: ANM-112N-05-06. [16] Moreira, P.M.G.P., Santos, T., Tavares, S.M.O., Richter-Trummer, V., Vilaca, P. and De Castro, P.M.S.T. (2009), “Mechanical and Metallurgical Characterization of Friction Stirs Welding Joints of AA6061-T6 with AA6082-T6”, Materials and Design, Vol. 30, pp. 180–187. [17] Cavaliere, P., Campanile, G., Panella, F. and Squillace, A. (2006), “Effect of Welding Parameters on Mechanical and Microstructural Properties of AA6056 Joints Produced by Friction Stir Welding”, Journal of Materials Processing Technology, Vol. 180, pp. 263–270. [18] ASTM Standard E8 / E8M (2013a), “Standard Test Methods for Tension Testing of Metallic Materials”, ASTM International, West Conshohocken, PA. [19] ASTM Standard E8-11(2011), “Standard Guide for Preparation of Metallographic Specimens”, ASTM International, West Conshohocken, PA. [20] ASTM Standard E384 (1999), “Standard Test Methods for Microindentation Hardness of Materials”, ASTM International, West Conshohocken, PA.

10

[21] Palanivel, R., Koshy Mathews, P., Murugan, N. and Dinaharan, I. (2012), “Effect of Tool Rotational Speed and Pin Profile on Microstructure and Tensile Strength of Dissimilar Friction Stir Welded AA5083-H111 and AA6351-T6 Aluminium Alloys”, Materials and Design, Vol. 40, pp. 7–16. [22] Rajakumar, S., Muralidharan, C. and Balasubramanian, V. (2011), “Influence of Friction Stir Welding Process and Tool Parameters on Strength Properties of AA7075-T6 Aluminium Alloy Joints”, Materials and Design, Vol. 32, pp. 535–549. [23] Wang, X.H. and Wang, K.S. (2006), “Microstructure and Properties of Friction Stir Butt-weldedAZ31 Magnesium Alloy”, Material Science and Engineering, Vol. 431, pp. 114–117. [24] Thomas, W.M. and Nicholas, E.D. (1997), “Friction Stir Welding for the Transportation Industries”, Materials and Design, Vol. 18, pp. 269–273. [25] Lima, E.B.F., Wegener, J., Dalle Donne, Goerigk, G., Wroblewski, T. and Buslaps, T. (2003), “Dependence of the Microstructure, Residual Stresses and Texture of AA6013 Friction Stir Welds on the Welding Processes”, ZMetallkd, Vol. 94(8), pp. 908–915. [26] Oosterkamp, A., Djapic Oosterkamp, L. and Nordeide, A. (2004), “Kissing Bondphenomena in Solid State Welds of Aluminium Alloys”, Weld Joints, Vol. 1, pp. 225s–231s. [27] Elangovan, K. and Balasurbramanian, V. (2008), “Influences of Tool Pin Profile and Tool Shoulder Diameter on Friction Stir Processing Zone in AA6061 Aluminium Alloy”, Materials and Design, Vol. 29, pp. 362–373. [28] Scialpi, A., De Filippis, L.A.C. and Cavaliere, P. (2007), “Influence of Shoulder Geometry on Microstructure and Mechanical Properties of Friction Stir Welded 6082 Aluminium Alloy”, Materials and Design, Vol. 28, pp. 1124–1129.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Improving Wear Behaviour of GIHC Grade of Grey Cast Iron by Using WC-12CO and Stellite-6 Coatings Baljit Singh Ph.D. Research Scholar, Mechanical Engineering, SBS State Technical Campus, Ferozepur, Punjab–152004, India E-mail: [email protected] Abstract—The wear is very severe problem in industry and it is also common phenomenon in sliding parts. Wear leads economic loss to society. In the present research study efforts will be made to reduce the wear rate of Braking Disc Rotor with surface modifications techniques and comparison made on the basis of wear properties of grade of cast iron GIHC coated by WC-12CO and Stellite-6 deposited by Detonation spray processes is presented. The WC-12CO on GIHC grey cast iron performs slightly better than the Stellite-6 coating as shown in the results. The WC12CO-GIHC coating substrate combination has shown minimum Cumulative Volume loss among all the two combinations. The wear resistance for coating–substrate combinations in their decreasing order is WC-12CO-GIHC>Stellite6-GIHC. Keywords: Wear, Detonation Spray, Brake Disc

INTRODUCTION The optimization of automotive vehicles braking systems, subjected to mechanical and thermal stresses, depends on a combination of properties. In general, a complex state of stress is found and it is practically impossible to select a material and design a component based only on one of these properties. The material used in brake rotors should be able to bear thermal fatigue and should absorb and dissipate, as soon as possible, the heat generated during braking [1].

PREPARATION OF SAMPLES Small cylindrical pins having diameter of circular cross-section equal to 8mm and length equal to 30 mm were prepared from GIHC grey irons. A total of 9 pins of GIHC grade were prepared. Pins were given the Sample No. from 1 to 9. The grinding of end faces (to be coated) of the pins done using emery papers of five different grades 220, 400, 600, 800, 1000 in the same order. Grinding was followed by polishing with 1/0, 2/0, 3/0 and 4/0 grades polishing papers. Table 1: Chemical Composition (Wt %) of the GIHC Grey Cast Iron C Si Mn P S GIHC 3.54 1.66 0.524 0.816 0.124 Table 2: Hardness Values of GIHC Grey Iron Material GIHC Hardness (HB) 207

coatings which are very dense and homogeneous. The various constituents of the powder are Carbon (1%), Tungsten (6 %), Fe (5%), Chromium (30%), and Cobalt (58%). The powder particle size is 10–45 microns. Tungsten Carbide (WC-12CO): WC-12CO when sprayed using the Awaaz Detonation spray coating process, it produces coatings, which are very hard, dense and excellent bonded. Coatings can be built upto higher thickness. WC-12CO when sprayed using the detonation spray coating process, it produces coatings, which are very hard, dense and its bonding strength is very high. Coatings can be built up to higher thickness. These are coatings which are gives better protection against wear. The powder particle size is 15–35 microns. The various constituents of the powder are Chromium (32.72%), Cobalt (46.58%) and Tungsten (20.70%). Table 3: Detonation Spray Parameters for Two Coatings

(a) Stellite-6 Gases Oxygen Flow Rate 2990 Spraying Distance 165 mm

Acetylene 2410

Nitrogen 720

Acetylene 2300

Nitrogen 960

(b) WC-12CO Gases Oxygen Flow Rate 4350 Spraying Distance 150 mm

Characterization of Coatings

Deposition of Coatings

Specimen Preparation

Two types of coating powders namely (1) Stellite-6 (2) WC-12CO are selected for Detonation Spray Coating Process after the literature survey. Research papers shows that the above coatings have excellent wear resistance.

Two specimens having dimensions 20 mm *15 mm * 5 mm were cut from the substrate material GIHC. The specimens were grinded using sand papers of 220, 400, 600 and 1000 grit sizes and subsequently polished on 1/0, 2/0, 3/0 grades. Samples were well polished until it shines like mirror. One sample of GIHC substrate was

Stellite-6: Stellite powder when sprayed using the Awaaz Detonation spray coating process, it produces

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

11

Singh

Improving Wear Behaviour of GIHC Grade of Grey Cast Iron by Using WC-12CO and Stellite-6 Coatings

coated with Stellite-6 coating and other is coated with WC-12CO coating.

Sliding wear Study using Pin-on-disc Configuration Experimental Set Up Dry sliding wear tests for the uncoated and detonation spray coated cylindrical specimens were conducted using a pin-on-disc machine (Wear and Friction Monitor Tester TR-201made by M/S DUCOM, Bangalore, India) conforming to ASTM G 99 standard. The tests were conducted in air with a room temperature of 30–32 °C. Wear tests were performed on the pin specimens that had flat surfaces in the contact regions and the rounded corner. The pin was held stationery against the counter face of a rotating disc made of carbon steel (EN-31) at 40 mm track diameter. EN-31 steel is a plain carbon steel; case hardened 62 to 65 HRC as provided with the pin-on-disc machine. The composition of the material of the steel disc is given in Table 4. Table 4: Chemical Composition (wt %) of the En-31 Carbon Steel Disc C Si Mn S P 0.42 (max) 0.05-0.35 0.40-0.70 0.05 (max) 0.05 (max)

Sliding Wear Studies The pins were polished with emery paper and both disc and the pin were cleaned and dried before carrying out the test. The pin was loaded against the disc through a dead weight loading system. The wear tests for coated as well as uncoated specimens were conducted under three normal loads of 40 N, 50 N and 60 N and a fixed sliding velocity of 1 m/s. The track radii for the pins were kept at 40 mm. The speed of the rotation of the disc (477 rpm) for all the cases was so adjusted so as to keep the linear sliding velocity at a constant value of 1 m/s. A variation of ±5 rpm was observed in the rpm of the disc. Wear tests have been carried out for a total sliding distance of 5400 m (6 cycles of 5 min, 5 min, 10 min, 10 min, 20 min, 40 min duration), so that only top coated surface was exposed for each detonation sprayed sample. Tangential force was monitored continuously during the wear tests. Weight losses for pins were measured after each cycle to determine the wear loss. The pin was removed from the holder after each run, cooled to room temperature, brushed lightly to remove lose wear debris, weighed and fixed again in exactly the same position in the holder so that the orientation of the sliding surface remains unchanged. The weight was measured by a micro 12

balance to an accuracy of 0.0001 gm. The coefficient of friction has been determined from the friction force and the normal loads in all the cases. Wear Rate The wear rate data for the coated as well as uncoated specimens were plotted with respect to sliding distance to establish the wear kinetics. The specific wear rates for the coated and uncoated material were obtained by W = δw/ LρF Where W denotes specific wear rates in, Bowden (B) (1B = 10-6 mm3/N-m) [Recommendation from IRG OECD meeting with about 30 participants to introduce a new unit for wear rate: Bowden (B) equal to 10-6 mm3/N.m] δw is the weight loss measured in, g L the sliding distance in, m ρ the density of the worn material in g/ mm3 and F the applied load in N. Wear Volume The wear volume loss was also calculated from the weight loss and density of the coatings as well as substrate material for all the investigated cases. These data were reported in the form of plots showing the cumulative wear volume loss Vs sliding distance for all the cases. Bar charts were also drawn to show net Volume = mass / density Wear Volume Loss = (δw/9.81) /ρ Where δw is the weight loss in, g And ρ is the density of material, g/mm3. Coefficient of Friction The coefficient of friction (μ) determined from the frictional force and the normal load has been plotted against the sliding time to give the friction behavior of the coated as well as the uncoated material. The coefficient of friction (μ) was calculated as below: μ = Frictional Force (N) / Applied Normal Load (N)

RESULTS SEM/EDS Analysis of the D-gun as Sprayed Coatings The Scanning electron microscope micrographs as well as Energy Dispersive Spectrum (EDS) with element composition for Detonation sprayed Stellite-6 and

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Improving Wear Behaviour of GIHC Grade of Grey Cast Iron by Using WC-12CO and Stellite-6 Coatings

WC-12CO coatings on GIHC shown in Fig. 1. The microstructure of these coatings is hardly bonded, homogeneous and free from surface cracks, pores and voids. The SEM/EDS analysis of the stellite-6 coating

Singh

and minor phases of C and O which together to make Co which is also desired element of coating WC-12CO. The no. of peaks corresponding to elements of coatings can be seen from diffraction patterns of different coatings for GIHC.

(a)

Fig. 1: Surface Morphology and EDS Patterns from Different Spots on as Coated Samples (a) Stellite-6 (b) WC-12CO

showed in Fig. 1(a). The elements for stellite-6 coating corresponding to (spectrum 1) for load 40 N and 50 N are C, O, Cr, CO Si etc. The color of the surface at this spectrum is dull grey and near this point surface is white which may be due to presence of oxygen and at (spectrum 4) the coating is more uniform. The elemental composition for WC-12CO coating corresponding to spectrum 1 & 4 for load 40N and 50N is shown in (Fig. 1(b)). The spectrum 4 of WC-12CO coating also confirms the presence of desired coating elements Chromium (32.72%), Cobalt (46.58%) and Tungsten (20.70%). At (spectrum 4) the color is naturally white which is may be due presence of excess oxygen on the surface.

X-ray Diffraction (XRD) Analysis The X-ray diffraction patterns for detonation sprayed stellite-6 and WC-12CO ON GIHC are shown in Fig. 2. Fig. 2(a) shows the X-ray diffraction patterns for as coated samples of Stellite-6 coating on GIHC and Fig. 2(b) shows the X-ray diffraction patterns for as coated samples of WC-12CO coating on GIHC. From Fig. 2(a) it is identified that coating stellite-6 shows the excess of desired coating elements such as C, Cr, O, Co and small amount of Fe and Si. Similarly, from Fig. 2(b) it is evident that coating WC-12CO shows the major phases of tungsten which is desired element of coating

(b)

Fig. 2: X-ray Diffraction Patterns of as Coated GIHC Material; (a) Stellite-6 (b) WC-12CO

Wear Behavior of Two Coatings vs GIHC Substrate Three samples of each coating i.e. Stellite-6 and WC-12COon GIHC were subjected to wear on Pin-On– Disc-wear test rig at normal loads of 40N, 50N and 60N respectively. Three samples of bare GIHC substrate were also subjected to wear on Pin-On Disc-wear test rig at the same loads. The cumulative volume loss vs time for each coating is plotted as shown in Fig. 3. From the results of (Fig. 3) it is investigated that cumulative volume loss for two detonation sprayed wear resistant coatings show better wear resistant in comparison to bare GIHC. Also from Fig. 3(a) it is observe that Stellite-6 coating has wear resistant but this resistant is less as compared toWC12CO coating which has more wear resistant as shown in

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

13

Singh

Improving Wear Behaviour of GIHC Grade of Grey Cast Iron by Using WC-12CO and Stellite-6 Coatings

(Fig. 3). The bar chart (Fig. 4) showing the Cumulative Volume Loss (CVL) in onecomplete cycle (90 min) is also drawn for each coating and GIHC substrate. From (Fig. 3 & Fig. 4) it is observed that with increase in loadwear loss increases for the detonation sprayed coatings Stellite-6 and WC-12CO and bare GIHC the observation is same as that of the observation of (Cueva, 2003) in which also the wear loss of GIHC substrate increases with increase in load.

Comparative Wear Behaviour for Two Coatings Figure 5 shows Comparative Volumetric Wear Loss (mm3) for two coatings at (a) 40 N (b) 50 N and (c) 60 N. It is also observed from the results that WC-12CO is showing the minimum cumulative volume loss as compared to other two coatings. Therefore the wear resistance of Detonation sprayed coatings on GIHC in their decreasing order can be given as WC12CO>Stellite-6.

Fig. 3: Cumulative Volume Loss (mm3) with Time for (a) Stellite-6 (b) WC-12CO Coatings and GIHC Substrate

Fig. 5: Comparative Volumetric Wear Loss (mm3) for Two Coatings on GIHC Substrate at (a) 40 N (b) 50 N and (c) 60 N

DISCUSSION Fig. 4: Cumulative Volume Loss (mm3) in One Cycle for D-gun Sprayed Coatings and Bare GIHC at 40 N, 50 N and 60 N

14

Selection of material, coating and the wear behaviour of the uncoated GIHC and detonation sprayed Stellite-6 and

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Improving Wear Behaviour of GIHC Grade of Grey Cast Iron by Using WC-12CO and Stellite-6 Coatings

WC-12CO coatings have been discussed. From the study of Mohanty, 1996 it was observed that it is possible to deposit almost any material on any substrate by D-gun spray process to considerably extend the life of parts; also it is observed in the present study that the Stellite-6 and WC-12CO coatings powders have been successfully deposited on GIHC substrate by the detonation spray process. It was further confirmed by characterization of coatings using SEM and EDAX analysis of as coated specimens. SEM/EDAX results of WC-12CO and Stellite-6 (Fig. 1(a&b)) shows the presence of O and C on the surface which may be due to formation of carbide. Also XRD analysis of study did (Fig. 2) which supports the results of scanning electron microscope (SEM). There is always the material loss of bare material greater than the as coated material. From (Fig. 3) it is observed that the detonation sprayed wear resistant coatings Stellite-6, WC-12CO coated GIHC specimens showed significantly lower cumulative volume loss as compared to bare GIHC material under the normal load of 40 N, 50 N and 60 N. It is investigated with the help of Pin-on-Disk Wear testing machine. There are many studies; Murthy & Venkataraman, 2006, Sundararajan, 2005 and Jun Wang 2000 which support the above finding that Detonation sprayed coatings increases the wear resistant and wear loss of bare material is always greater than the as coated material which is also found in present study in (Fig. 4). Also Fig. 5 shows comparison of two coatings in which WC-12CO has minimum wear loss as compared to Stellite-6 and therefore WC-12CO can be used for coating the grey cast iron material which is used in light truck pads. WC-12CO coating has more bonding strength than stellite-6. From (Fig. 4) it is observed that the wear loss is increase with increase load. From this Fig. it is clear that the wear loss for two coatings and bare GIHC also increases with load which is same observation as Cueva (2003), The CVL for WC-12CO coating was found to be minimum in present study as shown in Fig. 4. it is may be due to the presence of W, CO, FE and O also tungsten and carbide increases the property of wear resistant. Identical results have been reported by Mohanty, 1996. It may be due to carbide formation due to diffusion of the Fe from the substrate. The WC-12CO-GIHC coating substrate combination has shown minimum wear loss among all two combinations as shown in Fig. 5.

CONCLUSION

Singh

1. GIHC is best grade of grey cast iron. 2. Stellite-6 and WC-12CO are best coating powders to deposit on grey cast irons. 3. Stellite-6 and WC-12CO wear resistant coatings have successfully been deposited on GIHC grade of grey cast iron. 4. The Stellite-6 and WC-12CO coating based GIHC specimens showed lower cumulative volume loss as compared to bare GIHC specimens. 5. Wear loss for detonation sprayed wear resistant coatings Stellite-6 and WC-12CO coated samples and bare GIHC samples increases with increase in load. 6. The Cumulative Volume loss for WC-12CO coating was minimum in the present study. Therefore WC-12CO is best coating to deposit on GIHC grade of grey cast irons. 7. The wear resistance for coating–substrate combinations in their decreasing order is WC-12CO-GIHC>Stellite-6-GIHC. Therefore out of these combinations WC-12CO-GIHC coating substrate combination is the best combination.

