Biosynthesized CuO nano-platelets: Physical properties & enhanced thermal conductivity nanofluidics

2.1 Biosynthesis process via Callistemon viminalis’ aqueous extracts

Callistemon viminalis commonly known as bottle brush in view of its red flowers shaped in the form of traditional bottle brushes is a plant originally from Australia. Several studies showed that the red-dye extract obtained from the red Callistemon viminalis flowers is rich in flavonoids, saponins, steroids, alkaloids and triterpenoids (Salem et al., 2013). A summary of the isolated active compounds is reported in Table 1. The acid compounds such as the Betulic or the Oleanolic acids can act as bioactive agents in reducing metal-ion based precursors.

For the synthesis, fresh red flowers of Callistemon viminalis were collected from the Western Cape site of iThemba LABS, South Africa. 6.65 g of the cleaned red parts of the bottle-brush shaped flowers was heated in 250 ml of de-ionized water between 75 and 85 °C for 2.5 h yielding a red coloured extract of pH ∼3.5. Upon cooling, the filtered red aqueous extract was added to 2.8 g of CuSO₄—5H₂O (Sigma–Aldrich, 97%). At room temperature, while swirling the Cu salt was observed, within 5 min, to dissolve completely in the red aqueous extract forming a greyish brown suspension. The resultant suspension was allowed to settle over a period of 1–2 h after which a green coloured supernatant over a brown/grey precipitate (presumably a mixture of CuO and Cu) was observed. The precipitate was separated from the aqueous solution, first by decanting and then by centrifuging at 3500 rpm 3 times over successive additions of de-ionized water. The final precipitate was dried at 200 °C (15 min) yielding a brown-coloured precipitate which when heated in air at 500 °C for 2 h gave rise to a brown-black/grey-black powder (presumably CuO). A schematic illustration of the biosynthesis process is seen in Fig. 1. In the following experimental sections, the focus will be on the sample annealed at 500 °C as it is at this threshold value that the formation of crystalline CuO could be confirmed. Annealing the powders at 500 °C in air would help to compensate for O deficiencies if any.

Close

Figure 1

Schematic representation of the biosynthesis of CuO nanoparticles using aqueous extracts of Callistemon viminalis.

Export to PPT

2.2 Structural, luminescent and thermal conductivity characterization techniques

Surface morphology of the prepared powders was carried out using an Auriga High Resolution Scanning Electron Microscope (HRSEM) coupled to an Oxford instruments X-Max solid state Silicon drift detector (20 keV) used for carrying out elemental analysis by Energy Dispersive X-ray Spectroscopy (EDS). High Resolution Transmission Electron Microscopy (HRTEM) was carried out in a TECNAI F2 G20 HRTEM. The sample subjected to HRTEM was obtained by taking a swab of the thin film dissolved in ethanol using a Cu grid holder. Electron Dispersive X-ray Spectroscopy (EDS) was carried out using a specialized detector coupled to an OXFORD instruments nano-Scanning Electron Microscope. To determine the crystal structure of the prepared samples X-ray Diffraction (XRD) analysis was carried out using a Bruker Advanced D8 diffractometer with monochromated Cu Kα radiation of 1.5406 Å operating in the Bragg–Brentano geometry. The zeta potential of the samples annealed at 500 °C was determined using the Malvern Zetasizer Nano ZS. Smoluchowski’s model was used to calculate the zeta potential from measurements of the electrophoretic mobility.

To ascertain the presence of metal-oxo, metal-hydroxy bonds as well as detect the presence of other compounds surface bound to the expected CuO particles, Fourier-Transform Infrared spectroscopy was carried out in Attenuated Total Reflection (ATR) mode with a diamond crystal employed for measurements. The Attenuated Total Reflection was obtained within a 400–4000 cm⁻¹ range using a Perkin Elmer 100 Spectrometer. For determining the vibrational modes in Cu—O based samples Raman spectroscopy was carried out using the 514.5 nm excitation line of an Ar⁺ laser source in the spectral range of 300–700 nm. X-ray Photoelectron Spectroscopy useful in determining the chemical states of Cu and O ions in the samples as well as ascertaining their purity was acquired using a constant 50 eV pass energy mode, in 0.1 eV increments at a 50 ms dwell time with the signal averaged for, at several regular scans. The XPS system was equipped with a dual Mg Kα–Al Kα anode for photoexcitation. The spectrum was calibrated with binding energy (BE) 284.5 eV for C1s electron.

Photoluminescence measurements were carried using a fibre-optics linked OCEANOPTICS system consisting of a high-powered UV Light-emitting diode source (300 nm) coupled to a high sensitivity QE Pro-FL spectrometer that had a signal-noise ratio >1000:1 and operated in a wavelength range of 185–1100 nm. Thermal conductivity of the CuO-water nanofluids was performed on a standard KD2 (Decagon Devices, Inc., USA). The measurements were performed at room temperature (25 °C) following initial tests on high quality ethylene glycol (Sigma–Aldrich) and de-ionized water for the calibration phase.

3.1 Surface morphology, elemental analysis and growth mechanism

SEM-EDS results (Fig. 2) on the brown CuO powder were obtained from drying at ∼200 °C, the brown precipitate emanating from the CuSO₄-containing extract suggested the presence of agglomerated CuO, and possibly Cu nanoparticles were still held by some amount of glazed amorphous content. SEM analysis using Image J software gave the average particle size distribution (inset in Fig. 1) to be 14.9 ± 3.8 nm for about 71% of the measured particles and 42.4 ± 4.9 nm for the rest. Annealing of this brown CuO powder in air at 500 °C yielded a grey-black powder which under SEM revealed the formation of dense cuboids and hierarchical platelet-containing agglomerates. EDS confirmed the presence of Cu and Oxygen in the powders. The presence of C and S observed in the powder could be ascribed to the carbon film used to support the CuO samples as well as residual contributions from the CuSO₄ precursor and flower-based organic compounds that emanated from the aqueous extract.

Close

Figure 2

SEM micrograph for biosynthesized CuO nanoparticles annealed at (a) 200 °C for 2 h and (b) 500 °C, for 2 h.

Export to PPT

Fig. 3 reports a Transmission Electron Microscopy (TEM) image while Fig. 4a reports Electron Dispersive X-ray Spectroscopy (EDS) results of the brown CuO_x powder obtained from drying at ∼200 °C and heating in air at 500 °C for ∼2 h. The formed particles are highly agglomerated and nanoscaled within the size range from 19 to 200 nm as observed through Transmission Electron Microscopy (SEM). This effective agglomeration could be a severe disadvantage for potential technological applications. In this regard, a Zeta potential investigation was carried out after ultrasonication of the annealed CuO powders in water for 3–5 min. From this the zeta potential (Fig. 4b) was found to be 0.0758 ± 4.63 mV for CuO powder in water. This value shows that the CuO nanoparticles are of very low stability in water as was confirmed by their tendency to flocculate and sediment over time under 15 min, hence the need to further stabilize them using Polyvinylpyrrolidone. The tendency to aggregate/flocculate is due to strong attractive electrostatic forces that occur between the CuO nanoparticles.

Close

Figure 3

Transmission electron micrograph of the biosynthesized CuO powder annealed at 500 °C in air, for 2 h.

Export to PPT

Close

Figure 4

(a) Electron Dispersive X-ray Spectroscopy spectrum of biosynthesized nanoplatelet-containing CuO powder annealed at 500 °C, for 2 h. (b) Typical zeta potential measurement for biosynthesized CuO powder annealed at 500 °C, for 2 h.

Export to PPT

3.2 Crystal structure elucidation using XRD analysis

The powdered samples of the as centrifuged precipitate, dried at 200 °C and annealed at 500 °C respectively were deposited onto float-glass substrates. The as-centrifuged and dried-at-200 °C samples both exhibited a semi-amorphous state while the annealed powder at 500 °C displayed a crystalline structure (Fig. 5). The XRD spectrum (Fig. 5) observed presents 3 major Bragg peaks at 2 $ϴ$ ° ∼32.11°, ∼35.76°, and ∼38.96° which are ascribed to reflections from the (−1 1 1), (0 0 2) and (1 1 1) reticular planes respectively according to the JCPDS card 41-0254. Minor peaks include those at 48.86°, 53.35°, 58.39°, 61.93°, 66.96° and 67.99° which can be indexed to reflections from (−2 0 2), (0 2 0), (2 0 2) (1 1 −3), (−3 1 1) and (1 1 3) planes. Hence the crystallographic structure of the CuO_x nanocrystals is pure Tenorite CuO belonging to the bcc monoclinic Bravais lattice with space group C2/c (C_2h⁶) and unit cell lattice parameters a = 4.65, b = 3.41 and c = 5.11 where a/b = 1.36 and c/b = 1.50 ± 0.02. The broadness of the Bragg peaks sustains the nanoscaled aspects of the Tenorite CuO crystallites. The crystallite size of the CuO nanoparticles could be estimated using the Debye–Scherrer approximation. The average size of the CuO nanocrystals is ∼22 nm. It is worth mentioning that in addition to the broad peak observed between 20° and 35° 2 $ϴ$ emanating from the glass substrate, 2 other peaks observed at ∼25.74° and ∼15.58° may be attributed to the presence of carbon.

