1: \documentstyle[multicol,aps,prl,epsf]{revtex}
2:
3: \pagestyle{empty}
4: \begin{document}
5:
6: \draft
7:
8: \title{Fabrication of stable Pd nanowire assisted by hydrogen in solution}
9:
10: \author{Manabu Kiguchi and Kei Murakoshi}
11:
12: \address{Department of Chemistry, Graduate School of Science,
13: Hokkaido University, Sapporo, 060-0810, Japan}
14:
15: \date{\today}
16:
17: \maketitle
18:
19: \begin{abstract}
20:
21:
22: We have mechanically fabricated a Pd nanowire in solution under electrochemical potential
23: control. A clear feature appeared in the conductance histogram when the electrochemical
24: potential of the Pd wire was kept at the hydrogen evolution potential. Conductance traces
25: showed the Pd wire was stretched 0.4 nm in length just before breaking, suggesting that at
26: least two Pd atoms might contribute to the formation of the Pd wire. The results indicate that a
27: certain atomic configuration of the Pd nanowire is stabilized by hydrogen. We discuss the
28: stabilization mechanism due to changes in bond strengths caused by hydrogen adsorption or
29: incorporation.
30:
31: \end{abstract}
32:
33: \medskip
34:
35:
36:
37: \begin{multicols}{2}
38: \narrowtext
39:
40: Fabrication and characterization of metal atomic wires have attracted attention due to their
41: potential applications in future electronic devices \cite{1}. Recently, Ir, Pt, Au mono atomic wires
42: have been fabricated by pulling nano contacts using a scanning tunneling microscope (STM)
43: or a mechanically controllable break junction (MCBJ) in ultra high vacuum (UHV) at low
44: temperatures \cite{1,2,3}. In addition to methods in UHV, the electrochemical method has been
45: recognized as another powerful approach in the fabrication of metal nanowires \cite{4,5,6,7,8}. Recently,
46: we have demonstrated that Fe, Co, and Ni metal nanowires were stabilized in solution at room
47: temperature via the hydrogen evolution reaction \cite{6,7}. Furthermore, we observed results
48: suggesting the formation of a mono atomic Ni wire in solution, while mono atomic wires of
49: 3d and 4d metals have not been prepared in UHV up to now.
50:
51: In the present study, we have studied the hydrogen assisted stabilization of the Pd (4d
52: metal) nanowire in the fabrication of a mono atomic wire. A mono atomic Pd wire is predicted
53: to be ferromagnetic \cite{9}, and to test this hypothesis it is useful to fabricate this wire. In addition
54: to the stabilization of the Pd nanowire by hydrogen, the interaction between hydrogen and the
55: Pd nanowire should be interesting. Since the interaction between hydrogen and Pd is large,
56: the nanowire of the palladium hydride formed in solution should have electrical properties \cite{10}.
57:
58: In our study of the Pd nanowire, we have chosen to measure the conductance. Electrical
59: conductance through a metal nanowire on an atomic scale is expressed by $G=2e^{2}/h \sum T_{i}$
60: where $T_{i}$ is the transmission probability of the $i$-th conductance channel, $e$ is the
61: electron charge, and $h$ is Plank's constant \cite{1}. Since conductance quantization depends on the
62: atomic structure of a nanowire and the inherent properties of metals, we can study the atomic
63: and electronic structures of metal nanowires by the conductance measurements. We have
64: already studied conductance quantization of a Pd nanowire in solution \cite{8}. The conductance
65: histogram displayed a clear feature peak at 1 $G_{0}$, while conductance quantization was not
66: observed in UHV at room temperature \cite{11}. At low temperatures in UHV, the appearance of the
67: 1 $G_{0}$ and 0.5 $G_{0}$ features were reported in the conductance histogram for Pd under hydrogen
68: dosing \cite{12}. At the present stage, it is clear that hydrogen stabilizes the structure with a 1 $G_{0}$ or
69: 0.5 $G_{0}$ both in UHV and solution. However, there has been little discussion about the structure,
70: that is, it is not clear whether the conductance originates from the bridging hydrogen
71: molecules or the Pd nanostructure stabilized by hydrogen.
