Int J Adv Manuf Technol (2012) 60:611–623 DOI 10.1007/s00170-011-3642-6
ORIGINAL ARTICLE
Surface roughness optimization of cold-sprayed coatings using Taguchi method Tarun Goyal & Ravinderjit Singh Walia & T. S. Sidhu
Received: 11 October 2010 / Accepted: 13 September 2011 / Published online: 27 September 2011 # Springer-Verlag London Limited 2011
Abstract In this paper, Taguchi L 18 orthogonal array have been employed for depositing the electro-conductive coatings by varying various process parameters, i.e., substrate material, type of powder feeding arrangement, stagnation gas temperature, stagnation gas pressure, and stand-off distance. The response parameter of the coatings so produced is measured in terms of surface roughness. The optimum process parameters are predicted on the basis of analyses (ANOVA) of the raw data and signal to noise ratio. The significant process parameters in order of their decreasing percentage contribution are: stagnation pressure, stand-off distance, substrate material, stagnation temperature of the carrier gas, and feed arrangement of the powder particles, respectively. Keywords Low-pressure cold spray (LPCS) . Coating . Surface roughness . Taguchi optimization . Signal to noise ratio (S/N)
1 Introduction Surface treatment procedures are used to modify the surface properties of various materials without altering their bulk T. Goyal Punjab Technical University, Jallandhar, Punjab, India e-mail:
[email protected] R. S. Walia (*) PEC University of Technology, Chandigarh, Punjab, India e-mail:
[email protected] T. S. Sidhu SBSCET, Ferozepur, Punjab, India e-mail:
[email protected]
characteristics. These techniques are used within industrial environments to improve resistance to corrosion, wear, fatigue, and heat. A comparison of competitive spraying technologies clearly shows that cold gas dynamic spraying has been established as a viable coating technology in the thermal spray processes family. The dense and oxide-free coatings that may be produced through gas dynamic spraying (GDS) have brought about a multitude of new applications which, up to now, have not been feasible using traditional processes. Practical solutions may be readily observed within such industries as automotive and electronics manufacturing. Despite this great progress, however, the full potential of high-pressure and low-pressure GDS has not yet been fully realized [1]. Cold gas dynamic spray process is emerging as a boon in the thermal spray category for producing coatings so as to avoid material degradation in the field of surface engineering. The cold spray process was originally developed in the mid-1980s at the Institute of Theoretical and Applied Mechanics of the Russian Academy of Sciences in Novosibirsk by Dr. Anatolii Papyrin and his colleagues. They successfully deposited a wide range of pure metals, metal alloys, and composites onto a variety of substrate materials, and demonstrated the feasibility of the cold spray process for various applications. A US patent was issued in 1994, and the European patent in 1995 [2]. Cold spray coatings may be classified as high-pressure cold spray (HPCS) and low-pressure cold spray (LPCS) on the basis of the stagnation pressure of the working gas. Table 1 shows classification of the two versions of the cold spray process on the basis of their operating process parameters. Cold gas dynamic spray (or simply cold spray) is a process of applying coatings by exposing a metallic or dielectric substrate to a high velocity (300–1,200 m/s) jet of small (1–50 μm) particles accelerated by a supersonic jet of compressed gas. This process is based on the selection of
612
Int J Adv Manuf Technol (2012) 60:611–623
Table 1 Classification of cold spray process on the basis of their process parameters [3] Process characteristic
High-pressure system
Low-pressure ystem
Working gas
N2, He, air
N2, air
Gas pressure, MPa Gas preheat, °C Gas flow rate, m3/h
2.5–4.5 20–800 50–150
0.5–1.0 20–550 15–30
Maximum gas mach no. Powder flow rate, g/s
1–3 0.1–1.0
1–3 0.1–1.0
Particle size, μm
5–100
10–80
the combination of particle temperature, velocity, and size that allows spraying at the lowest temperature possible. In the cold spray process, powder particles are accelerated by the supersonic gas jet at a temperature that is always lower than the melting point of the material, resulting in coating formation from particles in the solid state. As a consequence, the deleterious effects of high-temperature oxidation, evaporation, melting, crystallization, residual stresses, debonding, gas release, and other common problems for traditional thermal spray methods are minimized or eliminated [4]. Eliminating the deleterious effects of high temperature on coatings and substrates offers significant advantages and new possibilities and makes cold spray promising for many industrial applications. Figure 1 shows the operating principle of low-pressure cold spray process. In this paper, surface coatings have been produced by cold spray process by selecting Taguchi L 18 OA design. This research work is an attempt for development of coatings by low-pressure cold spray process for electrotechnical applications. The process parameters selected for producing the coatings are: stagnation pressure, stagnation Fig. 1 Operating principle of low-pressure cold spray process [5]
temperature of the carrier gas, type of powder feeding arrangement, substrate material, and stand-off distance. The coatings so produced are analyzed on the basis of surface roughness obtained on the coated specimens. The aim of the present work is to optimize process parameters for surface roughness of low-pressure cold-sprayed copper coatings. An attempt has been made to do so in the present research by coating copper powder on the substrate— aluminum; brass and nickel alloy using low-pressure cold spray process so that the frequent contact of dissimilar materials may be avoided. The substrates have been selected keeping in view their actual applications in the industrial components. The information arising out from the investigation will be useful to explore the possibility of use of the low-pressure cold spray coatings on different materials so as to achieve a lower surface roughness value of the so-formed coatings.
2 Literature review The cold spray process was developed in 1994 and is still in infancy but various researchers and scientists all over the world have been contributing significantly towards the improvement in the process for wide range of application area. The research on the process is focused on process attributes, design and modeling, process parameter effects, nozzle design parameters, bonding mechanism, bow-shock phenomenon, adiabatic shear instability, post spraying treatments, and exploring new combinations and advancements of the process to suit wide-spread application areas. Some of the important research contributions in regard to the effect of various process parameters on the improvement of the process are presented in this section.
Int J Adv Manuf Technol (2012) 60:611–623
2.1 Effect of carrier gas temperature Lee et al. [6] presents the effects of gas temperature on critical velocity and deposition characteristics in kinetic spraying. This study was carried out to determine the influence of process gas pressure and temperature on the kinetic spray deposition of bronze powder onto aluminum and mild steel substrates. It was found that increasing the gas pressure caused an increase in particle velocity, while increasing the gas temperature not only affected the particle velocity but also the particle temperature. Increasing the particle temperature could enhance thermal softening, which is important for bonding. They found that there exists two critical velocities—particle bonding on substrate (Vcr1) and particle–particle bonding (Vcr2). The critical velocity decreased by 50 m/s when the process gas temperature was increased by 100°C. The difference between the critical velocity for particle deposition onto the aluminum substrate (Vcr1) and for particle–particle bonding (Vcr2) was about 160 m/s. The difference between the critical velocity for particle deposition onto the mild steel substrate (Vcr1) and for particle–particle bonding (Vcr2) was higher than about 330 m/s. Bond strength at the substrate/coating interface mainly depends on the first stage critical velocity (Vcr1). 2.2 Effect of carrier gas pressure Li et al. [7] studied the effect of gas pressure on Al coating by cold gas dynamic spray. They found that gas pressure as one of the processing parameters can have an influence on the coating properties. They suggested that the peening effect of low-pressure cold spray be suitably considered. 2.3 Coating powder properties Ning et al. [8] studied the effects of powder properties on in-flight particle velocity and deposition process during low-pressure cold spray process. The results showed that the irregular shape particle presents higher in-flight velocity than the spherical shape particle under the same condition. Critical velocities of about 425 m/s and more than 550 m/s were estimated for the feedstock copper powder with spherical and irregular shape morphology, respectively. For irregular shape particles, the in-flight velocity decreased from 390 to 282 m/s as the particle size increases from 20 to 60 mm. For the irregular shape particles, the critical velocity decreased from more than 550 to 460 m/s after preheating at 390°C for 1 h. It was also found that the larger size powder presents a lower critical velocity in this study. Li et al. [9] presented significant influence of particle surface oxidation on deposition efficiency, interface
613
microstructure, and adhesive strength of cold-sprayed coatings. They found that gas and particle temperature and particle oxygen content significantly affect the deposition efficiency which increases with increasing gas and particle temperature and decreases with increase in oxygen content. The hardness for low and high oxygen content copper powder was comparable but the oxide inclusions at the coating interface was more with high oxygen content powder. The adhesive strength of the coatings were found to decrease with increasing oxygen content as it inhibited the effective bonding of fresh metals. 2.4 Process conditions Shin et al. [10] presented a study on the influence of process parameters such as feed rate, spray distance, and particle velocity on deposition characteristics of soft/hard composite coating in kinetic spray process. The results showed that the high diamond fraction in the coating can be achieved using a low feed rate, intermediate spray distance, and high impact particle velocity. The possibility of impact between hard brittle diamond particles is the main factor affecting the diamond fraction in the coating. Although the deposition efficiency, diamond fraction, and bond strength of the coating increase with particle velocity, a slight decrease of cohesive strength between diamond particle and bronze base was also observed. 2.5 Spray parameters Steenkiste et al. [11] studied kinetic spray coatings. They found that new high-velocity spray apparatus has provided the capability of controlling process parameters important to the kinetic spray process. This process forms coatings via conversion of particulate kinetic energy to mechanical and thermal deformation of particles upon impact with a substrate. These coatings were found to have relatively low porosity values, hardness comparable with the corresponding bulk materials, adhesive strengths as high as 68–82 MPa, and oxide contents essentially the same as in the powders. In DOE the nozzle air inlet pressure, inlet air temperature, nozzle-substrate stand-off distance, and powder feed rate were varied. Al, Cu, and Fe powders were sprayed onto Al and brass substrate. Substrate effects appeared to be relatively weak in this experiment. Porosity and maximum coating thickness in a single pass were measured. Porosity values were found to be less than 1% for Cu and Fe, but in the case of Al porosity was sensitive to the spray parameters. Al and Fe had similar average single-pass thicknesses, while Cu thicknesses were much higher. In fact, Cu thicknesses were sensitive primarily only to powder feed rates.
