Haloplumbate Salts as Reagents for the Non-Aqueous Electrodeposition of Lead

Cyclic voltammetry experiments on the Pb(II) salts, [PPh4][PbX3] (X = Cl, Br, I) in CH2Cl2 solution ([PPh4]X supporting electrolyte) at a Pt disk electrode show reproducible nucleation and stripping features consistent with reduction to elemental Pb. The reduction potential shifts less cathodic from Cl (0.40 V)  Br (0.27 V)  I (0.19 V vs. Ag/AgCl), in line with the PbX bond strengths decreasing. Potentiostatic electrodeposition using [PPh4][PbCl3] in CH2Cl2 leads to growth of a thin film of crystalline Pb onto planar TiN electrodes, confirmed by SEM, EDX and XRD analysis. Electrodeposition under similar conditions onto a planar Au electrode leads to deposition of elemental Pb, accompanied by some alloying at the substrate/film interface, with XRD analysis confirming the formation of AuPb2 and AuPb3. Transferring the [PPh4][PbCl3] reagent into supercritical CH2F2 (17.5 MPa and 360 K) containing [PPh4]Cl led to very limited solubility of the Pb reagent; using [NBu4]Cl as supporting electrolyte caused an increase in solubility, although still lower than in liquid CH2Cl2. Cyclic voltammetry experiments (Pt disk) using this electrolyte also shows voltammetry consistent with Pb deposition, however the low solubility of the lead salt in scCH2F2 meant that electrodeposition onto a planar TiN substrate was not possible.


Introduction
The Group 14 elements have long been important to the electronics industry. Since the first experimental observation of transistor action in n-type polycrystalline germanium in 1947, and subsequently in polycrystalline silicon, both of these semiconductor elements have remained fundamental to transistor technology and integrated circuitry. 1 Meanwhile, graphene has recently emerged as highly attractive material for numerous applications, including electronics, sensors and energy storage devices. 2 Though used in various compound semiconductors (e.g. SnTe, PbTe and PbSnTe), 3 the heavier Group 14 elements, tin and lead, have received considerably less attention, although recent work has predicted new and exciting properties for these elements at the extreme nanoscale. 4,5 For example, a theoretical study 4 has predicted that α-tin will undergo a semimetal to semiconductor transition in nanowires with diameters less than ~ 4 nm.

2
The stable phase of lead is face centred cubic and metallic. Bulk lead is ductile, soft, highly malleable, has a high density and is a relatively poor electrical conductor. 6 Bulk lead is an elementary superconductor below Tc = 7.2 K and so there has been a lot of interest in investigating and tailoring its superconducting behaviour at the nanoscale. Lead nanowires with diameters of ~ 5 nm, synthesised using a template-assisted chemical vapour deposition (CVD) approach, have been shown to increase the Tc by 3-4 K above that of bulk Pb, while the upper critical field was almost two hundred times higher. 7 Other reported methods of fabricating Pb nanowires (of varying crystallinities and with diameters typically > 20 nm) include pressure casting into templates, 8 template-assisted electrodeposition from aqueous electrolytes 9,10,11 and solution phase synthesis involving the thermal decomposition of a lead precursor salt. 12 Lead also forms important compound semiconductors; the narrow band gap binary semiconductors PbS, PbSe and PbTe have potential uses in sensors, infrared photodetectors, solar photovoltaics and thermoelectric devices. Nanowires of these materials show improved carrier mobilities, larger absorption cross sections and better absorption and emission polarisation sensitivities. 13 A field effect transistor has been fabricated using a single PbTe nanowire. 14 Previously, we reported versatile precursor systems for the electrodeposition of several main-group metals from both liquid dichloromethane 15 and supercritical difluoromethane (scCH2F2). 16 Non-aqueous solvents such as these offer several advantages for the electrodeposition of complex nanostructured materials over conventional aqueous electrolytes. These include: (i) the use of a much wider range of reagents which can be tailored to the application; (ii) access to more reactive alloy compositions; (iii) a wider electrochemical window, enabling the deposition of reactive materials; and (iv) access to higher temperature electrodeposition.
Studies on halometallate complexes of Ge(II) and Ge(IV) in CH2Cl2 have revealed that the reduction potential is more accessible (less cathodic) for the lower oxidation state 17 and hence Pb(II) compounds would appear to be promising candidates for the non-aqueous electrodeposition of lead. Coordinatively saturated reagents can be preferable for electrodeposition since they can simplify the solution speciation, however, Pb(II) tends to form complexes with large and varying coordination numbers that are very labile in solution, making it difficult to design a precursor that is coordinatively saturated. We therefore chose to keep the systems as simple as possible, using well-defined halometallate salts incorporating organic cations to aid solubility in the low dielectric solvent.
Previous studies 15 have shown that p-block halometallates can be readily synthesised in high yields and purity, are easily handled and are stable at high temperatures. They typically show a high solubility and stability in organic solvents, allowing high-concentration electrolytes to be prepared. By using well-defined molecular species it is easy to control the speciation in solution. Furthermore, the tetrabutylammonium chlorometallate salts exist for a wide range of the p-block elements, presenting the prospect of co-deposition from these mutually compatible reagents to form compound semiconductors and alloys. Investigations into 3 the electrochemistry of [N n Bu4][MCl4] (M = In, Sb, Bi) and [N n Bu4]2[MCl6] (M = Se, Te) in CH2Cl2, using [N n Bu4]Cl as the supporting electrolyte, found that they form stable and reproducible electrochemical systems from which the element can be electrodeposited. 15 Due to the poor solubility of the Pb(II) halides in organic solvents, 18 work with lead has focused on halometallate salts with solubilising cations, specifically using the [PPh4] + cation.

