Electrodeposition of Ag Alloys with Ni and W from ... - Semantic Scholar
ELECTRODEPOSITION OF AG ALLOYS WITH NI AND W FROM A THIOUREA-CITRATE ELECTROLYTE
A Dissertation Prospectus Presented By Avinash Kola to The Department of Chemical Engineering
In partial fulfillment of the requirements For the degree of Doctor of Philosophy In the field of Chemical Engineering Northeastern University Boston, Massachusetts
June 6th, 2014 1
ABSTRACT
Tungsten alloys, such as Ni-W are well recognized for their outstanding corrosion resistance, wear resistance and catalytic properties towards hydrogen generation. Nickeltungsten alloys also have the potential to act as a barrier layer in the semiconductor industry, to prevent diffusion of conducting metals (e.g.,Cu, Au) into the substrate. Tungsten, cannot be reduced alone, and requires the presence of certain inducing elements (e.g., Ni, Co, Fe) exhibiting induced codeposition to form alloys of W. This mechanism is not well understood. Electrodeposition of binary alloys of silver, such as tin-silver, silver-nickel and electroless deposition of silver-tungsten alloys are potential alternatives for lead-free materials in electronic packaging, printed circuit boards and other electronic components. However, the electrodeposition mechanism of Ag-Ni-W alloys has never been examined. Interest to electrodeposit all three elements, Ag-Ni-W comes from the motivation to tailor desired properties, including the increase of Ag hardness, corrosion resistance at high temperatures, while maintaining the favorable electrical properties of pure Ag. The examination and fundamental understanding of such a system will contribute towards developing a superior alloy with combined properties (Ni-W and Ag) The first report of the electrodeposition of a ternary Ag-Ni-W alloy is presented. The addition of Ag was found to lower the deposition rate of Ni and W. Chemical equilibria calculations were used to estimate the concentration of possible complexed species present in the electrolyte within a pH range 2 to 8. The most dominant species of Ag present were [AgTu4]+ and [AgTu3]+ irrespective of the pH, while NiHCit2 was the dominant at lower pH, while NiCit2 was negligible below pH 3 and was found to increase 2
with pH and was highest at pH 8. The concentration of Ni-Tu was found to be present only at low pH and above pH 4 was negligible. Only one W-cit species was found to present in the pH range 2-4 and was negligible over pH 4. Ag-Ni-W nanowire deposition was attempted in polycarbonate templates with 50 nm diameter and length of 6 um. The nanowires we released by dissolution of the membrane in dichloromethane. Transmission electron microscopy (TEM) showed that the Ag-Ni-W nanowires had smooth morphology, however they were non-uniform in length. The non-uniformity and weak strength of the nanowires is a strong function of the deposition conditions and the electrolytic hydrogen production, due to the low pH which is high in H+ ion concentrations. Understanding the reaction mechanism of the ternary alloys deposition and increased pH can help improve the current efficiency during deposition thereby resulting in alloys with different compositions suitable for a variety of technological applications.
3
1.0 INTRODUCTION
Tungsten alloys are well known for their outstanding properties, for example, NiW alloys are useful catalysts for electrolytic hydrogen generation, W imparts superior hardness and wear resistance to its alloys with Ni and Co, and W alloys have improved corrosion resistance compared to their codeposited counterparts [1-20]. The electrodeposition behavior of
W alloys was coined by Brenner as “induced
codeposition”, as W ions cannot be electrodeposited by itself, but can be fully reduced if codeposited with iron group elements, such as Ni, Co and Fe [21]. The recent advancement in multilevel interconnects technology is key for signal routing in ultra-large scale integrated (ULSI) circuits. Typical metallic conductors include an adhesion layer, a barrier layer, a conducting layer, a capping layer, and possibly an antireflective coating. Currently, aluminum and copper are used for on-chip interconnects. The integration of electroless technology in integrated circuit production was reviewed by Shacham-Diamand et al. [22]. Barrier layers of cobalt-tungstenphosphorus (Co-W-P) and nickel-tungsten-phosphorus (Ni-W-P) layers were deposited on silicon and silicon dioxide. The Co-W-P and Ni-W-P layers have higher hardness and melting point than similarly deposited Co-P and Ni-P [23]. Therefore, films with tungsten are expected to have better reliability and to act as better diffusion barriers for Cu interconnects when compared to similar films without tungsten. The excellent conductivity of silver and the relative simple procedure of its electroless deposition is a motivation to produce stable and corrosion-resistant silver films. Electroless silver plating baths are very unstable and thus short lived. Not at all surprising is the fact that the electrolyte stability and plating rate are markedly affected by 4
the pH [24]. Electrodeposition has the ability to better control the rate and achieve a higher rate than electroless deposition, and very stable electrolytes with cyanide have been the traditional norm. Over the years, many other electrolytes have been proposed, such as the ones involving nitrate, iodide, thiourea, thiocyanate, sulfamate, and thiosulfate. Thiourea is considered to be an effective chelating agent for Ag, and various groups have investigated the deposition of Ag with thiourea [25-29]. Silver being a noble metal, there exists a large difference in the reduction potential between Ni (-0.29 VSHE) and Ag (+0.799 VSHE), and a complexing agent is one way to achieve deposits with different alloy compositions, as it lowers the difference in reduction potentials. One advantage of thiourea, for deposition involving Ag and Ni from the same electrolyte, is the specific complexation of thiourea with Ag compared to Ni ions. The present work is devoted towards the replacement of cyanide electrolytes and to develop a single ternary stable electrolyte for the electrodeposition of Ag-Ni-W alloys that has good properties and a non-polluting nature. This proposal will address the electrodeposition parameters for Ag-Ni-W electrodeposition. As a first step in developing a single electrolyte for the ternary alloy, Ag-Ni-W alloys have been fabricated to probe the reaction mechanism and gain a better understanding for a variety of technological applications.
5
2.0 CRITICAL REVIEW
Ni-W alloys and Ag alloys with Ni or W can be fabricated by a variety of methods such as sputtering, electron beam evaporation and electrodeposition. The electrodeposition route can be cost effective compared to other techniques. In this chapter a review of the literature relevant to the electrodeposition of Ni-W and Ag alloys, its applications and the various advances towards tailoring its properties will be presented. The electrodeposition of tungsten from aqueous solutions has been attempted since the early 1930s. However, the electrodeposition of tungsten was found to be hindered by the formation of an oxide layer, which could not be further reduced further [30]. Holt et al. [31] showed that tungsten electrodeposition could be achieved in the presence of certain metals (Co, Ni , Fe) with tungsten content as high as 50 wt %. This phenomenon is known as “induced codeposition” and was first coined by Brenner in 1963 [21]. 2.2 Electrodeposition of Ternary Ag-Ni-W Alloys 2.2.2
Interest in Ag-Ni Alloys Ag-Ni alloys also exhibit remarkable catalytic properties [32], and are promising
candidates for electrical contacts and switches [33, 34]. However, the Ag- Ni deposition system remains one of the least examined systems to date via electrodeposition synthesis [28, 35-37]. According to the phase diagram [38], Ag and Ni are immiscible in bulk form and even intermetallic phases were not observed at high temperatures. However, the alloying effect at the nano-scale level is quite different from bulk, since the heat of formation reduces with decreasing particle size and hence alloying these metals becomes a possibility. Another aspect of the Ag-Ni system is the large difference in reduction
6
potential Ag(+0.799 VSHE) and Ni (-0.25 VSHE) [28], an important feature of thiourea especially while considering Ag-Ni deposition from the same electrolyte is that, while thiourea binds strongly to Ag+ ions, it form a weak complex with Ni2+ ions. This selective complexation becomes crucial to lower the large difference in redox potentials between. Liang et al. [28] investigated the formation of metastable Ag-Ni solid solution and their phase separation to elemental form upon thermal annealing. Excessive thiourea (0.2 M) was used to stabilize the electrolyte, and act as a complexing agent to selectively complex Ag (10 mM) and lower the reduction potential between Ag and Ni. The addition of 0.2 M thiourea polarized the reduction potential of Ag from 0.03 V MSE to -0.650 V, which indicates a strong complexing effect between thiourea and Ag, while the addition of 0.2 M thiourea shifted the potential of Ni (0.15 M) from -1.04 V to -1.08 V due to a weak complexing effect. Eom et al. [35], examined the deposition of the alloy in sodium citrate electrolytes. A shift in the Ag ion deposition potential to more negative values was observed in the presence of citrate. They identified pH to be a dominant factor to control the composition of the samples, as it affected the complexing of the metal ions with citrate. At low current densities (0.5 mA/cm2), dendrite formation was observed with a film composition of Ag60Ni40, and the microstructure transformed to a granular deposit with increasing current density, which correspondingly increased the Ni content. The dendrite formation was attributed to the low concentration of Ag ions in the electrolyte. 2.2.3
Interest in Ag-W Alloys
The application of silver for ultra large scale integration is promising due to its low bulk resistivity (1.59 µΩcm), relatively high melting point and higher electromigration 7
resistance compared to Cu, which is widely used in the down-scaling of interconnects. Suitable conductivities were reported for Ag fabricated by sputtering [39, 40] and electroplated thin films [41], while some drawbacks of Ag, such as corrosion in air and diffusion in SiO2 , could be avoided by using a suitable binary alloy such as Ag-W instead of pure Ag [42]. Electroless deposition was intensively adapted to examine Ag-W thin films [23, 42-50] . Shacham-Diamand et al. [23], examined and compared the effect of tungsten concentration in the electrolyte, microstructure and morphology of Ag and Ag -W thin films, and related their influence on the electrical properties. Thin films containing tungsten with improved reliability and as effective diffusion barriers was first demonstrated in Co-W-P and Ni-W-P thin films, which exhibited higher hardness and melting points than films deposited without tungsten. Inberg et al. [48-50] examined the Ag-W electroless deposition system extensively. They found that increasing the tungsten concentration in the electrolyte decreased the deposition rate of Ag, resulting in smooth and high quality Ag-W films. A maximum of 3.2 at % W was achieved when the [WO42]/ [Ag+] molar ratio was unity. Higher concentrations of tungsten in the electrolyte did not increase the tungsten at % in the deposit. Glickman et al. [45] examined the factors contributing to the resistivity of Ag-W films and was able to achieve a considerable decrease in the resistivity for 100 and 50 nm Ag-W films. They achieved a resistivity of 4.5 µohm-cm and 6.5 µohm-cm for 100 nm and 50 nm films respectively for very low W at % (0.6 and 0.9). They reported grain boundary scattering to be a dominant factor in controlling the electrical resistivity of sub100 nm films. Using post vacuum annealing, at low temperatures ~ 150 ºC for 1 hr, a 8
considerable drop in resistivity could be achieved, while 75-80 % of the resistivity drop was attained at annealing temperatures of 200-250 ºC for 1 hr. In spite of the recent efforts in exploring Ag-W alloys as potential candidates for electronic applications, information on the mechanism of Ag-W electrodeposition is very scarce in the literature. A detailed understanding of the mechanism and the effect of one element on the other is crucial.
