2023 Volume 64 Issue 10 Pages 2466-2470
The structural changes in electrodeposited palladium (Pd) films with desorption of co-deposited hydrogen were investigated, and the existing states of hydrogen in the Pd film were analyzed. The Pd films were electrodeposited from an alkaline bath consisted of palladium (II) chloride, ammonium chloride, and citric acid. Two pronounced desorption peaks at around 500 K and 880 K were observed in the thermal desorption spectrum of hydrogen from as-deposited Pd film. For the Pd films with lower hydrogen concentrations, the lattice contraction proceeded concurrently with hydrogen desorption at room temperature. The lattice parameter of Pd film decreased with increasing heat treatment temperature up to 500 K, and then increased with grain growth above 550 K. An exothermic peak corresponding to the hydrogen desorption at around 500 K was observed in the differential scanning thermal analysis curve. A large number of nano-voids were observed in the Pd film. These results suggested that the hydrogen desorption peaks observed at around 500 K and 880 K were ascribed to the break-up of vacancy-hydrogen clusters and the desorption of hydrogen molecules from nano-voids, respectively.
This Paper was Originally Published in Japanese in J. Japan Inst. Met. Mater. 86 (2022) 172–175. Figures 1 and 2 were slightly modified.
Palladium (Pd) plating is used as surface treatments for jewelry due to its excellent decorative properties.1) It is also applied to the surface treatments of electronic components and circuit boards due to its excellent corrosion resistance and soldering properties.2) During the electrodeposition process of Pd films, hydrogen is generated concurrently with the deposition of Pd atoms, and most of which is released as hydrogen gas, however, some of which is co-deposited as hydrogen atoms in the Pd film. The co-deposited hydrogen is known to induce the internal stress and cracks in the Pd film.3) We previously reported that the grain growth of the Pd film and the formation of a Cu–Pd alloy layer due to the interdiffusion between the Cu substrate and the Pd film after hydrogen desorption at room temperature were observed in electrodeposited Pd films with higher concentration of co-deposited hydrogen (x = H/Pd, x ≥ 5.8 × 10−2).4) On the other hand, the lattice contraction was observed in Pd films with lower concentration of co-deposited hydrogen (x ≤ 0.4 × 10−2) after hydrogen desorption. Such grain growth associated with the hydrogen desorption from the plating films at room temperature has also been observed in the electrodeposited Cu films with high hydrogen concentration,5,6) and these diffusion phenomena are considered to be caused by the diffusion enhancement effect of hydrogen-induced superabundant vacancies (SAVs).7)
In this study, we have investigated the structural changes with hydrogen desorption for the electrodeposited Pd films with low hydrogen concentration in detail and analyzed the existing states of co-deposited hydrogen in Pd films.
An alkaline plating bath containing 0.03 M (M = mol dm−3) palladium chloride (II), 0.50 M ammonium chloride, and 0.05 M citric acid was used for electrodeposition of Pd films. The pH of the plating bath was adjusted to 8.3 with an aqueous ammonia solution. Stainless steel (SUS304) plates from which the Pd films can be peeled off and Cu plates were used as the substrates. These substrates were cleaned with acetone, and the Cu plates were chemically etched with nitric acid for a few seconds. The substrate was covered with masking tape except for the deposition area of Pd film of 10 mm × 10 mm. A platinum plated titanium mesh (25 mm × 20 mm) was used as a counter electrode. The bath temperature was maintained at 298 K using a thermostatic bath and the solution was stirred with a magnetic stirrer. The electrodeposition was performed at a constant current density of 50 A m−2. The film thickness and current efficiency were calculated by the gravimetric method.
Two specimens of approximately 5 mm square were cut from a 10 mm square of the Pd film peeled off from the stainless steel substrate, and hydrogen analysis by thermal desorption spectroscopy (TDS) and thermal analysis by differential scanning calorimetry (DSC) were performed, respectively. In the TDS, the specimen was placed in a quartz tube and evacuated for about 1 hour until the degree of vacuum stabilized at about 10−5 Pa. The hydrogen desorbed from the specimen in the temperature range of 300∼1000 K at a heating rate of 5 K min−1 was detected by a quadrupole mass spectrometer (Canon Anelva M-ATOIC). The amount of hydrogen in the Pd films was quantified comparing a calibration curve prepared with magnesium hydride (MgH2) as the standard sample. Thermal analysis was performed using a differential scanning calorimeter (Parkin Elmer, DSC-8500) with a heating rate of 5 K min−1 in an Ar gas flow of 20 mL min−1 in the temperature range of 223∼600 K. The TDS measurement and the DSC measurement were simultaneously performed. The electrodeposited Pd films were heat treated in the vacuum condition for 30 minutes at predetermined temperature in the range of 400∼1100 K. The lattice parameter of electrodeposited Pd films was calculated by Nelson-Riley extrapolation method from the diffraction lines in the (111), (200), (220), and (311) planes obtained using Mo-Kα radiation as the X-ray source in X-ray diffraction (XRD) measurements (Rigaku RINT-2200). The grain size of electrodeposited Pd films was calculated by Scherrer equation from the full width at half maximum of the (111) diffraction line. The microtexture of electrodeposited Pd films was observed by transmission electron microscopy (TEM, JEOL JEM-2100). The specimens were prepared by the ion milling equipment (GATAN PIPS Model 691D), and observed from the plane direction.
The thickness of electrodeposited Pd films was approximately 2 µm, and the current efficiency was about 80% due to the concomitant hydrogen evolution in the electrodeposition process. Figure 1 shows the time dependent changes in the lattice parameter at room temperature of the electrodeposited Pd films on the Cu substrate and the self-standing Pd films peeled off from the stainless steel substrate. The lattice parameter of Pd film on the Cu substrate immediately after electrodeposition was smaller than that of the standard Pd sample (astd = 0.3890 nm),8) and decreased with the passage of time. The lattice contraction was also observed in the self-standing Pd films, and the lattice parameter decreased to a = 0.3883 nm after 36 days and remained almost constant until 78 days. Both the electrodeposited Pd films on the Cu and stainless steel substrates exhibited random microcrystalline textures with the grain sizes less than 100 nm without differences between them. For the self-standing Pd films, the internal stress is considered to be released, however, the similar lattice parameter changes as in the Pd films on the Cu substrate were observed. Therefore, in the subsequent analysis, the self-standing Pd films were used to avoid alloying with the Cu substrate during the heat treatment process.
Time dependent changes in lattice parameter of electrodeposited Pd films. ○: Pd films peeled off from SUS304 substrate. ●: Pd films on Cu substrate.
Figure 2 shows the changes in lattice parameter and grain size of the self-standing Pd films with respect to the annealing temperature. The lattice contraction of the Pd film immediately after electrodeposition proceeded further with increasing annealing temperature up to 500 K, however it was relaxed from 550 K and the lattice parameter became almost the same as that of the standard Pd sample8) above 600 K. The grain size obtained using the Scherrer equation gradually increased with increasing annealing temperature from approximately 44 nm immediately after electrodeposition to approximately 124 nm at 900 K, and then increased rapidly between 900 K and 1100 K. Since the recrystallization temperature estimated from the melting point of Pd is around 914 K, such rapid grain growth is considered to be due to the recrystallization of the Pd film.
Lattice parameter (○) and grain size (□) of electrodeposited Pd films as a function of annealing temperature. XRD was measured at room temperature after heat treatments for 30 minutes at each temperature.
Figure 3 shows the plane-view TEM images of the electrodeposited Pd films before and after annealing. The Pd film immediately after electrodeposition showed a random texture consisting of microcrystals less than 100 nm (Fig. 3(a)), and a large number of nano-voids were observed within the grains (Fig. 3(b)). After annealing at 600 K, larger grains over 100 nm were observed (Fig. 3(c)), and after annealing at 900 K, all the grains became larger than 100 nm (Fig. 3(e)). The size and number of numerous nano-voids observed within the grains of Pd films increased with increasing annealing temperature (Fig. 3(d), (f)).
Plane-view TEM images of electrodeposited Pd films. (a), (b): As deposited, (c), (d): Annealed at 600 K, (e), (f): Annealed at 900 K.
Figure 4 shows the thermal desorption spectra of hydrogen from the Pd films 1 hour and 7 weeks after electrodeposition. In this measurement, the same samples as the self-standing Pd films shown in Fig. 1 were used. The hydrogen concentration x of the Pd film was determined as the atomic ratio (H/Pd), and the vertical axis shows the x differentiated by temperature T, that is, the hydrogen desorption rate (−dx/dT) from the film. In the Pd film 1 hour after electrodeposition (Fig. 4(a)), two pronounced desorption peaks were observed at around 500 K and 880 K, and a small peak was observed at around 630 K, and the hydrogen concentration was x = 1.9 × 10−2. In the Pd film 7 weeks after electrodeposition (Fig. 4(b)), the two pronounced desorption peaks shifted to the high-temperature side at around 525 K and 980 K, respectively. Their intensity decreased and the hydrogen concentration decreased to x = 1.0 × 10−2. Since the thickness of the Pd film was approximately 2 µm, which is sufficiently thin to hydrogen diffusion in the Pd film during the heating process, the observed desorption peak temperatures correspond to the binding energy of hydrogen in the electrodeposited Pd film. Therefore, the shift to the high-temperature side and the decrease in intensity of the desorption peak at around 500 K by room temperature aging is attributed to the desorption of weakly trapped hydrogen. The desorption peak at around 880 K in Fig. 4(a) is attributed to the desorption of hydrogen trapped in the grain boundaries and molecular hydrogen existed in the voids observed in the TEM image in Fig. 3, since the rapid grain growth of the Pd film occurred at around this temperature (Fig. 2). The increase in number and size of voids in the Pd film by annealing indicates the involvement of vacancy-hydrogen clusters. During the heating process of the Pd films, hydrogen began to desorb from around 400 K and the grain sizes gradually increased concurrently. Diffusion of Pd atoms at such low temperatures is considered to proceeded due to the migration of unstable vacancies generated by the hydrogen desorption from the vacancy-hydrogen clusters. During this process, it can be inferred that new nano-voids were generated by the coagulation of vacancies, subsequently the voids expanded due to the absorption of vacancies into the nano-voids and the coalescence of nano-voids. It is known that the vacancy-hydrogen clusters in face-centered cubic metals can trap up to six hydrogen atoms in one vacancy.9) Since there are many vacancy-hydrogen clusters with a high coordination number of hydrogen atoms in the Pd film immediately after electrodeposition, it is considered that a large number of hydrogen molecules were existed in the nano-voids generated by the aggregation of these clusters. On the other hand, in the Pd film after 7 weeks, weakly trapped hydrogen atoms in the 3rd to 6th positions9) in the vacancy-hydrogen clusters were desorbed at room temperature and the coordination number decreased, thus the desorption of strongly trapped hydrogen atoms were observed at the high temperature side during the heating process, and it is considered that the number of hydrogen molecules in the generated nano-voids decreased.
Thermal desorption spectra of hydrogen from electrodeposited Pd films measured 1 hour (a) and 7 weeks (b) after deposition. Heating rate was 5 K min−1 and x = H/Pd.
Figure 5 shows the hydrogen thermal desorption spectrum and a DSC curve of the electrodeposited Pd film. A desorption peak was observed at around 500 K in the hydrogen thermal desorption spectrum, and a corresponding exothermic peak was observed in the DSC curve. Hydrogen absorbed into bulk Pd using techniques such as high-pressure synthesis or electrolytic charging occupies interstitial sites in the Pd lattice. The desorption of interstitial hydrogen occurs up to 450 K and is an endothermic reaction.10,11) On the other hand, it has been reported that sample in which hydrogen was absorbed into a Pd wire after high-pressure torsion exhibited an exothermic peak at around 480 K, which was attributed to the break-up of vacancy-hydrogen clusters.12) The results of this study are consistent with the latter, indicating that the hydrogen desorbed at around 500 K from the electrodeposited Pd film is existed as vacancy-hydrogen clusters.
Hydrogen thermal desorption spectrum (a) and DSC curve (b) of electrodeposited Pd films.
In the case of hydrogen absorption into bulk Pd, the hydrogen atoms penetrate from the surface and occupy the interstitial sites in the Pd lattice, and consequently the lattice expansion increased with increasing hydrogen concentration.13,14) On the other hand, Fukai et al. have reported that the lattice contraction of Pd foils held under high hydrogen pressures and high temperatures was observed after lattice expansion due to hydrogenation, and such lattice contraction was explained to be due to the formation of hydrogen-induced superabundant vacancies (SAVs).15,16) Hydrogen atoms co-deposit with Pd atoms during the non-equilibrium electrodeposition process, thus it is considered that abundant vacancies are introduced into the Pd film. Here, assuming that the cause of the lattice contraction in the electrodeposited Pd film in this experiment is due to the formation of SAVs, the thermal equilibrium concentration of vacancies was estimated using eq. (1).9)
| \begin{equation} 3\frac{\Delta a}{a} = \frac{\varOmega_{R}}{\varOmega_{0}}x_{v} \end{equation} | (1) |
Here, xv, ΩR, and Ω0 represent the concentration of vacancies, the volume contraction of the lattice, and the atomic volume of Pd, respectively. The thermal equilibrium concentration of vacancies in the Pd film on the Cu substrate and the self-standing Pd film immediately after electrodeposition calculated using eq. (1) were xv = 2.6 × 10−3 and xv = 4.7 × 10−3, respectively. Assuming that the coordination number of hydrogen atoms around vacancy is six, the vacancy concentration estimated from the amount of desorbed hydrogen attributed to the break-up of the vacancy-hydrogen clusters in the thermal desorption spectrum shown in Fig. 4 was xv = 1.6 × 10−3, which was almost the same as that calculated from the lattice contraction. These values were about 10 times larger than the thermal equilibrium concentration of vacancies (≈10−4) in perfect crystalline Pd at around its melting point. Assuming the lattice contraction in the as-deposited Pd films was caused by the presence of such high concentrations of vacancies, then the progression of the lattice contraction observed upon both the room-temperature aging (Fig. 1) and increasing annealing temperature (Fig. 2) can be explained by the release of hydrogen from vacancy-hydrogen clusters. Furthermore, since the exothermic reaction due to the hydrogen desorption was observed at above 450 K, it is considered that the vacancies annihilated with the break-up of vacancy-hydrogen clusters, and consequently the lattice contraction was released.
Thus, the lattice parameter changes observed in the electrodeposited Pd films by room-temperature aging and heat treatments suggest the formation of hydrogen-induced superabundant vacancies in the Pd films.
In this study, we investigated the structural changes with hydrogen desorption in the electrodeposited Pd films with low hydrogen concentration, and analyzed the existing states of co-deposited hydrogen in the Pd films. The results are summarized as follows:
These results support the formation of hydrogen-induced superabundant vacancies in the Pd films during the electrodeposition process.
This study was partially supported by a grant from the Hyogo Science and Technology Association for the Academic Research of 2021.
