Porous gold-layered cubic and octahedral Cu-oxide nanocrystals: Dopamine sensing

Two morphologically different porous gold layered on Cu-oxide-based electrochemical sensors were developed for the selective detection of dopamine in the presence of uric acid, ascorbic acid or dextrose. The nanoparticles were prepared by layering Au onto either a cubic or octahedron-shaped Cu-oxide crystal via a galvanic reaction. These were characterized with scanning electron microscopy, energy dispersive X-ray and X-ray photoelectron spectroscopy. The porous structure of the gold over layer was clearly visible on the scanning electron microscopy image while the macro morphology was maintained. X-ray photoelectron spectroscopy confirmed the presence of metallic gold while both CuI (CuO) and CuII (Cu2O) were present in the samples. These two Au/Cu-oxide nanocomposites were used to modify glassy carbon electrodes and were tested for their dopamine sensing ability. Differential pulse voltammetry was used to investigate the selectivity towards dopamine in the presence of different interfering molecules uric acid, ascorbic acid and dextrose). From the differential pulse voltammetry, the lowest limit of detection was found to be 1.1 μM, with a sensitivity of 3.4 μA mM−1 mm−2 in the linear range of 10–250 μM for the porous gold layered covering the octahedron Cu-oxide-modified glassy carbon electrode.


Introduction
Dopamine (DA) is neurochemical, which transmits signals between neurons. 1 DA influences many physiological conditions (such as mood, memory, motor control, cognitive functions, etc.) 2 and organ systems (cardiovascular, central nervous, renal and endocrine systems). 3Irregular blood levels of DA induce the onset and advancement of a number of neurological diseases, such as Parkinson's, Alzheimer's, attention deficit and hyperactivity disorder, Huntington's, schizophrenia and addiction. 4Therefore, the detection in a highly sensitive and selective approach is critical for the early detection of the aforementioned diseases.
4][15] During the electrochemical oxidation of DA, dopaminoquinone is formed through a one-step two-electron transfer process. 16This process, however, this is not reproducible on bare electrodes. 17To circumvent this complication, nanoparticle-modified electrode surfaces have been employed due to the improved chemical, thermal and mechanical properties the nanoparticles provide as well as large their surface areas.Nano-sized metal oxides electrochemical sensors resulted in the detection of biomolecules with good sensitivity and stability.Cu 2 O nanomaterials have been used in the detection of glucose, 18,19 tryptophan, 20 and DA. 21,224][25] The octahedral exposing the (111) facet showed superior sensing ability towards glucose over the cubic morphology. 23ue to the high conductivity of gold nanoparticles, they have been successfully employed for DA sensing, [26][27][28] using nanoporous gold structures results in an even higher electrocatalytic activity due to the increased surface area. 29,30][33] Based on the background presented above, in this work, porous nano gold layered on a cubic or octahedral Cu 2 O nanocrystals is proposed for the electrocatalytic detection of DA.The two Cu 2 O nanocrystals were prepared by colloidal synthesis.The nanoporous gold structure was layered onto the Cu 2 O nanocrystals via galvanic reaction in a surfactant-free oil-in-water emulsion.The porous Au/ Cu-oxide nanoparticles-modified glassy carbon electrode (GCE) was tested for the sensing of DA alone and in the presence of interfering biological molecules.

Results and discussion
Cubic and octahedron shaped Cu-oxide nanocrystals (having an average edge length of 1.14 and 1.31 μm, respectively, see Supplemental Figure S1 for the scanning electron microscopy (SEM) images) were prepared by established colloidal synthesis. 34A porous gold structure was deposited onto the cube and octahedron morphology of the Cu-oxide nanocrystal templates during a galvanic reaction in a surfactant-free oil-in-water emulsion. 35Hexadecane were adsorbed onto the surface of the Cu-oxide surface forming interfacial nanodroplets, which protected the surface of the Cu-oxide.Only the uncapped Cu-oxide surface was involved in the galvanic reaction. 36The progression of nanodroplet adsorption and the galvanic gold layering on the cube and octahedron morphology is shown by the SEM images in Figure 1.
After 6-h reaction time, no visible changes were observed for the cubic morphology.The SEM image after 12-h reaction time showed nanodroplet adsorption with an average size of 190 nm adsorbed onto the surface of the (100) facet.The energy dispersive X-ray (EDS) of the sample after 12 h (Supplemental Figure S3) revealed the presence of approximately 5.5% Au (compared the 63% Cu), indicating that the galvanic reaction has commenced, whereas the EDS after 6 h (Supplemental Figure S2) showed no Au present.After 24 h, etching of the Cu and formation of a porous structure can be observed, with the EDS (Supplemental Figure S4) indicating 13.2% Au (as compared to 58.4% Cu).
The SEM image of the octahedron morphology in Figure 1 illustrates first signs of nanodroplet adsorption and formation of the porous gold structure at 6 h.The EDS (Supplemental Figure S5) confirmed this with 33.2% C (the hexadecane) and 2.8% Au (compared to the 43.4% Cu).The presence of the Cl (3%) implied that some of the Au molecules might still be ionic and has not deposited yet.The adsorption of the nanodroplets is significantly faster than the adsorption on the cubic morphology, which is in correlation with the reported surface energy of the (111) facet being lower than the (100) facet, 0.76 and 1.26 J m -2 . 37The SEM taken at 12 and 24 h (Figure 1) illustrates the advanced etching of the Cu-oxide nanoparticles making way for the porous gold structure to form, with the EDS showing an increase in Au concentration of 15.8% and 35.1% (compared to the 53.2% and 42.1% Cu present).Like the cubic structure, the global octahedron morphology of the nanoparticle was retained.
X-ray photoelectron spectroscopy (XPS) was used to analyse the elemental composition and to confirm the oxidation state of the Au and the Cu in the porous Au layer on the cubic (C) and octahedron (O) morphology of the Cu-oxide after 24-h reaction time.Figure 2(a) shows the XPS of the Cu 2p area (for both C and O), two intense sharp peaks can be seen at approximately 933 and 953 eV, with shape-up peaks at approximately 11 eV higher than the main peaks.Since the Cu 2p 3/2 and the Cu 2p 1/2 photoelectron lines provide the same information, only the Cu 2p 3/2 will be discussed.To accurately determine the oxidation state(s) of the Cu, the proposed fitting parameters of Biesinger was employed. 38From the deconvoluted Cu 2p 3/2 photoelectron lines shown in Figure 3(b) and the data summarized in Table 1, it can be seen that metallic Cu, Cu I (in CuO) and Cu II (in Cu 2 O) are present in both C and O.This suggests that the samples' surface is composed of metallic Cu, Cu 2 O and CuO.In analogy to reports, 39 during the galvanic reaction, the Au on the surface promotes the formation of CuO (Cu II ), explaining the presence of the mixed oxidation state present on the surface.From the Au 4f spectra in Figure 2(c), the Au 4f 7/2 and Au 4f 5/2 photoelectron lines are located at approximately 84.3 and 87.8 eV, respectively, these values are consistent with the Au being metallic. 40rom the atomic ratio between the Cu and Au (which is an indication of the amount of Au deposited on the Cu-oxide nanocrystal), presented in Table 1, it is clear that more Au deposited on O than on C. It is reported that the (100) facet is only terminated with O atoms, while for the (111) facet between two Cu atoms at the surface a dangling bond perpendicular to the (111) facet is present. 41This renders the Cu atoms in the (111) facet coordinatively unsaturated, possibly explaining why the Cu I on (111) surface is more inclined towards oxidation to Cu II .Since the Au on the surface promotes the formation of CuO (Cu II ), the atomic ratio between Cu I and Cu II on the surface might be influenced by the amount of Au present.This is in fact the case.The atomic concentration of the Cu I and Cu II on the surface of C, were the same, with the metallic Cu in the minority (see Table 1).For O, the atomic concentration of Cu II is almost double the amount of Cu I , with the metallic Cu again being in the minority.The higher relative concentration of Cu II  (CuO) present on the surface of O, is consistent with the presence of a higher concentration of Au (relative to the Cu on the surface).This implies that the increased amount of Au on the surface on O resulted in an increased formation of Cu II .
The cyclic voltammetry (CV) curves of the modified electrodes C/GE and O/GE in 0.1 M phosphate-buffered saline (PBS) solution at pH 7.0 without any analyte showed an increase in the current from 750 to 1500 mV, this increase in current was less than 1 μA and will thus not interfere with the detection of DA, which will be measured in the potential window of -100 to 700 mV (shown in the Supplemental material).
Cyclic voltametric studies were conducted to investigate the electrochemical response of DA, UA, AA and dextrose (DX) on both the C-and O-modified GC electrodes in 0.1 M PBS solution (pH = 7.0).The CV curves of DA, UA, AA and DX on the two different electrodes are shown in Figure 3. Apart from the UA, none of the other biological molecules showed any electrochemical response over the unmodified GCEs, this is in correlation with literature, which also showed an oxidation wave at approximately 300 mV of unmodified GCE. 42The oxidation of DA over the C-modified GCE (C/GC) showed the best electrochemical response, with the E pa = 180 mV and i pa = 7 μA, whereas the response over O-modified GCE (O/ Table 1.The binding energy as measured by XPS for all the simulated peaks for the Cu 2p 3/2 as well as the Au 3f 7/2 photoelectron lines, all measured in eV.The ratio between the Cu 0 , Cu I and Cu II as well as the atomic ratio between the Cu and Au are also reported.GC) resulted in E pa = 400 mV and i pa = 1 μA.Data reported on the detection of DA over an unmodified and a Cu x Omodified Au-electrode revealed that the modified electrode (containing both Au and Cu x O) resulted in a lower oxidation potential but the same current response as the unmodified Au-electrode. 43In correlation with this (a mixture of Cu x O and Au, thus less Au), we can possibly conclude that seeing as the C contains less Au, it resulted in better detection of the DA.The AA oxidation was observed at 162 and 105 mV over C/GC and O/GC, respectively.Similar to the AA, DX's electrochemical response occurred at a lower potential over O/GC (47 mV) than over C/GC (146 mV), in both cases, the peak current was poor.In the case of the UA oxidation, again a lower oxidation potential was observed over O/GC (86 mV) than over C/GC (198 mV).Compared to unmodified GCE, all the biological molecules displayed good electrochemical response over the modified electrodes.The results indicate that O/GC displayed better electrocatalytic activity for the oxidation of AA, DX and UA as compared to C/GC.

Cu
The detection of AA and UA over C/GE, displayed a small anodic wave related to the oxidation wave, this is not uncommon for the sensing of AA and UA over modified GCE. 44fferential pulse voltammetry (DPV) was used to obtain better sensitivity, since its results in enhanced signals by removing the non-Faradaic current.
To investigate the sensitivity and limit of detection (LOD) of the C/GC and O/GC, DPV was measured of DA solutions containing a set concentration of 10 μM of either UA, AA or DX, while the concentration of the DA was systematically increased (from 10 to 250 μM). Figure 4 shows these DPV curves alongside the corresponding calibration curves.The DPV peak maxima for DA, UA, AA and DX appeared at approximately 180, 48, 36 and 60 mV over C/ GC and 162, 16, 32 and -84 mV over O/GC, respectively.From this, it is clear that O/GC displayed better detection for DA, AA, DX and UA as compared to C/GC.It has been reported by Aditya et al. that the (100) surface of a cubic Cu 2 O particle is neutral, while the (111) surface of an octahedron Cu 2 O is positive (implying it is acidic). 41,45The pK a of the biological compounds used during this research are DA = 9.44, 10.75 and 12.80, 46 AA = 4.10 and 11.6, 47 DX = 12.43, 48and UA = 2.3 and 5.4. 46According to Gilli et al., 49 whose lab investigated hydrogen bond strength based on pK a , concluded from the pK a equalization principle that the hydrogen bonding (D-H . . .A) increases as the pK a separation (ΔpK a = pK a (D-H)-pK a (A-H + )) between the H-donor and H-acceptor increases.Considering that the C is neutral and O is acidic, the ΔpK a between O and the biological molecules is higher than for C.This implies that stronger hydrogen bonding exists between O and the biological molecules.
The oxidation potential of DA, AA, DX and UA was detected at a lower potential over O/GC.It can thus be concluded that stronger hydrogen bonding is associated with lower oxidation potential (better detection).This is in agreement with reported data, where an increased amount of hydrogen bonding between a biological molecule and the electrode resulted in a negative shift in the oxidation potential. 50he DPV peak separation (see Table 2) between the DA and the other biomolecules, such as UA, AA and DX, is sufficient large to selectively detect DA without interference of the other biomolecules.A linear directly proportional relationship was established between the peak current for the DA and its concentration.
The LOD was calculated by equation ( 1): 31 where s represents the standard deviation of the blank voltammogram and m is the slope of the calibration curve.The LOD (measured in μM) of C/GC and O/GC as well as the sensitivity (measured in μA mM -1 cm -2 ) are summarized in Table 2.
Sensing of DA over O/GC in the presence of interfering biomolecules UA and DX resulted in better peak separation between the DA and the interfering molecule in addition to being more sensitive towards detection in comparison to C/ GC.As for the LOD (in the presence of UA, AA and DX), O/GC is an order of magnitude better than C/GC.The superior sensing of DA of O/GC over C/GC is attributed to the higher percentage concentration of Au present in O/ GC.
The effectivity of O/GC and C/GC is comparable to some other NP-based sensors for the detection of DA (see Table 2).

Conclusion
In summary, cubic-and octahedron-shaped Cu-oxide nanocrystals were used as well-defined morphological templates for the galvanic etching and deposition of porous Au layers.The SEM images indicated that the surfactant-free hexadecane-in-water nanodroplet adsorption as well as formation of the Au layer occurred faster on the (111) facet of the octahedron morphology.This was affirmed by the % Au present in the samples at different times.The samples contained metallic Au and mixed oxidation states of Cu (0, I and II) were detected by XPS.GCEs modified with the Au/ Cu-oxide nanocomposites were tested for their DA detection capability in the presence of interfering molecules (UA, AA and DX).The GC electrode modified with the porous Au layered onto the octahedron shaped Cu-oxide crystals resulted in superior DA detection.

Experimental section
All reagents, both solid and liquid, were reagent grade, purchased from Sigma-Aldrich and used without further purification.Distilled water was used throughout.The cubic and octahedron Cu-oxide nanocrystals were prepared according to published procedures. 34,58

Preparation of porous Au layer on the Cuoxide nanocrystals
The preparation is adapted from the original procedure of Ma et al., 35 for nanoparticles in suspension.

Preparation of Hexadecane-in-H 2 O emulsion via a solvent-shifting method
Hexadecane (1 ml) was added to acetone (9 ml) resulting in a volume fraction of 0.1%.The above solution was then injected into a 200 ml water to create an emulsion.

Preparation of porous Au nanoparticles
The appropriate Cu-oxide nanocrystal (~20 mg) was dispersed into 20 ml of the hexadecane-in-H 2 O emulsion for 20 min, allowing time for nanodroplet adsorption onto the nanocrystals.This was followed by dropwise addition of HAuCl 4 (2 ml, 2.9 mM) to the mixture and subsequent incubation for 6, 12 and 24 h, respectively.The samples were centrifuged after the addition of ethanol to prevent the freezing of hexadecane while centrifuging.The collected nanoparticles were dried overnight at 60°C.

Characterization
The crystalline phase of the samples was characterized by the X-ray diffractometer (Bruker D8-ADVANCE) with Cu-Kα (λ = 1.54060Å) operated in a continuous θ-θ scan in locked coupled mode with Cu-K radiation.Measurements ran within a range in 2θ defined by the user with a typical step size of 0.034° in 2θ.A position sensitive detector, Lyn-Eye, was used to record diffraction data at a typical speed of 0.5 s/step, which is equivalent to an effective time of 92 s/step for a scintillation counter.
For SEM, a JEOL-JSM7800 Field Emission Scanning Electron Microscope was employed.The electron beam energies in the range of 5 keV.
XPS data were recorded on a PHI 5000 Versaprobe system with a monochromatic AlK X-ray source.Spectra were obtained using the aluminium anode (Al Kα = 1486.6eV) operating at 50 μm, 12.5 W and 15 kV energy (97 X-ray beam).The survey scans were recorded at constant pass energy of 187.85 eV and region scans at constant pass energy of 29.35 eV with analyser resolution ⩽ 0.5 eV.The background pressure was approximately 2 × 10 -8 mbar.The XPS data sets were analysed utilizing Multipak version 8.2c computer software with Gaussian-Lorentz fits.

DA detection
DA detection was investigated by means of electrochemical measurements.CV and DPV were conducted on a BAS 100 B/W electrochemical workstation.A three-electrode cell was used, utilizing a Pt auxiliary electrode, an Ag/Ag + reference electrode and a modified glassy carbon working electrode with a surface area of 3.14 mm 2 .The pre-electrocatalyst suspension is prepared by adding the Au/Cu-oxide nanoparticle (20 mg) in MeOH (methanol) (0.5 ml), followed by sonication.The pre-electrocatalyst suspension (20 µl) is drop-coated onto the electrode surface.After drying, Nafion (20 µl) was dropped over the modified surface and allowed to air-dry.CV was conducted in a solution of 10 µM DA and 0.1 M Dulbecco's PBS.The potential was scanned from -300 to 600 mV using a scan rate of 50 mV s -1 .DPV was used investigate the selectivity towards DA detection in the presence of interfering biomolecules.The DPV was again scanned from -300 to 600 mV, with a pulse width and amplitude of 0.05 s and 50 mV, the pulse period was 200 s.The solution for the DPV was a mixture of 0.1 M PBS and 10 µM of either UA, L-AA or DX, with varied amounts of DA (10,  50, 100, 150, 200 and 250 µM).

Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Figure 1 .
Figure 1.SEM images illustrating nanodroplet adsorption and the galvanic gold deposition on the Cu-oxide cubic morphology after (a) 6 h, (b) 12 h and (c) 24 h and on the Cu-oxide octahedron morphology after (d) 6 h, (e) 12 h and (f) 24 h.The scale bar applies to (a)-(f).

Figure 3 .
Figure 3.The CV curves of the response of (a) 10 μM DA, (b) 10 μM AA, (c) 10 μM DX and (d) 10 μM UA, over C/GC and O/GC in 0.1 M PBS solution at a scan rate of 50 mVs -1 .

Figure 4 .
Figure 4.The DPV curves showing the current response of increasing concentration of DA in (A) a 10 μM UA solution over C, (B) a 10 μM UA solution over O, (C) a 10 μM AA solution over C, (D) a 10 μM AA solution over O, (E) a 10 μM DX solution over C, (F) a 10 μM DX solution over O, with the corresponding calibration curves of DA in (a) a 10 μM UA solution over C, (b) a 10 μM UA solution over O, (c) a 10 μM AA solution over C, (d) a 10 μM AA solution over O, (e) a 10 μM DX solution over C and (f) a 10 μM DX solution over O.

Table 2 .
The DPV peak separation, limit of detection, linear range and the sensitivity of C/GC and O/GC.Also presenting comparative data with already reported results.
C/GC: C-modified glassy carbon electrode; UA: uric acid; O/GC: O-modified glassy carbon electrode; AA: ascorbic acid; DX: dextrose.a This paper.b Could be higher.