1202 ieee transactions on nanotechnology, vol. 10, no. … · 1202 ieee transactions on...

1202 IEEE TRANSACTIONS ON NANOTECHNOLOGY, VOL. 10, NO. 5, SEPTEMBER 2011

Spontaneous Formation of In-Whiskers on YIn3

Thin Films Deposited by CombinatorialMagnetron Sputtering

Tetsuya Takahashi, Ahmed Abdulkadhim, Denis Music, and Jochen M. Schneider

Abstract—We demonstrate the spontaneous In-whisker forma-tion from combinatorially sputtered YIn3 thin films. The mor-phology and extrusion kinetics of In-whiskers are affected by thelocal chemical composition. In-whiskers are extruded from the filmsurface upon exposure to atmosphere, but not in vacuum. Conse-quently, In-whisker growth is accompanied by an increase in oxy-gen content in the films. The results presented enable controlledprocessing of 1-D submicrometer/nanostructured materials withrespect to morphology and growth kinetics.

Index Terms—Combinatorial sputtering, metal whisker,nanofabrication, oxidation.

I. INTRODUCTION

1 -D NANOSTRUCTURED materials also referred to asnanowires, nanorods, and nanowhiskers are known to ex-

hibit unique physical and chemical properties compared to 3-D bulk materials. The attractive properties of nanowires pro-vide many potential applications as novel functional nanode-vices [1], [2]. While many different synthesis techniques forself-organized nanowires have been reported, such as vapor–liquid–solid (VLS) method [3], solution method [4], [5], laserablation [6], [7], and thermal evaporation [8], [9], the develop-ment of new synthesis concepts and techniques for nanowireswith well-controlled morphology, growth rate, size, crystallinity,and chemistry is a challenging task.

Nanowires may be classified into two groups based on theirgrowth behavior. First, wires growing from their tips, the wireforming species are supplied from vapor and/or liquid phase sur-rounding. 1-D growth in this case is most commonly facilitatedthrough confinement by a liquid droplet. Most synthesis tech-niques reported so far utilize this mechanism. Second, wires arealso formed by growth from the roots. A well-known exampleis the spontaneous formation of soft metal (SM)-whiskers, suchas Sn, Cd, and Zn used for electroplating materials for elec-

Manuscript received September 20, 2010; revised January 21, 2011; acceptedFebruary 28, 2011. Date of publication April 5, 2011; date of current versionSeptember 8, 2011. This work was supported by the Air Force Office ofScientific Research, and by the Air Force Material Command, U.S. Air Force,under Grant FA8655-07-1-3052. The review of this paper was arranged byAssociate Editor M. Merlyne.

The authors are with the Materials Chemistry, Rheinisch-Westfalische Tech-nische Hochschule (RWTH) Aachen University, Aachen D-52056, Germany(e-mail: [email protected]; [email protected];music@ mch.rwth-aachen.de; [email protected]).

Color versions of one or more of the figures in this paper are available onlineat http://ieeexplore.ieee.org.

Digital Object Identifier 10.1109/TNANO.2011.2129527

tronic components [10]–[12]. Although the SM-whisker growthmechanism has not been fully understood; yet, it is generally ac-cepted that a major driving force for the growth of SM-whiskersis related to compressive stresses developed in SM-base matri-ces [11], [13]–[15]. Namely, SM can be extruded out of the sur-face in the form of whiskers to release the compressive stressescaused by intrinsic (e.g., residual stress related to material pro-cessing, impurities, microstructural defects, and intermetallicformation) and/or extrinsic reasons (e.g., external stress and ox-idation, etc.).

Previously, the stress-induced whisker growth has been rec-ognized as a possible synthesis technique of 1-D nanostructuredmaterials. Shim et al. have reported the growth of Bi nanowireson as-sputtered films after thermal annealing, which was facili-tated by the relaxation of thermal stress developed due to a mis-match of thermal expansion between the film and substrate [16].Saka et al. have reported rapid growth of Cu nanowhiskers onevaporated polycrystalline Cu films with the aid of local stressgradients in the film [17]. Cheng et al. have proposed a methodof fabricating Bi nanowires from the surfaces of sputtered Bi-CrN composite thin films, where the driving force for the Binanowire formation was reported to be the high-compressiveresidual stress in these composite films [18]. The spontaneousformation of SM-whiskers from the surfaces of SM containingternary carbide and nitride bulk samples has been observed [19].The whisker growth was attributed to the compressive stressesoriginating within the samples due to the volume expansionassociated with the oxidation of SM.

Recently, rapid spontaneous Sn-whisker growth has beenreported in Sn-solder alloys containing rare earth (RE) ele-ments [20]–[24]. The addition of RE elements was argued tolead to the formation of RESn3 intermetallic phases, such asLaSn3 , CeSn3 , and YSn3 within Sn matrix phases. The drivingforce for rapid Sn-whisker growth was considered to be re-lated to the compressive stresses developed due to preferentialoxidation of RE elements in RESn3 phases under atmosphereexposure. The preferential oxidation of RE elements of RESn3phases resulted in the formation of RE element rich oxide lay-ers by leaving pure Sn. Such a reaction is accompanied by largevolume expansions, and hence, compressive stresses built up ina confined volume, which eventually acted as the driving forcefor extrusion of Sn-whiskers. Similarly, rapid spontaneous SM-whisker growth has been reported in bulk compounds of NdSn3 ,NdIn3 , and LaPb3 [25], [26]. The formation of RE(OH)3 com-pounds was identified after exposure to atmosphere instead ofsimple RE-oxides.

1536-125X/$26.00 © 2011 IEEE

TAKAHASHI et al.: SPONTANEOUS FORMATION OF IN-WHISKERS ON YIn3 THIN FILMS 1203

The growth rates of Sn-whiskers from RESn3 were reportedto be∼10 A/s [22], [26], which are two to three orders of magni-tude higher than those observed for Sn-whisker extrusion duringSn plating applications (e.g., 0.2–0.4 A/s [27], [28]). Moreover,those whiskers can be formed on the sample surface sponta-neously just upon exposure to atmosphere. This is attractivefrom a view point of submicrometer/nanostructure engineeringalthough the SM-whisker formation has commonly been consid-ered to be detrimental because of the potential risk for electricalshort cut, and hence, damaging components. The stress-inducedwhisker extrusion from RESM3 compounds could potentiallybe utilized as a novel synthesis route for the rapid fabrication ofself-organized 1-D submicrometer/nanostructured materials.

Among many possible combinations of SM–RE systems, theIn–Y binary system is one of the candidate of material systemsto be studied, since the intermetallic compound YIn3 is ther-modynamically stable [29], and therefore, expected to providerapid spontaneous In-whiskering similar to Sn-whisker forma-tion reported previously from Sn-solder alloys. Although thepossible application of pure In-whiskers seems to be limited inpractice, its oxide In2O3 is known to be a wide bandgap trans-parent semiconductor, and hence, In2O3 nanowires have manypotential applications, such as for gas sensors and ultravioletphotodetectors [30].

Herein, we report the spontaneous growth behavior of In-whiskers from YIn3 thin films deposited by combinatorial mag-netron sputtering. The analysis of morphology and extrusionkinetics along with the Y/In concentration gradient of the com-positionally graded thin films enables an efficient investigationof the composition dependence of whisker extrusion morphol-ogy and kinetics. Only within ∼7 at.% spread in composition,we find the In-whisker diameter ranging from ∼0.2 to a fewmicrometer or even larger. Occasionally, a growth rate of In-whiskers as high as∼1500 A/s is observed, which is even two or-ders of magnitude larger than rapid Sn-whisker formation fromRESn3 phases reported previously [22], [26]. Based on theseexperimental observations, we suggest an alternative synthesispathway for the growth of 1-D submicrometer/nanostructures,where whisker morphology and growth kinetics are controlledby the film composition.

II. EXPERIMENTAL

Compositionally graded In–Y thin films were deposited usingthe combinatorial magnetron sputtering platform schematicallyshown in Fig. 1. Two magnetron cathodes equipped with In andY targets of 39 mm in diameter were placed facing the substrateat an angle of approximately 15◦ with respect to the substratenormal. The target–substrate distance was approximately 6 cm.The deposition chamber was evacuated prior to depositions to abase pressure of∼10−4 Pa using a turbomolecular pump. The Inand Y targets were cosputtered onto a 2 in Si (1 0 0) substrate for30 min at an Ar pressure of 0.35 Pa with dc powers of 25 and50 W, respectively. The depositions were carried out at roomtemperature, namely without intentional substrate heating.

Detailed surface investigations were carried out in a scanningelectron microscope (SEM, JEOL JSM-6480) equipped with an

Fig. 1. Schematic illustration of combinatorial magnetron sputtering setup fordeposition of In–Y thin films with composition gradient.

Fig. 2. Photograph of In–Y thin film with lateral composition gradient de-posited on 2 in Si-wafer after exposure to atmosphere for 30 h.

energy dispersive X-ray analyzer (EDX, EDAX Genesis 2000).The investigations were focused on a film surface segment in-cluding compositions close to YIn3 at which an appreciableamount of whiskers was observed. For the EDX analysis, In-L,Y-L, and O-K characteristic X-ray lines were used for the iden-tification of In, Y, and O, respectively. In–Y compositions werequantified using the ZAF method. In addition to the morphologi-cal and compositional studies, structural analysis was performedby micro X-ray diffraction (XRD) using a general area diffrac-tion detector system (GADDS, Bruker D8) with a collimatedX-ray beam (Cu-Kα ) using a pin-hole collimator with 0.5 mmin diameter. The voltage and current settings were 40 kV and40 mA, respectively. The incident angle of the X-ray beam wasfixed at 15◦.

III. RESULTS AND DISCUSSION

After exposure of the as-deposited films to atmosphere, In-whiskers were found to form spontaneously on certain filmsurface segments. Fig. 2 shows a photograph of an In–Y filmkept in atmosphere for 30 h at room temperature. One can seea variation in film surface appearance along its compositiongradient. As it will be shown later, this is primarily due to avariation of surface topographical features including differentsize and morphology of In-whisker formation.

Fig. 3(a) shows a SEM micrograph obtained from a se-lected area of the film shown in Fig. 2. SEM micrographs of


Fig. 3. SEM micrographs showing a variety of In-whisker morphologies ob-tained from the marked area in Fig. 2: (a) large area, low-magnification image,and (b)–(g) high-magnification images taken from positions denoted by (b)–(g)in Fig. 3(a).

Fig. 3(b)–(g) were obtained from positions denoted by (b)–(g)in Fig. 3(a). After exposure of the film to atmosphere, whiskersare found to grow spontaneously on the film surface investigated.According to EDX measurements, no evidence for the presenceof Y in the whiskers could be detected. The morphology, size,and population of In-whiskers appear to be largely dependent onthe position along the In–Y concentration gradient, and hence,on the as-deposited film composition. The In-whisker diame-ter observed ranges from ∼0.2 μm to a few micrometer, andoccasionally even larger. It should be noted that such a smallsize of whiskers with the diameter of, in particular, <1 μm isnot common for typical spontaneous Sn-whiskers observed inplating. Generally speaking, thin In-whiskers with the diameterof <1 μm with large population are observed at the film com-positions close to stoichiometric YIn3 , while thicker whiskerswith smaller population are primarily located at the In-rich side.When the Y composition is reduced to approximately 18 at.%(balance to 100 at.% is In), no whisker formation is observed,instead, hillock or nodule-like structures appear. Thus, an ap-preciable amount of spontaneous In-whiskering is confined to anarrow concentration range of 18–25 at.% Y.

To determine the cause for In-whisker formation, film crosssections were studied. Fig. 4 provides cross-sectional SEM im-ages obtained from a segment of a film exhibiting the growthof In-whiskers after exposure to atmosphere. The substrate was

Fig. 4. SEM micrographs showing cross-sectional features of a film withcomposition gradient: (a) low-magnification image (left side corresponds toIn-rich area), high-magnification images of a top (b), and a bottom (c) part ofIn-whiskers grown. The micrographs were taken after one month exposure toatmosphere.

cleaved along the composition gradient. Fig. 4(b) and (c) dis-plays whiskers in the upper and lower region of the frame shownin Fig. 4(a), respectively. The growth of In-whiskers with thediameter of a few hundreds nanometers can be seen in Fig. 4(b).Fig. 4(c) shows cross-sectional as well as surface features of theIn–Y film after In-whisker formation. In-whiskers are connectedto the film surface at their roots. In addition, compared with alarge tangle of In-whiskers seen from a top view [see Fig. 4(b)],the actual population of roots at the film surface appears to berelatively small.

The spontaneous growth mechanism of In-whiskers fromYIn3 thin films presented here is believed to be similar tothe one proposed previously for rapid Sn-whisker formationfrom RESn3 compounds [22], [24]. Namely, the growth of In-whiskers is driven by the preferential reaction of Y with atmo-sphere to form Y oxides and/or Y hydroxides. In the case of Y,the following reactions might occur upon exposure

4YIn3 + 3O2 → 2Y2O3 + 12In (1)


Fig. 5. Series of SEM micrographs showing temporal growth behavior of In-whisker with different atmosphere exposure time: (a) as-deposited, (b) after 3min, (c) 6 min, (d) 10 min, (e) 20 min, and (f) 80 min.

2YIn3 + 6H2O → 2Y(OH)3 + 6In + 3H2 . (2)

These reactions are energetically preferred because of thelarge exothermic standard enthalpies of formation for Y2O3(−1919.4 kJ/mol [31]) and Y(OH)3 (–1472.3 kJ/mol [32]) com-pared to that for YIn3 (−41.8 kJ/mol [29]). While the formationof Y2O3 and/or Y(OH)3 provides free In atoms, a large vol-ume expansion occurs. Using molar volumes of In (1.57 × 10-5

m3 /mol), YIn3 (5.83 × 10-5 m3 /mol [33]), Y2O3 (4.49 × 10-5

m3 /mol [34]), and Y(OH)3 (3.62 × 10-5 m3 /mol [35]), one canestimate possible volume changes of +19.3% and +42.9% forthe formation of Y2O3 and Y(OH)3 , respectively. These largeexpansive volume strain values cause compressive stresses inthe film, which eventually act as the major driving force forIn-whisker extrusion.

It is also interesting to notice in Fig. 4(c) that a layer ofIn is formed at the film–substrate interface. This is expected tohappen if the In extrusion force is larger than the adhesive forcesbetween the film and the substrate.

To further verify the hypothesis of atmosphere exposure in-duced In-whisker growth, we have studied the temporal growthbehavior of In-whiskers during exposure. Fig. 5 illustrates a se-ries of SEM micrographs showing an initial stage of In-whiskergrowth. In order to capture a surface condition as close as possi-ble to as-deposited state, an In–Y film was transferred from thedeposition chamber to the SEM chamber within < 5 min expo-sure time. After the first micrograph was captured at a selectedposition on the film surface as shown in Fig. 5(a), the SEM cham-ber was vented, thus intentionally exposing the film surface toatmosphere for a certain time. Then, the chamber was evacuatedagain, and a subsequent micrograph was acquired at the sameposition as the previous one. This exposure-observation processwas repeated several times to reveal an initial stage of whiskergrowth. The exposure time given in each micrograph in the fig-ures represents the total exposure time measured from the firstventing process.

It is obvious from the images shown in Fig. 5 that In-whiskersare extruded spontaneously from the film surface upon atmo-sphere exposure. According to EDX measurement performedon the smooth surface shown in Fig. 5(a), the film compositionwas determined to be about 22 at.% Y. After an incubation timeof ∼3 min, the whisker nucleation occurs, and they grow con-tinuously as the exposure time is increased. Based on a roughmeasurement of the change in whisker length with time, the

Fig. 6. Temporal structural evolution on film surface with different atmosphereexposure time: SEM micrographs for (a) as-deposited, (b) after ∼2 h, (c) after∼52 h, and (d) corresponding XRD patterns.

whisker growth rate is estimated to be as high as ∼1500 A/scorresponding to the growth of a 30-μm-long whisker in 3 min,according to Fig. 5(c). This is even two orders of magnitudelarger than those observed for bulk RESM3 compounds (∼10A/s) [22], [26]. The reason for this rather extensive differencein growth rate between YIn3 thin films and bulk RESM3 com-pounds is unknown at this moment. As we will discuss later,the difference in microstructures between thin films and bulkmaterials may be responsible.

It is also of great importance to mention here that in highvacuum no measurable dimensional change of In-whiskers wasobserved when the film was kept in the SEM chamber for 24h. This suggests that, for In-whisker growth to occur, a reactionbetween the YIn3 films and atmosphere is essential as suggestedpreviously for SM-whisker formation from RESM3 compounds[20]–[23].

Fig. 6 shows a series of SEM micrographs [see Fig. 6(a)–(c)]and XRD patterns [see Fig. 6(d)] upon exposure to atmosphereobtained at the film composition of about 21 at.% Y. The colli-mated X-ray results in an irradiated surface area of ellipse withthe transverse diameter of 2.5 mm in the present measurementsetup. In order to minimize a concentration spread within theirradiated area, the measurement was carried out in such a waythat the transverse direction of the ellipse was perpendicularto the concentration gradient in the film. Hence, the positionfor the SEM observations and XRD measurements are almostidentical. Although the XRD measurements were carried outunder exposure, the use of the GADDS system with an arealdetector enabled fast acquisition of diffraction patterns. The ac-quisition time for patterns was only 2 min, which minimizedpossible changes of the diffraction patterns due to In-whiskergrowth during measurements. The XRD pattern for as-depositedstate, i.e., total exposure time < 15 min, indicates that the filmstructure is single-phase cubic L12-YIn3 [see Fig. 6(d)]. Hence,the formation of metastable cubic fcc Y1−x In3 phase with aslight Y deficiency appears likely. The formation of metastable


Fig. 7. SEM micrograph showing area of retardation of In-whisker growthafter being repeatedly exposed to an electron beam.

phases is commonly observed during low-temperature thin-filmgrowth [36]. In-whiskers grow continuously upon further expo-sure [see Fig. 6(b) and (c)] and correspondingly diffraction peaksoriginating from the tetragonal In phase appear [see Fig. 6(d)].Although we expected the formation of Y2O3 and/or Y(OH)3in the film after exposure, the presence of those compoundswas not detected by XRD. This may be due to the formationof amorphous oxides and/or hydroxides due to the fact that thereactions take place at room temperature.

The importance of reactions between the In–Y film surfaceand atmosphere for In-whisker growth is emphasized by a SEMmicrograph shown in Fig. 7. A frame at the center representsan area, which has been investigated repeatedly by SEM forstudying In-whisker growth process. It is obvious that less In-whiskers grow on the surface after being observed by SEM thanin the close vicinity thereof. This indicates that electron beamirradiation of the film surface during observation retards the In-whisker growth kinetics. Xian and Liu have also reported thatthe spontaneous Sn-whisker growth from NdSn3 compoundswas stopped after being observed in SEM [26]. It is commonthat carbon contamination thin films are formed on the samplesurface being investigated by SEM due to the interaction ofthe electron beam with the absorbed surface layers [37]. As apossible reason for retardation of In-whisker growth, as shownin Fig. 7; therefore, we suggest that a carbon layer forms on thesurface after being observed repeatedly by SEM, and acts as aprotective layer against reactions with atmosphere.

To further study possible reactions with atmosphere uponexposure to air and the correlation to In-whisker formation,temporal changes in film composition upon exposure have beeninvestigated as well. Fig. 8(a) and (b) shows the changes infilm composition as a function of exposure time at two differ-ent as-deposited compositions: 1) 22 and 2) 25 at.% Y. Thecorresponding SEM micrographs after each exposure time aregiven in Fig. 8(c)–(n). However, the investigation positions wereslightly shifted intentionally after each exposure to exclude theinfluence of electron beam radiation as discussed earlier. TheEDX analyzes were carried out using a spot-mode focused onlyon the film surface to avoid signal from the In-whiskers. Sincethe quantification of oxygen content using EDX is considered tobe less reliable without using a suitable standard sample, only

Fig. 8. Temporal changes of film compositions and In-whisker growth withdifferent atmosphere exposure time for two different as-deposited films with22 at.% Y, (a) EDX result, (c)–(h) SEM micrographs, and 25 at.% Y (b) EDXresult, (i)–(n) SEM micrographs.

a relative change in intensity of O-K characteristic X-ray peakwas used as an indicator for oxygen content in the film.

It should be noted that there is a large difference in mor-phology and extrusion kinetics of In-whiskers at two differentas-deposited film compositions of 22 and 25 at.% Y [see Fig. 8].In both cases, the oxygen content in the film increases as theexposure time is increased. On the other hand, the In content de-creases as In-whiskers grow upon exposure. This indicates thatthe diffusion of In atoms to the roots of growing In-whiskersoccurs upon exposure. Furthermore, as can be seen for 22 at.%Y, the nucleation and subsequent growth of In-whiskers seem tobe correlated with the onset of the increase in oxygen contentin the film. This supports the hypothesis that the formation ofY2O3 and/or Y(OH)3 is involved for the growth of In-whiskersupon exposure of In–Y thin films.

One of the most interesting phenomena observed in thepresent In–Y thin film study is that the morphology and extru-sion kinetics of In-whiskers are strongly affected by the chemicalcomposition, even within the narrow composition range of 18–25 at.% Y. This range could be related to a possible homogeneityrange for the formation of single-phase Y1−x In3 with an Y de-ficiency. However, a more systematic study for the correlationbetween the composition, phase constitution, and morphologyin as-deposited In–Y thin films is necessary for better under-standing. As for the size of whiskers, Tu et al. have proposedthat radius of Sn-whisker R is determined by a balance betweensurface energy per unit area γ and strain energy per unit volumeε as following [15]:

R =2γ

ε. (3)

Thus, qualitatively speaking, thin (thick) In-whiskers are ex-pected to form when the strain energy, hence, compressivestress, is large (small). It might be the case that the change in Y


Fig. 9. SEM micrograph of uniform, dense In-whisker structure forming on afilm surface. This micrograph was taken after three days exposure.

composition causes the change in reactivity between In–Y thinfilms and atmosphere, and hence, the change in strain energystored in the film. A lower strain energy should be associatedwith a lower Y content in the film simply because of the smallervolume of Y2O3 and/or Y(OH)3 which could be formed. Thisis consistent with the here reported observation that thickerIn-whiskers are formed at smaller Y concentrations than thenominal YIn3 . It should also be pointed out that the In-whiskergrowth behavior may be influenced not only by the chemicalcomposition, but also by the constitution and microstructure ofthe film. Sputtered thin films are known to exhibit quite differ-ent microstructural features compared to bulk materials [36].For instance, the grain size of sputtered thin films is often in therange of 10–100 nm, which is one or two orders of magnitudesmaller than those of typical bulk polycrystalline materials. Be-cause of the large grain boundary density, the film reactivity toatmosphere is expected to be enhanced due to grain boundarydiffusion as compared to bulk samples. The very large whiskergrowth rate, ∼1500 A/s, obtained in the present thin film studymay be understood in terms of an accelerated reaction due tofine grain structures of the thin films.

Finally, as already mentioned earlier, the use of composition-ally graded In–Y thin films enables the efficient investigationof interesting morphological In-whisker features. For instance,in Fig. 9, a cross-sectional feature of a film exhibiting uniquelyself-organized In-whiskers is presented. The thin In-whiskerswith the diameter of ∼0.2 μm are well aligned, highly denselyin a large area along the surface normal, and the length of eachwhisker is very similar and about 10 μm. This suggests that thewhiskers nucleate almost simultaneously. This type of whiskermorphology may be useful for application, where a large surfaceto volume ratio is required.

IV. CONCLUSION

In this study, we have presented the spontaneous formationof In-whiskers from YIn3 thin films deposited by combinatorialmagnetron sputtering. In-whiskers were found to be extrudedspontaneously from the roots only upon exposure of the films toatmosphere, but not in vacuum. In-whisker growth is accompa-nied by an increase in oxygen content in the films. It is suggested

that the In-whiskers are extruded from the film surface due tocompressive stresses developed by preferential reactions of Yto form more stable compounds, such as Y2O3 .

We also found that the morphology and extrusion kinetics ofIn-whiskers were strongly dependent on the Y composition ofas-deposited films. The diameter of the In-whiskers observedranges from ∼0.2 μm up to a few micrometer or even largerwithin a composition range between 18 and 25 at.% Y. Basedon these data, the whisker morphology and growth kinetics canbe controlled through tuning the film composition. The resultsprovide an alternative pathway toward the controlled growth of1-D submicrometer/nanostructured materials.

ACKNOWLEDGMENT

The U.S. Government is authorized to reproduce and dis-tribute reprints for Governmental purposes notwithstanding anycopyright notation thereon.

REFERENCES

[1] C. M. Lieber and Z. L. Wang, “Functional nanowires,” MRS Bull., vol. 32,pp. 99–108, 2007.

[2] Y. Xia, P. Yang, Y. Sun, Y. Wu, B. Mayers, B. Gates, Y. Yin, F. Kim,and H. Yan, “One-dimensional nanostructures: Synthesis, characteriza-tion, and applications,” Adv. Mater., vol. 15, pp. 353–389, 2003.

[3] R. S. Wagner and W. C. Ellis, “Vapor-liquid-solid mechanism of singlecrystal growth,” Appl. Phys. Lett., vol. 4, pp. 89–90, 1964.

[4] J. D. Holmes, K. P. Johnston, R. C. Doty, and B. A. Korgel, “Control ofthickness and orientation of solution-grown silicon nanowires,” Science,vol. 287, pp. 1471–1473, 2000.

[5] T. J. Trentler, K. M. Hickman, S. C. Goel, A. M. Viano, P. C. Gibbons,and W. E. Buhro, “Solution-liquid-solid growth of crystalline III–V semi-conductors: An analogy to vapor-liquid-solid growth,” Science, vol. 270,pp. 1791–1794, 1995.

[6] Y. F. Zhang, Y. H. Tang, N. Wang, D. P. Yu, C. S. Lee, I. Bello, and S.T. Lee, “Silicon nanowires prepared by laser ablation at high temperature,”Appl. Phys. Lett., vol. 72, pp. 1835–1837, 1998.

[7] A. M. Morales and C. M. Lieber, “A laser ablation method for the synthesisof crystalline semiconductor nanowires,” Science, vol. 279, pp. 208–211,1998.

[8] B. D. Yao, Y. F. Chan, and N. Wang, “Formation of ZnO nanostructuresby a simple way of thermal evaporation,” Appl. Phys. Lett., vol. 81,pp. 757–759, 2002.

[9] Y. C. Kong, D. P. Yu, B. Zhang, W. Fang, and S. Q. Feng, “Ultraviolet-emitting ZnO nanowires synthesized by a physical vapor deposition ap-proach,” Appl. Phys. Lett., vol. 78, pp. 407–409, 2001.

[10] U. Lindborg, “Model for the spontaneous growth of zinc, cadmium andtin whiskers,” Acta Metall., vol. 24, pp. 181–186, 1976.

[11] U. Lindborg, “Observations on the growth of whisker crystals from zincelectroplate,” Metall. Mater. Trans. A, vol. 6, pp. 1581–1586, 1975.

[12] P. L. Key, “Surface morphology of whisker crystals of tin, zinc and cad-mium,,” in Proc. 20th Electron. Components Conf., 1970, pp. 155–160.

[13] R. M. Fisher, L. S. Darken, and K. G. Carroll, “Accelerated growth of tinwhiskers,” Acta Metall., vol. 2, pp. 368–373, 1954.

[14] B. Z. Lee and D. N. Lee, “Spontaneous growth mechanism of tinwhiskers,” Acta Mater., vol. 46, pp. 3701–3714, 1998.

[15] K. N. Tu, C. Chen, and A. T. Wu, “Stress analysis of spontaneous Snwhisker growth,” J. Mater. Sci.: Mater. Electron., vol. 18, pp. 269–281,2007.

[16] W. Shim, J. Ham, K. Lee, W. Y. Jeung, M. Johnson, and W. Lee, “On-filmformation of Bi nanowires with extraordinary electron mobility,” NanoLett., vol. 9, pp. 18–22, 2009.

[17] M. Saka, F. Yamaya, and H. Tohmyoh, “Rapid and mass growth of stress-induced nanowhiskers on the surfaces of evaporated polycrystalline Cufilms,” Scr. Mater., vol. 56, pp. 1031–1034, 2007.

[18] Y. T. Cheng, A. M. Weiner, C. A. Wong, M. P. Balogh, and M. J. Lukitsch,“Stress-induced growth of bismuth nanowires,” Appl. Phys. Lett., vol. 81,pp. 3248–3250, 2002.


[19] M. W. Barsoum, E. N. Hoffman, R. D. Doherty, S. Gupta, andA. Zavaliangos, “Driving force and mechanism for spontaneous metalwhisker formation,” Phys. Rev. Lett., vol. 93, pp. 206104-1–206104-4,2004.

[20] T. H. Chuang, “Rapid whisker growth on the surface of Sn-3Ag-0.5Cu-1.0Ce solder joints,” Scr. Mater., vol. 55, pp. 983–986, 2006.

[21] T. H. Chuang, C. C. Chi, and H. J. Lin, “Formation of whiskers and hillockson the surface of Sn-6.6RE alloys,” Metall. Mater. Trans. A, vol. 39 A,pp. 604–612, 2008.

[22] T. H. Chuang and S. F. Yen, “Abnormal growth of tin whiskers in aSn3Ag0.5Cu0.5Ce solder ball grid array package,” J. Electron. Mater.,vol. 35, pp. 1621–1627, 2006.

[23] M. A. Dudek and N. Chawla, “Oxidation behavior of rare-earth-containingPb-free solders,” J. Electron. Mater., vol. 38, pp. 210–220, 2009.

[24] M. A. Dudek and N. Chawla, “Mechanisms for Sn whisker growth in rareearth-containing Pb-free solders,” Acta Mater., vol. 57, pp. 4588–4599,2009.

[25] M. Liu and A. P. Xian, “Spontaneous growth of whiskers on RE-bearingintermetallic compounds of Sn-RE, In-RE, and Pb-RE,” J. Alloys Compd.,vol. 486, pp. 590–596, 2009.

[26] A. P. Xian and M. Liu, “Observations of continuous tin whisker growth inNdSn3 intermetallic compound,” J. Mater. Res., vol. 24, pp. 2775–2783,2009.

[27] K. N. Tu, “Interdiffusion and reaction in bimetallic Cu-Sn thin films,”Acta Metall., vol. 21, pp. 347–354, 1973.

[28] V. K. Glazunova and K. M. Gorbunova, “Spontaneous growth of whiskersfrom electrodeposited coatings,” J. Cryst. Growth, vol. 10, pp. 85–90,1971.

[29] A. Palenzona and S. Cirafici, “Thermodynamic properties of Y, Th, andU, MX3 compounds with III-A and IV-A elements,” Thermochim. Acta,vol. 13, pp. 357–360, 1975.

[30] G. Cheng, E. Stern, S. Guthrie, M. A. Reed, R. Klie, Y. Hao, G. Meng,and L. Zhang, “Indium oxide nanostructures,” Appl. Phys. A: Mater. Sci.Process., vol. 85, pp. 233–240, 2006.

[31] E. G. Lavut and N. V. Chelovskaya, “Enthalpy of formation of diyttriumtrioxide,” J. Chem. Thermodyn., vol. 22, pp. 817–820, 1990.

[32] I. I. Diakonov, K. V. Ragnarsdottir, and B. R. Tagirov, “Standard thermo-dynamic properties and heat capacity equations of rare earth hydroxides:II. Ce(III)-, Pr-, Sm-, Eu(III)-, Gd-, Tb-, Dy-, Ho-, Er-, Tm-, Yb-, and Y-hydroxides. Comparison of thermochemical and solubility data,” Chem.Geol., vol. 151, pp. 327–347, 1998.

[33] K. H. J. Buschow, H. W. de Wijn, and A. M. van Diepen, “Magneticsusceptibilities of rare-earth-indium compounds : RIn3 ,” J. Chem. Phys.,vol. 50, pp. 137–141, 1969.

[34] L. Smrcok, “Rietveld refinement of Y2 O3 using the Pearson VII profileshape function,” Cryst. Res. Technol., vol. 24, pp. 607–611, 1989.

[35] G. W. Beall, W. O. Milligan, and H. A. Wolcott, “Structural trends inthe lanthanide trihydroxides,” J. Inorg. Nucl. Chem., vol. 39, pp. 65–70,1977.

[36] J. A. Thornton, “High rate thick film growth,” Annu. Rev. Mater. Sci.,vol. 7, pp. 239–260, 1977.

[37] W. A. Knox, “Contamination formed around a very narrow electron beam,”Ultramicroscopy, vol. 1, pp. 175–180, 1976.

Authors’ photographs and biographies not available at the time of publication.

1202 ieee transactions on nanotechnology, vol. 10, no. … · 1202 ieee transactions on...

Documents