Results and Discussion
The structural chemistry of the lead(II) halides is complex and varied due to the stereochemical activity (or not) of the lone pair and their tendency to polymerise in the solid state. 19 PbCl2 and PbBr2 are essentially isostructural, with a nine-coordinate metal centre best described as [7+2] coordinate, while PbI2 has the CdI2 hexagonal layer lattice structure. 17 Though formation of the halometallate anions in situ from MX2 may be possible due to the large excess of Xions in solution provided by the electrolyte, the poor solubility of PbX2 using either PbCl2 or Pb(NO3)2 as the lead source were unsuccessful, and so the [PPh4] + ion was selected as an alternative cation as its solubilising properties in organic solvents are well known and a number of lead(II) halide salts have already been isolated. 22,23,24,25 The reactions of PbX2 and   (6), Cyclic voltammetry and deposition studies were then performed on the Pb reagents in dry CH2Cl2 in order to evaluate their electrochemical properties. The supporting electrolyte used was [PPh4]X (where X matches the halogen present in the precursor) to avoid unnecessary complications through the introduction of other ions into the system, as Xis expected to be liberated during the reduction of the halometallate precursor.
To allow the efficient exclusion of air and moisture, the electrolyte preparation and electrochemical experiments were performed in a N2-filled glove box. The electrochemical experiments in CH2Cl2 were performed at room temperature using a three-electrode system in a single compartment electrochemical cell.  Extra features were observed on cyclic voltammograms recorded at an Au disk working electrode ( Figure 3).
The first scan on a fresh electrode shows two reduction waves, while on subsequent scans a third reduction process is also evident, presumably because the electrode surface has been altered. The underpotential deposition of Pb on gold in various media has been extensively studied, 27,28 while Pb is also known to form three Au alloys, Au2Pb, AuPb2 and AuPb3. 29 The charge associated with the first peak on the first scan 6 (2.8 µC) is comparable to the theoretical charge required for the formation of a Pb monolayer on the Au electrode (2.0 µC) based on the literature, 27 suggesting an underpotential deposition process. The peak deposition current for bulk Pb occurs at -0.38 V, while the deposition and stripping onset are both at -0.34 V.

Electrodeposition of lead from CH2Cl2 solution
The  Table 1. The deposited dark grey films on the electrodes were gently washed by immersing the films into CH2Cl2 before being analysed by SEM, EDX and XRD. SEM images of the deposited Pb films ( Figure 6) show that the material on the Au slide is formed of individual crystallites of differing sizes spread fairly uniformly across the electrode surface, while the crystallite morphology on the TiN slide is more regular. EDX analysis 9 shows Pb as the dominant peak, but, unsurprisingly, given the thinness of the deposited films, Au or Ti and Si peaks from the Au or TiN substrate also evident. No chloride was observed in any of the EDX spectra recorded. Grazing incidence X-ray diffraction (GIXRD) data were collected on the films deposited on TiN and gold under the conditions shown in Table 1 to understand their phase behaviour. Symmetric (θ-θ2) scans were also collected to test for preferred orientation of the deposit, though none was found and these patterns contained strong substrate reflections.
The GIXRD pattern of the film deposited on TiN contained the expected fcc lead peaks and weak reflections due to the substrate and to a small quantity of PbO, presumably due to air oxidation of the film. Rietveld PbO (litharge structure 31 ), although this is likely to be due to surface oxidation and hence the refined figure is probably higher than the real content due to the surface enhancement provided by the grazing incidence geometry. Some weak reflections due to the TiN substrate were also observed.
The film deposited on gold also underwent some alloying with the gold, and strong alloy reflections in the symmetric scans suggested this was mainly at the substrate-film interface. The formation of these alloy phases may explain the additional peaks observed in the voltammetry on gold disk electrodes (Figure 3). although these may be underestimates as they are likely to be mainly present at the substrate/film interface.
Peaks due to the gold substrate were also modelled, but due to preferred orientation of the substrate this was done in Le Bail mode. A weak reflection due to PbO was also observed and is marked in Figure 3b.

Conclusions
The series of tetraphenylphosphonium halometallate precursors for the electrodeposition of lead, The results from this work indicate that lead may be readily electrodeposited from the haloplumbate salts using liquid CH2Cl2. This broadens further the range of p-block elements accessible using these solvents, and may also offer the potential for electrodeposition of lead alloys and/or binary semiconductors, such as lead telluride (by combining [PbCl3]and [TeCl6] 2in a single electrolyte bath in CH2Cl2), in a similar manner to that which we used recently for the electrodeposition of functional Ge2Sb2Te5 phase change memory. 34

Experimental
Reactions were carried out under a dry dinitrogen atmosphere using Schlenk line and glove box techniques.
The products were stored, and spectroscopic samples prepared, in a N2-filled glove box. Solvents were dried and degassed prior to use. CH3CN and CH2Cl2 were distilled over CaH2; the former was stored over molecular sieves (4 Å). The lead dihalides were purchased from Sigma-Aldrich and used as received.
IR spectra were recorded as Nujol mulls between CsI plates using a Perkin Elmer Spectrum 100 spectrometer over the range 4000-200 cm -1 . NMR spectra were recorded using a Bruker AVII400 or AVIIIHD400 spectrometer. 1 H spectra were referenced to the residual solvent reference and 31 P{ 1 H} to external 85% H3PO4. Microanalytical measurements were performed by London Metropolitan University.
Single crystal X-ray data were collected using a Rigaku AFC12 goniometer equipped with an enhanced  Table 2 and the structure has been assigned CCDC number 1480713.

Electrochemical measurements in CH2Cl2
The electrolyte preparation and electrochemical experiments in dry CH2Cl2 were carried out under a dry dinitrogen atmosphere inside a glove box. The electrochemical measurements were performed at room temperature using a three-electrode system in a one-compartment electrochemical cell that had a volume of

Electrochemical measurements in scCH2F2
Supercritical fluid electrochemical studies were performed in a stainless steel high-pressure cell, the details of which have been described in previous publications. 36  were transferred into the cell inside a nitrogen-purged glove box. Once sealed, the cell was removed from the glove box, connected to a high-pressure rig and heated to the desired temperature using a band heater under PID (proportional-integral-derivative) control. The scCH2F2 was then introduced using a specialized carbon dioxide pump (PU-1580-CO2, JASCO) until the desired pressure was achieved. To ensure that the solution was homogeneous, the system was stirred during pumping using a magnetic stirrer. Stirring was stopped at least 5 minutes prior to electrochemical measurements. All experiments were carried out at 17.

Characterisation of electrodeposited lead
Electrodeposited films were investigated using SEM, EDX and XRD. A Jeol JSM 6500F field emission gun scanning electron microscope (FEG-SEM) equipped with an Oxford Instruments EDX detector was used for the SEM and EDX analyses of electrodeposited films, for which an accelerating voltage of 20 keV was employed.
XRD patterns of lead films were collected using a Riguku Smartlab Thin Film diffractometer with a 0.1 mm parallel line of Cu-Kα X-rays, 5° Soller slits and a DTex250 1D detector. Grazing incidence patterns were collected with a 1° incidence angle. Data were initially compared with databases using the Rigaku PDXL package, then refined using the GSAS package 38 and models from ICSD. 39