72:
73: In the present study, we have investigated the mechanical properties of the Pd nanowire
74: under electrochemical potential control. To obtain information about the structure of the
75: nanowire, we have analyzed each conductance trace, as the conductance traces provide
76: information about the nanowire such as wire length and thickness. Using the trace analysis
77: results, we discuss the formation mechanism of the Pd mono atomic wire in solution.
78:
79: The experimental design used in this study was the same as described in our previous
80: report \cite{5,6,7,8}. An electrochemical scanning tunneling microscope (EC-STM) was used to fabricate
81: the Pd nanowire. The electrochemical potential ($\Phi_{0}$) was controlled using a potentiostat
82: (Pico-Stat, Molecular Imaging Co.) with a Ag/AgCl reference electrode. The electrolyte
83: consisted of 1 mM PdCl$_{4}$, 2 mM HCl and 0.05 M H$_{2}$SO$_{4}$. The Pd nanowire was prepared as
84: follows. First, the electrochemical potential of both the STM tip and substrate were
85: maintained at less than +450 mV where Pd bulk deposition proceeds \cite{13}. After the sufficient
86: deposition of Pd onto the surfaces of the STM tip and the substrate, the tip was pressed into
87: the substrate and contacts were then pulled out. During contact breaking, a Pd nanowire is
88: formed between the tip and substrate. Conductance was measured during the breaking process
89: under an applied bias of 20 mV between the tip and substrate. Statistical data (conductance
90: histograms, etc.) was obtained for a large number (over 3000) of individual conductance
91: traces.
92:
93: Figure 1 (a) shows the cyclic voltammogram of the Au electrode in solution. Pd deposition
94: and hydrogen adsorption/evolution proceeded when the electrochemical potential of Au
95: electrodes were maintained under more negative conditions than $\Phi_{0}$= 450 mV and $\Phi_{0}$= -200
96: mV, respectively \cite{13,14,15}. Figure 1 (b) and (c) show the conductance histogram of the Pd
97: nanowire at (b) $\Phi_{0}$= -250 mV and (c) $\Phi_{0}$= 100 mV. There is no clear feature in the
98: conductance histogram at $\Phi_{0}$= 100 mV, suggesting that no preferential atomic configurations
99: were formed. The featureless conductance histogram agreed with the previous results in UHV
100: at room temperature \cite{11}. When the electrochemical potential of the nanowire was maintained
101: under more negative conditions ($\Phi_{0}$= -200 mV), a feature appeared near the unit of the
102: quantum conductance (0.9 $G_{0}$). These results agree with our previous results \cite{8}.
103:
104: Next, we analyzed the conductance traces to determine the structure of the Pd nanowire.
105: Figure 2 shows the typical conductance traces at (a) $\Phi_{0}$= 100 mV and (b,c) $\Phi_{0}$= -250 mV. As
106: a general tendency, the conductance changed in a stepwise fashion rather than continuously.
107: At $\Phi_{0}$= 100 mV, the conductance values of the plateaus were non-reproducible, resulting in
108: featureless conductance histograms (Fig. 1 (c)). On the other hand, the 0.6-1.2 $G_{0}$ plateau with
109: a stretched length ca. 0.5 nm appeared in the trace at $\Phi_{0}$= -250 mV as shown in Fig. 2(b).
110: These characteristic plateaus were the origin of the feature at 0.9 $G_{0}$ in the conductance
111: histogram (Fig. 1 (b)). The value of 0.9 $G_{0}$ may correspond to that of a Pd mono atomic
112: contact. This value is comparable to previously documented estimations of clean transition
113: metal mono atomic contacts in UHV systems \cite{1}. The length of the plateau also showed
114: dependence on the electrochemical potential. To evaluate characteristics of the plateau length
115: quantitatively, we examined the distribution of lengths for the last conductance plateau. For
116: comparison, we examined the Pt nanowire under the hydrogen evolution, in which a hydrogen
117: molecule would bridge between Pt electrodes. Smit et al. showed that the 1 $G_{0}$ feature
118: appeared in the conductance histogram under hydrogen dosing at low temperatures in UHV \cite{16}.