614
2.6 Incidence angle Gang et al. [12] discuss the effect of different incidence angles on bonding performance in cold spraying. It was found that the contact area between the deformed particle and substrate decreases and the crater depth in the substrate reduces with increasing the tilting angle at the same impact velocity. The normal component of impact velocity takes an important role in the impacting process and formation of bonding. Li et al. [13] presented numerical investigations of the effect of oblique impact on particle deformation in cold spraying. They showed that the tangential component of incident velocity create a gap between deformed particle and substrate, decreasing contact area, and deteriorating bonding. They felt that the maximum deposition efficiency may not be found at normal angle; however, at an angle range in which the deposition efficiency may be promoted under the combined positive effect of shear friction and negative effect of tangential movement. 2.7 Stand-off distance Li et al. [14] studied the effect of stand-off distance on coating deposition characteristics in cold spraying. It was found that the deposition efficiency was decreased with the increase of stand-off distance from 10 to 110 mm for both Al and Ti powders used in this study. However, for Cu powders, the maximum deposition efficiency was obtained at the stand-off distance of 30 mm, and then the deposition efficiency decreased with further increasing the stand-off distance to 110 mm. The stand-off distance had a little effect on coating microstructure and micro hardness for these three powders. Both the strain-hardening effect of the deposited particles and the shot-peening effect of the rebounded particles take the roles in coating hardness. It was also found that the surface of substrate or previously deposited coating could be exposed to a relatively high gas temperature at a short stand-off distance. Pattison et al. [15] discuss the effect of stand-off distance and bow-shock phenomena in cold spray process. The bow shock formed at the impingement zone plays a critical role in the cold spray process; not only does it reduce the velocity of the gas, but also that of the entrained particles. Therefore at small stand-off distances, when the strength of the bow shock is high, deposition performance is reduced. While at large stand-off distances, when the bow shock has disappeared, deposition can continue unhindered. 2.8 Impact velocity analysis Grujicic et al. [16] discusses the analysis of the impact velocity of powder particles in the cold gas dynamic spray
Int J Adv Manuf Technol (2012) 60:611–623
process. They concluded that using a nonlinear regression analysis and a numerical solution for the one-dimensional isentropic gas flow in a cold gas dynamic spray nozzle, a relatively simple function is defined which relates the gas velocity at the nozzle exit with the nozzle expansion ratio and the carrier gas stagnation properties. Further, they found that to compute the velocity at which particles impact the substrate surface, deceleration of the particles in a stagnant subsonic region adjacent to the substrate surface must be considered. Wu et al. [17] present the measurement of particle velocity and characterization of deposition in aluminum alloy by kinetic spraying process. The results showed that particle velocity increases with increasing the process gas pressure and temperature. At higher temperature (or pressure), gas pressure (or temperature) do more effect particle velocities. They found that the substrate surface roughness do not have great effect on deposition efficiency. In this research, parameters such as particle and substrate temperatures effect were not studied. Helfritch et al. [18] in their paper gave a model study of powder particle size effect in cold spray deposition. It has often been thought that the smallest particles attain highest velocities in the cold spray process, thus achieving good deposition efficiency. While it is true that small particles exit the nozzle at high velocity, their velocity at impact can be significantly lower. Modeling efforts showed that the low gas velocity following the bow shock wave decreases particle velocities, especially for the smallest particles. It was shown that impact velocity increases as the particle diameter decreases until a diameter of 4 to 8 μm is reached. Impact velocity then decreases as the particle diameter is further reduced. Schmidt et al. [19] present development of a generalized parameter window for cold spray deposition. Calculations and experimental results demonstrate that size effects in impact dynamics can have a significant influence on the critical velocity, which has to be exceeded for bonding, and thus on the amount of bonded area and coating quality. The results obtained for copper and steel 316 L clearly demonstrate that critical velocities decrease with increasing particle size, which can be attributed to effects by heat conduction or strain rate hardening, respectively. Based on that analysis, a simple expression was supplied which allows an easy estimation of critical velocities for different metallic materials. The generalized approach developed in this study supplies the tools needed to predict optimum spray conditions and required powder size cuts for successful cold spraying of various materials. Li et al. [20] studied the high velocity impact of micro size particles in cold spraying. They concluded that both the compression ratio and flattening ratio of the particles increase with increase in particle velocity. They further
Int J Adv Manuf Technol (2012) 60:611–623
615
Table 2 Nominal chemical composition of the substrate materials chosen [23] Nominal chemical composition of ASTM B 221 (Al alloy) %Si
%Fe
%Cu
%Mn
%Mg
0.4–0.8 0.7 0.15–0.40 0.15 0.8–1.2 Nominal chemical composition of ASTM B 36 (Brass) % Cu 64–68.5
%Cr
%Zn
%Ti
%Others
%Al
0.04–0.35
0.25
0.15
0.15
Rem.
% Pb 0.15
% Fe 0.05
% Zn Remainder
Nominal chemical composition of ASTM B 435 (Ni alloy) %Fe 31
%Ni 20
%Co 18
%Cr 22
%Mo 3
%W 2.5
found that the compression ratio is more convenient for characterizing the extent of particle deformation than the flattening ratio owing to easy estimation and its dependency on meshing size. 2.9 Impact phenomena Klinkov et al. [21] analyzed the significance of particle impact phenomenon. In this study, a classification of impact phenomena is made based on particle size and impact velocity. The study showed that results of particle impacts depend not only on impact velocity but also on particle size. One can separate some typical ranges of impact velocities and particle sizes, where features of impacts are similar. Impacts result in two contrasting phenomena: destruction (also called erosion, cratering etc.) and adhesion (also called attachment, sticking). At hyper-velocity impacts, cratering and destruction are typical features which exhibit minor scale effects. At low impact velocities, there is a transition from erosion to adhesion (sticking) when the particle size decreases. Here, the nature of adhesion is due to Van der Waals and electrostatic forces. At high velocities, the results of impact depend not only on size and velocity but also on other parameters (e.g., plasticity, Fig. 2 Principle scheme of low-pressure portable cold spray machine (SST, Centreline, Windsor, Canada) [1]
%Mn 1
%Si 0.4
%Ta 0.6
%Al 0.2
%C 0.1
%N 0.2
%Zr 0.02
%La 0.02
particle flux concentration, etc.). In the regime of cold spray there is a competition between adhesion and erosion. Under these conditions a temperature variation in a rather minor range can result in a transition from erosion to adhesion. Wu et al. [22] studied the rebound phenomenon in kinetic spraying deposition. They concluded that a rebound phenomenon was observed in which a high particle velocity caused a high fraction of rebound particles. A maximum impact velocity was found for the particle deposition onto the substrate. The deposition of individual particles was controlled by the adhesion energy and the rebound (elastic recovering) energy. The impacting particles could only be attached to the substrate when the adhesion energy was higher than the rebound energy.
3 Experimental procedure 3.1 Substrate material The substrate materials selected were Al (ASTM B 221), brass (ASTM B 36), and Ni (ASTM B 435) in the rolled sheet form. The substrate materials selected for the study finds application in the manufacture of electrical contact points, fuse element of
616
Int J Adv Manuf Technol (2012) 60:611–623
Table 3 Design parameters and their chosen levels considered for the Taguchi experiment
Nozzle type, converging– diverging; carrier gas, air; powder size, <45 μm
Symbol
Process parameter
Range
Level 1
Level 2
Level 3
A
Feed type
Gravity, argon
Gravity
Argon
–
B C
Substrate material Stagnation pressure
Al, brass, Ni 104–120 psi
Aluminum 104
Brass 112
Nickel 120
D
Stagnation temperature
350–400°C
350
375
400
E
Stand-off distance
2.5–7.5 mm
2.5
5.0
7.5
electric mains plug, battery terminals, bimetallic joints, heat sinks, waste incinerators, gas turbines, brazing and soldering alloys, and many more. The nominal composition of the substrate material used for the study is mentioned in Table 2. The samples were cut from the alloy sheet to form approximately 25×15×5-mm sized specimens. The specimens were polished and grit blasted with alumina powders (grit 60) before being cold sprayed. 3.2 Coating formulation The coating powder selected was commercially available copper (less than 45 μm diameters) obtained from Centreline (Windsor), Ltd. (Windsor, ON, Canada; material ID, 440-00251 and the Catalogue number, SST-C 5001). The carrier gas used in the system was compressed air. The
coating deposition was done at Surface and Coatings Laboratory, University of Alberta, Edmonton, Canada using low-pressure cold spray equipment—SST LPCS Model # SSM-P3800-001, produced by Centreline Windsor, Canada (Fig. 2). 3.3 Development of coatings Based on the literature review, it was concluded that the processing parameters that influence quality of the coatings in the most significant way [24, 25] are working gas temperature, working gas pressure, working gas composition (He, N2, or air), size distribution of powder particles, spray distance, and nozzle type. In this study, the following parameters were selected for investigating their effect on surface roughness in LPCS
Table 4 Taguchi L 18 OA with measured surface roughness values (raw data and S/N ratios) Trial no. Input parameters
Surface roughness (μm)
Feed type Substrate material Air pressure (psi) Air temp. (°C) Stand-off distance (mm) R1
R2
R3
Response S/N ratio
1 2 3
GF GF GF
Al Al Al
104 112 120
350 375 400
2.5 5 7.5
13.56 10.87 8.05
13.71 10.91 7.81
13.83 10.36 7.73
−22.73 −20.60 −17.91
4 5 6 7 8
GF GF GF GF GF
Brass Brass Brass Ni Ni
104 112 120 104 112
350 375 400 375 400
5 7.5 2.5 2.5 5
9.2 8.17 8.7 11.17 6.69
10.04 8.02 7.5 11.22 6.68
10.68 10.23 8.6 11.56 7.35
−19.99 −18.95 −18.36 −21.07 −16.79
9 10 11 12 13 14 15 16 17 18 TOTAL
GF AF AF AF AF AF AF AF AF AF
Ni Al Al Al Brass Brass Brass Ni Ni Ni
120 104 112 120 104 112 120 104 112 120
350 400 350 375 375 400 350 400 350 375
7.5 7.5 2.5 5 7.5 2.5 5 5 7.5 2.5
6.89 6.97 7.02 8.31 8.26 9.59 9.24 9.6 9.52 8.42 8.29 8.92 9.04 9.78 9.62 7.19 7.79 7.82 6.18 6.97 6.39 7.94 8.13 8.15 7.19 8.62 7.34 7.94 7.11 7.84 154.75 157.41 162.55
−16.85 −18.83 −19.51 −18.63 −19.54 −17.62 −16.28 −18.14 −17.77 −17.66 −337.29
GF (SST) gravity feeder (Supersonic Spray Technologies), AF (SM) argon feeder (Sulzer Metco), Al aluminum (ASTM B221), Brass brass (ASTM B36), Ni nickel (ASTM B435), ASTM American Society of Testing Materials, R1, R2, R3 represent the surface roughness values for three repetitions of each trial, respectively
Int J Adv Manuf Technol (2012) 60:611–623
617
Table 5 Average values and main effects of surface roughness (raw data) Level
Feed type
L1 L2
9.390 8.192
L3
*
L2−L1 L3−L2
−1.197 *
Substrate material
Air pressure
Air temperature
Stand-off distance
9.832 8.440
10.211 8.533
9.053 9.415
9.661 8.454
8.101
7.629
7.905
8.258
−1.392 −0.339
−1.678 −0.903
0.362 −1.510
−1.207 −0.196
L1, L2, and L3 represent levels 1, 2, and 3, respectively, of parameters. L2–L1 is the average main effect when the corresponding parameter changes from level 1 to level 2. L3–L2 is the average main effect when the corresponding parameter changes from level 2 to level 3
process: feed type, substrate material, air stagnation pressure, air stagnation temperature, and stand-off distance and the parameters kept constant during the entire experiments are: nozzle type, working gas (air), size distribution of powder particles. The design parameters as well as their chosen levels considered for the Taguchi experiment are listed in Table 3. Table 4 shows the Taguchi L18 OA followed for conducting the experiments at each condition and the response measured at each condition in the form of surface roughness of the developed coatings. 3.4 Measurement of surface roughness for the coatings The surface roughness was measured of the samples with the help of Surface Roughness Tester, Mitutoyo, Japan make, Model SJ 400 for a resolution of 0.000125 μm and maximum measuring range of 800 μm. The surface roughness data were analyzed to determine the effect of various process parameters. The quality characteristic for calculation of S/N ratio of surface roughness was taken as of lower-the-better type [26, 27]. The experimental results of the surface roughness were transformed into S/N ratio at each condition of measured value. Taguchi recommends the use of S/N ratio to measure the quality characteristics deviating from the desired values. A greater S/N ratio corresponds to better-quality characteristics. Therefore the optimal level of the process parameters was the
level with the higher S/N ratio. The S/N ratio for the lowerthe-better type of response can be computed [26, 27] as: "
ðS=NÞLB
R 1 X ¼ 10 log Y2 R j¼1 j
# ð1Þ
where Yj, j=1, 2…n are the response values under the trial conditions repeated R times. Analysis of variance (ANOVA) was performed to identify the process parameters that were statistically significant. With the S/N ratio and ANOVA analyses, the optimal combination of the process parameters was predicted.
4 Results and discussions The average values of surface roughness and the S/N ratio for each parameter at level L1, L2, and L3 were calculated and are given in Tables 5 and 6, respectively. These values have been plotted in Fig. 3. Figure 3a shows the variation of surface roughness with respect to the feed arrangement of the powder particles. These results shows that surface roughness obtained when coatings are made by spraying powder through gravity feeder (SST, Canada) is comparatively higher. However, the surface roughness decreases with the use of Argon Feeder arrangement (Sulzer Metco, USA).The S/N ratio is higher
Table 6 Average values and main effects of surface roughness (S/N ratio) Level
Feed type
L1 L2 L3 L2−L1 L3−L2
−19.253 −18.223 * 1.029 *
Substrate material
Air pressure
Air temperature
Stand-off distance
−19.705 −18.460 −18.050 1.244 0.409
−20.052 −18.543 −17.619 1.509 0.924
−18.859 −19.411 −17.944 −0.551 1.466
−19.495 −18.409 −18.311 1.086 0.097
L1, L2, and L3 represent levels 1, 2, and 3, respectively, of parameters. L2−L1 is the average main effect when the corresponding parameter changes from level 1 to level 2. L3−L2 is the average main effect when the corresponding parameter changes from level 2 to level 3
618
a
b -17.0
16.0
-18.0 Gravity(SST)
14.0 B221
Argon(SM)
B36
B435
-17.5 S/N ratio
-19.0 14.0
Raw data
12.0
Raw data
-21.0
12.0 Raw data (µm)
S/N ratio
S/N ratio
-20.0
Raw data (µm)
-18.0 S/N ratio
Fig. 3 Variation of surface roughness (raw data and S/N ratio) with process parameters a feed arrangement, b substrate material, c air pressure (psi), d air temperature (°C), and e stand-off distance (mm)
Int J Adv Manuf Technol (2012) 60:611–623
-18.5
10.0
-19.0
10.0 -19.5
-22.0
Substrate material (ASTM specifications)
c -16.0
-17.0
120
10.0 350
-17.5
8.0
-18.0
-18.0 6.0 -19.0
S/N ratio
10.0
Raw data (µm)
112
-17.0
S/N ratio
d
12.0
375
400
9.0 S/N ratio
-18.5
4.0
-19.0
2.0
-19.5
0.0
-20.0
raw data
8.0
Raw data (µm)
Type of feed
104
8.0
-20.0
8.0
-23.0
S/N ratio -20.0
raw data
-21.0
2.5
5
7.5
16.0
-18.0
S/N ratio
-19.0 14.0 -20.0 S/N ratio Raw data
12.0
-21.0
Raw data (µm)
e
7.0 Air Temperature (o C)
Air pressure (psi)
10.0 -22.0
8.0
-23.0 Stand-off distance (mm)
with the argon fed powder arrangement as compared to that for gravity feeding. The above results show that as the surface roughness value is lower, when the powder particles are fed by means of argon feeding arrangement and a higher value of S/N ratio at this feeding arrangement suggests a stronger signal and less noise and thus provides an optimum result.
It can be observed from Fig. 3b that the surface roughness obtained on the substrate material B435 is the lowest. However, it is higher for the substrate material B36 and highest for the substrate B221. The values of S/N ratio correspond to raw data values for those obtained for different substrates. The above results show that the surface roughness of the coating depends on the substrate material
Int J Adv Manuf Technol (2012) 60:611–623
619
Fig. 4 a–r SEM microstructure of the coated specimens (as per design Table 4)
and is not only dependent on the impact velocity of the striking particles. The property of high hardness of the material B435 may help in achieving the sufficient plastic deformation of the striking particles on impact resulting in better surface finish of the coatings [20]. From the Fig. 3c, it can be noted that as the air pressure increases from 104 to 120 psi, there is decrease in the value of surface roughness obtained. The best surface roughness (lowest value) is obtained for the highest pressure of 120 psi for the chosen range. The surface roughness obtained is highest at 104 psi air pressure. The S/N ratio corresponds to raw data values of surface roughness obtained at different levels of air pressure. S/N ratio, being
highest for a pressure of 120 psi, decreases with decreasing pressure of the carrier gas, air. The same was expected as an increase in stagnation pressure of the carrier gas increases the impacting velocity of the particles striking the substrate. The higher the impacting velocity of the striking particles better is the adhesion to the substrate and lower is the surface roughness value of the so formed coating. Figure 3d shows the variation of surface roughness with respect to the stagnation temperature of the carrier gas. These results show that the surface roughness obtained is minimum at temperature of 400°C and increases with decrease in temperature. However, the highest value of surface roughness is obtained at a level of temperature of
620
Int J Adv Manuf Technol (2012) 60:611–623
Fig. 4 (continued)
air, i.e., 375°C. The S/N value is maximum at a temperature of 400°C, followed by 350°C and 375°C. Therefore, the S/N ratio values are in agreement with the experimental obtained Table 7 Pooled ANOVA (raw data)
Source Feed type
a
Significant at 95% confidence level
SS sum of squares, DOF degree of freedom, V variance, SS′ pure sum of squares
Substrate material Air pressure Air temperature Stand-off distance Error Total
SS
raw data values. The stagnation temperature of the carrier gas helps in obtaining the required impact velocity and plastic deformation of the striking particles, as the particles are DOF
19.35
1
30.31 61.75 22.37 38.52 0.01 172.34
2 2 2 2 44 53
V
F ratio
SS′
P (%)
19.35
50,005.51a
15.15 30.87 11.18 19.26 0.00038 –
39,156.13a 79,775.03a 28,898.73a 49,765.51a – –
19.35 30.31 61.75 22.37 38.52 0.02 172.34
11.23 17.58 35.83 12.98 22.35 0.01 100
Int J Adv Manuf Technol (2012) 60:611–623 Table 8 Pooled ANOVA (S/N data)
621
Source
SS
Feed type
a
Significant at 95% confidence level
SS sum of squares; DOF degree of freedom; V variance; SS′ pure sum of squares
Substrate material Air pressure Air temperature Stand-off distance Error Total
F ratio
SS′ 4.51 8.39 17.58 6.06 4.66
10.45 19.52 39.68 14.42 11.35
4.41 45.63
4.54 100
1
4.77
18.39a
8.91 18.10 6.58 5.18 2.07 45.63
2 2 2 2 8 17
4.45 9.05 3.29 2.59 0.25 –
17.17a 34.90a 12.68a 9.98a – –
Table 9 Average values of various responses at optimal levels
A2 B3 C3 D3 E3 T
V
4.77
attached to the substrate resulting in coating only if they achieve a velocity in the critical range for the chosen parameters and plastic deformation at the impact. The above results suggest that the higher the stagnation temperature of the carrier gas, the better is the coating quality in terms of surface roughness as it also helps the striking particles to achieve the velocity higher than the critical velocity of the particles and also aids in plastic deformation/thermal softening of particles. From Fig. 3e, it can be observed that the surface roughness value decreases as the stand-off distance is increased from 2.5 to 7.5 mm. The surface roughness of the coating is lowest at a stand-off distance of 7.5 mm. The S/N ratio being maximum at a stand-off distance of 7.5 mm and decreases with decrease in stand-off distance value. The lower value of stand-off distance decreases the impacting velocity of the striking particles due to the presence of adiabatic shear instability and shock wave phenomena [14, 15, 28]. As surface roughness is the-lower-the-better type quality characteristic, lower values of surface roughness were sought. From Fig. 3a–e, it can be observed that the optimum level of process parameters, (A2), (B3), (C3), (D3), and (E3) provide minimum value of surface roughness. However, in the case of all the parameters A, B, C, D, and E, the lowest values of mean response correspond to the highest values of S/N ratio.
Levels
DOF
P (%)
The SEM microstructures shown in Fig. 4a–r represent the set of 18 experiments conducted as per the design in Table 4, respectively. The SEM microstructures were taken at IIC, IIT, Roorkee, India with help of a scanning electron microscope (JSM-5800, JEOL, New York, Resolution— 5 nm). The micrographs were taken at following parameter values: HV 20 kV, magnification ×1,000, WD 9.7 mm, Detector ETD, HFW 0.30 mm. The impact velocity and plastic deformation of the striking particles are the main factors influencing the quality of the coatings in regard to low-surface roughness values. The effect of these factors can be noticed in the SEM micrographs shown in Fig. 4a–r. In order to study the significance of the process parameters towards the surface roughness, ANOVA was performed. The pooled versions of ANOVA of the raw data and the S/N ratio for surface roughness are given in Tables 7 and 8. From these tables, it can be concluded that parameters A, B, C, D, and E significantly affect both the mean and variation in surface roughness values. The percentage contribution of air pressure was highest (35.83%), followed by stand-off distance (22.35%), substrate material (17.58%), air temperature (12.98%), and feed arrangement (11.23%). 4.1 Estimation of optimum performance characteristics The estimated mean of the surface roughness was determined [26, 27] as follows:
Surface roughness 8.192 8.101 7.629 7.905 8.258 8.790
Where A 2—second level of feed arrangement, B 3—third level of substrate material, C 3—third level of air pressure, D 3—third level of air temperature, E 3—third level of stand-off distance, and T—average of surface roughness value for all set of experiments
4.1.1 Optimal values of response characteristics (predicted means) The average values of all the quality characteristics at the optimum levels of significant parameters are recorded in Table 9. The optimal value of the predicted mean (μ) of response characteristic can be obtained from the following equation [26, 27]: m ¼ A2 þ B3 þ C3 þ D3 þ E3 4 T ¼ 4:921
ð2Þ
622
Int J Adv Manuf Technol (2012) 60:611–623
The 95% confidence interval of confirmation experiments (CICE) and population (CIPOP) can be computed [27] by using the following equation:
CICE
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi 1 1 ¼ Fa ð1;fe ÞVe þ neff R
CIPOP
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi 1 ¼ Fa ð1;fe ÞVe neff
ð3Þ
5 Conclusions Coatings were successfully deposited on different substrates by using low-pressure cold spray process using air as the carrier gas and copper as the powder and a low value of surface roughness was achieved using cold spray process as compared to conventional thermal spray processes. The following conclusions can be drawn from the study: &
ð4Þ
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 variance; neff = (N/(1+DOF)), N=total number of trials; and DOF=total degrees of freedom associated in the estimate of mean response. The following values have been obtained by the ANOVA. N=54, fe =44, ve =0.000387079, neff =5.4, R=3, F0.05(1, 44)=4.064 From Eq. 3, CICE =±0.02856 From Eq. 4, CIPOP =±0.01706 The predicted optimal range for surface roughness is given by CICE : 4:89243 < m < 4:94956 CIPOP : 4:90394 < m < 4:93806 The optimal values of surface roughness were predicted at the selected levels of the significant parameters. Powder feeding arrangement (A, second level)=argon type (Sulzer Metco, USA). Substrate material (B, third level)=nickel alloy (ASTM B 435) Air pressure (C, third level)=120 psi Air temperature (D, third level)=400°C Stand-off distance (E, third level)=7.5 mm 4.2 Conformation experiments Three conformation experiments were conducted at the optimum setting of the process parameters. The powder feeding arrangement was set at second level (A2), substrate material at third level (B3), air pressure at third level (C3), air temperature at third level (D3), and stand-off distance at third level (E3). The average surface roughness of the lowpressure cold spray process was measured as 4.92 μm, which was within the confidence interval of the predicted optima of surface roughness for the coatings.
&
&
&
& & &
The carrier gas stagnation pressure helps in imparting the high velocity to impacting particles and thus aids in the formation of the coating and a better surface roughness value of the deposited coating. Stand-off distance inversely affects the surface roughness. The resultant impact velocity of the striking particles is reduced due to the presence of adiabatic shear bands at low stand-off distance. The surface roughness obtained depends on the type of substrate material used—the minimum for ASTM B 435 (Ni) alloy followed by ASTM B36 (Brass) alloy and ASTM B 221 (Al) alloy, respectively. The carrier gas temperature helps in the plastic deformation of the particles. So, the plastic deformation and high impact velocity of the powder particles are the main factors contributing towards a better quality of the coating. It is possible to achieve a comparatively lower value of surface roughness by using argon powder feed arrangement as compared to gravity powder feed arrangement. The predicted optimal range for surface roughness is CIPOP: 4.90394<μ<4.93806. The 95% confidence interval of the predicted mean for surface roughness is CICE: 4.89243<μ<4.94956.
References 1. Maev RGr, Leshchynsky V (2008) Introduction to low pressure gas dynamic spray. WILEY-VCH, Weinheim 2. Champagne VK (2007) The cold spray materials deposition process: fundamentals and applications. Wood head Publishing Limited, Cambridge 3. Materials Deposition, Cold Spray- Department of Defense Manufacturing Process Standard; MIL-STD-3021, 4 August, 2008. 4. Ghelichi R, Guagliano M (2009) Coating by cold spray process: a state of the art. Frattura ed Integrità Strutturale 8:30–44. doi:10.3221/IGF-ESIS.08.03 5. Villafuerte J (2010) Current and future applications of Cold Spray Technology. http://metalfinishing.com: 37–39 Accessed 18 December 2010 6. Lee J, Shin S, Kim HJ, Lee C (2007) Effects of gas temperature on critical velocity and deposition characteristics in kinetic spraying. Appl Surf Sci 253:3512–3520 7. Li H, Shin H, Lee S, Ko K (2008) Effect of gas pressure on Al coating by cold gas dynamic spray. Mater Lett 62:1579–1581 8. Ning XJ, Jang JH, Kim HJ (2007) Effects of powder properties on in-flight particle velocity and deposition process during low pressure cold spray process. Appl Surf Sci 253:7449–7455
Int J Adv Manuf Technol (2012) 60:611–623 9. Li WY, Li CJ, Liao H (2010) Significant influence of particle surface oxidation on deposition efficiency, interface microstructure and adhesive strength of cold sprayed coatings. Appl Surf Sci 256:4953–4958 10. Shin S, Xiong Y, Ji Y, Kim HJ, Lee C (2008) The influence of process parameters on deposition characteristics of soft/hard composite coating in kinetic spray process. Appl Surf Sci 254:2269–2275 11. Steenkiste et al (1999) Kinetic spray coatings. Surf Coat Technol 111:62–71 12. Li G, Wang XF, Li WY (2007) Effect of different incidence angles on bonding performance in cold spraying. Trans Nonferrous Met Soc China 17:116–121 13. Li WY, Yin S, Wang XF (2010) Numerical investigations of the effect of oblique impact on particle deformation in cold spraying by SPH method. Appl Surf Sci 256:3725–3734 14. Li WY, Zhang C, Guo XP, Zhang G, Liao HL, Li CJ, Coddet C (2008) Effect of stand-off distance on coating deposition characteristics in cold spraying. Mater Des 29:297–304 15. Pattison J, Celotto S, Khan A, O’Neill W (2008) Stand-off distance and bow shock phenomena in the cold spray process. Surf Coat Technol 202:1443–1454 16. Grujicic M, Zhao CL, Tong C, DeRosset WS, Helfritch D (2004) Analysis of the impact velocity of powder particles in the cold-gas dynamic-spray process. Mater Sci Eng A 368:222–230 17. Wu J, Fang H, Yoon S, Kim HJ, Lee Ch (2005) Measurement of particle velocity and characterization of deposition in aluminum alloy kinetic spraying process. Appl Surf Sci 252:1368–1377
623 18. Helfritch D and Champagne V (2008) A model study of powder particle size effects in cold spray deposition. 26th Army Science Conference Proceedings 19. Schmidt T, Gartner F, Assadi H, Kreye H (2006) Development of a generalized parameter window for cold spray deposition. Acta Mater 54:729–742 20. Li WY, Liao H, Li CJ, Li G, Coddet C, Wang X (2006) On high velocity impact of micro sized metallic particles in cold spraying. Appl Surf Sci 253:2852–2862 21. Klinkov SV, Kosarev VK, Rein M (2005) Cold spray deposition: significance of particle impact phenomenon. Aerosp Sci Technol 9:582–591 22. Wu J, Fang H, Yoon S, Kim H, Lee C (2006) The rebound phenomenon in kinetic spraying deposition. Scr Mater 54:665– 669 23. http://hydro.com/northamerica/optigrain. Accessed 17 September 2009 24. Freslon (2003) Handbook of thermal spray technology-coatings equipment and theory. Thermal Spray Society and ASM International, USA 25. Steenkiste, Smith (2004) Evaluation of coatings produced via kinetic and cold spray process. J Therm Spray Technol 13:274–282 26. Ross PJ (1988) Taguchi techniques for quality Engineering. McGraw-Hill Book Company, New York 27. Roy RK (1990) A primer on taguchi method. Van Nostrand Reinhold, New York 28. Assadi H, Gartner F, Stoltenhoff T, Kreye H (2003) Bonding mechanism in cold gas spraying. Acta Materialia 51:4379–4394