1. Introduction
With the fast development of nanotechnology in the last decades, the realm of nanostructures has unveiled all its beauty, giving unimaginable quantum systems, spanning all fields of physics, chemistry, and biology.
The autonomous ordering and self-assembly of atoms and molecules on atomically well-defined surfaces [
1], two dimensional (2D) elemental materials such as graphene [
2], silicene [
3] and all those 2D beyond graphene [
4,
5,
6], including van der Waals 2D heterostructures [
7], transition metal dichalcogenides [
8], perovskite systems [
9,
10] and bio-polydopamine composite materials [
11] are only a small part of this, inaccessible to naked eye, extraordinary world.
The close conjugation between the atomic and nanoscopic scale of materials and their physical properties, mainly derived from the 3D, 2D, 1D, or 0D quantum size confinement [
12,
13,
14], directly takes us into the mechanics applied to nanoscience and its complexity [
15,
16].
So far, the world of chemistry and biology has been able to strongly embrace that of physics, revealing unexpected structural, electronic, and catalytic properties of nanostructured materials that have ubiquitously flowed into the daily life, through devices and sensors.
To date, cupric oxide (CuO) has been diffusively studied thanks to its great potential applications in various fields, such as catalysis [
17], solar cells [
18], optical and photoconductive response [
19]. It could be used, also, for its field emission properties [
20], being a p-type semiconductor with a narrow band gap of 1.2 eV. One of the CuO fascinating features is its ability to form nanostructures that include many shapes, such as flower-like structures composed of hierarchical 2D nanosheets and spherical architectures. They are constituted of ultrathin nanowalls, of about 10 nm in thickness [
21,
22], or large honeycombs assembly on copper foils [
23]. In the context of gas sensing, these CuO flower-like nanostructures are applied as electrode sensors capable of having a high non-enzymatic electrocatalytic activity for hydrogen peroxide detection [
24,
25].
To produce these extraordinary flower-like CuO nanostructures, chemical methods were typically used [
21,
22,
23,
24,
25] with copper mainly treated in alkaline conditions by persulfate [
21,
22,
23,
24,
25]. In these studies, the (-111) X-ray diffraction (XRD) reflection at 2θ = 35.744° [
21,
24,
25] and the Raman peak at 295 cm
−1 [
26] were utilized to identify the run products.
CuS nanoparticle clusters and sponge-like structures of nanoflakes were also synthesized by the chemical method by using copper chloride and thioacetamide as reagents [
27,
28]. These CuS were studied by scanning electron microscopy (SEM), XRD, and Raman spectroscopy to characterize their nanostructure morphology. In particular, XRD peaks specific to the hexagonal CuS structure [
27] and the sharp Raman peak at 473 cm
−1 assigned to the S-S stretching mode of S
2 ions at the 4e sites [
27,
28,
29,
30] were used for featuring the obtained compound.
Peculiar copper sulphide nanostructures, such as spheres and nanotubes, were successfully synthesized, by using a microwave-assisted solvothermal method on Cu complexes [
29]. Indeed, pure CuS (hcp) with flower-like, spherical hollows and tubular structures were obtained by means of CuCl
2•2H
2O and CH
3CSNH
2 dissolved in ethylene glycol, by varying the pH by the addition of NaOH [
30]. An interesting and promising application of CuS metal sulphide involves its use as anode material in lithium ion batteries [
31].
Interestingly, hybrid organic–inorganic nanoflowers were accidently discovered by adding copper sulphate (CuSO
4) to phosphate-buffered saline solutions, containing 0.1 mg/mL bovine serum albumin at pH 7.4 and 25 °C [
32]. A few days after, a precipitate appeared containing porous flower-like structures. This was the first example of a hybrid nanoflower made of both organic materials (in the form of proteins) and inorganic materials, i.e., copper (II) phosphate [
32]. As further observed in [
32], when an enzyme is used as the protein component, the nanoflowers showed enhanced catalytic activity, stability, and durability compared with free enzymes and other immobilization systems. These important results are probably due to either the large nanoflower surface-to-volume ratio or to the cooperative effects between the immobilized laccase molecules used and their interactions with copper (II) ions in the nanoflowers [
32].
