Structural and Magnetic Properties of Co-Mn Codoped ZnO Nanoparticles Obtained by Microwave Solvothermal Synthesis

Zinc oxide nanoparticles codoped with Co2+ and Mn2+ ions (Zn(1−x−y)MnxCoyO NPs) were obtained for the first time by microwave solvothermal synthesis. The nominal content of Co2+ and Mn2+ in Zn(1−x−y)MnxCoyO NPs was x = y = 0, 1, 5, 10 and 15 mol % (the amount of both ions was equal). The precursors were obtained by dissolving zinc acetate dihydrate, manganese (II) acetate tetrahydrate and cobalt (II) acetate tetrahydrate in ethylene glycol. The morphology, phase purity, lattice parameters, dopants content, skeleton density, specific surface area, average particle size, average crystallite size, crystallite size distribution and magnetic properties of NPs were determined. The real content of dopants was up to 25.0% for Mn2+ and 80.5% for Co2+ of the nominal content. The colour of the samples changed from white to dark olive green in line with the increasing doping level. Uniform spherical NPs with wurtzite structure were obtained. The average size of NPs decreased from 29 nm to 21 nm in line with the increase in the dopant content. Brillouin type paramagnetism and an antiferromagnetic interaction between the magnetic ions was found for all samples, except for that with 15 mol % doping level, where a small ferromagnetic contribution was found. A review of the preparation methods of Co2+ and Mn2+ codoped ZnO is presented.

Spintronics is the science of electron spin and the related magnetic moment in semiconductors.Many researchers share the view that the 21st century will be remembered as the age of spintronic revolution due to the construction of the first spintronic devices for quantum computation or communication [40].Spintronics enables the creation of next-generation optoelectronic devices and data carriers [41].New materials, e.g., diluted magnetic semiconductors (DMS), where spin-polarised charge carriers can be obtained to enable performance of operations in spintronic devices, are being intensively sought.DMS are materials that are characterised by semiconductor (tunable conductivity) and ferromagnetic (controlled spin polarisation) properties.Dietl et al. [42] discussed the Zener model of ferromagnetism in zinc-blend magnetic semiconductors including ZnO by calculating Curie temperature Tc higher than 100 K. Doped ZnO has long been considered a promising material for applications as DMS as a result of theoretical calculations, implying that doped ZnO might display ferromagnetic properties at the room temperature [43,44].Additionally, such ZnO properties as the band gap value and the conductivity could be controlled through doping ZnO NPs with ions of transition metals (e.g., Co, Mn, Cr, Ni, Fe, V).The issue of doped ZnO NPs has been quite extensively examined by various research groups [6,[44][45][46][47].
ZnO NPs doped with Co 2+ (Zn (1−y) Co y O) and Mn 2+ (Zn (1−x) Mn x O) ions have gained much interest from scientists [4,6].Despite numerous studies, the magnetic properties of Zn (1−x) Mn x O NPs and Zn (1−y) Co y O NPs have been a fairly controversial subject so far .The nature of the origin of the ferromagnetic properties of doped ZnO NPs has not been unambiguously described yet.Most researchers explain a change in the magnetic properties of Zn (1−x) Mn x O NPs and Zn (1−y) Co y O NPs with the formation of, e.g., a secondary phase such as Co metal clusters, Co(OH) 2 , CoO, Co 3 O 4 , Co 2 O 4 , Mn 2 O 3 , Mn 3 O 4 , ZnMnO 3 [48][49][50][51][52][53][54][55][56][57][58][59][60] and a spinel phase (ZnMn 2 O 4 , ZnCo 2 O 4 ) [53,54,[57][58][59][60][61][62][63][64], which are characterised by different magnetic properties (Table 1).However, a change in the magnetic properties of doped ZnO NPs, apart from a change in the level of the oxidation state of dopants and the presence of foreign phases, can also be explained by the presence of oxygen vacancies, the formation of dopant clusters, the exchange interaction between the local spin-polarised electrons and associated with specific defects and adsorbed elements at the surface of the NPs [66-71].Martínez et al. [72], in turn, correlated the role of the microstructure with the change of the magnetic properties of Zn (1−y) Co y O NPs NPs, which demonstrates the complex nature of the magnetic properties of NPs.Based on the results of our own research [73], we believe that if the synthesis method employed permits the achievement of uniform substitution of Zn 2+ ions in the ZnO crystalline lattice with dopant ions (Co 2+ or Mn 2+ ), such a material will be paramagnetic with some antiferromagnetic coupling.Currently, new methods of modifying ZnO properties are being sought to enable obtaining ZnO properties that satisfy the criteria of a DMS [46].One of the solutions could be, e.g., codoping of ZnO with ions of Mn 2+ and Co 2+ (Zn (1−x−y) Mn x Co y O).Codoping is a promising strategy that enables an effective modification of the magnetic and electronic properties of DMS materials.Research concerning codoping has been carried out for more than 20 years now [74].It is believed that codoping might overcome the difficulties in bipolar doping and compensation in semiconductors [74].Thanks to codoping, it is possible to increase the solubility of dopants, improve the stability of the expected defects, increase the carrier mobility and increase the activation rate by lowering the ionisation energy of acceptors and donors [74,75].Paramagnetic [88] The relevant literature has reported a dozen or so methods of obtaining Zn (1−x−y) Mn x Co y O nanostructures, an overview of which is presented in Table 2.
ethanol solution as a solvent, and DEA as a stabilizing agent 500  The magnetic properties of codoped ZnO depend chiefly on the method of obtaining it, the synthesis parameters and the sample preparation process (Table 2).Most authors modified the properties of the obtained Zn (1−x−y) Mn x Co y O samples by selecting the temperature and gas atmosphere of the applied method or by soaking the samples additionally.An appropriate selection of the parameters for obtaining or preparing Zn (1−x−y) Mn x Co y O samples permits the oxidation or reduction of dopant ions, which might result in the formation of a secondary phase of, e.g., dopant oxides and oxygen vacancies.A good example might be a study of Naeem et al. [99], where Zn (1−x−y) Mn x Co y O NPs were obtained with the use of a chemical route and then were deliberately soaked at a temperature of 600 • C in an oxidising atmosphere and a reducing atmosphere.It was only Zn (1−x−y) Mn x Co y O NP samples soaked in a reducing atmosphere that displayed ferromagnetic properties; their nature was explained by the formation of oxygen vacancies because Naeem et al. did not observe any foreign phase inclusions within the detection limit of the X-ray diffraction method.In order to understand the nature of the magnetic properties of Zn (1−x−y) Mn x Co y O, the first-principles method based on the density functional theory is employed for their analysis [116][117][118].
Our research conducted to date has shown that the microwave solvothermal synthesis of ZnO enables: obtaining homogenous ZnO NPs with controlled size from circa 15 nm to 120 nm [140,141]; -controlling the average size of ZnO NPs aggregates of 60, 90 and 120 nm at the same time preserving the constant size of single NPs, which are sized 27 nm [148]; -obtaining homogenous ZnO NPs doped with Co 2+ or Mn 2+ ions [73,149,150]; -obtaining ZnO NPs doped with Co 2+ ions with controlled particle size between at least 28 nm and 53 nm [151].
The aim of this study is to prove that the microwave solvothermal synthesis enables obtaining homogeneous spherical Zn (1−x−y) Mn x Co y O NPs with paramagnetic properties.

Synthesis of Zn (1−x−y) Mn x Co y O NPs
Zn (1−x−y) Mn x Co y O NPs were obtained with the use of the MSS method in accordance with the procedure described in our earlier publications [73,[149][150][151][152].Precursors of Zn (1−x−y) Mn x Co y O were prepared by dissolving mixtures of crystalline cobalt acetate, manganese acetate, and zinc acetate in ethylene glycol.The composition of Zn (1−x−y) Mn x Co y O NPs precursors summarised in Table 1 was calculated based on Equations ( 1) and (2), with the molar concentration of zinc acetate in ethylene glycol being constant and amounting to 0.3037 mol/dm 3 .
x Mn 2+ = n Mn 2+ n Zn 2+ + n Mn 2+ + n Co 2+ (1) (2) Precursor solutions (Table 3) were obtained by dissolving strictly defined compositions of mixtures of the acetates in glycol (150 mL) using a hot-plate magnetic stirrer (70 • C, 450 rpm, SLR, SI Analytics GmbH, Mainz, Germany).After the acetates dissolved, each solution was poured into a bottle (250 mL, PP), sealed and cooled down until it reached the ambient temperature.The parameters of NPs syntheses carried out in a microwave reactor (Model 02-02600 W, 2.45 GHz, ERTEC, Poland) were as follows: volume of the precursor poured into a Teflon reaction container-75 mL, heating duration-25 min; maximum temperature of the bottom of the outer wall of the reaction vessel [140]-190 ± 5 • C; microwave power-100%, cool-down duration-20 min.The reaction duration was chosen based on our earlier research concerning the optimisation of parameters of the microwave solvothermal synthesis of ZnO NPs in ethylene glycol.Our primary assumption for the optimisation was to achieve the maximum reaction efficiency, while in the case of a reaction duration of 25 min the efficiency of zinc acetate conversion with identical synthesis parameters was circa 100% [141].These setpoints of synthesis parameters were entered using the reactor's control panel window (see the Supplementary Materials).An example profile of the course of the microwave solvothermal synthesis in the reactor model 02-02 is presented in Figure 1.The microwave heating in the employed reactor can be divided into the following stages: 1.
Continuous microwave heating of the feedstock until the preset maximum temperature is reached (T max )-195 • C.

2.
Once the feedstock reaches the maximum temperature of 195 • C, the reactor's controller switches off the microwave heating and the feedstock temperature drops to the preset minimum temperature (T min )-185 • C.

3.
Once the feedstock reaches T min 185 • C, the reactor's controller switches on the continuous microwave heating and re-heats the sample to T max 195 • C.

4.
Cyclic repetition of stages 2 and 3 until the preset reaction duration of 25 min is reached.

5.
The reactor is cooled down by activating cold water flow in the metal body, inside which the Teflon reaction vessel containing the obtained post-reaction suspended matter is placed.
Post-reaction suspended matter was centrifuged (MPW-350, MPW Med Instruments, Warsaw, Poland) and the supernatants were decanted.Afterwards, the sediments were rinsed intensively three times with water and centrifuged.Fifty millilitres of water were added to the obtained pastes; they were intensively stirred and then cooled down with liquid nitrogen and dried in a freeze dryer (Lyovac GT-2, SRK Systemtechnik GmbH, Riedstadt, Germany).

Characterisation Methods
The measurement procedures employed were described in our earlier publications [149,151].The analysis of the phase composition was performed with the X-ray diffraction method (XRD) (2 theta from 10° to 150°, room temperature, CuKα1, X'Pert PRO, Panalytical, The Netherlands).The parameters of the crystalline lattice, a and c, were determined by the Rietveld method in Fityk software, version 0.9.8.Crystallite size was calculated by means of Scherrer's formula [149] and an analysis of the profile of XRD peaks with the FW15/45M method [153][154][155], which also enabled the calculation of the crystallite size distribution.
The morphology of NPs was defined using scanning electron microscopy (SEM) (ZEISS, Ultra Plus, Oberkochen, Germany).
The specific surface area of NPs was determined through an analysis of nitrogen adsorption isotherm by the BET (Brunauer-Emmett-Teller) method (ISO 9277:2010, Gemini 2360, V 2.01, Micromeritics).Based on the results of the specific surface area and density, the average particle size was calculated with the assumption that all particles were spherical and identical [149].
The chemical composition of the samples dissolved in nitric acid (V) was analysed with the use of argon inductively coupled plasma optical emission spectrometry (ICP-OES) (Thermo Fisher Scientific, iCAP model 6000, Waltham, MA, USA) [73].The quantitative microanalysis of the Zn, Mn and Co content in the pressed NP samples was performed using energy-dispersive spectrometry (EDS) (Quantax 400, Bruker, Billerica, MA, USA) [151].
The colour analysis by means of the Red-Green-Blue (RGB; value: 0 to 1023) and Hue-Saturation-Luminance (HSL; value: 0 to 1000) colour model was carried out with the RGB-2000 metric by VOLTCRAFT (Conrad Electronic SE, Wernberg-Köblitz, Germany) in accordance with the manufacturer's recommendations.The synthesis of ZnO NPs codoped with Mn 2+ and Co 2+ ions in EG is presented by Equation (3): Post-reaction suspended matter was centrifuged (MPW-350, MPW Med Instruments, Warsaw, Poland) and the supernatants were decanted.Afterwards, the sediments were rinsed intensively three times with water and centrifuged.Fifty millilitres of water were added to the obtained pastes; they were intensively stirred and then cooled down with liquid nitrogen and dried in a freeze dryer (Lyovac GT-2, SRK Systemtechnik GmbH, Riedstadt, Germany).

Characterisation Methods
The measurement procedures employed were described in our earlier publications [149,151].The analysis of the phase composition was performed with the X-ray diffraction method (XRD) (2 theta from 10 • to 150 • , room temperature, CuK α1 , X'Pert PRO, Panalytical, The Netherlands).The parameters of the crystalline lattice, a and c, were determined by the Rietveld method in Fityk software, version 0.9.8.Crystallite size was calculated by means of Scherrer's formula [149] and an analysis of the profile of XRD peaks with the FW15/45M method [153][154][155], which also enabled the calculation of the crystallite size distribution.
The morphology of NPs was defined using scanning electron microscopy (SEM) (ZEISS, Ultra Plus, Oberkochen, Germany).
The specific surface area of NPs was determined through an analysis of nitrogen adsorption isotherm by the BET (Brunauer-Emmett-Teller) method (ISO 9277:2010, Gemini 2360, V 2.01, Micromeritics).Based on the results of the specific surface area and density, the average particle size was calculated with the assumption that all particles were spherical and identical [149].
The chemical composition of the samples dissolved in nitric acid (V) was analysed with the use of argon inductively coupled plasma optical emission spectrometry (ICP-OES) (Thermo Fisher Scientific, iCAP model 6000, Waltham, MA, USA) [73].The quantitative microanalysis of the Zn, Mn and Co content in the pressed NP samples was performed using energy-dispersive spectrometry (EDS) (Quantax 400, Bruker, Billerica, MA, USA) [151].
The colour analysis by means of the Red-Green-Blue (RGB; value: 0 to 1023) and Hue-Saturation-Luminance (HSL; value: 0 to 1000) colour model was carried out with the RGB-2000 The magnetisation measurements are performed with the use of a SQUID-type magnetometer (liquid helium cooled MPMSXL device manufactured by Quantum Design, Inc., San Diego, CA, USA) in the temperature range 2-300 K and magnetic fields up to 7 T.

Morphology
Figures 2 and 3 present selected representative SEM images of Zn(1−x−y)MnxCoyO NP samples.An impact of the content of dopants on the morphology of Zn(1−x−y)MnxCoyO is noticeable.Powders of ZnO and Zn(0.98)Mn0.01Co0.01Owere composed of loose homogeneous spherical particles sized 20-50 nm.Powders of Zn0.90Co0.05Mn0.05Oand Zn0.8Co0.1Mn0.1O, in turn, were composed of compact homogeneous spherical NPs sized 20-40 nm, which formed conglomerates sized between 1 µm and 3 µm resembling a "cauliflower" in terms of shape and structure.A similar impact of a dopant on a change in NPs morphology was observed in our earlier studies on doped ZnO [149,151].In order to eliminate the effect of aggregation of the obtained Zn(1−x−y)MnxCoyO NPs, the synthesis parameters must be individually optimised for each composition of Zn(1−x−y)MnxCoyO NPs.We have shown that microwave solvothermal synthesis permits controlling the size of ZnO NPs aggregates through a change of the microwave power used for ZnO NPs synthesis [148].Zhang et al. [156] described the impact of a change in the solvothermal synthesis temperature on the size of ZnO NPs aggregates.

Phase Composition and Lattice Parameters
Despite the presence of the lamellar structure visible in SEM images in the Zn0.7Co0.15Mn0.15Osample, all diffraction peaks in Figure 4 were attributed exclusively to the hexagonal phase ZnO (JCPDS No. 36-1451).The XRD test did not show the presence of the secondary phase, which could be M5(OH)(10−z)(CH3COO)Z•nH2O (lamellar structure) [157].The hexagonal phase of M5(OH)(10−z)(CH3COO)Z•nH2O has a diffraction peak below 10° (2 theta angle) [141].In order to verify the presence of a foreign phase in the Zn0.7Co0.15Mn0.15Osample, XRD measurement was performed again with the range from 5° to 100° (Figure 5).All diffraction peaks in Figure 5 were attributed exclusively to the hexagonal phase ZnO (JCPDS No. 36-1451).The absence of the secondary phase (Figure 3) in XRD results may mean that:

Phase Composition and Lattice Parameters
Despite the presence of the lamellar structure visible in SEM images in the Zn 0.7 Co 0.15 Mn 0.15 O sample, all diffraction peaks in Figure 4 were attributed exclusively to the hexagonal phase ZnO (JCPDS No. 36-1451).The XRD test did not show the presence of the secondary phase, which could be M 5 (OH)

Phase Composition and Lattice Parameters
Despite the presence of the lamellar structure visible in SEM images in the Zn0.7Co0.15Mn0.15Osample, all diffraction peaks in Figure 4 were attributed exclusively to the hexagonal phase ZnO (JCPDS No. 36-1451).The XRD test did not show the presence of the secondary phase, which could be M5(OH)(10−z)(CH3COO)Z•nH2O (lamellar structure) [157].The hexagonal phase of M5(OH)(10−z)(CH3COO)Z•nH2O has a diffraction peak below 10° (2 theta angle) [141].In order to verify the presence of a foreign phase in the Zn0.7Co0.15Mn0.15Osample, XRD measurement was performed again with the range from 5° to 100° (Figure 5).All diffraction peaks in Figure 5 were attributed exclusively to the hexagonal phase ZnO (JCPDS No. 36-1451).The absence of the secondary phase (Figure 3) in XRD results may mean that:   (2 theta angle) [141].In order to verify the presence of a foreign phase in the Zn 0.7 Co 0.15 Mn 0.15 O sample, XRD measurement was performed again with the range from 5 • to 100 • (Figure 5).All diffraction peaks in Figure 5 were attributed exclusively to the hexagonal phase ZnO (JCPDS No. .The absence of the secondary phase (Figure 3) in XRD results may mean that: the amount of the secondary phase was below the detection limit of the XRD method, -the secondary phase was an amorphous material, -the secondary phase is lamellar-shaped ZnO.
At this stage of the research, we are unable to identify what the lamellar structure product in the Zn 0.7 Co 0.15 Mn 0.15 O sample is (Figure 3) and we do not know whether its presence can be eliminated through optimisation of microwave solvothermal synthesis processes.This requires further research.
Catalysts 2018, 8, x FOR PEER REVIEW 11 of 28 -the amount of the secondary phase was below the detection limit of the XRD method, -the secondary phase was an amorphous material, -the secondary phase is lamellar-shaped ZnO.
At this stage of the research, we are unable to identify what the lamellar structure product in the Zn0.7Co0.15Mn0.15Osample is (Figure 3) and we do not know whether its presence can be eliminated through optimisation of microwave solvothermal synthesis processes.This requires further research.Theorem-type environments (including propositions, lemmas, corollaries, etc.) can be formatted as follows: ZnO is characterised by the hexagonal wurtzite structure (295 K, h-ZnO space group P63mc) with the lattice parameters of a = 3.2498 Å and c = 5.2066 Å [6], while the c/a lattice parameter ratio in ZnO is 1.6021 and is close to the c/a value = 1.6330 for the theoretical close-packed hexagonal structure (hcp) [151,158].The ionic radius of Zn 2+ is 0.74 Å [159].Manganese oxide (II) MnO has a cubic rock salt structure (295 K, c-MnO, space group Oh5-Fm3m) with the lattice parameter of a = 4.4475 Å [160], and the ionic radius of Mn 2+ is 0.83 Å [161].Cobalt oxide (II) CoO crystallises in two crystalline phases: cubic rock salt CoO (295 K, c-CoO, space group Fm3m) with the lattice parameter of a = 4.2581 Å, and hexagonal wurtzite CoO (77 K, h-CoO, space group P63mc) with lattice parameters of a = 5.183 Å i c = 3.017 Å [160].The ionic radius of Co 2+ is 0.745 Å [162].
The results of our earlier studies [73,149] proved that the lattice parameters a and c in ZnO change in line with the change in Co 2+ or Mn 2+ ion content, which we explained with the impact of changes in the values of ion radii of dopants in relation to Zn 2+ .However, each dopant gave rise to a distinct change in the value of the lattice parameters of doped ZnO depending on the content and type.For Co 2+ dopant ions, whose ion radius is almost identical to that of Zn 2+ , the value of a lattice parameter in Zn(1−y)CoyO samples grew in line with the increase in the dopant content from 1% to 15% [73].The value of the c lattice parameter, in turn, grew when the nominal dopant content ranged from 1% to 5%, and dropped when the content was 5 to 15%.In the case of the dopant of Mn 2+ ions, where their ionic radius is greater than Zn 2+ radius by as much as 0.09 Å, both lattice parameters, a and c, in Zn(1−x)MnxO samples increased their values within the dopant content range of 1 to 25 mol % [149].
The calculated a and c lattice parameters of Zn(1−x−y)MnxCoyO samples are presented in Table 4 and in Figure 6.The value of both lattice parameters a and c grew in line with the increase in the content of Mn 2+ and Co 2+ dopants in ZnO (Figure 6).Table 4 also presents the lattice parameters of Zn0.85Mn0.15ONPs and Zn0.85Co0.15ONPs as comparative samples for the Zn0.70Mn0.15Co0.15Osample.According to our assumptions, the principal impact on the value of ZnO lattice parameters was exerted by the dopant with the greater ionic radius since a and c lattice parameters for the Zn0.70Mn0.15Co0.15Osample are comparable to the values for the Zn0.85Mn0.15Osample.The change in a and c lattice parameters in the obtained Zn(1−x−y)MnxCoyO samples proves an effective substitution of Theorem-type environments (including propositions, lemmas, corollaries, etc.) can be formatted as follows: ZnO is characterised by the hexagonal wurtzite structure (295 K, h-ZnO space group P63mc) with the lattice parameters of a = 3.2498 Å and c = 5.2066 Å [6], while the c/a lattice parameter ratio in ZnO is 1.6021 and is close to the c/a value = 1.6330 for the theoretical close-packed hexagonal structure (hcp) [151,158].The ionic radius of Zn 2+ is 0.74 Å [159].Manganese oxide (II) MnO has a cubic rock salt structure (295 K, c-MnO, space group Oh5-Fm3m) with the lattice parameter of a = 4.4475 Å [160], and the ionic radius of Mn 2+ is 0.83 Å [161].Cobalt oxide (II) CoO crystallises in two crystalline phases: cubic rock salt CoO (295 K, c-CoO, space group Fm3m) with the lattice parameter of a = 4.2581 Å, and hexagonal wurtzite CoO (77 K, h-CoO, space group P63mc) with lattice parameters of a = 5.183 Å i c = 3.017 Å [160].The ionic radius of Co 2+ is 0.745 Å [162].
The results of our earlier studies [73,149] proved that the lattice parameters a and c in ZnO change in line with the change in Co 2+ or Mn 2+ ion content, which we explained with the impact of changes in the values of ion radii of dopants in relation to Zn 2+ .However, each dopant gave rise to a distinct change in the value of the lattice parameters of doped ZnO depending on the content and type.For Co 2+ dopant ions, whose ion radius is almost identical to that of Zn 2+ , the value of a lattice parameter in Zn (1−y) Co y O samples grew in line with the increase in the dopant content from 1% to 15% [73].The value of the c lattice parameter, in turn, grew when the nominal dopant content ranged from 1% to 5%, and dropped when the content was 5 to 15%.In the case of the dopant of Mn 2+ ions, where their ionic radius is greater than Zn 2+ radius by as much as 0.09 Å, both lattice parameters, a and c, in Zn (1−x) Mn x O samples increased their values within the dopant content range of 1 to 25 mol % [149].
The calculated a and c lattice parameters of Zn (1−x−y) Mn x Co y O samples are presented in Table 4 and in Figure 6.The value of both lattice parameters a and c grew in line with the increase in the content of Mn 2+ and Co 2+ dopants in ZnO (Figure 6).Table 4

Impact of Chemical Composition on the Colour of Codoped ZnO NPs
The results of the dopant content analysis are presented in Table 5.The differences between the results of the analysis carried out by the EDS method and the ICP-OES method arise from the sensitivity, accuracy and limitations of these analytical techniques as regards the quantitative determination of manganese, cobalt and zinc.The NP samples for the EDS analysis were pressed to form pastilles so that the sample surface was flat and smooth and the porosity of the examined material was reduced as much as possible (ISO 22309:2011).The ICP method is considered a source of the most representative results owing to its numerous advantages, e.g., low detection limits, a wide linear dynamic range, high precision [163].Our interesting findings include efficiencies of doping for the obtained samples (Table 6).In line with the increase in the nominal contents of two dopants in precursor solutions, the efficiency of doping with Mn 2+ ions decreased and at the same time the efficiency of doping with Co 2+ ions in codoped ZnO NPs increased.The achieved low efficiency of doping with Mn 2+ , which was between 16% and 25%, can be explained by the difference in the sizes of the ionic radii of Mn 2+ and Zn 2+ being 0.09 Å.The efficiency of doping with Co 2+ ions, in turn, was as many as 3-4 times greater than the efficiency of doping with Mn 2+ ions, which is explained by the negligible difference in the values of Co 2+ and Zn 2+ ionic radii.An advantage of the method we have developed is the use of a solution of salts dissolved in ethylene glycol, which enables obtaining homogenous codoped NPs.The mechanism of the MSS reaction of Zn(1−x−y)MnxCoyO NPs permits the

Impact of Chemical Composition on the Colour of Codoped ZnO NPs
The results of the dopant content analysis are presented in Table 5.The differences between the results of the analysis carried out by the EDS method and the ICP-OES method arise from the sensitivity, accuracy and limitations of these analytical techniques as regards the quantitative determination of manganese, cobalt and zinc.The NP samples for the EDS analysis were pressed to form pastilles so that the sample surface was flat and smooth and the porosity of the examined material was reduced as much as possible (ISO 22309:2011).The ICP method is considered a source of the most representative results owing to its numerous advantages, e.g., low detection limits, a wide linear dynamic range, high precision [163].Our interesting findings include efficiencies of doping for the obtained samples (Table 6).In line with the increase in the nominal contents of two dopants in precursor solutions, the efficiency of doping with Mn 2+ ions decreased and at the same time the efficiency of doping with Co 2+ ions in codoped ZnO NPs increased.The achieved low efficiency of doping with Mn 2+ , which was between 16% and 25%, can be explained by the difference in the sizes of the ionic radii of Mn 2+ and Zn 2+ Crystals 2018, 8, 410 13 of 28 being 0.09 Å.The efficiency of doping with Co 2+ ions, in turn, was as many as 3-4 times greater than the efficiency of doping with Mn 2+ ions, which is explained by the negligible difference in the values of Co 2+ and Zn 2+ ionic radii.An advantage of the method we have developed is the use of a solution of salts dissolved in ethylene glycol, which enables obtaining homogenous codoped NPs.The mechanism of the MSS reaction of Zn (1−x−y) Mn x Co y O NPs permits the substitution of Zn 2+ ions with the optimum number of Mn 2+ and Co 2+ ions in the crystalline lattice of the ZnO being formed, depending on the precursor composition.The remaining quantity of Mn 2+ and Co 2+ , which has not been integrated into the crystalline lattice of ZnO, in turn, remains in the form of unreacted acetate salts dissolved in ethylene glycol [73].In the sol-gel co-precipitation method, the codoping efficiency is 100% [102], which considerably reduces the costs of synthesis and chemical waste disposal.Nevertheless, thanks to the employment of ethylene glycol, a solvent with poor reducing properties, the MSS method prevents a change in the oxidation state of Mn 2+ and Co 2+ dopants [73,149,151], which considerably limits the possibility of formation of foreign phase inclusions.Another advantage of the MSS method is the use of a low synthesis temperature (190-220 • C) and a short synthesis time compared to other methods enumerated in Table 2, which prevents uncontrolled NPs growth, formation of NPs aggregates as a consequence of sintering, and formation of foreign phase inclusions during NPs growth.It is generally known that ZnO codoping with Mn 2+ and Co 2+ ions causes a change in the optical properties and a reduction of the band gap width [103,105].Figures 7 and 8 present a visual comparison of changes in the colours of Zn (1−x−y) Mn x Co y O samples depending on the content of dopants, while Table 7 summarises the results of the analysis of colours by the RGB and HSL colour models.ZnO codoping with Mn 2+ and Co 2+ ions resulted in a change of sample colour from light green to dark olive green.A similar change of Zn (1−x−y) Mn x Co y O powder colours is reported by Yin-Hua et al. [100].The colour of the codoped ZnO is a resultant of the impact of the presence of two dopants in the ZnO crystalline lattice.Our earlier studies [149,150] showed that ZnO doped with Mn 2+ ions changes colour from yellow to orange (Figure 7c) when the quantity of the Mn 2+ dopant increases, while Co-doped ZnO NPs are green and the intensity of the colour depends on the content of the Co 2+ dopant (Figure 7b).Based on the impact of a single dopant on the colour of ZnO, it can be inferred that the green colour of Zn (1−x−y) Mn x Co y O is attributed mainly to the presence of the Co 2+ dopant.The modification of the colour of ZnO NPs through their codoping permits the application of Zn (1−x−y) Mn x Co y O nanopowders as colourful pigments.Obviously, when using Zn (1−x−y) Mn x Co y O NPs as pigments, their applications should be selected such that their potential toxicity is irrelevant.The obtained results are summarised in Table 8.The density of an undoped ZnO NP sample is 5.25 g/cm 3 and is smaller than the value of 5.61 g/cm 3 for the theoretical density of ZnO [164].The difference between theoretical densities of the materials and real densities of these materials in the nano form is well known [140,151,165,166] and arises mainly from the imperfection of the real crystalline structure of nanomaterial surface, deviations from the stoichiometric composition, or the presence of hydroxides.With the theoretical density of CoO (6.45 g/cm 3 ) and MnO (5.37 g/cm 3 ) taken into consideration, it could be assumed that the density of the samples would increase in line with the growth of the dopant content in Zn (1−x−y) Mn x Co y O.However, the density of Zn (1−x−y) Mn x Co y O samples dropped from 5.25 g/cm 3 to 5.06 g/cm 3 in line with the increase in the dopant content, which could be explained by three causes: a different volume and packing of the unit cell of ZnO than in the case of MnO and CoO [6]; -lower atomic mass of the dopants (Co 2+ -≈58.93 u; Mn 2+ -≈54.94u) in comparison to the substituted Zn 2+ (≈65.38 u) atoms in the ZnO crystalline structure; -increasing number of defects in the ZnO crystalline structure resulting from the growth of the content of Co 2+ and Mn 2+ dopants.In line with the increase in the dopant content in Zn (1−x−y) Mn x Co y O (Table 8), the specific surface area of the samples increased from 39.8 m 2 /g to 56.4 m 2 /g, while the average particle size calculated based on the specific surface area and density decreased from 29 nm to 21 nm.The average crystallite size calculated by the two methods fell within the range of standard deviations of these results.The ratio of the sizes of d c i d a crystallites proves a change in the asymmetry of the obtained Zn (1−x−y) Mn x Co y O crystallites, and similar d c /d a values were obtained for Zn (1−x) Mn x O NPs depending on the dopant content [149].The obtained crystallite size distribution of the Zn (1−x−y) Mn x Co y O samples showed that the growth of the dopant content resulted in a wider distribution (Figure 9).The differences in the results between the average particle size and the average crystallite size might be caused by the adopted assumptions of the calculation methods employed [149,151,153,167], and the similar values of average particle and crystallite sizes prove that Zn (1−x−y) Mn x Co y O particles are made of single crystallites.similar values of average particle and crystallite sizes prove that Zn(1−x−y)MnxCoyO particles are made of single crystallites.

Magnetic Characterisation of the Zn(1−x−y)MnxCoyO NPs
The representative magnetisation data are shown in Figure 10, where magnetisation as measured versus magnetic field is depicted at T = 2 K (a) and T = 300 K (b).Low-temperature data overall show typical paramagnetic behaviour: magnetisation increases monotonously in line with the increasing magnetic field and tends to saturate for the highest fields.For the samples with low concentrations of magnetic ions, saturation is nearly perfect, whereas for the highest concentrations magnetisation does not saturate in the applied field range.Such behaviour is typical for a paramagnetic system of localised magnetic moments of transition metals (TM) ions, coupled by an antiferromagnetic exchange interaction and was commonly observed for different.For lightly Mn 2+ /Co 2+ -doped samples this picture is confirmed by high-temperature data (T = 300 K), where magnetisation is a linear function of the magnetic field (cf. Figure 10b).On the other hand, hightemperature data for high Mn 2+ /Co 2+ concentrations reveal a different behaviour: magnetisation rises fast at low fields (B < 1 T), with a tendency to saturation, and then shows a linear field dependence (for B > 1 T).This suggests two components of the measured magnetisation: one purely paramagnetic, responsible for the linear field dependence and the second originating from ferro/ferrimagnetic phase arising during the growth process of the samples (e.g., Co3O4, Mn3O4).We note that such a situation was widely encountered for DMS synthesised with TM concentration close to the solubility limits [168][169][170].The representative magnetisation data are shown in Figure 10, where magnetisation as measured versus magnetic field is depicted at T = 2 K (a) and T = 300 K (b).Low-temperature data overall show typical paramagnetic behaviour: magnetisation increases monotonously in line with the increasing magnetic field and tends to saturate for the highest fields.For the samples with low concentrations of magnetic ions, saturation is nearly perfect, whereas for the highest concentrations magnetisation does not saturate in the applied field range.Such behaviour is typical for a paramagnetic system of localised magnetic moments of transition metals (TM) ions, coupled by an antiferromagnetic exchange interaction and was commonly observed for different.For lightly Mn 2+ /Co 2+ -doped samples this picture is confirmed by high-temperature data (T = 300 K), where magnetisation is a linear function of the magnetic field (cf. Figure 10b).On the other hand, high-temperature data for high Mn 2+ /Co 2+ concentrations reveal a different behaviour: magnetisation rises fast at low fields (B < 1 T), with a tendency to saturation, and then shows a linear field dependence (for B > 1 T).This suggests two components of the measured magnetisation: one purely paramagnetic, responsible for the linear field dependence and the second originating from ferro/ferrimagnetic phase arising during the growth process of the samples (e.g., Co 3 O 4 , Mn 3 O 4 ).We note that such a situation was widely encountered for DMS synthesised with TM concentration close to the solubility limits [168][169][170].Having the above in mind, the measured magnetic moment of each sample can be regarded as a sum of paramagnetic contributions of localised magnetic moments of Mn 2+ or Co +2 ions, diamagnetic contributions of the NP lattice and the sample holder (parafilm wrapper, glue), as well as possible contributions of unintentional impurities/secondary phases.Therefore, the measured magnetic moment can be expressed in the form: where MNP(B, T) is the total magnetisation of the magnetic moments of NPs, Xdia is the sum of diamagnetic susceptibility of NP, parafilm and glue (assumed to be temperature independent in the studied temperature range) and C represents the contribution from possible ferro/ferrimagnetic phases, as well as ferro/ferrimagnetic contributions of the ingredients used in the synthesis of the samples, parafilm and glue.As may be noticed from Figure 10b, the diamagnetic contribution for low concentrations of Mn/Co is sizeable and dominates the measured magnetic moment at high temperatures, e.g., at T = 300 K the measured magnetic moment is diamagnetic.In such a case, precise values of Xdia and C for the sample in question are crucial.Instead of applying a standard way to evaluate Xdia, i.e., measuring the undoped sample (ZnO), it is proposed to eliminate Xdia*B + C contributions by subtracting the high-temperature magnetic moment (where the paramagnetic contribution of Mn/Co magnetic moments is largely quenched) from the low-temperature one.
Assuming that Xdia and C are temperature-independent (which is true to a large extent), the following is obtained: Given the previous results for Mn 2+ -and Co 2+ -base DMS, MNP(B, T) can be assumed in the following form: Having the above in mind, the measured magnetic moment of each sample can be regarded as a sum of paramagnetic contributions of localised magnetic moments of Mn 2+ or Co +2 ions, diamagnetic contributions of the NP lattice and the sample holder (parafilm wrapper, glue), as well as possible contributions of unintentional impurities/secondary phases.Therefore, the measured magnetic moment can be expressed in the form: where M NP (B, T) is the total magnetisation of the magnetic moments of NPs, X dia is the sum of diamagnetic susceptibility of NP, parafilm and glue (assumed to be temperature independent in the studied temperature range) and C represents the contribution from possible ferro/ferrimagnetic phases, as well as ferro/ferrimagnetic contributions of the ingredients used in the synthesis of the samples, parafilm and glue.As may be noticed from Figure 10b, the diamagnetic contribution for low concentrations of Mn/Co is sizeable and dominates the measured magnetic moment at high temperatures, e.g., at T = 300 K the measured magnetic moment is diamagnetic.In such a case, precise values of X dia and C for the sample in question are crucial.Instead of applying a standard way to evaluate X dia , i.e., measuring the undoped sample (ZnO), it is proposed to eliminate X dia *B + C contributions by subtracting the high-temperature magnetic moment (where the paramagnetic contribution of Mn/Co magnetic moments is largely quenched) from the low-temperature one.
Assuming that X dia and C are temperature-independent (which is true to a large extent), the following is obtained: Given the previous results for Mn 2+ -and Co 2+ -base DMS, M NP (B, T) can be assumed in the following form: M NP (B, T) = A g µ B S B S (B, T), (6) where B S (B, T) is the Brillouin function for spin S = 5/2 (Mn 2+ ) or S = 3/2 (Co 2+ ), g = 2.00 is the g-factor, µ B is the Bohr magneton and A is the number of spins (magnetic moments) in the sample.A possible interaction between spins can be taken into account by assuming effective temperature T eff = T − T 0 , instead of experimental temperature T [171,172].We recall that T 0 < 0 corresponds to antiferromagnetic (AFM) interactions, while T 0 > 0 means ferromagnetic (FM) interactions.
In order to demonstrate how the proposed method works, we start with NPs doped only with one type of ions, i.e., Mn 2+ or Co 2+ .The results concerning the characterisation of Zn 0.99 Mn 0.01 O, Zn 0.85 Mn 0.15 O, Zn 0.99 Co 0.01 O and Zn 0.85 Co 0.15 O samples are included in the Supplementary Materials Figure 11a shows M exp (B, T = 2 K)-M exp (B, T = 300 K), as well as the fit with Equations ( 5) and ( 6), where A was the only adjustable parameter (spin S = 5/2 and T 0 = 0 were fixed).
Catalysts 2018, 8, x FOR PEER REVIEW 18 of 28 MNP(B, T) = A g µB S BS(B, T), (6) where BS(B, T) is the Brillouin function for spin S = 5/2 (Mn 2+ ) or S = 3/2 (Co 2+ ), g = 2.00 is the g-factor, µB is the Bohr magneton and A is the number of spins (magnetic moments) in the sample.A possible interaction between spins can be taken into account by assuming effective temperature Teff = T − T0, instead of experimental temperature T [171,172].We recall that T0 < 0 corresponds to antiferromagnetic (AFM) interactions, while T0 > 0 means ferromagnetic (FM) interactions.
The slower saturation of magnetisation MNP(B, T) than the Brillouin function reflects AFM interaction between Mn 2+ ions, which is expected for the system with about 1% of Mn 2+ ions [172].As mentioned above, this effect can be taken into account by introducing effective temperature Teff = T − T0 and considering T0 as an adjustable parameter.Figure 11b shows Mexp(B, T = 2 K)-Mexp(B, T = 300 K), as well as the fits with Equations ( 5) and ( 6), where A and T0 are adjustable parameters (spin S = 5/2 or S = 3/2 are fixed).
Satisfactory matching was obtained for all the samples.Effective temperatures T0 are all negative, indicating AFM exchange interactions between Mn 2+ and Co 2+ ions, as expected [171,172].5) and ( 6); T 0 was set to 0 K; (b) Zn 0.99 Mn 0.01 O and Zn 0.85 Mn 0.15 O (red), solid lines: fits with Equations ( 5) and ( 6), with S = 5/2 and T 0 as an adjustable parameter (T 0 = −0.17The slower saturation of magnetisation M NP (B, T) than the Brillouin function reflects AFM interaction between Mn 2+ ions, which is expected for the system with about 1% of Mn 2+ ions [172].As mentioned above, this effect can be taken into account by introducing effective temperature T eff = T − T 0 and considering T 0 as an adjustable parameter.Figure 11b shows M exp (B, T = 2 K)-M exp (B, T = 300 K), as well as the fits with Equations ( 5) and (6), where A and T 0 are adjustable parameters (spin S = 5/2 or S = 3/2 are fixed).
Crystals 2018, 8, 410 19 of 28 Satisfactory matching was obtained for all the samples.Effective temperatures T 0 are all negative, indicating AFM exchange interactions between Mn 2+ and Co 2+ ions, as expected [171,172].
For the samples codoped with both Mn 2+ and Co 2+ the following function was used: M NP (B, T) = A Mn g µ B S Mn B S (B, T eff ) + A Co g µ B S Co B S (B, T eff ), (7) where A Mn (A Co ) corresponds to number of Mn 2+ (Co 2+ ) ions, S Mn = 5/2, S Co = 3/2.In order to avoid a large number of fitting parameters only one effective temperature was used, which means that T 0 is a common parameter for all interactions in the NP, i.e., Mn 2+ -Mn 2+ , Co 2+ -Co 2+ and Mn 2+ -Co 2+ .The results of the fittings are shown in Figure 12.For all the samples T 0 is negative, suggesting preferred antiferromagnetic interactions between Mn and Co ions.For the samples codoped with both Mn 2+ and Co 2+ the following function was used: MNP(B, T) = AMn g µB SMn BS(B, Teff) + ACo g µB SCo BS(B, Teff), (7) where AMn (ACo) corresponds to number of Mn 2+ (Co 2+ ) ions, SMn = 5/2, SCo = 3/2.In order to avoid a large number of fitting parameters only one effective temperature was used, which means that T0 is a common parameter for all interactions in the NP, i.e., Mn 2+ -Mn 2+ , Co 2+ -Co 2+ and Mn 2+ -Co 2+ .The results of the fittings are shown in Figure 12.For all the samples T0 is negative, suggesting preferred antiferromagnetic interactions between Mn and Co ions.7), where ratio AMn/ACo was kept as resulting from Table 5.The resulting T0 are the following: T0 = −1.78K for Zn0.70Mn0.15Co0.15O(red), T0 = −0.42K for Zn0.80Mn0.10Co0.10O(blue), T0 = −0.37K for Zn0.90Mn0.05Co0.05O(green), T0 = −0.35K for Zn0.98Mn0.01Co0.01O(black).
To summarise the magnetic properties, it can be stated that the discussed NP samples can be generally described as the systems of localised magnetic moments arising from Mn 2+ or Co 2+ d-shell electrons located at Zn 2+ lattice sites.The paramagnetic properties of these systems can be reasonably well described by the effective Brillouin function, with an indication of AFM exchange interactions between Mn 2+ or Co 2+ ions.Moreover, for the Zn0.70Mn0.15Co0.15Osamples with the highest TM ions concentrations, an additional ferromagnetic-type magnetic phase is observed, most probably originating from crystalline phases other than Zn(1−x−y)MnxCoyO.
The Zn0.70Mn0.15Co0.15ONP sample is a good example, confirming our earlier research results [73].Namely, we stated that if a ZnO sample displays ferromagnetic properties, this may be caused by the presence of a foreign phase or a change in the dopant oxidation state.XRD tests did not indicate the To summarise the magnetic properties, it can be stated that the discussed NP samples can be generally described as the systems of localised magnetic moments arising from Mn 2+ or Co 2+ d-shell electrons located at Zn 2+ lattice sites.The paramagnetic properties of these systems can be reasonably well described by the effective Brillouin function, with an indication of AFM exchange interactions between Mn 2+ or Co 2+ ions.Moreover, for the Zn 0.70 Mn 0.15 Co 0.15 O samples with the highest TM ions concentrations, an additional ferromagnetic-type magnetic phase is observed, most probably originating from crystalline phases other than Zn (1−x−y) Mn x Co y O.

Figure 1 .
Figure 1.Diagram showing the course of the microwave synthesis of codoped ZnO NPs and undoped ZnO NPs on the example of Zn0.98Mn0.01Co0.01Osample.Experimental data obtained in the microwave reactor Model 02-02.

Figure 1 .
Figure 1.Diagram showing the course of the microwave synthesis of codoped ZnO NPs and undoped ZnO NPs on the example of Zn 0.98 Mn 0.01 Co 0.01 O sample.Experimental data obtained in the microwave reactor Model 02-02.

Figures 2 and 3
Figures 2 and 3 present selected representative SEM images of Zn (1− x−y) Mn x Co y O NP samples.An impact of the content of dopants on the morphology of Zn (1−x−y) Mn x Co y O is noticeable.Powders of ZnO and Zn (0.98) Mn 0.01 Co 0.01 O were composed of loose homogeneous spherical particles sized 20-50 nm.Powders of Zn 0.90 Co 0.05 Mn 0.05 O and Zn 0.8 Co 0.1 Mn 0.1 O, in turn, were composed of compact homogeneous spherical NPs sized 20-40 nm, which formed conglomerates sized between 1 µm and 3 µm resembling a "cauliflower" in terms of shape and structure.A similar impact of a dopant on a change in NPs morphology was observed in our earlier studies on doped ZnO[149,151].In order to eliminate the effect of aggregation of the obtained Zn (1−x−y) Mn x Co y O NPs, the synthesis parameters must be individually optimised for each composition of Zn (1−x−y) Mn x Co y O NPs.We have shown that microwave solvothermal synthesis permits controlling the size of ZnO NPs aggregates through a change of the microwave power used for ZnO NPs synthesis[148].Zhang et al.[156] described the impact of a change in the solvothermal synthesis temperature on the size of ZnO NPs aggregates.

Figure 3
Figure 3 presents the morphology of Zn 0.70 Co 0.15 Mn 0.15 O powder.SEM images show two products of synthesis: ZnO NPs aggregates, and a lamellar structure, which might be an unreacted intermediate in the form of hydroxide metal acetates with the general formula of M 5 (OH) (10−z) (CH 3 COO) Z •nH 2 O, where M = (Zn, Co and Mn) [157], or a compound of codoped hydroxide metal acetates Zn 5(1−x−y) Mn (5x) Co (5y) (OH) (10−z) (CH 3 COO) z •nH 2 O [141,151].Information regarding the phase composition of the Zn 0.70 Co 0.15 Mn 0.15 O sample will be provided by XRD.The product prevalent in the obtained Zn 0.70 Co 0.15 Mn 0.15 O sample is NPs, which is illustrated by the SEM image in Figure 3c.

Figure 4 .
Figure 4. XRD diffraction patterns of Zn (1−x−y) Mn x Co y O NPs (where x = y), with the nominal content of dopants in the solution being 0, 1, 5, 10, 15 mol %.The hexagonal phase of M 5 (OH) (10−z) (CH 3 COO) Z •nH 2 O has a diffraction peak below 10 • (2 theta angle)[141].In order to verify the presence of a foreign phase in the Zn 0.7 Co 0.15 Mn 0.15 O sample, XRD measurement was performed again with the range from 5 • to 100 • (Figure5).All diffraction peaks in
also presents the lattice parameters of Zn 0.85 Mn 0.15 O NPs and Zn 0.85 Co 0.15 O NPs as comparative samples for the Zn 0.70 Mn 0.15 Co 0.15 O sample.According to our assumptions, the principal impact on the value of ZnO lattice parameters was exerted by the dopant with the greater ionic radius since a and c lattice parameters for the Zn 0.70 Mn 0.15 Co 0.15 O sample are comparable to the values for the Zn 0.85 Mn 0.15 O sample.The change in a and c lattice parameters in the obtained Zn (1−x−y) Mn x Co y O samples proves an effective substitution of Zn 2+ ions with Mn 2+ and Co 2+ dopant ions in the ZnO crystalline lattice.Similar dependencies on changes to Mn-Co codoped ZnO lattice parameters were described by Adbullahi et al. [105].

Figure 6 .
Figure 6.Lattice parameters versus nominal dopants content of Zn (1−x−y) Mn x Co y O NP samples.

Figure 8 .
Figure 8. Photos of dry powders of Zn(1−x−y)MnxCoyO.Figure 8. Photos of dry powders of Zn (1−x−y) Mn x Co y O.

Figure 8 .
Figure 8. Photos of dry powders of Zn(1−x−y)MnxCoyO.Figure 8. Photos of dry powders of Zn (1−x−y) Mn x Co y O.

Table 1 .
Magnetic properties of the secondary phase observed in Mn 2+ -doped ZnO and Co 2+ -doped ZnO.
HLP 20UV, Hydrolab, Straszyn, Poland).The reagents were not subjected to any purification processes and were used as received.Only deionised water was used in the sample preparation process.

Table 3 .
Compositions (nominal content) of precursors of Zn (1−x−y) Mn x Co y O synthesis.

Table 4 .
Lattice parameters and ratio of lattice parameters of the obtained Zn (1−x−y) Mn x Co y O NPs.

Table 4 .
Lattice parameters and ratio of lattice parameters of the obtained Zn(1−x−y)MnxCoyO NPs.

Table 5 .
Results of the analysis of the chemical composition of Zn (1−x−y) Mn x Co y O samples.

Table 7 .
Results of the colour analysis of dry NP samples (RGB and HSL colour model).

Table 8 .
Characteristics of the NP samples.