REFERENCES [1]

[2]

[3]

[4] [5]

[6]

[7]

Jimbo, Y (1990), “Development of High Thermal Conductivity Cast Iron for Brake Disk Rotors”, SAE Technical Paper Series, International Congress and Exposition, Detroit, MI. Singh, H., Grewal, M.S., Sekhon, H.S. and Rao, R.G. (2008), “Sliding wear Performance of High-velocity oxy-fuel Spray Al2O 3/ TiO2 and Cr2O3 Coatings”, Published Vol. 222, Issue: 4, pp. 601–610. Singla, Manoj Kumar, Singh, Harpreet and Chawla, Vikas (2011), “Thermal Sprayed CNT Reinforced Nanocomposite Coatings”, Vol. 10, No. 8, pp. 717–72. Heath, R., Heimgartner, P., Irons, G., Miller, R., and Gustafsson S. (1997), Material Science Forum, Vol. [251]-54, pp. 809–816. Cueva, G., Sinatora, A., Guesser, W.L. and Tschiptschin, A.P. (2003), “Wear Resistance of Cast Irons Used in Brake Disc Rotors”, Wear, Vol. 255, pp. 1256–1260. Wang, Jun, Sun, Baode, Guo, Qixin, Nishio, Mitsuhiro and Ogawa, Hiroshi (2000), “Wear Resistance of a Cr3C2-NiCr Detonation Spray Coating”, Vol. 11, No. 2, pp. 261–265. Murthy, J.K.N., Rao, D.S. and Venkataraman, B. (2001), “Effect of Grinding on the Erosion behaviour of a WC–Co–Cr Coating Deposited”, HVOF and Detonation Gun Spray Processes, Vol. 249, Issue 7, pp. 592–600.

Based upon experimental results obtained in the present study, the following conclusions have been drawn:

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

15

Mechanical Characterization of Polypropylene (PP) and Polyethylene (PE) Based Natural Fiber Reinforced Composites Deepak Varshney, Kishore Debnath* and Inderdeep Singh Department of Mechanical and Industrial Engineering, Indian Institute of Technology, Roorkee–247667, India E-mail: *[email protected] Abstract—Natural fiber reinforced composites have attained a commanding position in various industries owing to their superior physical and mechanical properties as compared to the conventional materials. The present research work is focused on the development of light weight, environmental friendly, cost-effective composite materials based on sisal, nettle, hemp and jute fibers. Two different types of thromboplastic polymers such as, polypropylene (PP) and polyethylene (PE) has been used for fabricating the composite laminates. The mechanical behavior of the developed composite laminates has been compared under different fibers loading (15 to 25 wt.%). The results show that the PP based composites have superior mechanical properties as compared to the PE based composites. The mechanical properties of the sisal fiber reinforced composites have been found superior to all the developed composites. The morphological analysis of the fractured test specimens has also been carried out using scanning electron microscopy (SEM). Keywords: Natural Fibers, Polypropylene, Polyethylene, Composites, Mechanical Properties, SEM Analysis

INTRODUCTION Synthetic fiber (glass, carbon, aramid etc.) reinforced polymer composites have been extensively used in numerous structural applications where high strength and stiffness are required. But, the major problem associated with these types of fiber reinforced composites is that these materials are non-biodegradable in nature. When the issues such as reuse or recycling are involved at the end of the service of product, these materials pose significant problems. Due to the growing environmental awareness, the use of synthetic fiber reinforced composites is decreasing progressively. Therefore, the conceptualization and development of biodegradable composite materials has received an attention from the researchers and technologists [1–4]. The development of partially biodegradable composites made of natural fibers with thermosetting resin started in 1980s. A variety of biodegradable composites have been developed with reasonable mechanical properties using different types of natural fibers such as, sisal, hemp, flax, ramie, jute, banana etc. and biodegradable polymers such as starch, cellulose or vegetable oil derivatives. Unlike the traditional synthetic fiber composites, natural fiber reinforced composites are not only renewable and biodegradable but, also possess some other merits such as, low cost, high modulus, high strength, depleting tool wear, lightweight and safer manufacturing process when compared with synthetic fiber reinforced composites. When natural fiber is used as reinforcement, they lowers the energy required almost 80% for production and reduces the weight by 10%, where, the cost of natural fiber based composites is 5 to 10% lower than the glass fiber reinforced composites [5]. The mechanical properties of the developed natural fiber reinforced composites are significantly influenced by many factors, such as properties of the natural fibers 16

and polymers, bonding strength between the fibers and matrix, orientation of fibers, volume fraction of fibers etc. Though, natural fiber reinforced composites have low modulus of elasticity and high moisture absorption tendency but, the most important characteristics of natural fiber reinforced composites is that they are biodegradable in nature. Due to the biodegradable nature of natural fiber reinforced composites, they can easily decompose in the environment. Moreover, the increasing number of publications in the area of development, processing and characterization of natural fiber reinforced composites shows the significance of these materials in the modern manufacturing industries [6–8]. A number of research initiatives have been made to study the effect of various natural fibers such as, sisal, flax, abaca, hemp, bamboo, grewia optiva and jute on mechanical properties of the resultant composites [9–15]. It has been established that the weight percentage of the fiber significantly influence the mechanical properties of the developed composites. Therefore, the effect of fiber content on the mechanical properties of the fiber reinforced composite is of particular interest. It has been investigated that the tensile, flexural and impact strength of 2% maleic anhydride grafted PP based hybrid composites is reached to an optimum value at the fiber loading of 15% [16]. The tensile and flexural modulus of the kenaf fiber reinforced PP composites was found maximum at the optimum fiber loading of 30% by weight. But, the optimum value of fiber weight fraction is 40% for the jute fiber reinforced PP composites [17]. The variation of tensile properties with fiber content of micro winceyette fiber reinforced thermoplastic corn starch composites was investigated [18]. It was observed that with an increase in fiber content from 0 to 20 wt. %, the tensile strength of the developed composites was approximately trebled to 150 MPa. But, the percentage elongation at break of the developed composites is decreased with an increase in fiber content.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Mechanical Characterization of Polypropylene (PP) and Polyethylene (PE)

It was also observed that the percentage elongation dropped significantly when fiber content lies in between 0–10% by weight. A slight decrease in percentage elongation was observed after 10% fiber content. The effects of fiber content on the mechanical properties of the natural fiber reinforced composites have not been reported extensively [19–22]. From the above discussion, it is quite clear that the natural fiber reinforced composites have the potential to replace traditional synthetic fiber composites. The experimental investigation of mechanical behavior of these materials is an important aspect in order to further expand their application spectrum. Therefore, in the present experimental investigation, several types of composite laminates have been fabricated by reinforcing polypropylene (PP) and polyethylene (PE) with sisal, jute, nettle and hemp fibers. The objective of this study is to evaluate the effect of fiber loading on the mechanical and morphological properties of sisal, jute, nettle and hemp fiber reinforced PP and PE composites.

MATERIALS AND METHODS Fibers and Polymers Natural fibers such as, sisal, hemp, nettle and jute (Fig. 1) were obtained in mat form were obtained from Uttarakhand Bamboo and Fiber Development Board, Dehradun, India and Women’s Development Organization, Dehradun, India. PP and PE in a ‘ready to use’ pellet form was supplied by Reliance Industries Limited, Mumbai, India. The density of PP and PE at room temperature is 0.905 and 0.91 g/cm3, respectively. The glass transition temperature (Tg) and melting temperature (Tm) of PP are 100 °C and 165 °C, respectively. PE has melting temperature (Tm) of 125 °C. The melt flow index (MFI) of PP and PE are 10.5 and 2 g/10 min, respectively.

Varshney, Debnath and Singh

Laminates Fabrication All composites were manufactured by hot compression using film-stacking method. The fiber weight fraction varies from 15 to 25% in each type of composites. To remove the absorbed moisture from fibers and polymers, all the natural fibers and polymers are pre-heated in an oven at 80 °C for 6 hours. At first stage, PP and PE pellets are converted into thin films of 1 mm thickness by hot pressing. To prepare the films of PP and PE, respective pallets are placed in the mold cavity and pressure and temperature is applied. The maximum pressure of 3 MPa was applied when the temperature reached to the melting point of polymers. After that the mold is allowed to cool under constant pressure for the curing of polymers. In the next stage, natural fibers and polymers with desired weight percentage are stacked alternatively in the mold cavity. The application of pressure and temperature results in the mixture to take the shape of mold cavity. The stacked polymer films and woven fibers are then hot pressed at a melting temperature of polymer. The consolidation of the PP and PE based composite were performed at a temperature of 165 °C and 125 °C, respectively. The time taken to process the PP and PE based composites are 16 and 12 minutes, respectively. Teflon sheets and mold release agent are applied at the top and bottom surface of the mold to prevent sticking of the polymers with the mold plates. The melting point of mold release agent is greater than the thermoplastic polymers. The thickness of the prepared composite was maintained to 4 mm. The prepared composites were stored in a desiccator until further use to avoid moisture absorption.

Measurement of Mechanical Properties The tensile and flexural strength (three point bend test) of the developed composites have been conducted in universal testing machine (Instron, Model 1011, UK) as per ASTM 3039 [23] and ASTM D790 [24], respectively. During tensile and flexural test the extension rate was set at a rate of 2 mm/min. In the impact test (Unnotched Charpy impact test), the composite specimens were tested according to ASTM D6110. The impact testing machine (Fine Testing Machines, Pune, India) was used to perform impact test has a maximum impact energy of 300 J with the weight of pendulum of 20 kg and striking velocity of 5 m/sec. All the resulted values are taken as the average of three experimental values.

Morphological Analysis

Fig. 1: Natural Fiber Mats

The fracture surface of the composite specimens was investigated using SEM (LEO, Model 435VP, LEO Electron Microscopy Ltd., England). SEM has an accelerating voltage of 30kV with resolution 4 nm in HV and 6 nm in VP and magnification 10 X to 300000 X. SEM specimens are sputter coated with gold in order to enhance the conductivity of the test specimen.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

17

Varshney, Debnath and Singh

Mechanical Characterization of Polypropylene (PP) and Polyethylene (PE)

RESULTS AND DISCUSSION The tensile, flexural and impact strength of the developed composite laminates have been found experimentally as per ASTM standards. The effect fiber loading on the mechanical and morphological properties of the developed natural fiber reinforced composites has also been investigated. It has been found that the types of natural fibers and polymers, and the weight fraction of fibers has a significant effect on the mechanical properties of the natural fiber reinforced composites. It is quite clear from the current experimental investigation that the response of composite laminates under different fiber loading conditions is different for different types of developed laminates. The mechanical properties of the neat resin and the developed natural fiber reinforced composite laminates are presented in Table 1 to 3, respectively.

Tensile Behavior of the Developed Natural Fiber Reinforced Composites The tensile properties of the PP and PE based composites are shown in Table 2 and Table 3, respectively. From the result it is quite clear that the incorporation of natural fibers to the neat polymers results in significant improvement in the tensile properties. It has been observed that sisal fiber reinforced composites have superior tensile strength as compared to the other type of composite laminates. The highest value of tensile strength (76.8 MPa) was found in sisal/PP containing fiber content of 25 wt.%. The second highest value of tensile strength is 60.1 MPa that was achieved during the tensile testing of sisal/PE composite laminates. The tensile strength performance of PP and PE based nettle and jute fiber

Material Properties Neat PP Neat PE

Composite Properties Fiber (wt.%) PP/Hemp PP/Sisal PP/Nettle PP/Jute

Composite Properties Fiber (wt.%) PE/Hemp PE/Sisal PE/Nettle PE/Jute

18

Tensile Strength (MPa) 19.3 11.7

composites was found inappreciable as compared to the other two type composites. For PP based composites, the tensile strength of the nettle/PP composites is least though, the tensile strength of nettle/PP was magnificently improved as compared to the neat PP. The least tensile strength for PE based composites was found in jute/PE composites. It is worthy to mention that the tensile strength of the developed composite laminates increases with an increase in fiber loading. Neat PP and neat PE showed tensile modulus of 0.96 and 0.21 GPa, respectively. A minor improvement in the tensile modulus was observed for all types of developed composite laminates, whereas PP based composite laminates showed relatively higher improvement in their tensile modulus. The increase in tensile modulus implies that the stiffness of the polymers enhanced after incorporating natural fibers. The percentage elongationat break for neat PP and neat PE is 13.38% and 15.43%, respectively. It is interesting to note the percentage elongationat break of polymers is significantly decreased after incorporating natural fibers. The microscopic images of the fractured specimen after tensile testing are shown in Fig. 2 and Fig. 3, respectively. From the micrographs it is quite evident that the fiber fracture, fibers pulled out and matrix cracking is the major failure mode that occurred during the tensile testing of composite laminates. It is well established that the tensile properties of the fiber reinforced composites is mainly depends on the interfacial bonding strength between the fiber and matrix. The interfacial bonding characteristic between the fiber and matrix is governed by the surface characteristics or surface roughness of the fibers. The mechanical properties of the sisal fiber reinforced composites is superior to the other types of developed composites because, sisal fiber is relatively rougher than other fibers which may results in good interfacial bonding properties between sisal fibers and polymers [7].

Table 1: Mechanical Properties of the Neat Polymer Tensile Modulus Elongation at Flexural (GPa) Break (%) Strength (MPa) 0.96 13.38 27.5 0.21 15.43 24.5

Table 2: Mechanical Properties of the PP Based Natural Fiber Reinforced Composites Tensile Strength Tensile Modulus Elongation at Flexural Strength (MPa) (GPa) Break (%) (MPa) 15% 20% 25% 15% 20% 25% 15% 20% 25% 15% 20% 25% 28.6 32.2 33.4 1.4 2.1 1.1 2.1 1.8 2.1 56.3 59.7 95.9 47.7 58.2 76.8 1.8 2.5 2.9 2.7 2.4 2.7 58.8 81.5 108 24.1 28.9 32.1 1.3 1.2 2.9 1.8 1.2 0.9 49.9 58.8 105 25.6 30.1 33.1 1.1 1.9 2.2 2.4 2.1 1.5 64.5 94.4 137 Table 3: Mechanical Properties of the PE Based Natural Fiber Reinforced Composites Tensile Strength Tensile Modulus Elongation at Break Flexural Strength (MPa) (GPa) (%) (MPa) 15% 20% 25% 15% 20% 25% 15% 20% 25% 15% 20% 25% 19.1 19.7 28.4 0.59 0.46 0.40 1.81 1.82 2.03 14.8 58.4 36.9 47.5 51.1 60.1 1.51 1.58 0.66 5.45 3.63 3.09 75.9 105 101 14.3 16.7 18.6 0.49 0.81 0.61 1.81 0.91 1.73 21.3 33.1 47.6 12.3 13.1 14.3 0.79 0.53 0.85 1.82 2.42 1.52 44.5 44.5 53.9

Impact Strength (KJ/m2) 24.5 33.3

Impact Strength (KJ/m2) 15% 20% 25% 13.2 13.3 13.1 12.7 13.6 13.9 12.5 12.5 13.2 12.5 13.8 13.4

Impact Strength (KJ/m 2) 15% 20% 25% 13.7 13.6 13.4 14.3 14.5 14.9 13.8 14.3 14.8 14.0 13.9 13.7

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Mechanical Characterization of Polypropylene (PP) and Polyethylene (PE)

Fig. 2: SEM Images of Fractured Surface of PP Based Composites under Tensile Failure Mode (a) Sisal/ PP, (b) Nettle/ PP, (c) Hemp/ PP and (d) Jute/ PP

Varshney, Debnath and Singh

Fig. 3: SEM Images of Fractured Surface of PE Based Composites under Tensile Failure Mode (a) Sisal/ PE, (b) Nettle/ PE, (c) Hemp/PE and (d) Jute/ PE

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

19

Varshney, Debnath and Singh

Mechanical Characterization of Polypropylene (PP) and Polyethylene (PE)

Flexural Strength of the Developed Natural Fiber Reinforced Composites The flexural strength results for neat PP and PP based composites are presented in Table 1 and Table 2, respectively. Neat PP showed the flexural strength of 27.5 MPa. It was found that the flexural strength of the PP based composites increases linearly with an increase in the fiber content. Jute/PP composites have better flexural properties among all the developed composite laminates. The flexural strength results for PE based composites are presented in Table 3. Neat PP showed the flexural strength of 24.5 MPa. It was found that flexural strength of the PE based composites increases with an increase in the fiber content up to 20 wt%, further if the fiber content increases up to 25 wt%, hemp and sisal fiber reinforced PE composites experiences lower flexural strength. On the other hand, flexural strength of nettle and jute fiber based composites increases linearly with an increase in fiber loading from 15 to 25 wt.%. This is clear from the experimental results that sisal/PE composites have the promising flexural properties among all the PE based composites. The microscopic images of the fractured specimen after flexural testing are shown in Fig. 4 and Fig. 5, respectively. Fiber fracture, debonding and matrix cracking were detected in the micrographs of the fractured test specimens after flexural testing of the developed natural fiber reinforced composites.

Impact Strength of the Developed Natural Fiber Reinforced Composites Charpy impact strength of the PP and PE based composites was also evaluated experimentally. There is no significant variation observed in the impact strength values with an increase in fiber content. The impact strength of the PP and PE based composites has been found lower than that of neat PP and neat PE. It was found that impact strength of hemp/ PP and jute/ PP based composites increases with an increase in the fiber content up to 20 wt.% and then decreases. It was also found that impact strength of sisal and nettle based composites increases linearly with an increase of fiber content up to 25 wt.%. It is clear that sisal fiber based composites have better impact strength in comparison with other composites. Sisal/ PE composites showed maximum impact strength among the all developed composite laminates. It is interesting to note that, jute/PE and hemp/PE composites have experienced decreased impact strength with fiber loading. It has been reported that the unnotched impact resistance of thermoplastics generally decreases in the presence of agro based fibers [25]. SEM analysis of the fractured specimens after impact testing was carried out to observe failure mechanisms of composite constituents. The microscopic images of the fractured composite specimens are shown in Fig. 6 and 20

Fig. 4: SEM Images of Fractured Surface of PP Based Composites under Flexural Failure Mode (a) Sisal/ PP, (b) Nettle/ PP, (c) Hemp/ PP and (d) Jute/ PP

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Mechanical Characterization of Polypropylene (PP) and Polyethylene (PE)

Fig. 5: SEM Images of Fractured Surface of PP Based Composites under Flexural Failure Mode (a) Sisal/ PP, (b) Nettle/ PP, (c) Hemp/ PP and (d) Jute/ PP

Varshney, Debnath and Singh

Fig. 6: SEM Images of Fractured Surface of PP Based Composites under Impact Failure Mode (a) Sisal/ PP, (b) Nettle/ PP, (c) Hemp/PP and (d) Jute/PP

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

21

Varshney, Debnath and Singh

Mechanical Characterization of Polypropylene (PP) and Polyethylene (PE)

Fig. 7, respectively. From the SEM micrographs, it is clear that matrix cracking, debonding and fiber pullout is the predominant failure modes detected in the fractured surface of the test specimens.

CONCLUSION The mechanical characterization of the PP and PE based natural fiber reinforced composites was investigated and the results were compared. The mechanical properties of the PP based composites was found relatively higher than PE based composites. It can also be concluded that sisal/PP composites have higher tensile properties among all the developed composites. However, jute/ PE composites have the maximum flexural strength. The maximumimpact strength was obtained during the imapct testing of sisal/PE composites. The fractured surface of the PP and PE based composites after mechanical testing were also investigated using SEM. The results of the SEM analysis show that the fiber fracture, fibers pulled out and matrix cracking is the major failure mode that occurred during the mechanical testing of composite laminates.