It has been known since a long time that carbon (C) provides an excellent variety of nanostructured systems, such as the C
60 [
33], C
70 [
34], C
76 [
35,
36], nanotubes [
37], and, last but not least, the abovementioned graphene sheet [
38], occupying a dominant role either for fundamental or applied physics.
The present work is part of an ongoing study on C nanostructures synthesized by green pyrolysis at different temperatures (500, 700, and 900 °C). The main purpose of this investigation was twofold: (i) The exploitation of metallic crystalline nanostructures derived from the impurities present in red onion peels and obtained after their annealing [
39,
40]; and (ii) the possible application of these porous C nanostructures as sensors for toxic gases.
We obtained pure metallic nanocrystalline Cu agglomeration in C powders during green bio-waste pyrolysis from the peels of red onion [
40],
Allium cepa, L., [
41]. In particular, the Cu nanostructures were found only on the samples pyrolyzed at the highest temperature (900 °C) and not on those obtained at 500 and 700 °C (not further described here).
The aim of the present manuscript is to report the discovery of a new Cu nanostructure coined here as Cu nano-rose, due to the particular characteristic exhibited by its layers, which are spirally rolled-up like the petals of a rose.
High-resolution scanning electron microscopy (HR-SEM), energy dispersive X-ray spectroscopy (EDS), energy dispersive X-ray diffraction (ED-XRD) and Raman spectroscopy were used to exploit the morphology, chemical, and structural properties of this new arrangement of copper.
3. Results and Discussion
The morphologies and structures of the pyrolyzed carbon powders were determined using HR- field emission scanning electron microscopy equipped with selected area electron energy dispersive X-ray spectroscopy.
The SEM images at 53.54 × 66.33 μm
2 and that of the overlapped false color table from EDS total yield spectra, taken with the Zeiss system FEG-SEM, LEO 1530 operating at an accelerating high voltage (HV) of 20.0 KV, are reported in
Figure 1a,b.
At this magnification, the C powders display the coexistence of structures at various scales, from large flakes a few μm across to small spheres of some nanometres.
They are inconsistently arranged, even if somehow, they keep the memory of their original shape attributed to the bio-waste, [
40],
Allium cepa, L., [
41] (onion vulgaris) peels from which they derive. On the other hand,
Figure 1b evidences the different chemical composition, Na, O, Ca, Cl, K, Si, Al, and Mg, in addition to the predominant carbon, in close agreement with what reported in [
40] and to the fact that the red onion vulgaris already contains these elements as precursors and, probably, in a conjugate form.
The false color EDS images of C (red), Ca (light blue), K (green), O (dark blue), Mg (magenta), and S (yellow) and their corresponding Kα
1 and Kα
2 shells are shown in
Figure 1c–h, whereas
Figure 1i shows the intensity of the Lα
1, Lα
2 and Kα
1, Kα
2 emission lines as a function of increasing energy for these elements. We can easily note, from the inset, that the presence of chemical species different from carbon can mainly be assigned to impurities, including the Cu Lα
1, Lα
2 emission lines that are clearly visible at around (0.93–0.95) KeV of energy, as well as the others Kα
1, Kα
2, and Kα
3 less intense around (8.0–8.9) KeV, proving C to be over 80%. The Si peak originated from the substrate.
A typical HR-SEM image, (58.56 × 77.91) μm
2, from C nanopowders supported on the gold substrate at magnifications of 3.54 K× and an electron acceleration HV of 2.0 KV, collected by using the Zeiss FE-SEM, is shown in
Figure 2a. At first sight, we observe very defined and large C flakes (from 5 to 50 μm in length) randomly oriented and occupying most of the image area. Smaller structures scattered over the entire surface of these C flakes are clearly visible. Among these, three vaguely spherical objects (indicated with blue arrows) can be seen, the largest (red arrow) having a diameter of ~7 μm and the other smaller two of approximately 0.5 μm (yellow arrows). These particles were studied in a detailed morphological and chemical analysis, described hereafter.
