Comparative Study of Ni-Sn Alloys Electrodeposited from Choline Chloride-Based Ionic Liquids in Direct and Pulsed Current

: This work presents, for the ﬁrst time, the electrodeposition of Ni-Sn alloys in pulse current, from deep eutectic solvents (choline chloride: ethylene glycol eutectic mixture). Additionally, in this study, we report a comparison of the electrodeposition methods known as pulse and direct current. The elemental composition of the ﬁlms, evaluated from EDX, remained almost constant independently on the electrodeposition parameters. The XRD data revealed the presence of the NiSn metastable phase, which has been conﬁrmed by DSC analysis. This phase shows a nanocrystalline structure with crystallite sizes between 12 and 20 nm. The use of pulse current electrodeposition method has led to an improvement of alloys’ mechanical properties. Moreover, by controlling the electrodeposition parameters, we succeeded in tuning the mechanical properties of the coatings prepared through the PC method. We showed that the hardness parameters exhibited by the Ni-Sn alloys are strongly dependent on their crystallite sizes.


Introduction
Metal coatings have been widely used in surface protection or as decorative applications. Tin is a soft metal and easy to melt (223 • C melting point). However, its low melting point makes it difficult to use in its pure state. When tin is combined with other metals such as nickel, the resulting alloys are suitable for a wide range of applications such as surgical instruments, watch pars, printed circuits boards, and electrical connectors, which represent an alternative to chromium decoration due to their good anticorrosive properties. Recently, they have been investigated as anode material in lithium ion batteries [1][2][3].
Nickel-tin alloys are usually electrodeposited from aqueous electrolytes containing chloride/fluoride or pyrophosphates compounds. Parkinson in 1950 reported, for the first time, the electrodeposition The Ni-Sn alloys were obtained by plating using two different methods: direct current (DC) and pulse plaiting (PC). For DC electrodeposition, three different current densities (5, 10, and 15 mA/cm 2 ) were applied during 30 min at 75 ± 5 • C. In PC, the average current density was fixed at 6.67 mA/cm 2 , as well as the duty cycle (θ = 0.33), and different frequencies were applied: 0.067, 0.67, and 6.7 Hz for 120 min, at the same temperature. The obtained Ni-Sn deposits were rinsed with acetone, rinsed with deionized water, and air dried.
Morphological and compositional property studies of the Ni-Sn alloys were performed by Scanning Electron Microscopy (SEM) (Tokyo, Japan) using a Hitachi SU 8230 system equipped with an Oxford Energy Dispersive X-ray detector (EDX) (High Wycombe, United Kingdom). The structural properties of the coatings were investigated by Rigaku SmartLab X-ray Diffractometer (XRD) system (Tokyo, Japan) equipped with Cu K α radiation (λ = 0.15406 nm), operating at room temperature in the 2θ range 5 • -90 • .
The phase transformations of Ni-Sn alloys under heat treatment were studied by Differential Scanning Calorimetry (DSC) (Selb, Germany) with a Netzsch 404C Pegasus instrument in a purified argon atmosphere, at a scanning rate of 20 • C/min, from room temperature up to 500 • C.
The mechanical properties of the samples were evaluated by nanoindentation tests performed on a NanoTest TM Vantage system (Wrexham, United Kingdom) using a three-sided pyramidal Berkovich diamond tip. In order to avoid the interference of the substrate, the electrodeposited coatings were indented in a cross-section. At least 10 measurements were performed on each sample. A load of 20 mN was applied and the hardness H was determined, according to the following expression [30].
where P max is the maximum applied load measured at the maximum depth of penetration and A is the projected area determined by the equation below.
where h f is the contact indentation depth obtained from the total depth (h max ) using the following formula.
In Equation (3), ε is a geometrical parameter having a value of 0.75 for a pyramidal indenter, P is the maximum applied load, and S is the contact stiffness determined from the following relationship. where β is a correction factor and E r is the reduced elastic modulus defined by the equation below.
E s is the specimen elastic modulus, and E i and υ i,s are the elastic modulus and the Poisson's ratio, respectively.

Direct Current Electrodeposition of Nickel-Tin Alloys
Different amounts of nickel and tin salts have been added to a DES of choline chloride: ethylen glycol (ILEG), which results in three different types of electrolytes: ILEG-NiSn 1 , ILEG-NiSn 2 , and ILEG-NiSn 3 , as shown in Table 1. All the mixtures were in a liquid state at room temperature, with a green colour, and electrical conductivities in the range of 2-12 mS/cm for temperatures between 25-80 • C.
Nickel tin alloys have been electrodeposited in DC on Cu substrates in a galvanostatic mode, at different current densities. The obtained deposits were bright, grey, and adherent to the copper substrates. The coatings exhibited different calculated thicknesses (using Faraday's law) in the range of 4 to 10 µm, depending on the applied current density. Figure 1 shows examples of SEM micrographs of Ni-Sn deposits obtained from the different ILEG-NiSn systems. Es is the specimen elastic modulus, and Ei and i,s are the elastic modulus and the Poisson's ratio, respectively.

Direct Current Electrodeposition of Nickel-Tin Alloys
Different amounts of nickel and tin salts have been added to a DES of choline chloride: ethylen glycol (ILEG), which results in three different types of electrolytes: ILEG-NiSn1, ILEG-NiSn2, and ILEG-NiSn3, as shown in Table 1. All the mixtures were in a liquid state at room temperature, with a green colour, and electrical conductivities in the range of 2-12 mS/cm for temperatures between 25-80 °C.
Nickel tin alloys have been electrodeposited in DC on Cu substrates in a galvanostatic mode, at different current densities. The obtained deposits were bright, grey, and adherent to the copper substrates. The coatings exhibited different calculated thicknesses (using Faraday's law) in the range of 4 to 10 µ m, depending on the applied current density. Figure 1 shows examples of SEM micrographs of Ni-Sn deposits obtained from the different ILEG-NiSn systems. As shown in Figure 1, no dendritic growth was observed. Different surface morphologies defined by acicular (ILEG-NiSn2), circular (ILEG-NiSn3), and quasi-circular (ILEG-NiSn1) features have been identified, as a result of the variation of current density and the electrolyte nature. The Ni-Sn coatings prepared from the ILEG-NiSn3 system did not exhibited any significant difference in morphology with the variation of the applied current density, where all the coatings seem to be formed from relatively ordered circular grains. Jellesen and Møller observed a similar morphology for electrodeposited Ni-Sn alloys from an aqueous fluoride bath [6]. As shown in Figure 1, no dendritic growth was observed. Different surface morphologies defined by acicular (ILEG-NiSn 2 ), circular (ILEG-NiSn 3 ), and quasi-circular (ILEG-NiSn 1 ) features have been identified, as a result of the variation of current density and the electrolyte nature. The Ni-Sn coatings prepared from the ILEG-NiSn 3 system did not exhibited any significant difference in morphology with the variation of the applied current density, where all the coatings seem to be formed from relatively ordered circular grains. Jellesen and Møller observed a similar morphology for electrodeposited Ni-Sn alloys from an aqueous fluoride bath [6].
The chemical composition of the deposited films was determined by EDX. The EDX elemental maps for the samples prepared at 15 mA/cm 2 from different types of electrolytes reveal a good chemical homogeneity of the Ni-Sn coatings (see Figure 2). Moreover, we noticed that the composition of the deposits is not influenced by the electrodeposition parameters such as current density and mole ratio of salts in the electrolytes. As can be observed in Table 2, the weight percentages determined by EDX are around 32 wt.% Ni and 68 wt.% Sn (approximately 1:1 atomic Ni:Sn). It should be mentioned that our results are in good agreement with the studies reported by Parkinson, who observed that variations between 20%-30% in the molar ratio of metal ions and/or in the temperature do not induce significant changes in the composition of the alloys electrodeposited from aqueous-based electrolytes [4]. The chemical composition of the deposited films was determined by EDX. The EDX elemental maps for the samples prepared at 15 mA/cm 2 from different types of electrolytes reveal a good chemical homogeneity of the Ni-Sn coatings (see Figure 2). Moreover, we noticed that the composition of the deposits is not influenced by the electrodeposition parameters such as current density and mole ratio of salts in the electrolytes. As can be observed in Table 2, the weight percentages determined by EDX are around 32 wt.% Ni and 68 wt.% Sn (approximately 1:1 atomic Ni:Sn). It should be mentioned that our results are in good agreement with the studies reported by Parkinson, who observed that variations between 20%-30% in the molar ratio of metal ions and/or in the temperature do not induce significant changes in the composition of the alloys electrodeposited from aqueous-based electrolytes [4].  The XRD patterns obtained from different types of electrolytes are represented in Figure 3. For the Ni-Sn alloy prepared from ILEG-NiSn1 electrolyte, the XRD pattern was plotted using a logarithmic scale in order to show the low intensity peaks. The diffraction peaks at 2θ = 30.50°, 43.77°, 54.61°, 60.15°, 63.61°, 72.66°, 80.69°, and 89.92° were assigned to NiSn metastable phase and the peaks at 2θ = 44.32°, 50.48°, and 74.10° were attributed to the copper substrate (card No.: 00-004-0836) [31]. The formation of the metastable phase NiSn is independent on the concentration ratio of metal salts in the electrolyte. The formation of the NiSn metastable phase has been explained by several authors. This phase has a hexagonal network with a structure similar to Ni3Sn2 stable phase, where additional tin is incorporated into the alloy by inducing modifications in the lattice constants, which was found by Rooksby [31]. Parkinson tried to find an explanation of the formation of this compound suggesting that the free energy for its formation is lower than the one of the other possible nickel-tin compounds [4]. Additionally, a change in the occupancy of Ni atoms in the unit cell of the stable phase Ni3Sn2 can lead to the NiSn metastable phase as proposed by Watanable et al. [32]. Another explanation proposed by Lo is that the NiSn alloy is actually formed by the Ni3Sn2 phase with Sn atoms segregated at the grain boundaries [33].  The XRD patterns obtained from different types of electrolytes are represented in Figure 3. For the Ni-Sn alloy prepared from ILEG-NiSn1 electrolyte, the XRD pattern was plotted using a logarithmic scale in order to show the low intensity peaks.  [31]. The formation of the metastable phase NiSn is independent on the concentration ratio of metal salts in the electrolyte. The formation of the NiSn metastable phase has been explained by several authors. This phase has a hexagonal network with a structure similar to Ni 3 Sn 2 stable phase, where additional tin is incorporated into the alloy by inducing modifications in the lattice constants, which was found by Rooksby [31]. Parkinson tried to find an explanation of the formation of this compound suggesting that the free energy for its formation is lower than the one of the other possible nickel-tin compounds [4]. Additionally, a change in the occupancy of Ni atoms in the unit cell of the stable phase Ni 3 Sn 2 can lead to the NiSn metastable phase as proposed by Watanable et al. [32]. Another explanation proposed by Lo is that the NiSn alloy is actually formed by the Ni 3 Sn 2 phase with Sn atoms segregated at the grain boundaries [33].  The presence of copper peaks has been observed only for the sample electrodeposited from the ILEG-NiSn1 system, which means that this prepared film has a lower thickness as compared to the other samples, which can be seen in Figure 3b. To confirm this assumption, the current efficiency of the NiSn electrodeposited films was determined. The current efficiency refers to the ratio between the actual mass deposited vs. the one predicted theoretically. The deposit obtained from the ILEG-NiSn 1 system showed, as expected, the lowest current efficiency, at around 50% ( Figure 4). Meanwhile, the samples prepared based on the other systems, ILEG-NiSn 2 and ILEG-NiSn 3 , revealed a current efficiency between 75% to 78%. The lowest current efficiency was observed when the nickel to tin ratio was below one.  The presence of copper peaks has been observed only for the sample electrodeposited from the ILEG-NiSn1 system, which means that this prepared film has a lower thickness as compared to the other samples, which can be seen in Figure 3b. To confirm this assumption, the current efficiency of the NiSn electrodeposited films was determined. The current efficiency refers to the ratio between the actual mass deposited vs. the one predicted theoretically. The deposit obtained from the ILEG-NiSn1 system showed, as expected, the lowest current efficiency, at around 50% ( Figure 4). Meanwhile, the samples prepared based on the other systems, ILEG-NiSn2 and ILEG-NiSn3, revealed a current efficiency between 75% to 78%. The lowest current efficiency was observed when the nickel to tin ratio was below one.   For all ILEG-NiSn systems, a decrease in the current efficiency with the increase of the current density was noticed (see in Figure 5 an example for the ILEG-NiSn 1 system) caused by the faster depletion of the active metal ions near the cathodic surface at higher current densities [34]. For all ILEG-NiSn systems, a decrease in the current efficiency with the increase of the current density was noticed (see in Figure 5 an example for the ILEG-NiSn1 system) caused by the faster depletion of the active metal ions near the cathodic surface at higher current densities [34]. The formation of the metastable phase NiSn was confirmed by heat treatment studies using differential scanning calorimetry. As seen in Figure 6, one exothermic peak is observed at 292 °C. This peak is associated with the transformation of NiSn in both Ni3Sn2 and Ni3Sn4 phases, which confirms that the metastable phase was obtained by electrodeposition, as described by Dutta and Clarke [35]. After completing the phase transformation, the sample was subjected to a second thermal cycle (2 nd DSC measurement), where no peak was observed, which indicates the irreversibility of the process.

Pulse Current Electrodeposition of Nickel-Tin Alloys
The second part of the study was focused on the electrodeposition of Ni-Sn coatings in pulse current plating. To the best of our knowledge, this is the first time when nickel tin alloys were electrodeposited from deep eutectic solvents in pulse current. The influence of pulse current in the morphology, composition, and mechanical properties of the deposited Ni-Sn coatings has been investigated. The schematic diagram of the pulse plating wave is represented in Figure 7, where ip indicates the peak current density (mA/cm 2 ) and TON and TOFF refer to the on and off time durations of the pulse. The average current density has been determined, according to the following expression. The formation of the metastable phase NiSn was confirmed by heat treatment studies using differential scanning calorimetry. As seen in Figure 6, one exothermic peak is observed at 292 • C. This peak is associated with the transformation of NiSn in both Ni 3 Sn 2 and Ni 3 Sn 4 phases, which confirms that the metastable phase was obtained by electrodeposition, as described by Dutta and Clarke [35]. After completing the phase transformation, the sample was subjected to a second thermal cycle (2 nd DSC measurement), where no peak was observed, which indicates the irreversibility of the process. For all ILEG-NiSn systems, a decrease in the current efficiency with the increase of the current density was noticed (see in Figure 5 an example for the ILEG-NiSn1 system) caused by the faster depletion of the active metal ions near the cathodic surface at higher current densities [34]. The formation of the metastable phase NiSn was confirmed by heat treatment studies using differential scanning calorimetry. As seen in Figure 6, one exothermic peak is observed at 292 °C. This peak is associated with the transformation of NiSn in both Ni3Sn2 and Ni3Sn4 phases, which confirms that the metastable phase was obtained by electrodeposition, as described by Dutta and Clarke [35]. After completing the phase transformation, the sample was subjected to a second thermal cycle (2 nd DSC measurement), where no peak was observed, which indicates the irreversibility of the process.

Pulse Current Electrodeposition of Nickel-Tin Alloys
The second part of the study was focused on the electrodeposition of Ni-Sn coatings in pulse current plating. To the best of our knowledge, this is the first time when nickel tin alloys were electrodeposited from deep eutectic solvents in pulse current. The influence of pulse current in the morphology, composition, and mechanical properties of the deposited Ni-Sn coatings has been investigated. The schematic diagram of the pulse plating wave is represented in Figure 7, where ip indicates the peak current density (mA/cm 2 ) and TON and TOFF refer to the on and off time durations of the pulse. The average current density has been determined, according to the following expression.

Pulse Current Electrodeposition of Nickel-Tin Alloys
The second part of the study was focused on the electrodeposition of Ni-Sn coatings in pulse current plating. To the best of our knowledge, this is the first time when nickel tin alloys were electrodeposited from deep eutectic solvents in pulse current. The influence of pulse current in the morphology, composition, and mechanical properties of the deposited Ni-Sn coatings has been investigated. The schematic diagram of the pulse plating wave is represented in Figure 7, where i p indicates the peak current density (mA/cm 2 ) and T ON and T OFF refer to the on and off time durations of the pulse. The average current density has been determined, according to the following expression.
where the frequency f (Hz) and the duty cycle θ are given by the equations below.
where the frequency f (Hz) and the duty cycle θ are given by the equations below.
Both the average current density in PC and the current density in DC were set to 6.67 mA/cm 2 . Moreover, in this study, the value of the duty cycle was fixed at 0.33. Due to the fact that the composition of the Ni-Sn coating is independent of the concentration of the metal salts in ILEG, we have continued our studies considering only the electrolyte containing an equal amount of nickel and tin salts (ILEG-NiSn2 system). The other operating parameters are summarized in Table 3.  The surface morphology and the composition of the obtained Ni-Sn deposits were characterized by SEM-EDX. As shown in Figure 8, there are no noticeable differences in the surface morphology between the coating prepared in direct current and the ones obtained in pulse current electrodeposition conditions. All the deposits showed circular grains distributed over the entire surface. Additionally, no considerable differences have been observed in the composition of the coatings. The thicknesses of the coatings varied between 9 to 16 µ m, as can be observed from the cross-section SEM images, illustrated in Figure 9. Both the average current density in PC and the current density in DC were set to 6.67 mA/cm 2 . Moreover, in this study, the value of the duty cycle was fixed at 0.33. Due to the fact that the composition of the Ni-Sn coating is independent of the concentration of the metal salts in ILEG, we have continued our studies considering only the electrolyte containing an equal amount of nickel and tin salts (ILEG-NiSn 2 system). The other operating parameters are summarized in Table 3. Table 3. Operating parameters for the electrodeposition of Ni-Sn alloys from ILEG-NiSn 2 system under direct current (sample DC) and pulse current (samples P1, P2, and P3). The surface morphology and the composition of the obtained Ni-Sn deposits were characterized by SEM-EDX. As shown in Figure 8, there are no noticeable differences in the surface morphology between the coating prepared in direct current and the ones obtained in pulse current electrodeposition conditions. All the deposits showed circular grains distributed over the entire surface. Additionally, no considerable differences have been observed in the composition of the coatings. The thicknesses of the coatings varied between 9 to 16 µm, as can be observed from the cross-section SEM images, illustrated in Figure 9. The XRD patterns ( Figure 10) indicate the presence of the diffraction peaks assigned to the same metastable phase NiSn, as described in the previous section. As can be observed in the patterns, all the analysed coatings reveal a preferred orientation known as the (110)   The XRD patterns ( Figure 10) indicate the presence of the diffraction peaks assigned to the same metastable phase NiSn, as described in the previous section. As can be observed in the patterns, all the analysed coatings reveal a preferred orientation known as the (110) plane. The position of this peak (θ-Bragg's angle) and the values of the full width at half maximum (β-FWHM) were used in the Scherrer's equation in order to determine the crystallite size, D, for each sample. The XRD patterns (Figure 10) indicate the presence of the diffraction peaks assigned to the same metastable phase NiSn, as described in the previous section. As can be observed in the patterns, all the analysed coatings reveal a preferred orientation known as the (110) plane. The position of this peak (θ-Bragg's angle) and the values of the full width at half maximum (β-FWHM) were used in the Scherrer's equation in order to determine the crystallite size, D, for each sample. D = kλ β cos(θ) (9) where k is the Scherrer's constant (considered as 0.94) and λ is the X-ray wavelength of the Cu K α line [36].

Electrolysis Parameter NiSn-DC
where k is the Scherrer's constant (considered as 0.94) and λ is the X-ray wavelength of the Cu Kα line [36]. The estimated crystallite sizes obtained by Scherrer's equation were: 20 nm for NiSn-DC, 17.6 nm for NiSn-P1, 14.4 nm for NiSn-P2, and 12.4 nm for NiSn-P3. All samples prepared under pulse electrodeposition conditions showed a smaller crystallite size as compared to NiSn-DC. In the direct plating, the current is always in on-time and the metal ions are reduced continuously at the cathodic site, which leads to an increase of the crystallite domains. The applied frequency influences the crystallite size. By increasing the frequency, from 0.067 Hz to 6.67 Hz, a reduction of the crystallite size was observed. The use of high pulse frequencies, with short off-time, results in a crystallite refinement because the metallic ions do not have enough time to reduce at the cathodic surface. Moreover, the cathodic current efficiency determined for the samples prepared under pulse electrodeposition conditions was lower than the one observed in DC (see Figure 11). Several factors can lead to a decrease in the current efficiency in pulse current electrodeposition such as the electrodissolution of metal during TOFF or side reactions (hydrogen evolution, metal degradation, fails of the embodiment of the adatoms in the crystal structure, etc.) during the on-time [34]. Vicenzo et al. found that the cathodic current efficiency of the pulse plating of Sn decreased with either the increase of the pulse frequency or decrease of the duty cycle [37]. For tin electrodeposition, the decrease of the current efficiency is dependent on the on-time. At shorter on-time, the discharge of the metal cation cannot be sustained because the overpotential is not fully fixed on the cathodic surface [34]. Ecker et al. have found that the current efficiency presents a maximum value for a specific ratio between the on-time and off-time for pulse plating of molybdenum and chromiummolybdenum alloy [38]. In our study, the maximum current efficiency for pulse electrodeposition conditions was observed for TON = 500 ms and TOFF = 1000 ms. The estimated crystallite sizes obtained by Scherrer's equation were: 20 nm for NiSn-DC, 17.6 nm for NiSn-P1, 14.4 nm for NiSn-P2, and 12.4 nm for NiSn-P3. All samples prepared under pulse electrodeposition conditions showed a smaller crystallite size as compared to NiSn-DC. In the direct plating, the current is always in on-time and the metal ions are reduced continuously at the cathodic site, which leads to an increase of the crystallite domains. The applied frequency influences the crystallite size. By increasing the frequency, from 0.067 Hz to 6.67 Hz, a reduction of the crystallite size was observed. The use of high pulse frequencies, with short off-time, results in a crystallite refinement because the metallic ions do not have enough time to reduce at the cathodic surface.
Moreover, the cathodic current efficiency determined for the samples prepared under pulse electrodeposition conditions was lower than the one observed in DC (see Figure 11). Several factors can lead to a decrease in the current efficiency in pulse current electrodeposition such as the electrodissolution of metal during T OFF or side reactions (hydrogen evolution, metal degradation, fails of the embodiment of the adatoms in the crystal structure, etc.) during the on-time [34]. Vicenzo et al. found that the cathodic current efficiency of the pulse plating of Sn decreased with either the increase of the pulse frequency or decrease of the duty cycle [37]. For tin electrodeposition, the decrease of the current efficiency is dependent on the on-time. At shorter on-time, the discharge of the metal cation cannot be sustained because the overpotential is not fully fixed on the cathodic surface [34]. Ecker et al. have found that the current efficiency presents a maximum value for a specific ratio between the on-time and off-time for pulse plating of molybdenum and chromium-molybdenum alloy [38]. In our study, the maximum current efficiency for pulse electrodeposition conditions was observed for T ON = 500 ms and T OFF = 1000 ms.
The use of pulse plating has numerous advantages, such as the decrease of hydrogen embrittlement, reduction of the porosity, and internal stresses as well as the increase in the hardness of the coatings [39][40][41][42]. Mohan et al. reported an increase in the hardness of a nickel deposit from 190 HV when direct electrodeposition conditions were used up to a 300 HV value when nickel was plated in the pulse current [42]. Moreover, the increase in the pulse frequency was found to improve the hardness of tin electrodeposited from a sulphate bath, as Mohan and Rajasekaran reported [42]. The use of pulse plating has numerous advantages, such as the decrease of hydrogen embrittlement, reduction of the porosity, and internal stresses as well as the increase in the hardness of the coatings [39][40][41][42]. Mohan et al. reported an increase in the hardness of a nickel deposit from 190 HV when direct electrodeposition conditions were used up to a 300 HV value when nickel was plated in the pulse current [43]. Moreover, the increase in the pulse frequency was found to improve the hardness of tin electrodeposited from a sulphate bath, as Mohan and Rajasekaran reported [42].