9
3.0 EXPERIMENTAL A thiourea-citrate electrolyte was used for the electrodeposition of Ag-Ni-W alloys. The electrolyte conditions and procedure, deposit characterization, and cell designs are presented in this section. 3.1 Electrolyte, Conditions and Procedure The composition of the electrolyte for the deposition of Ag-Ni-W alloys is listed in Table 1. All electrolytes were prepared using de-ionized ultra-filtered (D.U.I.F) water from Fischer Scientific. The pH of the electrolytes was measured maintained at a value of 2. Copper plates from ESPI metals were used as working electrodes and a rectangle piece of Pt was used as the anode. All substrates were cleaned in dilute H2SO4 (10 vol %) in order to remove any copper oxides and cleaned in D.I.U.F water before deposition. The deposition was galvanostatically controlled using a Solartron potentiostat/function generator model 1287A. For partial current density measurements, the current density applied were 8 mA/cm2, 20 mA/cm2, 40 mA/cm2 and 80 mA/cm2 .The area in this case was 0.39 cm2. The Ag-Ni-W nanowires were electrodeposited, under the same deposition conditions mentioned above, in a polycarbonate membrane having the smallest region of the pore 50 nm in diameter and 6 µm in length. The nanowires were then dissolved in dichloromethane to remove the supporting template and release the nanowires. The dissolved nanowires were then subjected to centrifuge, and then dichloromethane was replaced with fresh solution. This procedure was repeated 3 times. The deposit thickness and composition was analyzed with a KEVEX Omicron energy dispersive X-ray fluorescence analyzer (XRF), at 40 keV, 2 mA in air with an acquisition time of 60 sec. 10
SEM analysis was done on a Hitachi S4800 at 3.0 KV and 4.5 K and 15.0 K magnification. TEM images were taken on a JEOL, JEM 1010 at 80 KV. Table 1. Electrolyte composition for Ag-Ni-W alloys. Chemical
Concentration (M)
Nickel Sulfate(M)
0.05
Sodium Tungstate (M)
0.015
Silver Sulfate
0.005
Sodium Citrate (M)
0.285
Thiourea (M)
0.650
Agitation of the electrolyte in the Hull cell is done by using air, bubbling close to the surface of the cathode to ensure uniform mixing. A flow meter is used to monitor the entering feed rate of air into the Hull cell. Five different flow rates, 1-5 L/min, were examined. For the conventional Hull cell experiments, two agitation conditions were examined, no agitation and 5 L/min. The primary purpose of any form of agitation to the electrolyte is to eliminate or minimize concentration gradients near the electrode surface to avoid mass transport limitations. In the case of Ag-Ni-W electrodeposition there is an order of magnitude lower amount of Ag ions so that mass transport effects may be considerable in that case. Air agitation can help then to control the boundary layer thickness during deposition, which is a crucial factor when depositing alloys involving diffusion limited species.
11
4.0 RESULTS AND DISCUSSION
4.5 Electrodeposition of Ni-W with/without Ag To examine the effect of Ag on the codeposition mechanism of Ni-W, partial current density measurements were done on a Cu substrate in a parallel configuration. Due to the instability of the electrolyte at pH 8, it was not examined further for this electrolyte concentration and deposition parameters. Polarization curves of the Ni-W electrolytes with and without Ag at an agitation rate of 5 L/min are shown in Figure 1. The deposition potential of Ni-W begins at -0.77 V and rises rapidly with increasing potential. The addition of Ag shows a clear variation in the deposition characteristics, with a start in deposition at -0.55 V, which is an indication of Ag deposition, reaches it’s limiting current density and then rises in current density around -0.88 V. We can also observe a slight shift in the Ni-W reaction. 20
NiW only
2
i (mA/cm )
15
10
5
NiW+ 5mMAg
+
0 0.4
0.5
0.6
0.7
0.8
0.9
1.0
E Vs Ag/AgCl
Figure 1. Polarization curves for Ag-Ni-W electrolyte with and without Ag (I). 12
Galvanostatic deposition was done on a Cu substrate in a parallel configuration for both the electrolytes at applied current densities of 8 mA/cm2, 20 mA/cm2, 40 mA/cm2 and 80 mA/cm2. Upon addition of Ag (I) into the electrolyte, as expected, Ag rich deposits are obtained at current densities, 8 mA/cm2 and 20 mA/cm2. As the current density increases, Ni composition increases in the deposit, due to the onset of Ni deposition at higher current densities, which is also observed from the polarization curves. However, the tungsten composition remains constant irrespective of the applied current density. With the addition of Ag in the electrolyte, nodules formation can be seen on the substrate and is uniform across the surface of the deposit (Figure 2 (a,b)), these nodules tend to get bigger with an increase in current density (Figure 2 (c,d)). Table 2 shows the composition variation with Ag (I) ions in the electrolyte. At 80 mA/cm2, the Ag composition drops drastically, indicating the onset of kinetic reduction of Ni.
Table 2. Composition of Ni, W and Ag. Current density (mA/cm2) 8 20 40 80
Ni wt %
W wt %
0.7 4.1 7.3 35.7
0.2 0.3 1.0 0.5
13
Ag wt % 99.0 95.4 91.6 63.7
(a)
(b)
(c)
(d)
Figure 2. Optical images of Ag-Ni-W deposits at 8 mA/cm2, 20 mA/cm2, 40 mA/cm2 and 80 mA/cm2: (a), (b), (c), (d) respectively.
Along with the composition data we are able to calculate the partial current densities of each species A semi-log plot (i vs EAgCl) is used because kinetic rates are typically exponential with potential. In Figures 3 (a and b), the reaction rate of Ni and W drop in the presence of Ag (I), however, also the deposition potential has been shifted to more positive values. No deposit can be seen at the low current density end in the absence of Ag (I), by physical observation of the samples. This thermodynamic shift in potential indicates that even though the reaction rate slows down, the presence of Ag (I) could in turn induce the deposition of Ni and W due to a more energetically favorable 14
condition, such as a change in the activity of the solid state. In the case of Ni deposition we observe a rise in Ni deposition rate at lower potential of -1.0 V, than observed in the absence of Ag (I), whereas in the case of W we see a very flat profile in the presence of Ag (I).
14 12
log iNi (mA/cm2)
10 8 6 4 2 0 0.4
0.6
0.8
1.0 1.2 1.4 E vs. Ag/AgCl (V)
0.6
0.8
1.0 1.2 1.4 E vs. Ag/AgCl (V)
1.6
1.8
2.0
0.5
log iW (mA/cm2)
0.4 0.3 0.2 0.1 0.0
0.4
1.6
1.8
2.0
Figure 3. Partial current density, with and without Ag (I) for (a) Ni and (b) W.
15
Figure 4 shows the side reaction partial current density. In the presence of Ag the side reaction occurs at lower potentials -0.6 V, and consumes most of the current, thereby indicating a low efficiency of the deposition process. The partial current density increases drastically compared to the Ag (I). This increase in partial current density could hinder the deposition of less noble Ni and WO42- ions.
80
2
log ij (mA/cm )
60
40
20
0 0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
E (V)
Figure 4. Partial current density of side reactions, with and without Ag (I).
4.3 Electrolyte Stability and Complex Species Distribution The stability of the electrolyte at pH 2 was much higher compared to the electrolyte at pH 8. However, the electrolyte at pH 2 would decompose over a period of time (~ 6 hrs). The electrolyte at pH 8 would precipitate even during deposition, even though the average current density was low 1.7 mA/cm2. This instability could be related to different species forming at different pH. Equilibria calculations, help in determining
16
the species distribution in an electrolyte, depending on their complex stability constants. Table 3 shows the mass balance and equilibria calculations used to calculate the distribution of the complexing species. Table 3. Mass Balance and Equilibria Equations. Species
Stability constant Log K
Equations
Mass Balance Eq
𝑪𝑵𝒊
𝑪𝑪𝒊𝒕
𝑪𝑵𝒊 + 𝑪𝑵𝒊𝑪𝒊𝒕 + 𝑪𝑵𝒊𝑯𝑪𝒊𝒕 + 𝑪𝑵𝒊𝑪𝒊𝒕𝟐 + 𝑪𝑵𝒊𝑯𝑪𝒊𝒕𝟐 + 𝑪𝑵𝒊𝑻𝒖 = 𝟎. 𝟎𝟓
Mass Balance Eq
𝑪𝑾𝑶𝟒
𝑪𝑾𝑶𝟒 + 𝑪𝑾𝑶𝟒𝑯𝑪𝒊𝒕𝑯 + 𝑪𝑾𝑶𝟒 𝑪𝒊𝒕𝑯𝟐 + 𝑪𝑾𝑶𝟒 𝑪𝒊𝒕𝑯𝟑 − 𝟎. 𝟎𝟏𝟓 = 𝟎
Mass Balance Eq
Mass Balance Eq Mass Balance Eq Mass Balance Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq
𝑪𝑪𝒊𝒕 + 𝑪𝑵𝒊𝑪𝒊𝒕 + 𝑪𝑵𝒊𝑯𝑪𝒊𝒕 + 𝑪𝑵𝒊𝑪𝒊𝒕𝟐 + 𝑪𝑵𝒊𝑯𝑪𝒊𝒕𝟐 + 𝑪𝑾𝑶𝟒𝑯𝑪𝒊𝒕𝑯 + 𝑪𝑾𝑶𝟒 𝑯𝑪𝒊𝒕𝑯𝟐 + 𝑪𝑾𝑶𝟒 𝑯𝑪𝒊𝒕𝑯𝟑 − 𝟎. 𝟐𝟖𝟓 = 𝟎
𝑪𝑻𝒖
𝑪𝑻𝒖 + 𝑪𝑵𝒊𝑻𝒖 − 𝟎. 𝟔𝟓𝟎 = 𝟎
𝑪𝑨𝒈 𝑪𝑯
𝑪𝑨𝒈 + 𝑪𝑨𝒈𝑻𝒖 + 𝑪𝑨𝒈𝑻𝒖𝟐 + 𝑪𝑨𝒈𝑻𝒖𝟑 + 𝑪𝑨𝒈𝑻𝒖𝟒 − 𝟎. 𝟎𝟎𝟓 = 𝟎
𝑪𝑵𝒊𝑪𝒊𝒕
5.35
𝑪𝑵𝒊𝑪𝒊𝒕𝟐
8.11
𝑪𝑵𝒊𝑻𝒖
1.47
𝑪𝑾𝑶𝟒 𝑯𝑪𝒊𝒕𝑯𝟐
17.03
𝑪𝑨𝒈𝑻𝒖
7.11
𝑪𝑵𝒊𝑯𝑪𝒊𝒕
9.13
𝑪𝑵𝒊𝑯𝑪𝒊𝒕𝟐
13.5
𝑪𝑾𝑶𝟒𝑯𝑪𝒊𝒕𝑯
10.2
𝑪𝑾𝑶𝟒 𝑯𝑪𝒊𝒕𝑯𝟑
21.67
𝑪𝑨𝒈𝑻𝒖𝟐
10.61
𝑪𝑨𝒈𝑻𝒖𝟒
13.57
𝑪𝑨𝒈𝑻𝒖𝟑
𝑪𝑯 − 𝟏𝟎𝒑𝑯 = 𝟎
𝟏𝟎𝟓.𝟑𝟓 × (𝑪𝑵𝒊 × 𝑪𝑪𝒊𝒕 ) − 𝑪𝑵𝒊𝑪𝒊𝒕 = 𝟎
𝟏𝟎𝟗.𝟏𝟑 × (𝑪𝑵𝒊 × 𝑪𝑪𝒊𝒕 × 𝑪𝑯 ) − 𝑪𝑵𝒊𝑯𝑪𝒊𝒕 = 𝟎 𝟏𝟎𝟖.𝟏𝟏 × �𝑪𝑵𝒊 × 𝑪𝑪𝒊𝒕 𝟐 � − 𝑪𝑵𝒊𝑪𝒊𝒕𝟐 = 𝟎
𝟏𝟎𝟏𝟑.𝟓 × �𝑪𝑵𝒊 × 𝑪𝑪𝒊𝒕 𝟐 × 𝑪𝑯 � − 𝑪𝑵𝒊𝑯𝑪𝒊𝒕𝟐 = 𝟎 𝟏𝟎𝟏.𝟒𝟕 × (𝑪𝑵𝒊 × 𝑪𝑻𝒖 ) − 𝑪𝑵𝒊𝑻𝒖 𝟎
𝟏𝟎𝟏𝟎.𝟐 × �𝑪𝑾𝑶𝟒 × 𝑪𝑪𝒊𝒕 × 𝑪𝑯 𝟐 � − 𝑪𝑾𝑶𝟒𝑯𝑪𝒊𝒕𝑯 = 𝟎
𝟏𝟎𝟏𝟕.𝟎𝟓 × �𝑪𝑾𝑶𝟒 × 𝑪𝑪𝒊𝒕 × 𝑪𝑯 𝟑 � − 𝑪𝑾𝑶𝟒 𝑪𝒊𝒕𝑯𝟐 = 𝟎 𝟏𝟎𝟐𝟏.𝟔𝟕 × �𝑪𝑾𝑶𝟒 × 𝑪𝑪𝒊𝒕 × 𝑪𝑯 𝟒 � − 𝑪𝑾𝑶𝟒 𝑪𝒊𝒕𝑯𝟑 = 𝟎 𝟏𝟎𝟕.𝟏𝟏 × �𝑪𝑨𝒈 × 𝑪𝑻𝒖 � − 𝑪𝑨𝒈𝑻𝒖 = 𝟎
𝟏𝟎𝟏𝟎.𝟔𝟏 × �𝑪𝑨𝒈 × 𝑪𝑻𝒖 � − 𝑪𝑨𝒈𝑻𝒖𝟐 = 𝟎 𝟏𝟎𝟏𝟐.𝟕𝟑 × �𝑪𝑨𝒈 × 𝑪𝑻𝒖 � − 𝑪𝑨𝒈𝑻𝒖𝟑 = 𝟎
12.73
𝟏𝟎𝟏𝟑.𝟓𝟕 × �𝑪𝑨𝒈 × 𝑪𝑻𝒖 � − 𝑪𝑨𝒈𝑻𝒖𝟒 = 𝟎
17
A plot of the distribution of the species in an electrolyte mentioned in table 1 is shown in Figure 5. The entire Ag ions complex with thiourea to form [AgTu4]
+
and
