research papers\(\def\hfill{\hskip 5em}\def\hfil{\hskip 3em}\def\eqno#1{\hfil {#1}}\)

Journal logoSTRUCTURAL SCIENCE
CRYSTAL ENGINEERING
MATERIALS
ISSN: 2052-5206

Resolving the structure of V3O7·H2O and Mo-substituted V3O7·H2O

crossmark logo

aChemistry and Physics of Materials, University of Salzburg, Jakob-Haringer-Straße 2a, Salzburg 5020, Austria, bMaier-Leibnitz Zentrum MLZ Forschungsreaktor Munchen FRM-II, Lichtenbergstr. 1, Garching, Bavaria 85748, Germany, cInstitute of Surface Chemistry and Catalysis, Ulm University, Albert-Einstein-Allee 47, 89081-Ulm, Germany, and dInstitute of Theoretical Chemistry, Ulm University, Albert-Einstein-Allee 11, 89081-Ulm, Germany
*Correspondence e-mail: juergen.schoiber@sbg.ac.at

Edited by J. Hadermann, University of Antwerp, Belgium (Received 11 April 2022; accepted 22 June 2022; online 13 July 2022)

Vanadate compounds, such as V3O7·H2O, are of high interest due to their versatile applications as electrode material for metal-ion batteries. In particular, V3O7·H2O can insert different ions such as Li+, Na+, K+, Mg2+ and Zn2+. In that case, well resolved crystal structure data, such as crystal unit-cell parameters and atom positions, are needed in order to determine the structural information of the inserted ions in the V3O7·H2O structure. In this work, fundamental crystallographic parameters, i.e. atomic displacement parameters, are determined for the atoms in the V3O7·H2O structure. Furthermore, vanadium ions were substituted by molybdenum in the V3O7·H2O structure [(V2.85Mo0.15)O7·H2O] and the crystallographic positions of the molybdenum ions and their oxidation state are elucidated.

1. Introduction

Vanadate-based compounds are under investigation as potential next-generation electrode materials due to their ability to electrochemically insert various ions (Zhu et al., 2014[Zhu, K., Yan, X., Zhang, Y., Wang, Y., Su, A., Bie, X., Zhang, D., Du, F., Wang, C., Chen, G. & Wei, Y. (2014). ChemPlusChem, 79, 447-453.]; Huang et al., 2015[Huang, X., Rui, X., Hng, H. H. & Yan, Q. (2015). Part. Part. Syst. Charact. 32, 276-294.]; Moretti & Passerini, 2016[Moretti, A. & Passerini, S. (2016). Adv. Energy Mater. 6, 1600868.]; Yan et al., 2016[Yan, Y., Li, B., Guo, W., Pang, H. & Xue, H. (2016). J. Power Sources, 329, 148-169.]; Olszewski et al., 2018[Olszewski, W., Isturiz, I., Marini, C., Avila, M., Okubo, M., Li, H., Zhou, H., Mizokawa, T., Saini, N. L. & Simonelli, L. (2018). Phys. Chem. Chem. Phys. 20, 15288-15292.]). Therefore, interest arose from the battery materials community for the compound V3O7·H2O (also written as H2V3O8, HVO), which can intercalate Li+, Na+, K+, Mg2+ or Zn2+ (Simões et al., 2014[Simões, M., Mettan, Y., Pokrant, S. & Weidenkaff, A. (2014). J. Phys. Chem. C, 118, 14169-14176.]; Zhu et al., 2014[Zhu, K., Yan, X., Zhang, Y., Wang, Y., Su, A., Bie, X., Zhang, D., Du, F., Wang, C., Chen, G. & Wei, Y. (2014). ChemPlusChem, 79, 447-453.]; Wang et al., 2016[Wang, D., Wei, Q., Sheng, J., Hu, P., Yan, M., Sun, R., Xu, X., An, Q. & Mai, L. (2016). Phys. Chem. Chem. Phys. 18, 12074-12079.]; He et al., 2017[He, P., Quan, Y., Xu, X., Yan, M., Yang, W., An, Q., He, L. & Mai, L. (2017). Small, 13, 1702551.]; Tang et al., 2017[Tang, H., Xu, N., Pei, C., Xiong, F., Tan, S., Luo, W., An, Q. & Mai, L. (2017). ACS Appl. Mater. Interfaces, 9, 28667-28673.]; Rastgoo-Deylami et al., 2019[Rastgoo-Deylami, M., Heo, J. W. & Hong, S. T. (2019). ChemistrySelect, 4, 11711-11717.]). This insertion behaviour is of high interest since there are only a few compounds that have this ability. Furthermore, the stability during electrochemical cycling is better for V3O7·H2O than for related materials, e.g. V2O5, due to hydrogen bonds in the structure (Gao et al., 2009[Gao, S., Chen, Z., Wei, M., Wei, K. & Zhou, H. (2009). Electrochim. Acta, 54, 1115-1118.]). Fig. 1[link] shows the crystallographic structure of V3O7·H2O determined by Oka et al. from powder X-ray diffraction (Oka et al., 1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]). The structure provides three symmetry-independent vanadium positions highlighted as coloured spheres [Fig. 1[link](a)], in turquoise (V1) with an oxidation state of +5, orange (V2) with an oxidation state of +4 and green (V3) with an oxidation state of +5, and with two different oxygen coordination chemistries [Fig. 1[link](b)].

[Figure 1]
Figure 1
Crystal structure of V3O7·H2O based on data from Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]) viewed along the c axis, space group Pnam: (a) in ball-and-stick form and (b) polyhedron style. The colours turquoise (V1), orange (V2) and green (V3) represent the three different vanadium positions in the structure. Red spheres represent oxygen atoms. The grey spheres represent the oxygen of the water molecule in the structure.

So far, it has been suggested that one position (V3) has a square pyramidal coordination geometry, while two positions have octahedral ones (V1 and V2) (Oka et al., 1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]). However, the actual coordination geometries are still up for discussion. Considering bond valence sums, square pyramidal coordination rather than octahedral is present for the V2 position. This has also been stated by Rastgoo-Deylami et al. (Rastgoo-Deylami et al., 2018[Rastgoo-Deylami, M., Chae, M. S. & Hong, S.-T. (2018). Chem. Mater. 30, 7464-7472.]). The possible position of the hydrogen atoms was assigned to an oxygen atom (O6) bonded to the V2 vanadium centre (Mettan et al., 2015[Mettan, Y., Caputo, R. & Chatterji, T. (2015). RSC Adv. 5, 106543-106550.]). This leads to the hydrogen-bonding scheme mentioned above, which interconnects the vanadium oxide layers. Very recently, a crystal data set for V3O7·H2O has been reported, including experimentally determined hydrogen-atom positions. However, it included fixed atomic displacement parameters for the various atom positions (Kuhn et al., 2022[Kuhn, A., Pérez-Flores, J. C., Prado-Gonjal, J., Morán, E., Hoelzel, M., Díez-Gómez, V., Sobrados, I., Sanz, J. & García-Alvarado, F. (2022). Chem. Mater. 34, 694-705.]). As shown in Fig. 1[link](b), an open sheet-like topology is present, where the insertion of different ions can occur between the V-site layers (Simões et al., 2014[Simões, M., Mettan, Y., Pokrant, S. & Weidenkaff, A. (2014). J. Phys. Chem. C, 118, 14169-14176.]; Zhu et al., 2014[Zhu, K., Yan, X., Zhang, Y., Wang, Y., Su, A., Bie, X., Zhang, D., Du, F., Wang, C., Chen, G. & Wei, Y. (2014). ChemPlusChem, 79, 447-453.]; Wang et al., 2016[Wang, D., Wei, Q., Sheng, J., Hu, P., Yan, M., Sun, R., Xu, X., An, Q. & Mai, L. (2016). Phys. Chem. Chem. Phys. 18, 12074-12079.]; He et al., 2017[He, P., Quan, Y., Xu, X., Yan, M., Yang, W., An, Q., He, L. & Mai, L. (2017). Small, 13, 1702551.]; Tang et al., 2017[Tang, H., Xu, N., Pei, C., Xiong, F., Tan, S., Luo, W., An, Q. & Mai, L. (2017). ACS Appl. Mater. Interfaces, 9, 28667-28673.]; Rastgoo-Deylami et al., 2019[Rastgoo-Deylami, M., Heo, J. W. & Hong, S. T. (2019). ChemistrySelect, 4, 11711-11717.]). However, possible positions in the HVO structure have been determined experimentally or calculated (by DFT) so far for Li+, Mg2+ and Zn2+ (Kuhn et al., 2022[Kuhn, A., Pérez-Flores, J. C., Prado-Gonjal, J., Morán, E., Hoelzel, M., Díez-Gómez, V., Sobrados, I., Sanz, J. & García-Alvarado, F. (2022). Chem. Mater. 34, 694-705.]; Kundu et al., 2018[Kundu, D., Hosseini Vajargah, S., Wan, L., Adams, B., Prendergast, D. & Nazar, L. F. (2018). Energ. Environ. Sci. 11, 881-892.]; Rastgoo-Deylami et al., 2018[Rastgoo-Deylami, M., Chae, M. S. & Hong, S.-T. (2018). Chem. Mater. 30, 7464-7472.]). Li+ ions have a sixfold coordination, while Mg2+ ions seem to have square pyramidal coordination in the HVO structure, and tetrahedral and/or octahedral coordination was attributed to Zn2+ ions (Kuhn et al., 2022[Kuhn, A., Pérez-Flores, J. C., Prado-Gonjal, J., Morán, E., Hoelzel, M., Díez-Gómez, V., Sobrados, I., Sanz, J. & García-Alvarado, F. (2022). Chem. Mater. 34, 694-705.]; Rastgoo-Deylami et al., 2018[Rastgoo-Deylami, M., Chae, M. S. & Hong, S.-T. (2018). Chem. Mater. 30, 7464-7472.]; Kundu et al., 2018[Kundu, D., Hosseini Vajargah, S., Wan, L., Adams, B., Prendergast, D. & Nazar, L. F. (2018). Energ. Environ. Sci. 11, 881-892.]). These differences in coordination geometries hint at different insertion positions between the layers in the V3O7·H2O structure. Still, some questions remain unanswered. For example, as mentioned before, the position of the hydrogen atom in the structure is still under discussion. Furthermore, recently synthesized Mo-substituted HVO has been described, but it remains unclear whether Mo substitutes V sites randomly or whether specific unit-cell positions are favoured (Söllinger et al., 2021[Gao, S., Chen, Z., Wei, M., Wei, K. & Zhou, H. (2009). Electrochim. Acta, 54, 1115-1118.]). With this work, we are able to confirm the proposed position of the hydrogen atom in the V3O7·H2O structure from previous works and determine a complete crystallographic data set by correlating two diffraction data sets obtained by powder neutron and powder X-ray diffraction. In addition, the importance of atomic displacement parameters obtained for each atom on the different crystallographic positions is discussed in comparison to previously reported ones. Based on these results, we determined the Mo position in the V3O7·H2O structure for substituted materials and discussed the findings taking into account the oxidation state and the coordination chemistry of Mo.

2. Experimental

2.1. Hydro­thermal synthesis of V3O7·H2O and (V1–xMox)3O7·H2O

Materials. Vanadium(V) oxide (V2O5, 99.2%), anhydrous oxalic acid (H2C2O4, 98%) and L(+)-ascorbic acid (99+ %) were purchased from Alfa Aesar. Ammonium molybdate tetrahydrate [(NH4)6Mo7O24·4H2O, 99.98%] was purchased from Sigma Aldrich. All chemical reagents were used as received.

Compounds were synthesized according Söllinger et al. (2021[Söllinger, D., Karl, M., Redhammer, G. J., Schoiber, J., Werner, V., Zickler, G. A. & Pokrant, S. (2021). ChemSusChem, 14, 1112-1121.]).

V3O7·H2O. H2C2O4 (10.4 g) was dissolved in deionized water (100 ml) at room temperature. After complete dissolution of the oxalic acid, V2O5 (5 g) was added, followed by stirring for 3 h and further stirring at 80 °C for 5 h. The obtained solution was transferred into an evaporating dish and dried at 100 °C for 10 h followed by a calcination step at 400 °C for 10 h. The obtained as-synthesized V2O5 (VO) (2 g) was added to an aqueous ascorbic acid solution (20 ml, 0.025 M). The obtained suspension was stirred under reflux at 110 °C for 16 h in a round bottom flask. After this step, a hydro­thermal process was initiated. The whole suspension was transferred into an autoclave with a 100 ml polytetra­fluoro­ethyl­ene inlet. Additionally, deionized water (30 ml) was added to the suspension and brought to reaction at 220 °C for 6 h. The obtained precipitate was collected, washed with deionized water and iso­propanol, and dried at 80 °C for 3 h. Pristine V3O7·H2O without further modifications is denoted HVO in the following.

(V2.85Mo0.15)O7·H2O. Mo-substituted V3O7·H2O, was synthesized similar to HVO. To incorporate Mo6+ into VO, the soft chemistry process leading to VO was modified by adding an aqueous solution containing the Mo6+ ion. Therefore, (NH4)6Mo7O24·4H2O (1.77 g) was dissolved in deionized water (100 ml) to obtain an aqueous (0.1 M) Mo stock solution. To obtain 5 at% Mo in the V3O7·H2O structure, 4.75 g instead of 5 g V2O5 with Mo stock solution (27.5 ml, 0.1 M) were added to the oxalic acid solution and 9.88 g oxalic acid instead of 10.4 g were introduced to the precursor solution, respectively. The obtained compound with the formula (V0.95Mo0.05)3O7·H2O is labelled as HVO-Mo-5.

2.2. X-ray and neutron diffraction

2.2.1. X-ray diffraction

The crystalline phases of the synthesized powders were studied by powder X-ray diffraction (PXRD) on a Bruker D8 Advance diffractometer with a goniometer radius of 280 mm with a fast-solid state LynxEye detector and an automatic sample changer. The measurements were carried out with Cu Kα1,2 radiation in the 2θ range from 5° to 95° with a step size of 0.015°. All samples were prepared on zero-background single-crystal silicon sample holders. Full pattern refinement (Rietveld method) was performed with FullProf (Rodríguez-Carvajal, 1993[Rodríguez-Carvajal, J. (1993). Phys. B, 192, 55-69.]). The three-dimensional visualization of the crystal structure was constructed via VESTA (Momma & Izumi, 2011[Momma, K. & Izumi, F. (2011). J. Appl. Cryst. 44, 1272-1276.]).

2.2.2. Neutron powder diffraction

The neutron diffraction (ND) experiment was performed at the Research Neutron Source Heinz Maier-Leibnitz (FRM II) Garching bei München, Germany, within the Rapid Access Program. Powder diffraction data were acquired in Debye–Scherrer geometry using the high-resolution powder diffractometer SPODI (Hoelzel et al., 2012[Hoelzel, M., Senyshyn, A., Juenke, N., Boysen, H., Schmahl, W. & Fuess, H. (2012). Nucl. Instrum. Methods Phys. Res. A, 667, 32-37.]). Monochromatic neutrons with wavelength λ = 1.5482 Å were chosen from 551 reflection of a vertically focused composite germanium monochromator. The sample was contained in a 14 mm-diameter vanadium container and measured with constant rotation at 300 K. The high-resolution neutron powder diffraction data were collected in a 2θ range of 3° ≤ 2θ ≤ 152°, with a resolution step of 0.05°.

2.2.3. Diffraction data evaluation

Data refinement was performed with the FullProf suite of programs. For V3O7·H2O: Final results were extracted from a joint refinement on neutron and X-ray diffraction data using equal weights. For (V2.85Mo0.15)O7·H2O: Mo-ion positions and the amount of occupancy were determined using XRD data only. The determined HVO structure from this work served as the starting structure model.

2.3. X-ray photoelectron spectroscopy analysis

The X-ray photoelectron spectroscopy (XPS) measurements were carried out at a PHI 5800 MultiTechnique ESCA System (Physical Electronics) using monochromated Al Kα (1486.6 eV) radiation (250 W, 15 kV). The measurements were performed with a detection angle of 45° and a pass energy of 29.35 eV at the analyser to obtain detailed spectra. The samples were neutralized with electrons from a flood gun (current 3 µA) to compensate for charging effects at the surface. For binding energy calibration, the C(1s) peak was set to 284.8 eV. The peak fit of the detail spectrum in the Mo(3d) region was done with CasaXPS (Walton et al., 2010[Walton, J., Wincott, P., Fairley, N. & Carrick, A. (2010). Peak Fitting with CasaXPS: A Casa Pocket Book. Knutsford: Accolyte Science.]), using a Shirley-type background and a peak doublet (Gaussian–Lorentzian peak shape, intensity ratio 3:2, spin-orbit splitting 3.1 eV).

3. Results and discussion

3.1. The V3O7·H2O structure

V3O7·H2O was first described by Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]). Rastgoo-Deylami et al. (2018[Rastgoo-Deylami, M., Chae, M. S. & Hong, S.-T. (2018). Chem. Mater. 30, 7464-7472.]) published a resolved HVO structure, again, but still without any hydrogen positions, which had been proposed earlier by Mettan et al. (2015[Mettan, Y., Caputo, R. & Chatterji, T. (2015). RSC Adv. 5, 106543-106550.]). Although Mettan et al. (2015[Mettan, Y., Caputo, R. & Chatterji, T. (2015). RSC Adv. 5, 106543-106550.]) performed structure refinements on neutron diffraction data acquired at 4 K and 300 K. Moreover, no crystallographic data set was published in that work. Recently, Kuhn et al. (2022[Kuhn, A., Pérez-Flores, J. C., Prado-Gonjal, J., Morán, E., Hoelzel, M., Díez-Gómez, V., Sobrados, I., Sanz, J. & García-Alvarado, F. (2022). Chem. Mater. 34, 694-705.]) evaluated lithium-inserted HVO and determined only one hydrogen position, bonded to the same oxygen stated by Mettan et al. (2015[Mettan, Y., Caputo, R. & Chatterji, T. (2015). RSC Adv. 5, 106543-106550.]). Hence, our article aims to re-evaluate the work by combining ND and XRD powder patterns and providing a crystallographic data file (see supporting information).

Fig. 2[link] shows the obtained pattern from ND Rietveld refinement yielding an RBragg value of 4.6% in space group setting Pnam, which was originally proposed by Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]). The refinement led to the structure presented in Fig. 3[link].

[Figure 2]
Figure 2
Neutron diffraction (ND) pattern of Rietveld refined V3O7·H2O. The black dots represent the experimentally obtained ND pattern. The red line with dots represents the calculated pattern. The blue line shows the difference between the experimental and calculated patterns. The green vertical lines represent the Bragg peak positions.
[Figure 3]
Figure 3
Crystal structure of V3O7·H2O with view along the c axis, space group Pnam, in ball and sticks form without (a) and with hydrogen position (b); data sets for (a) are from Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]); data sets for (b) are from the refined structure. The colours turquoise (V1), orange (V2) and green (V3) represent the three different vanadium positions in the structure. Red spheres represent oxygen atoms, while the grey sphere represents the oxygen atom of the crystal water molecule without hydrogen bonds. The pinkish spheres represent the hydrogen atoms of the water molecule in the structure.

Fig. 3(a) shows the HVO structure determined by Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]) without the hydrogen-atom position determined. In this case, we highlighted the oxygen of the crystal water molecule in grey to symbolise the missing H bonds. In comparison, the structure we obtained from Rietveld analysis from combined ND and XRD results is shown in Fig. 3(b). Note that we kept the stated coordination chemistry from Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]) with V1 (turquoise) and V2 (orange) in octahedral and V3 (green) in square pyramidal coordination.

The basic unit of the structure is a band of V1 and V2 octahedra. The V1 octahedra are connected to each other by sharing two edges forming a non-planar zigzag chain extending in c direction. The V2 octahedra share three edges with V1 octahedra and are laterally attached to this chain along the a direction. The V3 sites are described to have a square pyramidal oxygen coordination; by sharing two edges within the base-plane of the pyramid, they form a zigzag chain with the apex of the pyramids alternately pointing up and down, and which interconnects the V1–V2 band to a layer of V sites extending in the bc plane. The hydrogen atoms are bonded to the O6 oxygen atom of the V2 octahedron and interconnect these layers via hydrogen bonding to the O5 oxygen atom of a V2 octahedron of the next layer, stacked along the crystallographic a axis.

The positions of the three vanadium atoms in the present refinement of the HVO structure show less distorted octahedra than in the result from Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]), leading to a smaller variance of the different bond lengths (Table 1[link]). In particular, the square pyramidal coordination chemistry of the V3 position, hence, the bond lengths between V3 and surrounding oxygen atoms, changes substantially compared to the initially proposed structure of Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]). The bond length between V3 and O7 changes from 1.8975 (3) to 1.82644 (5) Å, which is the bond in the square plane of the pyramidal geometry, while the axial bond between V3 and O8 changes from 1.5824 (13) to 1.86009 (6) Å. Comparing the positions of V1 and V2, bond lengths between O1 and V1/V2 and O2 and V1 (Table 1[link]) show substantial changes, too. Additionally, the positions of the vanadium atoms V1 and V2 are less displaced from the centre of the coordination octahedron. Overall, the obtained bond lengths between the vanadium centres and the surrounding oxygens are typical for (hydrated) vanadates (Liu et al., 2010[Liu, X., Yin, X.-H. & Zhang, F. (2010). Z. Naturforsch. 65, 1451-1456.]; Wu et al., 2009[Wu, C., Dai, J., Zhang, X., Yang, J. & Xie, Y. (2009). J. Am. Chem. Soc. 131, 7218-7219.]). The distance between V2 and O1 is in general the longest compared to all other bonds in te structure. It changed from 2.4731 (9) Å [structure from Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.])] to 2.20580 (5) Å (this work). This distance is at the limit of attributing the two atoms to a bond or not. As seen in Fig. 3[link], we decided to create awareness of the potential chemical interaction between these sites and present it as a bond and the coordination of the V2 atom as a distorted octahedron.

Table 1
Bond lengths (in Å) between vanadium and oxygen atoms

    V3O7·H2O (V2.85Mo0.15)O7·H2O
Vx Oy Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]) Rastgoo-Deylami et al. (2018[Rastgoo-Deylami, M., Chae, M. S. & Hong, S.-T. (2018). Chem. Mater. 30, 7464-7472.]) This work This work
V1 O1 1.9470 (3) 1.9425 (16) 1.84584 (5) 1.82951 (10)
  O2 1.5900 (10) 1.581 (5) 1.81928 (5) 1.55687 (10)
  O3 ax 2.0718 (8) 2.067 (5) 2.02833 (5) 2.350014 (12)
  O4 ax 1.9101 (11) 1.897 (5) 1.95870 (5) 1.98430 (11)
V2 O1 ax 2.4731 (9) 2.460 (5) 2.20580 (5) 2.38819 (11)
  O3 1.9068 (3) 1.8904 (13) 1.85642 (5) 1.82911 (10)
  O4 2.0288 (11) 2.019 (5) 2.01154 (5) 1.96378 (9)
  O5 ax 1.6058 (10) 1.597 (5) 1.67657 (4) 1.62805 (10)
  O6 2.0557 (9) 2.044 (5) 2.01568 (5) 2.00464 (9)
V3 O4 1.8159 (9) 1.806 (4) 2.05096 (7) 1.97249 (11)
  O7 1.8975 (3) 1.8793 (14) 1.82644 (5) 1.83821 (10)
  O8 ax 1.5824 (13) 1.573 (5) 1.86009 (6) 1.53280 (10)
†Results obtained from available cif-file from the ICSD.
‡ax means the oxygen atom Oy is in axial coordination to the vanadium central atom.

Evaluating the position of the hydrogen atoms in the structure by residual electron/nuclear density analysis, we determined from experimental data the same crystallographic position for the hydrogen atom at the oxygen atom O6 as Mettan et al. (2015[Mettan, Y., Caputo, R. & Chatterji, T. (2015). RSC Adv. 5, 106543-106550.]). However, Mettan et al. (2015[Mettan, Y., Caputo, R. & Chatterji, T. (2015). RSC Adv. 5, 106543-106550.]) discussed a second position of the H atom attached to the same O6 atom, since a value of 0.6 was found for hydrogen occupancy, indicating that slightly more than a half of the position was occupied by hydrogen atoms. We obtained similar results. In the work of Mettan et al. (2015[Mettan, Y., Caputo, R. & Chatterji, T. (2015). RSC Adv. 5, 106543-106550.]), no values for atomic displacement parameters (U) were mentioned but set to be equal for all atoms. In this work, we determined isotropic displacement parameters (Uiso) by combining ND and XRD patterns during the Rietveld refinement for all independent atomic positions present. With that, we obtained varying Uiso values, listed in Table 2[link], for the vanadium, oxygen and hydrogen positions. By doing so, the occupancy value of the hydrogen atom changed to 1 after evaluating reasonable Uiso values (see Table 2[link]). In particular, oxygen atoms with multifold bond situations (three to four) have lower Uiso values than those with fewer bonds. These values are in good agreement with the condition that strongly bonded oxygen atoms, e.g. O1, O3 or O4, are locked into their particular crystallographic positions. In contrast, less connected oxygens, i.e. O2, O5 and O8, will vibrate more strongly around the given position. These results are in contrast to the previous results of Kuhn et al. (2022[Kuhn, A., Pérez-Flores, J. C., Prado-Gonjal, J., Morán, E., Hoelzel, M., Díez-Gómez, V., Sobrados, I., Sanz, J. & García-Alvarado, F. (2022). Chem. Mater. 34, 694-705.]), who kept Uiso values constant for each atom species. Hence, the obtained values for isotropic displacement parameters give new insight into the oxygen positions in the HVO structure and show the dependence on the multifold bond situation.

Table 2
Isotropic displacement parameters of hydrogen, vanadium and oxygen atoms in the V3O7·H2O structure

Atom No. of bonds Uiso value (Å2)
H1 1 0.09780
V1 6 0.02398
V2 6 0.01384
V3 5 0.01281
O1 4 0.00820
O2 1 0.01289
O3 3 0.01182
O4 3 0.00924
O5 1 0.01412
O6 3 0.01238
O7 3 0.0065
O8 1 0.01731

Interestingly, the H atom shows a very high Uiso value. Since Mettan et al. described that there are two possible positions for the hydrogen atom, as mentioned above, a high Uiso value might also indicate that the hydrogen atoms of the water molecule cannot be precisely located and shows some positional disorder, possibly induced by the hydrogen-bond configuration. Hence, while a full occupancy of the hydrogen atom at the determined position was obtained, an exact position cannot be stated because of the high isotropic atom displacement parameter.

3.2. The Mo-substituted V3O7·H2O structure

In order to carry out Rietveld analysis, an excellent starting crystal structure data set is necessary if structurally analogous, chemically modified compounds, or crystal structure changes during (electro)chemical processes are investigated. We demonstrate the necessity and potential of using the obtained HVO crystallographic data set in that context.

Fig. 4[link](a) shows the obtained refinement plot of (V2.85Mo0.15)O7·H2O using starting values from our HVO structure file. We obtained a fit with an RBragg of 7.07% and a reliable structure [Fig. 4[link](b)]. Unit-cell parameters of refined HVO and Mo-substituted HVO, as well as from literature, are summarized in Table 3[link]. The obtained parameters for HVO are between the previously reported ones; e.g. we obtained a value of 16.865 (5) Å for unit-cell parameter a, while Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]) and Rastgoo-Deylami et al. (2018[Rastgoo-Deylami, M., Chae, M. S. & Hong, S.-T. (2018). Chem. Mater. 30, 7464-7472.]) reported 16.930 and 16.844 Å, respectively. Comparing the unit-cell parameters of Mo-substituted HVO with unsubstituted HVO, an increase of the unit-cell parameters is observed. This increase corroborates the results of a recent study (Söllinger et al., 2021[Söllinger, D., Karl, M., Redhammer, G. J., Schoiber, J., Werner, V., Zickler, G. A. & Pokrant, S. (2021). ChemSusChem, 14, 1112-1121.]).

Table 3
Unit-cell parameters of (Mo-substituted) V3O7·H2O structures

  V3O7·H2O (V2.85Mo0.15)O7·H2O
Unit-cell parameter Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]) Rastgoo-Deylami et al. (2018[Rastgoo-Deylami, M., Chae, M. S. & Hong, S.-T. (2018). Chem. Mater. 30, 7464-7472.]) This work This work
a (Å) 16.92979 (20) 16.844 (1) 16.8650 (5) 16.8790 (12)
b (Å) 9.3589 (1) 9.309 (1) 9.3290 (3) 9.3325 (5)
c (Å) 3.64432 (4) 3.625 (1) 3.6332 (8) 3.64357 (19)
V3) 577.42 568.40 571.96 (3) 573.95 (6)
†Results obtained from available cif-file from the ICSD.
‡The values have been transposed from Pnma to Pnam for better comparison.
[Figure 4]
Figure 4
(a) XRD powder pattern of (V2.85Mo0.15)O7·H2O. The black dots represent the obtained XRD pattern. The red line with dots represents the calculated pattern. The blue line shows the difference between the experimental and calculated patterns. The green vertical lines represent the Bragg peak positions. (b) Refined Mo-substituted HVO structure (V2.85Mo0.15)O7·H2O. (c) HVO structure.

Our attempts to refine the Mo-substituted structure starting with values obtained from Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]) or Rastgoo-Deylami et al. (2018[Rastgoo-Deylami, M., Chae, M. S. & Hong, S.-T. (2018). Chem. Mater. 30, 7464-7472.]) failed as soon as atomic positions were refined. Specifically, the position O2 shifted towards the V1 position, resulting in a V1–O2 distance below 1 Å, which is unreasonable. This shift is observed in the refinement with the novel HVO data set, too. Still, a reasonable bond distance was obtained compared to refinements with starting data sets from Oka et al. (1990[Oka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372-377.]) or Rastgoo-Deylami et al. (2018[Rastgoo-Deylami, M., Chae, M. S. & Hong, S.-T. (2018). Chem. Mater. 30, 7464-7472.]) (see Table 1[link]). Hence, the interaction between these two positions became more robust with the substitution of Mo, leading to a shorter bond distance.

Additionally, with the previously published HVO structures from Oka et al. or Rastgoo-Deylami et al. (2018[Rastgoo-Deylami, M., Chae, M. S. & Hong, S.-T. (2018). Chem. Mater. 30, 7464-7472.]), no reliable model for the Mo occupation at any of the three possible sites could be determined. However, with our set of parameters, Mo ions were detected at the V1 and V2 sites. An all-over Mo-substitution of 5.2 at% was derived, resulting in the chemical composition of (V2.842Mo0.158)O7·H2O. This Mo atomic concentration is close to the target value of the synthesis strategy, that is 5 at% substitution of vanadium with molybdenum. Furthermore, the result indicates that the combined refinement of X-ray and neutron diffraction data yields reliable results. Around two-thirds of the Mo content are present at the V1 site, while the other third is located at the V2 site. No Mo-content could be detected at the V3 site. Furthermore, the hydrogen bond broke as the oxygen atom from the O5 position shifted away from the coordinated water molecule's hydrogen atoms.

Since mild reductive synthesis procedures were applied, we also checked for the possibility that, in addition to the reduction of V5+ to V4+ species, Mo6+ was reduced to Mo5+ or lower. Hence, the oxidation state of Mo in the HVO structure was determined by XPS analysis. The detail of the spectrum in the Mo(3d) region (Fig. 5[link]) shows a single peak doublet at 233.0/236.1 eV, which we assign to Mo6+ (Moulder et al., 1992[Moulder, J. F., Stickle, W. F., Sobol, P. E. & Bomben, K. D. (1992). Handbook of X-ray Photoelectron Spectroscopy. Eden Prairie (MN): Perkin Elmer Corp.]).

[Figure 5]
Figure 5
XPS spectrum of Mo-substituted HVO in the Mo(3d) domain.

Hence, no reduction of Mo6+ occurs during the synthesis. Mo6+ occupies preferentially the octahedrally coordinated V1 (V5+) and V2 (V4+) sites (Shannon, 1976[Shannon, R. D. (1976). Acta Cryst. A32, 751-767.]) instead of the pyramidally coordinated V3 (V5+) site. This preference for octahedral coordination cannot be explained based on simple steric arguments alone, since the ionic radius difference between V5+ and Mo6+ is similar for octahedral and pyramidal coordination (Table 4[link]). We believe that quantum chemical calculations are necessary to elucidate this preference.

Table 4
Ionic radii of selected vanadium and molybdenum ions

Species Radii in CN 5 (Å) Radii in CN 6 (Å)
V4+ n.r. 0.58
V5+ 0.46 0.54
Mo6+ 0.5 0.59
†CN means coordination number.
‡n.r. means not relevant due to the coordination situation of the ion in the HVO structure.

4. Conclusion

Phase pure V3O7·H2O and (V2.85Mo0.15)O7·H2O compounds were synthesized via a hydro­thermal synthesis route. The pristine vanadate compound was investigated by powder ND and XRD techniques. We resolved the hydrogen position in the crystal structure, leading to an parameter set which has the potential to be used for further Rietveld analysis of modified compounds. We demonstrated this by investigating (V2.85Mo0.15)O7·H2O by XRD. Our data set led to reasonable structural refinements. This allowed us to determine the Mo content in the structure and evaluate the preferred site occupancy of the substituting ion. Furthermore, we could evaluate the oxidation state of molybdenum, namely +6. As an outcome of this work, we provide a high-quality structure file that can be used to analyse further V3O7·H2O-related mechanisms or compounds, such as ion-insertion mechanisms or vanadium substitution.

Supporting information


Computing details top

For both structures, program(s) used to refine structure: FULLPROF.

(VO) top
Crystal data top
V3O7.H2Oc = 3.63523 (10) Å
Mr = 282.84V = 571.92 (3) Å3
Orthorhombic, PnamZ = 4
Hall symbol: -P 2c 2nDx = 3.285 Mg m3
a = 16.8658 (5) ÅConstant Wavelength Neutron Diffraction radiation
b = 9.3282 (3) ÅT = 298 K
Data collection top
SPODI
diffractometer
2θmin = 1.003°, 2θmax = 151.953°, 2θstep = 0.050°
Radiation source: nuclear reactor
Refinement top
Rp = 1.489Profile function: pseudo-voigt
Rwp = 1.86178 parameters
Rexp = 1.1430 restraints
RBragg = 4.645Background function: linear extrapolation
3020 data pointsPreferred orientation correction: none
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/Ueq
H10.243100.142390.04250.09780*
V10.038940.103390.250000.02398*
V20.853410.063930.250000.01384*
V30.469400.074100.250000.01281*
O10.961600.931070.250000.00820*
O20.094370.270700.250000.01289*
O30.133370.968740.250000.01182*
O40.939770.212660.250000.00924*
O50.782680.190220.250000.01412*
O60.780110.893260.250000.01238*
O70.520120.922660.250000.00650*
O80.361060.036790.250000.01731*
(VMoO) top
Crystal data top
V2.85Mo0.15O7.H2Oc = 3.64357 (19) Å
Mr = 289.58V = 573.95 (6) Å3
Orthorhombic, PnamZ = 4
Hall symbol: -P 2c 2nDx = 3.355 Mg m3
a = 16.8790 (12) Å'Cu Kα radiation
b = 9.3325 (5) ÅT = 298 K
Data collection top
Bruker D8 Advance
diffractometer
2θmin = 5.045°, 2θmax = 94.982°, 2θstep = 0.010°
Radiation source: rotating-anode X-ray tube
Refinement top
Rp = 6.7198784 data points
Rwp = 9.2458 parameters
Rexp = 2.8380 restraints
RBragg = 7.076
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/UeqOcc. (<1)
V10.033780.097370.250000.03189*0.89288
Mo10.033780.097370.250000.03189*0.10712
V20.852970.076810.250000.02196*0.95020
Mo20.852970.076810.250000.02196*0.04980
V30.462630.072880.250000.03893*
O10.956280.901960.250000.00694*
O20.091120.228040.250000.01108*
O30.141230.937220.250000.00654*
O40.937950.220530.250000.00349*
O50.765660.150950.250000.01289*
O60.782920.903350.250000.00775*
O70.538920.901000.250000.00001*
O80.376690.019810.250000.01553*
H10.260550.638270.001340.09569*
 

Acknowledgements

The authors acknowledge G. Tippelt for X-ray diffraction measurements.

References

First citationGao, S., Chen, Z., Wei, M., Wei, K. & Zhou, H. (2009). Electrochim. Acta, 54, 1115–1118.  Web of Science CrossRef CAS Google Scholar
First citationHe, P., Quan, Y., Xu, X., Yan, M., Yang, W., An, Q., He, L. & Mai, L. (2017). Small, 13, 1702551.  Web of Science CrossRef Google Scholar
First citationHoelzel, M., Senyshyn, A., Juenke, N., Boysen, H., Schmahl, W. & Fuess, H. (2012). Nucl. Instrum. Methods Phys. Res. A, 667, 32–37.  Web of Science CrossRef CAS Google Scholar
First citationHuang, X., Rui, X., Hng, H. H. & Yan, Q. (2015). Part. Part. Syst. Charact. 32, 276–294.  Web of Science CrossRef CAS Google Scholar
First citationKuhn, A., Pérez-Flores, J. C., Prado-Gonjal, J., Morán, E., Hoelzel, M., Díez-Gómez, V., Sobrados, I., Sanz, J. & García-Alvarado, F. (2022). Chem. Mater. 34, 694–705.  Web of Science CrossRef CAS Google Scholar
First citationKundu, D., Hosseini Vajargah, S., Wan, L., Adams, B., Prendergast, D. & Nazar, L. F. (2018). Energ. Environ. Sci. 11, 881–892.  Web of Science CrossRef CAS Google Scholar
First citationLiu, X., Yin, X.-H. & Zhang, F. (2010). Z. Naturforsch. 65, 1451–1456.  Web of Science CrossRef CAS Google Scholar
First citationMettan, Y., Caputo, R. & Chatterji, T. (2015). RSC Adv. 5, 106543–106550.  Web of Science CrossRef CAS Google Scholar
First citationMomma, K. & Izumi, F. (2011). J. Appl. Cryst. 44, 1272–1276.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationMoretti, A. & Passerini, S. (2016). Adv. Energy Mater. 6, 1600868.  Web of Science CrossRef Google Scholar
First citationMoulder, J. F., Stickle, W. F., Sobol, P. E. & Bomben, K. D. (1992). Handbook of X-ray Photoelectron Spectroscopy. Eden Prairie (MN): Perkin Elmer Corp.  Google Scholar
First citationOka, Y., Yao, T. & Yamamoto, N. (1990). J. Solid State Chem. 89, 372–377.  CrossRef ICSD CAS Web of Science Google Scholar
First citationOlszewski, W., Isturiz, I., Marini, C., Avila, M., Okubo, M., Li, H., Zhou, H., Mizokawa, T., Saini, N. L. & Simonelli, L. (2018). Phys. Chem. Chem. Phys. 20, 15288–15292.  Web of Science CrossRef CAS PubMed Google Scholar
First citationRastgoo-Deylami, M., Chae, M. S. & Hong, S.-T. (2018). Chem. Mater. 30, 7464–7472.  CAS Google Scholar
First citationRastgoo-Deylami, M., Heo, J. W. & Hong, S. T. (2019). ChemistrySelect, 4, 11711–11717.  CAS Google Scholar
First citationRodríguez-Carvajal, J. (1993). Phys. B, 192, 55–69.  Google Scholar
First citationShannon, R. D. (1976). Acta Cryst. A32, 751–767.  CrossRef CAS IUCr Journals Web of Science Google Scholar
First citationSimões, M., Mettan, Y., Pokrant, S. & Weidenkaff, A. (2014). J. Phys. Chem. C, 118, 14169–14176.  Google Scholar
First citationSöllinger, D., Karl, M., Redhammer, G. J., Schoiber, J., Werner, V., Zickler, G. A. & Pokrant, S. (2021). ChemSusChem, 14, 1112–1121.  Web of Science PubMed Google Scholar
First citationTang, H., Xu, N., Pei, C., Xiong, F., Tan, S., Luo, W., An, Q. & Mai, L. (2017). ACS Appl. Mater. Interfaces, 9, 28667–28673.  Web of Science CrossRef CAS PubMed Google Scholar
First citationWalton, J., Wincott, P., Fairley, N. & Carrick, A. (2010). Peak Fitting with CasaXPS: A Casa Pocket Book. Knutsford: Accolyte Science.  Google Scholar
First citationWang, D., Wei, Q., Sheng, J., Hu, P., Yan, M., Sun, R., Xu, X., An, Q. & Mai, L. (2016). Phys. Chem. Chem. Phys. 18, 12074–12079.  Web of Science CrossRef CAS PubMed Google Scholar
First citationWu, C., Dai, J., Zhang, X., Yang, J. & Xie, Y. (2009). J. Am. Chem. Soc. 131, 7218–7219.  Web of Science CrossRef ICSD PubMed CAS Google Scholar
First citationYan, Y., Li, B., Guo, W., Pang, H. & Xue, H. (2016). J. Power Sources, 329, 148–169.  Web of Science CrossRef CAS Google Scholar
First citationZhu, K., Yan, X., Zhang, Y., Wang, Y., Su, A., Bie, X., Zhang, D., Du, F., Wang, C., Chen, G. & Wei, Y. (2014). ChemPlusChem, 79, 447–453.  Web of Science CrossRef CAS PubMed Google Scholar

This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.

Journal logoSTRUCTURAL SCIENCE
CRYSTAL ENGINEERING
MATERIALS
ISSN: 2052-5206
Follow Acta Cryst. B
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds