Online monitoring of thermo-cycles and its correlation with microstructure in laser cladding of nickel based super alloy

Optics and Lasers in Engineering 88 (2017) 139–152

Contents lists available at ScienceDirect

Optics and Lasers in Engineering journal homepage: www.elsevier.com/locate/optlaseng

Online monitoring of thermo-cycles and its correlation with microstructure in laser cladding of nickel based super alloy Gopinath Muvvala, Debapriya Patra Karmakar, Ashish Kumar Nath n Department of Mechanical Engineering, Indian Institute of Technology, Kharagpur 721302 India

art ic l e i nf o

a b s t r a c t

Article history: Received 19 April 2016 Received in revised form 20 July 2016 Accepted 14 August 2016

Laser cladding, basically a weld deposition technique, is finding applications in many areas including surface coatings, refurbishment of worn out components and generation of functionally graded components owing to its various advantages over conventional methods like TIG, PTA etc. One of the essential requirements to adopt this technique in industrial manufacturing is to fulfil the increasing demand on product quality which could be controlled through online process monitoring and correlating the signals with the mechanical and metallurgical properties. Rapid thermo-cycle i.e. the fast heating and cooling rates involved in this process affect above properties of the deposited layer to a great extent. Therefore, the current study aims to monitor the thermo-cycles online, understand its variation with process parameters and its effect on different quality aspects of the clad layer, like microstructure, elemental segregations and mechanical properties. The effect of process parameters on clad track geometry is also studied which helps in their judicious selection to deposit a predefined thickness of coating. In this study Inconel 718, a nickel based super alloy is used as a clad material and AISI 304 austenitic steel as a substrate material. The thermo-cycles during the cladding process were recorded using a single spot monochromatic pyrometer. The heating and cooling rates were estimated from the recorded thermocycles and its effects on microstructures were characterised using SEM and XRD analyses. Slow thermocycles resulted in severe elemental segregations favouring Laves phase formation and increased γ matrix size which is found to be detrimental to the mechanical properties. Slow cooling also resulted in termination of epitaxial growth, forming equiaxed grains near the surface, which is not preferred for single crystal growth. Heat treatment is carried out and the effect of slow cooling and the increased γ matrix size on dissolution of segregated elements in metal matrix is studied. & 2016 Elsevier Ltd. All rights reserved.

Keywords: Laser cladding Online monitoring Thermo-cycles IR pyrometer

1. Introduction Laser cladding is a deposition technique in which metal powder or wire is fused to a base material using high energy laser beam as a heat source. The process offers the potential to deposit new materials with required mechanical and physical properties onto different low value substrates with minimum dilution, heat effect zone, distortion and good bond strength. Fast cooling rates by selfquenching attainable in laser cladding process give rise to extremely refined microstructure, leading to improved mechanical properties. Laser cladding finds its applications in aerospace, nuclear, rail, oil rig, mining industries etc. for repairing as well as direct generation of functional components [1–3]. Based on the application and method of powder feeding, this processes is called by several names like selective laser sintering n

Corresponding author. E-mail address: [email protected] (A.K. Nath).

http://dx.doi.org/10.1016/j.optlaseng.2016.08.005 0143-8166/& 2016 Elsevier Ltd. All rights reserved.

(SLS) or direct metal laser sintering (DMLS), laser engineered net shaping (LENS), laser metal forming (LMF), laser additive manufacturing (LAM) etc [2,4]. SLS or DMLS uses pre-placed powder technique wherein a thin layer of powder is placed on a part building table, scanned by a laser beam in selected areas and the table is displaced down and a new layer is placed. The process is repeated till the final component is built up [5]. Furthermore, preplaced powder technique is largely used in the field of laser surface coatings to study the metallurgical properties of different materials upon irradiating with laser beam, optimum mixing ratio of metal powders and ceramic particles to obtain a defect-free coating with enhanced surface properties like wear resistance, corrosion resistance etc [6,7]. Control of final characteristics of deposited clad track like geometry which includes clad height, width and contact angle, percentage of dilution, surface roughness, microstructure and mechanical properties requires a good understanding of the relationship among these characteristics, processes parameters and

140

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

thermal history of molten pool during cladding. Furthermore, application of this process in industry requires proper monitoring system to diagnose the process online. Laser cladding involves more than 19 process parameters [8] which include laser beam power, scan speed, beam spot diameter, powder layer thickness, powder mass flow rate, carrier and shielding gas pressure, powder particle size and shape, standoff distance, thermal conductivity of powder and substrate, hatch spacing, percentage of overlapping, interlayer ideal time, operating environment, geometric dimensions of the substrate etc. To a greater extent these process parameters are machine or system dependent which include the nozzle design (number of outlets, powder exit angle etc), type of laser system integrated, it is operating wavelength and beam intensity distribution etc. Thus, the monitoring system or the parameter should be such, which is independent of the system, can capture and provide all the necessary information to assess the metallurgical and mechanical properties of the deposited layer or component. Thermal history of the molten pool is one such parameter which is an outcome of the combination of many of the above critical process parameters and affects the mechanical and metallurgical properties of the deposited component. Therefore, monitoring the molten pool temperature and thermal history can facilitate in real-time non-destructive evaluation of the process. 1.1. Background Several researchers developed analytical and numerical models to estimate the effect of process parameters on the surface temperature and to predict the extent of heat effected zone. Fang et al. [9] developed a three dimensional finite element model to estimate the temperature distribution and cooling rate surrounding the molten pool, and correlated it with the heat effect zone in laser cladding process. Lie et al. [10] developed a three dimensional model to simulate high power laser cladding of TiC/NiCrBSiC composite coating on Ti6Al4V alloys and could predict the temperature distribution, shape and size of the molten pool and the heat affected zone using the model. Tahmasbi et al. [11] also carried out FEM modelling to study the effect of layer number on the temperature distribution in a vertical build up of IN-738 layers. All the above analytical and numerical models are focused on predicting the clad geometry, heat effected zone and surface temperature with respect to the process parameters. Summing up the published results, the analytical models could help in understanding the process and the effect of process parameters on temperature distribution. However, the temperature history predicted from these models fail to capture some of the important events like shift in phase transformation point upon addition of foreign materials, melt pool and solidification shelf lifetime which controls the microstructure and mechanical properties of the clad layer and vaporisation from the clad surface. This is because, all the above studies consider the effect of latent heat of fusion and Marangoni flow by modifying the specific heat capacity and thermal conductivity, and neglect any chance of vaporisation because of which surface temperature predicted in many cases is near 4000 °C which in real case is not possible. Thus, online monitoring system which can record all the above mentioned events is of significant importance. Since many coupled physical events takes place in a very short period of time during laser cladding, involving a wide temperature range, small processing zone, sharp temperature gradients i.e. fast heating and cooling rates (104–106 K/s), it requires a non-contact type temperature measuring system, not affected by laser radiation and having short signal acquisition time [12]. Feasibility studies on the application of pyrometer in online monitoring and control in laser machining have been carried out by several researchers [12–16]. Ignatiev et al. [13,14] developed a two

wavelength pyrometer, demonstrated that the measurement within a spectral band free from spectral lines of irradiated material provides accurate monitoring of brightness temperature. In case of 1-spot pyrometers the temperature signal may fluctuate from the mean value in the range of several hundred degrees which may be due to the spikes of free-running laser pulse, plume dynamics, and melt hydrodynamics, etc. Doubenskaia et al. [15,16] studied the variation in brightness temperature profile with laser beam power, pulse shape and powder feed rate. Bi et al. [17] applied different measuring systems like photodiode, pyrometer and CCD camera to detect the IR radiation from processing zone and studied the effect of process parameters on temperature signal. Temperature signal detected with photodiode arranged coaxially with the cladding nozzle increased with increase of powder feed rate up to certain extent and became constant. However, pyrometer signals showed a decreasing trend with increase of powder feed rate, similar to the results of Doubenskaia et al. [16]. Bi et al. also developed a cladding head integrated with different sensors like contact thermometers, photodiodes and CCD cameras to monitor the process and simultaneously the condition of various optical components incorporated in the head [18]. In multilayer thin wall deposition, the heat conduction mechanism transforms from 3D to 2D as the number of layers increases, resulting in heat accumulation and inhomogeneous wall thickness. Further, absence of material at the edges to conduct the heat results in edge build up as well as shrinkage of the wall. The authors further extended their work in which laser beam power was varied during the process with change in IR-temperature signal from a pre-set value using a PID controller based closed loop control system [19]. Pavlov et al. [20] studied the effect of scan speed and composition of MMC's on brightness temperature signal. It was observed that with the increase of TiC content in metal matrix, the brightness temperature increased which was attributed to the shift of critical phase transitions point, optical and thermal properties. Also, the TiC content resulted in decreasing the height of multi layer deposition. Presence of MMC's like TiC on the surface of clad acts as a thermal barrier and also has high melting point. This results in increase of surface temperature, decrease in temperature gradients and melt pool size and thus powder catchment efficiency [21]. Emamian et al. [22] estimated the variation in surface temperatures using a 3D model during deposition of multiple tracks at a particular distance from one other. They observed that compared to the first track the later ones were less prone to micro-cracks due to the relatively lower thermal stresses as a result of slow cooling rates owing to substrate heating caused by previously deposited tracks. The temperature variation along multiple tracks could also influence the dimension and shape of clad layers as they depend on the surface tension between the solid-liquid-gas which is dictated by the melt pool lifetime and the surface temperature of substrate as well as the melt pool. Yu et al. [23] studied the effect of process parameters viz. laser beam power, scan speed and powder mass flow rate on melt pool lifetime and its effect on spreading of molten pool on substrate surface. They used a two wavelength pyrometer to detect the temperature signal and found that in case of multilayer deposition, the melt pool lifetime increases with increase of layer number, decreasing the dimensional accuracy. All the above studies are focused on the application of various detection sensors and techniques, their industrial feasibility and effect of process parameters on the clad geometry and detected signal in laser cladding process. However, the ultimate mechanical properties of components fabricated by laser cladding process depend on the thermal-history-dependent microstructure. Therefore, it is vital to understand the changes in thermo-cycles with various input parameters and their effect on the resulting microstructures and mechanical properties.

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

(a)

(b) 1285

1600

1185

Zone 1

1200

Temperature (°C)

Temperature (°C)

1400

Zone 2

1000 800

1085 985 885

600 400

141

0

250

500 750 Time (ms)

1000

785 0

100 200 Time (ms)

300

Fig. 1. Typical thermo-cycle (a) without and (b) with notch filter (P¼ 800 W and V ¼ 4000 mm/min).

The current study focuses on the online monitoring of thermocycles and the effect of process parameters on it, followed by estimation of cooling rates in laser deposited nickel based super alloy, Inconel 718 clad layers. This is widely used in elevated temperature applications like gas turbines, liquid fuelled rockets etc. It exhibits high strain-age cracking resistance and maintains high strength up to 700 °C. Inconel 718 contains aluminium, titanium and niobium which precipitate as hard phases during appropriate heat treatment, increasing the strength and hardness of the alloy. It is basically strengthened by the formation of γ″ intermetallic precipitate (Ni3Nb). However, during laser cladding process, Nb is known to segregate in coarse dendrite as eutectic phase at the termination of solidification facilitating the formation of Laves phase. This retards the precipitation of the main strengthening phase during heat treatment, which deteriorates the strengthening effect of Nb addition in Inconel 718 alloy. Several studies were carried out on the microstructural aspects wherein details of different phases that occur in Inconel 718 were discussed in different processes like laser metal deposition, remelting, laser welding, electron beam welding etc [24–30]. However, irrespective of the process, elemental segregations depend mainly on the rate at which liquid metal is solidifying and reaching to the room temperature. Zhang et al. [24] observed in laser-remelted Inconel 718 coatings that the dendritic spacing of columnar dendrite and the equiaxed grain size decreased with the increase of laser scan speed. It is well known that the overall efficiency of gas turbines improves with increasing operating temperature. Single crystal turbine blades of Ni- based super-alloys are often used for high temperature operation. The life of these turbine blades is limited by tip erosion and they need to be repaired. Gäumann [31] established a process called epitaxial laser metal forming (E-LMF) based on solidification theory for refurbishing the single crystal blades. They related the expected solidification microstructures and growth morphologies to the processing conditions. The formation of single crystal or the epitaxial growth in laser cladding process is interrupted by the equiaxed grain growth which is reported to be a function of cooling rate. Therefore, the effect of cooling rate on microstructural aspects viz. elemental segregation, size of γ matrix and its effect on dissolution of segregated elements during standard solution heat treatment [27], termination of epitaxial growth and formation of equiaxed grains also have been studied.

2. Experimental setup Laser cladding is carried out using a Yb-Fibre laser (IPG photonics, Model no. YLR 2000) operating at 1.07 mm wavelength with maximum output power of 2 kW. This can be operated in CW as well as in modulated mode at 50–1000 Hz frequency range with 5–100% duty cycle with rectangular shaped temporal power profile. The laser has multimode beam intensity profile with high intensity at the centre and two annular rings of relatively low intensity [32]. The laser beam delivery system is mounted on a 5-axis CNC machine capable to move at speeds up to 20 m/min effectively. Single spot monochromatic pyrometer (Micro Epsilon, model: CTLM-2HCF3-C3H) operating at 1.6 mm, and having 700 mm vision zone and 1 ms acquisition time with 385 °C to 1600 °C working temperature range has been used to record thermo-cycles during the laser cladding process. The pyrometer can be aimed at the required point with the help of a pair of guide laser beams provided in it. In the current study, all the thermo-cycles were recorded keeping the pyrometer static. In case of CW mode laser cladding the laser beam was scanned keeping the substrate static, while in case of pulse mode the substrate was moved keeping laser beam static. Thus, in CW mode fixed-point temperature monitoring technique is used whereas in pulse mode continuous monitoring technique is used. Though the pyrometer is specified to operate at 1.6 mm wavelength, it was observed that the reflected laser radiation from the processing zone was also interfering with the IR signal from the molten pool, due to which pyrometer produced thermo-cycles as shown in Fig. 1(a), when the laser beam was scanned on a bare SS-304 substrate. The IR signal in zone 1 and zone 2 is generated by the laser radiation reflected from the processing zone at the trailing and leading ends of the area where pyrometer is focused. For laser scan speed of 4000 mm/min and spot diameter of 3 mm the interaction time should be approximately 45 ms. However, it is observed in Fig. 1(a) that the detected signal duration is much longer than the expected interaction time. Moreover, with no sign of melting in the processing zone, the peak temperature recorded was above the melting point of 304 stainless steel (
1400 °C) substrate material. In order to eliminate the effect of reflected laser radiation, a notch filter of 1064 nm 725 nm spectral range with optical density of 3 was used to block the 1070 nm laser radiation, and the resulting signal recorded is as shown in Fig. 1(b). The notch filter blocked a part of

142

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

Fig. 2. (a) Morphology and (b) X-ray diffraction pattern of Inconel 718 powder.

Table 1 Chemical composition (wt%) of cladding and substrate materials. Element (wt%)

Cr

Fe

Nb

Mo

Al

Ti

Mn

Si

B

Ni

C

P

S

IN 718 AISI 304

19.02 18.97

18.1 Bal.

4.92 –

3.19 0.224

0.54 –

0.97 –

0.04 1.731

0.20 0.753

0.004 –

Bal. 8.554

– 0.067

– 0.045

0.031

Table 2 Laser cladding process parameters. CW mode Laser beam power (W)

Scan speed (mm/ min)

Thickness of preplaced powder coating (mm)

Beam spot diameter (mm)

400–2000

200–2000

1, 2

3

Pulse (Modulated laser power) mode Thickness of Scan Laser preplaced speed beam powder (mm/ power coating min) (W) (mm) 1250 1800 1

Beam spot diameter (mm)

Laser modulating Frequency (Hz)

Duty cycle (%)

3

50

50, 80

V

Plano-convex lens Window Shielding gas

Shrouding gas

Pyrometer

Clad layer

Substrate

Pre-placed powder Processing Zone

Fig. 3. A schematic of experimental set-up.

the 1.6 mm radiation also; therefore the pyrometer was calibrated with and without the notch filter at the melting temperature of different metals, e.g. steel, copper and aluminium. With the notch filter the temperature measurement range is 785–3260 °C. AISI 304 austenitic steel plates of 50 mm  25 mm  8 mm

dimensions were used as substrates and gas atomised Inconel 718 (MetcoClad 718) powder as the clad material. The morphology and the X-ray diffraction pattern of powder particles are shown in Fig. 2(a) and (b) respectively. Powder particles were in the size range of 43–100 mm. Chemical compositions of the clad material and the substrate are presented in Table 1. The ground surface of as received AISI 304 substrates were polished with emery paper (220 mesh) and ultrasonically cleaned in acetone to remove contaminants. Inconel 718 powder was mixed with aqueous solution of 2% polyvinyl alcohol (Alfa Aesar, 87–89% hydrolysed) which was used as a binder. Mixing was done mechanically by alternate cycles of mechanical stirring (Spinot Model MC-02, Tarsons) [7,33]. After proper mixing a predefined thickness of powder-PVA blend is coated on the substrate surface using a coating machine (Model no: K101 control coater, RK Print Coat Instruments Ltd., UK,). The coated samples were then baked in furnace in Argon atmosphere at 100 °C for a period of 15 min to remove moisture from the coating and for proper bonding, and then cooled to room temperature in the furnace. Laser cladding was carried out on these specimens by varying different process parameters as shown in Table 2, and thermo-cycles were recorded during the process. Argon gas was used as a shielding as well as shrouding gas to protect the optical components of the laser processing head from fumes and the molten pool from oxidation respectively. Shrouding gas was directed towards the processing zone at 5 l/min flow rate through a 6 mm diameter copper pipe inclined at 45° and 5 cm away from the molten pool. The laser beam spot diameter was maintained 3 mm throughout the experiment. Three replicates were taken for each experiment. Fig. 3 shows the schematic of the experimental setup. The pyrometer was sighted on the specimen at 45° to the axis of laser beam. The laser deposited specimens were cross sectioned, mirror polished and etched with a solution of 50 ml HCl, 25 ml HNO3, 0.2 g CuCl2 and 100 ml deionised water. Clad geometry and microstructures were characterised by optical and scanning electron microscopy. EDS analysis was utilised to determine the chemical composition. Phase identification was performed using X-ray diffractometer with Cu Kα radiation at 40 kV and 40 mA using a continuous scan mode. A quick scan at 1.4°/min was performed over a wide range of 30–100°. Vickers

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

143

Hc

θ

Hf

(a)

W

(b)

(c)

Fig. 4. (a) A schematic diagram of clad track geometry, (b) and (c) clad track defects.

1 mm

1 mm thick preplaced

2000mm/min

1 mm

1800mm/min

0.5 mm

0.5 mm

600mm/min

400mm/min

1 mm

1600mm/min

1 mm

1800W

800mm/min

1400mm/min

0.5 mm

0.5 mm

1400W

2 mm thick preplaced

1000mm/min

Fig. 5. Typical cross-section of clad tracks deposited with different processing conditions.

hardness and fracture test were performed in order to study the effect of elemental segregation on the mechanical properties. Heat treatment was carried out to decompose the segregated alloying elements into the metal matrix and the results were analysed using SEM.

3. Results and discussion 3.1. Effect of process parameters on clad track geometry The primary objective of the laser cladding process is to obtain porosity- and crack- free clad layers with good metallurgical bonding and low dilution with substrate. Clad track geometry is an important aspect which determines the dimensional accuracy and quality of the deposited component. Clad track geometry (Fig. 4a) shows the clad height (Hc), width (W), contact or clad angle (θ) and melt depth (Hf), and the dilution is defined as Hf/(Hc þHf) [34]. The forming characteristics are affected mainly by the laser process parameters. Improper selection of process parameters like laser beam power density or scan speed can lead to deepening of melt depth at the centre of clad track or balling phenomenon as shown in Fig. 4(b) and (c) respectively. An understanding of relation between process parameters and clad track geometry helps in defining the hatching distance or percentage of overlap, number of layers required to cover a given area and the processing time, and also in the selection of process parameters to obtain a predefined coating thickness. Fig. 5 shows the typical cross-section of the laser cladded tracks deposited with different set of process parameters. In most cases the dilution is low which differentiates it from the laser alloying process [35]. Fig. 6 shows the effect of CW laser beam power (P)

and scan speed (V) on different aspects of clad track geometry for 1 mm thick pre-placed powder layer. It can be observed that the clad height remained unchanged with changing of the scan speed from 1000 mm/min to 2000 mm/min at higher CW laser beam power of 2000 W and 1800 W, Fig. 5(a). However, the same was observed to increase with increasing scan speed at relatively lower power of 1600 W and 1400 W. A closer observation of Fig. 6 (a) revealed that the height remained unchanged as long as the line energy which is defined as laser energy supplied per unit length, i.e. P/V (J/mm) was greater than 1 J/mm, and at line energy below this the clad height tended to increase. Similar is the case with cladding angle as shown in Fig. 6(b). With line energy greater than 1 J/mm, enough amount of laser energy gets coupled with the specimens, rising its temperature which causes fall in surface tension and wide spreading of molten pool [23] resulting in little change in clad height and angle. On the other hand, clad width and dilution showed increasing trend with the decrease of scan speed as shown in Fig. 6(c) and (d) respectively. However, it may be noted that the clad width is less than the laser beam spot diameter at lower line energy. This could be because of the non-uniform intensity distribution of the multimode laser beam having relatively high intensity central lobe surrounded by low intensity annular rings [32]. With the decrease of scan speed at a constant CW laser beam power the line energy and the lateral heat conduction in molten pool increases due to increased laser interaction time, leading to an increase in clad width beyond laser beam diameter as shown in Fig. 6(c). 3.2. Detection of thermo-cycles with IR pyrometer Fig. 7 shows a typical thermo-cycle recorded using the IR pyrometer during laser cladding of 2 mm thick preplaced powder

144

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

2000W 0.8

1800W

1600W

1400W

2000W

1800W

1600W

1400W

170 Clad angle (°)

Clad height (mm)

0.6 0.4 0.2

150 130 110

0

800

1200 1600 2000 Scan speed (mm/min)

(a)

1800W

1600W

800

(b)

1400W

2000W

4

80

3

60

% of Dilution

Clad width (mm)

2000W

90

2 1

0

1200 1600 2000 Scan speed (mm/min)

1800W

1600W

1400W

40 20 0

800

1200 1600 2000 Scan speed (mm/min)

(c)

800

(d)

1200 1600 2000 Scan speed (mm/min)

Fig. 6. Effect of CW laser beam power and scan speed on clad (a) Height, (b) Angle, (c) Width and (d) Percentage of dilution (1 mm thick preplaced powder layer).

part of the paper. The duration of AC represents the molten pool lifetime. The duration and the cooling rate between CD would dictate the formation of precipitates, grains size and different phases like in laser surface modification. Thus, the online monitoring of thermo-cycles can provide a real time insight into the microstructural and resulting mechanical and metallurgical properties of the deposited clad layer.

1785 L – Liquid S – Solid T – Liquidus Temperature T – Solidus Temperature

Temperature (°C)

A

1585 T

B

1385

C T

1185 985 785

L

L+S

S

D

0

1000

2000

3000

4000

5000

6000

7000

Time (ms) Fig. 7. Typical thermo-cycle in cladding of 2 mm pre-placed powder layer, where 0A is the heating cycle, AB is the cooling of liquid phase, BC is the solidification shelf, CD is the cooling of solid phase, and AC is the melt pool life time (P¼ 1200 W, V ¼200 mm/min).

layer of Inconel 718 on AISI 304 austenitic steel substrate at a CW laser beam power and scan speed of 1200 W and 200 mm/min respectively. It consists of a heating cycle (0A) and a cooling cycle (AD). One may distinguish several typical stages during the cooling cycle in Fig. 7. AB represents the time interval for which the molten pool is in complete liquid state. From ‘B’, where the molten metal reaches the liquidus temperature, solidification initiates and ends at ‘C’ when it reaches to solidus temperature. Between these two points, liquid and solid phase exists in equilibrium. It can be clearly observed from Fig. 7 that the cooling rate (∂T/∂t) reduces drastically to near zero between BC, due to the release of latent heat of solidification. BC is generally termed as freezing zone or the solidification shelf and its duration would dictate the extent of segregations in super alloys which will be discussed in the later

3.2.1. Effect of process parameters on thermo-cycles Experiments were carried out to analyse the effect of various process parameters viz. CW laser beam power, scan speed, preplaced powder layer thickness, and duty cycle on the thermo-cycle. Figs. 8 and 9 shows the effect of scan speed and CW laser beam power on the thermo-cycles where they are varied from 2000 to 200 mm/min and 1200–400 W respectively for 2 mm thick preplaced powder layers. It can be clearly observed from Fig. 8 that at higher scan speeds, the temperature increased sharply above the melting point followed by fast cooling. At slow scan speeds, there exists a quasi-steady state at the peak temperature. Further, the peak temperature was found to increase with decrease of scan speed from 2000 to 1400 mm/min and remained almost constant up to 800 mm/min, followed by a decreasing trend up to 200 mm/ min as shown in Fig. 10. This indicates that after 800 mm/min vaporisation from the molten pool surface becomes significant because of which temperature falls [36]. Thus, the monitoring of thermo-cycle can help in avoiding excess heating and vaporisation by minimising the duration of quasi-steady state at the peak temperature. Unlike scan speed, CW laser beam power did not show any significant effect on the heating rate, but the peak temperature was found to increase marginally with increasing CW laser beam power. An increase of CW laser beam power from

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

1785

2000 mm/min 1800 mm/min 1600 mm/min 1400 mm/min 1200 mm/min 1000 mm/min 800 mm/min 600 mm/min 400 mm/min

Temperature (°C)

1585

Melting point of Inconel 718 (1336 °C)

1385

145

1185 985 785

0

500

1000

1500

2000

2500

3000

Time (ms) Fig. 8. Variation in thermo-cycles with change in laser scan speed (P ¼1200 W, pre-placed powder layer thickness¼ 2 mm).

1785

1200 W 1000 W 800 W 600 W 400 W

Temperature (°C)

1585 1385

Melting point of Inconel 718 (1336 °C)

1185 985 785 0

400

800

1200

1600

2000

Time (ms) Fig. 9. Variation in thermo-cycles with change in CW laser beam power (V ¼600 mm/min, pre-placed powder layer thickness¼ 2 mm).

Temperature (°C)

1700 1650 1600 1550 1500 2000

1600 1200 800 400 Scan speed (mm/min)

Fig. 10. Effect of scan speed on surface temperature (P ¼1200 W, pre-placed powder layer thickness¼ 2 mm).

400 W to 1200 W showed an increase of 110 °C only in the surface temperature. Heating and cooling rates are found to increase linearly with the increase of scan speed as shown in Fig. 11(a) and (b). However, the heating rate is nearly 10 times of the cooling rate in the present experimental range of the scan speed. At lower scan speeds more amount of energy gets coupled with the specimens and the interaction time also increases. At higher interaction time more heat from interaction zone conducts into the substrate increasing its temperature, decreasing the thermal gradient and thereby decreasing the cooling rate. Similar trend was observed with the increase of CW laser beam power. However, compared with scan speed the change in cooling rate with CW laser beam

power is relatively small as the interaction time is constant, Fig. 11 (c). Thus, it can be concluded that the scan speed plays a more important role in controlling the cooling rate than the CW laser beam power. The melt pool lifetime was also found to increase with the decrease of scan speed as shown in Fig. 11(d). A change in scan speed from 2000 mm/min to 200 mm/min increased the lifetime of melt pool from 0.11 s to 2.19 s. Apart from CW laser beam power and scan speed, preplaced powder layer thickness was also found to affect the cooling rate. It decreased when the preplaced powder layer thickness was increased from 1 mm to 2 mm as shown in Fig. 12 at different scan speeds in 600– 1200 mm/min range. This could be due to relatively low thermal conductivity of preplaced Inconel 718 powder layer coating than the steel substrate, because of which heat cannot get conducted into the substrate readily in thicker layer. Laser cladding in pulse mode can have certain advantages over continuous mode laser cladding because of the faster heating and cooling rates, relatively less rise in average temperature, lower heat input and thermal distortion of substrate, reduced heat affected zone, and proper mixing of powder blend etc. The extent of these advantages depends mainly on the pulse on-time and duty cycle. These two parameters will dictate the thermo-cycles of the molten pool. Unlike CW mode laser cladding where the molten pool undergoes one heating and cooling cycle, in pulse mode this would undergo a series of thermo-cycles within a given interaction time. Therefore it is important to monitor the thermal history which will dictate the microstructure and mechanical properties of the deposited layer. The temperature signals were monitored in pulse mode laser cladding keeping the pyrometer and laser beam stationary, and moving the substrate using another X-Y CNC table. Fig. 13(a) and (b) shows typical real time temperature signals recorded while irradiating specimens with 1 mm thick pre-placed powder layer at 50 Hz laser beam power modulation frequency

146

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

1.6

y = 4.359 x + 482 .6 R² = 0.993

8

Cooling rate ( 10 3 °C/s)

Heating rate (10 3 °C/s)

10

6 4 2

1.2 0.8 0.4

0

0

0

400

(a)

800 1200 1600 Scan speed (mm/min)

Y = 0.683X + 16.33 R2 = 0.993

2000

0

400

800 1200 1600 Scan speed (mm/min)

0

400

800 1200 1600 Scan speed (mm/min)

(b)

2000

2500 Melt pool life time (ms)

Cooling rate (10 3 °C/s)

0.66 0.63 0.6 0.57

0.54 200

2000 1500 1000 500 0

400

(c)

600 800 1000 Laser power (W)

1200

(d)

2000

Fig. 11. Effect of scan speed on (a) Heating rate and (b) Cooling rate for CW laser beam power ¼ 1200 W; (c) Cooling rate verses CW laser beam power at a scan speed of 600 mm/min and (d) Melt pool lifetime versus scan speed for CW laser beam power¼ 1200 W; (Pre-placed powder layer thickness¼ 2 mm).

1 mm

2 mm

Cooling rate ( °C/s)

1800

1500 1200 900 600 300

0

600

800 1000 1200 Scan Speed (mm/min)

Fig. 12. Effect of preplaced powder layer thickness on cooling rate (P¼ 1200 W).

with 50% and 80% duty cycle respectively. The laser beam peak power and the scan speed were kept constant at 1250 W and 1800 mm/min respectively. It can be observed in Fig. 13(c) that the laser irradiated zone undergoes through cyclic melting and resolidification phases during interaction with the laser beam modulated at 50% duty cycle, but at 80% duty cycle the temperature continuously oscillates above the melting point, maintaining the molten state during the entire interaction time as shown in Fig. 13(d). Small amplitude oscillation in surface temperature during laser irradiation could be because of the onset of vaporisation and the effect of recoil pressure of vapour on the molten pool which could increase the boiling temperature.

3.3. Effect of thermo-cycles on microstructures in Inconel 718 laser cladding 3.3.1. Comparison of microstructures in CW and pulsed mode laser cladding Fig. 14(a) and (b) shows the typical microstructures obtained in CW mode and pulsed mode laser cladding respectively at constant laser beam power of 1250 W and scan speed of 1800 mm/min in 1 mm thick pre-placed powder layer specimens. The repeated melting and solidification of clad layer with a moving pulsed laser source having multimode laser beam profile changes the microstructure to a greater extent compared to that in continuous mode laser cladding. Solidification starts from the molten pool and solid metal interface. Solidification front then moves towards the top surface of molten pool. In CW mode laser cladding, there exists a steep temperature gradient from the top surface of molten pool to the interface, and thus a columnar dendritic structure growing epitaxially from the interface is observed, Fig. 14(a). Further, the dendritic growth from the interface is found to be oriented towards the centre of clad bead surface where the temperature is expected to be the maximum. Similarly, in case of pulse mode cladding also, columnar dendritic structure is observed. However, there are stacks of columnar dendrites which are aligned in different directions with one another as shown in Fig. 14(b). To a greater extent they are found to follow the moving laser beam. This is because of the following reason: In pulse mode cladding as the laser beam moves, there is certain overlap between two consecutive pulses, depending upon the scan speed. Under the pulse mode laser cladding condition in Fig. 13(c) the laser irradiated zone undergoes through cyclic melting and re-solidification

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

147

Temperature (°C)

(a) 1685 1385 1085 785

0

300

600

900

1200

1400

Time (ms)

Temperature (°C)

(b) 1600 1400 1200 1000 0

200

400

600

800 Time (ms)

1000

1200

1400

1600

(d) 1650

(c)

Temperature (°C)

Temperature (°C)

1500 1400 1300

1550

1450

1200 1100

310

315

320 325 Time (ms)

330

1350 345

350

355 360 Time (ms)

365

Fig. 13. Real time surface temperature signals in pulse mode laser cladding at 50 Hz modulation frequency, P¼ 1250 W, V ¼1800 mm/min, 1 mm pre-placed powder layer thickness (a) 50% duty cycle (b) 80% duty cycle, (c) and (d) enlarged individual thermo-cycle of (a) and (b) respectively.

Fig. 14. Microstructure structure evolution in (a) CW mode and (b) Pulse mode laser cladding at 50 Hz with 50% Duty cycle (P¼ 1250 W, V ¼1800 mm/min, pre-placed powder layer thickness¼ 1 mm).

phases and the depth of the molten layer progressively reduces as the laser beam moves forward. This results in stacks of columnar dendrites rather than a continuous growth from the interface. 3.3.2. CW mode laser cladding Fig. 15 shows a typical microstructure revealing the transformation of epitaxial growth into equiaxed grains near the surface of laser cladding carried out at 1200 W CW laser beam power and 600 mm/min scan speed on 2 mm pre-placed powder layer sample. Three distinct zones can be identified clearly where zone 1 shows the pure epitaxial growth followed by transition zone

2 where the epitaxial growth terminates and equiaxed grains initiate and zone 3 consisting of fully defined equiaxed grains. As the formation of stray grains is detrimental to the mechanical properties of the formed component, secondary operations like grinding or laser remelting are generally carried out (Zhang et al. [24]) to remove the layer with stray grains. Now, the amount of material or the thickness up to which remelting has to be carried out depends upon the thickness of equiaxed grain layer and a prior knowledge of approximate layer thickness helps in determining this. As discussed in Section 1, laser cladding is a complex process where the cooling rate depends on several process parameters,

148

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

Granules

Zone 3

Laves γ Matrix

Zone 2 Zone 1

20 μm Fig. 15. Transition from epitaxial growth to equiaxed grains in clad zone near the surface (P ¼1200 W, V ¼ 600 mm/min, pre-placed layer thickness¼2 mm).

and incorporating all these parameters in a single analytical model is a complex task. Thus online monitoring of thermo-cycles and interpreting the cooling rate could help in estimating the stray grain layer thickness, and also to adjust the parameters to minimise or eliminate the formation of stray grains. Fig. 16(a)– (c) shows typical microstructure near the surface of laser cladded specimens, processed at a constant CW laser beam power of 1200 W and three different scan speeds of 1000, 800 and 600 mm/ min, respectively on a 2 mm pre-placed powder layer specimens. The cooling rates under these processing conditions were 680, 430 and 260 °C/s. It can be seen that with the decrease of cooling rate, fully dendritic structure with small traces of equiaxed grains, as in Fig. 16(a) is transforming into complete stray grains near the surface as presented in Fig. 16(b) and (c). Further, the thickness of stray grains layer and the size of each grain are found to increase with decreasing cooling rate. The stray grain layer thickness in case of Fig. 16(b) and (c) was found to increase from
60 mm to
85 mm with a decrease of cooling rate by approximately 40%.

5 μm

Fig. 17. Microstructure of Inconel 718 clad layer (P¼ 1200 W, V ¼ 200 mm/min, preplaced powder layer thickness¼ 2 mm). Table 3 Elemental composition (wt%) of different phases obtained from EDS analysis. Element (wt%)

Al Ti Cr Fe Co Ni Nb Mo

γ matrix

Granules

Laves

1

2

3

1

2

3

1

2

3

0.00 0.00 19.60 37.30 0.00 43.01 0.09 0.00

0.08 0.00 20.02 33.81 0.00 46.04 0.05 0.00

0.03 0.03 20.42 46.53 0.00 32.93 0.06 0.00

17.85 12.26 15.74 22.77 0.00 21.22 10.17 0.00

10.60 10.12 15.37 15.90 0.00 32.90 15.11 0.00

10.03 7.47 19.15 18.96 0.00 35.89 8.50 0.00

0.00 2.91 17.41 25.34 1.82 36.56 15.97 0.00

0.00 0.86 18.32 27.68 0.00 40.55 12.58 0.00

2.71 1.06 18.41 31.44 0.00 37.53 8.85 0.00

Thus, online monitoring of thermo-cycles and interpreting the cooling rates could help in controlling the microstructure. Fig. 17 shows the microstructure of a cladded specimen with 1200 W CW laser beam power and 200 mm/min scan speed on 2 mm pre-placed powder layer specimen where the segregation of the alloying elements can be clearly observed, forming different phases like Laves, γ matrix and granules. Point EDS analysis was carried out on all these phases at three different locations to determine the chemical composition and results are tabulated in Table 3. The γ matrix was found to be deficient in alloying elements which are segregated in Laves and granules. The solidification initiates in γ austenitic phase, pushing the alloying elements into liquid solution, resulting in increase of concentration of Nb, Al, Ti in the liquid solution [37,38]. This results in formation of Laves. Thus, the amount of segregation within a material basically

Fig. 16. Effect of cooling rate on stray grains layer thickness (a) 680 °C/s, (b) 430 °C/s and (c) 260 °C/s corresponding to scan speeds of 1000, 800 and 600 mm/min respectively, (P ¼1200 W, pre-placed layer thickness¼ 2 mm).

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

(a)

(b)

10 μm

149

(c)

10 μm

10 μm

Fig. 18. Variation in grain size with cooling rate at different scan speeds (a) 600 mm/min, (b) 1000 mm/min and (c) 1400 mm/min (P ¼1200 W, pre-placed powder layer thickness¼2 mm).

Fig. 19. XRD peaks of specimens cladded at a scan speed of (a) 1400 mm/min and (b) 600 mm/min (P¼ 1200 W, pre-placed powder layer thickness¼ 2 mm).

1400 mm/min

600 mm/min

Hardness (HV0.05 )

400 350 300 250

200 150 0 200 400 600 800 Distance from clad surface to substrate (μm)

Fig. 20. Hardness evolution from substrate to deposited material (P¼ 1200 W, preplaced powder layer thickness¼2 mm).

depends on the solidification rate which is controlled by the cooling rate and the temperature gradient. Faster the cooling rate lesser will be the segregation. Rapid solidification results in entrapping of these alloying elements within the γ matrix reducing the segregation. Fig. 18 shows the microstructures of the specimens cladded at 1200 W CW laser beam power and three different scan speeds of 600, 1000 and 1400 mm/min. The cooling rates measured experimentally corresponding to these scan speeds are 260, 550 and 700 °C/s respectively. It can be clearly observed that as the scan speed decreases from 1400 to 600 mm/min, the γ matrix size, which is basically Ni–Cr–Fe phase, is increasing with segregation of phase strengthening elements at the grain

boundary as Laves phase. This increase in the size of γ phase results in retardation of Laves phase dissolution during the solution heat treatment as distance to be travelled by segregations increases. Fig. 19 shows the XRD results of specimens cladded with scan speed of 1400 mm/min and 600 mm/min respectively. Domination of Laves phases at low scan speed or low cooling rate can be clearly observed. from the XRD peaks. Further, it can be observed that the phase strengthening peaks of γ′ and γ″ are missing because of consumption of Nb and Ti in the Laves phase in the form of (Ni, Fe, Cr)2 (Nb, Mo, Ti) [39] which is also evident from the EDS results in Table 3. In order to investigate the effect of elemental segregation on mechanical properties, Vickers hardness test and fracture test were carried out. Further, the specimens were heat treated and the effect of γ phase grain size on dissolution of segregated elements was investigated. Fig. 20 shows the hardness along the cross-section of the clad track from the clad surface to the substrate interface, cladded at 1200 W CW laser beam power and scan speed of 1400 mm/min and 600 mm/min. It is observed that in case of 1400 mm/min where the cooling rate is relatively high (700 °C/s) and Laves phases are relatively less, hardness is high compared to that at 600 mm/min (260 °C/s). However, the fracture surface showed a brittle mode failure in all the cases which indicates that Laves phases are brittle in nature. Since throughout the study, analysis was carried out on single clad tracks deposited with different combination of process parameters, the fractured surfaces were obtained by cutting the substrate from behind till the clad layer is approached using wire EDM and then clad layer was pulled out/fractured mechanically. Fig. 21 (a) and (b) shows the fractured surface of the as deposited specimens cladded at 1000 mm/min and 200 mm/min respectively. It is observed that both the specimens showed brittle mode of fracture.

150

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

(b)

(a)

9 μm

9 μm (d)

(c)

9 μm

9 μm

(f)

(e)

9 μm

9 μm

Fig. 21. Fractured surface of laser cladded specimens: (a), (b) before heat treatment; (c), (d) after heat treatment and (e), (f) microstructure after heat treatment (P ¼1200 W, scan speed 1000 mm/min for (a), (c) and (e); 200 mm/min for (b),(d),and (f), pre-placed powder layer thickness¼ 2 mm).

Further, the specimen cladded at 200 mm/min is also having intermetallics (spherical granules) as revealed in the fracture surface shown in Fig. 21(b). As-deposited specimens were solution heat treated at 1050 °C in Argon environment for 2 h and water quenched to dissolve the segregated elements in γ matrix. Then the specimens were precipitation hardened for 8 h at 750 °C, followed by aging at 650 °C for 8 h more and quenched with argon to room temperature. After the heat treatment the specimen cladded at 1000 mm/min showed a clear ductile mode fracture where as the specimen cladded at 200 mm/min did not show complete ductile fracture which is evident from Fig. 21(c) and (d) respectively. However, the spherical granules were found to disappear which signifies their dissolution during solution heat treatment. Upon inspecting the microstructure of the heat treated specimens it was observed that elemental segregations in case of 1000 mm/min got

completely dissolved as shown in Fig. 21(e) whereas that in 200 mm/min, there is only partial dissolution and it is observed from Fig. 21(f) that even after complete heat treatment there exists segregated elements in reduced amount. Fig. 22 shows the XRD peaks of these two specimens where the 2θ location of different phases found to coincide with γ matrix within large range of 2θ (30–100°) which was in good agreement with other investigations [40,41] . The presence of Laves phases even after heat treatment in case of 200 mm/min scan speed can be clearly observed, whereas in case of 1000 mm/min they are absent. Thus, it can be concluded that slow cooling results in heavy elemental segregations and increased γ matrix size. This, in turn retards the dissolution process of segregated elements during the heat treatment.

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

151

Fig. 22. XRD peaks of heat treated specimens cladded, (a) V ¼ 1000 mm/min (b) V ¼200 mm/min (P¼ 1200 W, pre-placed powder layer thickness¼2 mm).

4. Conclusion Through experiments, the effects of process parameters viz. laser beam power, scan speed, powder layer thickness and pulse on-time on the thermo-cycles in laser cladding of Inconel 718 were studied. The cooling rates, solidification time and the existence of quasi-steady state of molten pool and their effects on mechanical properties of the cladding were investigated and following conclusions are drawn: (1) The cooling rate which influences the resulting microstructure and the mechanical properties of the deposited component depends more on scan speed rather than laser beam power. (2) The surface temperature showed an oscillatory trend with increasing laser interaction time indicating the onset of evaporation of the molten pool. (3) The heating and cooling rates increase almost linearly with increasing scan speed. (4) With the increase of powder layer thickness, cooling rate decreases. (5) Cladding in pulse mode results in stacks of columnar dendrites aligned in different orientation due to the cyclic remelting and solidification of irradiated volume. (6) With the decrease of cooling rate, equiaxed grain layer thickness at the clad surface increases. (7) Slow cooling rate results in severe elemental segregation, formation of Laves phase and increase of γ matrix size which causes decrease in clad layer hardness and brittle mode of fracture. (8) In relatively large size γ matrix the segregated elements at its grain boundaries cannot be dissolved in the matrix even after standard heat treatment cycle, resulting in brittle mode of fracture.

Acknowledgement Authors gratefully acknowledge the financial support from the Department of Science and Technology, Ministry of Science and Technology, Government of India, under the FIST Program-2007 (SR/FIST/ETII-031/2007). They would also like to thank Professor Asimava Roy Choudhury, Department of Mechanical Engineering, IIT Kharagpur for providing cladding materials for the experiment. References [1] Sextona L, Lavin S, Byrne G, Kennedy A. Laser cladding of aerospace materials. J

Mater Process Technol 2002;22:63–8. [2] Costa L, Vilar R. Laser powder deposition. Rapid Prototyp J 2009;15:264–79. [3] Dubourg L, Archambeault J. Technological and scientific landscape of laser cladding process in 2007. Surf Coat Technol 2008;202:5863–9. [4] Toyserkani E, Khajepour A, Corbin S. Laser cladding.New York: CRC press; 2004. [5] Bandyopadhyay PP, Chicot D, Kumar CS, Decoopman X, Lesage J. Influence of sinking-in and piling-up on the mechanical properties determination by indentation: a case study on rolled and DMLS stainless steel. Mat Sci Eng A 2013;576:126–33. [6] Manoj M, Ganesh P, Rakesh K, Roy Choudhury A. Microstructure and mechanical properties of TiB2–TiC–Al2O3–SiC composite coatings developed by combined SHS, sol–gel and laser technology. Surf Coat Technol 2010;204:3471–80. [7] Chatterjee Satyajit, Ganesh P, Palai R, Wu JA, Kaul R, Dutta Majumdar J, Roy Choudhury A. Effect of h-BN addition on the properties of nanostructured Al2O3–TiB2–TiN based coatings developed by combined SHS and laser surface alloying. Surf Coat Technol 2010;204:1702–9. [8] Kahlen FJ, Kar A. Tensile strengths for laser-fabricated parts and similarity parameters for rapid manufacturing. J Manuf Sci Eng 2001;123:38. [9] Luo F, Yao JH, Hu XX, Chai GZ. Effect of laser power on the cladding temperature field and the heat affected zone. J Iron Steel Res Int 2011;18(1):73–8. [10] Lei YW, Sun R, Tang Y, Niu W. Numerical simulation of temperature distribution and TiC growth kinetics for high power laser clad TiC/NiCrBSiC composite coatings. Opt Laser Technol 2012;44:1141–7. [11] Tahmasbi HR, Fayaz GR. Three dimensional finite element modeling of laser solid freeform fabrication of turbine blades. Optik 2015;126:3382–4. [12] Smurov I, Ignatiev M. Real time pyrometry in laser surface treatment. In: Proceedings of the conference of laser processing: surface treatment and film deposition. The Netherlands; 1996, p. 529–64. [13] Ignatiev M, Smurov I, Flamant G, Senchenko V. Surface temperature measurements during pulsed metallic and ceramic materials laser action on metallic and ceramic materials. Appl Surf Sci 1996;96–98:505–12. [14] Ignatiev M, Smurov I, Flamant G, Senchenko V. Two-dimensional resolution pyrometer for real-time monitoring of temperature image in laser materials processing. Appl Surf Sci 1997;109/110:498–508. [15] Doubenskaia M, Bertrand pH, Smurov I. Pyrometry in laser surface treatment. Surf Coat Technol 2006;201:1955–61. [16] Doubenskaia M, Bertrand pH, Smurov I. Optical monitoring of Nd: YAG laser cladding. Thin Solid Films 2004;453–454:477–85. [17] Bi G, Gasser A, Wissenbach K, Drenker A, Poprawe P. Identification and qualification of temperature signal for monitoring and control in laser cladding. Opt Laser Eng 2006;44:1348–59. [18] Bi G, Schurmann B, Gasser A, Wissenbach K, Poprawe R. Development and qualification of a novel laser-cladding head with integrated sensors. Int J Mach Tool Man 2007;47:555–61. [19] Bi G, Gasser A, Wissenbach K, Drenker A, Poprawe R. Characterization of the process control for the direct laser metallic powder deposition. Surf Coat Technol 2006;201:2676–83. [20] Pavlov M, Novichenko D, Doubenskaia M. Optical diagnostics of deposition of metal matrix composites by laser cladding. LiM Phys Procedia 2011;12:674–82. [21] Smurov I, Doubenskaia M, Zaitsev A. Comprehensive analysis of laser cladding by means of optical diagnostics and numerical simulation. Surf Coat Technol 2013;220:112–21. [22] Emamian A, Alimardani M, Khajepour A. Correlation between temperature distribution and in situ formed microstructure of Fe–TiC deposited on carbon steel using laser cladding. Appl Surf Sci 2012;258:9025–31. [23] Yu J, Lin X, Wang J, Chen J, Huang W. Mechanics and energy analysis on molten pool spreading during laser solid forming. Appl Surf Sci 2010;256:4612–20. [24] Zhang Y, Li Z, Nie P, Wu Y. Effect of cooling rate on the microstructure of laserremelted INCONEL 718 coating. Met Mater Trans A 2013;44A:5513–21.

152

G. Muvvala et al. / Optics and Lasers in Engineering 88 (2017) 139–152

[25] Odabasi A, Necipunlu, Goller G, Eruslu MN. A study on laser beam welding (LBW) technique: effect of heat input on the microstructural evolution of superalloy inconel 718. Met Mater Trans A 2010;41A:2357–65. [26] Liu WC, Yao M, Chen ZL, Wang SG. Niobium segregation in Inconel 718. J Mater Sci 1999;34:2583–6. [27] Chang SH, Lee SC, Huang KT. Influences of γ″ and δ precipitations on the microstructural properties of 718 alloy through HIP, solid-solution, and different. Aging Heat Treat Mater Trans 2010;51(9):1683–8. [28] Zhang YN, Cao X, Wanjara P. Microstructure and hardness of fiber laser deposited Inconel 718 using filler wire. Int J Adv Manuf Technol 2013;69:2569– 81. [29] Janaki Ram GD, Venugopal Reddy A, Prasad Rao K, Madhusudhan Reddy G. Microstructure and mechanical properties of Inconel 718 electron beam welds. Mater Sci Technol 2013;21:1132–8. [30] Azadian Saied, Wei Liu-Ying, Warren Richard. Delta phase precipitation in Inconel 718. Mater Charact 2004;53:7–16. [31] Gäumann M, Bezencon C, Canalis P, Kurz W. Single-crystal laser deposition of superalloys: processing–microstructure maps. Acta Mater 2001;49:1051–62. [32] Madhukar Yuvraj K, Mullick Suvradip, Shukla Dinesh K, Kumar Shailesh, Nath Ashish K. Effect of laser operating mode in paint removal with a fiber laser. Appl Surf Sci 2013;264:892–901. [33] Meng QW, Geng L, Zhang BY. Laser cladding of Ni-base composite coatings onto Ti–6Al–4V substrates with pre-placed B4C þNiCrBSi powders. Surf Coat Technol 2006;200:4923–8. [34] Kwok CT, Cheng FT, Man HC. Laser-fabricated Fe–Ni–Co–Cr–B austenitic alloy on steels. Part I. Microstructures and cavitation erosion behaviour. Surf Coat Technol 2001;145:194–205. [35] Piasecki A, Kulka M, Kotkowiak M. Wear resistance improvement of 100CrMnSi6-4 bearing steel by laser boriding using CaF2 self-lubricating addition. Tribol Int 2016;97:173–91. [36] Thompson SM, Bian L, Shamsaei N, Yadollahi A. An overview of direct laser deposition for additive manufacturing; Part I: transport phenomena, modeling and diagnostics. Addit Manuf 2015;8:36–62. [37] Cieslak MJ, Knorovsky GA, Headley TJ, Romig AD. The solidification metallurgy of alloy 718 and other Nb containing superalloys. In: Loria EA, editor. Conference proceedings on superalloy 718 – Metallurgy and applications. TMSAIMS, Warrendale, PA: The Minerals, Metals and Materials Society; 1989. p. 59–68. [38] Baker H. Alloy phase diagrams.Materials Park, OH: ASM international; 1992. [39] Schirra JJ, Caless RH, Hatala RW. The effect of Laves phase on the mechanical properties of wrought and cast þ HIP Inconel 718. JOM 1991:375–88. [40] Jia Qingbo, Gu Dongdong. Selective laser melting additive manufacturing of Inconel 718 superalloy parts: densification, microstructure and properties. J Alloy Compd 2014;585:713–21. [41] Amato KN, Gaytan SM, Murr LE, Martinez E, Shindo PW, Hernandez J, et al. Microstructures and mechanical behavior of Inconel 718 fabricated by selective laser melting. Acta Mater 2012;60:2229–39.

Gopinath Muvvala is currently a Ph.D. student in the Department of Mechanical Engineering of Indian Institute of Technology, Kharagpur. He is working in the area of ‘Laser additive manufacturing’ and also on ‘Laser surface modification’. He obtained his M.Tech degree in Manufacturing Science and Engineering in 2013 from the same institute, and B.Tech degree in Mechanical Production and Industrial Engineering from GITAM University, Visakhapatnam, India in 2011. He has coauthored several conference papers in the area of laser cladding.

Debapriya Patra Karmakar is currently a Ph.D. student in the Department of Mechanical Engineering of Indian Institute of Technology, Kharagpur. He is working in the area of ‘Laser cladding of hardfacing alloy on hot-work tool steel’ and also on ‘Laser surface modification’. He obtained his M.Tech degree in Manufacturing Science and Engineering in 2013 from the same institute, and B. E. degree from Bengal Engineering and Science University, Shibpur (currently known as IIEST, Shibpur), India in 2011.

Dr. Nath started his scientific career in BARC, Mumbai, India after graduating from BARC Training School in 1972. He worked mainly on the development of high power lasers and their scientific and industrial applications. He obtained Doctoral degree in 1982. He was a PDF in University of Alberta, Canada during 1982–84. During 1986–07 he headed Industrial CO2 laser Programme and also Solid State and Semiconductor Lasers Program (2005–07) in RRCAT, Indore. Since 2008 he is a Professor in Mechanical Engineering Department, IIT Kharagpur. He has co-authored more than 135 journal papers. His present research interests include laser additive manufacturing, laser surface engineering, and underwater laser material processing.