remains constant irrespective of pH. The NiHCit2 complex (cit –citrate) is dominant at pH 2 and reduces in concentration with increasing pH. An inverted trend is observed for NiCit2 species which increases with pH and is dominant at pH 8. Different tungstate citrate species are present at different pH. The WO4HCit2 is dominant at pH 2 and reduces after pH 4. The WO4HCitH species increases from pH 3, reaches a peak at pH 5.5 and drops in concentration with increasing pH. The WO4CitH species increases from pH 6 and is highest at pH 8. The inset of Figure 5 also shows that Ni-Tu complex is very insignificant, indicating preferential complexation of thiourea with Ag ions. The stability issue at pH 8 could be due to the different complexed species forming at a higher pH which are unstable during deposition.
0.05 NiCit2 NiCit2 0.04
NiHCit2 NiHCit Concentration (M)
Concentration (M)
2
WO4HCitH2 [1,1,2] Wo4HCitH2
0.03
WO4HCitH3 [1,1,3] Wo4HCitH3 2-
WO4 Wo4
AgTU AgTU4
0.02
3E-08 2.5E-08 2E-08 1.5E-08 1E-08 5E-09 0
N…
1 2 3 4 5 6 7 8 pH
0.01
0 1
2
3
4
5
pH
6
7
8
9
Figure 5. Species distribution of different complexed species in an Ag-Ni-W electrolyte, inset shows the Ni-Tu complex species. 18
4.6 Electrodeposition of Nanowires The development of nanowires into porous templates is a challenge in systems where there is a substantial side reaction of gas evolution. If the side reaction is too voluminous then gas bubbles can block the template pores and prevent deposition. Guided by the conditions of the thin films fabricated in preceding sections a current density was selected to deposit Ag-Ni-W nanowires, into the polycarbonate membranes. Figure 6 shows the resulting Ag-Ni-W nanowires deposited under the same current density of 1.7 mA/cm2 and then released from the template. From the thin film results it is expected to have a composition that is Ag rich. What is notable is that the nanowires have different lengths. Thus, they easily break when released from the membrane. The longest length achievable after a deposition time of 1800 s was 5.0 microns long, which provides an estimate of the potential deposition rate (~0.0027 microns/s). Nanowires that are more robust are desired.
Figure 6. TEM of Ag-Ni-W alloy nanowires deposited at 1.7mA/cm2 for 1800 s into a PC membrane. 19
6.0 PROPOSAL The electrodeposition of ternary alloy Ag-Ni-W system has a variety of technological applications especially in the semiconductor industry. For these alloys to be viable for large scale industrial production, few key limitations need to be addressed. As mentioned earlier one goal of this project is to develop an environmentally friendly and stable electrolyte for silver deposition. Thus far we have achieved this goal and deposited Ag-rich Ni-W alloys from a stable, non-cyanide, pH 2 electrolyte. However, one drawback of the low pH is the high concentration of H+ ions in the electrolyte which increases the side reactions and lowers current efficiency. In order to address the issue of low current efficiency and to deposit such alloys in deep recessed substrates, for example, nanowires, a better understanding of the electrodeposition mechanism of Ag-Ni-W alloys is necessary. Analysis from the conventional Hull cell, in which agitation by air bubbling is employed, does give us some information on the effect of Ag (I) addition to Ni and W deposition rates. But, since Ag (I) is a noble metal species and is mostly under diffusion limited control, better control over the boundary layer thickness is crucial to overcome certain limitations such as, poor boundary layer control, the inconsistency due to the variation in bubble flow pattern and operable limit (maximum flow rate 5 L/min).The rotating cylinder setup gives a well-defined control over the hydrodynamic conditions within the electrolyte system. For the next set of experiments, we propose to use the RCE setup to examine the effect of Ag-Ni-W alloy deposition under different mixing conditions, and use the complexation model to identify the optimum pH for the deposition of Ag-Ni-W alloys. The effect of concentration of metal ions and different additives in the electrolyte on the current efficiency will also be examined. 20
The following tasks will be focused on for future work: AIM 1: Improve the overall current efficiency for practical applications. Although, a pH 2 electrolyte was found to be stable for the electrodeposition of Ag-Ni-W alloys over the examined concentration range, low pH electrolytes lead to very low current efficiencies due to the high concentration of H+ ions in the electrolyte leading to the following reaction below 2𝐻 + + 2𝑒 − → 𝐻2
The hydrogen evolution reaction due to high H+ ion concentration can be dealt with by increasing the pH. A look at the complexation model Figure (5), we can examine the pH range we can focus on, since we have shown that pH 8 is not stable, and leads to precipitation during electrodeposition. We notice different species present at pH 2 and pH 8, at pH 2 we know that WO4HCitH3 and NiHCit2 are stable while at pH 8 either NiCit2 or WO42- leads to instability. For this task electrodeposition of the Ag-Ni-W alloy at three different pH (3, 4, 5) will be done and this will contribute towards answering this question of which species causes the instability – NiCit2 or Wo42. A constant rotation rate of 2825 rpm will be used at an applied current density of 1.7 mA/cm2 for 30 min. Samples will be weighed before and after deposition, XRF analysis will be done to measure the
composition of the alloy, and calculate current efficiencies. Another
approach towards improving current efficiency is to increase the overall concentration of the metal ions in the electrolyte. For this aim, concentration of each species will be double to 0.1 M nickel sulfate, 0.01 M silver sulfate, 0.03 M sodium tungstate, 0.570 M sodium citrate and 1.3 M thiourea. Three experiments at different current densities 1.7mA/cm2, 5 mA/cm2 and 10 mA/cm2 will be applied at a constant rotation rate of 2825 21
rpm. Samples will be weighed before and after deposition and XRF analysis will be used to analyze the composition of the deposited alloy and calculate current efficiency. AIM 2: Identify the deposition behavior of Ag induced Ni, W reduction. Analysis from the partial current density experiments on the Hull cell indicated an induced effect of Ag ions on Ni and W deposition rate. This is known as under potential deposition (UPD) behavior of Ni and W species in the presence of Ag ions. The goal in this aim is to establish if the reduction behavior between Ag ions is coupled with Ni and W, i.e. if, the rate of Ni or W will increase with Ag deposition rate. Since Ni and W are under kinetic control in the applied current density range examined, rotation rate should not affect their deposition rate. If an increase in the deposition rate of Ni or W is observed, along with an increase in Ag deposition rate, then this is purely due to a coupled effect between Ag and Ni/W. For this task six current density experiments, 0.5 mA/cm2, 5 mA/cm2, 10 mA/cm2, 20 mA/cm2, 40 mA/cm2 and 80 mA/cm2 for 30 mins each and at three different rotation rates 706, 1412 and 2824 rpm will be done. XRF will be used to analyze the composition of the deposited alloys. AIM 3: Effect of different additives on the deposition of Ag-Ni-W alloys. The effect of different concentrations of additives, such as thiourea, boric acid, sodium gluconate and citrate on the deposition of Ag-Ni-W alloys will be examined. One approach to examine a large set of variables effectively is by using Factorial design experiments. A 2k design for k =3 factors/variables will be considered. This design has 8 experiments, which enables us to examine the effect of individual variables (each at two levels), and in addition three binary and one ternary interaction on one single parameter, for example in our case wt % of W. A comparison of the values of each effect gives us an 22
idea on what parameters play a crucial role in obtaining a desired outcome. Thiourea (0, 0.650 M), boric Acid (0, 0.5M), sodium gluconate (0, 0.5 M) and citrate (0, 285 M) will be considered for this study, using the optimum electrolyte for Ag-Ni-W after completion of aim 1 and 2. Each sample will be weighed before and after deposition for calculating the current efficiency, XRF will be used to analyze the composition of the alloy and SEM will be done to examine the surface morphology of the deposits. AIM 4: Investigate the growth mechanism of Ag-Ni-W nanowires. Template synthesized nanowires have received a great deal of attention over the past decade because they show great promise in a wide range of applications such as electronics, sensing, drug delivery and fabrication of solar cells. The electrodeposition of 1D nanostructures such as, Ag nanowires are attractive for their superior electrical and thermal conductivity. The electrodeposition of Ag-Ni-W ternary alloy nanowires is a novel aspect to this work. From preliminary experiments we were able to obtain Ag-Ni-W nanowires under the deposition conditions examined, however due to the low pH the hydrogen side reaction causes the nanowires to become brittle and break during release. Results from aim 1-3 will help towards solving the issue of low current efficiency and develop more robust Ag-Ni-W nanowires.
23
8.0 REFERENCES 1. Zayats, A. I., and Perekhrest, N. A., "Electrodeposition of Iron-Tungsten Alloys from Pyrophosphate Electrolytes.", J. Appl. Chem. USSR, 44, 1286-1291, (1971). 2. Chassaing, E., Quang, V., and Qiart, R., "Mechanism of Nickel-molybdenum alloy electrodeposition in citrate electrolytes.", J. Appl. Electrochem., 19, 839, (1989). 3. Akiyama, T., and Fukushima, H., "Recent Study on the Mechanism of the Electrodeposition of Iron Group Metal Alloys.", ISIIJ Int., 32, 787-789, (1992). 4. Stepanova, L. I., and Purovskaya, O. G, "Electrodeposition of Nickel-Based alloys with tungsten and molybdenum.", Metal. Finish., 96, 50,52-53, (1998). 5. Svensson, M., Wahlstrom, U., and Holmbom, G., "Compositiolnally Modulated Cobalt-tungsten Alloys Deposited from a Single Ammonical electrolyte.", Surf. Coat. Technol., 105, 218-223, (1998). 6. Yamasaki, T., "High-Strength Nanocrystalline Ni-Alloys Produced by Electrodeposition.", Mater. Phys. Mech., 1, 98-102, (2000). 7. Cesiulis, H., Donten, M., Donten, M. L., and Stojek, Z., "Electrodeposition of Ni-W, Ni-Mo and Ni-W-Mo Alloys from Pyrophosphate Baths.", Mater. Sci.(Medziagotyra), 7, 237-241, (2001). 8. Itoh, K., Wang, F., and Watanabe, T., "Relationship between the Crystallographic Structure of Electrodeposited Ni-W Alloy Film and Its Thermal Equilibrium Diagram ", J. Japan Inst. Metals,, 65, 1023-1028, (2001). 9. Wang, Y., Chen, M., Zhou, F., and Ma, E., "High tensile ductility in a nanostructured metal.", Nature, 419, 912-915, (2002). 10. Younes, O., and Gileadi, E., "Electroplating of Ni/W Alloys I. Ammoniacal Citrate Baths.", J. Electrochem. Soc., 149, C100-C111, (2002). 11. Jiles, D. C., "Recent advances and future directions in magnetic materials.", Acta Materialia, 51, (19), 59075939, (2003). 12. Karpralova, I. G., Perelygin,Y. P., and Semchenko, T. K. , "Electrodeposition of Nickel-tungsten Alloy from Acetate Electrolyte.", J. Appl. Chem. USSR, 76, 1524-1526, (2003). 13. Liu, R., Wang, H., Yao, J. Y., Li, X. P., and Ding, G. F., "Preparing Ni-W alloy films with low internal stress and high hardness by heat treating.", Surf. Rev. Lett., 14, 1107-1112, (2007). 14. Sriraman, K. R., Ganesh Sundara Raman, S. and Seshadri, S. K., "Corrosion Behaviour of Electrodeposited Nanocrystalline Ni-W and Ni-Fe-W Alloys.", Mater. Sci. Eng. A, A460-461, 39-45, (2007). 15. Wang, H., Liu, R., Cheng, F. J., Cao, Y., Ding, G. F., and Zhao, X. L., "Electrodepositing amorphous Ni-W alloys for MEMS.", Microelectron. Eng., 87, 1901-1906, (2010). 16. Weston, D. P., Harris, S. J, Capel, H., Ahmed, N., Shipway, P. H, and Yellup, J. M, , "Nanostructrured Co-W Coatings Produiced by Electrodeposition to Replace Hard Cr on Aerospace Components.", Trans. Inst. Met. Finish., 88, 47-56, (2010). 17. Chianpairot, A., Lothongkum, G., Schuh, C. A., and Boonyongmaneerat, Y., "Corrosion of nanocrystalline Ni-W alloys in alkaline and acidic 3.5wt.% NaCl solutions.", Corr. Sci., 53, 1066-1071, (2011). 18. Wei, G. Y., Lou, J. W., Ge, H. L., Yu, Y. D., Jiang, L., and Sun, L. X.,, "Co-W Films Prepared from Electrolplating Baths with Different Complexing Agents ", Surf. Eng., 28, 412-417, (2012). 19. Sun, S., Bairachna, T., and Podlaha, E. J., "Induced Codeposition Behavior of Electrodeposited NiMoW Alloys.", J. Electrochem. Soc., 160, (10), D434- D440, (2013). 20. Saraç, H., Patrick, M. A., Wragg, A. A., "Physical properties of the ternary electrolyte potassium ferriferrocyanide in aqueous sodium hydroxide solution in the range 10–90°C.", J.Appl. Electrochem., 23, (1), 5155, (1993). 21. Brenner, A., Electrodeposition of Alloys: Principles and Practice, Academic Press, New York, (1963). 22. Shacham-Diamand, Y., Dubin,V., and Angyal,M., "Electroless copper deposition for ULSI.", Thin Solid Films, 262, 93-103, (1995). 23. Shacham-Diamand, Y., Inberg, A., Sverdlov, Y., and Croitoru, N., "Electroless silver and silver with tungsten thin films for microelectronics and microelectromechanical system applications.", J. Electrochem. Soc., 147, 3345-3349, (2000). 24. Schlesinger, M. (2010). Electroless and Electrodeposition of Silver, in Modern Electroplating, John Wiley & Sons, Inc.
24
25. Azzaroni, O., Schilardi, P. L., Salvarezza, R. C., and Arvia, A. J., "Smoothening mechanism of thiourea on silver electrodeposition. Real time imaging of the growth front evolution.", Langmuir, 15, 1508-1514, (1999). 26. de Oliveira, G. M., and Carlos, I. A., "Silver-zinc electrodeposition from a thiourea solution with added EDTA or HEDTA.", Electrochim. Acta, 54, 2155-2163, (2009). 27. Gómez, E., García-Torres, J., and Vallés, E., "Study and preparation of silver electrodeposits at negative potentials.", J. Electroanal. Chem., 594, 89-95, (2006). 28. Liang, D., Liu, Z., Hilty, R. D., and Zangari, G., "Electrodeposition of Ag-Ni films from thiourea complexing solutions.", Electrochim. Acta., 82, 82-89, (2012). 29. Reents, B., Plieth, W., Macagno, V.A., and Lacconi, G.I, "Influence of thiourea on silver deposition: Spectroscopic investigation.", J Electroanal Chem, 453, 121-127, (1998). 30. Lassner, E., and. Schubert, W. D, "Tungsten-Properties Chemistry,Technology of the Element,Alloys and Chemical Compounds.", (1999). 31. Holt, M. L., and Kahlenberg, L., "The Deposition of Tungsten from Aqueous Alkaline Solutions.", Quart. Rev. Am. Electroplaters' Soc., 9, 41-52, (1933). 32. Guo, H., Chen, Y., Chen, X., Wen, R., Yue, G. H., and Peng D. L., "Facile synthesis of near-monodisperse Ag@Ni core-shell nanoparticles and their application for catalytic generation of hydrogen.", Nanotechnology, 22, (2011). 33. Tyler, E. H., Clinton, J. R., and Luo, H. L., "Electrical Resistivity of Metastable Ag-Ni Alloys.", Solid State Commun., 13, 1409-1411, (1973). 34. Swingler, J., "Performance and arcing characteristics of Ag/Ni contact materials under DC resistive load conditions.", IET Sci. Meas. Technol., 5, 37-45, (2011). 35. Eom, H., Jeon, B., and Kim, D., "Electrodeposition of Silver-nickel Thin Films with a Galvanostatic Method.", Mater. Trans., 51, 1842-1846, (2010). 36. Bdikin, I. K., Strukova, G. K., Strukov, G. V., Kedrov, V. V., Matveev, D. V., Zver'kov, S. A., and Kholkin, A. L., "Growth, crystal structure and stability of Ag-Ni/Cu films.", Mater. Sci. Forum, 514-516, 1166-1170, (2006). 37. Santhi, K., Karthick, S. N., Kim, H. J., Nidhin, M., Narayanan, V., and Stephen, A., "Microstructure analysis of the ferromagnetic Ag-Ni system synthesized by pulsed electrodeposition.", Appl. Surf. Sci., 258, 3126-3132, (2012). 38. Singleton, M., and Nash, P., Bull. Alloy Phase Diagrams, American Society for Metals, 8, 119-121, (1987). 39. Wang, Y., and Alford, T. L., "Formation of aluminum oxynitride diffusion barriers for Ag metallization.", Appl. Phys. Lett., 74, 52-54, (1999). 40. Hauder, M., Hansch, W., Gstöttner, J., and Schmitt-Landsiedel, D., "Electromigration resistance of sputtered silver lines using different patterning techniques.", Microelectron. Eng., 60, 51-57, (2002). 41. Baker, B. C., Freeman, M., Melnick, B., Wheeler, D., Josell, D., and Moffat, T.P., "Superconformal electrodeposition of silver from a KAg(CN)2-KCN-KSeCN electrolyte.", J. Electrochem. Soc., 150, C61-66, (2003). 42. Inberg, A., Bogush, V. Croitoru, N., and Shacham-Diamand, Y., "Electrochemical study of the mechanism of Ag(W) electroless deposition.", J. Electrochem. Soc., 154, D1-4, (2007). 43. Bogush, V., Inberg, A., Croitoru, N., Dubin, V., and Shacham-Diamond, Y., "Electroless deposition of novel Ag-W thin films.", Microelectron. Eng., 70, (489-494), (2003). 44. Bogush, V., Inberg, A., Croitoru, N., Dubin, V., and Shacham-Diamand, Y., "Material properties of very thin electroless silver-tungsten films.", Thin Solid Films, 426, 288-295, (2003). 45. Glickman, E. E., Bogush, V., Inberg, A., Shacham-Diamand, Y., and Croitoru, N., "Electrical resistivity of thin electroless Ag–W films for metallization.", Microelectron. Eng., 70, 495-500, (2003). 46. Glickman, E., Inberg, A., Bogush, V., Aviram, G., Croitoru, N., and Shacham-Diamand, Y., "On the mechanism of annealing effect in electrical resistivity of sub-100 nm Ag (1% W) films.", Microelectron. Eng., 76, 182189, (2004). 47. Glickman, E., Inberg, A., Bogush, V., Aviram, G., Popovitz, R., Croitoru, N., and Shacham-Diamand, Y., "Role of local microchemistry and surface structure in electrical resistivity of 50nm electroless films Ag-Woxygen.", Microelectron. Eng., 82, 307-313, (2005). 48. Inberg, A., Shacham-Diamand, Y., Rabinovich, E., Golan, G., and Croitoru, N., "Electroless-deposited Ag–W films for microelectronics applications.", Thin Solid Films, 389, 213-218, (2001). 49. Inberg, A., Shacham-Diamand, Y., Rabinovich, E., Golan, G., and Croitoru, N., "Material and electrical properties of electroless Ag-W thin film.", J. Electron. Mater., 30, 355-359, (2001). 50. Inberg, A., Zhu, L., Hirschberg,G., Gladkikh, A., Croitoru, N., Shacham-Diamand,Y., and Gileadi, E., "Characterization of the Initial Growth Stages of Electroless Ag(W) Films Deposited on Si(100).", J. Elechrochem. Soc., 148, C784-C789, (2001).
