Introduction

It is known from the literature that phosphates(V) of di- and trivalent metals have been for many years the objects of intensive studies, because they can be applied as inorganic pigments, as components of catalysts of different chemical processes or as electrode materials [1,2,3,4,5,6]. Taking into account this information, it can be assumed that also double phosphates(V), i.e. containing mixed di- and trivalent cation, can be very important for designing of new materials to be applied in different areas of technology. However, physicochemical properties of most of the double phosphates(V) have been so far investigated only in a fragmentary way, which makes it impossible to know their potential possibilities.

Merely two papers [7, 8] were found referring to double phosphates(V) Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6; what is more, both concern only the structures of these compounds. The above-mentioned compounds are the members of the Fe32+Fe43+(PO4)6 structure type family [7, 8], comprising a large number of phases, including not only phosphates, but also vanadates, molybdates and arsenates [9]. For example, the solid solution Mn3−xMgxFe4(PO4)6 (0 ≤ x ≤ 3) has the above structure and this solution produces interesting brown colourations in ceramic glazes [9]. On the other hand, phosphates are also a good alternative for the withdrawn anticorrosive chromates(VI). The best example is the conventionally used zinc orthophosphate(V) with its toxicity several times lower than that of chromates(VI) [3]. Unfortunately, an urgent problem becomes the gradual introduction of too much zinc ions to the natural environment. As a result, the other phosphates are studied in order to find new environmentally friendly pigments [3]. It has been confirmed that inter alia chromium(III) phosphates(V) show high anticorrosion efficiency [3].

In the light of the above literature data, Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6, i.e. double phosphates containing trivalent chromium and divalent magnesium or zinc, can be interesting for designing of new inorganic pigments. As mentioned above, except for the structure, there is no information about Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6. Therefore, the aim of the presented work was to synthesize and to investigate some of their physicochemical properties, including their thermal stability, size of grains and colour (very important for potential pigments), but also their infrared spectra and results of spectrophotometric measurements, which have not been presented until now. In our study for the first time, Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6 were obtained by solid-state reaction method. This method, in comparison with “wet” method and chemical transport applied by the authors of [7], is admittedly time-consuming, but non-waste and relatively simple. The additional aim of the work was to find out whether Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6 form a solid solution. Such a possibility is very important, because it enables partial replacement of zinc ions with more environmentally friendly ones, e.g. magnesium ions, and modification of physicochemical properties of the obtained potential pigments. So, if Zn3Cr4(PO4)6 shows high anticorrosion efficiency (as Zn3(PO4)2), a necessary reduction in zinc content could be achieved.

Experimental

The initial reagents used in the study were: 3MgCO3·Mg(OH)2·3H2O (p.a., POCh, Poland), ZnO (99.9%, Sigma-Aldrich, USA), (NH4)2HPO4 (p.a., Chempur, Poland), Cr2O3 (99%, Acros Organics, Germany). Syntheses were conducted by the conventional method of calcining samples described inter alia in the papers [10,11,12,13]. Stoichiometric mixtures of reacting substances were homogenized by grinding in an automated agate mortar, and then, they were heated in air atmosphere in several stages. After each heating stage, the samples were gradually cooled down in the furnace to room temperature, ground and examined by XRD method with respect to their composition. This procedure was repeated until the monophasic samples were obtained. To avoid the melting of the samples, the maximum temperature of their heating was a few ten degrees lower than their melting points found from DTA curves. The exceptions were the first stages of syntheses, when oxide precursors were used. In such cases, the temperature in the furnace was gradually increased from 350 to 650 °C, which was applied mainly to remove NH3, CO2 or steam as gas products of the occurring reactions [10, 14]. After gas removal, the samples were heated in the form of pastilles.

The XRD patterns of samples were recorded with the aid of the Empyrean II diffractometer (PANalytical, Netherlands) with CuKα radiation. The identification of phases, present in the samples, was carried out on the basis of their XRD characteristics contained in the PDF cards.

