Synthesis of spherical copper-platinum nanoparticles by sonoelectrochemistry followed by conversion reaction

Electrochimica Acta 176 (2015) 567–574

Contents lists available at ScienceDirect

Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta

Synthesis of spherical copper-platinum nanoparticles by sonoelectrochemistry followed by conversion reaction Samuel Levia , Valérie Manciera , Céline Roussea , Omar Lozano Garciab , Jorge Mejiab , Maribel Guzmanc , Stéphane Lucasb , Patrick Fricoteauxa,* a

LISM, EA 4695, UFR Sciences Exactes et Naturelles, BP 1039, F-51687 Reims Cedex 2, France Namur Nanosafety Center (NNC), NAmur Research Institute for LIfe Sciences (NARILIS), Research Centre for the Physics of Matter and Radiation (PMR), University of Namur (UNamur), rue de Bruxelles 61, B-5000 Namur, Belgium c PUCP, Engineering Department, Av. Universitaria 1801, Lima-32, Peru b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 8 December 2014 Received in revised form 25 February 2015 Accepted 25 June 2015 Available online 17 July 2015

Cu-Pt nanopowders were prepared by sonoelectrochemistry followed by a displacement reaction. The first method provides pure copper particles. In the second method the surface copper atoms are replaced by platinum atoms. The influence of dissolved oxygen during the conversion reaction was also studied. Nanoparticles (NPs) were observed by transmission electron microscopy (TEM) and their size distribution was determined by centrifugal liquid sedimentation (CLS). The mean diameter of isolated particles was found to be around 8 nm. Their composition was studied by energy dispersive X-ray spectroscopy (EDXS) and X-Ray photoelectron spectrometry (XPS). These analyses and X-ray diffraction (XRD) patterns showed that the particle shell is a solid solution of Cu and Pt. The shell composition is heterogeneous with a richer Pt percentage on its surface. ã 2015 Elsevier Ltd. All rights reserved.

Keywords: nanoparticles copper platinum sonoelectrochemistry dissolved oxygen

1. Introduction Over the last two decades the interest for nanotechnology has increased markedly. This interest for nanoparticles can be explained by their various fields of application. Among them, medicine (cancer therapy [1,2], IRM contrasting agent [3] or drug delivering [4]), sensors [5], energy storage [6] and catalysis [7,8] can be cited. Nanoparticles are characterized by a high surfaceto-volume ratio. As the properties of a material mainly depend on its surface, nanoparticles synthesis is an interesting way to enhance the materials characteristics. Nowadays a large range of nanoparticles production methods are available: chemical or physical vapor deposition (CVD [9] and PVD [10]), laser ablation [11], sol-gel and solvothermal synthesis [12], reduction in emulsion [13,14] and thermal decomposition [15]. Other simple ways of producing nanoparticles are classical electrochemistry - under pulsed current [16] or not [17] - and sonoelectrochemistry. High intensity ultrasounds have been used in chemistry [18] since 1934 and are still used now to synthesize advanced materials such as nanomaterials and nanoalloys [19,20]. The

* Corresponding author. Tel.: +33 326 91 85 82; fax: +33 326 91 89 15. E-mail address: [email protected] (P. Fricoteaux). http://dx.doi.org/10.1016/j.electacta.2015.06.155 0013-4686/ ã 2015 Elsevier Ltd. All rights reserved.

combination of electrodeposition and ultrasound was first employed around 1950 [21]. The major effect of ultrasound propagation in an aqueous solution is acoustic cavitation i.e. formation, growth and collapse of microbubbles. During the negative pressure phase, bubbles are formed and grow to a size of 5 to 20 mm [22]. During the compression cycle, these bubbles rapidly collapse, local temperature rises to a thousand kelvin and pressure to hundreds of bars. Strong shock waves and very high cooling rate can also be observed around the bubbles. When they collapse a liquid jet forms. If the cavitation takes place nearby, the jet erodes the surface. Other physical and chemical effects have been observed such as: mass transport improvement, diffusion layer decrease, cleaning and degreasing of the electrode surface and formation of radicals like OH
[23–25]. Several sonoelectrochemical setups were elaborated. The first one used a classical electrochemical cell dipped into an ultrasound bath with the power transmitted to the electrochemical cell varying according to the spatial configuration of the setup [26]. Next, another configuration was developed in which the cathode (working electrode) is coupled to the transducers; this new electrode was named sonotrode [27]. With this specific device, Reisse et al. [28] studied copper electrodeposition and showed the interest of ultrasound to produce copper nanopowders. This setup was later used to produce various nanoparticles such as: pure metals (Zn [29], Ni [29], Pt [30], Au [31], Ag [32]), alloys

568

S. Levi et al. / Electrochimica Acta 176 (2015) 567–574

[33,34], oxides [35] and core-shell structures (a material covered by another material) [36]. At nanometric scale, platinum (catalytic material) has been extensively studied because the nanostructures enhance the active surface. Usual applications are electrocatalyst for oxygen reduction reaction in fuel cells [37], cyclopropanation [38], cycloisomerization [39] or Suzuki coupling [40]. Displacement of platinum ions on bulk copper can be employed to lead Pt deposition [41], a very convenient way to deposit a noble metal. In 2010, Sarkar and Manthiran achieved synthesis of Cu-Pt NPs using a chemical reduction to synthesize the copper core and a displacement reaction to create the platinum shell [42]. Those nanostructures have several interests. Firstly, core-shell nanostructures may combine the properties of the two metallic components [43,44]. Secondly, in the case of expensive material (as platinum) the use of a less expensive material as a core decreases the production cost for a same active surface. Moreover it has been shown that Cu-Pt NPs are four times more efficient as a catalyst for the oxygen reduction reaction than pure Pt NPs [45]. For this paper, the researches were conducted in order to define new parameters for the synthesis of Cu-Pt NPs combining sonoelectrochemistry and galvanic displacement. 2. Experimental section 2.1. Electrolytic solutions The solution for copper electrolysis was made with copper sulfate pentahydrate (CuSO4, 5H2O) at 0.2 mol.L1 and sulfuric acid (H2SO4) at 0.9 mol.L1. The platinum bath was a mixture of Na2PtCl6 (0.001 mol.L1) and sulfuric acid (0.45 mol.L1). Copper sulfate was bought from Panreac, sulfuric acid and platinum salt were bought from Chimie Plus Laboratoires. The three reagents were used without any further purification. The copper electrolyte was regulated at 30  C. The conversion of platinum was done at ambient temperature. 2.2. Preliminary study In order to synthesize copper nanoparticles, the reduction potential of the copper was determined by recording a current density versus potential curve. A disc of titanium (S = 0.785 cm2) was used as working electrode and a copper wire as soluble counter electrode. The reference electrode was a mercury saturated sulfate electrode noted SSE (0.64 V vs SHE). The potentiostat employed was a home-made one. The current density versus potential curve presented in Fig. 1 was recorded with a scan rate of 20 mV.s1. In the cathodic part of the curve, the reduction of copper appeared at 600 mV vs SSE (Eq. (1)) while the hydrogen evolution (Eq. (2)) happened around 1200 mV vs SSE. Since the

Potential (mV/SSE) -1500

-1300

-1100

-900

-700

-500 0 -20

-60 -80 -100

Intensity (mA/cm 2 )

-40

-120 -140

Fig. 1. Current density-potentiel curve for Cu(II) solution reduction at T = 30  C. Scanning rate = 20 mV.s1.

hydrogen evolution is an adverse reaction, the electrolysis potential has to be set between 600 and 1200 mV vs SSE. The potential of 1000 mV vs SSE was arbitrarily chosen to carry out the production of copper nanoparticles. Cu2+ + 2 e ! Cu

(1)

2 H+ + 2 e ! H2

(2)

2.3. Copper nanoparticles synthesis Copper nanoparticles were produced using out-of-phase pulsed sonoelectrochemical setup (Fig 2a). Two generators are necessary. A potentiostat, connected to the working electrode of the sonotrode, manages electrolysis. An ultrasonic generator connected to tranducers, placed at the top of the sonotrode, allows their vibrations. The latter induce surface working electrode vibrations that pull off the electrodeposit. These two systems are independant and never work at the same time. A rest time is also applied between ultrasounds generation and electrolysis in order to reduce acoustic streaming in solution, diffusion layer and possible electronic interactions. This setup, patented by Winand et al. [28,46], has been widely and successfully employed to produce numerous powders. In this work, the previous homemade potentiostat was linked via a trigger to a titanium sonotrode (Linea S23-10-1/2 manufactered by Sinaptec) powered by a Nexus P198-R generator (Sinaptec). The sonotrode was supplied with an electrical current set at a frequency around 24 kHz. To avoid overheating, the piezoelectric transducers were cooled with compressed air. The Cu NPs synthesis was controlled using a classical three electrode electrochemical setup. The reference electrode was the SSE, the soluble counter electrode was made with a copper wire and the working electrode was the sonotrode. The side parts of the sonotrode were insulated with a sleeve to let only a planar circular surface at the bottom of the horn as electroactive surface (0.785 cm2). Since high intensity ultrasounds can damage the sintered material, the reference electrode was placed in a second beaker filled with saturated potassium sulfate K2SO4. The link between the beaker and the electrochemical cell was a bridge filled with the copper electrolyte. The counter electrode was covered with a polyethylene net to avoid mixing of the copper particles coming from the counter electrode ultrasonic erosion with the copper nanoparticles produced by electrolysis. Before experiments, the oxygen was removed from electrolyte by nitrogen bubbling during 20 minutes. During electrolysis, the nitrogen bubbling was stopped to avoid the change of the ultrasonic waves frequency and because a sonicated solution is de facto degassed. In order to produce Cu NPs, a pulse of polarization (1000 mV vs SSE; pulse duration = 100 ms) was applied which produces nuclei at the surface of the sonotrode. Those nuclei were immediately pulled off the surface using a 100 W.cm2 power ultrasonic pulse for 100 milliseconds. Applied potential for electrolysis and ultrasound pulses were used out of phase, without any latency between them. Then a free time of 100 ms was applied to let the diffusion of electrochemical species. The sequence of these pulses was repeated to produce large amounts of Cu NPs (Fig. 2b). After the sonoelectrochemical synthesis, the copper nanoparticles suspension was filtered under vacuum using a hydrophilic polyethersulfone membrane (Sartorius Biolab Products 15458) before several washings in large volumes of deionized water and with deoxygenated ethanol. Powders were removed from the filter by immersion in degassed ethanol and sonication in an ultrasonic tank. The obtained copper nanopowders were dried after sedimentation and placed under vacuum to avoid oxidation.

S. Levi et al. / Electrochimica Acta 176 (2015) 567–574

569

a

b Telectrolysis

Tultrasound

Tfree

Telectrolysis

Tultrasound

Tfree

1

2

3

1

2

3

Time of electrolysis

1: Pulse of potential at - 1000 mV/ESS (100 ms) 2: Pulse of ultrasonics waves (100 ms) 3: Free time: replenishment of the diffusion layer (100 ms) Fig. 2. (a) Schematics of the pulsed electrochemical setup. WE= working electrode; CE= counter electrode; RE: reference electrode (SSE). (b) Out-of-phase electrochemical and ultrasound pulses.

2.4. Platinum shell synthesis The copper nanoparticles were immersed in the platinum solution in order to provoke a spontaneous galvanic displacement reaction due to the respective potential of copper and platinum (E0 = 0.74 V vs SHE for PtCl64/Pt; E0 = 0.34 V vs SHE for Cu2+/Cu) (Eq. (3)). After the galvanic displacement under stirring, the Cu-Pt NPs were filtered under vacuum and rinsed with deionized water and deoxygenated ethanol. After washing, the nanopowders obtained were stored in deoxygenated pure ethanol. Pt(IV) + 2 Cu ! 2 Cu(II) + Pt

(3)

2.5. Characterizations X-Ray diffraction (XRD) analysis were carried out using a BRUCKER D8 ADVANCE X-Ray diffractometer equipped with a copper anticathode (lCuKa=1.54056 Å). XRD diffractograms were recorded in Bragg Brentano configuration. The chemical composition was investigated by energy-dispersive X-ray spectroscopy (EDXS) using a JEOL 1300 microprobe coupled with a JEOL JSM 6440LA scanning electron microscope (SEM). Studies of the morphology and approximate size of powders were performed at 200 kV on a Philips CM20- and on a Philips CM30 TEM/STEM electron microscopes. Particle size distributions were measured with a disc centrifuge DC24000 (CPS Instruments). The measurement is based on the centrifugal liquid sedimentation (CLS) method according to Stokes’ law using a 405 nm wavelength laser. This method is also known as

differential centrifugal sedimentation. The measured diameters are hydrodynamic diameters. A certified calibration standard of polyvinyl chloride nanoparticles (226 nm) was used to calibrate the measurements. Samples were diluted in milliQ water and stirred during 30 min at 400 rpm before mesaurement. Each measurement was done by injecting 0.2 mL of 2 mg.mL1 stock dispersions. Surface composition was analyzed with an X-ray Photoelectron Spectroscopy (XPS) system. The apparatus used was a SSX100 system using Al K-a X-rays, with spectra recorded at 35 take-off angle. The analysis depth of XPS is around 5 nm. Core-level lines (C1s, Si2p) were calibrated to the C 1s peak (284.6 eV) and Au 4f7/2 peak (84.0 eV). Spectra were analyzed, fitting the Gaussian function to the experimental curve, with a non-linear least squares scheme using a Shirley background. Nominal resolution was measured as full width at half maximum of 1.0 eV (core-level spectrum) to 1.5 eV (survey spectrum). A droplet of stock dispersions (idem to CLS) was deposited and left to dry naturally on a gold slab. 3. Results and discussion 3.1. Characterizations of copper nanoparticles The copper nanoparticles were first characterized by XRD analysis (Fig. 3). The X-Ray diffractogram of sample shows peaks at 43.30; 50.43; 74.13; 89.90; 116.92; 136.51 and 144.70 . All these peak values are attributed to pure copper crystallized in a face centered cubic system (JCPDS 004-0836). No copper oxide is detected.

570

S. Levi et al. / Electrochimica Acta 176 (2015) 567–574

Fig. 3. Diffractogram of copper nanoparticles between 40 and 148 in BraggBrentano conditions. Fig. 5. Transmission electron micrograph of Cu nanoparticles.

CLS analyses were carried out to determine the diameter of Cu NPs. Their size were found equal to 7  2 nm (Fig. 4). An example of bright field TEM picture in Fig. 5 shows the Cu nanoparticles with roughly around 10 nm. 3.2. Copper-platinum nanoparticles The platinum quantity can be achieved choosing the displacement reaction time and/or choosing the amount of introduced platinum ions. To describe the initial experimental conditions (molecular amount of copper and platinum), the ratio x (Eq. (4)) was defined according to the stoichiometry of Eq. (3):

x¼

2:nPt0 nCu0

(4)

with nPt0 : introduced moles number of Pt(IV) nCu0 : introducedmolesnumberofmetalliccopper = (molecular weight of Cu = 63,55 g.mol1) For example a x value of 1 means that the copper and the platinum were introduced in the stoichiometric proportions. In this case and for higher values of x (platinum ions in excess with regard to the copper), and in the hypothesis of a total reaction, pure platinum nanoparticles would be obtained according to Eq. (3). However, if the copper is covered by a dense platinum layer, displacement reaction (Eq. (3)) is stopped and bilayer nanoparticles (copper core-platinum shell) can be obtained. In the case of the previous hypothesis (total conversion reaction), a theoretical maximum platinum atomic percentage (theoretical max Pt at. %) can be calculated according to Eq. (5).

at: % Pt max ¼

Particle size distribution (height normalized)

(5)

with nPt: moles number of metallic platinum obtained in case of a total displacement reaction nCu: moles number of remaining metallic copper in case of a total displacement reaction For example, when x  1, max Pt at. % = 100 % since all the copper can be converted into metallic platinum. Table 1 summarizes the different cases of maximum platinum atomic percentage according the x values. In case of x < 1, max Pt at. % can be expressed directly in function of x (Eq. (6)) as following:

x¼

2: nPt0 2: nPt0 ) nCu0 ¼ nCu0 x

at:%Pt max ¼

initial introduced mass of Cu NPs Molecular weight of Cu

100

nPt  100 nCu þ nPt

nPt0 nPt
 100 ¼  100 nCu þ nPt nCu0 2:nPt0 þ nPt0

nPt0   100 at:%Pt max ¼  n Pt0 2: x  2:nPt0 þ nPt0 nPt0   100 at:%Pt max n Pt0 2: x  nPt0 1   100 at:%Pt max ¼  2:x1  1

and then at%Pt max ¼

x

2x

 100

(6)

80

60

Table 1 Theoretical maximum platinum atomic percentage according to x values.

40

20

0 1

10

Diameter (nm)

100

1000

Fig. 4. Hydrodynamic diameter of copper nanopowder (7  2 nm) using CLS in number mode.

Initial stage Final stage when x<1 Final stage when x=1 Final stage when x>1

Pt(IV) +

2Cu !

Pt + 2 Cu (II)

nPt0 0

nCu0 nCu0  2: nPt0

0 nPt0

0 2: nPt0

0

nPt0

2: nPt0

nPt0 nPt0

 100 ¼ 100

0

nCu0 2

nCu0

nCu 0 2 nCu 0 2

 100 ¼ 100

0 nPt0 

nCu0 2

Theoretical max Pt at. %

x

2x

 100 (formula 3)

S. Levi et al. / Electrochimica Acta 176 (2015) 567–574

70

60

b (with N2)

50

Atomic percentage of Pt (%)

60

Atomic percentage of Pt (%)

a (without N2)

a (ratio = 0.20) b (ratio = 0.85)

50 40 30 20

571

40

30 20 10

10

0 0

50

0 0

50

100

150

200

Displacement reaction time (min)

100

150

200

250

300

Displacement reaction time (min)

Fig. 7. Evolution of the atomic percentage of platinum during the conversion reaction (for x = 0.20) a) without and b) with nitrogen bubbling.

Fig. 6. Evolution of the atomic percentage of platinum during the conversion reaction (a) for x=0.20, (b) for x=0.85.

Fig. 6 exhibits the evolution of Pt atomic percentage during the conversion reaction (Eq. (3)) for two values of x: 0.20 (Fig. 6a; theoretical max Pt at. % = 11) and 0.85 (Fig. 6b; theoretical max Pt at. % = 74). The experimental atomic percentage of platinum in each sample was determined by EDXS analysis. It can be noticed on this graph that the kinetic is very fast at the beginning of displacement reaction, that is in accordance with a large available area of copper. However the processes implied are probably different according to initial conditions because the curve obtained with x = 0.20 (Fig. 6a) exhibits two parts unlike the curve obtained with x = 0.85 (Fig. 6b). Nevertheless, the speed of displacement reaction (Eq. (3)) becomes very slow after two hours in both cases. It must also be noticed that after three hours of conversion, the obtained platinum percentage in series a (50%) is largely superior to the theoretical maximal value (11%) while it is lower for series b (56% compare to 74%). These observations led us to further investigations with different values of x with a longer conversion time (7 hours). The corresponding theoretical maximal atomic percentage, calculated via Eq. (6), are reported in Table 2 and compared with experimental results. No logical evolution seems to appear from this table. For x equal to 0.85, like for x = 2.0, the final percentage is lower to the theoretical maximum calculated value. This can be attributed to a full coverage of copper by platinum. For x lower to 0.85, the experimental values are superior to the theoretical maximum atomic percentage values. This point is very surprising because the initial conditions do not allow such a final percentage of platinum. A hypothesis may explain this result: for low values of x, the surface of the copper nanoparticles is not fully covered by dense platinum or is fully covered by a porous platinum layer, allowing the copper atoms to react with a second reagent. Indeed, as the conversion experiments are done in ambiant conditions, without controled atmosphere, dissolved oxygen is present inside

Fig. 8. X-Ray diffractograms of Cu-Pt nanopowders recorded in Bragg-Brentano conditions between 40 and 95

solutions. The dissolved oxygen (E0 = 1.23 V vs SHE for O2/H2O) can react then with copper nanoparticles following the Eq. (7): Cu + 1/2 O2 + 2 H+ ! Cu2+ + H2O

(7)

So, Cu can be simutaneously oxydized by Pt (IV) according to Eq. (3) and by O2 according to Eq. (7). Oxidation of copper by oxygen decreases the number of metallic copper atoms inside the nanoparticle without substitution by metallic platinum atoms. As the number of metallic copper decreases and the number of Pt increases. metallic platinum stays constant, the atomic ratio nPtnþn Cu Consequently a Pt atomic percentage higher than the maximal caculated one can be obtained.

Table 2 Atomic percentage of platinum of Cu-Pt NPs for different x values after 7 hours of conversion reaction (estimated error  4%).

x value

Theoretical max. at.% Pt

EDX results (at.% Pt) after 7 hours of conversion

0.20 0.40 0.70 0.85 2.00

11 25 54 74 100

58 45 75 58 65

572

S. Levi et al. / Electrochimica Acta 176 (2015) 567–574 Table 3 Comparison of at.% of Pt in nanoparticles between EDXS ( 4 %) and XPS ( 2 %) method. EDXS results (at.% Pt)

XPS results (at.% Pt)

45 75

87 100

Particle size distribution (height normalized)

100 80 60 40 20 0 1

10

100

1000

Diameter (nm)

Fig. 9. CLS measurement in number mode of Cu42-Pt58 nanoparticles.

To show the influence of dissolved oxygen on the Pt atomic percentage in the nanoparticles, experiments were conducted with or without nitrogen bubbling before starting the conversion reaction for x = 0.20 (corresponding to the more important difference between the theoritical maximun calculated at. % Pt and the experimental value). The results (Fig. 7) show a great difference according to the cases. Regardless of those experiments, it can be seen, that after 300 minutes, the platinum atomic percentage reached is always superior (about 50% for experiment a and about 20% for experiment b) to the maximal caculated one (11%). Nevertheless, the Pt atomic percentage increases with the dissolved oxygen quantity. Without nitrogen bubbling (experiment a), two stages are clearly observed with a change after about one hour of conversion. The first stage of Fig. 7, experiment a, is attributed to the simultaneous oxidation of copper by platinum ions and by O2. When almost all the platinum ions have been reduced, the conversion (Eq. (3)) becomes negligeable but oxygen continues to react with accessible copper according to Eq. (7). It results in a break on the curve.

After nitrogen bubbling (experiment b, Fig. 7), the dissolved oxygen quantity is very low and the displacement reaction (Eq. (3)) is the most important reaction. However, as conversion is made under stirring, O2 dissolution inside solution appears gradually and leads to a weak oxidation of Cu according Eq. (7). This phenomenon explains why the platinum atomic percentage value is slightly superior to the maximal one. It is important to note that the co-oxidation of Cu by Pt (IV) and O2 exists also for higher values of x, but its impact seems less visible. Indeed, for high x ration, the platinum layer can be sufficently covering and dense to block or to slow down the oxidation reactions and leads to at. % Pt lower to the theoritical one. The competition between the rate of a dense shell Pt formation (depending on the x ratio) and the copper oxidation by O2 can explain the illogical evolution of platinum final atomic percentage values presented in Table 2. XRD analyses were next performed on some Cu-Pt NPs corresponding to different compositions determinated by EDXS analyses. An example of three X-Ray diffractograms of copperplatinum nanoparticles is presented in Fig. 8. These diffractograms correspond respectively to Cu100-Pt0, Cu88-Pt12 and Cu35-Pt65 NPs. For the X-Ray diffractogram of copper nanoparticles, thin peaks can be observed at 43.30; 50.43; 74.13; 89.90 . They can be attributed to face-centered cubic system copper (JCPDS 004-0836). The narrow diffraction peaks of copper can be explained by a high degree of crystallinity. On the X-Ray diffractogramm of Cu88-Pt12 nanoparticles, two types of diffraction peaks can be observed: the narrow peaks localized at 43.30; 50.34; 74.13 and 89.93 are attributed to the copper face-centered cubic system while the large diffraction peaks centered on 42.73 ; 49.56 ; 72.18 and 87.67 can be attributed to a solid solution between copper (JCPDS 004-0836) and platinum (JCPDS 004-0802) both cristallized in a cubic face centered sytem. The X-Ray diffractogram of Cu35-Pt65 particles shows only large diffraction peaks centered on 40.62; 47.09; 68.99 and 83.33 . They are attributed to a solid solution of copper and platinum. The enlargement of the CuPt solid solution diffraction peaks may have two explanations: the nanosize of the domains or the low degree of crystallinity. At low content of platinum inside material, both non-alloyed metallic copper and solid solution of CuPt can be observed. For high content of platinum the nanoparticles are composed entirely with a solid solution of CuPt. The displacement of the solid solution peaks from X-Ray diffraction values of copper to X-Ray diffraction values of platinum follows the evolution of the atomic percentage of platinum. For high atomic percentage of platinum, XRD analyses showed the presence of a CuPt solid solution and the

Fig. 10. Transmission electron micrographes of Cu65-Pt35 nanoparticles: (a) global view; (b) focused view.

S. Levi et al. / Electrochimica Acta 176 (2015) 567–574

disappearance of the Cu bulk. For this sample, the composition can be obtained by Vegard’s law application too. The experimental lattice parameter was calculated for the sample with the peak (111) of the CuPt solid solution. The obtained result is consistent with the EDXS one: 67 at% Pt is found by this method (to compare to 65 at% Pt obtained by EDXS analysis). Homogeneity of the CuPt solid solution was studied on some samples comparing XPS and EDXS analyses. Two representative specimens are exhibited in Table 3 (Cu55Pt45 and Cu25Pt75). It shows that the values obtained by XPS are larger than the ones obtained by EDXS. Given that XPS is a surface analysis technique, this highlights that the platinum percentage is higher on the surface than in the center of the particle. This proves that the particles obtained are richer in copper in the core and richer in platinum at the surface, that suggesting a core-shell structure. This result and the ones obtained by XRD analyses are in agreement with previous works from Sarkar [42]. These authors have reported that the conversion reaction between copper nanoparticles and Pt (IV) ions leads to a core-shell nanostructure in which the core is composed of Cu-Pt alloy and the shell of Pt. To complete the characterization, particle size distribution of the copper-platinum nanoparticles was determined by CLS. An example is presented on Fig. 9 for the sample Cu42-Pt58. It shows a mean diameter of 9  4 nm. TEM analyses were finally done to achieve nanoparticles characterizations. As an example of TEM image of Cu65-Pt35 nanoparticles is shown on Fig. 10. The observed diameters, roughly around 10 nanometers, are consistent with CLS results. 4. Conclusion Copper-platinum nanoparticles were easily synthesized using sonoelectrochemistry followed by conversion reaction. The first stage leads to pure copper nanopowder, without oxide. The second stage is the copper covering by platinum using conversion reaction. The study of the displacement reaction kinetic demonstrates the main role of oxygen in the final composition of nanoparticles. Without any control of the atmosphere, oxygen may oxidize a part of metallic copper that enhances “artificially” the final platinum percentage inside the nanomaterial. XRD and XPS analyses show that for low percentage of platinum, a CuPt solid solution surrounds the Cu core while for high percentages, a pure platinum external layer covers a CuPt solid solution. The following works will be to characterize the catalytic properties of these new particles. These investigations will be particularly useful because no surfactant has been employed during synthesis which avoids a modification of the external layer. Acknowledgment This research received support from the QNano Project http:// www.qnano-ri.eu which is financed by the European Community Research Infrastructures under the FP7Capacities Programme (Grant No. INFRA-2010-262163) and the Conseil Régional de la Marne for its financial support. The authors thank Sylvie Ricord for the English corrections of this document. References [1] H. He, A. David, B. Chertok, A. Cole, K. Lee, J. Zhang, J. Wang, Y. Huang, V.C. Yang, Magnetic nanoparticles for tumor imaging and therapy: a so-called theranostic system, Pharm. Res. 30 (2013) 2445–2458. [2] T.-M. Sun, Y.-C. Wang, F. Wang, J.-Z. Du, C.-Q. Mao, C.-Y. Sun, R.-Z. Tang, Y. Liu, J. Zhu, Y.-H. Zhu, X.-Z. Yang, J. Wang, Cancer stem cell therapy using doxorubicin conjugated to gold nanoparticles via hydrazone bonds, Biomaterials 35 (2014) 836–845.

573

[3] M.A. Bruckman, X. Yu, N.F. Steinmetz, Engineering Gd-loaded nanoparticles to enhance MRI sensitivity via T 1 shortening, Nanotechnology 24 (2013) 462001. [4] A.O. Elzoghby, Gelatin-based nanoparticles as drug and gene delivery systems: Reviewing three decades of research, J. Control. Release 172 (2013) 1075–1091. [5] S. Prakash, T. Chakrabarty, A.K. Singh, V.K. Shahi, Polymer thin films embedded with metal nanoparticles for electrochemical biosensors applications, Biosens. Bioelectron. 41 (2013) 43–53. [6] S. Liu, S. Sun, X.-Z. You, Inorganic nanostructured materials for high performance electrochemical supercapacitors, Nanoscale 6 (2014) 2037–2045. [7] A.M. Kalekar, K.K.K. Sharma, A. Lehoux, F. Audonnet, H. Remita, A. Saha, G.K. Sharma, Investigation into the catalytic activity of porous platinum nanostructures, Langmuir 29 (2013) 11431–11439. [8] R. Carrera-Cerritos, V. Baglio, A.S. Aricò, J. Ledesma-Garcìa, M.F. Sgroi, D. Pullini, A.J. Pruna, D.B. Mataix, R. Fuentes-Ramìrez, L.G. Arriaga, Improved Pd electro-catalysis for oxygen reduction reaction in direct methanol fuel cell by reduced graphene oxide, Appl. Catal. B-Environ. 144 (2014) 554–560. [9] N.-M. Hwang, D.-K. Lee, Charged nanoparticles in thin film and nanostructure growth by chemical vapour deposition, J. Phys. D Appl. Phys. 43 (2010) 483001. [10] Z. Deng, Y. Tian, X. Yin, Q. Rui, H. Liu, Y. Luo, Physical vapor deposited zinc oxide nanoparticles for direct electron transfer of superoxide dismutase, Electrochem. Commun. 10 (2008) 818–820. [11] M.A. Gondal, T.F. Qahtan, M.A. Dastageer, T.A. Saleh, Y.W. Maganda, D.H. Anjum, Effects of oxidizing medium on the composition, morphology and optical properties of copper oxide nanoparticles produced by pulsed laser ablation, Appl. Surf. Sci. 286 (2013) 149–155. [12] Z. Li, B. Hou, Y. Xu, D. Wu, Y. Sun, W. Hu, F. Deng, Comparative study of sol–gelhydrothermal and sol–gel synthesis of titania–silica composite nanoparticles, J. Solid State Chem. 178 (2005) 1395–1405. [13] H.Y. Koo, S.T. Chang, W.S. Choi, J.-H. Park, D.-Y. Kim, O.D. Velev, Emulsion-based synthesis of reversibly swellable, magnetic nanoparticle-embedded polymer microcapsules, Chem. Mater. 18 (2006) 3308–3313. [14] C.E. Astete, C.S.S.R. Kumar, C.M. Sabliov, Size control of poly(d,l-lactide-coglycolide) and poly(d,l-lactide-co-glycolide)-magnetite nanoparticles synthesized by emulsion evaporation technique, Colloid. Surface A 299 (2007) 209–216. [15] Z. Mirghiasi, F. Bakhtiari, E. Darezereshki, E. Esmaeilzadeh, Preparation and characterization of CaO nanoparticles from Ca(OH)2 by direct thermal decomposition method, J. Ind. Eng. Chem. 20 (2014) 113–117. [16] Y.-C. Hsieh, L.-C. Chang, Y.-C. Tseng, P.-W. Wu, J.-F. Lee, Structural characterization of PtRu nanoparticles by galvanostatic pulse electrodeposition, J. Alloy Compd. 583 (2014) 170–175. [17] J.J. Burk, S.K. Buratto, Electrodeposition of Pt Nanoparticle Catalysts from H2Pt (OH)6 and their Application in PEM Fuel Cells, J. Phys. Chem. C 117 (2013) 18957–18966. [18] N. Moriguchi, The effect of supersonic waves on chemical phenomena, (III). The effect on the concentration polarization, J. Chem. Soc. Jpn. 55 (1934) 749–750. [19] D.G. Shchukin, D. Radziuk, H. Möhwald, Ultrasonic fabrication of metallic nanomaterials and nanoalloys, Annu. Rev. Mater. Res. 40 (2010) 345–362. [20] D.V. Radziuk, W. Zhang, D. Shchukin, H. Möhwald, Ultrasonic alloying of preformed gold and silver nanoparticles, Small 6 (2010) 545–553. [21] O. Lindström, A Study of Some Electrochemical Effects in a Field of Stationary Ultrasonic Waves, Acta Chem. Scand. 6 (1952) 1313–1323. [22] O. Louisnard, J. Gonzalez-Garcia, in: H. Feng, J. Weiss, G.V. Barbosa Cánovas (Eds.), Ultrasound Technologies for Food and Bioprocessing, chapter 2, Springer, United States, 2010 ISBN 9781441974716. [23] R.G. Compton, J.C. Eklund, S.D. Page, G.H.W. Sanders, J. Booth, Voltammetry in the presence of ultrasound. Sonovoltammetry and surface effects, J. Phys. Chem. 98 (1994) 12410–12414. [24] J.P. Lorimer, B. Pollet, S.S. Phull, T.J. Mason, D.J. Walton, The effect upon limiting currents and potentials of coupling a rotating disc and cylindrical electrode with ultrasound, Electrochim. Acta 43 (1998) 449–455. [25] V. Sáez, A. Frías-Ferrer, J. Iniesta, J. González-García, A. Aldaz, E. Riera, Characterization of a 20kHz sonoreactor. Part II: Analysis of chemical effects by classical and electrochemical methods, Ultrason. Sonochem. 12 (2005) 67–72. [26] D.J. Walton, S.S. Phull, A. Chyla, J.P. Lorimer, T.J. Mason, L.D. Burke, M. Murphy, R.G. Compton, J.C. Eklund, S.D. Page, Sonovoltammetry at platinum electrodes: surface phenomena and mass transport processes, J. Appl. Electrochem. 25 (1995) 1083–1090. [27] E. Namgoong, J.S. Chun, The effect of ultrasonic vibration on hard chromium plating in a modified self-regulating high speed bath, Thin Solid Films 120 (1984) 153–159. [28] J. Reisse, H. Francois, J. Vandercammen, O. Fabre, A. Kirsch-de Mesmaeker, C. Maerschalk, J.-L. Delplancke, Sonoelectrochemistry in aqueous electrolyte: A new type of sonoelectroreactor, Electrochim. Acta 39 (1994) 37–39. [29] J.-L. Delplancke, V. Di Bella, J. Reisse, R. Winand, Production of metal nanopowders by sonoelectrochemistry, Materials Research Society Symposium - Proceedings 372 (1995) 75–81. [30] V. Zin, B.G. Pollet, M. Dabala, Sonoelectrochemical (20kHz) production of platinum nanoparticles from aqueous solution, Electrochim. Acta 54 (2009) 7201–7206.

574

S. Levi et al. / Electrochimica Acta 176 (2015) 567–574

[31] A. Aqil, H. Serwas, J.-L. Delplancke, R. Jérôme, C. Jérôme, L. Canet, Preparation of stable suspensions of gold nanoparticles in water by sonoelectrochemistry, Ultrason. Sonochem. 15 (2008) 1055–1061. [32] L.-P. Jiang, A.-N. Wang, Y. Zhao, J.-R. Zhang, J.-J. Zhu, A novel route for the preparation of monodisperse silver nanoparticles via a pulsed sonoelectrochemical technique, Inorg. Chem. Commun. 7 (2004) 506–509. [33] M. Guzman, J.-L. Delplancke, G.J. Long, J. Delwiche, M.-J. Hubin-Franskin, F. Grandjean, Morphologic and magnetic properties of Pd100-xFex nanoparticles prepared by ultrasound assisted electrochemistry, J. Appl. Phys. 92 (2002) 2634–2640. [34] V. Mancier, J.L. Delplancke, J. Delwiche, M.J. Hubin-Franskin, C. Piquer, L. Rebbouh, F. Grandjean, Morphologic, magnetic, and Mössbauer spectral properties of Fe75Co25 nanoparticles prepared by ultrasound-assisted electrochemistry, J. Magn. Magn. Mater. 281 (2004) 27–35. [35] V. Mancier, A.-L. Daltin, D. Leclercq, Synthesis and characterization of copper oxide (I) nanoparticles produced by pulsed sonoelectrochemistry, Ultrason. Sonochem. 15 (2008) 157–163. [36] V. Mancier, C. Rousse-Bertrand, J. Dille, J. Michel, P. Fricoteaux, Sono and electrochemical synthesis and characterization of copper core–silver shell nanoparticles, Ultrason. Sonochem. 17 (2010) 690–696. [37] J.-Y. Choi, D.U. Lee, Z. Chen, Nitrogen-doped activated graphene supported platinum electrocatalyst for oxygen reduction reaction in PEM fuel cells, ECS Trans. 50 (2012) 1815–1822. [38] R. Rajeev, R.B. Sunoj, Mechanistic insights on platinum- and palladium-pincer catalyzed coupling and cyclopropanation reactions between olefins, Dalton Trans. 41 (2012) 8430–8440.

[39] S.Y. Kim, Y. Park, Y.K. Chung, Sequential platinum-catalyzed cycloisomerization and cope rearrangement of dienynes, Angew. Chem. Int. Edit. 49 (2010) 415–418. [40] X. Li, X. Zhao, S. Gao, S. Marqués-González, D.S. Yufit, J.A.K. Howard, P.J. Low, Y. Zhao, N. Gan, Z. Guo, The structure and coordinative self-assembly of films based on a palladium compound of pyridyl-acetylene platinum and its application in Suzuki and Heck coupling reactions, J. Mater. Chem. A 1 (2013) 9164–9172. [41] R.J. Tremont, G. Cruz, C.R. Cabrera, Pt electrodeposition on a copper surface modified with 3-mercaptopropyltrimethoxysilane and 1-propanethiol, J. Electroanal. Chem. 558 (2003) 65–74. [42] A. Sarkar, A. Manthiram, Synthesis of Pt@Cu Core-shell nanoparticles by galvanic displacement of Cu by Pt4+ ions and their application as electrocatalysts for oxygen reduction reaction in fuel cells, J. Phys. Chem. C 114 (2010) 4725–4732. [43] H. Jing, Z. Yu, L. Li, Antibacterial properties and corrosion resistance of Cu and Ag/Cu porous materials, J. Biomed. Mater. Res.-A 87 (2008) 33–37. [44] Y. Liu, J. Zhou, J. Gong, W.-P. Wu, N. Bao, Z.-Q. Pan, H.-Y. Gu, The investigation of electrochemical properties for Fe3O4@Pt nanocomposites and an enhancement sensing for nitrite, Electrochim. Acta 111 (2013) 876–887. [45] Z. Yang, Z. Geng, Y. Zhang, J. Wang, S. Ma, Improved oxygen reduction activity on the Ih Cu@Pt core–shell nanoparticles, Chem. Phys. Lett. 513 (2011) 118– 123. [46] R. Winand, J. Reisse, J-L. Delplancke, Device for the production of ultrafine powders, belgian patent n 09400555 (06/03/94); european patent n W095/ 33871 (12/14/95).