REFERENCES [1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10] Fig. 7: SEM Images of Fractured Surface of PP Based Composites under Impact Failure Mode (a) Sisal/PP, (b) Nettle/PP, (c) Hemp/PP and (d) Jute/PP

22

Debnath, K., Singh, I. and Dvivedi, A. (2014), “Rotary Mode Ultrasonic Drilling of Glass Fiber-reinforced Epoxy Laminates”, Journal of Composite Materials. DOI: 10.1177/ 0021998314527857. Debnath, K., Singh, I., Dvivedi, A. and Kumar, P. (2013), “Recent Advances in Composite Materials for Wind Turbine Blades”, World Academic Publishing-Advances in Materials Science and Applications, Hong Kong, pp. 25–40. Debnath, K., Singh, I. and Dvivedi, A. (2013), “Dry Sliding Wear behaviour of Glass Fibre Reinforced Epoxy Composites Filled with Natural Fillers”, Reason-A Technical Journal, Vol. XII, pp. 61–68. Debnath, K., Dhawan, V., Singh, I. and Dvivedi, A. (2013), “Effect of Natural Fillers on wear Behavior of Glass-fiberReinforced Epoxy Composites”, Proceedings of the International Conference on Research and Innovations in Mechanical Engineering, Ludhiana, India, pp. 441–450. Venkateshwaran, N., Perumal, A.E., Alavudeen, A. and Thiruchitrambalam, M. (2011), “Mechanical and water absorption behaviour of banana/sisal reinforced hybrid composites”, Materials and Design, Vol. 32, pp. 4017–4021. Bajpai, P.K., Singh, I. and Madaan, J. (2014), “Development and characterization of PLA-based green composites: A review”, Journal of Thermoplastic Composite Materials, Vol. 27, pp. 52–81. Bajpai, P.K., Singh, I. and Madaan, J. (2012), “Comparative Studies of Mechanical and Morphological Properties of Polylactic Acid and Polypropylene Based Natural Fiber Composites”, Journal of Reinforced Plastics and Composites, Vol. 31, pp. 1712–1724. Bajpai, P.K., Meena, D., Vatsa, S. and Singh, I. (2013), “Tensile behavior of Nettle Fiber Composites Exposed to Various Environments”, Journal of Natural Fibers, Vol. 10, pp. 244–256. Cyras, V.P., Martucci J.F., Iannace, S. and Vazquez, A. (2002), “Influence of the Fiber Content and the Processing Conditions on the Flexural Creep behavior of Sisal-pcl-starch Composites”, Journal of Thermoplastic Composite Materials, Vol. 15, pp. 253–265. Romhány, G., Karger‐Kocsis, J. and Czigány, T. (2003), “Tensile Fracture and Failure behavior of Thermoplastic Starch with Unidirectional and Cross‐ply Flax Fiber Reinforcements”, Macromolecular Materials and Engineering, Vol. 288, pp. 699–707.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Mechanical Characterization of Polypropylene (PP) and Polyethylene (PE) [11]

[12]

[13]

[14]

[15]

[16]

[17]

Rahman, M.R., Huque, M.M., Islam, M.N. and Hasan, M. (2009), “Mechanical Properties of Polypropylene Composites Reinforced with Chemically Treated Abaca”, Composites Part A: Applied Science and Manufacturing, Vol. 40, pp. 511–517. Sawpan, M.A., Pickering, K.L. and Fernyhough, A. (2011), “Improvement of Mechanical Performance of Industrial Hemp Fibre Reinforced Polylactidebio Composites”, Composites Part A: Applied Science and Manufacturing, Vol. 42, pp. 310–319. Okubo, K., Fujii, T. and Yamamoto, Y. (2004), “Development of Bamboo-based Polymer Composites and their Mechanical Properties”, Composites Part A: Applied Science and Manufacturing, Vol. 35, pp. 377–383. Singha, A.S. and Thakur, V.K. (2009), “Mechanical, Thermal and Morphological Properties of Grewia Optiva Fiber/ Polymer Matrix Composites”, Polymer-Plastics Technology and Engineering, Vol. 48, pp. 201–208. Gowda, T.M., Naidu, A.C.B. and Chhaya, R. (1999), “Some Mechanical Properties of Untreated Jute Fabric-reinforced Polyester Composites”, Composites Part A: Applied Science and Manufacturing, Vol. 30, pp. 277–284. Nayak, S.K. and Mohanty, S. (2010), “Sisal Glass Fiber Reinforced PP Hybrid Composites: Effect of MAPP on the Dynamic Mechanical and Thermal Properties”, Journal of Reinforced Plastics and Composites, Vol. 29, pp. 1551–1568. Lee, B.H., Kim, H.J. and Yu, W.R. (2009), “Fabrication of Long and Discontinuous Natural Fiber Reinforced Polypropylene Biocomposites and their Mechanical Properties”, Fibers and Polymers, Vol. 10, pp. 83–90.

Varshney, Debnath and Singh [18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

Ma, X., Yu, J. and Kennedy, J.F. (2005), “Studies on the Propertied of Natural Fibres-reinforced Thermoplastic Starch Composites”, Carbohydrate Polymers, Vol. 62, pp. 19–24. Li, X., Tabil, L.G., Panigrahi, S. and Crerar, W.J. (2009), “The Influence of Fiber Content on Properties of Injection Molded Flax Fiber-HDPE Biocomposites”, Canadian Biosystems Engineering, Vol. 8, pp. 1–10. Hajnalka, H., Racz, I. and Anandjiwala, R.D. (2008), “Development of HEMP Fibre Reinforced Polypropylene Composites”, Journal of Thermoplastic Composite Materials, Vol. 21, pp. 165–74. Facca, A.G., Kortschot, M.T. and Yan, N. (2007), “Predicting the Tensile Strength of Natural Fibre Reinforced Thermoplastics”, Composites Science and Technology, Vol. 67, pp. 2454–2466. Facca, A.G., Kortschot, M.T. and Yan, N. (2007), “Predicting the Elastic Modulus of Natural Fiber Reinforced Thermoplastics”, Composites Part A: Applied Science and Manufacturing, Vol. 37, pp. 1660–1671. ASTM Standard D3039. Tensile Properties of Polymer Matrix Composite Materials. West Conshohocken, PA: ASTM International, 2008. ASTM Standard D790. Standard Test Methods for Flexural Properties of Unreinforced and Reinforced Plastics and Electrical Insulating Materials. West Conshohocken, PA: ASTM International, 2010. Sanadi, A.R., Caulfield, D.F. and Jacobsen, R.E. (1997), Paper and Composites from Agro-based Resources, CRC Press, New York, pp. 377–402.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

23

Optimization of Process Parameters of Gas Metal ARC Welding by Taguchi’s Experimental Design Method Deepak Kumar and Sandeep Jindal* Department of Mechanical Engineering, M.M. Engineering College, Maharishi Markandeshwar University, Mullana, Ambala–133207, India E-mail: *[email protected] Abstract—Gas metal arc welding is a fusion welding process having wide applications in industry. The process parameters play a very significant role in determining the quality of a welded joint in Gas Metal Arc Welding (GMAW). So proper selection of process parameters is necessary to obtain weld joint with increased tensile strength. In the work, experiments were carried out on 1018 mild steel plates using gas metal arc welding (GMAW) process. L9 orthogonal array of Taguchi`s experimental design was used for optimization of welding current, voltage and gas flow rate on welded joints. Keywords: GMAW, Taguchi’s Experimental Design, Orthogonal Array, Tensile Strength

INTRODUCTION Gas Metal Arc Welding is a process in which arc, molten metal and the heat affected zone are protected from contamination by the atmosphere of inert gas fed through the GMAW torch. GMAW is one of the most widely used processes in industry. The input parameters play a very significant role in determining the quality of a welded joint. Sapakal and Telsang [1] presents study and optimization of influence parameters current, voltage and welding speed on penetration depth of MS material with the help of Taguchi’s design. The predicted parameters were analyzed by analysis of variance (ANOVA). Patel and Chaudhary [2] evaluated the parameters; welding current, wire diameter and wire feed rate to investigate their influence on weld bead hardness for MIG welding and TIG welding by Taguchi’s method and Grey Relational Analysis (GRA). From the study it was concluded that the welding current was most significant parameter for MIG and TIG welding. By use of GRA optimization technique the optimal parameter combination was found to bewelding current, 100 Amp; wire diameter 1.2 mm and wire feed rate, 3 m/min for MIG welding. Haragopal et al. [3] usedtaguchi’smethod to design process parameters MIG welding that optimize mechanical properties of weld specimen foraluminum alloy (Al-65032). The process parameters considered weregas pressure, current, groove angle and pre-heat. L-9 orthogonal array was used for designing experiments. It was concluded from the work that current is most influencing parameter on ultimate tensile strength (UTS) and pressure is most influencing parameter for proof stress, % age elongation and impact energy.The effect of current, voltage and welding speed on depth of penetration for MS has been studied by authors [4] for MIG welding process. The depth of penetration was measured for each specimen after the welding operation for closed butt joint. The voltage higher than 26.5 v and current higher than 150 amp causes abrupt rise in 24

penetration depth but welding speed higher than 0.16 mm/min decreases the depth of penetration. Tewari et al. [5] studied the effect of parameters on weldability of Mild Steel plates. The effect of current, voltage, weldingspeed and heat input rate on depth of penetration was studied. They concluded that increasing the speed of travel and maintaining constant arc voltage and current will increase penetration until an optimum speed is reached at which penetration will be maximum. Increasing the speed beyond this optimum value will result in decreasing penetration.Singh [6] did the work on MS plates by varying the parameters gas flow rate, voltage and welding position gap to find the tensile strength and find the optimum parameters with the help of Taguchi orthogonal array, the signal-to-noise (S/N) ratio and analysis of variance. Conformation experiments were conducted to verify the effectiveness of the Taguchi optimization method.

TAGUCHI’S EXPERIMENTAL DESIGN Taguchi’s experimental design method tells us how we can get the optimum results by varying all the parameters at a time. As a researcher in Electronic Control Laboratory in Japan, Dr. Genechi Taguchi has developed a method based on “orthogonal array”.Taguchi's orthogonal arrays are highly fractional orthogonal designs which can be used to estimate main effects using only a few experimental runs.Taguchi's Signal-to-Noise ratios (S/N) serve as objective functions for optimization which help in data analysis and prediction of optimum results. Signal to Noise Ratio: The following three types of S/N ratio are employed in practice: Smaller - The-Better: n = -10log10 [mean of sum of squares of measured data] Larger-The-Better: n = -10log10 [mean of sum of squares of reciprocal of measured data]

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Optimization of Process Parameters of Gas Metal ARC Welding

Kumar and Jindal

Nominal-The-Best:

Selection of Orthogonal Array

n = 10log10 [square of mean/variance]

To select an appropriate orthogonal array for experiments, the total degrees of freedom need to be computed. In this study each three level parameter has 2 degree of freedom (DOF = Number of level–1), the total DOF required for three parameters each at three levels is 8. Once the degrees of freedom required are known, the next step is to select an appropriate orthogonal array to fit the specific task. Basically, the degrees of freedom for the orthogonal array should be greater than or at least equal to those for the process parameters. In this study, an L9 Orthogonal array (a standard 3-level OA) having 8 degree of freedom was selected from the Taguchi’s special set of predefined arrays.

EXPERIMENTATION The base metal used for present work is 1018 Mild Steel with dimensions of the workpiece as 150 mm x 100 mm x 12.5 mm. Consumable electrode (ER 70 S6) of 1.2 mm diameter and carbon dioxide as inert gaswas used for gas metal arc welding. Welding of specimens has been carried out by GMAW setup at Saini Engineering Works, Hisar. Chemical composition of base metal and filler wire or electrode is given in Table 1 and Table 2, respectively. C 0.18

Table 1: Chemical Composition of Base Metal Mn P S Fe 0.6-0.9 0.04 max 0.05 max 98.81-99.26

Si

Mn

P

S

Cu

Cr

V

Mb

Fe

0.06-0.15

0.8-0.15

1.40-1.85

0.025

0.35

0.05

0.15

0.03

0.15

Bal

Table 2: Chemical Composition of Electrode# C

The input welding process parameters selected for this work were welding current, voltage and gas flow rate. Tensile strength was taken as the output quality characteristic. Each of these parameters was varied at 3 levels. The range and levels of these parameters were decided on the basis of preliminary experiments conducted by using one variable at a time approach. The feasible range for the machining parameters was defined by varying current (200–240A), voltage (36–40V), gas flow rate (15–19 lpm) for welding of selected base metal and electrode. The input welding parameters and their levels are given in Table 3. According to the number of factors and their levels; L9 Orthogonal array is selected from the Taguchi’s special set of predefined arrays presented in Table 4. Table 3: Process Parameters and their Levels Parameters Current (Amp) Voltage (Volt) Gas Flow Rate (l pm)

Code A B C

Level 1 200 36 15

Level 2 220 38 17

Table 4: Standard L9 Orthogonal Array No of Runs Control Factors A B 1 L1 L1 2 L1 L2 3 L1 L3 4 L2 L1 5 L2 L2 6 L2 L3 7 L3 L1 8 L3 L2 9 L3 L3

Level 3 240 40 19

C L1 L2 L3 L2 L3 L1 L3 L1 L2

The tensile strength of the welded samples is measured on Universal Testing Machine (UTM). The specimens are machined according to ASME standards as shown in Fig. 1.

Fig. 1: Standard Tensile Test Specimen

RESULT AND DISCUSSION A class of statistics called signal-to-noise(S/N) ratios has been defined to measure the effect of noise factors on performance characteristics. S/N ratio takes into account both the variability in the response data and the closeness of the average response to the target value. There are several signal-to noise ratios available depending on the type of performance characteristic. As mentioned earlier, the evaluation characteristic of each weldment is tensile strength. In the Taguchi’s method the term signal represents the desirable value (mean) for the output characteristic and the term noise represents the undesirable value (standard Deviation) for the output characteristic. Therefore, the S/N ratio to the mean to the S.D. S/N ratio is used to measure the quality characteristic deviating from the desired value. The S/N ratio n is defined as n = -10 log (M.S.D.) where M.S.D. is the mean square deviation for the output characteristic. Higher tensile strength leads to a stronger weld so, larger-the-better quality characteristic was taken. For larger the better quality characteristic S/N ratio is expressed by: S/N = -10 log10 [mean of sum of squares of reciprocal of measured data]

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

25

Kumar and Jindal

Run

1 2 3 4 5 6 7 8 9

Optimization of Process Parameters of Gas Metal ARC Welding

Table 5: Result for Tensile Strength and S/N Ratio Current Voltage GFR Tensile S/N (Amp) (Volt) (lpm) Strength Ratio (MPa) 200 36 15 348.3 50.8391 200 38 17 369.9 51.3617 200 40 19 386.9 51.7520 220 36 17 371.9 51.4085 220 38 19 376.7 51.5199 220 40 15 394.0 51.9099 240 36 19 360.2 51.1309 240 38 15 367.8 51.3122 240 40 17 392.4 51.8746

In Taguchi method signal-to-noise ratios is used to determine the optimum level of each factor. This is done by collecting levels with high signal-to-noise ratio. The response table shows the average of each response characteristic (S/N ratios, means) for each level of each factor. Table 6 include ranks based on Delta statistics, which compare the relative magnitude of effects. The Delta statistic is the highest minus the lowest average for each factor. Minitab assigns ranks based on Delta values; rank 1 to the highest Delta value, rank 2 to the second highest, and so on.Finally the optimum level of each factor is given in Table 7. These levels are the peak values of each factor as shown in Fig. 2. From Fig. 1, it can be seen that the tensile strength first increases with the current upto the optimum level i.e. 220A and after that it decreases whereas in case of voltage tensile strength increases with increases in its Table 6: Response Table for Signal to Noise Ratio Level 1 2 3 Delta Rank

Current 51.32 51.61 51.44 0.29 2

Voltage 51.13 51.40 51.85 0.72 1

Gas Flow Rate 51.35 51.55 51.47 0.20 3

Table 7: Optimum Values for Each Factor Current Level 2 220A 51.35 Voltage Level 3 40V 51.55 Gas flow rate Level 2 17LPM 51.47

Fig. 2: Graph for S/N Ratio of Different Parameters

26

Table 8: Result of Analysis of Variance for Tensile Strength

value. But in case of gas flow rate tensile strength suddenly increases upto the optimum level i.e. 17 lpm and after that it decreases gradually. The analysis of variance was carried out at 95% confidence level. The purpose of ANOVA is to investigate which welding process parameters significantly affect the tensile strength. This is accomplished by separating the total variability of the S/N Ratios, which is measured by the sum of squared deviations from the total mean of the S/N ratio, into contributions by each welding process parameter and the error (Table 8). In the experimentation work, for S/N ratios, voltage(p = 0.13) has the significant effect on tensile strength at an α-level of 0.05, other parameters current (p = 0.071) and gas flow rate (p = 0.150) are nonsignificant because their p-values are greater than 0.05. The percentage contribution by each of the welding process parameters in the total sum of the squared deviations can be used to evaluate the importance of each parameter change on the tensile strength. From the Fig. 3 we can see that voltage has the greatest percentage contribution of 79.87%, current has 13.33% and gas flow rate has 5.78%. It can be concluded that voltage has greatest effect followed by current and voltage.

Regression Mathematical Model Equation In this approach, a single equation relating one response to the process parameters for the whole domain of

Fig. 3: Pie Chart for % Age Contribution of Different Parameters for Tensile Strength

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Optimization of Process Parameters of Gas Metal ARC Welding

investigation is represented. By use of Minitab 15, regression analysis equation for the work was found shown as below: Tensile Strength = 32.6 + 0.127 Current + 7.74 Voltage + 1.14 GFR

CONCLUSION Optimization of the process parameters in GMAW by Taguchi’s experimental design method has been performed. An L9 Orthogonal Array was selected to study the relationships between the tensile strength and the three controllable input welding parameters such as voltage, current and gas flow rate. The following conclusions can be drawn based on the experimental results of this research work: 1.

Taguchi’s experimental design method provides a simple, systematic and efficient methodology forthe optimization of the GMAW parameters.

2.

The optimum values for each factor are shown as below: Current Voltage Gas flow rate

Level 2 Level 3 Level 2

220 A 40 V 17l pm

51.35 51.55 51.47

3.

Tensile strength increases with increase in voltage. But in case of current and gas flow rate, it increases upto the optimum level and decreases on further increasing these values.

4.

Voltage has the greatest percentage contribution followed by current and gas flow rate.

5.

Voltage is the significant factor for tensile strength but current and gas flow rate are the non-significant parameters in GMAW.