The DTA measurements were performed in air atmosphere using an instrument of a SDT 2960 type (TA Instruments, USA). Samples of ~ 25 mg each were heated in corundum crucibles from room temperature up to ~ 1450 °C at a heating rate of 10 °C min−1. The investigations up to ~ 1500 °C were conducted using an optical pyrometer (Raytek, model RAYMMIMHSF2V, Germany).

The densities of the obtained phases were determined with the help of an Ultrapyc 1200e ultrapycnometer (Quantachrome Instruments, USA); argon (5 N purity) was used as a pycnometric gas. The measurements were performed in five repetitions using ~ 2 g of each sample for the test.

The IR measurements (the Specord M 80 spectrometer, Carl Zeiss, Germany) were conducted applying the technique of pressing pastilles of the sample with KBr at the ratio 1:300 by mass.

The shape and size of the obtained crystallites were observed by means of SEM method (FE-SEM Hitachi SU-70 microscope).

The UV–Vis diffuse reflectance spectra of obtained phases were recorded using a UV–Vis spectrometer V-670 (JASCO, Japan) equipped with a reflecting attachment for solid-state investigation (integrating sphere attachment with horizontal sample platform PIV-756/(PIN-757)).The spectra were recorded in the wavelength region of 200–800 nm at room temperature using barium sulphate for reference. The spectral bandwidth was 5 nm.

Results and discussion

Solid-state synthesis and XRD characterization

Mg3Cr4(PO4)6 and Zn3Cr4(PO4)6 were obtained according to the following summary reactions:

$$3\left( {3{\text{MgCO}}_{3} \cdot{\text{Mg}}\left( {\text{OH}} \right)_{2} \cdot3{\text{H}}_{2} {\text{O}}} \right)_{{({\text{s}})}} + 8{\text{Cr}}_{2} {\text{O}}_{{3({\text{s}})}} + 24\left( {{\text{NH}}_{4} } \right)_{2} {\text{HPO}}_{{4({\text{s}})}} = 4{\text{Mg}}_{3} {\text{Cr}}_{4} \left( {{\text{PO}}_{4} } \right)_{{6({\text{s}})}} + 9{\text{CO}}_{{2({\text{g}})}} + 48{\text{NH}}_{{3({\text{g}})}} + 48{\text{H}}_{2} {\text{O}}_{{({\text{g}})}}$$
(1)
$$3{\text{ZnO}}_{{({\text{s}})}} + 2{\text{Cr}}_{2} {\text{O}}_{{3({\text{s}})}} + 6\left( {{\text{NH}}_{4} } \right)_{2} {\text{HPO}}_{{4({\text{s}})}} = {\text{Zn}}_{3} {\text{Cr}}_{4} \left( {{\text{PO}}_{4} } \right)_{{6({\text{s}})}} + 12{\text{NH}}_{{3({\text{g}})}} + 9{\text{H}}_{2} {\text{O}}_{{({\text{g}})}}$$
(2)

Stoichiometric mixtures of reagents were heated in the following stages:

  • Synthesis of Mg3Cr4(PO4)6: 350 °C(8 h) → 550 °C(8 h) → 650 °C(12 h) + 750 °C(12 h) + 1000 °C(12 h) + 1050 °C(12 h) + 1100 °C(12 h) × 3 + 1150 °C(12 h)

  • Synthesis of Zn3Cr4(PO4)6: 350 °C(8 h) → 550 °C(8 h) → 650 °C(12 h) + 750 °C(12 h) + 800 °C(12 h) + 900 °C(12 h) + 950 °C(12 h) + 1000 °C(12 h) + 1100 °C(12 h) + 1150 °C(12 h)

The change in the phase composition of samples during the synthesis, depending on the heating temperature, is presented in Table 1.

Table 1 Phase composition of the samples after subsequent stages of their heating
Full size table

The data show that in the first stage of the synthesis of Mg3Cr4(PO4)6 compound, in the temperature range from 350 to 650 °C, a reaction occurs to form MgP2O6:

$$3{\text{MgCO}}_{3} \cdot{\text{Mg}}\left( {\text{OH}} \right)_{2} \cdot3{\text{H}}_{2} {\text{O}}_{{({\text{s}})}} + 8\left( {{\text{NH}}_{4} } \right)_{2} {\text{HPO}}_{{4({\text{s}})}} = 4{\text{MgP}}_{2} {\text{O}}_{{6({\text{s}})}} + 16{\text{NH}}_{{3({\text{g}})}} + 16{\text{H}}_{2} {\text{O}}_{{({\text{g}})}} + 3{\text{CO}}_{{2({\text{g}})}}$$
(3)

The synthesis of Mg3Cr4(PO4)6 began at a temperature of about 750 °C, at which magnesium metaphosphate(V) reacts with chromium(III) oxide, present in the reagent mixture, according to the reaction:

$$3{\text{MgP}}_{2} {\text{O}}_{{6({\text{s}})}} + 2{\text{Cr}}_{2} {\text{O}}_{{3({\text{s}})}} = {\text{Mg}}_{3} {\text{Cr}}_{4} \left( {{\text{PO}}_{4} } \right)_{{6({\text{s}})}}$$
(4)

After subsequent heating stages of the sample in the temperature range of 1000–1100 °C, three compounds were identified in it, whereas the Mg3Cr4(PO4)6 content gradually increased with a decreasing content of MgP2O6 and Cr2O3. A monophasic sample containing Mg3Cr4(PO4)6 was obtained at 1150 °C.

As can be seen from the data presented in Table 1, after the first two stages of Zn3Cr4(PO4)6 synthesis, zinc pyrophosphate(V) and unreacted chromium(III) oxide were identified in the sample. The diffractograms contained also a set of lines that could not be assigned to any of the known phases forming in the side systems of the ZnO–P2O5–Cr2O3 system. It was therefore concluded that the set of lines is related to the presence of a previously unknown phase (marked as X), which forms as an intermediate product during synthesis. Research aimed at determining the composition of phase X will be continued.

Thus, in the case of the synthesis of Zn3Cr4(PO4)6, two reactions take place in parallel in the temperature range of 350–750 °C. The first of these leads to the formation of Zn2P2O7 according to the equation:

$$2{\text{ZnO}}_{{({\text{s}})}} + 2\left( {{\text{NH}}_{4} } \right)_{2} {\text{HPO}}_{{4({\text{s}})}} = {\text{Zn}}_{2} {\text{P}}_{2} {\text{O}}_{{7({\text{s}})}} + 4{\text{NH}}_{{3({\text{g}})}} + 3{\text{H}}_{2} {\text{O}}_{{({\text{g}})}}$$
(5)

On the basis of the phase composition of this sample, it can be additionally thought that Zn2P2O7 reacts with the chromium(III) oxide and (NH4)2HPO4, present in the initial mixture, to form the X phase.

Under the conditions of the experiment, the synthesis of Zn3Cr4(PO4)6 began at a temperature of about 800 °C. With the increase in temperature in the subsequent stages of heating, the content of Cr2O3, Zn2P2O7 and X phase in the sample decreased gradually and the Zn3Cr4(PO4)6 content increased simultaneously. Zinc pyrophosphate(V) reacted completely at 950 °C. Further steps of the synthesis performed in the temperature range 1000–1150 °C indicate that compound X reacts with Cr2O3 to form Zn3Cr4(PO4)6. Monophasic sample, containing only Zn3Cr4(PO4)6, was obtained at 1150 °C.

The diffraction patterns, taken after the last heating stage of both samples, were consistent with the XRD characteristics of Mg3Cr4(PO4)6 and Zn3Cr4(PO4)6, respectively, contained in the PDF cards No. 86-1997 and 86-1996.

Thermal stability

Thermal stability up to 1100 °C enables the use of the obtained compounds for colouring of commercially available middle-temperature ceramic glazes. To determine thermal stability of Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6, both compounds were studied by DTA method. In the DTA curve of Zn3Cr4(PO4)6 (Fig. 1), only one endothermic effect was recorded, with onset temperature of 1340 °C. In order to determine the type of the transformation that the effect is due to, a sample of Zn3Cr4(PO4)6 was heated for 1 h at 1360 °C (i.e. at the temperature of the maximum of this effect), and then, it was rapidly cooled to room temperature, ground and subjected to XRD study. On removal from the furnace at 1360 °C, the liquid was observed, so the recorded endothermic effect is connected with melting of Zn3Cr4(PO4)6. In the “frozen” sample, two solid phases were identified, i.e. α-CrPO4 and Cr2O3 (stable up to 1450 °C and 2335 °C, respectively [15, 16]). α-Zn2P2O7 was also identified, but taking into account its melting point (i.e. about 1000 °C), it can be concluded that zinc pyrophosphate(V) crystallized from the liquid. The obtained results point to incongruent melting of Zn3Cr4(PO4)6.

Fig. 1
figure 1

Fragment of DTA curve of Zn3Cr4(PO4)6

Full size image

In the DTA curve of Mg3Cr4(PO4)6, no thermal effects were recorded, so it was concluded that this compound undergoes melting or decomposition at temperature very close or higher than 1450 °C. To establish its thermal stability, the sample containing Mg3Cr4(PO4)6 was heated in air atmosphere in a horizontal tube furnace equipped with an optical pyrometer. According to the result of the performed experiment, it was estimated that the compound under study is stable in air up to ~ 1450 °C. XRD analysis, performed for the sample after experiment, indicates that Mg3Cr4(PO4)6 melts incongruently with the formation of Cr2O3. Magnesium pyrophosphate(V) was also identified in the sample. Mg2P2O7 with its melting point at 1395 °C [2], like Zn2P2O7, most likely crystallized from the liquid.

SEM investigation

Sizes of grains rank among fundamental properties of solids when applying them, for example, as inorganic pigments, because they determine the intensity of colour and dispersion of pigment in the binding system applied. The recommended mean sizes of inorganic pigment grains depend on the particular application, although in general they lie in the range 0.01–10 µm [17]. To gain the information about the obtained compounds on the microscale, their characterization by SEM method was made. The grains of Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6 (Figs. 2, 3) are similar in their shape, and both are much differentiated in sizes. The sizes of the larger crystallites are of the order of 6 µm (for Zn3Cr4(PO4)6) or 4 µm (for Mg3Cr4(PO4)6), whereas the sizes of the smaller crystallites do not often exceed 1 µm or 0.5 µm for Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6, respectively. It can be therefore concluded that the recommended mean grain size was attained for the prepared samples.

Fig. 2
figure 2

SEM image of Zn3Cr4(PO4)6

Full size image
Fig. 3
figure 3

SEM image of Mg3Cr4(PO4)6

Full size image

Infrared spectra

Figure 4 presents the IR spectra of Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6 in the wave number range 1350–300 cm−1. The structures of both double phosphates are known [7, 8], so probable attribution of the recorded absorption bands to the vibrations in concrete polyhedra can be made in the light of the literature data. As the obtained compounds are isostructural, both spectra exhibit much similarity. Three regions of absorption bands can be distinguished in these spectra. The first one between 1350 and 700 cm−1 exhibits absorption maxima at 1220, 1155, 1110, 1020, 960, 925 cm−1. This wide band is characteristic for orthophosphate ions [18]. In this wave number range, the absorption is assigned to the stretching symmetric and asymmetric vibration of P–O bonds in the PO4 tetrahedra [18,19,20,21,22,23]. The second absorption band in the wavenumber range 700–490 cm−1 probably comes from the stretching vibrations of M–O bonds in the MO5 (M = Mg, Zn) and MO6 (M = Cr, Mg, Zn) polyhedra coupled with the vibrations in PO4 tetrahedra [19, 20, 23]. The absorption bands recorded in the remaining wave number range, i.e. 490–300 cm−1, can be ascribed to the deformation vibrations of O–M–O and O–P–O bonds in PO4 groups [19,20,21,22, 24], or they can have a combined nature [21, 23].

Fig. 4
figure 4

IR spectra of Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6

Full size image

New solid solution

Considering approximate values of Zn2+ and Mg2+ radii in MO6 octahedra (i.e. 0.074 and 0.072 nm, respectively) [25], it can be expected that Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6 will form a substitution solid solution. Replacement of zinc ions by magnesium ions is realized, for example, in the solid solution of the formula Zn2–xMgxFeV3O11 [26]. This type of modification is also reported in the case of pigments, for example mentioned earlier Mn3−xMgxFe4(PO4)6 [9] or solid solution Mn2-xMgxP2O7, which can be used as a pink pigment for conventional ceramic glazes [27].

In order to find out whether Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6 form a solid solution, an equimolar mixture of these compounds was heated in two 12-h stages at 1100 °C. XRD analysis results, performed for the sample after reaction, show it to be monophasic, which indicates that a new solid solution is formed, according to the equation:

$${\text{Zn}}_{3} {\text{Cr}}_{4} \left( {{\text{PO}}_{4} } \right)_{{6({\text{s}})}} + {\text{Mg}}_{3} {\text{Cr}}_{4} \left( {{\text{PO}}_{4} } \right)_{{6({\text{s}})}} = 2{\text{Zn}}_{1.5} {\text{Mg}}_{1.5} {\text{Cr}}_{4} \left( {{\text{PO}}_{4} } \right)_{{6({\text{s}})}}$$
(6)

The new solid solution was obtained also according to another reaction, i.e.

$$3\left( {3{\text{MgCO}}_{3} \cdot{\text{Mg}}\left( {\text{OH}} \right)_{2} \cdot3{\text{H}}_{2} {\text{O}}} \right)_{{({\text{s}})}} + 12{\text{ZnO}}_{{({\text{s}})}} + 16{\text{Cr}}_{2} {\text{O}}_{{3({\text{s}})}} + 48\left( {{\text{NH}}_{4} } \right)_{2} {\text{HPO}}_{{4({\text{s}})}} = 8{\text{Zn}}_{1.5} {\text{Mg}}_{1.5} {\text{Cr}}_{4} \left( {{\text{PO}}_{4} } \right)_{{6({\text{s}})}} + 9{\text{CO}}_{{2({\text{g}})}} + 96{\text{NH}}_{{3({\text{g}})}} + 84{\text{H}}_{2} {\text{O}}_{{({\text{g}})}}$$
(7)

The stoichiometric mixture of reactants was heated in the following stages: 350 °C(8 h) → 550 °C(8 h) → 650 °C(12 h) + 750 °C(12 h) + 950 °C(12 h) + 1000 °C(12 h) + 1050 °C(12 h) + 1100 °C(12 h) × 2.

The powder diffraction pattern of the sample after its last heating stage did not differ from that obtained for Zn1.5Mg1.5Cr4(PO4)6, synthesized according to Eq. (6).

Formation of the new solid solution was corroborated by DTA method and by density measurements. On the DTA curve of Zn1.5Mg1.5Cr4(PO4)6 (Fig. 5), one endothermic effect was recorded (only its end was not registered). The effect has its onset at 1412 °C. Therefore, it is higher than melting temperature of Zn3Cr4(PO4)6 (i.e. 1340 °C) and lower than that of the estimated melting point of Mg3Cr4(PO4)6 (i.e. ~ 1450 °C). Therefore, incorporation of Mg2+ ions into the crystal lattice of Zn3Cr4(PO4)6 enhanced the thermal stability of the obtained phase. Density determined for Zn1.5Mg1.5Cr4(PO4)6 equals to 3.84 ± 0.01 g cm−3, so it is between the values 4.02 and 3.59 g cm−3, determined for Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6, respectively. The change in density of the studied phases is consistent with the expectation; namely, it is connected with a big difference in atomic masses between zinc and magnesium.

Fig. 5
figure 5

Fragment of DTA curve of Zn1.5Mg1.5Cr4(PO4)6

Full size image

Spectrophotometric measurements

The monophasic samples containing Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6 as well as the solid solution Zn3−xMgxCr4(PO4)6 for x = 1.50 were subjected to UV–Vis (DRS) study that permitted estimation of the energy gaps as described inter alia in the paper [28]. Figure 6 presents the plots of K2 = f(E), where K is obtained from the Kubelka–Munk transformation (Spectra Analysis Program) and in the plot is marked as (K/M)2. The energy gap values were read off as the points of intersection of the tangent to a given curve and the abscissa axis. The energy gap (Eg) determined for Zn3Cr4(PO4)6 was ~ 4.00 eV, the energy gap (Eg) determined for Mg3Cr4(PO4)6 was ~ 3.60 eV, while the energy gap (Eg) determined for the solid solution Zn1.5Mg1.5Cr4(PO4)6 was ~ 3.75 eV. On the basis of the obtained energy gap values, the phases were classified as wide band gap semiconductors.

Fig. 6
figure 6

Kubelka–Munk transformation of the UV–Vis–DR spectra of compounds: Mg3Cr4(PO4)6, Zn3Cr4(PO4)6 and the solid solution Zn1.5Mg1.5Cr4(PO4)6

Full size image

The physicochemical characterization of the all obtained phases was supplemented with determination of their colour expressed in the coordinates of CIE L*C*h* colour space, i.e. in the system which is a variant of CIE L*a*b* system. The L*C*h colour space is preferred by some industry professionals because its system correlates well with how the human eye perceives colour. In this system, the parameter L* represents the lightness or darkness of colour as related to a neutral grey scale which is described by numbers from 0 (black) to 100 (white), the parameter C* (chroma) represents saturation of the colour, and the parameter h* (colour hue) represents the hue angle which characterizes the colour shade. The values of C* and h* are indicated by the values of a* (the green → red axis) and b* (the blue → yellow axis) parameters of CIEL*a*b* system: C* = (a*2 + b*2)0.5; h* = arctan(b*/a*)−1. On the basis of the UV–Vis diffuse reflectance spectrum, the values of chromaticity coordinates L* C* h* are given in Table 2. On the basis of the a*, b*, C*, h* and L* values, the colours of the obtained phases are very similar, and according to the RAL system, they can be described as boulder brown (RAL090 40 10).

Table 2 Chromaticity coordinates a*, b*, C* and h* and lightness of the resulting phases
Full size table

Conclusions

  • For the first time, non-waste solid-state reaction method was applied to synthesize Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6.

  • The obtained compounds of boulder brown colour have high thermal stability in air (above 1300 °C) and form crystallites of the sizes suitable for their application as inorganic pigments.

  • Obtaining of a new solid solution of the formula Zn1.5Mg1.5Cr4(PO4)6 indicates the possibility of ion replacement in the crystal lattice of Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6. It is very important for designing of new materials, for example new inorganic pigments, because it enables modification of composition and therefore properties of such materials.

  • On the basis of the estimated energy gap values, all the phases were classified as wide band gap semiconductors.

Studies of Zn3Cr4(PO4)6 and Mg3Cr4(PO4)6 as well as the new solid solution Zn1.5Mg1.5Cr4(PO4)6 in the aspect of their colour properties and anticorrosion efficiency as the potential ceramic pigments are planned.