On the (non-)existence of Tutton salts with formula types [M(H₂O)₆](ClO₄)₂(H₂O)₂ and Na₂M(SO₄)₂(H₂O)₆ (M is a first-row transition metal)

Two structures of the Tutton salt family, originally reported with composition [Cu(H₂O)₆](ClO₄)₂(H₂O)₂ and Na₂Cu(SO₄)₂(H₂O)₆, have been redetermined based on the original intensity data. With respect to the original [Cu(H₂O)₆](ClO₄)₂(H₂O)₂ model, the perchlorate anion and the noncoordinating water mol­ecule are replaced by a sulfate anion and an ammonium cation. With respect to the original Na₂Cu(SO₄)₂(H₂O)₆ model, the sodium site is replaced by a mixed-occupancy potassium/ammonium site. The resulting revised formulae are (NH₄)₂Cu(SO₄)₂(H₂O)₆ [di­ammonium hexa­aqua­copper(II) di­sulfate] and [(NH₄)_1.176K_0.824]Cu(SO₄)₂(H₂O)₆, respectively. In both cases, the redetermination led to chemically more sensible structure models, accompanied by lower reliability factors. Three other reported structures with formula types [M(H₂O)₆](ClO₄)₂(H₂O)₂ or Na₂M(SO₄)₂(H₂O)₆ (M is a first row transition metal) have also been re-examined. From crystal-chemical considerations, their existence is likewise questioned. It is shown that the deposition of structure factors is beneficial for detailed re-examinations of problematic structure models.

S1. Introduction

Double salts of the general formula M^I₂M^II(XO₄)₂(H₂O)₆, where M^I is a large monovalent cation (K, NH₄, Rb, Cs, Tl), M^II is Mg or a divalent transition-metal (V, Mn, Fe, Co, Ni, Cu, Zn, Cd, Ru) and X is hexavalent S, Se, or Cr, have been the subject of crystallographic examinations for more than 120 years. Prominent examples of these salts are Mohr's salt, (NH₄)₂Fe(SO₄)₂(H₂O)₆, which is one of the few stable ferrous inorganic compounds, or the mineral picromerite (also named schönite in the older literature) with composition K₂Mg(SO₄)₂(H₂O)₆. Numerous other synthetic salts of this formula type were systematically investigated by A. E. H. Tutton at the end of the 19th century. In honour of this pioneering work, compounds of the strictly isotypic M^I₂M^II(XO₄)₂(H₂O)₆ series are also referred to as Tutton salts for which numerous structure refinements are compiled in the Inorganic Crystal Structure Database (ICSD, 2013). The first crystal structure determination of a Tutton salt was carried out on (NH₄)₂Mg(SO₄)₂(H₂O)₆ (Hofmann, 1931) in the nonstandard setting P2₁/a of space group No. 14. Most of the subsequent structure refinements of other members of this series were performed by using this setting, but also numerous structures were refined in the standard setting P2₁/c. In general, the Tutton salt unit cell contains two formula units and is made up of one M^II site situated on an inversion centre and surrounded by six water molecules in a slightly distorted o­cta­hedral environment, one XO₄ tetra­hedron and one distorted M^IO₈ polyhedron. The [M^II(H₂O)₆] units are directly linked to four M^IO₈ units via their equatorial O atoms, and to eight XO₄ tetra­hedra through medium-strength O—H···O hydrogen bonds, thus generating a three-dimensional network structure (Fig. 1). A detailed discussion of the crystal chemistry of Tutton salts in general and of K₂M^II(XO₄)₂(H₂O)₆ members in particular was given recently by Bosi et al. (2009). The copper- and ammonium-containing Tutton salts show some peculiar structural features. Because of the Jahn–Teller effect of the Cu^II cation (Falvello, 1997), the [CuO₆] o­cta­hedron is considerably distorted, typically with longer axial Cu—O distances, and additional medium-strength to weak N—H···O hydrogen-bonding inter­actions for the tetra­hedral NH₄⁺ cation.

S2. Experimental

S2.1. Synthesis and crystallization

Crystal growth was described in the original publications (Wu et al., 2008; Li et al., 2004).

S2.2. Refinement

Crystal data, data collection and structure refinement details are summarized in Table 1. For the rerefinement of [Cu(H₂O)₆](ClO₄)₂(H₂O)₂, the coordinates and atom labels of the original model (Li et al., 2004) were used, with the Cl site replaced by S, the O atom of the noncoordinating water molecule replaced with N, and the H atoms attached to this site removed. The remaining H-atom positions of the coordinating water molecules of the [Cu(H₂O)₆] unit were retained. From difference Fourier maps, four H-atom sites surrounding the N atom in a tetra­hedral manner were clearly discernible (Fig. 2). They were included in the refinement with N—H distance restraints of 0.900 (1) Å. The original model can be rejected as the correct model at the 0.5% significance level according to Hamilton s R factor ratio test (Hamilton, 1965).

For the rerefinement of Na₂Cu(SO₄)₂(H₂O)₆, the coordinates and atom labels of the original model (Wu et al., 2008) were used, with the Na⁺ site replaced by the N atom of an ammonium group. The refinement converged with considerably higher reliability factors, and the N atom showed physically meaningless displacement parameters. Free refinement of the occupancy of the N site resulted in a site-occupation factor (s.o.f.) of 1.8, pointing to an element or a mixed-occupied site with a higher number of total electrons. From the chemically reasonable elements only potassium was considered, and free refinement of the occupancy of the K⁺ site led to an s.o.f. of 0.61. Based on these results, a statistical occupancy of ammonium and potassium on this site was finally considered, with the position and displacement parameters constrained to be equal, resulting in an occupancy ratio of NH₄:K = 1.176 (5):0.824 (5), also resulting in reasonable anisotropic displacement parameters of this site. Examination of the difference Fourier maps revealed two of the H positions describable to an ammonium site, but a sensible refinement including all H atoms was not possible for the overall NH₄ tetra­hedron. Therefore these protons were excluded from the final refinement. Again, Hamilton s R factor ratio test indicated at the 0.5% significance level that the original model is incorrect. The [CuO₆] o­cta­hedron with Cu—O distances of 1.9616 (18), 2.130 (2) and 2.178 (2) Å shows a less-distinct distortion as expected from the crystal structures of the pure potassium compound (1.944, 2.064 and 2.275 Å; Bosi et al., 2009) and the pure ammonium compound (1.961, 2.096 and 2.219 Å; Montgomery & Lingafelter, 1966). This tendency might be related to the partial replacement of NH₄⁺ by K⁺ (or vice versa), because the Jahn–Teller elongation shifts from one set of Cu—O bonds in K₂Cu(SO₄)₂(H₂O)₆ to another set, perpendicular to the first, in (NH₄)₂Cu(SO₄)₂(H₂O)₆. Hence the structural adjustment of the Cu—O bond lengths in [(NH₄)_1.176K_0.824]Cu(SO₄)₂(H₂O)₆ points to dynamic Jahn–Teller disorder (Falvello, 1997) of the Cu^II ion. In fact, the Jahn–Teller radius was calculated as 0.32 Å, which is in perfect agreement with the mean of 0.32 (3) Å for 12 [CuN₆] chromophores in varied crystalline environments (Falvello, 1997). For the final revised model, the metal site was therefore assumed to be Cu with full occupancy.

S3. Results and discussion

During a close examination of the structure family of Tutton salts some exceptional members came to attention, because they differ substanti­ally from the characteristic chemistry. This includes the perchlorate compounds [Cu(H₂O)₆](ClO₄)₂(H₂O)₂ (Li et al., 2004) and [Ni(H₂O)₆](ClO₄)₂(H₂O)₂ (Staples et al., 1998), as well as the sodium-containing compounds Na₂Ni(SO₄)₂(H₂O)₆ (Zhao et al., 2006), Na₂Co(SO₄)₂(H₂O)₆ (Lu et al., 2009) and Na₂Cu(SO₄)₂(H₂O)₆ (Wu et al., 2008). Table 2 compiles lattice parameters and other structural data of these compounds. From a crystal–chemical point of view, both types of compounds appear suspicious.

Examination of the bond lengths in the structures of the two [M(H₂O)₆](ClO₄)₂(H₂O)₂ perchlorate compounds revealed unexpected geometries such as unusually long Cl—O bond lengths in the ClO₄ tetra­hedron. The average Cl—O bond length is 1.475 Å for the Cu-containing compound and 1.479 Å for the Ni-containing compound. Since the expected mean Cl—O bond length in perchlorates is only 1.44 Å (for typical Cl—O bond lengths in perchlorates, see, for example: Marabello et al., 2004; Rahman et al., 2003; Zhang et al., 2006), the observed X—O bond lengths match much better with those expected for a sulfate tetra­hedron, the average of which is 1.473 Å (Hawthorne et al. 2000, and references therein). The results of a bond valence analysis for the Cl atom in the two ClO₄ tetra­hedra, with an expected value of 7.0 v.u. The calculated values of 6.12 v.u. for the Cu- and 5.98 v.u. for the Ni-containing structure, clearly supports this assumption. In the structures of the three sodium-containing compounds Na₂M(SO₄)₂(H₂O)₆ the M^I position that is usually occupied by a comparatively large cation with a radius greater than 1.5 Å is then occupied by Na⁺ with its much smaller ionic radius that can range from 1.0 Å for coordination number 4 to a maximum of 1.4 Å for coordination number 12 (Shannon, 1976). Hence the Na—O distances in all Na₂M(SO₄)₂(H₂O)₆ structures are much longer than usually observed. Consequently, in the original structure models, the respective Na⁺ site is surrounded by eight O atoms with Na—O bond lengths in the range 2.76–3.34 Å. As has been shown for a similar case where a water molecule in the correct model was confused with a Na⁺ cation in the erroneous model (Brown, 2012), calculation of bond valence sums (Brown, 2002) is a convenient method for checking the validity of a structure model. In comparison with the expected value of 1 valence unit (v.u.) for Na⁺, the corresponding bond valence sums, calculated with the parameter provided by Brese & O'Keeffe (1991), are only 0.29 v.u., 0.34 v.u. and 0.32 v.u. for the Ni, Co and Cu compounds, respectively. The very long Na—O distances and the very low bond valence sums indicate that the reported models are chemically not meaningful. The environment of the Na⁺ site is one that would rather be expected for a much larger cation like K⁺, or most probably corresponds to an ammonium cation that forms medium-strength to weak hydrogen bonds to the surrounding O atoms.

Based on these findings, redeterminations of the crystal structures of all suspicious compounds seemed nercessary. For that purpose, at first, single-crystal growth of the five compounds in question was attempted. In one case experimental synthesis conditions were not given at all (Staples et al., 1998). None of the other cases led to crystal growth of the described phases using the described syntheses. Nevertheless, in the case of [Cu(H₂O)₆](ClO₄)₂(H₂O)₂ and of Na₂Cu(SO₄)₂(H₂O)₆ re-examinations of the structural models were possible, because structure factors from the original refinements had been deposited. The current rerefinements revealed that the originally given compositions [Cu(H₂O)₆](ClO₄)₂(H₂O)₂ and Na₂Cu(SO₄)₂(H₂O)₆ are in fact incorrect and the published structure models implausible.

The originally reported composition [Cu(H₂O)₆](ClO₄)₂(H₂O)₂ must be reformulated as (NH₄)₂Cu(SO₄)₂(H₂O)₆, the crystal structure of which had already been refined more than 50 years ago from three-dimensional X-ray single-crystal intensity data by Montgomery & Lingafelter (1966). It has been redetermined several times afterwards from X-ray or neutron data to study details of hydrogen bonding, or the temperature- and pressure-dependence of Jahn–Teller distortions of the [Cu(H₂O)₆] o­cta­hedron (Brown & Chidambaram, 1969; Alcock et al., 1984; Maslen et al., 1988; Simmons et al. 1993; Figgis et al., 2000). The bond lengths and angles in the structure of the revised [Cu(H₂O)₆](ClO₄)₂(H₂O)₂ model are virtually in the same ranges as in the previously refined (NH₄)₂Cu(SO₄)₂(H₂O)₆ structures. The S—O bond lengths (Table 3) are typical (Hawthorne et al., 2000), resulting in a reasonable bond valence sum of 6.02 v.u. for the S atom (expected 6.0 v.u.). Likewise, the O-atom environment of the ammonium cation is typical (García-Rodríguez et al., 2000), with six acceptor O atoms located at distances from 2.863 (2) to 3.174 (3) Å (Table 4). There are two additional O atoms at distances of 3.175 (2) and 3.276 (3) Å, but their N—H···O angles of 98 (2) and 118 (2)° do not support their contribution in hydrogen bonding.

Based on the current rerefinement, the Na site in the original Na₂Cu(SO₄)₂(H₂O)₆ model needs to be replaced by a statistically occupied NH₄⁺/K⁺ site, leading to a refined composition of [(NH₄)_1.176K_0.824]Cu(SO₄)₂(H₂O)₆. The formation of mixed crystals in NH₄⁺/K⁺ systems has already been observed by Tutton (Morrow, 1969). Because the radii of the cations NH₄⁺ and K⁺ are similar [for a coordination number of eight they amount to 1.66 Å (Khan & Baur, 1972) and 1.51 Å (Shannon, 1976), respectively], these ions can replace each other during crystallization. Numerous crystallographic studies have been devoted to such NH₄⁺/K⁺ solid solutions, e.g. for miscibilities of potassium halide and ammonium halide systems (Havighurst et al., 1925), or structure refinements of phases in the series [(NH₄)_1-xK_x]NO₃ (Coates & Crewe, 1961; Holden & Dickinson, 1975). In accordance with the partial replacement of ammonium with the slightly smaller potassium ion in the structure of [(NH₄)_1.176K_0.824]Cu(SO₄)₂(H₂O)₆, the unit-cell volume is also smaller, and the corresponding eight M^I—O distances (Table 5) are shorter [2.825 (2)–3.188 (3) Å] than in (NH₄)₂Cu(SO₄)₂(H₂O)₆.

In the face of the current redeterminations, the question arises what cam be learned with respect to structure refinements and how can erroneous structure models be avoided in general. Since all original structure models were reported to have converged with acceptable reliability factors (Table 2), these factors alone are not a definitive measure for the correctness of a structure model. Hence the application of other analytical methods, e.g. chemical analysis or spectroscopic measurements, is certainly desirable and should accompany the diffraction experiment. For these five reported structures, in four cases no chemical analysis was reported and only in one case a decomposition product was analysed. In many ambiguous cases, a simple comparison of the results of complementary methods gives very useful hints to a wrong composition or a wrong structure model. In addition, it is indispensable to use one's (crystal–)chemical intuition for inter­pretation of a structure model, e.g. in matters of bond lengths and coordination numbers. In the present cases, the similar scattering factors for Na, N and O, and for Cl and S, respectively, resulted in the confusion of the elements in the five structures. The mistaken identity of some atoms in these structures apparently had no dramatic effect on the reliability factors. Although the checkCIF routine (Spek, 2009) is of invaluable help reporting various errors and pitfalls in a structure model, in the cases of [Cu(H₂O)₆](ClO₄)₂(H₂O)₂ and Na₂Cu(SO₄)₂(H₂O)₆ no specific alerts pointing to a wrong composition or an unreasonable structure model were indicated. However, the erroneous models were clearly revealed by examination of bond lengths and substanti­ated by bond valence calculations and subsequent rerefinements. Unfortunately, at the moment only IUCr journals and very few other journals request structure factors to be deposited along with the crystallographic information file (CIF) for publication of a crystal structure. The presence of structure factors is a pre-condition for rerefinements based on the original intensity data and therefore is beneficial for a detailed (retroactive) examination of problematic structure models, as shown for the two examples of [Cu(H₂O)₆](ClO₄)₂(H₂O)₂ and Na₂Cu(SO₄)₂(H₂O)₆. Therefore it is desirable that the deposition of structure factors, or even original intensity (raw) data becomes mandatory in general, irrespective what scientific journal is chosen for publication of the results of a crystal structure analysis. Although for the other three reported Tutton salts with composition [Ni(H₂O)₆](ClO₄)₂(H₂O)₂ (Staples et al., 1998), Na₂Ni(SO₄)₂(H₂O)₆ (Zhao et al., 2006) and Na₂Co(SO₄)₂(H₂O)₆ (Lu et al., 2009), rerefinements based on original intensity data were not possible, the same crystal–chemically driven arguments as for the two test cases are valid. The too short Cl—O distances of the ClO₄ tetra­hedron, as well as the very unusual environment of the Na⁺ site on the M^I position resulting in very low bond valence sums, make all these structure models doubtful. Hence it is most likely that Tutton salts containing perchlorate anions as XO₄ anions or the comparatively small Na⁺ ion on the M^I position do not exist at all.

Related literature

For related literature, see: Alcock et al. (1984); Bosi et al. (2009); Brown (2002, 2012); Brown & Chidambaram (1969); Coates & Crewe (1961); Falvello (1997); Figgis et al. (2000); García-Rodríguez, Rute-Pérez, Piñero & González-Silgo (2000); Hamilton (1965); Havighurst et al. (1925); Hawthorne et al. (2000); Hofmann (1931); Holden & Dickinson (1975); ICSD (2013); Khan & Baur (1972); Li et al. (2004); Lu & Feng (2009); Marabello et al. (2004); Maslen et al. (1988); Montgomery & Lingafelter (1966); Morrow (1969); Rahman et al. (2003); Shannon (1976); Simmons et al. (1993); Spek (2009); Staples et al. (1998); Wu et al. (2008); Zhang et al. (2006); Zhao et al. (2006).

Computing details

Data collection: SMART (Bruker, 2002) for CI6335_redetermined; PROCESS (Rigaku, 1996) for FI2054_redetermined. Cell refinement: SMART (Bruker, 2002) for CI6335_redetermined; PROCESS (Rigaku, 1996) for FI2054_redetermined. Data reduction: SAINT (Bruker, 2002) for CI6335_redetermined; PROCESS (Rigaku, 1996) for FI2054_redetermined. For both compounds, program(s) used to solve structure: coordinates taken from the previous model; program(s) used to refine structure: SHELXL97 (Sheldrick, 2008). Molecular graphics: ATOMS for Windows (Dowty, 2006) and PLATON (Spek, 2009) for CI6335_redetermined; ATOMS for Windows (Dowty, 2006) for FI2054_redetermined. For both compounds, software used to prepare material for publication: publCIF (Westrip, 2010).

Figures

The crystal structure of the Tutton salt family M^I₂M^II(XO₄)₂(H₂O)₆ in a projection along [001] (setting in P2₁/c). The M^I position is displayed as a sphere, the [M^IIO₆] units as octahedra and the XO₄ units as tetrahedra.

Difference contour maps (PLATON; Spek, 2009) around the N atom in the revised [Cu(H₂O)₆](ClO₄)₂(H₂O)₂ model [correct composition (NH₄)₂Cu(SO₄)₂(H₂O)₆], revealing positive electron density (continuous lines) at distances of ca 0.8 Å that can be associated with four tetrahedrally arranged H atoms. Contours are shown at intervals of 0.1 e^- Å^-3 with cut-offs for minimum and maximum values of -0.2 and 0.5 e^- Å^-3, respectively. The planes in the upper and lower diagram are shifted by 0.9 Å in the z direction; the distance between two bars at the axes is 1.5 Å.

(CI6335_redetermined) Diammonium hexaaquacopper(II) disulfate

Crystal data

(NH4)2Cu(SO4)2(H2O)6	F(000) = 414
Mr = 399.84	Dx = 1.922 Mg m^−3
Monoclinic, P21/c	Mo Kα radiation, λ = 0.71073 Å
Hall symbol: -P 2ybc	Cell parameters from 3539 reflections
a = 6.2908 (3) Å	θ = 2.8–25.2°
b = 12.4043 (6) Å	µ = 1.95 mm^−1
c = 9.2181 (5) Å	T = 273 K
β = 106.146 (2)°	Prism, pale blue
V = 690.94 (6) Å^3	0.60 × 0.52 × 0.29 mm
Z = 2

Data collection

Bruker APEX area-detector diffractometer	1244 independent reflections
Radiation source: fine-focus sealed tube	1214 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.0000
ϕ and ω scans	θmax = 25.2°, θmin = 2.8°
Absorption correction: multi-scan (SADABS; Sheldrick, 2004)	h = −7→7
Tmin = 0.290, Tmax = 0.555	k = 0→14
1244 measured reflections	l = 0→11

Refinement

Refinement on F^2	Primary atom site location: structure-invariant direct methods
Least-squares matrix: full	Secondary atom site location: difference Fourier map
R[F^2 > 2σ(F^2)] = 0.024	Hydrogen site location: inferred from neighbouring sites
wR(F^2) = 0.070	H atoms treated by a mixture of independent and constrained refinement
S = 1.14	w = 1/[σ^2(Fo^2) + (0.0409P)^2 + 0.3339P] where P = (Fo^2 + 2Fc^2)/3
1244 reflections	(Δ/σ)max < 0.001
104 parameters	Δρmax = 0.33 e Å^−3
4 restraints	Δρmin = −0.49 e Å^−3

Special details

Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes.

Refinement. Refinement of F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > σ(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq
Cu1	0.0000	0.0000	0.0000	0.02060 (14)
O5	−0.2833 (2)	0.06496 (11)	0.00418 (16)	0.0272 (3)
H5	−0.3105	0.1270	−0.0255	0.041*
H5A	−0.3128	0.0625	0.0850	0.041*
O6	−0.0319 (3)	−0.10911 (12)	0.16483 (17)	0.0336 (4)
H6	−0.0069	−0.1702	0.1410	0.050*
H6A	0.0389	−0.0979	0.2519	0.050*
O7	0.1744 (3)	0.11558 (13)	0.17354 (17)	0.0364 (4)
H7	0.3034	0.0998	0.2116	0.055*
H7A	0.1137	0.1246	0.2395	0.055*
S1	0.74474 (8)	0.36089 (4)	0.90889 (5)	0.02449 (16)
O1	0.5996 (3)	0.26780 (13)	0.91394 (19)	0.0378 (4)
O2	0.6316 (2)	0.42946 (12)	0.77947 (16)	0.0309 (4)
O3	0.7802 (3)	0.42346 (15)	1.04748 (18)	0.0461 (5)
O4	0.9568 (3)	0.32162 (13)	0.88950 (19)	0.0358 (4)
N1	0.6412 (3)	0.65141 (16)	0.8664 (2)	0.0326 (4)
H1N	0.640 (5)	0.5830 (8)	0.834 (3)	0.053 (9)*
H2N	0.612 (5)	0.701 (2)	0.792 (3)	0.068 (10)*
H3N	0.545 (4)	0.665 (3)	0.920 (3)	0.070 (10)*
H4N	0.779 (2)	0.654 (3)	0.930 (3)	0.055 (9)*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
Cu1	0.0215 (2)	0.0211 (2)	0.0188 (2)	0.00154 (11)	0.00514 (15)	−0.00044 (11)
O5	0.0287 (7)	0.0266 (7)	0.0280 (7)	0.0041 (6)	0.0107 (6)	0.0029 (6)
O6	0.0404 (9)	0.0304 (8)	0.0280 (8)	0.0045 (7)	0.0061 (7)	−0.0024 (6)
O7	0.0295 (8)	0.0465 (9)	0.0326 (8)	0.0039 (7)	0.0076 (7)	−0.0023 (7)
S1	0.0271 (3)	0.0241 (3)	0.0217 (3)	0.00241 (19)	0.0057 (2)	0.00303 (18)
O1	0.0340 (9)	0.0303 (8)	0.0497 (10)	−0.0016 (7)	0.0125 (7)	0.0113 (7)
O2	0.0345 (8)	0.0318 (8)	0.0259 (8)	0.0050 (6)	0.0075 (6)	0.0085 (6)
O3	0.0604 (12)	0.0486 (10)	0.0254 (8)	0.0076 (9)	0.0054 (8)	−0.0059 (7)
O4	0.0279 (8)	0.0336 (8)	0.0464 (9)	0.0041 (6)	0.0110 (7)	0.0020 (7)
N1	0.0355 (11)	0.0320 (11)	0.0326 (11)	0.0016 (8)	0.0132 (9)	0.0003 (8)

Geometric parameters (Å, º)

Cu1—O5^i	1.9662 (14)	O7—H7	0.8135
Cu1—O5	1.9662 (14)	O7—H7A	0.8109
Cu1—O6	2.0857 (15)	S1—O3	1.4580 (17)
Cu1—O6^i	2.0857 (15)	S1—O4	1.4767 (16)
Cu1—O7^i	2.2005 (16)	S1—O2	1.4769 (14)
Cu1—O7	2.2005 (16)	S1—O1	1.4808 (16)
O5—H5	0.8188	N1—H1N	0.8999 (11)
O5—H5A	0.8167	N1—H2N	0.8999 (11)
O6—H6	0.8166	N1—H3N	0.8998 (11)
O6—H6A	0.8145	N1—H4N	0.8998 (11)
O5^i—Cu1—O5	180.00 (11)	O7^i—Cu1—O7	180.00 (9)
O5^i—Cu1—O6	91.13 (6)	O3—S1—O4	111.00 (11)
O5—Cu1—O6	88.87 (6)	O3—S1—O2	108.85 (10)
O5^i—Cu1—O6^i	88.87 (6)	O4—S1—O2	109.93 (9)
O5—Cu1—O6^i	91.13 (6)	O3—S1—O1	109.45 (11)
O6—Cu1—O6^i	180.00 (11)	O4—S1—O1	109.40 (9)
O5^i—Cu1—O7^i	89.43 (6)	O2—S1—O1	108.15 (9)
O5—Cu1—O7^i	90.57 (6)	H1N—N1—H2N	114 (3)
O6—Cu1—O7^i	88.70 (6)	H1N—N1—H3N	115 (3)
O6^i—Cu1—O7^i	91.30 (6)	H2N—N1—H3N	106 (3)
O5^i—Cu1—O7	90.57 (6)	H1N—N1—H4N	100 (3)
O5—Cu1—O7	89.43 (6)	H2N—N1—H4N	115 (3)
O6—Cu1—O7	91.30 (6)	H3N—N1—H4N	108 (3)
O6^i—Cu1—O7	88.70 (6)

Symmetry code: (i) −x, −y, −z.

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
O5—H5···O1^ii	0.82	1.87	2.688 (2)	174
O5—H5A···O2^iii	0.82	1.92	2.736 (2)	174
O6—H6···O4^iv	0.82	1.94	2.749 (2)	172
O6—H6A···O3^v	0.81	1.90	2.714 (2)	176
O7—H7···O2^vi	0.81	2.02	2.824 (2)	172
O7—H7A···O4^iii	0.81	2.02	2.821 (2)	168
N1—H1N···O2	0.90 (1)	1.97 (1)	2.863 (2)	175 (3)
N1—H2N···O1^vii	0.90 (1)	2.16 (2)	2.980 (3)	151 (3)
N1—H2N···O7^viii	0.90 (1)	2.62 (3)	3.174 (3)	121 (3)
N1—H4N···O4^ix	0.90 (1)	2.02 (1)	2.899 (3)	164 (3)
N1—H3N···O1^x	0.90 (1)	2.15 (1)	3.018 (3)	161 (3)
N1—H3N···O3^x	0.90 (1)	2.41 (2)	3.113 (3)	135 (3)

Symmetry codes: (ii) x−1, y, z−1; (iii) x−1, −y+1/2, z−1/2; (iv) −x+1, −y, −z+1; (v) −x+1, y−1/2, −z+3/2; (vi) x, −y+1/2, z−1/2; (vii) −x+1, y+1/2, −z+3/2; (viii) −x+1, −y+1, −z+1; (ix) −x+2, −y+1, −z+2; (x) −x+1, −y+1, −z+2.

(FI2054_redetermined) Di(ammonium/potassium) hexaaquacopper(II) disulfate

Crystal data

[(NH4)1.176K0.824]Cu(SO4)2(H2O)6	F(000) = 427
Mr = 417.20	Dx = 2.038 Mg m^−3
Monoclinic, P21/c	Mo Kα radiation, λ = 0.71073 Å
Hall symbol: -P 2ybc	Cell parameters from 289 reflections
a = 6.2345 (12) Å	θ = 2–25.1°
b = 12.333 (3) Å	µ = 2.23 mm^−1
c = 9.1822 (18) Å	T = 291 K
β = 105.56 (3)°	Prismatic, blue
V = 680.1 (2) Å^3	0.20 × 0.17 × 0.17 mm
Z = 2

Data collection

Rigaku R-AXIS IV image-plate diffractometer	1322 independent reflections
Radiation source: fine-focus sealed tube	1257 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.0000
Oscillation frames scans	θmax = 26.0°, θmin = 2.8°
Absorption correction: multi-scan (SADABS; Sheldrick, 2004)	h = −7→7
Tmin = 0.607, Tmax = 0.709	k = 0→15
1322 measured reflections	l = 0→11

Refinement

Refinement on F^2	Secondary atom site location: difference Fourier map
Least-squares matrix: full	Hydrogen site location: inferred from neighbouring sites
R[F^2 > 2σ(F^2)] = 0.030	H-atom parameters constrained
wR(F^2) = 0.079	w = 1/[σ^2(Fo^2) + (0.0379P)^2 + 0.610P] where P = (Fo^2 + 2Fc^2)/3
S = 1.17	(Δ/σ)max < 0.001
1322 reflections	Δρmax = 0.38 e Å^−3
90 parameters	Δρmin = −0.40 e Å^−3
0 restraints	Extinction correction: SHELXL97 (Sheldrick, 2008), Fc^*=kFc[1+0.001xFc^2λ^3/sin(2θ)]^-1/4
Primary atom site location: structure-invariant direct methods	Extinction coefficient: 0.054 (4)

Special details

Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes.

Refinement. Refinement of F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > σ(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq	Occ. (<1)
K1	0.8481 (2)	0.35099 (10)	0.62651 (14)	0.0387 (5)	0.412 (5)
N1	0.8481 (2)	0.35099 (10)	0.62651 (14)	0.0387 (5)	0.588 (5)
Cu1	0.5000	0.5000	1.0000	0.01986 (18)
O1	0.5347 (3)	0.38741 (16)	0.8251 (2)	0.0344 (5)
H1E	0.5031	0.3269	0.8462	0.052*
H1F	0.4391	0.4009	0.7342	0.052*
O2	0.3314 (3)	0.61377 (17)	0.8339 (2)	0.0362 (5)
H2E	0.4091	0.6201	0.7650	0.054*
H2F	0.1921	0.5941	0.7860	0.054*
O3	0.7862 (3)	0.56493 (14)	0.9975 (2)	0.0257 (4)
H3E	0.8249	0.5629	0.9109	0.038*
H3F	0.8009	0.6269	1.0249	0.038*
S1	0.76452 (11)	0.63851 (5)	0.59658 (7)	0.0252 (2)
O4	0.8778 (3)	0.56887 (15)	0.7248 (2)	0.0293 (4)
O5	0.7358 (4)	0.5769 (2)	0.4569 (2)	0.0491 (6)
O6	0.9066 (4)	0.73372 (17)	0.5944 (3)	0.0417 (5)
O7	0.5467 (3)	0.67430 (16)	0.6151 (2)	0.0359 (5)

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
K1	0.0395 (8)	0.0397 (8)	0.0393 (8)	−0.0010 (5)	0.0147 (5)	−0.0019 (5)
N1	0.0395 (8)	0.0397 (8)	0.0393 (8)	−0.0010 (5)	0.0147 (5)	−0.0019 (5)
Cu1	0.0203 (3)	0.0202 (3)	0.0188 (3)	−0.00122 (15)	0.00477 (17)	0.00202 (15)
O1	0.0384 (11)	0.0316 (10)	0.0305 (10)	−0.0060 (8)	0.0044 (8)	0.0042 (8)
O2	0.0281 (10)	0.0462 (12)	0.0331 (10)	−0.0049 (9)	0.0060 (8)	0.0046 (9)
O3	0.0271 (9)	0.0251 (9)	0.0264 (9)	−0.0028 (7)	0.0101 (7)	−0.0013 (7)
S1	0.0272 (4)	0.0247 (3)	0.0225 (3)	0.0025 (2)	0.0046 (3)	0.0043 (2)
O4	0.0319 (10)	0.0311 (10)	0.0250 (9)	0.0041 (8)	0.0080 (8)	0.0082 (7)
O5	0.0637 (16)	0.0558 (14)	0.0239 (10)	0.0098 (12)	0.0053 (10)	−0.0066 (10)
O6	0.0356 (11)	0.0321 (11)	0.0559 (13)	−0.0021 (9)	0.0098 (10)	0.0153 (10)
O7	0.0264 (10)	0.0337 (11)	0.0477 (12)	0.0045 (8)	0.0101 (9)	0.0036 (9)

Geometric parameters (Å, º)

K1—O4	2.825 (2)	Cu1—O2	2.130 (2)
K1—O7^i	2.855 (3)	Cu1—O1^v	2.178 (2)
K1—O6^ii	2.982 (3)	Cu1—O1	2.178 (2)
K1—O5^iii	3.029 (3)	O2—H2E	0.8974
K1—O6^iii	3.034 (3)	O2—H2F	0.8954
K1—O1	3.042 (2)	O3—H3E	0.8911
K1—O5	3.177 (3)	O3—H3F	0.8025
K1—O2^iv	3.188 (2)	S1—O5	1.460 (2)
Cu1—O3^v	1.9616 (18)	S1—O6	1.474 (2)
Cu1—O3	1.9616 (18)	S1—O4	1.4750 (18)
Cu1—O2^v	2.130 (2)	S1—O7	1.480 (2)
O4—K1—O7^i	108.73 (7)	O3—Cu1—O1	88.90 (8)
O4—K1—O6^ii	102.26 (6)	O2^v—Cu1—O1	88.88 (8)
O7^i—K1—O6^ii	138.70 (7)	O2—Cu1—O1	91.12 (8)
O4—K1—O5^iii	79.30 (7)	O1^v—Cu1—O1	180.0
O7^i—K1—O5^iii	117.01 (7)	K1^vi—O3—K1	136.21 (5)
O6^ii—K1—O5^iii	94.77 (7)	Cu1—O3—K1^vii	147.73 (7)
O4—K1—O6^iii	122.65 (7)	K1^vi—O3—K1^vii	77.98 (4)
O7^i—K1—O6^iii	85.81 (7)	K1—O3—K1^vii	96.79 (4)
O6^ii—K1—O6^iii	99.98 (6)	O5—S1—O6	109.58 (15)
O5^iii—K1—O6^iii	46.59 (6)	O5—S1—O4	108.51 (13)
O4—K1—O1	70.47 (6)	O6—S1—O4	108.30 (12)
O7^i—K1—O1	85.60 (7)	O5—S1—O7	110.85 (14)
O6^ii—K1—O1	79.65 (6)	O6—S1—O7	109.70 (12)
O5^iii—K1—O1	147.09 (7)	O4—S1—O7	109.84 (11)
O6^iii—K1—O1	166.23 (7)	S1—O4—K1	108.83 (10)
O4—K1—O5	46.32 (6)	S1—O4—K1^vii	80.40 (8)
O7^i—K1—O5	71.58 (7)	K1—O4—K1^vii	158.82 (6)
O6^ii—K1—O5	147.28 (7)	S1—O4—K1^iii	56.66 (7)
O5^iii—K1—O5	73.12 (8)	K1—O4—K1^iii	89.06 (5)
O6^iii—K1—O5	93.36 (6)	K1^vii—O4—K1^iii	80.16 (4)
O1—K1—O5	94.12 (7)	S1—O5—K1^iii	99.07 (13)
O4—K1—O2^iv	146.26 (7)	S1—O5—K1	93.83 (11)
O7^i—K1—O2^iv	74.47 (6)	K1^iii—O5—K1	106.88 (8)
O6^ii—K1—O2^iv	64.65 (6)	S1—O5—K1^i	86.41 (11)
O5^iii—K1—O2^iv	130.48 (7)	K1^iii—O5—K1^i	138.98 (9)
O6^iii—K1—O2^iv	90.89 (6)	K1—O5—K1^i	113.31 (8)
O1—K1—O2^iv	76.44 (6)	S1—O6—K1^vii	123.09 (12)
O5—K1—O2^iv	145.35 (7)	S1—O6—K1^iii	98.47 (11)
O3^v—Cu1—O3	180.000 (1)	K1^vii—O6—K1^iii	120.48 (8)
O3^v—Cu1—O2^v	89.90 (8)	S1—O7—K1^i	120.30 (12)
O3—Cu1—O2^v	90.10 (8)	S1—O7—K1	48.17 (7)
O3^v—Cu1—O2	90.10 (8)	K1^i—O7—K1	101.15 (6)
O3—Cu1—O2	89.90 (8)	S1—O7—K1^viii	150.23 (11)
O2^v—Cu1—O2	180.000 (1)	K1^i—O7—K1^viii	89.01 (6)
O3^v—Cu1—O1^v	88.90 (8)	K1—O7—K1^viii	136.85 (5)
O3—Cu1—O1^v	91.10 (8)	S1—O7—K1^vii	62.27 (8)
O2^v—Cu1—O1^v	91.12 (8)	K1^i—O7—K1^vii	153.99 (7)
O2—Cu1—O1^v	88.88 (8)	K1—O7—K1^vii	97.79 (4)
O3^v—Cu1—O1	91.10 (8)	K1^viii—O7—K1^vii	89.26 (5)

Symmetry codes: (i) −x+1, −y+1, −z+1; (ii) −x+2, y−1/2, −z+3/2; (iii) −x+2, −y+1, −z+1; (iv) −x+1, y−1/2, −z+3/2; (v) −x+1, −y+1, −z+2; (vi) −x+2, −y+1, −z+2; (vii) −x+2, y+1/2, −z+3/2; (viii) −x+1, y+1/2, −z+3/2.

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
O1—H1E···O7^iv	0.81	1.96	2.760 (3)	173
O1—H1F···O5^i	0.90	1.82	2.720 (3)	175
O2—H2E···O7	0.90	1.93	2.801 (3)	164
O2—H2F···O4^ix	0.90	1.91	2.791 (3)	166
O3—H3E···O4	0.89	1.83	2.714 (2)	174
O3—H3F···O6^x	0.80	1.89	2.676 (3)	167

Symmetry codes: (i) −x+1, −y+1, −z+1; (iv) −x+1, y−1/2, −z+3/2; (ix) x−1, y, z; (x) x, −y+3/2, z+1/2.