Dissolution Kinetics of Oxidative Etching of Cubic and Icosahedral

Feb 10, 2017 - We report here the application of in situ liquid transmission electron microscopy for quantitative analysis of oxidative etching of cub...
1 downloads 3 Views 8MB Size
Download PDF
Dissolution Kinetics of Oxidative Etching of Cubic and Icosahedral Platinum Nanoparticles Revealed by in Situ Liquid Transmission Electron Microscopy Jianbo Wu,†,§,∥,‡,⊥ Wenpei Gao,†,∥,‡ Hong Yang,*,§ and Jian-Min Zuo*,†,∥ †

Department of Materials Science and Engineering, University of Illinois at Urbana−Champaign, 1304 West Green Street, Urbana, Illinois 61801, United States ∥ Frederick Seitz Materials Research Laboratory, University of Illinois at Urbana−Champaign, 104 South Goodwin Avenue, Urbana, Illinois 61801, United States § Department of Chemical and Biomolecular Engineering, University of Illinois at Urbana−Champaign, MC-712, 600 South Mathews Avenue, Urbana, Illinois 61801, United States S Supporting Information *

ABSTRACT: Dissolution due to atom-level etching is a major factor for the degradation of Pt-based electrocatalysts used in low-temperature polymer electrolyte membrane fuel cells. Selective surface etching is also used to precisely control shapes of nanoparticles. Dissolution kinetics of faceted metal nanoparticles in solution however is poorly understood despite considerable progress in understanding etching of two-dimensional surfaces. We report here the application of in situ liquid transmission electron microscopy for quantitative analysis of oxidative etching of cubic and icosahedral Pt nanoparticles. The experiment was carried out using a liquid flow cell containing aqueous HAuCl4 solution. The data show that oxidative etching of these faceted nanocrystals depends on the location of atoms on the surface, which evolves with time. A quantitative kinetic model was developed to account for the mass lost in electrolyte solutions over time, showing the dissolutions followed the power law relationship for Pt nanocrystals of different shapes. Dissolution coefficients of different surface sites were obtained based on the models developed in this study. KEYWORDS: corrosion, in situ liquid TEM, oxidative etching, kinetics, platinum

O

nanoparticles, on the other hand, is important for highperformance catalysts, because surface atoms, including those on corners, edges, and facets have different reactivity. Dissolution of atoms from nanoparticles is thus expected to be different from that on extended 2D surfaces. For instance, preferential atomic etching may take place at corners or edges or both. Previous studies on etching of nanoparticles under reaction environments relied mostly on ex situ characterizations, which are not ideal for studying the dissolution dynamics.13,14

xidative etching is a major cause for the degradation of catalyst performance under harsh reaction conditions.1−7 Etching leads to change in surface structure and eventual dissolution of catalyst nanoparticles. The process is also important in the preparation of shape-controlled nanoparticles, where atoms are selectively removed, resulting in certain preferred morphologies.8−10 Understanding the etching mechanism therefore benefits greatly the design of nanoparticles for catalysis. While surface etching has been studied by scanning tunneling microscopy (STM) under ultrahigh-vacuum (UHV) conditions11 and atomic force microscopy (AFM) in liquids,12 those studies typically focus on extended twodimensional (2D) surfaces. Few have reported on etching processes in 3D nanoparticles in solution. The reason is mainly because of the technical challenges to characterize nanostructure surfaces directly in solution. The 3D morphology of metal © XXXX American Chemical Society

Received: November 8, 2016 Accepted: February 10, 2017 Published: February 10, 2017 A

DOI: 10.1021/acsnano.6b07541 ACS Nano XXXX, XXX, XXX−XXX

Article

www.acsnano.org

Article

ACS Nano

cube were etched slowly in the first 10 min, a major change in morphology started on both {100} terraces and corner sites after about 20 min. These changes led to the formation of a truncated and concave cube. Meanwhile, the adjacent icosahedron turned into a pseudospherical shape as the corners became rounded. After 1 h of observation, the cube was dissolved completely, while a small volume remained for the icosahedron (Figure 2). Separately, a control experiment was carried out by using HCl (0.01 M) or H2SO4 (0.01 M) instead of the HAuCl4 solution. Pt nanoparticles were stable in these two acids, which indicates that H3O+ should not be the main chemical species involved in the observed etching of Pt surfaces. Thus, etching should be due to mainly the oxidation by [AuCl4]− ion, which reacted with surface Pt atoms to form Pt2+ through a dissolution process.11 Halogen ions were also shown to promote oxidative etching of Pt metal into [PtCl4]2−.11 To quantify the morphological changes of the Pt nanoparticles based on in situ LTEM data, we systematically analyzed the change in two distances for both the icosahedron and the cube: corner-to-corner distance for icosahedron (Dcorner‑ih), side-to-side distance for icosahedron (Dside‑ih), corner-to-corner distance for cube (Dcorner‑c), and side-to-side distance for cube (Dside‑c) (Figure S2). These distances were obtained from the TEM micrographs recorded at a rate of 1 frame/s. Particles in contact with each other could influence the growth of core−shell nanoparticles because of possible atom diffusion between the two particles.32 This effect is not strong in etching, except that the contact could help protect some of the facets and corners from the etchant species. Thus, all distances were measured from the noncontacted area of the nanoparticles to avoid the possible influence. Figure 3 shows the change in distance for the icosahedral particle. Dcorner‑ih was initially larger than Dside‑ih and then decreased rapidly to become smaller than Dside‑ih after 1800−2800 s. Both Dside‑ih and Dcorner‑ih dropped rapidly afterward. A decrease in Dcorner‑ih and Dside‑ih followed the similar pattern after 3400 s, likely because the icosahedral nanoparticle lost its specific geometry at this point and became pseudospherical. The results show a given nanoparticle underwent different kinds of etching kinetics during the process because of the evolution of surface geometry. Further evidence of etching at the corners of icosahedral Pt nanoparticles was provided by an ex situ experiment using Pt icosahedral nanoparticles in HAuCl4 solution (Figure 4). The average size of as-made Pt icosahedral nanoparticles was about 36 nm for the distance between the two opposite corners (Figure 1). The ex situ TEM micrographs show that the six corners of the icosahedron were etched away after 10 min (Figure 4a). Along the edges, the regions near the corners were etched more extensively than the other areas. After 30 min, the entire icosahedron evolved into an approximate spheroid (Figure 4b). Figure 5 shows the measured distances as a function of time for the Pt cube. Etching at the corners was observable in the first 200 s, after which the cubic nanoparticle was bounded with {111} facets at the corners and {110} facets on the edges. The {100} facets were etched after about 540 s, as indicated by the change in distance (Dcorner‑c and Dside‑c). The truncated cube evolved into a spherical particle after 3100 s and was then etched away rapidly afterwards, as indicated by the large drop in its dimension. The above in situ TEM results indicate that the cubic and icosahedral Pt nanoparticles were etched, involving

Recently, significant progress has been made in in situ liquid transmission electron microscopy (LTEM). Use of liquid cells provides a solution to the study of dynamic events occurring in liquid phase by TEM.15−28 Among them, a considerable amount of work has been devoted to understanding the growth of nanoparticles. However, quantitative analysis of dissolution kinetics is still lacking.19−21,23 For dissolution or etching of noble metal nanoparticles in solution,29−31 it has been shown that, at a dose rate of >280 e−/Å2·s in STEM, the electron probe could act as oxidizing agent, leading to the etching of cubic Pd nanoparticles into a nanospheroid in the presence of Br− ion.29 This electron beam effect was found to assist the dissolution of Au nanorods and nanoparticles in FeCl3solution in a graphene liquid cell.31 Increasing the concentration of FeCl3 was shown to affect the etching kinetics, so the dissolution process became dominate. Here we present a time-resolved in situ LTEM study of oxidative etching of cubic and icosahedral Pt nanoparticles in aqueous solutions using HAuCl4 as the oxidative agent. Using in situ data, we developed a quantitative model accounted for the shape-dependent dissolution processes of Pt nanoparticles based on the Lifshitz−Slyozov−Wagner (LSW) theory.

RESULTS AND DISCUSSION Dissolution of Pt Icosadedron and Nanocube. The assynthesized Pt icosahedron is bounded by {111} facets, while a nanocube is enclosed by its {100} facets (Figure 1a,b). TEM

Figure 1. TEM micrographs of representative (a) icosahedral and (b) cubic Pt nanocrystals used in this study.

micrographs of the ensemble of nanoparticles in the liquid cells are shown in Figure S1. Both the icosahedral and cubic Pt nanoparticles had an edge length of approximately 20 nm. The dissolution kinetics at different surface sites of these two geometries was captured in liquid inside the SiN-based liquid cell at a flow rate of 5 μL/min. The electron dose rate was set at 30 e−/Å2·s. Figure 2 shows the morphological changes of the icosahedral and cubic Pt nanoparticles over a period of 1 h (also see the video in the Supporting Information). The reaction started after the mixture of HAuCl4 (0.1 M) and KCl (0.01 M) aqueous solution was introduced into the liquid cell at a flow rate of 5 μL/min. The video in the Supporting Information shows changes of the two nanoparticles over the whole process. The TEM micrograph shown in Figure 2 (designated as 0 min) was taken ∼10 min after the HAuCl4 aqueous solution was pumped into the liquid flow cell. Dramatic changes in the nanoparticle shapes were observed. Specifically, as dissolution occurred, the corners became rounded, compared with edges of the original Pt cube and icosahedron. Although corners of the B

DOI: 10.1021/acsnano.6b07541 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 2. Morphological changes of icosahedral and cubic nanoparticles of platinum in 1 h in the presence of a mixed solution of HAuCl4 and KCl.

Figure 3. Change in distance associated with the morphological evolution of a Pt icosahedron. Points within the blue oval correspond to the distance for atoms at corner regions. The points within the red oval show the sides and corners started to dissolve at the similar rate, where the icosahedral nanoparticle lost its specific geometry and became spheroid.

Figure 5. Change in distance associated with the morphological evolution of a Pt cube.

etching of the entire nanoparticle took place with a rapid decrease in size. For the cubic Pt nanocrystal, the process was dominated by a relatively slow dissolution of {100} facets and a rapid global dissolution (Figure 5). Under the same experimental condition, other etching sequences were also observed in the in situ LTEM study. In Figure S3, the dissolution started from the center of a facet on one of the icosahedra. Thus, there appeared to be no preferred etching for these nanoparticles in the same batch. Site-Dependent Multimode Dissolution Kinetics. In what follows, we show that the in situ data can be quantitatively predicted, using the LSW theory, for the observed dynamics of shape-dependent multimode dissolutions. In this model, Pt atoms are assumed to react with [AuCl4]− ions, and the concentration of etchant ([AuCl4]−) is well above that of the Pt species throughout the reaction. This solution process is thus reaction limited, similar to the metal deposition on a metal nanoparticle in solution observed previously.18 Both icosahedral and cubic nanoparticles feature different sites on corners, edges, and surfaces, which is 0D, 1D, and 2D in geometry. If etching predominantly occurs in one type of geometric sites, it should exhibit a distinct pattern in terms of size (or volume) as a

Figure 4. TEM micrographs of icosahedral Pt nanocrystals after the etching process under a flow of HAuCl4 aqueous solution for (a) 10 and (b) 30 min, respectively. The images show preferred etching at corner regions.

multiple steps. For the icosahedral Pt nanoparticle, corners were first etched, followed by the {111} facets at a slower rate, resulting in a small plateau, as shown in Figure 3. Finally, C

DOI: 10.1021/acsnano.6b07541 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 6. (a, b) Log and (c, d) linear plots for oxidative etching of icosahedral Pt nanocrystals at different reaction stages: log plots for sites at the (a) corner and (b) side (edge and terrace); linear plots for sites on the corner (inset of (a)) and the (c) edge and (d) terrace. All measurements were based on in situ LTEM data.

function of time for a nanocrystal. For example, if etching happens at corners, the corner−corner distance can be used to account for the change in volume of a nanoparticle; thus dissolution rates can be determined quantitatively. Under this model, the change in the number of Pt atoms in a nanoparticle resulting from etching can be calculated according to the following equation: dVNP dN = dt vPt ·dt

r 3 − r0 3 = −

(1)

where ae is the dissolution rate coefficient of atoms on the edges, lPt is the specific length of single atom along the edge. For the terrace atoms on (111) surface of an icosahedron, the change can be obtained: r − r0 = −

(2)

(3)

where Vico and r are the volume and edge length of the icosahedron, respectively. By combining eqs 1, 2, and 3, the dissolution rate at corner can be determined using the following equation: dr 6 2 vPt =− ac dt 5 r2

6 vPt at t 2 sPt

(7)

where at is the dissolution rate coefficient of Pt atoms on {111} facets, and sPt is the specific area of single Pt atom on the {111} facet. Similar analysis can be applied to analyze the dissolution of Pt cubes. The detailed derivation for dissolution in each site (corner, edge and facet) for both icosahedron and cube can be found in the Supporting Information. The above rate equations suggest that the etching of shapedefined nanoparticles follows the scaling laws, depending on the dominant etching sites, whether it includes corner (0D), edge (1D), and terrace (2D), according to the following equations:

where Nc is the number of corners of a nanoparticle and ac is the rate coefficient at these corners. For an icosahedron, there are 12 corners, and ⎛ 2 3⎞ dVico = d⎜20· ·r ⎟ = 5 2 ·r 2·dr ⎝ 12 ⎠

(5)

where r is the measured edge length of the icosahedron and r0 is the starting edge length. Similarly, dissolution along the edges of an icosahedron can be obtained using the following equation: v r 2 − r0 2 = −6 2 Pt aet lPt (6)

where N is the total number of atoms, VNP is the volume of the nanoparticle, vpt is the specific volume of single atom. There are three different types of surface sites in a nanoparticle, namely, corner (0D), edge (1D), and terrace (2D). For etching occurring predominantly at the corners of a nanoparticle, the dissolution rate can be written as

dN = −Ncac dt

18 2 vPtact 5

r 3 − r0 3 = −N0A 0t ,

atom around corner (0D)

(8)

r 2 − r0 2 = −N1A1t ,

atom around edge (1D)

(9)

r − r0 = −N2A 2 t ,

(4)

atom on terrace (2D)

(10)

where r is the dimension of a nanocrystal, An is the shapedependent etching coefficient, and Nn is the parameter dependent on the dimension of the etching site, where n is 0

By integrating the eq 4, we obtain the following relationship between the size of a particle and etching time: D

DOI: 10.1021/acsnano.6b07541 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 7. Analysis of etching modes near different sites of a Pt cube in the flow of HAuCl4 aqueous solution: (a−c) Dedge and (d−f) Dterrace. Dissolution near the edge sites was most pronounced in the blue line region (1800−2500 s) and the terrace in the green line region (2800− 3400 s).

to be 1233 Å3/s at the corners of icosahedron based on eq 5. After ∼2880 s, the slope changed to about 1 (green regions in Figure 6a), which corresponds to 2D dissolution. The change in slope indicates the dissolution around corner regions of the Pt icosahedron started from point etching, followed by the terrace between 2880 and 3500 s. In 2D projection, a side of icosahedron can be either the edge or terrace depending on the angle between electron beam and the orientation of nanoparticle. Two types of Dside‑ih were observed in this sample, representing etching at the different regions (Figure 6): one highlighted in blue color had a slope of 1/2, that is, a 1D etching along the edges, and the other in green color had a slope of 1, a 2D etching. To be more specific, the change in Dside‑ih‑1 was along one edge, and the change in Dside‑ih‑2 was on the (111) surface (Figures 6c,d and S4). The dissolution rate coefficient was calculated to be 4.53 Å2/s along the edge and 0.0390 Å/s on the (111) facets based on the 1D and 2D dissolution equations, respectively (eqs S6 and S9). The cubic Pt nanocrystal used in this study did not have sharp corners (Figure 1b). After about 1800 s, the slope of log plot was ∼1/2 for Dcorner, indicating the edge etching became dominant (Figure 7a−c). D2 could be fitted roughly linearly as a function of time between 1800 and 2500 s, further indicating the atom dissolution from edges was pronounced (Figure 7c, eq 8). The etching then occurred predominantly on {100} facets from ∼2800 s, where the slope was close to 1 in the log plot (Figure 7a). The side-to-side distance changed roughly linearly with time between 2800 and 3400 s (Figure 7f). After the

for corner, 1 for edge, and 2 for terrace sites. Based on these power law equations, the relationship between log r and t should be linear in a log plot, where the slope m is equal to 1, 1/2, and 1/3, respectively. Thus, slope of the log plots can be used to determine the dominant etching mechanism. A slope equal or close to 1/3 corresponds to corner etching, which is dominated by the reaction occurring at the corner sites, and 1/ 2 for edge sites, and 1 for facet or terrace sites. Both cube and icosahedron dissolved uniformly in all directions after they became spherical at the late stages. The change in size over time should then follow the scaling law: r − r0 = −NsA st

(11)

where As is the etching coefficient of a spherical surface, and Ns is the geometric parameter. Dissolution Rate and Rate Coefficient. The above models were used to analyze the dominated dissolution at different sites based on D−t plot. Figure 6 shows the analysis of oxidative etching at different stages during the dissolution of icosahedral Pt nanocrystals using both log and linear plots. In Figure 6a, the change of Dcorner was first plotted in logarithmic scale, with the inset showing the same data in linear scale. These in situ data show the corner regions were etched starting from 1500 s (the starting point of the red region in Figure 6a). The slope in the log plot was close to 1/3 for the region between 1500 and 2880 s (the red line in Figure 6a), suggesting that the corresponding process follows the point etching or dissolution (0D). The dissolution rate coefficient is estimated E

DOI: 10.1021/acsnano.6b07541 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Our data analysis shows Cl− and [AuCl4]− ions prefer to react with high-energy, low-coordination sites at the corner region of Pt icosahedron,9,34,35 so atoms located at these sites dissolve the fastest. Pt icosahedron becomes truncated with convex corners and eventually turns into a spherical particle. For the Pt cube, Cl− and [AuCl4]− ions reacted with edges preferentially, followed by atoms on the terraces.

dissolution of terrace atoms at 3100 s, Pt cube became spherical nanoparticles (Figure 5). When faceted nanoparticles evolved into different geometries, the corresponding dissolutions could occur preferably at the corner, along the edge, and on the terrace. Atoms located at these different surface sites were etched predominantly at the different stages. Dissolution coefficients were calculated for the etched nanostructures based on the linear fitting of in situ LTEM data (see the Supporting Information for detail and Figure S4, Tables S1 and S2). Table 1 summarizes the

CONCLUSIONS In summary, we developed a quantitative model to analyze the dissolution kinetics in a solution of Pt nanocrystals based on the scaling law relationship using in situ LTEM data. Previously, kinetic studies of dissolution of nanoparticles were almost exclusively based on the sphere model, in which the difference among corner, edge, and terrace atoms could not be considered. In situ LTEM results show that the dissolution of cubic and icosahedral Pt nanoparticles involves several steps and that our simple scaling law can be used to identify the dominant mode, such as dissolution of corner or edge atoms. The dissolution rate coefficients can be obtained for atoms at different sites by fitting the in situ LTEM data. Thus, this power law model makes it possible for examining all the major structural features including size, corner, edge, and terrace surface sites dynamically at the subnanometer level. The structural subtlety on dissolution, such as effect of twinning, can also be studied based on this approach.

Table 1. Dissolution Rate Coefficients on Various Pt Surface Sites in the Presence of Cl− Ion shape

site

icosahedron

corner edge facet {111} edge facet {110} facet {100}

cube

rate coefficient (atom s−1) 16.077 0.098 0.014 0.651 0.024 0.016

± ± ± ± ± ±

0.9 0.02 0.001 0.1 0.003 0.003

dissolution rate coefficients for Pt atoms at different surface sites. For the corner sites of Pt icosahedron, the dissolution coefficient was ∼16 atom s−1 at the corner region (0D), which is about 2 orders of magnitude of that around the edges (∼0.1 atom s−1) (1D). For the terrace sites (2D), Pt atoms had an estimated dissolution coefficient of 0.014 atom s−1, which is the slowest among the three regions. Among the flat surfaces, in situ LTEM data show facet selectivity in the order of a{110} (cube) > a{100} (cube) > a{111} (icosahedron) in the presence of Cl− ion. This order is consistent with the surface energy and anion adsorption on Pt, which follow the order of γ{110} > γ{100} > γ{111}.33 However, the etching rate on Pt {111} facets is close to the rate on Pt {100} facets, one possible reason could be Pt {111} facets of icosahedron experiencing tensile strain, thus allowing halogen ions to attack the Pt surface more easily than that without the stress. Dissolution of Pt atoms on {110} and {100} facets of corner-rounded cubic crystal appeared to follow the order of surface energy, though the difference was small and other factors such as capping ligand might play important roles as well.11 Figure 8 summarizes the modes for halogen-mediated oxidative etching of icosahedral and cubic Pt nanoparticles.

METHODS Synthesis of Pt Icosahedron. All chemicals were purchased from Aldrich and used as received except those specified otherwise. The synthesis is based on the GRAIL method reported previously.36−38 In a typical procedure, 20 mg of platinum acetylacetonate (Pt(acac)2, Strem Chemicals, purity: 98%), 1 mL of dodecylamine (DDA, 98%), and 50 μL of oleic acid (OA, 90%) were mixed and preheated at 160 °C to make the precursor solution. A mixture of 14 mg of Y(acac)3· H2O (Strem Chemicals, 99.9%), 1 mL of DPE (diphenyl ether, 99%), and 9 mL of DDA was degassed in a 25 mL flask under the protection of argon gas for about 10 min. Next, carbon monoxide gas (99.98%, Specialty Gases of America, Inc.) was introduced, and the COsaturated solution was heated at 210 °C for 15 min at a flow rate of 120 cm3/min (OMEGA FMA-A2305) at a pressure of 10 psi, followed by injecting the precursor solution with a syringe. The mixture in the closed flask turned dark within 1 min, which was kept at 210 °C while stirring under the CO atmosphere for 30 min. The black precipitate was washed with 10 mL of chloroform for three times and collected by centrifugation at 6500 rpm for 8 min. Synthesis of Pt Nanocubes. The procedure is similar to that for the synthesis of Pt icosahedron, except the two precursor solutions were made by room-temperature mixing, instead of hot injection. A mixture of Pt(acac)2 (20 mg), Y(acac)3·H2O (14 mg), DDA (10 mL), OA (50 μL), and DPE (1 mL) was bubbled with CO gas for 15 min after degassing in vacuum. The reaction flask was then transferred to a preheated oil bath at 210 °C. The mixture was exposed under the flow of CO gas throughout. The synthetic mixture turned black in about 5 min after the flask was transferred to the heated oil bath. It was kept in the oil bath for an additional 30 min. The final product was washed with chloroform for several times using the same procedure as stated above. In situ LTEM and ex situ TEM Studies. In situ LTEM observation was carried out on a JEOL2100Cryo TEM with a LaB6 emitter at 200 kV using a SiN-based liquid flow cell holder (Hummingbird Scientific). During LTEM observation, electron beam with a spot size of 200 nm in diameter was used, corresponding to a current density of 2 pA/cm2 on screen. Under this low electron dose condition, in conjunction with the use of liquid flow cell, the effect of electron irradiation was greatly suppressed, and electron beam-induced dendritic growth was no

Figure 8. Schematic illustration of modes of facet-selective oxidative etching of (top) icosahedral and (bottom) cubic Pt nanocrystals in the presence of Cl− and [AuCl4]− ions. The dissolution along the long green arrow directions is faster than that along the blue ones. F

DOI: 10.1021/acsnano.6b07541 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano longer observed.18 Particles used for both in situ LTEM and ex situ TEM observations were centrifuged and washed first before dispersed in DI water. A small amount (1.5 μL or 2 μg) of Pt nanocrystal suspension in water solution was added into the liquid cell before TEM observation. The nanoparticle suspension contained a mixture of Pt icosahedra and cubes at a concentration of 2 mg/mL of each type. We estimated the population ratio between cube and icosahedron was close to 1. An aqueous solution of HAuCl4 (0.1 M) and KCl (0.01 M) was introduced into the liquid flow cell at 5 μL/min via a syringe pump during the observation. TEM images were recorded using a Gatan Ultrascan CCD camera installed on the JEOL2100 TEM. The bright-field TEM micrographs were taken using an exposure time of 1 s. The entire process was also captured on video at 100 ms interval. Snapshots were recorded at 1 s exposure approximately every 15 s. The snapshots provided a better signal/noise ratio for the recorded images, which were used for further analysis. High-resolution TEM (HRTEM) micrographs were acquired using ACAT (the Argonne Chromatic Aberration corrected TEM), which is an FEI Titan TEM with a CEOS Cc/Cs imaging corrector, operated at 200 kV. The nanoparticles were washed at least three times in DI water before beginning the in situ experiment to minimize the amount of surface ligands on the surface. From the HRTEM images, nanoparticles exhibit a fairly clean surface without amorphous-like contrast generated from surface capping ligands. During the entire observation, the nanoparticle movement was limited to up to ∼30 nm in distance, and small in-plane rotations were occasionally observed. These movements were corrected in our distance measurements. Experimentally, the focus was adjusted before taking the TEM micrograph; thus error due to the movement of particles should not exceed a 3pixel length or 0.4 nm in all measured distances. In the ex situ experiment, 10 mg of Pt nanoparticles was added into 2 mL of mixed solution of HAuCl4 (0.1 M) and KCl (0.01 M). The reactant solution was stirred at 500 rpm at room temperature for different lengths of reaction time.

was partially supported by Department of Energy (50%, grant no.: DEFG02-01ER45923 to J.M.Z.) and the Shen Fellowship from Department of Chemical and Biomolecular Engineering at University of Illinois (50%, to J.B.W./H.Y.). W.P.G. was supported by NSF (grant no.: DMR-1410596 to J.M.Z.). Use of the Center for Nanoscale Materials, including the resources of the Electron Microscopy Center at Argonne National Laboratory, is supported in part by the U.S. Department of Energy, Office of Science under contract no. DE-AC0206CH11357. We thank Dr. Jianguo Wen and Dr. Dean J. Miller for help with HRTEM.

REFERENCES (1) Wu, J.; Yang, H. Platinum-Based Oxygen Reduction Electrocatalysts. Acc. Chem. Res. 2013, 46, 1848−1857. (2) Lim, B.; Jiang, M.; Camargo, P. H. C.; Cho, E. C.; Tao, J.; Lu, X.; Zhu, Y.; Xia, Y. Pd-Pt Bimetallic Nanodendrites with High Activity for Oxygen Reduction. Science 2009, 324, 1302−1305. (3) Hu, M.; Chen, J.; Li, Z.-Y.; Au, L.; Hartland, G. V.; Li, X.; Marquez, M.; Xia, Y. Gold Nanostructures: Engineering Their Plasmonic Properties for Biomedical Applications. Chem. Soc. Rev. 2006, 35, 1084−1094. (4) Cordeiro, M. A. L.; Crozier, P. A.; Leite, E. R. Anisotropic Nanocrystal Dissolution Observation by in situ Transmission Electron Microscopy. Nano Lett. 2012, 12, 5708−5713. (5) Chee, S. W.; Duquette, D. J.; Ross, F. M.; Hull, R. Metastable Structures in Al Thin Films Before the Onset of Corrosion Pitting as Observed using Liquid Cell Transmission Electron Microscopy. Microsc. Microanal. 2014, 20, 462−468. (6) Hatty, V.; Kahn, H.; Heuer, A. H. Fracture Toughness, Fracture Strength, and Stress Corrosion Cracking of Silicon Dioxide Thin Films. J. Microelectromech. Syst. 2008, 17, 943−947. (7) Mayrhofer, K. J. J.; Meier, J. C.; Ashton, S. J.; Wiberg, G. K. H.; Kraus, F.; Hanzlik, M.; Arenz, M. Fuel Cell Catalyst Degradation on the Nanoscale. Electrochem. Commun. 2008, 10, 1144−1147. (8) Burda, C.; Chen, X.; Narayanan, R.; El-Sayed, M. A. Chemistry and Properties of Nanocrystals of Different Shapes. Chem. Rev. 2005, 105, 1025−1102. (9) Xia, Y.; Xiong, Y.; Lim, B.; Skrabalak, S. E. Shape-Controlled Synthesis of Metal Nanocrystals: Simple Chemistry Meets Complex Physics? Angew. Chem., Int. Ed. 2009, 48, 60−103. (10) Zhang, H.; Jin, M.; Liu, H.; Wang, J.; Kim, M. J.; Yang, D.; Xie, Z.; Liu, J.; Xia, Y. Facile Synthesis of Pd−Pt Alloy Nanocages and Their Enhanced Performance for Preferential Oxidation of CO in Excess Hydrogen. ACS Nano 2011, 5, 8212−8222. (11) Doná, E.; Cordin, M.; Deisl, C.; Bertel, E.; Franchini, C.; Zucca, R.; Redinger, J. Halogen-Induced Corrosion of Platinum. J. Am. Chem. Soc. 2009, 131, 2827−2829. (12) Gratz, A. J.; Manne, S.; Hansma, P. K. Atomic Force Microscopy of Atomic-Scale Ledges and Etch Pits Formed During Dissolution of Quartz. Science 1991, 251, 1343−1346. (13) Zheng, Y.; Zeng, J.; Ruditskiy, A.; Liu, M.; Xia, Y. Oxidative Etching and Its Role in Manipulating the Nucleation and Growth of Noble-Metal Nanocrystals. Chem. Mater. 2014, 26, 22−33. (14) Long, R.; Zhou, S.; Wiley, B. J.; Xiong, Y. Oxidative Etching for Controlled Synthesis of Metal Nanocrystals: Atomic Addition and Subtraction. Chem. Soc. Rev. 2014, 43, 6288−6310. (15) Zheng, H.; Smith, R. K.; Jun, Y.-W.; Kisielowski, C.; Dahmen, U.; Alivisatos, A. P. Observation of Single Colloidal Platinum Nanocrystal Growth Trajectories. Science 2009, 324, 1309−1312. (16) Liao, H.-G.; Cui, L.; Whitelam, S.; Zheng, H. Real-Time Imaging of Pt3Fe Nanorod Growth in Solution. Science 2012, 336, 1011−1014. (17) Liao, H.-G.; Zherebetskyy, D.; Xin, H.; Czarnik, C.; Ercius, P.; Elmlund, H.; Pan, M.; Wang, L.-W.; Zheng, H. Facet Development during Platinum Nanocube Growth. Science 2014, 345, 916−919.

ASSOCIATED CONTENT S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.6b07541. TEM images, figures showing the analysis of corrosion dynamics, tables summarizing the parameters of dissolution kinetics (PDF) The dissolution process of the particles presented in Figure 2. (AVI)

AUTHOR INFORMATION Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Hong Yang: 0000-0003-3459-4516 Present Address ⊥

State Key Lab of Metal Matrix Composites and School of Materials Science and Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China. Author Contributions ‡

These authors contributed equally to this work.

Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS This work was supported in part by the U.S. National Science Foundation (grant nos.: CHE-1213926 to H.Y. and DMR1410596 to J.M.Z.) and the University of Illinois (H.Y.). J.B.W. G

DOI: 10.1021/acsnano.6b07541 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano (18) Wu, J.; Gao, W.; Wen, J.; Miller, D. J.; Lu, P.; Zuo, J.-M.; Yang, H. Growth of Au on Pt Icosahedral Nanoparticles Revealed by LowDose in situ TEM. Nano Lett. 2015, 15, 2711−2715. (19) Ross, F. M. Opportunities and Challenges in Liquid Cell Electron Microscopy. Science 2015, 350, aaa9886. (20) Ngo, T.; Yang, H. Toward Ending the Guessing Game: Study of the Formation of Nanostructures Using in situ Liquid Transmission Electron Microscopy. J. Phys. Chem. Lett. 2015, 6, 5051−5061. (21) Wu, J.; Shan, H.; Chen, W.; Gu, X.; Tao, P.; Song, C.; Shang, W.; Deng, T. In Situ Environmental TEM in Imaging Gas and Liquid Phase Chemical Reactions for Materials Research. Adv. Mater. 2016, 28, 9686−9712. (22) Zheng, H.; Meng, Y. S.; Zhu, Y. Frontiers of in situ Electron Microscopy. MRS Bull. 2015, 40, 12−18. (23) Park, J.; Elmlund, H.; Ercius, P.; Yuk, J. M.; Limmer, D. T.; Chen, Q.; Kim, K.; Han, S. H.; Weitz, D. A.; Zettl, A.; Alivisatos, A. P. 3D Structure of Individual Nanocrystals in Solution by Electron Microscopy. Science 2015, 349, 290−295. (24) Smeets, P. J. M.; Cho, K. R.; Kempen, R. G. E.; Sommerdijk, N. A. J. M.; De Yoreo, J. J. Calcium Carbonate Nucleation Driven by Ion Binding in a Biomimetic Matrix Revealed by in situ Electron Microscopy. Nat. Mater. 2015, 14, 394−399. (25) Nielsen, M. H.; Aloni, S.; De Yoreo, J. J. In situ TEM Imaging of CaCO3 Nucleation Reveals Coexistence of Direct and Indirect Pathways. Science 2014, 345, 1158−1162. (26) Loh, N. D.; Sen, S.; Bosman, M.; Tan, S. F.; Zhong, J.; Nijhuis, C. A.; Král, P.; Matsudaira, P.; Mirsaidov, U. Multistep Nucleation of Nanocrystals in Aqueous Solution. Nat. Chem. 2017, 9, 77−82. (27) Qin, F.; Wang, Z.; Wang, Z. L. Growth and Coalescence Dynamics of Hybrid Perovskite Nanoparticles Observed by LiquidCell Transmission Electron Microscopy. ACS Nano 2016, 10, 9787− 9793. (28) Kim, J.; Jones, M. R.; Ou, Z.; Chen, Q. In situ Electron Microscopy Imaging and Quantitative Structural Modulation of Nanoparticle Superlattices. ACS Nano 2016, 10, 9801−9808. (29) Jiang, Y.; Zhu, G.; Lin, F.; Zhang, H.; Jin, C.; Yuan, J.; Yang, D.; Zhang, Z. In situ Study of Oxidative Etching of Palladium Nanocrystals by Liquid Cell Electron Microscopy. Nano Lett. 2014, 14, 3761−3765. (30) Tan, S. F.; Lin, G. H.; Bosman, M.; Mirsaidov, U.; Nijhuis, C. A. Real-Time Dynamics of Galvanic Replacement Reactions of Silver Nanocubes and Au Studied by Liquid-Cell Transmission Electron Microscopy. ACS Nano 2016, 10, 7689−7695. (31) Ye, X. C.; Jones, M. R.; Frechette, L. B.; Chen, Q.; Powers, A. S.; Ercius, P.; Dunn, G.; Rotskoff, G. M.; Nguyen, S. C.; Adiga, V. P.; Zettl, A.; Rabani, E.; Geissler, P. L.; Alivisatos, A. P. Single-Particle Mapping of Nonequilibrium Nanocrystal Transformations. Science 2016, 354, 874−877. (32) Weiner, R. G.; Chen, D. P.; Unocic, R. R.; Skrabalak, S. E. Impact of Membrane-Induced Particle Immobilization on Seeded Growth Monitored by in situ Liquid Scanning Transmission Electron Microscopy. Small 2016, 12, 2701−2706. (33) Arruda, T. M.; Shyam, B.; Ziegelbauer, J. M.; Mukerjee, S.; Ramaker, D. E. Investigation into the Competitive and Site-Specific Nature of Anion Adsorption on Pt Using in situ X-ray Absorption Spectroscopy. J. Phys. Chem. C 2008, 112, 18087−18097. (34) Xia, Y.; Yang, H.; Campbell, C. T. Nanoparticles for Catalysis. Acc. Chem. Res. 2013, 46, 1671−1672. (35) Yu, T.; Kim, D. Y.; Zhang, H.; Xia, Y. Platinum Concave Nanocubes with High-Index Facets and Their Enhanced Activity for Oxygen Reduction Reaction. Angew. Chem., Int. Ed. 2011, 50, 2773− 2777. (36) Zhou, W.; Wu, J.; Yang, H. Highly Uniform Platinum Icosahedra Made by Hot Injection-Assisted GRAILS Method. Nano Lett. 2013, 13, 2870−2874. (37) Wu, J.; Gross, A.; Yang, H.; Shape. and Composition-Controlled Platinum Alloy Nanocrystals Using Carbon Monoxide as Reducing Agent. Nano Lett. 2011, 11, 798−802.

(38) Wu, J.; Qi, L.; You, H.; Gross, A.; Li, J.; Yang, H. Icosahedral Platinum Alloy Nanocrystals with Enhanced Electrocatalytic Activities. J. Am. Chem. Soc. 2012, 134, 11880−11883.

H

DOI: 10.1021/acsnano.6b07541 ACS Nano XXXX, XXX, XXX−XXX