In the present study, the micro-mechanical properties of NiSn-DC, NiSn-P1, NiSn-P2, and NiSn-P3 samples were investigated in the cross-section. Table 4 shows the Ni-Sn films properties for samples obtained in both DC and PC conditions. As shown in this table, the use of pulse current improves the mechanical properties of the coatings as compared to DC. By increasing the frequency, a gradual increase in the hardness was observed from 9.92 GPa at 0.067 Hz, up to 12.73 GPa at 6.7 Hz. At short on-time, the diffusion layer is not extended completely on the solution, which leads to fine grain deposits that decrease the porosity of the coatings and improve their mechanical properties [42]. Moreover, a linear dependence between the hardness parameters and the inverse square root of crystallite sizes for Ni-Sn alloys has been observed, with a correlation factor of R 2 = 0.9, which can be seen in Figure 12.  In the present study, the micro-mechanical properties of NiSn-DC, NiSn-P1, NiSn-P2, and NiSn-P3 samples were investigated in the cross-section. Table 4 shows the Ni-Sn films properties for samples obtained in both DC and PC conditions. As shown in this table, the use of pulse current improves the mechanical properties of the coatings as compared to DC. By increasing the frequency, a gradual increase in the hardness was observed from 9.92 GPa at 0.067 Hz, up to 12.73 GPa at 6.7 Hz. At short on-time, the diffusion layer is not extended completely on the solution, which leads to fine grain deposits that decrease the porosity of the coatings and improve their mechanical properties [42]. Moreover, a linear dependence between the hardness parameters and the inverse square root of crystallite sizes for Ni-Sn alloys has been observed, with a correlation factor of R 2 = 0.9, which can be seen in Figure 12. The use of pulse plating has numerous advantages, such as the decrease of hydrogen embrittlement, reduction of the porosity, and internal stresses as well as the increase in the hardness of the coatings [39][40][41][42]. Mohan et al. reported an increase in the hardness of a nickel deposit from 190 HV when direct electrodeposition conditions were used up to a 300 HV value when nickel was plated in the pulse current [43]. Moreover, the increase in the pulse frequency was found to improve the hardness of tin electrodeposited from a sulphate bath, as Mohan and Rajasekaran reported [42].
In the present study, the micro-mechanical properties of NiSn-DC, NiSn-P1, NiSn-P2, and NiSn-P3 samples were investigated in the cross-section. Table 4 shows the Ni-Sn films properties for samples obtained in both DC and PC conditions. As shown in this table, the use of pulse current improves the mechanical properties of the coatings as compared to DC. By increasing the frequency, a gradual increase in the hardness was observed from 9.92 GPa at 0.067 Hz, up to 12.73 GPa at 6.7 Hz. At short on-time, the diffusion layer is not extended completely on the solution, which leads to fine grain deposits that decrease the porosity of the coatings and improve their mechanical properties [42]. Moreover, a linear dependence between the hardness parameters and the inverse square root of crystallite sizes for Ni-Sn alloys has been observed, with a correlation factor of R 2 = 0.9, which can be seen in Figure 12.  It seems that the hardness parameters revealed by the Ni-Sn coatings are dependent on their crystallite sizes, which showed a decreasing tendency with the increase of the electrodeposition frequency. There is a noteworthy similarity between this dependency ( Figure 12) and described by the classical Hall-Petch relation, which considers the grain size instead of the crystallite size. Therefore, we succeeded in tuning the mechanical properties of the prepared coatings by diminishing their crystallite sizes through the electrodeposition parameters.

Conclusions
We successfully synthetized Ni-Sn alloys from choline chloride:ethylen glycol (1:2) using two different methods known as DC and PC, where the last one mentioned was reported for the first time in the literature. In the present study, we demonstrated that the composition of the electrodeposited Ni-Sn alloys remains almost constant, regardless of the electrolyte composition and applied current density. The XRD analysis revealed the presence of the metastable phase NiSn (1:1 atomic). Additionally, this metastable phase was confirmed by differential scanning calorimetry analysis where a peak was observed at 292 • C, which indicates the irreversible transformation of NiSn (1:1) into stable Ni 3 Sn 2 and Ni 3 Sn 4 phases. Moreover, we demonstrated that the use of pulse current electrodeposition conditions leads to a decrease of the crystallite size. The Ni-Sn alloy coatings obtained in pulse current showed an improvement in the mechanical properties as compared to the ones obtained in DC.