Kumar and Jindal

REFERENCES [1]

Sapakal, S.V. and Telsang, M.T. (2012), “Parametric Optimization of MIG Welding using Taguchi Design Method”, International Journal of Advanced Engineering Research and Study, Vol. 1, pp. 28–30. [2] Patel, C.N. and Chaudhary, S. (2013), “Parametric Optimization of Weld Strength of Metal Inert Gas Welding and Tungsten Inert Gas Welding by using Analysis of Variance and Grey Relational Analysis”, International Journal of Research in Modern Engineering and Emerging Technology, Vol. 1, No. 3. [3] Sathish, R. Naveen, B. Nijanthan, P. Geethan, K. and Rao, V. (2012), “Weldability and Process Parameter Optimization of Dissimilar Pipe Joints Using GTAW,” International Journal of Engineering Research and Applications (IJERA), Vol. 2, Issue 3, pp. 2525–2530. [4] Singh, V. (2013), “An Investigation for Gas Metal Arc Welding Optimum Parameters of Mild Steel AISI 1016 using Taguchis Method”, International Journal of Engineering and Advanced Technology (IJEAT), Vol. 2. [5] Aghakhani, M., Mehrdad, E. and Hayati E. (2011), “Parametric Optimization of Gas Metal Arc Welding Process by Taguchi Method on Weld Dilution”, International Journal of Modeling and Optimization, Vol. 1, pp. 216–220 [6] Das, B., Debbarma, B., Rai, R.N. and Saha, S.C. (2013), “Influence of Process Parameters on Depth of Penetration of Welded Joint in MIG Welding Process”, International Journal of Research in Engineering and Technology, Vol. 2, Issue 10, pp. 220–224. [7] Patil, S.R. and Waghmare, C.A. (2013), “Optimization of MIG Welding Parameters for Improving Strength of Welded Joints”, International Journal of Advanced Engineering Research and Studies, Vol. 2, pp. 14–16 [8] Kocher, G., Kumar, S. and Singh, G. (2012), “Experimental Analysis in MIG Welding with IS 2062 E250 (A Steel with Various Effects)”, International Journal of Advanced Engineering Technology. [9] Tewari, S.P., Gupta, A. and Prakash, J. (2010), “Effect of Welding Parameters on the Weldability of Material”, International Journal of Engineering Science and Technology, Vol. 2, No. 4, pp. 512–516. [10] Sureshkumar, L., Verma, S.M., Radhakrishna Prasad, P., Kiran Kumar, P. and Siva Shanker, T. (2011), “Experimental Investigation for Welding Aspects of AISI 304 & 316 by Taguchi Technique for the Process of TIG & MIG Welding”, International Journal of Engineering Trends and Technology, Vol. 2.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

27

Nano Science and Technology through Friction Stir Processing and Bulk Metallic Glass Routes Part II: Bulk Metallic Glass Brij K. Dhindaw1, Harpreet Arora2 , J. Eckert3, Nilam Barekar4 and Sundeep Mukherjee5 1,4

Brunel Center for Advanced Solidification Technology (BCAST), Brunel University, Uxbridge, Middlesex, UB8 3PH, UK 3 Leibniz-Institut fur Festko ̈rper- und Werkstoffforschung Dresden, Institut fur Metallische Werkstoffe, Postfach 270016, D–01171 Dresden, Germany 2,5 Department of Materials Science and Engineering, University of North Texas, Denton, Texas 76203, USA E-mail:

INTRODUCTION Bulk glassy materials have been defined that can be obtained in glassy forms in thicknesses more than 1mm. They have been perceived as the ideal material that can have high strength combined with good formability. This could open up their application as unique structural materials for engineering applications. Therefore in the last decades, bulk metallic glasses (BMGs) have attracted considerable interest due to their superior physical, mechanical and corrosion properties [1]. Strength and the elastic limit of different materials have been shown in Fig. 1. It clearly indicates that glassy alloys have optimum combination of strength and elastic limit. It has strength more than steels and elastic limit as polymers. Nevertheless, one of the main problems of BMGs is still the very low plasticity at room temperature. In fact, most BMGs show very low plastic strain (0–2%) under uniaxial compressive loading or tensile loading at room temperature. If this could be enhanced one can get ideal materials where both high strength can be realised with high ductility. The production routes of BMG can be divided into three different groups according to the state before glass formation, vapour deposition, quenching from the liquid state/deposition from the dissolve state (e.g., melt spinning and metal mould casting / electrochemical

Fig. 1: Amorphous Metallic Alloys Combine Higher Strength than Crystalline Metal Alloys with the Elasticity of Polymers

28

deposition and solid state amorphization reactions, e.g., ball milling. In the present studies industrially viable and less costly casting route has been adopted. However, to be viable as engineering material researchers are interested in developing BMGs which exhibit high glass forming ability (GFA) along with good mechanical properties, so that bigger size samples can be cast. The main obstacles for bulk metallic glass to be used in structural applications are its low plasticity and limited casting dimensions. It is well known that the plastic deformation of amorphous alloys is confined to highly localised regions, the so-called shear bands [2]. Methods for improving the plasticity of BMGs led to the development of BMG composites [3], in which the shear bands are hindered by the second phase particles resulting in high strains to failure. The dispersed second phase being itself ductile and capable of strain hardening distributes the plastic strain more homogeneously in the material [4]. The precipitation of nanocrystals in the glassy matrix [5], nanometre-scale medium-range ordering [6] and phase separation in the glass [7] has been proposed to account for the enhancement of plastic deformability. BMG with nano dispersed crystallites seems to be having good corrosion resistance as the other major requirements of structural material. Thus, nano crystallites dispersed BMG seems to be perspective high end structural material. In order to clarify the structure–property relationship in these metastable alloys, the present work reports [8] the evolution of elastic and tensile properties of (Cu0.5Zr0.5)100−xAlx (x = 5, 6, 8) alloys and the analysis of the respective fracture surfaces. The alloy group represents a very simple composition to prepare BMG that is industrially viable to produce in large quantities. The BMGs not only show high fracture strength, but also exhibit macroscopically detectable plasticity before failure under uniaxial tension at room temperature. Without knowledge of tensile properties it is difficult to decide the materials capability to be used in structural applications. Corrosion studies are also briefly dealt with.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Nano Science and Technology through Friction Stir Processing and Bulk Metallic

EXPERIMENTAL High purity elements Cu (99.9 wt.%), Zr (99.99 wt.%), and Al (99.99 wt.%,) were used as starting materials. Four composition (Cu50Zr50, Cu47.5Zr 47.5Al5, Cu47Zr47Al6 and Cu46Zr 46Al8) and 20 (5*4) pre-alloyed ingots of 20 gram each are prepared by mixing the appropriate weights of each element in the form of small lumps for the desired atomic composition. The small lumps of each element were ground on abrasive paper (silicon carbide P-1100) followed by cleaning with alcohol and dried to remove any trace of moisture. The alloying elements were melted

Dhindaw Et Al.

in arc melter The pre-alloy was melted in a arc melter and cast into 2 mm * 10 mm * 50 mm plates by in situ copper mould suction casting facility attached to the arc melter (Fig. 2). The pre-alloy was melted in a arc melter and cast into 2 mm * 10 mm * 50 mm plates by in situ copper mould suction casting facility attached to the arc melter. Proper precautions and procedures were followed to make sure that no Oxygen entered the melting and casting chamber so that no inclusions are formed creating nucleation sites. The samples were characterized using XRD, DSC, TEM, Tensile testing. Samples were prepared from bulk cast plate.

Tungsten electrode

Argon Flow

Air ventilation Plasma

Evacuation

Sample Water cooled chamber

Mould

Fig. 2: Schematic Diagram of an in-situ Copper Mould Suction Casting Device

RESULTS AND DISCUSSION

Tensile Properties

Mechanical Properties

Figure 3 shows true tensile stress verses true tensile strain plots of Cu47.5Zr47.5Al5, Cu47Zr 47Al6 and Cu46Zr46Al8 alloys. All the three compositions show tensile fracture stress above 1600 MPa with fracture strain around 2%. Small plasticity is observed particularly in Cu47.5Zr47.5Al5, and Cu47Zr47Al6 alloys, followed by catastrophic failure (Fig. 3(a) & (b)). Cu46Zr46Al8 alloy was broken in the fully brittle mode. Fig. 3 shows the tensile stress–strain curves of alloys which elucidate the fracture strength and strain to failure with different Al content as listed in Table 1. It can be seen that the metallic glassy samples display initial elastic deformation behaviour. There is a clear deviation from linearity accompanied by detectable ductility.

Ultrasonic Testing for Elastic Properties The elastic properties of the as cast samples were estimated using the ultrasonic sound velocity measurements. Obtained data are summarized in Table 1. The measured values of density (ρ), longitudinal wave velocity (V l), shear wave velocity (V s), Young modulus (E), shear modulus (K), bulk modulus (G), Poison’s ratio (v) and G/K values of Cu47.5Zr 47.5Al5, Cu47Zr47Al6 and Cu46Zr 46Al8 alloys are summarized in Table 1. The Young's modulus (E) increases with increasing Al content. All BMGs have a Poisson's ratio (ν) larger than 0.31–0.32, the value, which is connected to a brittle-toductile transition for BMGs [9].

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

29

Dhindaw Et Al.

Nano Science and Technology through Friction Stir Processing and Bulk Metallic

Table 1: Mechanical Properties of Cu 47.5Zr47.5Al 5, Cu47Zr47Al 6 and Cu46Zr46Al 8 Alloys, Measured by Ultrasonic Wave Composition Ρ (gcm- E K G v G/K 3 ) (Gpa) (GPa) (GPa) Cu47.5Zr47.5Al5 7.16 ± 2 88 ± 1 117 ± 1 32 0.370 0.27 Cu47Zr 47Al6 7.12 ± 2 90 ± 1 116 ± 1 32 0.360 0.28 Cu46Zr 46Al8 7.05 ± 2 93 ± 1 117 ± 1 34 0.355 0.29

(a)

(b) Fig. 3: True Tensile Stress-strain Curves of (a) Cu47.5Zr47.5Al5 (b) Cu47Zr47Al6 (c) Cu46Zr46Al8 Alloys

Transmission Electron Microscopy In order to reveal the mechanisms causing the rather unique ductility of the present Cu–Zr–Al BMGs, the microstructures of the samples before and after the tensile tests were examined in the TEM. The TEM image (Fig. 4(a)) obtained from the as-cast Cu47Zr47Al6 sample shows an isotropic maze pattern being typical of an amorphous structure. The selected area electron diffraction (SAED) pattern (inset to Fig. 4(a)) is only comprised of a set of diffuse halos and there is no hint for the presence of crystallites. This agrees very well with the X-ray diffraction results. Yet, in some regions sparse nanocrystals with diameters of 2–5 nm were found in the glassy matrix, which may be pre-existing nuclei remaining from the liquid during rapid solidification. Fig. 4(b) shows the HRTEM image of the tensile deformed sample. At a higher magnification the bright and dark contrast phases become obvious and the dark phase contains a lot of lattice fringes shown by circle. Consequently, it can be assumed that the sample partially crystallises during deformation and fracture. One possible explanation could be the heat release in shear band when the sample fractures. Second possible reason could be that the shear during deformation changes the short range order (SRO) in the glass and facilitates crystallisation. Third reason may be the deformation induced transformation that has been reported many times in BMGs during compressive loading. A TEM image 30

(c)

Fig. 4: (a) HRTEM Image and Corresponding SAED Patterns (Inset) for as-cast Cu47Zr47Al6 Sample (b) HRTEM Image of Deformed Cu47.5Zr47.5Al 5 alloy, Show lot of Lattice Fringes, Indicated by Circle, Diffraction Pattern in Inset (c) TEM High Resolution Image of Cu 47Zr47Al6 Sample after Tensile Deformation Revealing the Presence of Nanocrystallites in the Glassy Matrix; Inset: Micrograph Showing Deformation-Induced Twinning in the Nanocrystals

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Nano Science and Technology through Friction Stir Processing and Bulk Metallic

recorded from the fractured Cu47Zr47Al6 sample is depicted in Fig. 4(c). In the case of the fractured sample, abundant nanocrystallites are distributed homogeneously in the amorphous matrix with an average size of 10–50 nm. The micrograph presented in the inset clearly shows a contrast, which must be attributed to twinning. A strong interface between the nanocrystallites and the matrix is evident. The nanocrystals can be regarded as a perfectly elastic body with a theoretical strength of σ ≈ E/10, where E is the Young's modulus [10] and, thus can serve as a reinforcement phase. It was generally believed that nanocrystals embedded in an amorphous matrix can only provide high strength, but not strain hardening owing to a lack of substantial dislocation multiplication, while the amorphous phase generally even softens during straining.

Dhindaw Et Al.

(a)

(b)

Fracture Surface SEM Analysis The SEM observations show that the Cu47Zr47Al6 alloy fractured primarily in a shear mode, as shown in Fig. 5. The tensile fracture angle, θT, between the tensile axis and the fracture plane (as marked in Fig. 5(a)) is equal to 54°. The typical fracture surface of a Cu47Zr 47Al6 sample is depicted in Fig. 5(b). Three regions with different morphologies can be distinguished, corresponding to different stages of failure. Region (i) constitutes a strip of approximately 50–100 μm, which spans over the whole side of the specimen and is relatively rugged. This is where the crack initiates. In region (ii) the crack propagates in a stable manner and as a result the fracture surface is relatively smooth. In this region, one can observe features typical of BMGs, such as vein-like patterns (Fig. 5(c)). These patterns are believed to form when the elastic energy stored in the bulk metallic glass during loading is released as heat into the nascent shear band. This is a clear indicator for a shear-dominated fracture mechanism [11]. Besides the vein-like structures, there are round cores with radiating veins on the fracture surface (Fig. 5(c)). The fracture surface changes its morphology again as can be seen in region (iii) (Fig. 5(b)), which is extremely rugged and shows multiple smaller cracks. Summarizing the tensile fracture mechanism of these BMGs, it is clear that samples having a high tensile fracture strength exhibited fracture angles around 54°. In case of low tensile fracture strength the samples show rough fracture surface. The rough fracture surface of these samples is believed to be originating from the heterogeneously distributed nano crystals. This has been supported by the distribution of the shear bands near fracture surface (Fig. 5(d) inset 2). The following section presents some observations on the behavior of different metallic glasses against surface degradation phenomenon including wear, corrosion and erosion from the existing literature.

(c)

(d)

Fig. 5: (a) Appearance of the Side Surface of the Cu47Zr47Al6 Specimen Fractured at the Strain Rate of 1 × 10-4 s-1 (b) Typical Fracture Surface of a Cu 47Zr47Al6 Sample. One can Clearly Distinguish Three Different Regions of (i) Crack Initiation, (ii) Stable Crack Growth and (iii) Brittle Failure (c) SEM Image of Tensile Fracture Morphology of Cu 47Zr47Al6 alloy (d) Fracture Surface Showing the Surface Cracks and Shear Bands, and in Enlarge View in Inset (1) and (2)

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

31

Dhindaw Et Al.

Nano Science and Technology through Friction Stir Processing and Bulk Metallic

WEAR STUDIES Sliding wear behavior of a copper-based bulk metallic glass (Cu50Hf41.5Al8.5) was investigated for both as-cast and annealed samples by Maddala et al. [12]. The results of the study showed that the wear behavior of a metallic glass improved with the increase in the annealing time. The wear rate of the investigated BMG was compared with that of stainless steel 304 and it was observed that the BMG showed superior wear resistance than that of stainless steel 304. The results of this study showed that controlled annealing treatments of the Cu50Hf41.5Al8.5 metallic glass can be helpful in optimizing its hardness and sliding wear resistance. These authors proposed that the generation of nanocrystals during the annealing treatment significantly contributed towards the improved wear behaviour of the investigated BMG. Composite samples comprising Fe-based metallic glass particles with nominal composition of Fe61.5Cr11.5Al1.03Si2.64P4.34C7.72Mo1.45Mn0.84B9 and the crystalline Ni matrix were produced by Kwon et al. [13]. Wear behavior was investigated by a pin-on-disc type tester using hardened steel as the counter material. It was observed that composite samples containing less than 10 vol.% crystalline Ni exhibited higher wear resistance than the monolithic BMG sample. The authors proposed that the observed result may be attributed to an increase in the overall toughness of the composite with the addition of tough crystalline Ni. Ni60Nb20Zr20 glass composite coatings were developed by Suresh et al. [14] using gas tunnel type plasma spraying torch and the sliding wear behavior of the developed metallic glass composite coatings was evaluated under unlubricated condition using Ball-on-Disk (BOD) tribometer. Composite coatings were developed at different arc current values and it was found that highest percentage of glassy phase in the coating was obtained at an optimum current value of 300 A. The sliding wear rate of the metallic glass composite coating was found to vary according to the applied load and crystallinity. Wear rate of the coating produced with a plasma current of 300 A showed the lowest friction coefficient and wear rates. Presence of the optimized crystalline phase in the amorphous structure of the glassy phase produced at 300 a mainly attributed for the reduced wear rates of the BMG coating. Thus it can be summarized that presence of nano crystalline phases in glassy matrix provides a perspective rout for further development of these materials.

CORROSION STUDIES (Cu0.6Hf0.25Ti0.15)90Nb10 BMG composites with dispersed Nb rich crystalline phase were fabricated by Qin et al. 32

[15] using copper mold casting. It was observed that the (Cu0.6Hf0.25Ti0.15) 90Nb10 alloy exhibited much higher corrosion resistance in 1N HCl and 3 mass % NaCl solutions than the Nb-free alloy. XPS analysis revealed the formation of Hf-, Ti- and Nb-enriched surface film on the alloy which might have caused higher corrosion resistance of the investigated BMG composite in HCl and NaCl solutions. Hsu et al. [16] fabricated Ti based bulk metallic glass composites using carbon nanotubes (CNT) and evaluated their corrosion performance. High energy ball milling was used to prepare CNT/Ti50Cu28Ni15Sn7 metallic glass composite powders. The bulk metallic glass composite were prepared by vacuum hot pressing of the as-milled CNT/Ti50Cu28Ni15Sn7 metallic glass composite powders. Electrochemical corrosion experiments revealed that the investigated bulk metallic glass composites exhibited high corrosion resistance in Hanks solution. Electrochemical and tribocorrosion tests were performed on ZrCuAlNi (Zr55Cu30Ni5Al10) bulk metallic glasses in different media simulating body fluids by Espallargas et al. [17]. It was found that in different electrolytes, the BMG material with amorphous structure exhibited largest wear rates. The authors suggested that the observed behaviour may be attributed to the passivating character of the alloy resulting in the formation of ZrO2 which can promote abrasion. However, as suggested by these authors, the recrystallization process as well as the presence of the new intermetallic phase (Zr2Cu) in the BMG material with crystalline phase prohibited the oxidation of Zr and thereby resulted to lower wear rates. Erosion–corrosion behavior of Zr-based bulk metallic glass (Zr55Cu30Ni10Al5) was evaluated by Ji et al. [18] in saline-sand slurry using a slurry pot erosion tester. Erosion as well as corrosion was found to increase continually with the impact velocity. However, the synergism between erosion and corrosion was observed to decrease firstly and then increase with the impact velocity. The crystallite phase developed in the BMG during the erosion-corrosion testing and the microstructure of the BMG was found to be composed of amorphous as well as crystalline phases. Compared with 304 SS, Zr-BMG showed a better E–C resistance in saline-sand slurry. Zr based metallic glass composite coatings (Zr55Cu30Al10Ni5) were produced by Yugeswaran et al. [19] using gas tunnel type plasma spraying. The sliding and erosive wear behaviors of the developed coatings were studied using a pin-on-disc and erosive wear tester, respectively. The results showed that the percentage of

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Nano Science and Technology through Friction Stir Processing and Bulk Metallic

crystallinity increased with the increase in the arc current with the corresponding reduction in the percentage porosity in the BMG specimen. The wear resistance of the coatings was found to improve with the increase in the arc current used in the plasma spraying. Here also it has been shown that presence of crystalline phases in BMG has generally contributed to improvement in corrosion properties.

properties. In the case of Bulk Metallic Glasses, transformation of some amorphous structure to crystalline nano phase, improved ductility so vital to BMG being considered for structural applications. There for it is suggested that rather than external addition of nano particles, in situ generation through technological routes, can be much more technology friendly option.

REFERENCES

CONCLUSION

[1]

The paper presents significant results about the possibility of getting reasonable ductility in high strength high modulus BMG. A high resolution microscopy examination of the fractured Cu47Zr47Al6 amorphous alloy revealed abundant nanocrystallites that had polymorphically precipitated in an amorphous matrix. The ability of strain hardening and twinning are believed to contribute to the observed ductility under tension. The results demonstrate that tensile loading facilitates nanocrystallization resulting in plasticity. Bulk metallic glasses also demonstrate superior resistance against surface degradation phenomenon including wear, corrosion, erosion and erosion-corrosion. Thus these materials show the very promising path for nano composites rout to attain the ideal properties through BMG route.

[2] [3]

ACKNOWLEDGMENT Contribution from Mr. R.B. Kumar and Dr. Simon Pauly in this research is gratefully acknowledged.

General Summary

[4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]

The paper presented research where it has been shown that by engineering microstructures in terms of incorporating nano scale particles mechanical properties can be enhanced in remarkable fashion. In the case of FSP it was shown that refinement of the in situ inter-metallic by fsp showed significant increase in the mechanical

Dhindaw Et Al.

[18] [19]

Zhang, G.Q., Jiang, Q.K., Chen, L.Y., Shao, M., Liu, J.F. and Jiang, J.Z. (2006), J. Alloy Compd., Vol. 424(1–2), pp. 176–178. Spaepen, F. (1977), Acta Metall, Vol. 25(4), pp. 407–415. Kühn, U., Eckert, J., Mattern, N., Schultz and L. (2002), Appl. Phys. Lett., Vol. 80(14), pp. 2478–2480. Sun, Y.F., Wei, B.C., Wang, Y.R., Li, W.H., Cheung, T.L. and Shek, C.H. (2005), Appl. Phys. Lett., Vol. 87, 051905-1–051905-3. Lee, S.W., Huh, M.Y., Fleury, E. and Lee, J.C. (2006), Acta Mater. Vol. 54, pp. 349–355. Kim, K.B., Das, J., Baier, F., Tang, M.B., Wang, W.H., Eckert, J. (2006), Appl. Phys. Lett., Vol. 88, 051911-1–051911-3. Oh, J.C., Ohkubo, T., Kim, Y.C., Fleury, E., Hono, K. (2005), Scripta Mater., Vol. 53, pp. 165–169. Barekar, N.S., Pauly, S., Kumar, R.B., Kühn, U., Dhindaw, B.K. and Eckert, J. (2010), Mater. Sci. Eng. A, Vol. 527, 5867–5872. Lewandowski, J.J., Wang, W.H. and Greer, A.L. (2005), Phil. Mag. Lett., Vol. 85(2), pp. 77–87. Lee, S.W., Lee, C.M., Ahn, J.P., Kim, Y.C. and Lee, J.C. (2007), Mater. Sci. Eng. A, pp. 449–451 to. 172–175. Zhang, Z.F., He, G., Eckert, J. and Schultz, L. (2003), Phys. Rev. Lett., Vol. 91(4), 045505-1–045505-4. Maddala, D.R., Mubarok, A. and Hebert, R.J. (2010), Wear, Vol. 269, pp. 572–580. Kwon, D.H., Park, E.S., Huh, M.Y., Kim, H.J. and Bae, J.C. (2011), J. Alloy Compd, Vol. 509S, pp. S105–S108. Suresh, K., Yugeswaran, S., Rao, K.P., Kobayashi, A., Shum, P.W. (2013), Vol. 88, pp. 114–117. Qin, C., Zhang, W., Amiya, K., Asami, K. and Inoue, A. (2007), Mater. Sci. Eng., pp. 449–451 to 230–234. Hsu, C.F., Kai, W., Lin, H.M., Lin, C.K. and Yew, L.P. (2010), J. Alloy Compd., Vol. 504S, pp. S176–S179. Espallargas, N., Aune, R.E., Torres, C., Papageorgiou, N. and Munoz, A.I., Wear, http://dx.doi.org/10.1016/j.wear.2012.12.053. Ji, X., Zhao, J., Zhang, X. and Zhou, M. (2013), Tribol. Int., Vol. 60, pp. 19–24. Yugeswaran, S., Kobayashi, A., Suresh, K., Rao, K.P. and Subramanian, B. (2012), Appl. Surf. Sci., Vol. 258, pp. 8460–8468.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

33

Jet Impingement Heat Transfer: Stationary Disc Gus Nasif1*, Ron Barron2 and Ram Balachandar3 1,2,3

Department of Mechanical, Automotive & Materials Engineering, University of Windsor, ON, Canada N9B 3P4 2 Department of Mathematics & Statistics, University of Windsor, ON, Canada N9B 3P4 3 Department of Civil & Environmental Engineering, University of Windsor, ON, Canada N9B 3P4 E-mail: *[email protected]

Abstract—A numerical investigation to determine the thermal characteristics of an unsubmerged axisymmetric oil jet impinging in a confined space on a stationary disc with uniform heat flux has been undertaken. The volume of fluid (VOF) method utilizing a high resolution interface capturing (HRIC) scheme was used to perform the two-phase (air-oil) simulations. The governing 3D Navier-Stokes equations and energy equation were numerically solved using a finite volume discretization. The conjugate heat transfer (CHT) method was used to obtain a coupled heat transfer solution between the fluid and solid disc, yielding a more accurate prediction for the heat transfer coefficient. The effect of nozzle geometry on the thermal characteristics was also investigated in this study. For long jets impinging on a stationary boundary, the nozzle geometry has no significant effect on the thermal characteristics. Keywords: Volume of Fluid, Conjugate Heat Transfer, Nusselt Number, Jet

INTRODUCTION When an axisymmetric free jet strikes a fixed surface, the flow field can be divided into an outer in viscid region and an inner viscous boundary layer. A very thin stagnation zone forms normal to the impingement axis. This layer exhibits little resistance to heat flow, where the convective heat transfer coefficient can reach tens of kW/m2K. Following impingement, the flow spreads thinner as it travels radially, i.e., the thickness of the liquid film adjacent to the wall decreases with radius. This decrease brings the growing boundary layer into contact with the surface of the fluid film. At this point, the fluid film thickness begins to increase at larger radii due to the viscous drag, which slows down the flow and thickens the liquid layer.

Fig. 1: Jet and Film Flow Field Showing Hydrodynamic Evolution [1]

The flow field of the wall jet can be divided into five consecutive regions as shown in Fig. 1 [1]; (1) the stagnation zone region, (2) the laminar boundary layer region, in which the viscous layer thickness is less than the liquid film; in this region, the liquid film free surface is assumed to have the same velocity of the incoming jet, (3) the viscous similarity region, in which the viscous boundary layer extends through the liquid film and the 34

surface velocity decreases as radius increases due to the viscous drag, (4) the transition region and (5) the fully turbulent flow region. Both experimental [2–5] and numerical [6–9] investigations have been carried out to study heat transfer characteristics of impinging jets. However, there are still many important issues to be addressed, such as the effect of a confined space and heat transfer within the disc material. These parameters are crucial in providing better insight into engine performance and various industrial applications. A numerical study to investigate the thermal characteristics for long, turbulent and free liquid jets impinging normally ontoa fixed disc with finite thickness and subjected to uniform heat flux surface was carried out in [10], using different nozzle diameters and flow conditions. The investigations revealed a distinct dependence of the normalized local Nusselt number on the radiallocation and only slight dependence on Re. The extent of the stagnation region beneath the jet was found tobe variable and not constant as suggested by many experimental studies; it is a function of the velocity gradient at the stagnation point. The effect of the extent of the stagnation zone on the Nusselt number is small for larger nozzles. However, for small diameter nozzles at higher Reynolds number, the difference in predicted stagnation zone Nusselt number can be as much as 22%. For a given Reynolds number, the temperature distribution on the impinging surface, i.e., fluid-solid interface, was found to be more uniform for larger nozzles compared to smaller nozzles. Smaller nozzles provide more efficient cooling at the stagnation region and subsequently lower temperature. An expressions to predict local and stagnation zoneNusselt number are given in [10], the correlation forstagnation zone Nusselt number is:

(1)

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Jet Impingement Heat Transfer: Stationary Disc

Nasif, Barron and Balachandar

Equation (1) predicts the computational data with an average error less than 10%, for all nozzle sizes and jet Reynolds numbers investigated in the study. The spacing term H/d does not appear in the above equation, because this equation is used to predict the stagnation zone Nusselt numbers for long jets, i.e., H/d > 5, where nozzle-to-target spacing is insignificant [11]. Experiments were performed in [12] to study the characteristics of an impinging turbulent jet onto a fixed surface subjected to constant heat flux, using different nozzle diameters and a wide range of Reynolds number. The investigations showed an obvious dependence of the stagnation zone Nusselt number on Reynolds number, Prandtl number and velocity gradient, and less dependency on nozzle-to-plate-spacing. The stagnation point velocity gradient is nozzle diameter dependent and is a linear function of uf/d in laminar jets. However, the influence of turbulence may result in a nonlinear dependence. The drawback of using uf/d as a correlating parameter is its dimensional nature, and there is no obvious reference time scale for use in its normalization. If uf/d is employed as a correlating parameter with other dimensionless parameters (Nu, Red, Pr, H/d), the empirical relation for stagnation zone Nusselt number is given in [12] as: (2) Equation (2) is valid for 4000 < Red < 52000 and predicts the experimental data with an average and maximum error of 5% and 14%, respectively, for all nozzles sizes used in the experiments. The objective of the present study is to numerically investigate the jet impingement thermal characteristics when an axisymmetric, turbulent, free oil jet impinges orthogonally onto a finite fixed circular disc with uniform heat flux, placed in a confined cylindrical space. Three nozzle geometries with same issuing bulk velocity were chosen to investigate the effect of nozzle geometry on the thermal characteristics. Two nozzle-to-disc elevations were used with a 1.0 mm pipe nozzle to investigate the effect of nozzle-to-disc spacings on the thermal characteristics.

storage and computing operations per cell is compensated for by a higher accuracy. However, in the current study, the structured mesh was found to be more convenient since it produced less smearing in comparison with a polyhedral mesh, especially if the cells are aligned long with the jet flow. For this reason the structured mesh was applied in the region where the numerical diffusion is expected, while the polyhedral mesh was applied to the rest of the computational domain (see next section). The k-ω SST turbulence model which is a two equation eddy-viscosity model is used as the turbulence model in the present study. Since the flow field involves two different immiscible fluids (oil jet in air), a numerical model to handle two-phase flow is required. Volume of fluid (VOF) is a simplified and efficient method which provides an approach to capture the movement of the interface between the mixture phases. High Resolution Interface Capturing (HRIC) [13] is the most common scheme used for interface capturing with the VOF model.

MODEL SETUP The present transient numerical simulation is used to predict steady-state thermal characteristics when an axisymmetric free oil jet with temperature at 130 °C impinges onto a fixed finite aluminum disc with 10 mm thickness placed in a cylindrical confined space. The computational domain, with relevant boundary conditions used in this simulation, is shown in Fig. 2. As illustrated in this figure, the cells were clustered along the jet trajectory, with minimum cell size of 0.075 mm, to prevent the smearing associated with numerical diffusion and preserve the sharpness of the oil-air interface. Fine prism cells were employed adjacent to the disc to resolve the thin film and reduce the artificial dissipation of the oil at these locations. A grid sensitivity study to determine an optimum cell size was previously carried out for a stationary boundary [10]. The final cell number chosen for subsequent simulations was based on negligible changes in the stagnation zone Nusselt number and the average disc temperature. A pipe nozzle with 1.0 mm diameter has been used to produce a fully developed turbulent pipe flow profile, which has been implemented

COMPUTATIONAL METHOD Finite volume based computations using CDA dapco’s STAR-CCM+ with both polyhedral and structured cells are performed in the current study. The major advantage of polyhedral cells is that they generally have many neighbors (typically of order 10), so gradients can be much better approximated using linear shape functions. Along the wall and at corners, a polyhedral cell is likely to have at least a couple of neighbors, which allows for a reasonable prediction of both gradients and local flow distribution. The fact that more neighbours means more

Fig. 2: Computational Domain and Relevant Boundary Conditions

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

35

Nasif, Barron and Balachandar

Jet Impingement Heat Transfer: Stationary Disc

as an inlet boundary condition to the computational domain. Other nozzle geometries, which have similar bulk exit velocity to the aforementioned pipe nozzle, will be used to examine the effect of nozzle geometry on the thermal characteristics. The Reynolds number of the issuing jet lies in the turbulent regime (i.e., Red ≈ 3000). A one-half segment of the entire geometry was used as a computational domain to reduce the CPU calculation time. Therefore, a symmetry boundary condition is assumed on the center plane of the domain. The other boundary conditions, such as pressure outlet and adiabatic surfaces, are used as shown in Fig. 2. The top surface of the 90 mm diameter disc is subjected to a uniform heat flux of q" = 50 kW/m2 [6], while the cylinder and solid disc outer surfaces are kept at constant temperature of T = 130 °C. All fluid and air properties, i.e., dynamic viscosity, density, thermal conductivity and specific heat are evaluated as a function of the local temperature in the computational domain. The heat transfer coefficient and consequently the Nusselt number are evaluated based on a specific reference temperature, i.e., the nozzle exit temperature T = 130 °C.

Effect of Nozzle Geometry In the current study, the effect of nozzle geometry on thermal characteristics was examined. Three nozzle geometries with exit diameter of d = 1.0 mm were employed as shown in Fig. 4. They consist of a pipe nozzle, a converging nozzle and a converging nozzle with a short pipe exit. A nozzle-to-target spacing of H/d = 60 was used in each case. The computational domain and relevant boundary conditions shown in Fig. 2 were used for this simulation. Fig. 4 shows the exit velocity profile for each of the three different nozzles. Although each nozzle has its own velocity profile, the bulk velocity remains constant and the profiles are essentially top-hat.

VALIDATION The numerical results for local and stagnation zone Nusselt numbers are validated for a stationary boundary [10]. The validation process given in [10] was employed for three nozzle sizes, over a wide range of Reynolds number. The empirical correlations to predict the local and stagnation zone Nusselt number given in [11] are used to compare with the numerical results. These comparisons revealed average differences in stagnation zone Nusselt number of 3.5, 5.0 and 8.0% corresponding to nozzles sizes of d = 1.0, 2.0 and 4.0 mm, respectively.

RESULTS AND DISCUSSION In the current simulation, the fluid-solid interface is split into nine regions for comparison purposes, as shown in Fig. 3. Region-1 in this figure represents approximately the stagnation region due to the jet impingement.

Fig. 3: Delineation of Nine Regions at the Fluid-Solid Interface

36

Fig. 4: Exit Velocities Profiles for the Three Nozzles

According to the published literature, the stagnation zone heat transfer coefficient (HTC) and corresponding Nusselt number are functions of the velocity gradient at the stagnation point. The velocity gradient is only a function of the velocity profile in the jet if the turbulence intensity is not strong [11]. For long jets, the viscosity tends to eliminate the gradient in the radial direction and hence creates a more uniform velocity profile which results in a constant velocity gradient at the stagnation point. Fig. 5 illustrates the elimination of velocity gradient for long jets. It is obvious from this figure that the velocity gradient in the radial direction diminishes after H/d > 16.0. A constant velocity gradient at stagnation point means constant thermal characteristics.

Fig. 5: Suppression of Velocity Gradient in Radial Direction (Velocity Relaxation)

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Jet Impingement Heat Transfer: Stationary Disc

Nasif, Barron and Balachandar

More insight into the thermal characteristics can be obtained by extracting the stagnation zone Nusselt number, stagnation zone temperature, disc average temperature and the interface surface average temperature for all nozzle geometries used in the study, as shown in Table 1. The key observation from the data in Table 1 is that, for long jets impinging on a stationary boundary, the nozzle geometry has no significant effect on the thermal characteristics. Table 1: Summary of Thermal Characteristics for Different Nozzle Geometries

Fig. 6: Steady-State Temperature Profile for Stationary Disc with Cooling Jet, at Two Nozzle-to-Disc Distances: (a) 20 mm and (b) 100 mm Table 2: Surface Average Nusselt Number at Solid-Fluid Interface, at Two Nozzle-to-Disc Distances: (a) 20 mm and (b) 100 mm

Jet Impingement onto a Stationary Disc Two steady-state simlations were run in the current study, with the disc fixed at two elevations, i.e., 20 and 100 mm away from the nozzle exit. The temperature profiles in the disc are shown in Fig. 6. The surface average heat transfer coefficients and corresponding Nusselt numbers are given in Table 2, only for the first four regions defined in Fig. 3. Important observations can be drawn upon close examination of Fig. 6 and Table 2. First, the maximum Nusselt number and subsequent minimum temperature occur at the center of the disc. Second, although the HTCs are almost comparable at the four consecutive regions in either case, the overall surface average HTC at the interface is higher with the longer jet. Consequently, the volume average temperature and stagnation zone temperature are less with the 100 mm jet in comparison with the 20 mm one. Figure 7 shows the comparison of the velocity magnitude contours for the two nozzle-to-disc distances. It is clear that the magnitude of the velocity is very similar up to region 4 in both cases, contributing to similar values of HTC. Beyond region 4, the velocity is higher in the case of the longer jet. Furthermore, the VOF contours (not shown here) indicate that there are differences in the VOF distribution in the two cases. It appears that the dynamics of the interaction is different with more splattering (enhanced mist) occurring in the shorter jet. The persistent contact of liquid with the disc and increased radial momentum of the flow provides for additional opportunities to enhance the HTC values in the longer jet.

(a)

(b)

Fig. 7: Comparison of the Velocity Magnitude Contours for Two Nozzle-to-Disc Distances: (a) 20 mm and (b) 100 mm

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

37

Nasif, Barron and Balachandar

Jet Impingement Heat Transfer: Stationary Disc

CONCLUSION

τ

Time constant (sec)

A numerical study of a circular oil jet impinging on a stationary disc with uniform wall heat flux was carried out using the volume of fluid method. The conclusions can be summarized as follow:

ν

Kinematic viscosity of fluid (m2/s)







The effect of nozzle geometry is insignificant on thermal characteristics. The viscosity tends to eliminate the velocity gradient in the radial directions for long jets, which results in a constant velocity gradient at the stagnation point. The values of surface average heat transfer coefficients and corresponding Nusselt numbers are nearly unchanged in regions close to the stagnation zone for long and short jets. The overall surface average HTC and corresponding Nusselt numbers are higher with a longer jet compared to a shorter one. This results in lower temperature at the stagnation zone, as well as lower disc volume average temperature with the longer jet.