Figure 2b–d show the SEM images at a higher resolution (17.92, 46.64, 169.32) K× of the largest spherical structure located at the center of the image of
Figure 2a. We can interestingly remark the presence of rolled-up structures like nano-roses, distributed in a fractal like pseudo-morphism. The red dashed circle in
Figure 2b with the diameter B = 2.63 μm indicates one of the composite structures that in a mathematical similarity forms the entire larger sphere of diameter A = 7.54 μm marked with a dashed light-blue circle. The yellow dashed circle with a diameter C = 0.92 μm emphasizes a single small rounded structure that builds-up the entire larger object in the dashed red circle.
Although in our bi-dimensional SEM images (consider for example the dotted light-blue area), we collapsed the three-dimensional reality of the rounded nano-roses on two-dimensional projections, we could extract some geometrical relations of the similarity among the spheres. One should also bear in mind that the circles drawn in
Figure 2b must be considered as an average. In particular, with the dotted light-blue circle we tried to enclose the maximum possible nanostructures.
Despite these simplifications, a close inspection of
Figure 2 suggests some interesting points to consider: If we calculate the ratios between the diameters of the various circles, i.e., A:B and B:C, we get the same value of 2.86, which allows A:C to be expressed as about n times C, where n is an integer and is equal to 9. This means that the pseudo-fractal feature that is apparent at first visual inspection of the SEM images can be confirmed as a numerical relationship based on this simple geometric observation.
By increasing the magnification, as shown in
Figure 2c, we can observe the petals of the individual nano-rose better, indicated, as a guide for the eyes, by the dashed pink circle. Each single petal (
Figure 2d) has, on average, a lateral dimension of about 172 nm (pink arrow), with few larger or smaller petals.
Furthermore, it seems that the individual nano-rose (see those within the yellow and red circles in
Figure 2b) uniformly covers a hypothetical sphere, the largest, as a spherical shell of a smaller inner diameter. We can imagine that the two structures of a smaller diameter indicated in
Figure 2a by yellow arrows are also composed, in the same way, by nano-roses, considering that they are shown to be jagged in their external lines, having a similar gleaning profile. This is an additional intriguing aspect since it would suggest that these pseudo-fractal structures present on the external surface of the sphere also occur inside its volume, assuming a 3D fractal similitude, where the single nano-rose may be considered as the building block.
To identify the chemical composition of the nano-roses, we performed EDS microanalysis using the Bruker system.
Figure 3a shows the SEM image that outlines the area where the EDS data were collected (3.8 K×, HV = 15 KV).
Figure 3b–f display, respectively, the X-ray fluorescence maps for C, Ca, Mg, O, and Cu collected on the same area as in
Figure 3a.
Interestingly, we may observe from these elemental maps that the nano-roses are constituted of Cu, as shown in
Figure 3f, and in particular by comparison of
Figure 3e,f that are mainly composed of metallic Cu
0.
This represents a very intriguing point, since to date, no structures similar to our nano-roses have been described, considering both the morphology of the crystallites and their composition. Typically, the nanostructures of copper reported in the literature, which were mainly obtained from a liquid phase by the immersion of pure copper sheets, have been found to be systematically composed of oxidized copper, CuO [
17,
18,
19,
20,
21,
22,
23,
24,
25], showing nanostructures in the form of flowers but completely dissimilar from those reported in this paper [
25]. The only exception is the Cu nanoflowers, constituted of metallic Cu
0 according to X-ray diffraction and X-ray spectroscopy measurements [
43], obtained by a wet chemical process using Cu[(acac)
2], acac = acetylacetonate as the reagent, oleyl amine both as the solvent and reducing agent, and cetyltrimethylammonium bromide as the capping agent.
We extrapolated that the main driving force used to obtain the Cu nano-roses from red onion peels is the high temperature applied during the annealing process performed in the inert nitrogen atmosphere. This produced both the expulsion of copper from its original organic molecules and the prevention of the oxidation onset to finally give rise to these particular Cu nanostructures. One cannot, however, exclude that the presence of elemental carbon may catalyze the Cu
0 nano-rose formation. In this respect, experiments are in progress to elucidate the actual mechanism of the formation of this new kind of Cu agglomeration, addressing both the physical and chemical aspects involved. This is a crucial point to clarify considering that both our results and those of [
43] demonstrate that it is possible to obtain Cu
0 crystalline structures that are stable in air.
In the past, Ag structures have been obtained from the reaction of silver nitrate and ascorbic acid in an acidic aqueous solution of polynaphthalene [
44]. Such structures are composed of metallic Ag
0 nanoparticles of about 10 nm in diameter linearly ordered inside a filamentous polymer sheath with a mesometric morphological structure in the flower form of about 10 microns in diameter, having meso-petals of a lanceolate shape. This constitutes an example where the Ag nanoparticles are kept chemically inert by a polymeric protective sheath that surrounds them. On the contrary, in the case of our copper nano-roses, as also reported in [
43], the structures are self-protected against any oxidation or sulfurization, coming from air or from the precursors of the pyrolyzed carbon nanostructures.
To validate this result, the ED-XRD spectrum was collected on the carbon nanostructures as a function of the quantum momentum scattering parameter. In
Figure 4, we can easily observe the presence of four peaks, two of which are assigned to the (002) and (101) reflections of the graphitic form of carbon indexed according to ICCD card number 00-001-0640 and corresponding to interplanar spacings with d = 3.380 Å and d = 2.020 Å, respectively; the other two are attributed to the (111) and (200) diffraction peaks of metallic copper [ICCD card number 00-001-1241] and are due to the interplanar spacings with d = 2.080 Å and d = 1.810 Å, respectively [
21,
24,
25,
43].
It is worth noting that the (110), (002), (-111), (111), and (200) diffraction peaks from CuO, which are expected in the region between 2.1 and 2.9 Å
−1 [
21,
24,
25], are completely absent in
Figure 4, confirming the EDS results. Indeed, the most intense reflection of CuO, (-111) at 2θ = 35.744° (ICDD Reference code: 00-001-1117), would have been located at q = 2.50 Å
−1, and is clearly absent. Furthermore, the presence of crystalline CuS (ICDD Reference code: 00-07-0876) is also excluded: the main (103) reflection peak should have been located at q = 2.23 Å
−1 [
45] and is missing in our pattern.
Finally,
Figure 5 displays the Raman spectrum from the C nanostructures. Two main peaks labelled D and G, located at 1386 and 1587 cm
−1, are related to the E
2g in the plane vibrational mode of the hexagon atoms in the honeycomb C graphitic arrangement and to the defects peak [
46,
47,
48].
The G-mode peak is found at 1587 cm
−1. Such a value, slightly blue-shifted with respect to highly crystalline graphite (G-mode at about 1580 cm
−1) [
46,
47,
48], can be attributed to the nanometric-sized C-particles. This agrees with [
46,
47,
48], where a frequency shift of about 15 cm
−1 towards higher wavenumbers was found in some samples with extremely small graphite crystal sizes [
46]. Several Raman spectra were collected on different portions/grains of the C nanostructures in order to obtain appreciable reproducibility. Notably, the presence of the main Raman features from CuO (295 cm
−1) [
26] and CuS (473 cm
−1) [
27,
28,
29,
30] was never detected.
The red rectangle in
Figure 5 delimits a region between 250 and 820 cm
−1, enlarged in the inset, which shows a flat area without features, where typically the structures of Cu oxides and sulphides are found [
26,
27,
28,
29,
30]. This observation is in perfect agreement with all the described data, confirming that the copper nano-roses exposed to air have the composition of metallic Cu
0 and thus are self-protected against oxidation from the external atmosphere.
One hypothesis that could explain this important result lies in the possibility of interpreting the petals of the nano-roses as consisting of stacked layers of Cu [111] planes, such as [
43], graphitic-like arranged, and thus resistant to possible oxidation.