Close

Figure 5

X-ray diffraction pattern of the CuO nanoplatelet-containing powder indicating the pure Tenorite structure.

Export to PPT

3.3 Vibration spectroscopy investigations using FT-IR and Raman spectroscopy

To confirm the nature of the CuO Tenorite of the nanocrystals, both Attenuated Total Reflection-FT-IR (ATR-FTIR) and Raman spectroscopy investigations were conducted. Fig. 6 reports a typical ATR-FTIR spectrum of the brown CuO powder obtained from drying at ∼200 °C and heated in air at 500 °C for ∼2 h. The spectrum is split in 2 regions: the first and the second are lying between 400–800 cm⁻¹ and 2100–2700 cm⁻¹ respectively. The strong vibrational modes observed at 478.8, 529.3 and 578.6 cm⁻¹ are typical of Cu—O single bonds in the bending mode (Xu et al., 2010). The weak bands observed at 2458.5 and 2493 cm⁻¹ can be attributed to Cu—O bonds in the stretching mode. From the point of view of group theory the observed modes centred at 529 and 578.6 cm⁻¹ are attributed to E_u and A_2u modes respectively. These 2 modes are characteristic of the pure Tenorite CuO phase (Xu et al., 2010). The absence of bands in the 1600 and 3300 cm⁻¹ suggests the absence of waters of hydration/moisture within or at the surface of the CuO powders.

Close

Figure 6

ATR/FT-IR spectrum of the CuO nanoplatelet-containing powder indicating the pure Tenorite structure.

Export to PPT

Based on XRD analysis, CuO prepared by this green synthesis is in the base-centred monoclinic phase. The Cu²⁺ ions are found at centres of inversion symmetry in a single fourfold site 4c (1/4, 1/4, 0), while oxygen ions occupy site 4e (0, y, 1/4) (Zhang et al., 2014). Monoclinic CuO from group analysis therefore has twelve zone-centre optical phonon modes with symmetries 4Au + 5Bu + Ag + 2Bg, of which only the Ag and 2Bg modes are Raman active (Zhang et al., 2014; Mukherjee et al., 2011; Xu et al., 1999; Reimann and Syassen, 1990). Raman peaks for monoclinic CuO crystal with a space group of C2/c have thus been shown to be the Ag, B¹g, and B²g phonon modes which appear at 303, 350, and 636 cm⁻¹, respectively (Chrzanowski and Irwin, 1989). All three phonon modes can be observed in the Raman spectrum of the annealed CuO nanopowders (Fig. 7) we report. An intense peak at 293.2 cm⁻¹ can be attributed to A¹g vibrations while a shoulder peak at 341.6 cm⁻¹ can be attributed to B¹g vibrations. A broad peak of lower intensity at 627.9 cm⁻¹ can be attributed to B²g phonon modes. These results, which are close to those reported by Chou et al. (2008) and Das et al. (2013), exhibit a slight red shift of ∼9 cm⁻¹ from bulk CuO. This red shift as well as the broad feature of all three peaks could be ascribed to a quantum confinement effect (Xu et al., 1999) or to strain/stress surface phenomena. The absence of Raman modes that can be attributed to Cu₂O suggests the CuO nanopowder is of high purity.

Close

Figure 7

Room temperature Raman spectrum of the CuO nanostructured powder confirming the pure Tenorite structure.

Export to PPT

3.4 Chemical valence states via XPS studies

To confirm again the absence of Cu₂O phase and the purity of the Tenorite CuO nano-particles, X-ray Photoemission Spectroscopy (XPS) investigations (Fig. 8) were carried out. High resolution energy scans on the sample exhibited two main peaks at BE 933.5 and 953.5 eV corresponding respectively to Cu 2p3/2 and Cu 2p1/2 valence states of CuO (Krishnamoorthy and Kim, 2013). The splitting between these two states is about 20 eV. This is due to the formation of CuO nanostructures. The Cu 2p3/2 at 933.5 eV has a satellite peak with a doublet at 942 and 944 eV which is in agreement with the reported data (Krishnamoorthy and Kim, 2013). The BE at 539 eV corresponds to O1s of CuO and not Cu₂O and hence is in agreement with data reported by Krishnamoorthy and Kim (2013). The observed XPS profile strongly suggests the absence of Cu₂O, and also Cu(OH)₂ impurities within the investigated sample.

Close

Figure 8

High-resolution XPS spectra of the CuO nanoplatelet-containing powder: (a) Cu 2p and (b) O1s.