REFERENCES [1]

[2]

[3]

[4]

[5]

[6]

[7]

NOMENCLATURE B

velocity gradient (dimensionless)

d

Nozzle diameter (mm)

H

Nozzle-to-disc spacing (mm)

[8]

[9]

Nu Nusselt number = hd/koil (dimensionless)

[10]

Nuo Stagnation zone Nusselt number (dimensionless)

38

Pr Prandtl number (dimensionless)

[11]

Red Reynolds number of the jet = uf d/ν(dimensionless)

[12]

r

Radius measured from point of jet impact (mm)

T

Temperature (°C)

uf

Bulk velocity at the nozzle exit (m/s)

ur

Radial velocity above viscous layer (m/s)

[13]

Liu, X., Lienhard, J.H. and Lombara, S. (1991), “Convective Heat Transfer by Impingement of Circular Liquid Jets”, J. Heat Transfer, Vol. 113(3), pp. 571–582. Ichimiya, K., Takema, S., Morimoto, S., Kunugi, T. and Akino, N. (2001), “Movement of Impingement Heat Transfer by a Single Circular Jet with a Confined Wall”, Int. J. Heat and Mass Transfer, Vol. 44, pp. 3095–3102. Ashforth, S., Jambunathan, K. and Whitney, C.F. (1997), “Velocityand Turbulence Characteristics of Semiconfined Orthogonally Impinging Slot Jet”, Experimental Thermal and Fluid Science, Vol. 14, pp. 60–67. Liu, X., Gabour, L.A. and Lienhard, J.H. (1993), “Stagnation-Point Heat Transfer During Impinging of Laminar Liquid Jets: Analysis Including Surface Tension”, J. Heat Transfer, Vol. 115, pp. 99–105. Liu, X. and Lienhard, J.H., “Liquid Jet Impingement Heat Transfer on a Uniform Flux Surface”, ASME Heat Transfer Division, Book No. H00498, 1989, Vol. 106, pp. 523–530. Agarwal, A.K., Goyal, S.K. and Srivastava, D.K. (2011), “Time Resolved Numerical Modelling of Oil Jet Cooling of a Medium Duty Diesel Engine Piston”, International Communications in Heat and Mass Transfer, Vol. 38, pp. 1080–1085. Hewakandamby, B.N. (2009), “A Numerical Study of Heat Transfer Performance of Oscillatory Impinging Jets”, Int. J. Heat and Mass Transfer, Vol. 52, pp. 396–406. Fuchang, X. and Mohammed, S. (2006), “Heat Transfer Behaviour in the Impingement Zone Under Circular Water Jet”, Int. J. Heat and Mass Transfer, Vol. 49, pp. 3785–3799. Rahman, M.M., Bula, A.J. and Leland, J. (1999), “Conjugate Heat Transfer during Free Jet Impingement of a High Prandtl Number Fluid”, Numerical Heat Transfer, Part B, Vol. 36, pp. 139–162. Nasif, G., Barron, R., Balachandar, R. and Iqbal, O. (2013), “Simulation of Jet Impingement Heat Transfer”, ASME 2013 Internal Combustion Engine Division (ICED), Fall Technical Conference, Dearborn, MI. Lienhard, J.H. (2006), “Heat Transfer by Impingement of Circular Free-Surface Liquid Jets”, 18th National and 7th ISHMTASME, Heat and Mass Transfer Conference, Guwahati, India. Stevens, J. and Webb, B.W. (1991), “Local Heat Transfer Coefficients under an Axisymmetric, Single Phase Liquid Jet”, J. Heat Transfer, Vol. 113(1), pp. 71–78. Muzaferija, S., Peric, M., Sames, P. and Schelin, T. (1998), “A Two-Fluid Navier-Stokes Solver to Simulate Water Entry”, Proceedings Twenty-Second Symposium on Naval Hydrodynamics, Washington, D.C. USA, pp. 638–651.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Development of High Temperature Coatings for Wear and Oxidation Resistance Using Cold Spray and HVOF Coatings A.S. Khanna* and W.S. Rathore Department of Metallurgical Engineering, IIT Bombay, Powai, Mumbai–400076 E-mail: *[email protected] Abstract—Cold Gas Deposition Spray (CGDS), using helium and nitrogen as career gases have been used to form CoCrAlOY coatings. The coatings obtained, were of different microstructures and morphologies and also of different surface mechanical properties. While, the coating made using helium as career gas, gave a relatively dense coating, the coating using nitrogen as career gas was porous. This was also reflected in their oxidation behaviour. The former coated surface gave an oxidation rate which was an order of magnitude lower than the latter. Keywords:

INTRODUCTION Cold spray is a very recent addition to thermal spray family. In fact, cold spray should be considered as complementary to thermal spray, as the temperature of the spray material is usually less than 500 oC in cold spray. The most common thermal spray techniques are flame, arc, Plasma spray and HVOF processes. They are differentiated in terms of their difference in kinetic energy of sprayed particles and the temperature required to melt the powder/wire in the gun. Flame spray has lowest temperature of about 3000 oC and low KE, arc has almost same KE as flame, but the temperature in the gun is about 5000 oC. HVOF is differentiated in terms of very high kinetic energy with temperature of gun same as in the case of flame spray gun. Plasma has very high temperature but low KE. As a result, one gets a difference in the coating properties, in terms of heterogeneous coating, if the temperature is low (flame, arc and HVOF) and homogenous if the temperature is high (plasma), and porous at low KE (Flame, Arc and Plasma) and dense at high KE (HVOF). Cold spray uses very high KE and a gun temperature of around 500 oC. Here the particles are not liquid droplets, but solids and because of their very high KE, deposition mechanism is not a splat cooling, but deposition due to high impact, which results in plastic deformation of solid particles, making a very adherent coating. That is why one of the requirements of the coating is that the powder particles can deform plastically. That is why, initially, it was thought that this technique is limited to metals such as copper, aluminium etc. But today it is possible to use superalloy powders to deposit bond coat. In this work, CoNiCrAlOY powder was used. Presence of Al, creates suitable deformation properties on impact. The same powder was deposited on SS substrate using HVOF and also by Cold Gas Dynamic Spray (CGDS) deposition process. In latter, again there are two categories, which give two different kinetic energies, when the gas used is nitrogen or helium. Latter gives much higher KE. The coatings made were applied on stainless steel, characterized and the subjected to wear and oxidation. One of the interesting results was that oxidation resistance of CGDS coating was much superior to that of

HVOF coated stainless steel. This was attributed to the retention of more aluminium on the surface which resulted in the formation of faster alumina layer and hence faster protection oxidation. In HVOF, part of the aluminium is either lost due to evaporation or due to dilution in the matrix. That is an important result. The wear and friction behaviour was also superior, but it was a marginal better than when compared with HVOF coated samples.

EXPERIMENTAL Commercially available CoNiCrAlY alloy powder, purchased from M/s. Metallizing Equipment Pvt. Ltd. Jodhpur (MEC 9950 AMF), has the chemical composition given in Table 1. The powder was characterized using XRD and SEM. The surface morphology and elemental composition (wt.%) at selected points is shown in Fig. 1. The powder particles have spherical morphology with particle size ranging between 10–42 µm. The observed composition of the powder is in considerable agreement to that supplied by the supplier. Since, the γ phase is a solid solution of Co, Ni, Cr etc. and as a higher mean atomic number, it appears brighter in BSE mode. Hence, the brighter phase is represented by γ and the darker phase as β-NiAl/CoAl, which has a lower mean atomic number and hence lower contrast in BSE mode [1–3]. Table 1: Chemical Composition of Powder Elements → Co Ni Cr Al O 38.25 32.33 20.51 8.90 00 Wt % →

Y 0.01

Fig. 1: XRD Results of CoNiCrAlY Powder Revealing a Two-phases, γ+β Structure

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

39

Khanna and Rathore

Development of High Temperature Coatings for Wear and Oxidation Resistance

Fig. 3: SEM Images of CGDS Coating Sprayed with He Gas in As-sprayed Condition Fig. 2: Schematic of the Cold Spray System [4] Table 2: Cold Spray Process Parameter Sprayed with Helium and N 2 Gases Parameter He Processing N2 Processing Values Values Gun temperature 400 oC 450 oC Gun pressure 20 bars 38 bars Powder feed rate 18 grams/min 15 grams/min Carrier gas flow rate 3.0 m3/hr 3.0 m3/hr

Cold spraying of the fine, CoNiCrAlY powder, was carried out with supersonic velocity onto 316L substrate, using a spray gun with converging-diverging nozzle at ASB Industries, Ohio. The schematic of the cold-spray process is shown in Fig. 2 [4]. The powder was fed through a high-pressure hopper, mixed with the preheated gas in a mixing chamber inside the gun, and was deposited at a very high pressure on to the substrate. In order to understand the effect of different specific-heat ratios in terms of coating performance, two different gas compositions, i.e., He and N2 were employed to carry and accelerate the CoNiCrAlY powder. The 316L stainless steel was grit blasted before deposition. The plateswere then coated using Nitrogen and Helium gases with desired process parameters as given in Table 2. The coated plates were cut into small pieces of (10  10  2 mm) size using wire Electric Discharge Machine for characterization and oxidation tests.

formation. The first layer of the coating builds up when particles impact on the substrate. The deposition continues on the former layers. As a result, the increasing plastic deformation reduces the porosity of the former layers. The coating surface was rough in the as-sprayed condition due to the presence of solid particles pinned at the surface of partially deformed particles. The coating was dense and uniform with a thickness in the range of 320–360 μm. Sonic velocity (v) in a media is given as v = (γRT/A)1/2, where γ is the specific heat ratio, R is universal gas constant, T is gas temperature, and A is molecular weight of the carrier gas. Due to low carrier gas density (0.1785 kg/m3), higher specific heat ratio (1.660), lower molecular weight (4 g/mol) of helium and low density of CoNiCrAlY powder (7824 kg/m3), a dense structure is obtained. Figure 4 (a, b, c) show the as received powder cross-section, as-sprayed powder and single powder particle microstructure of CGDS coating sprayed with He. The process of plastic deformation and reconsolidation of the impacting particles is distinct in this coating. It is very clear from Fig. 4(b) to see the heavy impact of particles in Helium-processed coating with most of particles exceeding the critical velocity and getting deposited onto the substrate [1, 4, 5].

Surface morphologies of the as-deposited and oxidized samples were investigated to assess the properties in the as-sprayed and oxidized samples using SEM. The coated samples were sectioned and polished using standard metallographic techniques to observe the coating cross-section.

RESULTS

Fig. 4: (a) Secondary Electron SEM Images of Etched Cross Section of Powder Revealing Dendrite Structure (b) As-sprayed CGDS Cross-section Revealing Deformed Dendrites

CGDS with He Gas Figure 3 (a, b, c) show the surface morphology and cross-section of the coating, showing dendritic microstructure of theas-sprayed CGDS coating using He. The microstructure confirms the intensive plastic deformation of CoNiCrAlY particles during deposition. Mechanical bonding is the main mechanism of the coating

40

Figure 5, further shows EDX analysis of surface and cross-section showing the view of an individual particle, deposited in as-sprayed condition. The EDAX analysis at some points indicates the dominance of Co and Ni and Cr in the composition of the coating, which is almost same as the composition of the sprayed powder.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Development of High Temperature Coatings for Wear and Oxidation Resistance

Khanna and Rathore

Fig. 7: (a) Secondary Electron SEM Images of Etched Cross Section of Powder Revealing Dendritic Structure, (b) As-sprayed CGDS Cross-section Revealing Deformed Dendrites and (c) Single Powder Particle in Deformed Condition

Fig. 5: SEM Images and EDX Compositions of CGDS Coating Sprayed with He Gas in As-sprayed Condition

CGDS with N 2 Gas Figure 6 (a) shows individual powder particles, not properly adhered to the surface, perhaps due to not acquiring the critical velocity required for the deposition. Hence, multi-impacts under N2 carrier gas display relatively higher degree of porosity, portraying low carrying capability of accelerating gas. This behaviour is due to the increase in density of the carrier gas (N 2 density is 1.2506 kg/m3), and thereby reduced degree of plastic deformation of the spray particles is expected for the coating. Thus, higher porosity levels, larger splat size, and reduced plastic deformation was observed when compared to those of helium-processed coating. Fig. 6(c) shows the surface morphology of the coating, revealing pores and voids and the presence of higher splats attributing to the multi-impacts on the coated layer. Figure 7(a, b, c) show the as received powder cross-section, as-sprayed powder and microstructure of sprayed particle. It is interesting to note that the dendritic microstructure of the feedstock powder is also found on the as-sprayed sublayer of CGDS coating, consisting of interlocked dendrites, where most of the dendrites appear to be deformed significantly, giving a flattened appearance as compared to original dendrites of the feedstock powder.

Fig. 6: SEM Image of CGDS Coating Sprayed with N 2 Gas in As-sprayed Condition

Fig. 8: SEM Images and EDX Compositions of CGDS Coating Sprayed with N2 Gas in As-sprayed Condition

Figure 8 shows the surface morphology of the as prayed coating and that of cross-section of the coating along with their EDX analyses. The EDAX analysis at some points indicates the dominance of Co and Ni and Cr in the composition of the coating, almost same as that of the sprayed powder. Table 3 compares some of the properties, such as roughness, porosity and thickness of the coatings deposited by CGDS using He and Nitrogen Table 3: Properties of He and N2 Processed CoNiCrAlY Sprayed Deposition Deposition Properties He Processing N2 Processing Thickness of coating 320–360 μm 105–115 μm Microstructural features Dense and Pores and porosity compact coating Surfaces roughness Values 23.41 ± 1.30 16.04 ± 1 Ra (µm) Porosity level (%) 0.9 ± 0.8 5 ± 0.9 Extent of plastic Insufficient plastic Lower degree of deformation of powder deformation plastic deformation particles

Nano-indentation was carried out on two coatings to see the difference in their surface mechanical properties as shown in Table 4. Coating deposited by He gas shows slightly better properties.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

41

Khanna and Rathore

Development of High Temperature Coatings for Wear and Oxidation Resistance

Table 4: Nanoindentation of CGDS Sprayed with He and N2 in As-sprayed Condition Process

CGDS He CGDS N2

Load P max (mN) 11000 11000

Hardness (GPa) 6.61 6.26

Reduced Modules Er (GPa) 153.38 145.04

Depth of Penetration H max (nm) 245 253

Fig. 10: Surface Morphologies of the Oxide Scale Formed on CGDS Coating Sprayed with He and N2 Gases after the Oxidation at 900 C for Different Times

Fig. 9: Mass Gain Curve Obtained during Isothermal Oxidation of the CGDS Sprayed with He and N 2 Coatings after the Oxidation at 900 C for 1000 hrs

Oxidation Studies

shows that transformations of microstructure have taken place during deposition. In CGDS coating, the absence of the β-phase is in accordance with findings [1]. During oxidation onset process θ-Al2O3 transforms into α-Al2O3, it retains its acicular morphology and subsequent growth of α-Alumina appears to be unaffected by the phase transformation [1]. Fig. 11 for CGDS coating reveals the existence θ-Alumina as dominant oxide (blade-like crystal) on surface and α-Al2O3 is dominant based on peak intensity.

The oxidation behaviour of the coated stainless steel was investigated using discontinuous oxidation tests in air. The weight changes vs time plots for the oxidation of two coatings using N2 and He as carrier gas, in air at 900 C are given in Fig. 9. The results show that the CGDS sprayed coating with He, shows significant lower oxidation rate than the CGDS sprayed coating with N2as career gas. The linear plots appear to follow parabolic kinetics. The lower oxidation rate of the CGDS sprayed with He samples can be attributed to the more denser and less porous coating, compared to that formed using N2 as carrier gas. This appears to be a most important advantage of CGDS-He process over the CGDS-N2 process.

CHARACTERIZATION OF OXIDE SCALES Figure 10 illustrates the surface morphologies of CGDS coatings, oxidized at 900 C for durations, 200, 500 and 1000 h for using He and Nitrogen as carrier gas. It can be seen that the surface is fully covered with an oxide scale having an acicular morphology. It appears that initially θAl2O3 usually grows in a needle-like morphology, while αAl2O3 usually grows in a dense equiaxed structure as reported by others [1, 6]. XRD analysis confirms θ-Al2O3, αAl2O3 and γ-solid solution (Fig. 11). It is observed that the assprayed coating do not retain the typical two-phase microstructure (γ-matrix Co-Ni-Cr solid solution and βNiAl/CoAl precipitates) initially found in the feedstock powder [4, 6]. The absence of β-phase in as-sprayed coating 42

Fig. 11: XRD Patterns of Oxidized CGDS Coating Sprayed with He Gas at the Temperature of 900 C for Different Times

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Development of High Temperature Coatings for Wear and Oxidation Resistance

The main difference between the CGDS coating using He and N 2 is in terms of the fact that since former shows only α-Al2O3throughout the oxidation, it protects from further oxidation, while in case of coating, deposited using nitrogen as carrier gas, the alumina layer changes to less stable CoCrAl spinel which enhances the oxidation.

GENERAL DISCUSSION In the present work, microstructural, morphological, and compositional analyses were conducted on 316L coated with CoNiCrAlY powder using He and N2carrier gases using cold spray technique. Both the carrier gases produced coatings with different morphologies and microstructures. It was observed that none of the assprayed coatings had retained the typical two-phase microstructure (γ-matrix Co–Ni–Cr solid solution and βNiAl precipitates) initially found in the feedstock powder which was similar to the findings of P. Richer et al. [1, 7]. Isothermal oxidation at 9000C showed that coating with He as carrier gas gave better oxidation resistance than that formed using nitrogen as carrier gas. This was interpreted in two ways. First due to the formation of more dense layer with lower porosity, the oxidation was restricted. Nitrogen carrier gas resulted in a coating which was less dense and had more porosity and hence lead to faster oxidation rate. The second reason is the formation of αAl2O 3 throughout on coating formed using He as carrier gas compared to the coating formed using nitrogen as career gas which initially gave α-Al2O 3 but latter it changed to more inferior CoCrAl spinel, leading to faster oxidation rate.

CoNiCrAlY powder was successfully deposited by the cold-spray technique using He and N2 as career gases.



Cold-spray deposition of CoNiCrAlY powder using helium as carrier gas exhibited dense morphological structure. This effect is attributed to the higher ratio of specific heats and lower mass density of helium as carrier gas.



The CGDS sprayed with N2 gas coating featured significant amounts of undesirable CoCr2O 4 mixed spinel-type oxides in the early stages of oxidation.



Oxidation kinetics of CGDS sprayed with He was an order of magnitude lower than that CGDS coated with N 2. This improvement can be associated partially porosity in CGDS coating, but more due to higher Al concentration on the surface, which resulted in a better selectively an alumina coating.

ACKNOWLEDGMENT The authors wish to thank, ASB Industries, Ohio, USA, for thermal spraying the bond coats samples.

REFERENCES [1]

[2]

[3]

[4] [5]

CONCLUSION 

Khanna and Rathore

[6]

[7]

Richer, P., Yandouzi, M., Beauvais, L. and Jodoin, B. (2010), “Oxidation Behavior of CoNiCrAlY Bond Coats Produced by Plasma, HVOF an Cold Gas Dynamic Spraying”, Surface and Coatings Technology, Vol. 204, pp. 3962–3974. Zhang, Q., Li, C.J., Li, Y., Zhang, S., Wang, X.R., Zhang, Q., Yang, G.J., and Li, C.X. (2008), “Thermal Failure of Nanostructured Thermal Barrier Coatings with Cold-Sprayed Nanostructured NiCrAlY Bond Coat”, J. Therm. Spray Technol., n Vol. 17(5–6), pp. 838–845. Tang, F., Ajdelsztain, L. and Schoenung, J.M. (2004), “Influence of Cryomilling on the Morphology and Composition of the Oxide Scales Formed on HVOF CoNiCrAlY Coatings”, Oxidation of Metals, Vol. 61, No. 314, pp. 219–238. Alkimov, A.P. and Kosarev, V.F. et al. (1990), Sov. Phys. Dokl., Vol. 35(12), pp. 1047–1049. Balani, K., Laha, T., Agarwal, A., Karthikeyan, J. and Munroe, N. (2005), “Effect of Carrier Gases on Microstructural and Electrochemical Behavior of Cold-sprayed 1100 Aluminum Coating”, Surface & Coatings Technology, Vol. 195, pp. 272–279. Saeidi, S., Voisey, K.T., and McCartney, D.G. (2009), “The Effect of Heat Treatment on the Oxidation Behavior of HVOF and VPS CoNiCrAlY Coatings”, J. Therm. Spray Technol., Vol. 18(2), pp. 209–216. Qian, Linmao, Ming, Li, Zhongrong, Zhou, Hui, yang and Xinyu, Shi (2005), “Comparison of Nano-indentation Hardness to Microhardness”, Surface and Coating Technology, Vol. 195, pp. 264–271.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

43

Surface Engineering Analysis of HVOF Sprayed Cr3C2-NiCr Coating under High-Temperature Oxidation V.N. Shukla*, R. Jayaganthan and V.K. Tewari Department of Metallurgical and Materials Engineering, Indian Institute of Technology Roorkee, Roorkee–247667, India E-mail: *[email protected] Abstract—In the present investigation, HVOF process has been used to deposit Cr3C 2-NiCr coating on 310S steel. The oxidation behavior of HVOF sprayed Cr3C2-NiCr coatings as well as the bare substrate, subjected to high temperature (900ºC) in air for 500 hours has been carried out in the present work. X-ray diffraction, Field emission–scanning electron microscopy/energy dispersive spectroscopy techniques were used to analyze the scales formed on the surface of the oxidized samples. High volume fraction of carbides contributed for the higher microhardness of the coating (6401014HV). The Cr3C2-25%NiCr coated specimen showed good adherence to the 310S SS substrate with negligible microspalling of the scales. The scale remained intact and adherent to the substrate. The development of continuous and protective oxides of chromium, nickel and spinel of nickel and chromium have contributed for the better oxidation resistance for coated specimen as compared to bare substrate (310S). Keywords: HVOF; Cr3C2–NiCr; Hot Corrosion

INTRODUCTION The high oxidation resistance of Fe based high-chromium and nickel containing alloys, Cr3C2-25% NiCr cermets are extensively used to fabricate corrosion and wear resistant coatings on steel components used in hot section of waste incineration boilers, electric furnaces and fossil fuel-fired boilers. Corrosion is a serious problem in superheater of coal fired boilers. Large amounts of material wastage in super heaters due to combustion of fuels are reported [1–7]. Literature reveals the wide contribution of thermal sprayed coatings for the protection of turbine engine and boiler tubes against high temperature corrosion attack [8–10]. Cr3C2-NiCr coatings are very stable and show comparatively superior resistance at high working temperatures among the other cermet coatings. Therefore, these coatings are widely used in steam turbine blades and boiler tubes for power generation [11, 12]. The influence of microstructural features on the oxidation behavior of HVOF sprayed Cr3C2-NiCr coatings and the degradation mechanisms of coatings exposed to high temperature environment (900 C) for 500 hours in air is limited in literature. Knowledge of the reaction kinetics and microstructural changes of the oxide scales formed during oxidation at high temperature are important for evaluating the performance of materials in high temperature applications [13–14]. Therefore, the present work has been focused to study the microstructural changes of high velocity oxy-fuel (HVOF) sprayed Cr3C2-25%NiCr coatings after exposure of 500 hours at 900 C by using FE-SEM/EDS and XRD, respectively.

EXPERIMENTAL SECTION The austenitic stainless steel (310S) was used as substrate in the present study. The chemical composition (wt.%) of the substrate material is given in Table 1. Spectrometric analysis was used to determine the chemical composition 44

of the substrate (310S stainless steel).Specimens with dimensions of approximately 20 × 15 × 5 mm3 was cut from the sheets. The specimens were polished and grit blasted with alumina powders prior to HVOF spraying of coatings. The size of the alumina powder used for grit blasting the substrate specimens was ~350 µm. Praxair-Tafa JP5000 HVOF torch was used to apply the Cr3C2-25% NiCrcoating, at M/s Industrial Processors and Metallizers (IPM), Pvt. Ltd, New Delhi, India using the standard spray parameters. The aim of this study was to evaluate the microstructural changes of both coated and uncoated specimens exposed at 900 C in air for 500 hours. The surface of the specimens wascleaned by polishing with alumina powder followed by cleaning by acetone. To remove the moisture, polished samples were pre-heated for 1 hour at 250 C. After exposure for 500 hours, the oxidized samples were analysed by FE-SEM/EDAX for the surface and cross sectional analysis. Table 1: Chemical Composition of Substrate Material Chemical Composition (wt%) of 310S C Cr Ni 0.0409 25.18 19.14

Mn 1.32

Si 0.486

Cu 0.189

Mo 0.177

P 0.0202

S Fe 0.004 Balance

RESULTS AND DISCUSSION Coating Characterization SEM micrograph in Fig. 1(a, b), shows the grain size and shape of the alloy powder with irregular shape. The surface microstructures of the as sprayed coating are seen in the BSE image (Fig. 2). The microstructures revealed that HVOF sprayed Cr3C2-25%NiCr coating consist of irregular shaped splats which are interconnected. The coated samples were cut in cross-section, mounted in transoptic powder and subjected to mirror polishing. Optical microscopy image of cross-section of coated specimen is shown in Fig. 3(a). Micro hardness of the Cr3C2-25%NiCr coatings was

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Surface Engineering Analysis of HVOF Sprayed Cr3C2-NiCr

Shukla, Jayaganthan and Tewari

determined across the cross-sectioned samples. The micro hardness of the as sprayed Cr3C2-25%NiCr coatings was measured by using Micro hardness Tester (VHM-002

Walter UHL, Germany) using 300 grams (2.941N) load applied for a dwell time of 15 sec. The microhardness data of the coatings are shown in Fig. 3(b).

a

b

Fig. 1: (a, b) FE-SEM Micrograph Showing the Morphology of Cr3C2-NiCr Powder Used Fordeposition of Coatings

Cr-rich phase

Ni-rich phase

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

45

Shukla, Jayaganthan and Tewari

Surface Engineering Analysis of HVOF Sprayed Cr3C2-NiCr

Fig 2: (a, b) BSE Image of Morphology of As-sprayed Cr3C2–25%NiCr Coating

In terface

1200

C oa tin g

Microhardness (Hv)

Sub strate 1000

a

800

600

400

b

200

-12 0

-8 0

-40

0

40 8 0 1 20 D istance (µm )

1 60

20 0

24 0

Fig. 3: (a) Cross-sectional Image of the As-sprayed Coating, (b) Variation of Microhardness Across the Cross Section of the Coating

Surface Morphology FE-SEM micrographs with EDS analysis for the bare and coated specimens exposed to high temperature environment (900 C) for 500 hours in air are shown in Fig. 4 and Fig. 5, respectively. The scale formed on alloy substrate was porous and less adherent. The scale consists mainly of Fe, Cr, Ni and O. Spalling and cracking were observed in case of alloy

46

substrate as shown in Fig. 3(a). The Cr3C2-25NiCr coatings deposited on 310S by the HVOF-spraying process behaved very well. The scale was adherent and minor spalling observed at the edges of the coated specimen. The scale consists mainly of Cr, C, Ni and O. Similar observations were reported by Sidhu et al. [15]. Wang also observed severe spalling of the oxide scale during cyclic study of 2.25Cr-1Mo steel at 740 C in coalfired boiler [16].

International Journal of Surface Engineering & Materials Technology, Vol. 4 No. 1 July-Dec 2014, ISSN: 2249-7250

Surface Engineering Analysis of HVOF Sprayed Cr3C2-NiCr

Shukla, Jayaganthan and Tewari

Fig. 4: FE-SEM/ EDS Surface Analysis of Bare Substrate Exposed to High Temperature Environment (900 C) for 500 Hours in Air

Fig. 5: FE-SEM/ EDS Surface Analysis of Cr3C 2-25%NiCr Coated Alloy Substrate Exposed to High Temperature Environment (900 °C) for 500 Hours in Air International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

47

Shukla, Jayaganthan and Tewari

Surface Engineering Analysis of HVOF Sprayed Cr3C2-NiCr

Coating

Fig. 6: Morphology of Oxide Scale Across the Cross Section of Cr3C2-25%NiCr coated Alloy Substrate Exposed to High Temperature Environment (900 °C) for 500 Hours in Air - Cr3C2A- Cr2O3 C-Austenite   NiCr2O4, B- (Cr,Mn,Fe)3O4

Intensity (arbitrary units)

A A

A

A

A



A



A

A



Cr3C2NiCr Coating

B C A

B

B

A A A

A

A

C C

Bare Substrate

20

30

40

50

60

Diffraction angle 2 Fig. 7: (a) XRD Patterns for Bare Substrate and As-sprayed Coating Exposed to High Temperature Environment (900 °C) for 500 Hours in Air

48

International Journal of Surface Engineering & Materials Technology, Vol. 4 No. 1 July-Dec 2014, ISSN: 2249-7250

Surface Engineering Analysis of HVOF Sprayed Cr3C2-NiCr

Shukla, Jayaganthan and Tewari

Cross-sectional Analysis

REFERENCES

The EDS analysis of the scale of the coated specimen (Fig. 6) shows the presence of Cr, Ni, O, C, Mn, Si and Fe. The presence of Fe, Mn and Si on top surface of the coating is probably due to their diffusion in to the coating through substrate-coating interface. The dark gray regions were Cr rich where as white regions are NiCr, which play the role of binder. The coating is intact with the substrate and there is no sign of crack at the interface. Long term exposure of the Cr3C2-NiCr has not been reported so far.

[1]

X-ray Diffraction (XRD) Analysis The major and minor phases identified from the XRD patterns of the scale formed on bare and Cr3C2-NiCrcoated specimens during cyclic oxidation in air at 900 οC for 500 hours are shown in Fig. 7. The uncoated surfaces indicates the presence of Cr2O3 and austenite phase as major phase, whereas (Cr, Mn, Fe)3O4 as minor phase. The coated surfaces after oxidation for 500 hours indicates the presence of Cr 2O 3, Cr3C2, as major phases and NiCr2O4 as minor phases. The formation of nickel chromium oxide or chromium oxide-rich scale in the present study probably acted as barrier to the inward diffusion of oxygen into the coating.

CONCLUSION Cr3C2-25NiCr coatings were deposited successfully on 310S by the HVOF-spraying process.Cr3C2-25NiCr coating consisted of very dense structure. The microhardness of the Cr3C2-25NiCr coatings was in the range of 775–1014Hv.Cr3C2-25NiCr coated specimens exhibited excellent oxidation and wear/ spallation resistance. The oxide scale of Cr 3C2-25NiCr coated specimen showed the presence of mainly oxides of Cr and Nickel, which might be responsible for better corrosion resistance. There is perceptible diffusion of some elements from the substrate to the coating. The HVOF sprayed Cr3C2-25NiCr coating was found to be successful in maintaining its adherence with the substrate even after 500 hrs of exposure.

[2]

[3]

[4] [5] [6]

[7] [8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

ACKNOWLEDGMENTS The authors would like to express their thanks toM/s Industrial Processors and Metallizers (IPM), Pvt. Ltd., New Delhi, India for providing the powders and coating facilities.

[16]

Priyantha, N., Jayaweera, P., Sanjurjo, A., Lau, K., Lu, F., Krist, K., (2003), “Corrosion-resistant Metallic Coatings for Applications in Highly Aggressive Environments”, Surf. Coat. Technol., Vol. 163–164, pp. 31–36. Kaur, Manpreet, Singh, Harpreet and Prakash, Satya (2011), “Surface Engineering Analysis of Detonation-gun Sprayed Cr3C 2NiCr Coating under High-temperature Oxidation and Oxidation– Erosion Environments”, Surf. Coat. Technol., Vol. 206, pp. 530–541. Buta Singh Sidhu, S. Prakash (2006), “Studies on the behaviour of Stellite-6 as Plasma Sprayed and Laser Remelted Coatings in Molten Salt Environment at 9000C under Cyclic Conditions”, Journal of Materials Processing Technology, Vol. 172, pp. 52–63. Collins, S. (1997), “Biomass-fired Plants Face down Near-term Challenges”, Power, Vol. 137, pp. 51–53. Nielsen, H.P., Dam-Johansen, K. and Baxter, L.L. (2000), Progress in Energy and Combustion Science, Vol. 26, pp. 283–298. Wang, B.Q. (1996), “Effect of Alkali Chlorides on ErosionCorrosion of Cooled Mild Steel and Cr3C2-NiCr Coating”, Wear, Vol. 199, pp. 268–274. Wang, B. (1996), “Erosion-corrosion of Thermal Sprayed Coatings in FBC Boilers”, Wear, Vol. 199, pp. 24–32. Kamal, S., Jayaganthan, R. and Prakash, S. (2009), “Evaluation of Cyclic Hot Corrosion behaviour of Detonation Gun Sprayed Cr3C2–25%NiCr Coatings on Nickel-and Iron-based Superalloys”, Surface and Coatings Technology, Vol. 203, pp. 1004–1013. Kamal, S., Jayaganthan, R. and Prakash, S. (2009), “High Temperature Oxidation Studies of Detonation-gun-sprayed Cr 3C2– NiCr Coating on Fe-and Ni-based Superalloys in Air under Cyclic Condition at 9000C”, Journal of Alloys and Compounds, Vol. 472, pp. 378–389. Kaur, Manpreet, Singh, Harpreet and Prakash, Satya (2008), “High-Temperature Corrosion Studies of HVOF Sprayed Cr3C 2NiCr Coating on SAE 347H Boiler Steel”, Journal of Thermal Spray Technology, Vol. 18(4), pp. 619–631. Staia, M.H., Valente, T., Bartuli, C., Lewis, D.B., Constable, C.P., Roman, A., Lesage, J., Chicot, D. and Mesmacque, G. (2001), “Part II: Tribological Performance of Cr3C2-25%NiCr Reactive Plasma Sprayed Coatings Deposited at Different Pressures”, Surf. Coat. Technol., Vol. 146–147, pp. 563–570. Murthy, J.K.N. and Venkataraman, B. (2006), “Abrasive wear behavior of WC-CoCrand Cr3C2-20(NiCr) Deposited by HVOF and Detonation Spray Processes”, Surface and Coatings Technology, Vol. 200, pp. 2642–2652. Hussain, N., Shahid, K.A., Khan, I.H. and Rahman, S. (1994), “Oxidation of High Temperature Alloys (Super Alloys) at Elevated Temperatures in Air”, J. Oxid. Met., Vol. 41 (3–4), pp. 251–269. Matthews, S., Hyland, M. and James, B. (2003), “Microhardness Variation in Relation to Carbide Development in Heat Treated Cr3C2-NiCr Thermal Spray Coatings”, Acta Materialia, Vol. 51, pp. 4267–4277. Sidhu, B.S., Puri, D. and Prakash, S. (2004), “Mechanical and Metallurgical Properties of Plasma Sprayed and Laser Remelted Ni3Al Coatings on Boiler Tube Steels”, Materials Science and Engineering A, Vol. 368, pp. 149–158. Wang, D. (1988), “Corrosion Behaviour of Chromized and/ or Aluminized 21/4Cr-1Mo Steel in medium-BTU Coal Gasifier Environments”, Surf. Coat. Technol., Vol. 36, pp. 49–60.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

49

Investigation on Multi-Response Parameter Optimization of Cold Spray Coating Process Tarun Goyal1*, R.S. Walia2 and T.S. Sidhu3 1

Professor, SUS College of Engineering and Technology, Tangori, Mohali, Chandigarh, India–140306 2 Associate Professor, Delhi Technological University, Delhi, India–110042 3 Director, SBSCET, Ferozepur, Punjab, India–152004 E-mail: *[email protected], [email protected]

Abstract—Cold spray (CS) process is a relatively new coating deposition thermal spray process and a lot of research is being carried out throughout the world towards the optimization of the process with an aim towards the performance improvement of the process. For optimization of process parameters, most of the existing approaches for multi response optimization of process parameters focus upon the subjective and practical knowledge available about the process. Keeping in view these limitations, an approach based on a Utility theory and Taguchi quality loss function (TQLF) has been applied to low-pressure cold spray (LPCS) process to deposit copper coatings, for simultaneous optimization of more than one response characteristics. In the present paper two potential response parameters i.e. coating thickness (CT) and micro hardness (MH) have been selected. Utility values based upon these response parameters have been analyzed for optimization by using Taguchi approach. Keywords: Cold Spray (CS), Optimization, Taguchi Method, Utility Concept, Coating Thickness, Micro Hardness

INTRODUCTION The cold gas-dynamic spray method (CGSM), hereafter referred to simply as cold spray (CS), is a relatively new process by which coatings of ductile materials (or composite materials with significant ductile phase content) can be produced without significant heating of the sprayed powder. The kinetic energy of the particles is sufficient to produce large deformations and high interfacial pressures and temperatures, which appear to produce a solid-state bond [1]. Cold-spray processing was developed in the former Soviet Union more than a decade ago as an offshoot of supersonic wind tunnel testing [2, 3]. CS is a process of applying coatings by exposing a metallic or dielectric substrate to a high velocity (300–1200 m/s) jet of small (1–50 μm) particles accelerated by a supersonic jet of compressed gas. The two main clear cut distinctions of the Low pressure cold gas-dynamic spray (LPCGDS) system from the High pressure cold gas-dynamic spray (HPCGDS) system are: the utilization of low pressure gas (5–10 bars instead of 25–30 bars) and the radial injection of powder instead of axial injection. The accelerating gas (usually air or N 2) is injected at low pressure (5–10 bars) and preheated within the gas heater to temperatures up to about 400 °C to optimize its aerodynamic properties. Solid powder particles are radially introduced downstream of the throat section of the supersonic nozzle thus eliminating the need for a high pressure delivery system, which increases system portability, operational safety and significantly reduces spraying costs. Within the nozzle, static pressure is maintained below the atmospheric pressure ensuring that feedstock particles are effectively drawn in from the powder feeder by Venturi effect [4]. Fig. 1 shows a schematic of the LPCGDS system. In the modern competitive non-conventional manufacturing scenario, it is most vital to optimize the parameters of a process to exploit its full utility. Practically, 50

Fig. 1: A Typical LPCGDS Device [5]

it is seen that one particular setting of input parameters for a response characteristics may not be suitable for other characteristics of the process/product. In most of the manufacturing processes, more than one quality characteristics has to be considered for optimization of process parameters making it necessary that several response characteristics have to be simultaneously optimized. Based on the foregoing discussions, in this paper, Taguchi method is briefly reviewed for the multiresponse optimization. The multi-response optimization of the response parameters of LPCS process is presented by using the experimental data. Optimization models have been developed by combination of the Taguchi Method and the Utility concept. The multi-response optimization of quality characteristics i.e. coating thickness and micro hardness of LPCS has been carried.

EXPERIMENTAL PROCEDURE Table 1 show the process parameters that were identified as potential important in affecting the quality characteristics of the LPCS process under consideration [6]. The process parameters, their designated symbols and ranges are also given in Table 1. The Taguchi’s mixed level design was selected as it was decided to keep two levels of powder feeding arrangement. The rest four parameters were studied at three levels. The effect of selected process parameters was studied on the following response characteristics of LPCS process:

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Investigation on Multi-Response Parameter Optimization of Cold Spray Coating Process

Goyal, Walia and Sidhu

Table 1: Process Parameters and Their Range Process Parameters Range Level 1 Feed Type Gravity, Argon Gravity Substrate Material Al alloy, Brass, Ni Al alloy alloy C Stagnation pressure 104–120 psi 104 D Stagnation temperature 350–4000 C 350 E Stand-off distance 2.5–7.5 mm 2.5 Nozzle type: Converging-diverging, Carrier gas: Air, Powder size: < 45 µm Symbol A B

Exp No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Total

R1 28 52 74.5 58.4 44.7 25.8 14.7 55.1 53.9 68.9 38.6 58.2 43.2 28.3 64.8 56.9 57.9 17.3 841.2

T CT

Level 2 Argon Brass

Level 3 -------Ni alloy

112 375 5.0

120 400 7.5

Table 2: Experimental Results of Various Response Characteristics Coating Thickness (mils) CT Micro hardness (H v0.3) MH S/N Ratio (dB) R2 R3 R1 R2 R3 26.2 30.4 28.95 122 126 124 51.5 52.2 34.30 134 129 132 74.2 74.7 37.43 131 132 122 58.2 58.7 35.33 126 128 124 44.1 45.4 33.01 124 125 128 24.3 25.5 28.01 126 127 128 11.2 14.6 22.39 129 131 132 55.6 55.5 34.86 137 133 134 54.5 54.3 34.68 134 135 139 64.1 69.7 36.57 123 118 111 38.2 38.7 31.70 119 115 117 57.5 57.9 35.24 126 131 129 43.3 43.6 32.74 125 121 122 27.2 25.5 28.60 110 119 113 65.4 64.6 36.24 122 120 124 57.1 59.5 35.23 129 125 127 56.2 56.7 35.10 125 127 128 19.1 16.7 24.91 123 128 132 827.9 844.2 585.4 2265 2270 2266

= overall mean of CT= 46.54

T MH

S/N Ratio (dB) 41.86 42.38 42.15 42.00 41.98 42.07 42.32 42.58 42.66 41.36 41.36 42.18 41.77 41.12 41.72 42.07 42.05 42.11 755.80

= overall mean of MH = 125.94

R1, R2, and R3 represent repetitions

a.

Coating thickness (CT)

b.

Micro hardness (MH)

Coating thickness and micro hardness are “higher the better” type of quality characteristics. A simplified multicriterion methodology based on Taguchi’s approach and utility concept (given below) is used to achieve the objective of this study. The observed values of response parameters are given in Table 2.

Utility Concept Utility can be defined as the usefulness of a product or a process in reference to the expectations of the users. The overall usefulness of a process/product can be represented by a unified index termed as Utility which is the sum of the individual utilities of various quality characteristics of the process/ product. The methodological basis for Utility approach is to transform the estimated response of each quality characteristic into a common index. If Xi is the measure of effectiveness of an attribute (or quality characteristic) i and there are n attributes evaluating the outcome space, then the joint Utility function can be expressed [7] as:

UX1, X2 ,...Xn  f U1 (X1),U2 (X2 )....Un (Xn )

(1)

th

where U i (Xi) is the utility of the i attribute. The overall Utility function is the sum of individual utilities if the attributes are independent, and is given as follows: n

U X1 , X 2 , ...Xn    U i (Xi )

(2)

i 1

The attributes may be assigned weights depending upon the relative importance or priorities of the characteristics. The overall utility function after assigning weights to the attributes can be expressed as: n

U X1 , X 2 , ...Xn    Wi U i (X i )

(3)

i 1

where Wi is the weight assigned to the attribute i, the sum of the weights for all the attributes must be equal to 1.

Determination of Utility Value A preference scale for each quality characteristic is constructed for determining its utility value. Two arbitrary

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

51

Goyal, Walia and Sidhu

Investigation on Multi-Response Parameter Optimization of Cold Spray Coating Process

numerical values (preference number) 0 and 9 are assigned to the just acceptable and the best value of the quality characteristic respectively. The preference number (Pi) can be expressed on a logarithmic scale as follows [8, 9]: X Pi  A  log  i'  Xi

  

(4)

where Xi = value of any quality characteristic or attribute i

X i' = just acceptable value of quality characteristic

(a) Preference scale for CT (PCT):

A = constant The value of A can be found by the condition that if Xi = X* (where X* is the optimal or best value), then Pi = 9 A

X* = Optimal value of CT = 72.36 (refer Table III)

X i' = Just acceptable value of CT = 11 (All the observed values of CT are greater than 11)

9 X* log ' Xi

Following equation is obtained from equation 4:

X  PCT 11.00  log  CT   11 

The overall utility can be calculated as follows: n

U   Wi Pi

(5)

i 1

(6)

(b) Preference scale for MH (PMH): X* = Optimal value of MH = 138.36 (refer Table III)

n

subject to the condition:  Wi  1 i 1

Among various quality characteristics type viz. smaller the better, higher the better, and nominal the better, suggested by Taguchi, the Utility function would be higher the better type. Therefore, if the Utility function is maximized, the quality characteristics considered for its evaluation will automatically be optimized (maximized or minimized as the case may be).

ANALYSIS AND DISCUSSIONS Based upon the methodology developed in the previous section, following case have been considered to obtain the optimal settings of the process parameters of LPCS for predicting the optimal values of combined responses. Two quality characteristics i.e. Coating Thickness (CT) and Micro hardness (MH) have been included in utility response. Taguchi L18 orthogonal array (OA) [10] has been adopted for conducting the experiments. Powder feeding Table 3: Optimal Setting and Values of Process Parameters Response Optimal Level Significant Predicted Optimal Characteristics of Process Process Value of Quality Parameters Parameters Characteristics CT A2, B1, C3, A, B, C, D, E 72.36 mils D3, E2 MH A1, B3, C3, A, B, C, D, E 138.36 H v0.3 D2, E2

52

Following is the stepwise procedure for transforming experimental data into utility data.

Construction of Preference Scales

or attribute i

Therefore,

arrangement (A), Substrate material (B), air stagnation pressure (C), air stagnation temperature (D), and stand-off distance (E) were selected as input parameters. Response parameters (quality characteristics) were coating thickness and micro hardness, when they are optimized individually; the summary of results is produced in Table 3.

X i' = Just acceptable value of MH = 110 (All the observed values of MH are greater than 110) Following equation is obtained from equation 4:

X  PMH  90.34  log  MH   110 

(7)

Calculation of Utility Value Equal weights (1/2 each) have been assigned to the selected quality characteristics assuming both the quality characteristics, are equally important. However, these weights can be varied depending upon the case or user requirements, if any. The following relation was used to calculate the utility function based upon the experimental trials:

U(n, r)  PCT (n, r)  WCT  PMH (n, r)  WMH (8) where

1 WCT  ; 2

WMH 

1 2

n is the trial number (n = 1, 2, 3, …., 18) and r is the repetition number (r = 1, 2, 3). The calculated Utility values are shown in Table 4.

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Investigation on Multi-Response Parameter Optimization of Cold Spray Coating Process

Goyal, Walia and Sidhu

Analysis of Utility Data for Optimal Setting of Process Parameters

maximum values of the utility and S/N ratio within the experimental space.

The average and main response in terms of Utility values and S/N ratio (Tables 5 and 6) are plotted in Figure 2. It can be observed from Figure 2 (i) to (v) that first level of powder feed arrangement (A1), Third level of substrate material (B3), Third level of air stagnation pressure (C3), first level of air stagnation temperature (D1) and Second level of stand-off distance (E2) are expected to yield a

The pooled version of ANOVA for utility data and S/N ratio are given in Tables 7 and 8 respectively. It can be noticed from Table 7 that all the input parameters have significant effect (at 95% confidence level) on the utility function. Similarly, it had been found from Table 8 that all the chosen parameters in study have significant effect on the S/N ratio of utility function.

Table 4: Calculated Utility Data Based on Responses CT and MH Utility Values R1 R2 R3 1 4.26 4.74 4.78 2 7.58 6.81 7.30 3 8.00 8.14 6.61 4 6.65 6.95 6.35 5 5.70 5.82 5.36 6 3.70 3.71 3.98 7 3.82 3.47 4.25 8 8.16 7.60 7.74 9 7.67 7.84 8.40 10 6.57 5.59 4.59 11 4.54 3.85 4.22 12 6.64 7.38 7.09 13 5.18 5.14 5.02 14 2.26 3.71 2.54 15 6.77 6.97 6.58 16 7.05 6.44 6.85 17 6.48 6.72 6.89 18 4.27 4.29 4.57 R1, R2, R3 = repetitions of experiments against each of the trial conditions Trial Number

S/N Ratio (dB) 13.20 17.15 17.47 16.44 14.99 11.57 11.61 17.86 18.01 14.65 12.40 16.92 14.17 8.49 16.60 16.60 16.50 12.81

7.2

17 16.5

6.7

15.5

6.2

15 5.7

14.5 14

5.2

13.5

Raw Data 4.7

S/N Ratio

Utility values (Raw data)

16

13

S/N Ratio

12.5 4.2 12 3.7

(i)

11.5

(ii)

(iii)

(iv)

(v)

Fig. 2: Average and Main Response in Terms of Utility Values and S/N Ratio

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

53

Goyal, Walia and Sidhu

Investigation on Multi-Response Parameter Optimization of Cold Spray Coating Process

Table 5: Average and Main Effects (Raw Data: CT and MH) Average Utility Values Main Effects Difference L1 L2 L3 L2-L1 L3-L2 (L3-L2)-(L2-L1) A 6.13 5.49 -0.64 -0.64 B 6.04 5.13 6.25 -0.91 1.12 2.03 C 5.43 5.74 6.26 0.31 0.52 0.21 D 6.15 5.54 5.73 -0.61 0.19 0.80 E 3.94 7.05 6.43 3.11 -0.62 -3.73 L1, L2 and L3 represents average values of raw data of corresponding parameters at levels 1, 2 and 3 respectively. L2-L1 is the average main effect when the corresponding parameter changes from level 1 to level 2. L3-L2 is average main effect when the corresponding parameter changes from level 2 to level 3. A-Powder feed arrangement, B- Substrate material, C-air stagnation pressure, D-air stagnation temperature, E-stand-off distance. Process Parameter Designation

Table 6: Average S/N Values and Main Effects (Raw Data: CT and MH) Process Parameter Designation

S/N Average Values Main Effects (dB) Difference L1 L2 L3 L2-L1 L3-L2 (L3-L2)-(L2-L1) A 15.37 14.35 -1.02 -1.02 B 15.30 13.71 15.56 -1.59 1.85 3.45 C 14.44 14.56 15.56 0.12 1.00 0.88 D 15.52 14.61 14.44 -0.91 -0.17 0.75 E 11.68 16.93 15.96 5.25 -0.97 -6.21 L1, L2 and L3 represents average values of S/N data of corresponding parameters at levels 1, 2 and 3 respectively. L2-L1 is the average main effect when the corresponding parameter changes from level 1 to level 2. L3-L2 is average main effect when the corresponding parameter changes from level 2 to level 3. A-Powder feed arrangement, B- Substrate material, C-air stagnation pressure, D-air stagnation temperature, E-stand-off distance. Table 7: Pooled ANOVA (Raw Data: CT and MH) Source SS DOF V F-Ratio A 5.48 1 5.48 47.94* B 12.68 2 6.34 55.48* C 6.30 2 3.15 47.94* D 3.46 2 1.73 15.13* E 100.79 2 50.39 440.70* E (Pooled) 5.03 44 0.11 Total (T) 133.76 53 *Significant at 95% confidence level. SS= Sum of Squares, DOF= Degree of Freedom, V= Variance, SS’= Pure Sum of Squares

SS’ 5.37 12.46 6.08 3.23 100.56 6.06 133.76

P% 4.01 9.32 4.54 2.42 75.18 4.53 100

Table 8: S/N Pooled ANOVA (Raw Data: CT and MH) Source SS DOF V A 4.65 1 4.65 B 12.01 2 6.04 C 4.52 2 2.26 D 4.08 2 2.04 E 93.63 2 46.82 E (Pooled) 2.97 8 0.37 Total (T) 121.96 17 *Significant at 95% confidence level. SS= Sum of Squares, DOF= Degree of Freedom, V= Variance, SS’= Pure Sum of Squares

Optimal Values of Quality Characteristics (Predicted Means) The optimal values of utility and thus the optimal values of response characteristics in consideration are predicted at the above levels of significant parameters. The average values of all the response characteristics at the optimum levels of significant parameters with respect to Utility function are recorded in Table 9. The optimal values of the predicted means (µ) of different response characteristics can be obtained from the following equation:

54

F-Ratio 12.53* 16.31* 6.10* 5.50* 126.22* -

SS’ 4.28 11.36 3.78 3.34 92.90 6.30 121.96

  A1  B 3  C 3  D1  E 2  4 T

P% 3.51 9.31 3.10 2.74 76.16 5.17 100

(9)

where, A1-First level of powder feed arrangement, B3Third level of substrate material, C3- Third level of air stagnation pressure, D1- first level of air stagnation temperature and E2-Second level of stand-off distance. Table 9: Average Values of Various Responses at Optimal Levels of Utility Function Levels Coating Thickness, CT (mils) Micro hardness MH (HVN) A1 45.12 129.33 B3 42.60 130.44 C3 49.07 128.28 D1 50.21 125.28 E2 57.73 128.33 Note: The above average values are taken from experimental data

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

Investigation on Multi-Response Parameter Optimization of Cold Spray Coating Process

The 95% confidence interval of confirmation experiments (CICE) can be computed [10] by using the following equation:

 1 1 CI CE  Fα (1, f e ) Ve     n eff R 

average of the observed values of the response characteristics fall well within the 95% CICE of the optimal range of the respective response characteristics. Table 10: Observed Values of Quality Characteristics (Confirmation Experiment) Exp. No. CT (mils) MH (H v0.3) r1 r2 r3 r1 r2 r3 1 58.2 57.6 59.3 134 136 139 2 57.9 59.2 57.5 141 138 142 3 59.4 60.1 58.7 137 140 133 Overall Average 58.6 137.78

(10)

where, Fα (1, fe) = The F-ratio at the confidence level of (1-α) against DOF 1 and error degree of freedom fe , R = Sample size for conformation experiments, Ve = Error N variance, neff = 1 DOF N= total number of trials, and DOF= Total degrees of freedom associated in the estimate of mean response. (a) For Coating Thickness (CT)

CT  A1 B3  C3  D1 E2  4TCT  58.57 where A1 = 45.12, B3 = 42.60, C3 = 49.07, D1= 50.21, E2 = 57.73 (Table 9): T CT = 46.54 (Table 2) The following values have been obtained by the ANOVA: N = 54, fe = 44, ve = 2.00, neff = 5.4, R= 3, F0.05 (1, 44) = 4.064 From equation 10, CICE = ± 2.05 The predicted optimal range (for conformation runs of three experiments) for CT is given by CICE: 56.52 < µCT < 60.62 (b) For Micro hardness (MH)

 MH  A1  B 3  C 3  D1  E 2  4 TMH  137 .9 where A1 = 129.33, B3 = 130.44, C3 = 128.28, D1= 125.28, E2 = 128.33, (Table 9): TMH = 125.94 (Table 2) The following values have been obtained by the ANOVA: N = 54, fe = 44; ve = 19.77, neff = 5.4, R= 3, F0.05 (1, 44) = 4.064 From equation 10, CICE = ± 6.45 The predicted optimal range (for conformation runs of three experiments) for MH is given by CICE: 131.45 < µMH < 144.35

Confirmation Experiment For confirmation of experimental results, three experiments were performed at optimal settings as suggested by Taguchi analysis of Utility data. The observed values of various response characteristics have been given in Table 10. It can be noticed that overall

Goyal, Walia and Sidhu

CONCLUSION The important conclusions of this research work are enlisted below: 1. Quality of coatings produced by LPCS process can be improved using Taguchi and Utility approach. 2. The selected input parameter significantly improves the Utility function (raw data and S/N ratio) comprising of quality characteristics (coating thickness and micro hardness). 3. The optimal setting of the process parameters for a multi-characteristic product can be predicted using the model. 4. The decreasing order of percentage contribution of the parameters to achieve a higher value of utility function is: stand-off distance (75.18%), substrate material (9.32%), air stagnation pressure (4.54%), powder feeding arrangement (4.01%) and air stagnation temperature (2.42%).

REFERENCES Papyrin, A.N. (2006), “Cold Spray: State of the Art and Applications”, Cold Spray Technology, Albuquerque, NM, USA, pp. 1–2. [2] Davis, J.R. (Ed.) et al. (2004), Handbook of Thermal Spray Technology, Materials Park, OH, USA, ASM International®, 1st Ed. [3] Papyrin, A., Kosarev, V., Klinkov, S., Alkhimov, A. and Fomin, V. (2007), Cold Spray Technology, London, Elsevier. [4] Maev, R. Gr. and Leshchynsky, V. (2008), Introduction to Low Pressure Gas Dynamic Spray Physics and Technology, Weinheim, Wiley2VCH. [5] Grujicic, M., Zhao, C.L., Tong, C., DeRosset, W.S. and Helfritch, D. (2004), “Analysis of the Impact Velocity of Powder Particles in the Cold-gas Dynamic-spray Process”, Materials Science and Engineering A368, Elsevier, pp. 222–230. [6] Goyal, Tarun, Walia, R.S. and Sidhu, T.S. (2011), “Effect of Parameters on Coating Thickness for Cold Spray Process”, J. Materials and Manufacturing Processes, Vol. 27, No. 2, pp. 185–192. [7] Bunn, Derek W. (1982), Analysis for Optimal Decisions, New York, John Wiley and Sons, p. 275. [8] Gupta, V. and Murthy, P.N. (1980), An introduction to Engineering Design Methods, New Delhi, Tata McGraw Hill, pp. 78–98. [9] Kumar, P., Barua, P.B. and Gaindhar, J.L. (2000), “Quality Optimization (Multi-Characteristics) through Taguchi Technique and Utility Concept”, J. Quality and Reliability Engineering International, Vol. 16, pp. 475–485. [10] Roy, R.K. (1990), A Primer on Taguchi Method, New York, Van Nostrand Reinhold. [1]

International Journal of Surface Engineering & Materials Technology, Vol. 4, No. 1, July–Dec 2014, ISSN: 2249-7250

55