119: By measuring the conductance and vibration spectra, they showed that the 1 $G_{0}$ feature
120: originated from a single hydrogen molecule bridging Pt electrodes. In the breaking trace, the 1
121: $G_{0}$ plateau broke within 0.2 nm, which is consistent with the small size of the hydrogen
122: molecule \cite{16}. The inset of Fig. 3 (a) shows the conductance histogram of the Pt nanowire under
123: hydrogen evolution reaction in solution. A feature close to 1 $G_{0}$ appears in the histogram. The
124: feature appears to have originated from the bridging hydrogen molecule, although the
125: situation in solution might be different from that in UHV at low temperatures. Figure 3 (a)
126: shows the distribution of lengths for the last conductance plateau for Pt. The length of the last
127: plateau was defined as the distance between the points at which the conductance dropped
128: below 1.3 $G_{0}$ and 0.7 $G_{0}$, respectively. These values were fixed in the following analysis. The
129: contact broke at a short length of ~0.2 nm which is in agreement with previous results \cite{16}.
130:
131: Figure 3 (b) shows the distribution of lengths for the last conductance plateau for Pd at $\Phi_{0}$=
132: -250 mV. For comparison, we also show the distribution of the lengths for Pd at $\Phi_{0}$= 100 mV.
133: While the contact broke within 0.2 nm at $\Phi_{0}$= 100 mV, the plateau was stretched 0.4 nm in
134: length at $\Phi_{0}$= -250 mV. If a hydrogen molecule bridges between the Pd nanowire, the plateau
135: length would be restricted to be 0.2 nm using the Pt results. The long plateau suggests that the
136: 1 $G_{0}$ feature did not originate from the hydrogen bridges. Since the Pd-Pd distance is 0.27 nm
137: for bulk Pd, the 0.4 nm long 1 $G_{0}$ plateau corresponds to a mono atomic wire of about two Pd
138: atoms. A hydrogen atom would adsorb on the surface or be incorporated in the wire under the
139: hydrogen evolution reaction, and thus, a stable Pd nanowire would be formed in solution.
140:
141: The formation of a Pd mono atomic wire was also supported by the conductance trace. We
142: occasionally observed reversible transition of conductance between a 1 $G_{0}$ and 0.6-0.7 $G_{0}$ as
143: seen Fig. 2(c). A similar reversible transition of conductance was reported for Au mono
144: atomic wires in the presence of physically adsorbed hydrogen in UHV at low temperatures
145: (10-30 K) \cite{17,18}. Similar behavior was also observed for the Au mono atomic wire in solution
146: under electrochemical potential control \cite{19}. The fractional conductance and reversible
147: transition are explained by the formation of a dimerized structure in the Au mono atomic wire,
148: and the dynamic structural transition between a dimerized wire and an equal-spacing wire,
149: respectively. The dimerization is a characteristic of a mono atomic wire. Therefore, the
150: reversible transition of the conductance observed in the present study may reflect a similar
151: dynamic structural transition in the Pd mono atomic wire. Present observations strongly
152: suggest that the Pd mono atomic wire was fabricated in solution at room temperature under
153: the hydrogen evolution reaction. Hydrogen stabilized the specific structure of the Pd mono
154: atomic wire in solution at room temperature, leading to the characteristic conductance
155: quantization behavior which has not observed in UHV at room temperature. At the present
156: stage, it remains unclear whether hydrogen atom adsorbs on the wire or incorporates in the
157: wire. In-situ point contact spectroscopy should be indispensable to obtain an information on
158: the structure \cite{16}.
