Color Tuning in Garnet Oxides: The Role of

DOI: 10.1002/asia.201701040

Pigments

Full Paper

| Very Important Paper |

Color Tuning in Garnet Oxides: The Role of Tetrahedral Coordination Geometry for 3 d Metal Ions and Ligand–Metal Charge Transfer (Band-Gap Manipulation) Anupam Bhim, Sourav Laha, Jagannatha Gopalakrishnan,* and Srinivasan Natarajan*[a] Abstract: We explored garnet-structured oxide materials containing 3d transition-metal ions (e.g., Co2 + , Ni2 + , Cu2 + , and Fe3 + ) for the development of new inorganic colored materials. For this purpose, we synthesized new garnets, Ca3Sb2Ga2ZnO12 (I) and Ca3Sb2Fe2ZnO12 (II), that were isostructural with Ca3Te2Zn3O12. Substitution of Co2 + , Ni2 + , and Cu2 + at the tetrahedral Zn2 + sites in I and II gave rise to brilliantly colored materials (different shades of blue, green, turquoise, and red). The materials were characterized by optical absorption spectroscopy and CIE chromaticity diagrams. The Fe3 + -containing oxides showed band-gap narrowing (owing

Introduction Crystalline inorganic compounds with bright colors have been the object of interest from ancient times.[1] Naturally occurring colored compounds such as Egyptian blue (CaCuSi4O10), Lapis lazuli [(NaCa)8Al6Si6O24(S,SO)4], Han blue (BaCuSi4O10), and Azurite [Cu3(CO3)2(OH)2], for example, have been employed as blue pigments and coloring agents.[1] In addition, colored solids such as ruby [Cr3 + -doped Al2O3], emerald [Cr3 + -doped Be3Al2(SiO3)6], and sapphire [Ti4 + /Fe2 + -doped Al2O3] have found use as precious gemstones.[2] The continuing demand for new colored materials has generated a number of synthetic colored compounds. Many of these are based on oxide hosts containing transition elements. Thus, Cu-substituted apatites,[3] Mn-substituted YInO3,[4] and CaTaO2N-LaTaON2[5] perovskites have been explored recently with much promise. The intense color of inorganic solids arises from the presence of transition elements with partially filled d electrons in various coordination geometries (d–d transition), intervalence charge transfer

[a] A. Bhim, Dr. S. Laha, Prof. Dr. J. Gopalakrishnan, Prof. Dr. S. Natarajan Solid State and Structural Chemistry Unit Indian Institute of Science Bangalore-560012 (India) E-mail: [email protected] [email protected] Homepage: http://sscu.iisc.ernet.in/frameworkslab/index.html Supporting Information and the ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/ asia.201701040. Chem. Asian J. 2017, 12, 2734 – 2743

to strong sp–d exchange interactions between Zn2 + and the transition-metal ion), and this tuned the color of these materials uniquely. We also characterized the color and optical absorption properties of Ca3Te2Zn3@xCoxO12 (0 < x , 2.0) and Cd3Te2Zn3@xCoxO12 (0 < x , 1.0), which display brilliant blue and green-blue colors, respectively. The present work brings out the role of the distorted tetrahedral coordination geometry of transition-metal ions and ligand–metal charge transfer (which is manifested as narrowing of the band gap) in producing brilliantly colored garnet-based materials.

between two elements, and band-gap engineering through specific doping.[6, 7] Octahedral and tetrahedral coordinations are two common geometries, and the optical absorption spectra of the transition elements are well established.[6, 8] There has also been interest to study the distorted geometries of transition elements, as they produce optical transitions that are different from those of the regular coordination geometries. In the recent past, we explored distorted coordination geometries of the transition elements in many hosts and produced uniquely colored solids.[9–11] Many minerals have been known to form with bright colors,[1] and this is due to doping of transition elements of varying concentrations. Of the many minerals, Garnet is important, forms in a cubic symmetry (space group: Ia3¯d), and has the general formula [A3][B2][C3]O12, in which the A, B, and C cations are 8- (dodecahedron), 6- (octahedron), and 4- (tetrahedron) coordinate, respectively. Well-known garnets include Y3Al5O12 (YAG), Y3In5O12 (YIG), and Y3Ga5O12 (YGG), and they have been investigated as laser hosts over the years.[12] The coordination variability of the cations in the garnet structure has been gainfully exploited by producing many interesting compounds. Thus, substitution at the “A” site by rare-earth ions such as Eu3 + and Tb3 + ions gives rise to intense red and green emissions.[13, 14] Silicate garnet, Ca3Cr2Si3O12, displays an intense green color and is commercially known as Victoria Green.[15] Almandine, Fe3Al2Si3O12, with the garnet structure exhibits a dark-red color.[16] We were particularly interested in studying garnet compounds containing Te6 + and Sb5 + ions, which prefer octahedral

2734

T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper coordination in oxide hosts, and Ca3Te2Zn3O12 and Cd3Te2Zn3O12 were prepared with the garnet structure.[17] Substitution of cobalt in place of zinc in these compounds resulted in intense blue and blue-green color, respectively.[17] It occurred to us that other garnet analogues could be prepared by suitably modifying the composition of Ca3Te2Zn3O12 garnet. We explored replacing Te6 + ions with Sb5 + ions, as both have comparable ionic radii (Te6 + = 0.56 a, Sb5 + = 0.6)[18] and bonding preferences.[19] To maintain charge balance, we replaced two Zn2 + ions by Ga3 + ions or Fe3 + ions (ionic radii Zn2 + = 0.6 a, Ga3 + = 0.47 a, and Fe3 + = 0.49 a)[18] at the tetrahedral site. Thus, two new garnet hosts with the formulas Ca3Sb2Ga2ZnO12 (white) and Ca3Sb2Fe2ZnO12 (beige yellow) were realized. These new host compounds offer the possibility of substituting + 2 transition-metal ions in place of Zn2 + ions and + 3 transitionmetal ions in place of Ga3 + / Fe3 + in tetrahedral coordination. In this work, we prepared Ca3Sb2Ga2CoO12 (brilliant blue), Ca3Sb2Ga2NiO12 (sky blue), Ca3Sb2Ga2Zn0.5Cu0.5O12 (light red), Ca3Sb2Fe2CoO12 (jade green), and Ca3Sb2Fe2Zn0.5Ni0.5CoO12 (sea green) by suitable substitution at the tetrahedral sites. We performed a detailed study of the structure and optical absorption behavior of all these compounds. In this paper, we describe the synthesis, structure, and optical properties of these compounds.

Figure 1. PXRD patterns of different Ca3Sb2Ga2Zn1@xMIIxO12 (MII = Co, Ni, and Cu) samples.

We performed Rietveld refinements of Ca3Sb2Ga2Zn1@xMIIxO12 [MII = Co (x = 1), Ni (x = 1), and Cu (x = 0.5)] from the PXRD data. The refinement results for Ca3Sb2Ga2CoO12 are shown in Figure 2 and Table 1. Refinement details for the other com-

Results and Discussion Synthesis and structure To investigate the role of the transition elements in imparting color to the garnet host, we prepared two new compounds, Ca3Sb2Ga2ZnO12 (white) and Ca3Sb2Fe2ZnO12 (beige yellow), with the garnet structure. Both compounds were derived from the known Ca3Te2Zn3O12 structure.[17] During the present study, we attempted simple substitutions in this structure: ZnII (t) + TeVI (Oh)!MIII/IV (t) + SbV/SnIV (Oh) (M = AlIII, GaIII, FeIII, SiIV, and GeIV). We were successful in preparing single-phase compounds with GaIII and FeIII along with ZnII at the tetrahedral site and SbV at the octahedral site. The ZnII ions were substituted by CoII, NiII, and CuII ions. Thus, single-phase garnets were obtained for Ca3Sb2Ga2Zn1@xCoIIxO12 (x = 1) (brilliant blue), Ca3Sb2Ga2Zn1@xNiIIxO12 (x = 1) (sky blue), Ca3Sb2Ga2Zn1@xCuIIxO12 (0 < x , 0.50) (light red), Ca3Sb2Fe2Zn1@xCoIIxO12 (x = 1) (jade green), Ca3Sb2Fe2Zn1@xNiIIxO12 (0 < x , 0.5) (sea green), Ca3Sb2Ga2@xFexCoO12 (x = 0.5, 1) (cerulean and pine green), and Ca3Sb2Ga2@xFexNiO12 (x = 0.5, 1) (cyan and turquoise).

Figure 2. Rietveld refinement of Ca3Sb2Ga2CoO12 from the PXRD data. The observed (O), calculated (red line), and difference (bottom blue line) profiles are shown. The vertical bars (j) indicate Bragg reflections.

Table 1. Crystallographic parameters for Ca3Sb2Ga2CoO12 obtained from Rietveld refinement of the PXRD data.[a]

Ca3Sb2Ga2Zn1@@xMIIxO12 (MII = Co, Ni, and Cu) The observed powder X-ray diffraction (PXRD) patterns of the single-phase compounds were compared with that of the Ca3Te2Zn3O12 phase (Figure 1 and Figure S1 in the Supporting Information). The PXRD pattern along with energy-dispersive X-ray (EDX) spectroscopy analysis (Figure S2) reveal that the compounds, Ca3Sb2Ga2Zn1@xMIIxO12 (MII = Co, Ni, and Cu), are single phase and isostructural with the Ca3Te2Zn3O12 parent. Chem. Asian J. 2017, 12, 2734 – 2743

www.chemasianj.org

2735

Atom

Site

x

y

z

Uiso [a2]

Occupancy

Ca Sb Ga/Co O

24c 16a 24d 96h

0.125 0 0.375 0.102(1)

0.0 0 0 0.199(2)

0.25 0 0.25 0.278(1)

0.014(2) 0.012(1) 0.014(2) 0.013(1)

1 1 0.666/0.334 1

[a] Space group Ia3¯d: a = b = c = 12.552(4) a, a = b = g = 908; reliability factors: Rp = 2.04 %, Rwp = 3.24 %, c2 = 5.83; bond lengths [a]: Ga/Co@O 1.873(1), Sb@O 1.998(2), Ca@O 2.469(2) (average). T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper we could replace only half of the zinc atoms to give Ca3Sb2Fe2Zn0.5Ni0.5O12, whereas cobalt replaced zinc fully to give Ca3Sb2Fe2CoO12. PXRD (Figures 4 and S7) and EDX analysis (Figure S2) indicates the formation of single-phase materials that are isostructural with Ca3Sb2Ga2ZnO12. The Rietveld refinement data for

Table 2. Summary of crystal-structure data of Ca3Sb2(Ga/Fe)2Zn1@xMIIxO12 (MII = Co, Ni and Cu).

Compound

a [a]

V [a3]

Ca3Sb2Ga2ZnO12 Ca3Sb2Ga2CoO12 Ca3Sb2Ga2NiO12 Ca3Sb2Ga2Zn0.5Cu0.5O12 Ca3Sb2Fe2ZnO12 Ca3Sb2Fe2CoO12 Ca3Sb2Fe2Zn0.5Ni0.5O12 Ca3Sb2GaFeCoO12 Ca3Sb2GaFeNiO12 Ca3Te2Zn3O12[a]

12.546(5) 12.552(4) 12.546(4) 12.536(4) 12.607(6) 12.618(7) 12.612(6) 12.583(4) 12.573(3) 12.580(2)

1974.76(24) 1977.44(20) 1975.04(20) 1969.84(21) 2003.94(28) 2009.22(35) 2005.99(12) 1992.31(18) 1987.53(11) 1990.91(7)

[a] The data for Ca3Te2Zn3O12 are from Ref. [17].

pounds are provided in Figures S3–S5 and Tables S1–S3. The summary of the crystal-structure data is given in Table 2. The Rietveld analyses indicate that the Ca3Sb2Ga2Zn1@xMIIxO12 members essentially have the same structure as Ca3Te2Zn3O12. In Figure 3, the unit cell of the Ca3Sb2Ga2CoO12 crystal structure is shown. The structure consists of SbO6 octahedra and (Ga/ Co)O4 tetrahedra connected through their corners to form a Figure 4. PXRD patterns of different Ca3Sb2Fe2Zn1@xMIIxO12 (MII = Co and Ni) samples.

Ca3Sb2Fe2CoO12 are given in Figure 5 and Table 3. The refinement results for all the other compounds are given in Figures S8 and S9 and Tables S4 and S5. The Rietveld refinement studies suggest that the cell volume for the Fe-containing compounds (e.g., Ca3Sb2Fe2CoO12, 2009.22 a3) is slightly expanded relative to that of the Ga-containing compounds (e.g., Ca3Sb2Ga2CoO12, 1977.44 a3) (Figure 6). This may be expected, as the ionic radius of FeIII (0.49 a) is slightly larger than that of GaIII (0.47 a). Figure 3. a) Crystal structure of Ca3Sb2Ga2CoO12 drawn from the Rietveld refinement data. b) Ga/CoO4 tetrahedra with bond lengths and bond angles.

chain along the three crystallographic axes (Figure S6 a). The interconnectivity between the chains gives rise to 16-membered rings, in which the CaII cations are located in a distorted, 8-coordinated environment (Figure S6 b). The tetrahedral Ga/Co@O distances have an average value of 1.873 a. The O@Ga/Co@O bond angles exhibit values of 114.92 and 99.058 (Figure 3 b), which indicates that the Ga/CoO4 tetrahedra are distorted. Similar distortions of the tetrahedral sites in the garnet structure have been observed earlier.[20] Ca3Sb2Fe2Zn1@@xMIIxO12 (MII = Co and Ni) Ca3Sb2Fe2ZnO12 is a new garnet that we prepared in this work. The PXRD pattern indicates that the compound has the same structure as Ca3Te2Zn3O12. We substituted cobalt and nickel in place of zinc in this structure. In the case of NiII substitution, Chem. Asian J. 2017, 12, 2734 – 2743

www.chemasianj.org

Figure 5. Rietveld refinement of Ca3Sb2Fe2CoO12 from the PXRD data. The observed (O), calculated (red line), and difference (bottom blue line) profiles are shown. The vertical bars (j) indicate Bragg reflections.

2736


Full Paper Table 3. Crystallographic parameters for Ca3Sb2Fe2CoO12 obtained from Rietveld refinement of the PXRD data.[a]

Atom Ca Sb Fe/Co O

Site 24c 16a 24d 96h

x 0.125 0 0375 0.103(1)

y 0.0 0 0 0.200(1)

z 0.25 0 0.25 0.276(2)

Uiso [a2] 0.011(1) 0.011(2) 0.018(1) 0.014(1)

Occupancy 1 1 0.666/0.334 1

[a] Space group Ia3¯d: a = b = c = 12.618(7) a, a = b = g = 908; reliability factors: Rp = 1.91 %, Rwp = 3.20 %, c2 = 5.80; bond lengths [a]: Fe/Co@O 1.905(1), Sb@O 1.985(1), Ca@O 2.481(2) (average).

Figure 8. Optical absorption (UV/Vis) spectra of Ca3Sb2Ga2Zn1@xCoxO12 (0 < x , 1.0).

Figure 6. Unit-cell volume variation in Ca3Sb2(Ga/Fe)2Zn1@xMIIxO12 (MII = Co and Ni) compounds.

We also explored the formation of solid solutions of Ca3Sb2Ga2@xFexCoO12 (x = 0.5, 1) and Ca3Sb2Ga2@xFexNiO12 (x = 0.5, 1). PXRD (Figure S10) and EDX analysis (Figure S2) indicates that the compounds are single phase with a garnet structure. The refinement results for all the compounds are collected in Figures S11 and S12 and Tables S6 and S7. Color and optical properties Ca3Sb2Ga2Zn1@@xMIIxO12 (MII = Co, Ni, and Cu) As mentioned before, the parent compound, Ca3Sb2Ga2ZnO12, is white in color with a band gap (Eg) of approximately 4.2 eV. We find that on doping with Co2 + in place of Zn2 + , the compound turns sky blue to intense blue in color with increasing cobalt content (Figure 7). The optical absorption spectra (Figure 8) of Ca3Sb2Ga2Zn1@xCoxO12 (0 < x , 1.0) show a broad triplet absorption band that falls approximately in the 1.75– 2.50 eV range with a maximum at approximately 2.00 eV, which leaves a window of almost no absorption in the blue

Figure 7. Colors of the Ca3Sb2Ga2Zn1@xCoxO12 members in daylight. Chem. Asian J. 2017, 12, 2734 – 2743

www.chemasianj.org

region of the visible spectrum. The strong absorption observed at approximately 2.00 eV is in the orange region; the complementary color would be blue.[8] The Co2 + ions in the tetrahedral symmetry exhibit three spin-allowed transitions: 1) 4A2(F)!4T2(F), 2) 4A2(F)!4T1(F), and 3) 4A2(F)!4T1(P).[8, 21–23] Of these, transitions 1 and 2 are in the IR region, and transition 3 falls in the visible region. Thus, the triplet band centered at approximately 2.00 eV can be assigned to the tetrahedral Co2 + [4A2(F)!4T1(P)] transition. It may be noted that the tetrahedral sites in the garnet structure occupy a special position, (24d; x = 0.375, y = 0.0, z = 0.25). In our present structure, the Ga/Co@O distances are the same, but the O@Ga/Co@O bond angles show variations that result in distortion to an S4 symmetry.[20] This would account for the splitting of the 4A2(F)!4T1(P) transition in the compound. Considering a possible application as an inorganic coloring agent (pigment), we performed a stability test for Ca3Sb2Ga2CoO12 by keeping the sample for 24 h in hot (boiling) water as well as in HNO3. The samples were filtered, dried, and examined for their color and structure. We found that the materials retained their structures and colors after the acid/boiling water treatment (for the PXRD patterns and UV/Vis spectra, see Figures S13–S15). The NiII-substituted compounds, Ca3Sb2Ga2Zn1@xNixO12 (0 < x , 1.0), also exhibit a blue color. The end member, Ca3Sb2Ga2NiO12, shows a sky blue color (Figure 9). The optical absorption spectra of Ca3Sb2Ga2Zn1@xNixO12 (0 < x , 1.0) exhibit (Figure 10) a doublet absorption in the range of 1.60 to 2.5 eV with maxima at approximately 1.85 and 2.00 eV. This leaves a small valley in the blue region of the visible spectrum. In addition, we observe two weak absorptions at 1.68 and 2.25 eV. The doublet absorption observed at approximately 1.85 and 2.00 eV is in the orange-red region, and the complementary color would be blue. The combined result is the observed blue color.

2737


Full Paper usual for Cu2 + in tetrahedral coordination. Optical absorption studies indicate a broad absorption centered at approximately 2.5 eV along with another broad absorption at 3.6 eV. We do not observe any bands in the visible region. Generally, coppercontaining minerals and compounds exhibit absorption in the near IR region that tails into the visible region to give rise to blue, green, or blue-green colors. The broad absorption at 3.6 eV could be due to the ligand-to-metal charge-transfer (LMCT) band, which is comparable to that of Cu2O (3.5 eV).[29] To confirm the observations of the optical absorption spectral study, we performed an X-ray photoelectron spectroscopy (XPS) investigation of the copper-containing compound. The XPS studies indicate two peaks at 952.71 and 932.64 eV that compare well with the 2p1/2 and 2p3/2 peaks of CuI in Cu2O.[30] The optical absorption along with the XPS studies suggest that the copper in the present compound exhibits a 1 + oxidation state and that the formula of the compound would be Ca3Sb2Ga2Zn0.5CuI0.5O11.75. This formalism is more likely, as CuI ions generally occupy tetrahedral lattice sites.[31] In addition, our efforts to increase the concentration of copper in Ca3Sb2Ga2Zn1@xCuxO12 beyond x = 0.50 yielded mixed phases. It is likely that the garnet structure may not tolerate higher oxygen vacancy, which limits the substitution of copper (CuI) in this structure to 0.5 only.

Figure 9. Colors of the Ca3Sb2Ga2Zn1@xNixO12 members in daylight.

Ca3Sb2Fe2ZnO12 Figure 10. Optical absorption (UV/Vis) spectra of Ca3Sb2Ga2Zn1@xNixO12 (0 < x , 1.0).

It is known that three spin-allowed transitions can be expected for tetrahedral Ni2 + : 3T1(F)!3T2(F), 3A2(F), and 3 T1(P).[8, 24–26] Of these, the 3T1(F)!3T1(P) transition is the only spin-allowed transition in the visible region. The doublet absorption observed at approximately 1.85 and 2.00 eV in Ca3Sb2Ga2Zn1@xNixO12 (0 < x , 1.0) may be due to the 3T1(F)! 3 T1(P) transition within the tetrahedral symmetry. The two shoulders at 1.68 and 2.25 eV may be due to spin–orbit coupling, which is known for d8 ions such as Ni2 + in the tetrahedral symmetry.[8, 26] It may be noted that stabilizing Ni2 + in the tetrahedral coordination is not common. On the other hand, cobalt (Co2 + ) in tetrahedral coordination is known (e.g., spinel CoAl2O4,[27] willemite CoxZn2@xSiO4,[28] and a-LiZn0.5Co0.5BO3[9]). All the compounds exhibit intense blue color. Nickel (Ni2 + ) in tetrahedral coordination is limited to a few examples: Ni-doped aLiZnBO3,[9] Ni-doped gahnite,[25] and Ni-doped hibonite (CaAl12O19).[26] These compounds exhibit magenta, blue, and turquoise colors, respectively. In the present study, the garnet host stabilizes Ni2 + in the tetrahedral site with a sky blue color, which is unique and significant. We attempted to substitute copper in place of Zn in Ca3Sb2Ga2Zn1@xCuxO12 (0 < x , 0.50). Our studies indicate the formation of single-phase compounds only up to x = 0.5. The as-synthesized compounds have a reddish color, which is unChem. Asian J. 2017, 12, 2734 – 2743

www.chemasianj.org

This material displays a beige yellow color. The optical absorption spectrum (Figure 11) shows a broad absorption that covers the region from approximately 2.20 to 3.00 eV, with a maximum at approximately 2.50 eV. The broad band leaves a window of no absorption from the yellow-orange to red region of the visible spectrum, which gives a beige yellow color to the compound. The absorption band further gains intensity and extends into the UV region. The ligand-field transitions are spin forbidden in FeIII :d5 because of the sextet [6A1(S)]

Figure 11. Optical absorption (UV/Vis) spectra of Ca3Sb2Fe2Zn1@xCoxO12 (0 < x , 1.0).

2738


Full Paper ground state. We could assign the band centered at 2.50 eV to the 6A1(6S)!4T1(G)/4T2(G) transitions within the tetrahedral symmetry, similar to the assignments for Fe3 + :LiAlO2,[32] Fe3 + :SiO2,[33] and Fe3 + in LiZnBO3.[9] Ca3Sb2Fe2Zn1@@xMIIxO12 (MII = Co and Ni) Ca3Sb2Fe2Zn1@xCoxO12 (0 < x , 1.0) compounds possess an unusual green color (Figure 12). The intensity of the green color of Ca3Sb2Fe2Zn1@xCoxO12 increases progressively from turquoise green (0 < x , 0.25) to intense green (0.25 < x , 1.0) upon increasing the value of x; whereas the end members, Ca3Sb2Fe2Zn1@xCoxO12 (x = 0) and Ca3Sb2Fe2Zn1@xCoxO12 (x = 1.0), are beige yellow and jade green, respectively (Figure 12).

the NiII-substituted compounds, we also observe a decrease in the band gap, which tunes the color from beige yellow to sea green. We further explored another two series of compounds, Ca3Sb2Ga2@xFexCoO12 (0 < x , 2.0) (cerulean and pine green) (Figure 13) and Ca3Sb2Ga2@xFexNiO12 (0 < x , 1.0) (cyan and turquoise) (Figure 14), by tuning the concentrations of FeIII and GaIII in the host matrix.

Figure 13. Colors of the Ca3Sb2Ga2@xFexCoO12 (0 < x , 2.0) members in daylight.

Figure 12. Colors of the Ca3Sb2Fe2Zn1@xCoxO12 members in daylight.

The optical absorption spectra (Figure 11) of Ca3Sb2Fe2Zn1@xCoxO12 (0 < x , 1.0) show a broad triplet absorption band that ranges from approximately 1.75 to 2.50 eV with a maximum at approximately 2.00 eV that is due to a spin-allowed transition [4A2(F)!4T1(P)] in the visible region[8, 21–23] for tetrahedral Co2 + , similar to that observed for the Ca3Sb2Ga2Zn1@xCoxO12 (0 < x , 1.0) members. Interestingly, we observe an enormous redshift in the Eg edge as the value of x increases in Ca3Sb2Fe2Zn1@xCoxO12 (0 < x , 1.0). Our band-gap calculations show that there is a remarkable difference in the band gaps of Ca3Sb2Fe2ZnO12 (Eg & 3.21 eV) and Ca3Sb2Fe2CoO12 (Eg & 2.41 eV). The decrease in band gap is attributed to a strong sp–d exchange interaction between the band electrons of Zn (which is associated with Fe3 + ) and the localized d electrons of the Co2 + ions substituting for Zn2 + ions in the system.[7, 34–36] The strong exchange interaction (s–d and p–d) results in a positive and negative correction to the valence and conduction band edges, respectively, and leads to narrowing of the band gap (Figure S19).[36, 37] Consequently, the ligand-tometal charge-transfer process is shifted to lower energy. Divalent nickel substitution in Ca3Sb2Fe2ZnO12, that is, Ca3Sb2Fe2Zn1@xNixO12 (0 < x , 0.50), results in a single phase only up to x = 0.50, and the solid solution displays a unique sea green color (Figure S20, inset). The optical absorption spectra of Ca3Sb2Fe2Zn1@xNixO12 (0 < x , 0.50) show (Figure S20) a strong doublet absorption band that ranges from approximately 1.60 to 2.5 eV with maxima at photon energies of approximately 1.85 and 2.00 eV that are due to the 3T1(F)!3T1(P) transition within the tetrahedral symmetry.[8, 24–26] Two shoulders at approximately 1.68 and 2.25 eV are attributed to the spin–orbit coupling of d8 :Ni2 + in tetrahedral symmetry.[8, 26] In Chem. Asian J. 2017, 12, 2734 – 2743

www.chemasianj.org

Figure 14. Colors of the Ca3Sb2Ga2@xFexNiO12 (0 < x , 1.0) members in daylight.

The optical absorption spectra (Figure 15) of Ca3Sb2Ga2@xFexCoO12 (0 < x , 2.0) reveal the characteristic broad

Figure 15. Optical absorption (UV/Vis) spectra of Ca3Sb2Ga2@xFexCoO12 (0 < x , 2.0).

2739


Full Paper triplet absorption band that ranges from approximately 1.75 to 2.60 eV with a maximum at approximately 2.00 eV that is due to a spin-allowed transition [4A2(F)!4T1(P)] in the visible region[8, 21–23] for tetrahedral Co2 + , similar to that described above. On the other hand, the optical absorption spectra (Figure 16) of Ca3Sb2Ga2@xFexNiO12 (0 < x , 1.0) show the typical strong doublet absorption band of tetrahedral Ni2 + that ranges from approximately 1.60 to 2.5 eV due to the 3T1(F)! 3 T1(P) transition[8, 24–26] along with two shoulders at approximately 1.68 and 2.25 eV that can be attributed to the spin– orbit coupling of d8 :Ni2 + in tetrahedral symmetry.[8, 26]

Figure 17. CIE chromaticity diagram for the Ca3Sb2Fe2Zn1@xMIIxO12 (M = Co and Ni) and Ca3Sb2Ga2@xFexMIIO12 [(0 < x , 2.0) for CoII and (0 < x , 1.0) for NiII] compounds.

M3Te2Zn3@@xCoxO12 (M = Ca, Cd) The brilliant-blue garnet (Figure 18), Ca3Te2ZnCo2O12, was first prepared by Kasper[17] along with a green-blue garnet, Cd3Te2Zn2CoO12. However, these compounds were not considered as inorganic pigments. Cobalt-substituted Ca3Te2Zn3@xCoxO12 (0 < x , 2.0) and Cd3Te2Zn3@xCoxO12 (0 < x , 1.0) display a brilliant-blue color and a green-blue color with an intensity that increases with x (Figure 18). Figure 16. Optical absorption (UV/Vis) spectra of Ca3Sb2Ga2@xFexNiO12 (0 < x , 1.0).

As the concentration of FeIII increases in both systems of compounds, Ca3Sb2Ga2@xFexCoO12 (0 < x , 2.0) (Figure 15) and Ca3Sb2Ga2@xFexNiO12 (0 < x , 1.0) (Figure 16), we notice a tremendous reduction in the band gap mainly due to the strong exchange interaction between the localized d electrons of FeIII and the band electrons of CoII and NiII.[7, 34–36] Therefore, the basically blue-colored garnet transforms into a green-colored garnet mainly because of band-gap tuning of Ca3Sb2Ga2@xFexCoO12 (0 < x , 2.0), Ca3Sb2Ga2CoO12 (Eg & 3.21 eV) and Ca3Sb2Fe2CoO12 (Eg & 2.41 eV), and Ca3Sb2Ga2@xFexNiO12 (0 < x , 1.0), Ca3Sb2Ga2NiO12 (Eg & 3.25 eV) and Ca3Sb2GaFeNiO12 (Eg & 2.83 eV). We measured CIE 1931 chromaticity coordinates to determine the pigment quality of these materials. The CIE 1931 chromaticity diagram is given in Figure 17 and Table S8, and the color coordinates correspond to the green region of the chromaticity diagram. Similarly, the color coordinates of Ca3Sb2Fe2Zn0.5Ni0.5O12 correspond to the blue-green region of the CIE 1931 chromaticity diagram (Figure 17). Summaries of the cobalt- and nickel-containing uniquely colored garnets, Ca3Sb2Ga2@xFexCoO12 (0 < x , 2.0) and Ca3Sb2Ga2@xFexNiO12 (0 < x , 1.0), are provided in Figures S21 and S22 respectively. Chem. Asian J. 2017, 12, 2734 – 2743

www.chemasianj.org

Figure 18. Colors of the Ca3Te2Zn3@xCoxO12 and Cd3Te2Zn3@xCoxO12 members in daylight.

The optical absorption spectra (Figure S23) of Ca3Te2Zn3@xCoxO12 (0 < x , 2.0) show a broad triplet absorption band that ranges from approximately 1.75 to 2.75 eV with a maximum at approximately 2.00 eV, which leaves a window of almost no absorption in the blue region of the visible spectrum. This is very similar to the results discussed above for the antimony garnets. Thus, the brilliant-blue color can be attributed to distorted [(Zn/Co)O4] with tetrahedral Co2 + :[4A2 (F)! 4 T1(P)] transition in the visible region.[8, 21–23] The spin-forbidden transition is observed as a small band at approximately

2740


Full Paper 495 nm and is attributed to transitions between the octahedral and tetrahedral sites.[37, 38] On the other hand, the optical absorption spectra (Figure S24) of Cd3Te2Zn3@xCoxO12 (0 < x , 1.0) are different from those of the Ca3Te2Zn3@xCoxO12 (0 < x , 2.0) members. We observe a broad triplet absorption band that ranges from approximately 1.75 to 2.5 eV with a maximum at approximately 2.00 eV that is due to regular tetrahedral Co2 + [4A2(F)!4T1(P)] transition in the visible region.[8, 21–23] However, the band-gap edge is progressively shifted to a lower energy (x = 0.5, & 2.6 eV and x = 1, & 2.55 eV in Cd3Te2Zn3@xCoxO12) than that of pure Cd3Te2Zn3O12 ( & 3.3 eV). The redshift in the Eg edge upon inserting Co in the pure Zn matrix was already observed[7, 34] and explained as being mainly due to sp–d exchange interactions between the band electrons and the localized d electrons of the CoII ions substituting ZnII ions.[35] The s–d and p–d exchange interactions give rise to negative and positive corrections to the conduction-band and valence-band edges, respectively, and this leads to band-gap narrowing.[36] Thus, the Cosubstituted compounds leave a very small window in the green-blue region, which is the observed color of the compounds. We determined the CIE 1931 chromaticity coordinates to quantify the pigment quality of the Ca3Sb2Ga2CoO12, Ca3Te2ZnCo2O12, and Cd3Te2Zn2CoO12 samples. The results correspond to the blue region of the chromaticity diagram for Ca3Sb2Ga2CoO12 and Ca3Te2ZnCo2O12. On the other hand, the data for Cd3Te2Zn2CoO12 fall in the green-blue color region of the CIE 1931 chromaticity diagram (Figure 19 and Table S9). We find that the color coordinates of Ca3Sb2Ga2NiO12 correspond to the light-blue region of the CIE 1931 chromaticity diagram (Figure 19). The color coordinates (L*, a*, and b* parameters) for the blue-colored compounds were compared with those of cobalt aluminate[9] (CoAl2O4) as well as those of the recently discovered YIn0.90Mn0.10O3[9] and LiZn0.9Co0.1BO3[9] compounds. The data are presented in Figure 19 and Table 4. The present com-

Table 4. Color coordinates for Ca3Te2ZnCo2O12, Cd3Te2Zn2CoO12, Ca3Sb2Ga2CoO12, CoAl2O4, YIn0.9Mn0.1O3, and LiZn0.9Co0.1BO3.

Compound

L*

a*

b*

Ca3Te2ZnCo2O12 Cd3Te2Zn2CoO12 Ca3Sb2Ga2CoO12 CoAl2O4[a] YIn0.9Mn0.1O3[a] LiZn0.9Co0.1BO3[a]

28.27 39.46 32.46 44.80 39.71 29.87

9.63 @19.73 11.30 2.10 @5.78 32.10

@58.42 @21.56 @53.24 @32.70 @22.28 @62.67

[a] Data for CoAl2O4, YIn0.9Mn0.1O3, and LiZn0.9Co0.1BO3 were taken from Ref. [9].

pounds appear to be comparable to the reported blue-colored compounds. It may be noted that the number of available blue-colored compounds is limited, and the discovery of intense-blue-colored compounds in the present study not only increases the number of blue compounds but also the possibility of using them as active coloring agents in pigments based on inorganic compounds. IR reflectance spectra TiO2 is a white pigment having a strong near-IR reflectance.[39] The white-colored Ca3Sb2Ga2ZnO12 compound displays (Figure S25) near-IR reflectivity of approximately 80 %, and thus, it exhibits slightly better near-IR reflectivity than TiO2 ( & 70 %). Beige-yellow-colored Ca3Sb2Fe2ZnO12 shows near-IR reflectivity of approximately 70 %, which is comparable to that of TiO2. Therefore, the newly synthesized garnets could be good candidates as white pigments with high near-IR reflectivity.

Conclusions In this investigation, we successfully synthesized two new garnets, Ca3Sb2Ga2ZnO12 and Ca3Sb2Fe2ZnO12 (space group Ia3¯d), that are isostructural to the Ca3Te2Zn3O12 garnet. The host garnets were substituted with several 3d transition-metal ions (e.g., Co2 + , Ni2 + , and Cu2 + ). The optical absorption spectra of the transition-metal-ion-substituted Ca3Sb2Ga2ZnO12 and Ca3Sb2Fe2ZnO12 derivatives were interpreted on the basis of ligand-field transitions as well as ligand–metal charge-transfer transitions that occurred within the distorted tetrahedral Ga/ Fe/MIIO4 chromophores. The investigation also identified the crucial role of band-gap tuning, which was found to impart rare and unique colors to the materials. The parent Zn compounds, Ca3Sb2Ga2ZnO12 (band gap & 4.2 eV) and Ca3Sb2Fe2ZnO12 (band gap & 3.2 eV), appear to be good competitors to TiO2 for near-IR reflectance.

Experimental Section Reagents

Figure 19. CIE chromaticity diagram for the Ca3Te2ZnCo2O12, Cd3Te2Zn2CoO12 and Ca3Sb2Ga2Zn1@xMIIxO12 (M = Co and Ni) compounds. Chem. Asian J. 2017, 12, 2734 – 2743

www.chemasianj.org

The chemicals CdCO3, Ga2O3, ZnO, CuO (Sigma–Aldrich, 99.9 %), CaCO3, Fe2O3 (SD Fine, India), TeO2, Sb2O5, and MC2O4·2 H2O (M = Co, Ni) (Alfa Aesar, 99 %) were used as received.

2741


Full Paper Synthesis The syntheses of all the compounds were performed by using the conventional ceramic method. Stoichiometric mixtures of CaCO3, CdCO3, TeO2, Sb2O5, Ga2O3, Fe2O3, ZnO, MC2O4·2 H2O (M = Co, Ni), and CuO corresponding to the formula Ca3Sb2(Ga/Fe)2Zn1@xMIIxO12 were heated in air at 1000 to 1280 8C for 24 h with several intermittent grindings.

Characterization All compounds were characterized by PXRD, scanning electron microscopy (SEM), and energy-dispersive X-ray (EDX) spectroscopy. The PXRD patterns were recorded with a Philips X’pert diffractometer (Ni-filtered CuKa radiation, l = 1.5418 a). PXRD data for Rietveld refinement of the structures were collected at room temperature by employing the same diffractometer in the 2 q range of 5 to 1208 with a step size of 0.028 and a step duration of 50 s. The PXRD patterns were refined with the program GSAS-EXPGUI.[40] Lattice parameters, scale factors, background (Fourier polynomial background function), pseudo-Voigt (U, V, W, and X), and isothermal temperature factors (Uiso) were refined. Thermal parameters were constrained to be the same for different atoms occupying the same sites. PXRD patterns were simulated with the help of the program POWDERCELL.[41] The diffuse reflectance spectra for all powdered samples were recorded by using a PerkinElmer Lambda 750 UV/Vis double-beam spectrometer over the spectral region of l = 250 to 1200 nm. The reflectance data were converted into the Kubelka–Munk[42] function by using Equation (1): FðRÞ¼

ð1@RÞ2 a ¼ 2R S

ð1Þ

in which R is the reflectance, a is the absorptivity, and S is the scattering factor. The optical band gaps were calculated from Tauc plots.[43] The Tauc relation [Eq. (2): ahn ¼ Aðhn@E g Þn

ð2Þ

in which a is the absorption coefficient, A is a constant called the band tailing parameter, and n is the power factor of the transition mode. A plot of (ahn) (1/n) versus the photon energy (hn) gives a straight line in a certain region. Extrapolation of this straight line intercepts the hn axis to give the value of the optical band gap (Eg). Near-IR reflectance in the range of l = 500 to 2500 nm at RT was collected by using the same instrument. The pigment quality of the samples was obtained by using the CIE 1931 chromaticity coordinates in the l = 380–750 nm range. The CIE 1931 chromaticity coordinates were determined by using the gocie.exe program.[44] SEM images and EDX data were recorded with a FEI-ESEM Quanta 200 scanning electron microscope. X-ray photoelectron spectroscopy was performed with a laboratory-based XPS instrument.

Acknowledgements J.G. thanks the National Academy of Sciences, Allahabad, India (NASI), for the award of a Senior Scientist Fellowship. S.N. thanks the Science and Engineering Research Board (SERB), Government of India, for the award of a research grant (EMR/ 2016/002300/IPC) and a J. C. Bose National Fellowship. We are Chem. Asian J. 2017, 12, 2734 – 2743

www.chemasianj.org

thankful to Dr. Shyamashis Das and Dr. Sumanta Mukherjee for XPS measurements.

Conflict of interest The authors declare no conflict of interest. Keywords: dyes/pigments · garnets · solid-state structures · structure elucidation · tetrahedral geometry [1] P. Ball, Bright Earth: Art and the Invention of Color, The University of Chicago Press, Chicago, 2001. [2] H. Berke, Chem. Soc. Rev. 2007, 36, 15 – 30. [3] P. E. Kazin, M. A. Zykin, Y. V. Zubavichus, O. V. Magdysyuk, R. E. Dinnebier, M. Jansen, Chem. Eur. J. 2014, 20, 165 – 178. [4] A. E. Smith, H. Mizoguchi, K. Delaney, N. A. Spaldin, A. W. Sleight, M. A. Subramanian, J. Am. Chem. Soc. 2009, 131, 17084 – 17086. [5] M. Jansen, H. P. Letschert, Nature 2000, 404, 980 – 982. [6] R. J. D. Tilley, Color and the Optical Properties of Materials: An Exploration of the Relationship Between Light, the Optical Properties of Materials and Color, Wiley, Chichester, 2011. [7] J. J. Beltr#n, C. A. Barrero, A. Punnoose, J. Phys. Chem. C 2014, 118, 13203 – 13217. [8] A. B. P. Lever, Inorganic Electronic Spectroscopy, Elsevier, Amsterdam, 1968. [9] S. Tamilarasan, M. L. P. Reddy, S. Natarajan, J. Gopalakrishnan, Chem. Asian J. 2016, 11, 3234 – 3240. [10] S. Tamilarasan, D. Sarma, S. Bhattacharjee, U. V. Waghmare, S. Natarajan, J. Gopalakrishnan, Inorg. Chem. 2013, 52, 5757 – 5763. [11] S. Tamilarasan, S. Laha, S. Natarajan, J. Gopalakrishnan, Eur. J. Inorg. Chem. 2016, 288 – 293. [12] B. D. Bartolo, G. Armagan, Spectroscopy of Solid-State Laser-Type Materials, Plenum Press, New York, 1987. [13] X. Wang, Z. Zhao, Q. Wu, Y. Li, Y. Wang, Inorg. Chem. 2016, 55, 11072 – 11077. [14] W. Le, W. Lv, Q. Zhao, M. Jiao, B. Shaoab, H. You, J. Mater. Chem. C 2015, 3, 2334 – 2340. [15] S. G. Seo, B. H. Lee, J. Korean Ceram. Soc. 2010, 47, 608 – 612. [16] A. M. Ferrari, L. Valenzano, A. Meyer, R. Orlando, R. Dovesi, J. Phys. Chem. A 2009, 113, 11289 – 11294. [17] H. M. Kasper, Mater. Res. Bull. 1968, 3, 765 – 766. [18] R. D. Shannon, Acta Crystallogr. Sect. A 1976, 32, 751 – 767. [19] G. Blasse, J. Inorg. Nucl. Chem. 1965, 27, 993 – 1003. [20] A. F. Frau, J.-H. Kim, P. S. Halasyamani, Solid State Sci. 2008, 10, 1263 – 1268. [21] D. N. Sathyanarayana, Electronic Absorption Spectroscopy and Related Techniques, Universities Press (India) Limited, Hyderabad, 2001. [22] D. L. Wood, J. P. Remeika, J. Chem. Phys. 1967, 46, 3595 – 3602. [23] M. Llusar, A. Fores, J. A. Badenes, J. Calbo, M. A. Tena, G. Monros, J. Eur. Ceram. Soc. 2001, 21, 1121 – 1130. [24] T. C. Brunold, H. U. Gudel, E. Cavalli, Chem. Phys. Lett. 1997, 268, 413 – 420. [25] G. Lorenzi, G. Baldia, F. D. Benedettob, V. Faso, P. Lattanzi, M. Romanelli, J. Eur. Ceram. Soc. 2006, 26, 317 – 321. [26] G. Costa, M. J. Ribeiro, W. Hajjaji, M. P. Seabra, J. A. Labrincha, M. Dondi, G. Cruciani, J. Eur. Ceram. Soc. 2009, 29, 2671 – 2678. [27] M. Gaudon, L. C. Robertson, E. Lataste, M. Duttine, M. Menetrier, A. Demourgues, Ceram. Int. 2014, 40, 5201 – 5207. [28] A. For8s, M. Llusar, J. A. Badenes, J. Calbo, M. A. Tena, G. Monros, Green Chem. 2000, 2, 93 – 100. [29] Z. Kang, X. Yan, Y. Wang, Z. Bai, Y. Liu, Z. Zhang, P. Lin, X. Zhang, H. Yuan, X. Zhang, Y. Zhang, Sci. Rep. 2015, 5, 7882 – 7889. [30] M. Hara, T. Kondo, M. Komoda, S. Ikeda, J. N. Kondo, K. Domen, M. Hara, K. Shinohara, A. Tanaka, Chem. Commun. 1998, 357 – 358. [31] A. Buljan, M. Llunell, E. Ruiz, P. Alemany, Chem. Mater. 2001, 13, 338 – 344. [32] G. A. Waychunas, G. R. Rossman, Phys. Chem. Miner. 1983, 9, 212 – 215.

2742


Full Paper [33] D. M. Sherman, Phys. Chem. Miner. 1985, 12, 161 – 175. [34] M. Gaudon, O. Toulemonde, A. Demourgues, Inorg. Chem. 2007, 46, 10996 – 11002. [35] R. B. Bylsma, W. M. Becker, J. Kossut, U. Debska, D. Yoder-Short, Phys. Rev. B 1986, 33, 8207 – 8215. [36] K. J. Kim, Y. R. Park, Appl. Phys. Lett. 2002, 81, 1420 – 1422. [37] W. Wang, Z. Xie, G. Liu, W. Yang, Cryst. Growth Des. 2009, 9, 4373 – 4377. [38] W. Zheng, J. Zou, RSC Adv. 2015, 5, 87932 – 87939. [39] Y. Wang, J. Li, L. Wang, T. Qi, D. Chen, W. Wang, Chem. Eng. Technol. 2011, 34, 905 – 913. [40] A. C. Larson, R. B. Von Dreele, General Structure Analysis System (GSAS), Los Alamos National Laboratory Report LAUR, 2000, pp. 86 – 748.

Chem. Asian J. 2017, 12, 2734 – 2743

www.chemasianj.org

[41] [42] [43] [44]

W. Kraus, G. Nolze, J. Appl. Crystallogr. 1996, 29, 301 – 303. P. Kubelka, F. Munk, Z. Tech. Phys. 1931, 12, 593 – 601. J. Tauc, Mater. Res. Bull. 1968, 3, 37 – 46. https://code.google.com/archive/p/jtchem/downloads.

Manuscript received: July 20, 2017 Revised manuscript received: August 24, 2017 Accepted manuscript online: September 4, 2017 Version of record online: September 26, 2017

2743