25
A Dissertation Prospectus Presented By Avinash Kola to The Department of Chemical Engineering
In partial fulfillment of the requirements For the degree of Doctor of Philosophy In the field of Chemical Engineering Northeastern University Boston, Massachusetts
June 6th, 2014 1
ABSTRACT
Tungsten alloys, such as Ni-W are well recognized for their outstanding corrosion resistance, wear resistance and catalytic properties towards hydrogen generation. Nickeltungsten alloys also have the potential to act as a barrier layer in the semiconductor industry, to prevent diffusion of conducting metals (e.g.,Cu, Au) into the substrate. Tungsten, cannot be reduced alone, and requires the presence of certain inducing elements (e.g., Ni, Co, Fe) exhibiting induced codeposition to form alloys of W. This mechanism is not well understood. Electrodeposition of binary alloys of silver, such as tin-silver, silver-nickel and electroless deposition of silver-tungsten alloys are potential alternatives for lead-free materials in electronic packaging, printed circuit boards and other electronic components. However, the electrodeposition mechanism of Ag-Ni-W alloys has never been examined. Interest to electrodeposit all three elements, Ag-Ni-W comes from the motivation to tailor desired properties, including the increase of Ag hardness, corrosion resistance at high temperatures, while maintaining the favorable electrical properties of pure Ag. The examination and fundamental understanding of such a system will contribute towards developing a superior alloy with combined properties (Ni-W and Ag) The first report of the electrodeposition of a ternary Ag-Ni-W alloy is presented. The addition of Ag was found to lower the deposition rate of Ni and W. Chemical equilibria calculations were used to estimate the concentration of possible complexed species present in the electrolyte within a pH range 2 to 8. The most dominant species of Ag present were [AgTu4]+ and [AgTu3]+ irrespective of the pH, while NiHCit2 was the dominant at lower pH, while NiCit2 was negligible below pH 3 and was found to increase 2
with pH and was highest at pH 8. The concentration of Ni-Tu was found to be present only at low pH and above pH 4 was negligible. Only one W-cit species was found to present in the pH range 2-4 and was negligible over pH 4. Ag-Ni-W nanowire deposition was attempted in polycarbonate templates with 50 nm diameter and length of 6 um. The nanowires we released by dissolution of the membrane in dichloromethane. Transmission electron microscopy (TEM) showed that the Ag-Ni-W nanowires had smooth morphology, however they were non-uniform in length. The non-uniformity and weak strength of the nanowires is a strong function of the deposition conditions and the electrolytic hydrogen production, due to the low pH which is high in H+ ion concentrations. Understanding the reaction mechanism of the ternary alloys deposition and increased pH can help improve the current efficiency during deposition thereby resulting in alloys with different compositions suitable for a variety of technological applications.
3
1.0 INTRODUCTION
Tungsten alloys are well known for their outstanding properties, for example, NiW alloys are useful catalysts for electrolytic hydrogen generation, W imparts superior hardness and wear resistance to its alloys with Ni and Co, and W alloys have improved corrosion resistance compared to their codeposited counterparts [1-20]. The electrodeposition behavior of
W alloys was coined by Brenner as “induced
codeposition”, as W ions cannot be electrodeposited by itself, but can be fully reduced if codeposited with iron group elements, such as Ni, Co and Fe [21]. The recent advancement in multilevel interconnects technology is key for signal routing in ultra-large scale integrated (ULSI) circuits. Typical metallic conductors include an adhesion layer, a barrier layer, a conducting layer, a capping layer, and possibly an antireflective coating. Currently, aluminum and copper are used for on-chip interconnects. The integration of electroless technology in integrated circuit production was reviewed by Shacham-Diamand et al. [22]. Barrier layers of cobalt-tungstenphosphorus (Co-W-P) and nickel-tungsten-phosphorus (Ni-W-P) layers were deposited on silicon and silicon dioxide. The Co-W-P and Ni-W-P layers have higher hardness and melting point than similarly deposited Co-P and Ni-P [23]. Therefore, films with tungsten are expected to have better reliability and to act as better diffusion barriers for Cu interconnects when compared to similar films without tungsten. The excellent conductivity of silver and the relative simple procedure of its electroless deposition is a motivation to produce stable and corrosion-resistant silver films. Electroless silver plating baths are very unstable and thus short lived. Not at all surprising is the fact that the electrolyte stability and plating rate are markedly affected by 4
the pH [24]. Electrodeposition has the ability to better control the rate and achieve a higher rate than electroless deposition, and very stable electrolytes with cyanide have been the traditional norm. Over the years, many other electrolytes have been proposed, such as the ones involving nitrate, iodide, thiourea, thiocyanate, sulfamate, and thiosulfate. Thiourea is considered to be an effective chelating agent for Ag, and various groups have investigated the deposition of Ag with thiourea [25-29]. Silver being a noble metal, there exists a large difference in the reduction potential between Ni (-0.29 VSHE) and Ag (+0.799 VSHE), and a complexing agent is one way to achieve deposits with different alloy compositions, as it lowers the difference in reduction potentials. One advantage of thiourea, for deposition involving Ag and Ni from the same electrolyte, is the specific complexation of thiourea with Ag compared to Ni ions. The present work is devoted towards the replacement of cyanide electrolytes and to develop a single ternary stable electrolyte for the electrodeposition of Ag-Ni-W alloys that has good properties and a non-polluting nature. This proposal will address the electrodeposition parameters for Ag-Ni-W electrodeposition. As a first step in developing a single electrolyte for the ternary alloy, Ag-Ni-W alloys have been fabricated to probe the reaction mechanism and gain a better understanding for a variety of technological applications.
5
2.0 CRITICAL REVIEW
Ni-W alloys and Ag alloys with Ni or W can be fabricated by a variety of methods such as sputtering, electron beam evaporation and electrodeposition. The electrodeposition route can be cost effective compared to other techniques. In this chapter a review of the literature relevant to the electrodeposition of Ni-W and Ag alloys, its applications and the various advances towards tailoring its properties will be presented. The electrodeposition of tungsten from aqueous solutions has been attempted since the early 1930s. However, the electrodeposition of tungsten was found to be hindered by the formation of an oxide layer, which could not be further reduced further [30]. Holt et al. [31] showed that tungsten electrodeposition could be achieved in the presence of certain metals (Co, Ni , Fe) with tungsten content as high as 50 wt %. This phenomenon is known as “induced codeposition” and was first coined by Brenner in 1963 [21]. 2.2 Electrodeposition of Ternary Ag-Ni-W Alloys 2.2.2
Interest in Ag-Ni Alloys Ag-Ni alloys also exhibit remarkable catalytic properties [32], and are promising
candidates for electrical contacts and switches [33, 34]. However, the Ag- Ni deposition system remains one of the least examined systems to date via electrodeposition synthesis [28, 35-37]. According to the phase diagram [38], Ag and Ni are immiscible in bulk form and even intermetallic phases were not observed at high temperatures. However, the alloying effect at the nano-scale level is quite different from bulk, since the heat of formation reduces with decreasing particle size and hence alloying these metals becomes a possibility. Another aspect of the Ag-Ni system is the large difference in reduction
6
potential Ag(+0.799 VSHE) and Ni (-0.25 VSHE) [28], an important feature of thiourea especially while considering Ag-Ni deposition from the same electrolyte is that, while thiourea binds strongly to Ag+ ions, it form a weak complex with Ni2+ ions. This selective complexation becomes crucial to lower the large difference in redox potentials between. Liang et al. [28] investigated the formation of metastable Ag-Ni solid solution and their phase separation to elemental form upon thermal annealing. Excessive thiourea (0.2 M) was used to stabilize the electrolyte, and act as a complexing agent to selectively complex Ag (10 mM) and lower the reduction potential between Ag and Ni. The addition of 0.2 M thiourea polarized the reduction potential of Ag from 0.03 V MSE to -0.650 V, which indicates a strong complexing effect between thiourea and Ag, while the addition of 0.2 M thiourea shifted the potential of Ni (0.15 M) from -1.04 V to -1.08 V due to a weak complexing effect. Eom et al. [35], examined the deposition of the alloy in sodium citrate electrolytes. A shift in the Ag ion deposition potential to more negative values was observed in the presence of citrate. They identified pH to be a dominant factor to control the composition of the samples, as it affected the complexing of the metal ions with citrate. At low current densities (0.5 mA/cm2), dendrite formation was observed with a film composition of Ag60Ni40, and the microstructure transformed to a granular deposit with increasing current density, which correspondingly increased the Ni content. The dendrite formation was attributed to the low concentration of Ag ions in the electrolyte. 2.2.3
Interest in Ag-W Alloys
The application of silver for ultra large scale integration is promising due to its low bulk resistivity (1.59 µΩcm), relatively high melting point and higher electromigration 7
resistance compared to Cu, which is widely used in the down-scaling of interconnects. Suitable conductivities were reported for Ag fabricated by sputtering [39, 40] and electroplated thin films [41], while some drawbacks of Ag, such as corrosion in air and diffusion in SiO2 , could be avoided by using a suitable binary alloy such as Ag-W instead of pure Ag [42]. Electroless deposition was intensively adapted to examine Ag-W thin films [23, 42-50] . Shacham-Diamand et al. [23], examined and compared the effect of tungsten concentration in the electrolyte, microstructure and morphology of Ag and Ag -W thin films, and related their influence on the electrical properties. Thin films containing tungsten with improved reliability and as effective diffusion barriers was first demonstrated in Co-W-P and Ni-W-P thin films, which exhibited higher hardness and melting points than films deposited without tungsten. Inberg et al. [48-50] examined the Ag-W electroless deposition system extensively. They found that increasing the tungsten concentration in the electrolyte decreased the deposition rate of Ag, resulting in smooth and high quality Ag-W films. A maximum of 3.2 at % W was achieved when the [WO42]/ [Ag+] molar ratio was unity. Higher concentrations of tungsten in the electrolyte did not increase the tungsten at % in the deposit. Glickman et al. [45] examined the factors contributing to the resistivity of Ag-W films and was able to achieve a considerable decrease in the resistivity for 100 and 50 nm Ag-W films. They achieved a resistivity of 4.5 µohm-cm and 6.5 µohm-cm for 100 nm and 50 nm films respectively for very low W at % (0.6 and 0.9). They reported grain boundary scattering to be a dominant factor in controlling the electrical resistivity of sub100 nm films. Using post vacuum annealing, at low temperatures ~ 150 ºC for 1 hr, a 8
considerable drop in resistivity could be achieved, while 75-80 % of the resistivity drop was attained at annealing temperatures of 200-250 ºC for 1 hr. In spite of the recent efforts in exploring Ag-W alloys as potential candidates for electronic applications, information on the mechanism of Ag-W electrodeposition is very scarce in the literature. A detailed understanding of the mechanism and the effect of one element on the other is crucial.
9
3.0 EXPERIMENTAL A thiourea-citrate electrolyte was used for the electrodeposition of Ag-Ni-W alloys. The electrolyte conditions and procedure, deposit characterization, and cell designs are presented in this section. 3.1 Electrolyte, Conditions and Procedure The composition of the electrolyte for the deposition of Ag-Ni-W alloys is listed in Table 1. All electrolytes were prepared using de-ionized ultra-filtered (D.U.I.F) water from Fischer Scientific. The pH of the electrolytes was measured maintained at a value of 2. Copper plates from ESPI metals were used as working electrodes and a rectangle piece of Pt was used as the anode. All substrates were cleaned in dilute H2SO4 (10 vol %) in order to remove any copper oxides and cleaned in D.I.U.F water before deposition. The deposition was galvanostatically controlled using a Solartron potentiostat/function generator model 1287A. For partial current density measurements, the current density applied were 8 mA/cm2, 20 mA/cm2, 40 mA/cm2 and 80 mA/cm2 .The area in this case was 0.39 cm2. The Ag-Ni-W nanowires were electrodeposited, under the same deposition conditions mentioned above, in a polycarbonate membrane having the smallest region of the pore 50 nm in diameter and 6 µm in length. The nanowires were then dissolved in dichloromethane to remove the supporting template and release the nanowires. The dissolved nanowires were then subjected to centrifuge, and then dichloromethane was replaced with fresh solution. This procedure was repeated 3 times. The deposit thickness and composition was analyzed with a KEVEX Omicron energy dispersive X-ray fluorescence analyzer (XRF), at 40 keV, 2 mA in air with an acquisition time of 60 sec. 10
SEM analysis was done on a Hitachi S4800 at 3.0 KV and 4.5 K and 15.0 K magnification. TEM images were taken on a JEOL, JEM 1010 at 80 KV. Table 1. Electrolyte composition for Ag-Ni-W alloys. Chemical
Concentration (M)
Nickel Sulfate(M)
0.05
Sodium Tungstate (M)
0.015
Silver Sulfate
0.005
Sodium Citrate (M)
0.285
Thiourea (M)
0.650
Agitation of the electrolyte in the Hull cell is done by using air, bubbling close to the surface of the cathode to ensure uniform mixing. A flow meter is used to monitor the entering feed rate of air into the Hull cell. Five different flow rates, 1-5 L/min, were examined. For the conventional Hull cell experiments, two agitation conditions were examined, no agitation and 5 L/min. The primary purpose of any form of agitation to the electrolyte is to eliminate or minimize concentration gradients near the electrode surface to avoid mass transport limitations. In the case of Ag-Ni-W electrodeposition there is an order of magnitude lower amount of Ag ions so that mass transport effects may be considerable in that case. Air agitation can help then to control the boundary layer thickness during deposition, which is a crucial factor when depositing alloys involving diffusion limited species.
11
4.0 RESULTS AND DISCUSSION
4.5 Electrodeposition of Ni-W with/without Ag To examine the effect of Ag on the codeposition mechanism of Ni-W, partial current density measurements were done on a Cu substrate in a parallel configuration. Due to the instability of the electrolyte at pH 8, it was not examined further for this electrolyte concentration and deposition parameters. Polarization curves of the Ni-W electrolytes with and without Ag at an agitation rate of 5 L/min are shown in Figure 1. The deposition potential of Ni-W begins at -0.77 V and rises rapidly with increasing potential. The addition of Ag shows a clear variation in the deposition characteristics, with a start in deposition at -0.55 V, which is an indication of Ag deposition, reaches it’s limiting current density and then rises in current density around -0.88 V. We can also observe a slight shift in the Ni-W reaction. 20
NiW only
2
i (mA/cm )
15
10
5
NiW+ 5mMAg
+
0 0.4
0.5
0.6
0.7
0.8
0.9
1.0
E Vs Ag/AgCl
Figure 1. Polarization curves for Ag-Ni-W electrolyte with and without Ag (I). 12
Galvanostatic deposition was done on a Cu substrate in a parallel configuration for both the electrolytes at applied current densities of 8 mA/cm2, 20 mA/cm2, 40 mA/cm2 and 80 mA/cm2. Upon addition of Ag (I) into the electrolyte, as expected, Ag rich deposits are obtained at current densities, 8 mA/cm2 and 20 mA/cm2. As the current density increases, Ni composition increases in the deposit, due to the onset of Ni deposition at higher current densities, which is also observed from the polarization curves. However, the tungsten composition remains constant irrespective of the applied current density. With the addition of Ag in the electrolyte, nodules formation can be seen on the substrate and is uniform across the surface of the deposit (Figure 2 (a,b)), these nodules tend to get bigger with an increase in current density (Figure 2 (c,d)). Table 2 shows the composition variation with Ag (I) ions in the electrolyte. At 80 mA/cm2, the Ag composition drops drastically, indicating the onset of kinetic reduction of Ni.
Table 2. Composition of Ni, W and Ag. Current density (mA/cm2) 8 20 40 80
Ni wt %
W wt %
0.7 4.1 7.3 35.7
0.2 0.3 1.0 0.5
13
Ag wt % 99.0 95.4 91.6 63.7
(a)
(b)
(c)
(d)
Figure 2. Optical images of Ag-Ni-W deposits at 8 mA/cm2, 20 mA/cm2, 40 mA/cm2 and 80 mA/cm2: (a), (b), (c), (d) respectively.
Along with the composition data we are able to calculate the partial current densities of each species A semi-log plot (i vs EAgCl) is used because kinetic rates are typically exponential with potential. In Figures 3 (a and b), the reaction rate of Ni and W drop in the presence of Ag (I), however, also the deposition potential has been shifted to more positive values. No deposit can be seen at the low current density end in the absence of Ag (I), by physical observation of the samples. This thermodynamic shift in potential indicates that even though the reaction rate slows down, the presence of Ag (I) could in turn induce the deposition of Ni and W due to a more energetically favorable 14
condition, such as a change in the activity of the solid state. In the case of Ni deposition we observe a rise in Ni deposition rate at lower potential of -1.0 V, than observed in the absence of Ag (I), whereas in the case of W we see a very flat profile in the presence of Ag (I).
14 12
log iNi (mA/cm2)
10 8 6 4 2 0 0.4
0.6
0.8
1.0 1.2 1.4 E vs. Ag/AgCl (V)
0.6
0.8
1.0 1.2 1.4 E vs. Ag/AgCl (V)
1.6
1.8
2.0
0.5
log iW (mA/cm2)
0.4 0.3 0.2 0.1 0.0
0.4
1.6
1.8
2.0
Figure 3. Partial current density, with and without Ag (I) for (a) Ni and (b) W.
15
Figure 4 shows the side reaction partial current density. In the presence of Ag the side reaction occurs at lower potentials -0.6 V, and consumes most of the current, thereby indicating a low efficiency of the deposition process. The partial current density increases drastically compared to the Ag (I). This increase in partial current density could hinder the deposition of less noble Ni and WO42- ions.
80
2
log ij (mA/cm )
60
40
20
0 0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
E (V)
Figure 4. Partial current density of side reactions, with and without Ag (I).
4.3 Electrolyte Stability and Complex Species Distribution The stability of the electrolyte at pH 2 was much higher compared to the electrolyte at pH 8. However, the electrolyte at pH 2 would decompose over a period of time (~ 6 hrs). The electrolyte at pH 8 would precipitate even during deposition, even though the average current density was low 1.7 mA/cm2. This instability could be related to different species forming at different pH. Equilibria calculations, help in determining
16
the species distribution in an electrolyte, depending on their complex stability constants. Table 3 shows the mass balance and equilibria calculations used to calculate the distribution of the complexing species. Table 3. Mass Balance and Equilibria Equations. Species
Stability constant Log K
Equations
Mass Balance Eq
𝑪𝑵𝒊
𝑪𝑪𝒊𝒕
𝑪𝑵𝒊 + 𝑪𝑵𝒊𝑪𝒊𝒕 + 𝑪𝑵𝒊𝑯𝑪𝒊𝒕 + 𝑪𝑵𝒊𝑪𝒊𝒕𝟐 + 𝑪𝑵𝒊𝑯𝑪𝒊𝒕𝟐 + 𝑪𝑵𝒊𝑻𝒖 = 𝟎. 𝟎𝟓
Mass Balance Eq
𝑪𝑾𝑶𝟒
𝑪𝑾𝑶𝟒 + 𝑪𝑾𝑶𝟒𝑯𝑪𝒊𝒕𝑯 + 𝑪𝑾𝑶𝟒 𝑪𝒊𝒕𝑯𝟐 + 𝑪𝑾𝑶𝟒 𝑪𝒊𝒕𝑯𝟑 − 𝟎. 𝟎𝟏𝟓 = 𝟎
Mass Balance Eq
Mass Balance Eq Mass Balance Eq Mass Balance Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq Equilibria Eq
𝑪𝑪𝒊𝒕 + 𝑪𝑵𝒊𝑪𝒊𝒕 + 𝑪𝑵𝒊𝑯𝑪𝒊𝒕 + 𝑪𝑵𝒊𝑪𝒊𝒕𝟐 + 𝑪𝑵𝒊𝑯𝑪𝒊𝒕𝟐 + 𝑪𝑾𝑶𝟒𝑯𝑪𝒊𝒕𝑯 + 𝑪𝑾𝑶𝟒 𝑯𝑪𝒊𝒕𝑯𝟐 + 𝑪𝑾𝑶𝟒 𝑯𝑪𝒊𝒕𝑯𝟑 − 𝟎. 𝟐𝟖𝟓 = 𝟎
𝑪𝑻𝒖
𝑪𝑻𝒖 + 𝑪𝑵𝒊𝑻𝒖 − 𝟎. 𝟔𝟓𝟎 = 𝟎
𝑪𝑨𝒈 𝑪𝑯
𝑪𝑨𝒈 + 𝑪𝑨𝒈𝑻𝒖 + 𝑪𝑨𝒈𝑻𝒖𝟐 + 𝑪𝑨𝒈𝑻𝒖𝟑 + 𝑪𝑨𝒈𝑻𝒖𝟒 − 𝟎. 𝟎𝟎𝟓 = 𝟎
𝑪𝑵𝒊𝑪𝒊𝒕
5.35
𝑪𝑵𝒊𝑪𝒊𝒕𝟐
8.11
𝑪𝑵𝒊𝑻𝒖
1.47
𝑪𝑾𝑶𝟒 𝑯𝑪𝒊𝒕𝑯𝟐
17.03
𝑪𝑨𝒈𝑻𝒖
7.11
𝑪𝑵𝒊𝑯𝑪𝒊𝒕
9.13
𝑪𝑵𝒊𝑯𝑪𝒊𝒕𝟐
13.5
𝑪𝑾𝑶𝟒𝑯𝑪𝒊𝒕𝑯
10.2
𝑪𝑾𝑶𝟒 𝑯𝑪𝒊𝒕𝑯𝟑
21.67
𝑪𝑨𝒈𝑻𝒖𝟐
10.61
𝑪𝑨𝒈𝑻𝒖𝟒
13.57
𝑪𝑨𝒈𝑻𝒖𝟑
𝑪𝑯 − 𝟏𝟎𝒑𝑯 = 𝟎
𝟏𝟎𝟓.𝟑𝟓 × (𝑪𝑵𝒊 × 𝑪𝑪𝒊𝒕 ) − 𝑪𝑵𝒊𝑪𝒊𝒕 = 𝟎
𝟏𝟎𝟗.𝟏𝟑 × (𝑪𝑵𝒊 × 𝑪𝑪𝒊𝒕 × 𝑪𝑯 ) − 𝑪𝑵𝒊𝑯𝑪𝒊𝒕 = 𝟎 𝟏𝟎𝟖.𝟏𝟏 × �𝑪𝑵𝒊 × 𝑪𝑪𝒊𝒕 𝟐 � − 𝑪𝑵𝒊𝑪𝒊𝒕𝟐 = 𝟎
𝟏𝟎𝟏𝟑.𝟓 × �𝑪𝑵𝒊 × 𝑪𝑪𝒊𝒕 𝟐 × 𝑪𝑯 � − 𝑪𝑵𝒊𝑯𝑪𝒊𝒕𝟐 = 𝟎 𝟏𝟎𝟏.𝟒𝟕 × (𝑪𝑵𝒊 × 𝑪𝑻𝒖 ) − 𝑪𝑵𝒊𝑻𝒖 𝟎
𝟏𝟎𝟏𝟎.𝟐 × �𝑪𝑾𝑶𝟒 × 𝑪𝑪𝒊𝒕 × 𝑪𝑯 𝟐 � − 𝑪𝑾𝑶𝟒𝑯𝑪𝒊𝒕𝑯 = 𝟎
𝟏𝟎𝟏𝟕.𝟎𝟓 × �𝑪𝑾𝑶𝟒 × 𝑪𝑪𝒊𝒕 × 𝑪𝑯 𝟑 � − 𝑪𝑾𝑶𝟒 𝑪𝒊𝒕𝑯𝟐 = 𝟎 𝟏𝟎𝟐𝟏.𝟔𝟕 × �𝑪𝑾𝑶𝟒 × 𝑪𝑪𝒊𝒕 × 𝑪𝑯 𝟒 � − 𝑪𝑾𝑶𝟒 𝑪𝒊𝒕𝑯𝟑 = 𝟎 𝟏𝟎𝟕.𝟏𝟏 × �𝑪𝑨𝒈 × 𝑪𝑻𝒖 � − 𝑪𝑨𝒈𝑻𝒖 = 𝟎
𝟏𝟎𝟏𝟎.𝟔𝟏 × �𝑪𝑨𝒈 × 𝑪𝑻𝒖 � − 𝑪𝑨𝒈𝑻𝒖𝟐 = 𝟎 𝟏𝟎𝟏𝟐.𝟕𝟑 × �𝑪𝑨𝒈 × 𝑪𝑻𝒖 � − 𝑪𝑨𝒈𝑻𝒖𝟑 = 𝟎
12.73
𝟏𝟎𝟏𝟑.𝟓𝟕 × �𝑪𝑨𝒈 × 𝑪𝑻𝒖 � − 𝑪𝑨𝒈𝑻𝒖𝟒 = 𝟎
17
A plot of the distribution of the species in an electrolyte mentioned in table 1 is shown in Figure 5. The entire Ag ions complex with thiourea to form [AgTu4]
+
and
remains constant irrespective of pH. The NiHCit2 complex (cit –citrate) is dominant at pH 2 and reduces in concentration with increasing pH. An inverted trend is observed for NiCit2 species which increases with pH and is dominant at pH 8. Different tungstate citrate species are present at different pH. The WO4HCit2 is dominant at pH 2 and reduces after pH 4. The WO4HCitH species increases from pH 3, reaches a peak at pH 5.5 and drops in concentration with increasing pH. The WO4CitH species increases from pH 6 and is highest at pH 8. The inset of Figure 5 also shows that Ni-Tu complex is very insignificant, indicating preferential complexation of thiourea with Ag ions. The stability issue at pH 8 could be due to the different complexed species forming at a higher pH which are unstable during deposition.
0.05 NiCit2 NiCit2 0.04
NiHCit2 NiHCit Concentration (M)
Concentration (M)
2
WO4HCitH2 [1,1,2] Wo4HCitH2
0.03
WO4HCitH3 [1,1,3] Wo4HCitH3 2-
WO4 Wo4
AgTU AgTU4
0.02
3E-08 2.5E-08 2E-08 1.5E-08 1E-08 5E-09 0
N…
1 2 3 4 5 6 7 8 pH
0.01
0 1
2
3
4
5
pH
6
7
8
9
Figure 5. Species distribution of different complexed species in an Ag-Ni-W electrolyte, inset shows the Ni-Tu complex species. 18
4.6 Electrodeposition of Nanowires The development of nanowires into porous templates is a challenge in systems where there is a substantial side reaction of gas evolution. If the side reaction is too voluminous then gas bubbles can block the template pores and prevent deposition. Guided by the conditions of the thin films fabricated in preceding sections a current density was selected to deposit Ag-Ni-W nanowires, into the polycarbonate membranes. Figure 6 shows the resulting Ag-Ni-W nanowires deposited under the same current density of 1.7 mA/cm2 and then released from the template. From the thin film results it is expected to have a composition that is Ag rich. What is notable is that the nanowires have different lengths. Thus, they easily break when released from the membrane. The longest length achievable after a deposition time of 1800 s was 5.0 microns long, which provides an estimate of the potential deposition rate (~0.0027 microns/s). Nanowires that are more robust are desired.
Figure 6. TEM of Ag-Ni-W alloy nanowires deposited at 1.7mA/cm2 for 1800 s into a PC membrane. 19
6.0 PROPOSAL The electrodeposition of ternary alloy Ag-Ni-W system has a variety of technological applications especially in the semiconductor industry. For these alloys to be viable for large scale industrial production, few key limitations need to be addressed. As mentioned earlier one goal of this project is to develop an environmentally friendly and stable electrolyte for silver deposition. Thus far we have achieved this goal and deposited Ag-rich Ni-W alloys from a stable, non-cyanide, pH 2 electrolyte. However, one drawback of the low pH is the high concentration of H+ ions in the electrolyte which increases the side reactions and lowers current efficiency. In order to address the issue of low current efficiency and to deposit such alloys in deep recessed substrates, for example, nanowires, a better understanding of the electrodeposition mechanism of Ag-Ni-W alloys is necessary. Analysis from the conventional Hull cell, in which agitation by air bubbling is employed, does give us some information on the effect of Ag (I) addition to Ni and W deposition rates. But, since Ag (I) is a noble metal species and is mostly under diffusion limited control, better control over the boundary layer thickness is crucial to overcome certain limitations such as, poor boundary layer control, the inconsistency due to the variation in bubble flow pattern and operable limit (maximum flow rate 5 L/min).The rotating cylinder setup gives a well-defined control over the hydrodynamic conditions within the electrolyte system. For the next set of experiments, we propose to use the RCE setup to examine the effect of Ag-Ni-W alloy deposition under different mixing conditions, and use the complexation model to identify the optimum pH for the deposition of Ag-Ni-W alloys. The effect of concentration of metal ions and different additives in the electrolyte on the current efficiency will also be examined. 20
The following tasks will be focused on for future work: AIM 1: Improve the overall current efficiency for practical applications. Although, a pH 2 electrolyte was found to be stable for the electrodeposition of Ag-Ni-W alloys over the examined concentration range, low pH electrolytes lead to very low current efficiencies due to the high concentration of H+ ions in the electrolyte leading to the following reaction below 2𝐻 + + 2𝑒 − → 𝐻2
The hydrogen evolution reaction due to high H+ ion concentration can be dealt with by increasing the pH. A look at the complexation model Figure (5), we can examine the pH range we can focus on, since we have shown that pH 8 is not stable, and leads to precipitation during electrodeposition. We notice different species present at pH 2 and pH 8, at pH 2 we know that WO4HCitH3 and NiHCit2 are stable while at pH 8 either NiCit2 or WO42- leads to instability. For this task electrodeposition of the Ag-Ni-W alloy at three different pH (3, 4, 5) will be done and this will contribute towards answering this question of which species causes the instability – NiCit2 or Wo42. A constant rotation rate of 2825 rpm will be used at an applied current density of 1.7 mA/cm2 for 30 min. Samples will be weighed before and after deposition, XRF analysis will be done to measure the
composition of the alloy, and calculate current efficiencies. Another
approach towards improving current efficiency is to increase the overall concentration of the metal ions in the electrolyte. For this aim, concentration of each species will be double to 0.1 M nickel sulfate, 0.01 M silver sulfate, 0.03 M sodium tungstate, 0.570 M sodium citrate and 1.3 M thiourea. Three experiments at different current densities 1.7mA/cm2, 5 mA/cm2 and 10 mA/cm2 will be applied at a constant rotation rate of 2825 21
rpm. Samples will be weighed before and after deposition and XRF analysis will be used to analyze the composition of the deposited alloy and calculate current efficiency. AIM 2: Identify the deposition behavior of Ag induced Ni, W reduction. Analysis from the partial current density experiments on the Hull cell indicated an induced effect of Ag ions on Ni and W deposition rate. This is known as under potential deposition (UPD) behavior of Ni and W species in the presence of Ag ions. The goal in this aim is to establish if the reduction behavior between Ag ions is coupled with Ni and W, i.e. if, the rate of Ni or W will increase with Ag deposition rate. Since Ni and W are under kinetic control in the applied current density range examined, rotation rate should not affect their deposition rate. If an increase in the deposition rate of Ni or W is observed, along with an increase in Ag deposition rate, then this is purely due to a coupled effect between Ag and Ni/W. For this task six current density experiments, 0.5 mA/cm2, 5 mA/cm2, 10 mA/cm2, 20 mA/cm2, 40 mA/cm2 and 80 mA/cm2 for 30 mins each and at three different rotation rates 706, 1412 and 2824 rpm will be done. XRF will be used to analyze the composition of the deposited alloys. AIM 3: Effect of different additives on the deposition of Ag-Ni-W alloys. The effect of different concentrations of additives, such as thiourea, boric acid, sodium gluconate and citrate on the deposition of Ag-Ni-W alloys will be examined. One approach to examine a large set of variables effectively is by using Factorial design experiments. A 2k design for k =3 factors/variables will be considered. This design has 8 experiments, which enables us to examine the effect of individual variables (each at two levels), and in addition three binary and one ternary interaction on one single parameter, for example in our case wt % of W. A comparison of the values of each effect gives us an 22
idea on what parameters play a crucial role in obtaining a desired outcome. Thiourea (0, 0.650 M), boric Acid (0, 0.5M), sodium gluconate (0, 0.5 M) and citrate (0, 285 M) will be considered for this study, using the optimum electrolyte for Ag-Ni-W after completion of aim 1 and 2. Each sample will be weighed before and after deposition for calculating the current efficiency, XRF will be used to analyze the composition of the alloy and SEM will be done to examine the surface morphology of the deposits. AIM 4: Investigate the growth mechanism of Ag-Ni-W nanowires. Template synthesized nanowires have received a great deal of attention over the past decade because they show great promise in a wide range of applications such as electronics, sensing, drug delivery and fabrication of solar cells. The electrodeposition of 1D nanostructures such as, Ag nanowires are attractive for their superior electrical and thermal conductivity. The electrodeposition of Ag-Ni-W ternary alloy nanowires is a novel aspect to this work. From preliminary experiments we were able to obtain Ag-Ni-W nanowires under the deposition conditions examined, however due to the low pH the hydrogen side reaction causes the nanowires to become brittle and break during release. Results from aim 1-3 will help towards solving the issue of low current efficiency and develop more robust Ag-Ni-W nanowires.
23
8.0 REFERENCES 1. Zayats, A. I., and Perekhrest, N. A., "Electrodeposition of Iron-Tungsten Alloys from Pyrophosphate Electrolytes.", J. Appl. Chem. USSR, 44, 1286-1291, (1971). 2. Chassaing, E., Quang, V., and Qiart, R., "Mechanism of Nickel-molybdenum alloy electrodeposition in citrate electrolytes.", J. Appl. Electrochem., 19, 839, (1989). 3. Akiyama, T., and Fukushima, H., "Recent Study on the Mechanism of the Electrodeposition of Iron Group Metal Alloys.", ISIIJ Int., 32, 787-789, (1992). 4. Stepanova, L. I., and Purovskaya, O. G, "Electrodeposition of Nickel-Based alloys with tungsten and molybdenum.", Metal. Finish., 96, 50,52-53, (1998). 5. Svensson, M., Wahlstrom, U., and Holmbom, G., "Compositiolnally Modulated Cobalt-tungsten Alloys Deposited from a Single Ammonical electrolyte.", Surf. Coat. Technol., 105, 218-223, (1998). 6. Yamasaki, T., "High-Strength Nanocrystalline Ni-Alloys Produced by Electrodeposition.", Mater. Phys. Mech., 1, 98-102, (2000). 7. Cesiulis, H., Donten, M., Donten, M. L., and Stojek, Z., "Electrodeposition of Ni-W, Ni-Mo and Ni-W-Mo Alloys from Pyrophosphate Baths.", Mater. Sci.(Medziagotyra), 7, 237-241, (2001). 8. Itoh, K., Wang, F., and Watanabe, T., "Relationship between the Crystallographic Structure of Electrodeposited Ni-W Alloy Film and Its Thermal Equilibrium Diagram ", J. Japan Inst. Metals,, 65, 1023-1028, (2001). 9. Wang, Y., Chen, M., Zhou, F., and Ma, E., "High tensile ductility in a nanostructured metal.", Nature, 419, 912-915, (2002). 10. Younes, O., and Gileadi, E., "Electroplating of Ni/W Alloys I. Ammoniacal Citrate Baths.", J. Electrochem. Soc., 149, C100-C111, (2002). 11. Jiles, D. C., "Recent advances and future directions in magnetic materials.", Acta Materialia, 51, (19), 59075939, (2003). 12. Karpralova, I. G., Perelygin,Y. P., and Semchenko, T. K. , "Electrodeposition of Nickel-tungsten Alloy from Acetate Electrolyte.", J. Appl. Chem. USSR, 76, 1524-1526, (2003). 13. Liu, R., Wang, H., Yao, J. Y., Li, X. P., and Ding, G. F., "Preparing Ni-W alloy films with low internal stress and high hardness by heat treating.", Surf. Rev. Lett., 14, 1107-1112, (2007). 14. Sriraman, K. R., Ganesh Sundara Raman, S. and Seshadri, S. K., "Corrosion Behaviour of Electrodeposited Nanocrystalline Ni-W and Ni-Fe-W Alloys.", Mater. Sci. Eng. A, A460-461, 39-45, (2007). 15. Wang, H., Liu, R., Cheng, F. J., Cao, Y., Ding, G. F., and Zhao, X. L., "Electrodepositing amorphous Ni-W alloys for MEMS.", Microelectron. Eng., 87, 1901-1906, (2010). 16. Weston, D. P., Harris, S. J, Capel, H., Ahmed, N., Shipway, P. H, and Yellup, J. M, , "Nanostructrured Co-W Coatings Produiced by Electrodeposition to Replace Hard Cr on Aerospace Components.", Trans. Inst. Met. Finish., 88, 47-56, (2010). 17. Chianpairot, A., Lothongkum, G., Schuh, C. A., and Boonyongmaneerat, Y., "Corrosion of nanocrystalline Ni-W alloys in alkaline and acidic 3.5wt.% NaCl solutions.", Corr. Sci., 53, 1066-1071, (2011). 18. Wei, G. Y., Lou, J. W., Ge, H. L., Yu, Y. D., Jiang, L., and Sun, L. X.,, "Co-W Films Prepared from Electrolplating Baths with Different Complexing Agents ", Surf. Eng., 28, 412-417, (2012). 19. Sun, S., Bairachna, T., and Podlaha, E. J., "Induced Codeposition Behavior of Electrodeposited NiMoW Alloys.", J. Electrochem. Soc., 160, (10), D434- D440, (2013). 20. Saraç, H., Patrick, M. A., Wragg, A. A., "Physical properties of the ternary electrolyte potassium ferriferrocyanide in aqueous sodium hydroxide solution in the range 10–90°C.", J.Appl. Electrochem., 23, (1), 5155, (1993). 21. Brenner, A., Electrodeposition of Alloys: Principles and Practice, Academic Press, New York, (1963). 22. Shacham-Diamand, Y., Dubin,V., and Angyal,M., "Electroless copper deposition for ULSI.", Thin Solid Films, 262, 93-103, (1995). 23. Shacham-Diamand, Y., Inberg, A., Sverdlov, Y., and Croitoru, N., "Electroless silver and silver with tungsten thin films for microelectronics and microelectromechanical system applications.", J. Electrochem. Soc., 147, 3345-3349, (2000). 24. Schlesinger, M. (2010). Electroless and Electrodeposition of Silver, in Modern Electroplating, John Wiley & Sons, Inc.
24
25. Azzaroni, O., Schilardi, P. L., Salvarezza, R. C., and Arvia, A. J., "Smoothening mechanism of thiourea on silver electrodeposition. Real time imaging of the growth front evolution.", Langmuir, 15, 1508-1514, (1999). 26. de Oliveira, G. M., and Carlos, I. A., "Silver-zinc electrodeposition from a thiourea solution with added EDTA or HEDTA.", Electrochim. Acta, 54, 2155-2163, (2009). 27. Gómez, E., García-Torres, J., and Vallés, E., "Study and preparation of silver electrodeposits at negative potentials.", J. Electroanal. Chem., 594, 89-95, (2006). 28. Liang, D., Liu, Z., Hilty, R. D., and Zangari, G., "Electrodeposition of Ag-Ni films from thiourea complexing solutions.", Electrochim. Acta., 82, 82-89, (2012). 29. Reents, B., Plieth, W., Macagno, V.A., and Lacconi, G.I, "Influence of thiourea on silver deposition: Spectroscopic investigation.", J Electroanal Chem, 453, 121-127, (1998). 30. Lassner, E., and. Schubert, W. D, "Tungsten-Properties Chemistry,Technology of the Element,Alloys and Chemical Compounds.", (1999). 31. Holt, M. L., and Kahlenberg, L., "The Deposition of Tungsten from Aqueous Alkaline Solutions.", Quart. Rev. Am. Electroplaters' Soc., 9, 41-52, (1933). 32. Guo, H., Chen, Y., Chen, X., Wen, R., Yue, G. H., and Peng D. L., "Facile synthesis of near-monodisperse Ag@Ni core-shell nanoparticles and their application for catalytic generation of hydrogen.", Nanotechnology, 22, (2011). 33. Tyler, E. H., Clinton, J. R., and Luo, H. L., "Electrical Resistivity of Metastable Ag-Ni Alloys.", Solid State Commun., 13, 1409-1411, (1973). 34. Swingler, J., "Performance and arcing characteristics of Ag/Ni contact materials under DC resistive load conditions.", IET Sci. Meas. Technol., 5, 37-45, (2011). 35. Eom, H., Jeon, B., and Kim, D., "Electrodeposition of Silver-nickel Thin Films with a Galvanostatic Method.", Mater. Trans., 51, 1842-1846, (2010). 36. Bdikin, I. K., Strukova, G. K., Strukov, G. V., Kedrov, V. V., Matveev, D. V., Zver'kov, S. A., and Kholkin, A. L., "Growth, crystal structure and stability of Ag-Ni/Cu films.", Mater. Sci. Forum, 514-516, 1166-1170, (2006). 37. Santhi, K., Karthick, S. N., Kim, H. J., Nidhin, M., Narayanan, V., and Stephen, A., "Microstructure analysis of the ferromagnetic Ag-Ni system synthesized by pulsed electrodeposition.", Appl. Surf. Sci., 258, 3126-3132, (2012). 38. Singleton, M., and Nash, P., Bull. Alloy Phase Diagrams, American Society for Metals, 8, 119-121, (1987). 39. Wang, Y., and Alford, T. L., "Formation of aluminum oxynitride diffusion barriers for Ag metallization.", Appl. Phys. Lett., 74, 52-54, (1999). 40. Hauder, M., Hansch, W., Gstöttner, J., and Schmitt-Landsiedel, D., "Electromigration resistance of sputtered silver lines using different patterning techniques.", Microelectron. Eng., 60, 51-57, (2002). 41. Baker, B. C., Freeman, M., Melnick, B., Wheeler, D., Josell, D., and Moffat, T.P., "Superconformal electrodeposition of silver from a KAg(CN)2-KCN-KSeCN electrolyte.", J. Electrochem. Soc., 150, C61-66, (2003). 42. Inberg, A., Bogush, V. Croitoru, N., and Shacham-Diamand, Y., "Electrochemical study of the mechanism of Ag(W) electroless deposition.", J. Electrochem. Soc., 154, D1-4, (2007). 43. Bogush, V., Inberg, A., Croitoru, N., Dubin, V., and Shacham-Diamond, Y., "Electroless deposition of novel Ag-W thin films.", Microelectron. Eng., 70, (489-494), (2003). 44. Bogush, V., Inberg, A., Croitoru, N., Dubin, V., and Shacham-Diamand, Y., "Material properties of very thin electroless silver-tungsten films.", Thin Solid Films, 426, 288-295, (2003). 45. Glickman, E. E., Bogush, V., Inberg, A., Shacham-Diamand, Y., and Croitoru, N., "Electrical resistivity of thin electroless Ag–W films for metallization.", Microelectron. Eng., 70, 495-500, (2003). 46. Glickman, E., Inberg, A., Bogush, V., Aviram, G., Croitoru, N., and Shacham-Diamand, Y., "On the mechanism of annealing effect in electrical resistivity of sub-100 nm Ag (1% W) films.", Microelectron. Eng., 76, 182189, (2004). 47. Glickman, E., Inberg, A., Bogush, V., Aviram, G., Popovitz, R., Croitoru, N., and Shacham-Diamand, Y., "Role of local microchemistry and surface structure in electrical resistivity of 50nm electroless films Ag-Woxygen.", Microelectron. Eng., 82, 307-313, (2005). 48. Inberg, A., Shacham-Diamand, Y., Rabinovich, E., Golan, G., and Croitoru, N., "Electroless-deposited Ag–W films for microelectronics applications.", Thin Solid Films, 389, 213-218, (2001). 49. Inberg, A., Shacham-Diamand, Y., Rabinovich, E., Golan, G., and Croitoru, N., "Material and electrical properties of electroless Ag-W thin film.", J. Electron. Mater., 30, 355-359, (2001). 50. Inberg, A., Zhu, L., Hirschberg,G., Gladkikh, A., Croitoru, N., Shacham-Diamand,Y., and Gileadi, E., "Characterization of the Initial Growth Stages of Electroless Ag(W) Films Deposited on Si(100).", J. Elechrochem. Soc., 148, C784-C789, (2001).
25