159:
160: Adsorbed or incorporated hydrogen may play a decisive role in improving the stability of
161: Pd mono atomic wire in solution. Generally, the bond strength between atoms increases when
162: the coordination number of neighboring atoms becomes low \cite{3}. Further increments in the
163: stabilization becomes significant in 5d metals due to the relative effects of the valence
164: electron, leading to the formation of a very stable nanowire of 5d metals. In the case of 3d and
165: 4d metals, it is known that stabilization of the bond strength due to changes in the
166: coordination number as well as from the relative effect of electrons is not significant. These
167: characteristics are key in the difficulty of preparing 3d and 4d metal nanowires showing stable
168: conductance quantization. However, hydrogen adsorption/incorporation in the Pd nanowire is
169: expected to change this situation. Adsorption of hydrogen to Pd proceeds via electron transfer
170: from Pd to hydrogen \cite{20}. Since the top of the 4d band of bulk Pd consists of states with
171: anti-bonding characteristics, a decrease in the occupancy of the anti-bonding 4d band due to
172: hydrogen adsorption would increase the bond strength between Pd atoms. Thus, hydrogen
173: adsorption/incorporation may result in the stabilization of the Pd nanowire. This change in the
174: electronic structure of the Pd nanowire under the hydrogen evolution reaction resulted in the
175: fabrication of the stable Pd nanowire with 0.9 $G_{0}$. Significant changes in the conductance
176: value from 1.8 $G_{0}$ clean Pd to 0.9 $G_{0}$ hydrogen adsorption/incorporation Pd may also originate
177: from this change in the electronic structure of the Pd nanowire. Finally, we briefly compare
178: the present results with that in UHV at low temperature. While the 1 $G_{0}$ and 0.5 $G_{0}$ features
179: were reported in the conductance histogram for Pd in UHV at low temperature \cite{12}, the 0.5 $G_{0}$
180: feature was not observed in the conductance histogram under hydrogen evolution reaction in
181: solution. In UHV, a hydrogen molecule is introduced to the Pd nano wire. On the other hand,
182: in solution under hydrogen evolution reaction, an atomic hydrogen is formed on the Pd
183: surface as dominant intermediate of the reaction \cite{21}. The situation in solution, which is
184: different from that in UHV at low temperature, may lead to the absence of the structure of 0.5
185: $G_{0}$ peak in the histogram.
186:
187: In conclusion, we have observed conductance quantization behavior of mechanically
188: fabricated Pd nanowire under electrochemical potential control. Under the hydrogen evolution
189: reaction, a peak close to a 1 $G_{0}$ appeared in the conductance histogram, suggesting that a
190: certain atomic arrangement was maintained in solution. Relatively long 0.4 nm stretch lengths
191: and reversible conductance fluctuations during stretching suggests the formation of a
192: one-dimensional Pd mono atomic wire in solution at room temperature.
193: This work was partially supported by a Grant-in-Aid for Scientific Research A (No.
194: 16205026) and Grant-in-Aid for Scientific Research on Priority Areas (No. 17069001) from
195: MEXT.
196:
197:
198:
199:
200:
201:
202:
203: \begin{references}
204:
205: \bibitem{1} N. Agrait, A.L. Yeyati, J.M. van Ruitenbeek, Physics Reports {\bf 377}, 81 (2003).
206: \bibitem{2} S.R. Bahn and K.W. Jacobsen, Phys. Rev. Lett. {\bf 87}, 266101 (2001).
207: \bibitem{3} R. H. M. Smit, C. Untiedt, A. I. Yanson, and J. M. van Ruitenbeek, Phys. Rev. Lett. {\bf 87},
208: 266102 (2001).
209: \bibitem{4} C. Shu, C. Z. Li, H. X. He, A. Bogozi, J. S. Bunch, and N. J. Tao, Phys. Rev. Lett. {\bf 84}, 5196
210: (2000).
211: \bibitem{5} J. Li, T. Kanzaki, K. Murakoshi, Y. Nakato, Appl. Phys. Lett. {\bf 81}, 123 (2002).
212: \bibitem{6} M. Kiguchi, T. Konishi, K. Murakoshi, Appl. Phys. Lett. {\bf 87}, 043104 (2005).
213: \bibitem{7} T. Konishi, M. Kiguchi, K. Murakoshi, Jpn. J. Appl. Phys. {\bf 45}, 2000 (2006).
214: \bibitem{8} J. Li, Y. Nakato, and K. Murakoshi, Chem. Lett. {\bf 34} (2005) 374.
215: \bibitem{9} A. Delin, E. Tosatti, R. Weht. Phys. Rev. Lett. {\bf 92}, 57201 (2004).
216: \bibitem{10} A.G. Lipson, B.F. Lyakhov, D.M. Sakov, V.A. Kuznetsov, Phys. Solid State {\bf 38}, 915(1996).
217: \bibitem{11} A. Enomoto, S. Kurokawa, and A. Sakai, Phys. Rev. B {\bf 65}, 125410 (2002).
218: \bibitem{12} Sz. Csonka, A. Halbritter, G. Milhaly, O.I.Shklyarevskii, S. Speller, H.van Kepen,
219: Phys. Rev. Lett. {\bf 93}, 16802(2004).
220: \bibitem{13} H. Naohara, S. Ye, K. Uosaki, J. Phys. Chem. B {\bf 102}, 4366 (1998).
221: \bibitem{14} S.Y. Liu, Y.H. Kao, Y.O. Sy, T.P. Perng, J. Alloys and Compounds {\bf 293-295}, 468 (1999).
222: \bibitem{15} A. Czerwinski, R. Matassi, S. Zamponi, J. Electroanl. Chem. {\bf 316}, 211 (1991).
223: \bibitem{16} R. H. M. Smit, Y. Noat, C. Untiedt, N. D. Lang, M. C. van Hemert and J. M. van
224: Ruitenbeek, Nature {\bf 419}, 906 (2002).
225: \bibitem{17} Sz. Csonka, A. Halbritter, G. Mihaly, E. Jurdik, O. I. Shklyarevskii, S. Speller, and H. van
226: Kempen, Phys. Rev. Lett. {\bf 90}, 116803 (2003).
227: \bibitem{18} Sz. Csonka, A. Halbritter, and G. Mihaly, Phys. Rev. B {\bf 73}, 75405 (2006).
228: \bibitem{19} M. Kiguchi, T. Konishi, and K. Murakoshi, Phys. Rev. B {\bf 73}, 125406 (2006).
229: \bibitem{20} V. Ledentu, W. Dong, P. Sautet, G. Kresse and J. Hafner, Phys. Rev. B {\bf 57}, 12482 (1998).
230: \bibitem{21} G. Jerkiewicz, Prog. Surf. Sci. {\bf 57}, 137 (1998).
231: \end{references}
232:
233:
234: \begin{figure}
235: \begin{center}
236: \leavevmode\epsfysize=55mm \epsfbox{fig1.eps}
237: \caption{(a) Cyclic voltammogram observed in 1 mM PdCl$_{4}$, 2 mM HCl and 0.05 M H$_{2}$SO$_{4}$.
238: The Conductance histograms at (b) $\Phi_{0}$ = -250 mV, (c) $\Phi_{0}$ = 100 mV. }
239: \label{fig1}
240: \end{center}
241: \end{figure}
242:
243:
244:
245: \begin{figure}
246: \begin{center}
247: \leavevmode\epsfysize=50mm \epsfbox{fig2.eps}
248: \caption{Typical conductance traces of Pd nanowire at (a) $\Phi_{0}$ = 100 mV, (b,c) $\Phi_{0}$ = - 250 mV.}
249: \label{fig2}
250: \end{center}
251: \end{figure}
252:
253: \begin{figure}
254: \begin{center}
255: \leavevmode\epsfysize=50mm \epsfbox{fig3.eps}
256: \caption{The distribution of lengths for the last conductance plateau for Pt under hydrogen
257: evolution reaction ($\Phi_{0}$ = - 800 mV). Inset: Conductance histogram of Pt nanowire at $\Phi_{0}$ = - 800
258: mV. (b) Distribution of lengths for the last conductance plateau for Pd at $\Phi_{0}$ = 100 mV and $\Phi_{0}$ =
259: - 250 mV. }
260: \label{fig3}
261: \end{center}
262: \end{figure}
263:
264:
265:
266:
267: \end{multicols}
268: \end{document}
269:
270:
271:
272:
273:
274: