Salt-Templated Platinum-Copper Porous Macrobeams for Ethanol Oxidation

Platinum nanomaterials provide an excellent catalytic activity for diverse applications and given its high cost, platinum alloys and bi-metallic nanomaterials with transition metals are appealing for low cost and catalytic specificity. Here the synthesis of hierarchically porous Pt–Cu macrobeams and macrotubes templated from Magnus’s salt derivative needles is demonstrated. The metal composition was controlled through the combination of [PtCl4]2− with [Pt(NH3)4] and [Cu(NH3)4] ions in different ratios to form salt needle templates. Polycrystalline Pt–Cu porous macrotubes and macrobeams 10’ s–100’ s μm long with square cross-sections were formed through chemical reduction with dimethylamine borane (DMAB) and NaBH4, respectively. Specific capacitance as high as 20.7 F/g was demonstrated with cyclic voltammetry. For macrotubes and macrobeams synthesized from Pt2−:Pt2+:Cu2+ salt ratios of 1:1:0, 2:1:1, 3:1:2, and 1:0:1, DMAB reduced 3:1:2 macrotubes demonstrated the highest ethanol oxidation peak currents of 12.0 A/g at 0.5 mV/s and is attributed to the combination of a highly porous structure and platinum enriched surface. Salt templates with electrochemical reduction are suggested as a rapid, scalable, and tunable platform to achieve a wide range of 3-dimensional porous metal, alloy, and multi-metallic nanomaterials for catalysis, sensor, and energy storage applications.


Introduction
Platinum nanomaterials provide excellent catalytic activity for fuel cells, sensors, and automotive and petroleum applications [1,2]. Given the relative scarcity and high cost of platinum, platinum alloy nanostructures are appealing for materials scale-up and sustainability. Controlling platinum alloy composition and nanostructure shape allows the tuning of catalytic specificity [3][4][5][6]. High aspect ratio nanomaterials maximize catalytically active surface area while minimizing material volume, while alloy composition allows for control over chemical adsorbate binding and product release [7,8]. Furthermore, the relative solvent exposure of different platinum crystal faces also provides a means to control reaction kinetics [9]. While numerous platinum-palladium alloy nanostructures have been developed [10][11][12][13], platinum alloys and bi-metallic materials with abundant transition metals is appealing for low cost and tuning catalytic specificity [14][15][16].
Copper is an abundant, low-cost transition metal that has been used to synthesize platinum alloy and bimetallic nanomaterials. Pt-Cu nanomaterials synthesis techniques include dealloying [17], glucose-directed synthesis of alloy nanowire networks [18], electrodeposition [19],  As seen in Figure 2a, the salt needles present a purple color. Polarized optical microscopy was used to determine the salt needle lengths corresponding to each salt ratio shown in Figure 2b-d. Salts formed with a 1:0:1 ratio resulted in the longest salt needles ranging approximately 500 μm to more than 1000 μm. Salts formed with a 3:1:2 ratio are on the order of 100' s of μm with longer needles exceeding 500 μm. The 2:1:1 salt needles, are smaller ranging from 10' s of μm up to approximately 100 μm. 1:1:0 all platinum salt needles prepared from [PtCl4] 2− and [Pt(NH3)4] 2+ , 100 mM solutions were previously reported as 20-65 μm [42]. The salt needle formation trend is a decrease in length as As seen in Figure 2a, the salt needles present a purple color. Polarized optical microscopy was used to determine the salt needle lengths corresponding to each salt ratio shown in Figure 2b-d. Salts formed with a 1:0:1 ratio resulted in the longest salt needles ranging approximately 500 µm to more than 1000 µm. Salts formed with a 3:1:2 ratio are on the order of 100' s of µm with longer needles exceeding 500 µm. The 2:1:1 salt needles, are smaller ranging from 10' s of µm up to approximately 100 µm. 1:1:0 all platinum salt needles prepared from [PtCl 4 ] 2− and [Pt(NH 3 ) 4 ] 2+ , 100 mM solutions were previously reported as 20-65 µm [42]. The salt needle formation trend is a decrease in length as the relative copper ion content decreases. As the salt formation results from electrostatic attraction and square planar ion stacking, the larger salt needle precipitates with higher copper ion content is potentially attributed to an ion size mismatch between [Cu(NH 3 ) 4 ] 2+ with [PtCl 4 ] 2− .  As seen in Figure 2a, the salt needles present a purple color. Polarized optical microscopy was used to determine the salt needle lengths corresponding to each salt ratio shown in Figure 2b-d. Salts formed with a 1:0:1 ratio resulted in the longest salt needles ranging approximately 500 μm to more than 1000 μm. Salts formed with a 3:1:2 ratio are on the order of 100' s of μm with longer needles exceeding 500 μm. The 2:1:1 salt needles, are smaller ranging from 10' s of μm up to approximately 100 μm. 1:1:0 all platinum salt needles prepared from [PtCl4] 2− and [Pt(NH3)4] 2+ , 100 mM solutions were previously reported as 20-65 μm [42]. The salt needle formation trend is a decrease in length as

Scanning Electron Microscopy (SEM)
Salt needles with Pt 2− :Pt 2+ :Cu 2+ ratios of 1:0:1, 3:1:2, 2:1:1, and 1:1:0 were used as templates to synthesize hierarchically porous macrobeam structures through electrochemical reduction with either DMAB or NaBH 4 . DMAB reduced synthesis products are shown in Figure 3. In general, the resulting structures conform to the salt templates ( Figures S2-S4). A distinct square cross-section is seen in the scanning electron microscope images for 1:0:1 in Figure 3a-c, 3:1:2 in Figure 3d-f, and 2:1:1 in Figure 3g-i. Additionally, these structures present a hollow core yielding a macrotube structure. The 1:1:0 salt template, however, does not present a distinct square cross-section or hollow-core when reduced with DMAB, but rather interconnected particles approximately 50 nm in diameter formed from smaller aggregated nanoparticles as seen in Figure 3j-l.
The macrotubes formed from DMAB reduced 1:0:1 Pt 2− :Pt 2+ :Cu 2+ salt needles seen in Figure 3a-c have the largest and most distinct square cross-section with approximately 3 µm square sides. This larger square geometry, and corresponding linear dimension correlated with the polarized optical microscopy images in Figure 2b. The sidewalls appear to be aggregated nanoparticles forming a highly textured surface, but with a porosity that does not appear to extend laterally through the sidewalls. The macrotubes formed from 3:1:2 and 2:1:1 salts in Figures 3d-f and 3g-i, respectively, indicate a cross-section approximately 200 nm square, hollow cores, and sidewalls formed from interconnected nanoparticles that are porous from the exterior of the macrotubes to the inner hollow core. Using the same synthesis scheme, Pd(NH 3 ) 4 Cl 2 was used to precipitate salt needles with Pt 2− :Pt 2+ :Cu 2+ ratios of 2:1:1 and 3:1:2, subsequently reduced with 100 mM DMAB with 1:50 (v/v) salt needle solution to reducing agent ( Figure S4), and presented structures similar to those seen in Figures 3d-f and 3g-i.
When salt needles with Pt 2− :Pt 2+ :Cu 2+ ratios of 1:0:1, 3:1:2, 2:1:1, and 1:1:0 were reduced with NaBH 4 , the resulting structures shown in Figure 4 are distinctly different than those reduced with DMAB shown in Figure 3. In general, the macrobeam structures exhibit porosity from the network of interconnected nanoparticles, however, a distinct hollow core for the 1:0:1, 3:1:2, and 2:1:1 salt templates is not apparent. Two other features appear on the NaBH 4 reduced macrobeams that are not present on the DMAB reduced structures. The first feature, circled in yellow, is the appearance of linear striations that are orthogonal to the beam axis. The second feature, circled in red, is the appearance of spheres that are connected to but distinct from the macrobeam structure. Energy-dispersive X-ray mapping shown in Figure S5 indicates that the spheres are composed of predominantly platinum. Similar spheres were previously observed for Pd 2+ :Pt 2− salts at a 1:1 ratio reduced with NaBH 4 [43]. Salt needles with Pt 2− :Pt 2+ :Cu 2+ ratios of 1:0:1, 3:1:2, 2:1:1, and 1:1:0 were used as templates to synthesize hierarchically porous macrobeam structures through electrochemical reduction with either DMAB or NaBH4. DMAB reduced synthesis products are shown in Figure 3. In general, the resulting structures conform to the salt templates ( Figures S2-S4). A distinct square cross-section is seen in the scanning electron microscope images for 1:0:1 in Figure 3a-c, 3:1:2 in Figure 3d-f, and 2:1:1 in Figure 3g-i. Additionally, these structures present a hollow core yielding a macrotube structure. The 1:1:0 salt template, however, does not present a distinct square cross-section or hollowcore when reduced with DMAB, but rather interconnected particles approximately 50 nm in diameter formed from smaller aggregated nanoparticles as seen in Figure 3j-l. The macrotubes formed from DMAB reduced 1:0:1 Pt 2− :Pt 2+ :Cu 2+ salt needles seen in Figure 3a-c have the largest and most distinct square cross-section with approximately 3 μm square sides. This larger square geometry, and corresponding linear dimension correlated with the polarized optical microscopy images in Figure 2b. The sidewalls appear to be aggregated nanoparticles forming a highly textured surface, but with a porosity that does not appear to extend laterally through the sidewalls. The macrotubes formed from 3:1:2 and 2:1:1 salts in Figure 3d-f and 3g-i, respectively,   When salt needles with Pt 2− :Pt 2+ :Cu 2+ ratios of 1:0:1, 3:1:2, 2:1:1, and 1:1:0 were reduced with NaBH4, the resulting structures shown in Figure 4 are distinctly different than those reduced with DMAB shown in Figure 3. In general, the macrobeam structures exhibit porosity from the network of interconnected nanoparticles, however, a distinct hollow core for the 1:0:1, 3:1:2, and 2:1:1 salt templates is not apparent. Two other features appear on the NaBH4 reduced macrobeams that are not present on the DMAB reduced structures. The first feature, circled in yellow, is the appearance of linear striations that are orthogonal to the beam axis. The second feature, circled in red, is the  [42]. Linear striations are circled in yellow, and platinum nanospheres are circles in red.

Proposed Reduction-Dissolution Mechanism
The formation of a hierarchically porous macrobeam or macrotube from a salt needle template has been previously suggested to result from a charge balance dissolution mechanism during the reduction process [42]. In this study, the reduction of [PtCl 4 ] 2− to Pt 0 shown in Equation (1) releases four Cl − ions into solution near the salt template surface:  (2) and (3) results in the release of four neutral ammonia molecules: Ammonia's interaction with water is charge neutral in the formation of NH 4 + and OH − ions through its weakly basic nature. The proposed reduction-dissolution action and nanoparticle surface free energy minimization likely contributes to the porous structures observed in the SEM images in Figures 3 and 4.

X-ray Diffractometry
X-ray diffraction spectra for salt needles with Pt 2− :Pt 2+ :Cu 2+ ratios of 1:0:1, 3:1:2, 2:1:1, and 1:1:0 reduced in 0.1 M DMAB and NaBH 4 are shown in Figure 5a,b, respectively. Diffraction peaks were indexed to Joint Committee on Powder Diffraction Standards (JCPDS) reference number 01-087-0640 for platinum (red dashed lines), and 03-065-9026 for copper (blue dashed lines). For both DMAB and NaBH 4 reduced 1:1:0 salt templates (all platinum), the peaks indexed well to 01-087-0640. As the copper content progressively increases from 2:1:1, 3:1:2, to 1:0:1, the DMAB reduced macrobeam diffraction peaks progressively shift to more intermediate positions between the indexed platinum and copper peaks suggesting possible alloy formation. The NaBH 4 platinum peaks, on the other hand, do not shift to intermediate positions between indexed Pt and Cu, but rather distinct 111 and 220 Cu peaks appear for the 1:0:1 samples with peak convolution broadening observed for the 2:1:1 and 3:1:2 samples peaks suggesting bimetallic structures.  (2) and (3) results in the release of four neutral ammonia molecules: Ammonia's interaction with water is charge neutral in the formation of NH4 + and OH − ions through its weakly basic nature. The proposed reduction-dissolution action and nanoparticle surface free energy minimization likely contributes to the porous structures observed in the SEM images in  The elemental composition and phase of the Pt-Cu macrobeams formed with DAMB and NaBH4 reducing agents were assessed by XPS. Figure 6 shows high-resolution XPS scans of the a,b Pt 4f and c,d Cu 2p regions at each Pt 2− :Pt 2+ :Cu 2+ ratio reduced by DMAB and NaBH4. The Pt 4f7/2 and 4f5/2  The elemental composition and phase of the Pt-Cu macrobeams formed with DAMB and NaBH 4 reducing agents were assessed by XPS. Figure 6 shows high-resolution XPS scans of the a,b Pt 4f and c,d Cu 2p regions at each Pt 2− :Pt 2+ :Cu 2+ ratio reduced by DMAB and NaBH 4 . The Pt 4f 7/2 and 4f 5/2 photoelectron lines were measured at approximately 71 eV and 74 eV, respectively, with typical 3.33 eV spin-orbit separation, which indicated the metallic phase Pt (i.e., Pt 0 ) [47]. The Cu 2p 1/2 and 2p 3/2 photoelectron features exhibited metallic, mono-, and di-valent Cu species (i.e., Cu 0 , Cu 2 O, and CuO, respectively) [47] with relative concentrations dependent on the reduction agent. Satellite shake-up features observed at 942 eV, 944 eV, and 963 eV are predominantly associated with Cu + (i.e., Cu 2 O) [47]. The relative strength of DMAB versus NaBH 4 as a reductant for the Pt 2− , Pt 2+ , and Cu 2+ precursors resulted in varying distributions of presence in metallic and oxidized Pt/Cu phases with reductant-unique trends in measured binding energies.
Catalysts 2019, 9, x FOR PEER REVIEW 7 of 16 features observed at 942 eV, 944 eV, and 963 eV are predominantly associated with Cu + (i.e., Cu2O) [47]. The relative strength of DMAB versus NaBH4 as a reductant for the Pt 2− , Pt 2+ , and Cu 2+ precursors resulted in varying distributions of presence in metallic and oxidized Pt/Cu phases with reductantunique trends in measured binding energies. Use of DMAB as a reductant resulted in samples with predominantly metallic Pt and Cu, with traces of Cu2O confined to the 1:0:1 sample. The measured Pt 4f binding energies ranged from 70.7-71.1 eV for the 4f7/2 line and 74.0-74.4 eV for the 4f5/2 line, but did not appear to follow a linear trend with stoichiometric ratio. The DMAB-reduced 1:0:1 sample exhibited Pt + signatures, which was consistent with the concomitant presence of Cu2O (i.e., Cu + 2p1/2 and 2p3/2 and shake-ups) in the Cu 2p spectrum. The DMAB-reduced 3:1:2, 2:1:1, and 1:1:0 samples did not exhibit oxidized phases and appeared fully metallic. Compared to DMAB, the use of NaBH4 as a reductant resulted in samples with varying distributions of both metallic and oxidized Pt and Cu. Measured binding energies of the composite Pt 4f lines progressively decreased with Cu content, shifting the 4f7/2 line from 71.2 to 70.9 eV and the 4f5/2 line from 74.5 to 74.2 eV as the stoichiometric Cu percentage decreased from 50% to 0%. This was ostensibly attributable to the emergence of Pt + species underlying the composite Pt 4f lines because oxidized Cu was evident in all samples (except for the 1:1:0 all platinum sample). The NaBH4-reduced 2:1:1 and 3:1:2 samples exhibited Cu content that was predominantly in metallic Cu 0 form and Cu + state (i.e., Cu2O). The NaBH4-reduced 1:0:1 sample exhibited Cu that was less metallic in favor of the Cu + oxidation state, while some appeared in the Cu 2+ oxidation state (i.e., CuO).  The DMAB-reduced 3:1:2, 2:1:1, and 1:1:0 samples did not exhibit oxidized phases and appeared fully metallic.
Compared to DMAB, the use of NaBH 4 as a reductant resulted in samples with varying distributions of both metallic and oxidized Pt and Cu. Measured binding energies of the composite Pt 4f lines progressively decreased with Cu content, shifting the 4f 7/2 line from 71.2 to 70.9 eV and the 4f 5/2 line from 74.5 to 74.2 eV as the stoichiometric Cu percentage decreased from 50% to 0%. This was ostensibly attributable to the emergence of Pt + species underlying the composite Pt 4f lines because oxidized Cu was evident in all samples (except for the 1:1:0 all platinum sample). The NaBH 4 -reduced 2:1:1 and 3:1:2 samples exhibited Cu content that was predominantly in metallic Cu 0 form and Cu + state (i.e., Cu 2 O). The NaBH 4 -reduced 1:0:1 sample exhibited Cu that was less metallic in favor of the Cu + oxidation state, while some appeared in the Cu 2+ oxidation state (i.e., CuO).
Peak areas of the background-subtracted Pt 4f lines and Cu 2p 3/2 lines were used to determine relative Pt and Cu atomic percentages (at %) and calculated Pt:Cu atomic ratios for comparison with EDS and precursor stoichiometry, shown in Table 1. The surface Cu content increased in the series 1:0:1 > 3:1:2 > 2:1:1 with Cu at % of 31% (52%), 15% (24%), and 11% (20%), respectively, for the DMAB (NaBH 4 ) reducing agent with ±1% uncertainty. The 1:1:0 samples contained no Cu. The balance was Pt and no contaminants were detected in the XPS survey scans, aside from C and O. Corresponding XPS Pt:Cu atomic ratios using DMAB (NaBH 4 ) were 1:0 (1:0), 7.9:1 (4.0:1), 5.8:1 (3.1:1), and 2.2:1 (0.92:1) for the Pt 2− :Pt 2+ :Cu 2+ samples 1:1:0 > 2:1:1 > 3:1:2 > 1:0:1, respectively. This followed the same general trend measured in bulk EDS measurements, though suggests greater platinum enrichment of the surface for both DMAB and NaBH 4 reduced samples. For both NaBH 4 and DMAB reduced samples, the EDS Pt:Cu ratios indicated less metalized Pt than was present stoichiometrically in the salt-template suggesting that the [Cu(NH 3 ) 4 ] 2+ ion reduces preferentially in the given reducing conditions. This effect appears more pronounced for the NABH 4 reduced samples. The closer correspondence of the DMAB reduced EDS Pt:Cu ratios to the salt-stoichiometry also correlates with the lesser presence of oxide species seen with XPS. Further, the reduction-dissolution mechanism suggested in Section 2.2.2 would make Pt ions available for reduction on the macrobeam particle surface once the charge gradients dissipate providing a possible reason surface enrichment of platinum observed with XPS. After chemical reduction of the Pt 2− :Pt 2+ :Cu 2+ salt needles with DMAB, the resulting Pt-Cu macrotubes were pressed between glass slides with approximately 200 kPa into free-standing films for electrochemical characterization. Electrochemical impedance spectroscopy was performed in 0.5 M H 2 SO 4 in a frequency range of 1 MHz to 1 mHz, with the resulting Nyquist plots shown in Figure 7a. The macrobeam films from Pt 2− :Pt 2+ :Cu 2+ salt ratios of 1:1:0, 2:1:1, 3:1:2, and 1:0:1 exhibit characteristic high surface area capacitive behavior throughout the whole frequency range. Figure 7b shows the high-frequency resistance through the macrotube films. The variation in the real axis intercept is due to minor differences in contact between the sample film and working electrode. In general, the samples present high-frequency impedance speactra consistent with porous samples [48,49]. A larger semi-circle from the charge transfer resistance for the 1:1:0 sample is possibly due to the less well connected macrobeam particles seen in Figure 3j-l. The DMAB samples charge transfer resistance is also attributed to the larger internal surface area within films for charge transport through the open network of macrotubes. Cyclic voltammetry (CV) of the Pt 2− :Pt 2+ :Cu 2+ salt ratios of 1:1:0, 2:1:1, 3:1:2, and 1:0:1 was conducted in 0.5 M H 2 SO 4 ( Figure S8). Figure 7c shows the CV of Pt-Cu microbeams film from a 1:0:1 Pt 2− :Pt 2+ :Cu 2+ salt ratio, in 0.5 M H 2 SO 4 for scan rates of 1, 5, 10, 25, and 50 mV/s. The CV shows characteristic oxidation-reduction peaks and the hydrogen adsorption-desorption in the 1:0:1 macrotube film from the presence of catalytic platinum from the 2.2:1 atomic composition of Pt:Cu (as shown in the XPS results in Table 1). Moreover, Figure 7d shows the CV in 0.5 M H 2 SO 4 at a 0.5 mV/s scan rate for the various DMAB reduced macrotube films with Pt 2− :Pt 2+ :Cu 2+ salt ratios of 1:1:0, 2:1:1, 3:1:2, and 1:0:1. The CVs in Figure 7c,d indicate characteristic hydrogen adsorption and desorption peaks from the presence of platinum between −0.2 V and 0 V (vs Ag/AgCl). The electrochemical active surface areas determined from the capacitive region between 0.15-0.25 V for the 0.5 mV/s CV curves are 3.8, 10.5, 20.7, and 12.9 F/g for the macrotube films with Pt 2− :Pt 2+ :Cu 2+ salt ratios of 1:1:0, 2:1:1, 3:1:2, and 1:0:1, respectively. In Figure 7d, the pronounced oxidation peak occurring around at +0.96 V (vs Ag/AgCl) and reduction peaks more positive than +0.58 V (vs Ag/Ag/Cl) are associated with the overall catalytic bimetallic crystal structure in the different Pt:Cu films. The oxidation of Pt-H to Pt-OH and Pt-OH to Pt-O is evident in the strong oxidation peaks at + 0.96 V (vs Ag/AgCl) and + 1.1 V (vs Ag/AgCl), respectively [43]. The overall larger current density normalized for the mass of platinum (A/g pt ) in the 2:1:1 salt ratio is a result of the contribution of catalytic current from the high atomic composition of platinum within the microtube film (as shown in the EDS and XPS results in Table 1). The hydrogen adsorption and desorption more negative than 0 V (vs Ag/AgCl) were more distinct in the macrotubes with the higher elemental composition of platinum (as shown in the EDS and XPS results in Table 1) and surface accessible catalytic sites within the network. The oxidation peaks observed at +0.58 V (vs Ag/AgCl) in both Figure 7c,d is attributed to a hydrogen spill over, the migration of chemisorbed hydrogen from the metal electrode onto glass support surfaces, as in the case of the electrochemical cell used in this study [50].
Using the hydrogen adsorption of the CV curves at 0.5 mV/s scan rates, a nominal value of 210 µC/cm 2 for hydrogen adsorption on platinum [51], and the EDS determined sample mass composition, the electrochemically active surface areas (ECSA) for DMAB reduced 1:0:1, 3:1:2, 2:1:1, and 1:1:0 salts were 31.8, 22.6, 20.9, and 10.6 m 2 /g Pt , respectively. The corresponding ideal nanoparticle geometries based on the EDS mass averaged densities were 10.7, 13.8, 14.5, and 27.2 nm, respectively. These ideal nanoparticle geometries generally match the nanoparticle feature sizes observed in Figure 3 for each macrobeam Pt:Cu ratio. 0.5 M H2SO4 at a 0.5 mV/s scan rate for the various DMAB reduced macrotube films with Pt :Pt :Cu salt ratios of 1:1:0, 2:1:1, 3:1:2, and 1:0:1. The CVs in Figure 7c,d indicate characteristic hydrogen adsorption and desorption peaks from the presence of platinum between −0.2 V and 0 V (vs Ag/AgCl). The electrochemical active surface areas determined from the capacitive region between 0.15-0.25 V for the 0.5 mV/s CV curves are 3.8, 10.5, 20.7, and 12.9 F/g for the macrotube films with Pt 2− :Pt 2+ :Cu 2+ salt ratios of 1:1:0, 2:1:1, 3:1:2, and 1:0:1, respectively. In Figure 7d, the pronounced oxidation peak occurring around at +0.96 V (vs Ag/AgCl) and reduction peaks more positive than +0.58 V (vs Ag/Ag/Cl) are associated with the overall catalytic bimetallic crystal structure in the different Pt:Cu films. The oxidation of Pt-H to Pt-OH and Pt-OH to Pt-O is evident in the strong oxidation peaks at + 0.96 V (vs Ag/AgCl) and + 1.1 V (vs Ag/AgCl), respectively [43]. The overall larger current density normalized for the mass of platinum (A/gpt) in the 2:1:1 salt ratio is a result of the contribution of catalytic current from the high atomic composition of platinum within the microtube film (as shown in the EDS and XPS results in Table 1). The hydrogen adsorption and desorption more negative than 0 V (vs Ag/AgCl) were more distinct in the macrotubes with the higher elemental composition of platinum (as shown in the EDS and XPS results in Table 1) and surface accessible catalytic sites within the network. The oxidation peaks observed at +0.58 V (vs Ag/AgCl) in both Figure 7c,d is attributed to a hydrogen spill over, the migration of chemisorbed hydrogen from the metal electrode onto glass support surfaces, as in the case of the electrochemical cell used in this study [50].

Ethanol Oxidation in 1 M KOH/1 M Ethanol
The catalytic performance for ethanol oxidation of the macrotube films from chemical reduction with DMAB ( Figure 8a) and NaBH 4 (Figure 8b) from the various Pt 2− :Pt 2+ :Cu 2+ salt ratios of 1:1:0, 2:1:1, 3:1:2, and 1:0:1 was evaluated with CV in 1 M KOH and 1 CH 3 CH 2 OH. The electrooxidation of ethanol can result in the formation of acetylaldehyde and ultimately acetic acid [52]. Both of the CVs in Figure 8 have discrete oxidation peaks in the forward scan from the oxidation of the adsorbed ethanol and a reverse cathodic scan with another corresponding oxidation peak due to the oxidation of partially oxidized intermediate species such as CO, CH 3 COO − , and CH 3 CO − [34,53]. As seen in both Figures 8a and 8b, the ethanol oxidation peaks in both the forward and reverse scans for the different Pt 2− :Pt 2+ :Cu 2+ salt ratio macrobeam-macrotube films varied due to the differences in the amount of accessible catalytic platinum atomic sites as confirmed by EDS and XPS. Figure 8a shows the CV for the Pt-Cu macrotube films produced by chemical reduction with DMAB, the electrooxidation of ethanol for the Pt 2− :Pt 2+ :Cu 2+ salt ratio macrotube films varied between −0.27 V to −0.13 V (Ag/AgCl) in the forward anodic scans. The same variation was observed in the NaBH 4 reduced Pt 2− :Pt 2+ :Cu 2+ salt ratio macrobeam films where the ethanol oxidation peaks occurred in between −0.40 V (vs Ag/AgCl) to −0.26 V (Ag/AgCl) in the forward anodic scans. Meanwhile, on the reverse cathodic scan, the remaining surface oxides and intermediates are reduced until the potential is too negative to oxidize the remaining adsorbed species as represented by the sharp oxidation followed by reduction current peaks [52]. The sharp stripping peaks observed at approximately −0.2 V (vs Ag/AgCl) are attributed to the reduction of Pt(OH) 2 shown in Equation (4), Pt(OH) 2 + 2 e − ↔ Pt(s) + 2 OH − 0.14 V (SHE) (4) and adsorbed OH and O species similar to those observed on palladium catalysts [54]. The reverse cathodic scans on both Figures 8a and 8b for 1:0:1 have multiple oxidation and reduction current peaks likely due to having the highest atomic composition of copper.  The assembly of practical electrodes for actual device integration and performance is a key step in realizing the benefits of nanoscale materials. Some of the attractive aspects of the salt-templated Pt-Cu macrotubes-macrobeams are the rapid precipitation of the salt needle templates, the one-step reduction synthesis, and the ability to press the Pt-Cu nanostructures into free-standing films for direct device integration. These considerations are envisioned to facilitate scale-up of salt-templated nanomaterials. One key challenge in the scale-up process using mechanical pressing to assemble entangled macrotubes is to ensure homogeneous active material and pore size distribution throughout the thickness and volume of the electrode.

Copper-Platinum Macrobeam Synthesis
Magnus' Pt-Cu salt needle derivatives were prepared by mixing 100 mM square planar cation complexes to anion complexes at 1:1 (v/v). The 100 mM platinum and copper salts used were Pt(NH3)4Cl2H2O, K2PtCl4, and Cu(NH3)4SO4H2O (Sigma Aldrich, USA). Salt needle templates were formed by combining solutions in Pt 2− :Pt 2+ :Cu 2+ stoichiometric ratios of 1:1:0, 2:1:1, 3:1:2, and 1:0:1. A total of 1 mL salt solution was prepared, and then reduced with either DMAB or NaBH4 (Sigma Aldrich, USA) in a 1:50 (v/v) salt needle solution to reducing agent. Reduced platinum-copper macrotubes and macrobeams were rinsed in excess deionized water and dried at either ambient temperature or 100 °C. Using the same procedure Pd(NH3)4Cl2 was used to precipitate salt needles with Pt 2− :Pd 2+ :Cu 2+ ratios of 2:1:1 and 3:1:2, subsequently reduced with 100 mM DMAB with 1:50 (v/v)  Figure 8c shows a volcano plot with data from Figure 8a,b. Macrobeams-macrotubes prepared from the 3:1:2 salts show the highest ethanol oxidation currents relative to the other samples. There is a general increase in peak oxidation current from the 1:1:0 to 2:1:1 and then the 3:1:2 samples, with a significant drop-off with the 1:0:1 samples. This trend is attributed to the balance of high surface area, phase composition, and surface platinum enrichment see in SEM, XRD, and XPS data. DMAB reduced samples yielded higher ethanol oxidation currents than the NaBH 4 reduced samples and is attributed to the greater presence of surface oxide species on the NaBH 4 reduced samples seen with XPS.
While the ECSA values of the Pt-Cu samples in H 2 SO 4 electrolyte corresponds to the particle feature sizes observed with SEM, and is comparable to other Pt-Cu studies, ethanol oxidation performance comparison to previous Pt-Cu studies is challenging as the free-standing films and electrode conditions in this work are unique. Unlike dispersed materials in direct contact with a glassy carbon electrode conventionally used, the hierarchical porosity of the entangled macrobeams-macrotubes of the free-standing films, and sample mass of approximately 1-5 mg, introduced hysteresis attributable to current distribution, charging current, and mass transfer effects so that the performance at a slower scan rate of 0.5 mV/s is reported in Figure 8. The DMAB reduced 3:1:2 macrotubes exhibited the highest peak ethanol oxidation current of 12.0 A/g at 0.5 mV/s. As an indirect comparison, Pt/C and PtCu 1.8 screw thread-like platinum−copper nanowires had reported ethanol oxidation peak currents of 0.21 A/mg Pt (210 A/g Pt ) and 0.25 A/mg Pt (250 A/g Pt ) at 50 mV/s, respectively.
The assembly of practical electrodes for actual device integration and performance is a key step in realizing the benefits of nanoscale materials. Some of the attractive aspects of the salt-templated Pt-Cu macrotubes-macrobeams are the rapid precipitation of the salt needle templates, the one-step reduction synthesis, and the ability to press the Pt-Cu nanostructures into free-standing films for direct device integration. These considerations are envisioned to facilitate scale-up of salt-templated nanomaterials. One key challenge in the scale-up process using mechanical pressing to assemble entangled macrotubes is to ensure homogeneous active material and pore size distribution throughout the thickness and volume of the electrode.

Polarized Optical Microscopy
Polarized optical microscopy with an AmScope PZ300JC polarized light microscope (AmScope, Irvine, CA, USA) was used to image salt templates. Salt needle lengths were measured using ImageJ [55].

Scanning Electron Microscopy
Pt-Cu macrotubes and macrobeams were imaged with a FEI Helios 600 scanning electron microscope (SEM) (Thermo Fisher Scientific, Hillsboro, OR, USA). Energy dispersive X-ray spectroscopy (EDS) was performed with a FEI Helios 600 and a JEOL IT500HRLA SEM (JEOL USA, Peabody, MA USA).

X-ray Diffractometry
A PANalytical Empyrean diffractometer (PANalytical, Almelo, Netherlands) was used to collect X-ray diffractometry (XRD) spectra. Diffraction angles (2θ) from 5 • to 90 • , with a 2θ step size of 0.0130 • , and 20 s per step were used at 45 kV and 40 mA with Cu K α radiation (1.54060 Å). High Score Plus software (PANalytical) was used for XRD spectra analysis.

X-ray Photoelectron Spectroscopy
X-ray photoelectron spectroscopy (XPS) was performed on a ULVAC PHI VersaProbe III (Physical Electronics, East Chanhassen, MN USA) with a monochromated Al K α source with a 100 µm spot, 25 W X-ray beam and 45 • take-off angle. The operating pressure was < 6 × 10 −6 Pa. Surface charging was neutralized by a low-voltage Ar-ion beam and a barium oxide electron neutralizer. Analyzer pass energy was set to 55 eV for high-resolution scans. All analysis was conducted with PHI's software MultiPak v.9.6.0.15. Spectra were corrected according to the adventitious C 1s peak at 284.8 eV binding energy. Relative Pt and Cu ratios were evaluated for each Pt-Cu macrobeam sample using Shirley background-corrected Pt 4f and Cu 2p 3/2 peak areas.

Electrochemical Characterization
Macrobeams were pressed between glass slides with approximately 200 kPa into free-standing film electrochemical testing. A Bio-Logic VMP-3 potentiostat (Bio-Logic, Knoxville, TN, USA) was used to perform electrochemical impedance spectroscopy (EIS) and cyclic voltammetry (CV). An Ag/AgCl reference electrode, a 1 mm platinum wire counter-electrode, and a lacquer-coated 1 mm platinum wire with exposed tip in contact with samples were used in a three-electrode cell with 0.5 M H 2 SO 4 electrolyte. EIS was performed at open-circuit voltage with a frequency range of 1 MHz to 1 mHz with a 10 mV sine wave. CV was performed in a voltage range of −0.2 to 1.2 V (vs. Ag/AgCl) with scan rates of 0.5, 1, 5, 10, 25, 50, 75, and 100 mV/s. Ethanol oxidation cyclic voltammetry was performed using 1 M KOH and 1 M ethanol with scan rates of 0.5, 1, 5, 10, 25, and 50 mV/s.

Conclusions
Here we demonstrated the synthesis of hierarchically porous Pt-Cu bimetallic macrobeams and macrotubes templated from Magnus's salt derivative needles. The metal composition was controlled through the combination of [PtCl 4 ] 2− with [Pt(NH 3 ) 4 ] 2+ and [Cu(NH 3 ) 4 ] 2+ ions in different ratios to form the salt needle templates. Polycrystalline Pt-Cu porous macrotubes and macrobeams with square cross-sections were formed through chemical reduction with DMAB and NaBH 4 , respectively. Platinum enrichment on the nanoparticle surfaces was observed for DMAB reduced salts, whereas reduction with NaBH 4 resulted in small amounts of copper oxide species on the macrobeam surface. Electrochemical impedance spectroscopy indicated highly capacitive materials with electrochemical active surface area as high as 20.7 F/g demonstrated with cyclic voltammetry. For the characterized macrotubes and macrobeams synthesized from Pt 2− :Pt 2+ :Cu 2+ salt ratios of 1:1:0, 2:1:1, 3:1:2, and 1:0:1, DMAB reduced 3:1:2 macrotubes demonstrated the highest ethanol oxidation peak currents of 12.0 A/g at 0.5 mV/s and is attributed to the combination of a highly porous structure for favorable mass transfer properties, and the platinum enriched surface. The use of salt templates with electrochemical reduction is suggested as an aqueous-based, rapid, scalable, and tunable platform to achieve a wide range of 3-dimensional porous metal, metal alloy, and multi-metallic nanomaterials for catalysis, sensor, and energy storage applications.