Export to PPT

3.5 Photoluminescence of CuO nanoparticles

The photoluminescent properties of the CuO nanoparticles dried in air at 200 °C for 10–12 h were investigated with the aid of a fluorescence spectrometer by exciting with ultraviolet light of wavelength 300 nm. The resultant emission spectrum (Fig. 9) showed the occurrence of a strong UV emission peak at 356 nm which can be ascribed to band edge emissions that result from the recombination of electrons and holes of CuO free excitons (Dagher et al., 2014). When compared to bulk CuO UV-emission is blue shifted and can be ascribed to a quantum confinement effect which is a consequence of the presence of nanoplatelets in the CuO powders (Mohammed et al., 2014). A violet-blue luminescent band (shoulder peak) observed at ∼418 nm is a typical near-band-edge emission peak of CuO (Sathyamoorthy and Mageshwari, 2013; Chang et al., 2005). The shoulder peak observed at 430 nm may be ascribed to the existence of Cu vacancies in CuO which is a p-type semiconductor. The yellowish-green emission peak at 561 nm can be ascribed to deep level defects. These defects are said to occur only at low temperatures. The red emission peaks at 664 nm can be ascribed to multiple oxidation states of copper or the presence of singly ionized oxygen vacancies (Mohammed et al., 2014; Sathyamoorthy and Mageshwari, 2013). The broad nature of visible emissions encountered here in the violet-blue, yellowish-green and red regions of the visible spectrum suggests that CuO particles investigated here have a high surface-to-volume ratio with numerous surface-states and defects (vacancies and interstitials) which create trap levels responsible for the emissions observed (Mohammed et al., 2014; Huang et al., 2010).

Close

Figure 9

Photoluminescence spectrum of CuO nanoparticles annealed in air at 200 °C, with an excitation source of wavelength 300 nm.

Export to PPT

3.6 CuO—H₂O nanofluids for enhanced thermal conductivity

One of the most prominent recent applications of nano-scaled CuO is within the nanofluid field (Farbod et al., 2015; Khodadadi et al., 2013). Indeed as indicated in Fig. 10, the thermal conductivity of CuO is amongst the highest of oxides approaching nearly 90 W/m K. Its use in the form of nano-dispersions in several fluids such as water, ethylene glycol, and mineral oil has clearly demonstrated a significant enhancement of the corresponding thermal conductivity. In this regard, several dispersions of the CuO nanoparticles in de-ionized water were prepared with volume fraction varying from 0.10% to 0.9%. To stabilize them, 15% by weight of Polyvinylpyrrolidone relative to the solid nanoparticle powder was added. Their initial dispersion was carried out in a 400 W ultrasonic power unit (UP400S, 400 W, 24 kHz) for ∼2 h. Table 2 and Fig. 11 report the experimental values of k_eff versus the volume fraction at room temperature. As one can notice, there is a net increase in the effective thermal conductivity with the nanoplatelets’ concentration. Yet, it is not a linear relationship, and k_eff increases from ∼0.623 to approximately ∼0.817 W/m K. This corresponds to an enhancement from nearly 6% to 34%. These results are comparable to those of Khedkar et al. (2012) hence confirming the possibility of using the current green chemistry synthesized CuO nanoplatelets for enhanced thermofluid applications. It is necessary however, to mention that the CuO nanoparticles-water dispersions are stable up to 9% volume fraction. This could be due to the 2-D morphology of the CuO nanoplatelets present in these particles, and their surface/interface interaction with the polar H₂O molecules. The follow-up investigations would consist of studying their stability over time, their thermal cycling as well as duplicating the same approach with other nano-scaled oxides synthesized by a similar method (Sone et al., 2015; Thema et al., 2015; Thovhogi et al., 2015; Diallo et al., 2015a,b).

Close

Figure 10

Thermal conductivity of various organic materials, heat transfer fluids, metals and oxides including CuO.

Export to PPT

Close

Figure 11

Room temperature evolution of the effective thermal conductivity of nanoplatelet-containing CuO powder dispersed in PVP-water solution versus their volume fraction.

Export to PPT

Also, for future work, as with previously synthesized nano-oxides prepared using Callistemon viminalis, Agosthoma betulina, Aspalathus linearis and Hibiscus sabdarifa natural extracts (Sone et al., 2016; Thema et al., 2015; Thovhogi et al., 2015; Diallo et al., 2015a,b), it is necessary to use adequate techniques such as HPLC and GC–MS to identify the bioactive molecules present in the natural extracts as well as elucidate the reaction mechanisms that govern the interaction of the bioactive molecules with the precursor Cu salt.