A Preparation and Structural Characterization of the Hydrated Trifluoride Salt of the [Co(NH₃)₆]³⁺ Cation: [Co(NH₃)₆]F₃ · 2H₂O. The Unexpected Preparation of [Co(NH₃)₆](BF₄)(SiF₆) from Pyrex Glass, and the Syntheses and Structures of (NH₄)[Co(NH₃)₆](BF₄)₂(2Cl) and of (NH₄)[Co(NH₃)₆]₂(SiF₆)₃(Cl) · 3H₂O

Sandra Mikhael ¹, Maria Shawky ¹, Gulraiz Hashmi ¹, Ivan Bernal ^{1, 2}, Roger A. Lalancette ^{1, *}

1. Carl A. Olson Memorial Laboratories, Department of Chemistry, Rutgers University, 73 Warren St., Newark, NJ, USA
2. Molecular Sciences Institute, School of Chemistry, University of the Witwatersrand, Private Bag 3, 2050 Johannesburg ZA, USA

* Correspondence: Roger A. Lalancette

Academic Editor: Maxim L. Kuznetsov

Special Issue: Coordination Chemistry and Metal Complexes

Received: June 28, 2020 | Accepted: July 21, 2020 | Published: July 23, 2020

Advances in Chemical Research 2020, Volume 2, Issue 3, doi:10.21926/acr.2003006

Recommended citation: Mikhael S, Shawky M, Hashmi G, Bernal I, Lalancette RA. A Preparation and Structural Characterization of the Hydrated Trifluoride Salt of the [Co(NH₃)₆]³⁺ Cation: [Co(NH₃)₆]F₃ · 2H₂O. The Unexpected Preparation of [Co(NH₃)₆](BF₄)(SiF₆) from Pyrex Glass, and the Syntheses and Structures of (NH₄)[Co(NH₃)₆](BF₄)₂(2Cl) and of (NH₄)[Co(NH₃)₆]₂(SiF₆)₃(Cl) · 3H₂O. Advances in Chemical Research 2020; 2(3): 006; doi:10.21926/acr.2003006.

© 2020 by the authors. This is an open access article distributed under the conditions of the Creative Commons by Attribution License, which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is correctly cited.

1. Introduction

Tassaert in 1798 described the tri-chloride of [Co(NH₃)₆]⁺³ [1], and in the early days of Coordination Chemistry, Jørgensen and Werner isolated many of this cation’s other salts; however, the fluoride was never reported. There exists no entry for its structure in the Inorganic Structural Database [2]. In the 1960’s, attempts were made by Gimenez-Huget [3], who concluded that he was unsuccessful in (a) obtaining crystalline material of a reliable composition; (b) his efforts to obtain a structure from the isolated crystalline material were equally frustrated. Nonetheless, he described “fine yellow needles” of approximate composition [Co(NH₃)₆]F₃ · 2.5HF that “decomposed on exposure to light”. We believe this latter remark can be accounted for by the presence of metallic particles of Ag^o in the products of some of his attempts, inasmuch as he also recorded the presence of “black particles” [3], which can account for the sensitivity to light mentioned above. Gimenez-Huget was unable to prepare material of well-defined composition and quality usable for X-ray diffraction, even using Teflon-ware. His thesis [3] was never published in the scientific journals. In 1975, a report appeared announcing the synthesis of a substance with composition [Co(NH₃₎₆]F₃ that was further characterized by thermogravimetric analysis up to its decomposition point, where some of its gaseous products were determined. However, the crystal structure of this anhydrous complex was not reported [4].

Our early attempts were made by (a) addition of PbF₂ to [Co(NH₃)₆]Cl₃; (b) standard oxidation of cobaltous fluoride using H₂O₂ and heat; (c) attempting the precipitation of the desired salt from a solution of [Co(NH₃)₆]Cl₃ by adding a large excess of ammonium fluoride; (d) adding solid AgF to a solution of [Co(NH₃)₆]I₃, to hopefully precipitate AgI and isolate the trifluoride salt. This last one gave the desired product.

The crystals we isolated, when the reaction came in contact with glass, were unexpectedly found to contain the double salt [Co(NH₃)₆](BF₄)(SiF₆), (A), which crystallizes in space group F-43m. The same results were obtained from experiments using plastic ware (or Teflon), but then using a sintered-glass funnel for filtering, which was equally surprising, since a totally unexpected species crystallized, which we eventually discovered was the result of ammonium fluoride attacking the Pyrex glass frit even during the short time it took to filter the mother liquor. Eventually, we succeeded in obtaining the desired crystalline salt, [Co(NH₃)₆]F₃ · 2H₂O (B), using Teflon-ware, plastic funnels and filter paper. In separate attempts to isolate one of the components of the double salt (A), we obtained crystals of (NH₄)[Co(NH₃)₆](BF₄)₂(2Cl), (C), that crystallizes in space group R3, and (NH₄)[Co(NH₃)₆]₂(SiF₆)₃(Cl) · 3H₂O (D) (which belongs in space group P2₁/c). All three (A), (C), and (D) constitute examples of co-crystalline, or double salts. There is a related complex [Co(NH₃)₆]Cl₂ · BF₄, published in 2005 [5], which describes the first structure determination of a tetrafluoroborate salt of hexamminecobalt(III); hydrogen bonding is found between the amine H atoms and both the Cl^- and F^- in the counteranions of this salt.

2. Materials and Methods

All the chemicals were of analytical reagent grade and were obtained from Sigma Aldrich, Fisher Scientific or VWR, and used without purification.

2.1 Synthesis of the Metal Complexes

2.1.1 Synthesis of (A): The Unexpected Preparation of [Co(NH₃)₆](BF₄)(SiF₆)

Trial 1: To CoF₂ (0.76 g) dissolved in 50 mL water at 65^oC in a plastic beaker were added 0.2 g charcoal and 0.83 g NH₄F and this slurry was heated in a water both until all dissolved but the charcoal; after adding 5.0 mL concentrated ammonia (15M), the solution was cooled in an ice bath, then 10 mL of 10% H₂O₂ were added slowly until effervescence stopped; again, the solution was heated to 60^oC, for 10 minutes, and filtered through a sintered glass-frit funnel (hence the attack of the funnel’s Pyrex glass by the excess fluoride!). 120 mL water was added and this was heated and stirred until the product dissolved. The resulting solution was filtered once again and the filtrate set aside to evaporate and crystallize.

Trial 2: Same quantities and procedures as above, but in a Teflon beaker sitting in an oil bath at 90^oC for 5 days, which required adding water periodically. After crystallization, this was filtered using a glass-frit filter (hence, once more, the attack of the Pyrex glass by the excess fluoride!). Both of these procedures yielded crystals of (A). Thus, it became clear that the unexpected product, which we later determined to contain boron tetrafluoride and silicon hexafluoride, must come from the frit and/or glass filter.

2.1.2 Synthesis of (B): The Preparation of [Co(NH₃)₆]F₃ · 2(H₂O), Using Teflon and Other Plastic-Ware

Dissolve 0.504 g (0.93 mmol) [Co(NH₃)₆]I₃ in 50.0 mL of water in a Teflon beaker by heating in a water bath at 50^oC. Grind 0.390 g AgF (3.1 mmol, 10% mole excess) in a mortar and pestle, in the dark, and slowly add it, with stirring, to the solution of the [Co(NH₃)₆]I₃. White AgI precipitated. Filter through paper, using a plastic funnel. The filtrate was kept in the dark in a refrigerator and allowed to crystallize. (Note: It turned black from the Ag metal formed on exposure to light other than that of the dark room). Adding hot distilled water and then filtering through paper in a plastic funnel provided a solution of the desired tri-fluoro complex. It was necessary to repeat that procedure (hot water and filtering to remove Ag^o particles) many times, and the resulting orange clear solution was allowed to crystallize at room temperature. Bright orange crystals were analyzed: Calcd. for CoF₃H₂₀N₆O₂: F, 22.43%; H, 8.72%; N, 33.07%; Found: F, 22.69%; H, 8.10%; N, 32.18%.

2.1.3 Synthesis of (C): The Preparation of (NH₄)[Co(NH₃)₆](BF₄)(2Cl)

After failing multiple times to prepare crystals of [Co(NH₃)₆](BF₄)(SiF₆) from a solution containing [Co(NH₃)₆]Cl₃ plus SiF₆^2- and BF₄^- anions (the latter two in equimolar amounts), we decided to prepare crystals of the individual salts, as follows: in a beaker, we dissolved 0.5 g of [Co(NH₃)₆] Cl₃ (1.9 mmol) in 20 mL of water; then we added, dropwise, a solution of 0.59 g of NH₄BF₄ (5.6 mmol) (1:3 ratio) dissolved in 20mL of water. This solution yielded rod-like, orange crystals of (C), whose structure is reported here.

2.1.4 Synthesis of (D): The Preparation of (NH₄)[Co(NH₃)₆]₂(SiF₆)₃(Cl) · 3 H₂O

To 0.5 g of [Co(NH₃)₆] Cl₃ (1.9 mmol) dissolved in 20 mL of water, was added, dropwise, 0.52 g of Na₂SiF₆ (2.9 mmol) (1:1.5 ratio) dissolved in 25mL of water. Because of solubility problems, this mixture was put into a glass bomb, and the temperature was raised to about 150°C for approximately 2 hours with stirring; on cooling, crystals of (D) were observed, and filtered out of the solution.

2.2 X-Ray Diffraction Data Collection and Processing for (A), (B), (C), and (D)

A crystal of each was mounted on a Bruker APEXII X-ray diffractometer using graphite-monochromated CuKα (λ = 1.54178 Å) radiation, oriented, and data were collected. For (A), after data collection at 296K, the crystal was brought to 100K and reoriented. The cell was essentially the same, except for slight changes due to decreased temperature; however, because of instrument limitations, only a partial dataset was collected at 100K, processed, and the results were the same as those above. For (B), after data collection at 100K, the crystal was allowed to warm to RT (296K) and reoriented, and the cell was essentially the same except for the changes associated with the higher temperature. Data were not collected at RT. For (C) and (D), data were collected at 100K using ω and φ scans with the sample(s) maintained at a constant temperature of 100K using an Oxford Cryostream. Data processing, Lorentz-polarization, and face-indexed numerical absorption corrections were performed using SAINT, APEX, and SADABS computer programs [6,7,8,9,10]. The structures were solved by direct methods and refined by full-matrix least-squares methods on F², using the SHELXTL V6.14 programpackage. All non-hydrogen atoms were refined with anisotropic displacement parameters. For the three structures presented, structural and refinement parameters can be found in Table 1.

3. Results and Discussion

3.1 The Structure of [Co(NH₃)₆](BF₄)(SiF₆), (A)

The substance isolated from the synthetic procedure carried out in vitro (or in Teflon-ware, but then filtered through glass-frit) were identical and crystallized in the space group F-43m (No. 216) with partial occupancies of 18.00% BF₄^- and 82.00% SiF₆^2- in site (0.25, 0.75, 0.75) and vice-versa 82.00% BF₄^- and 18.00% SiF₆^2- in site (0.25, 0.25, 0.75). This is not surprising inasmuch as both [Co(NH₃)₆](BF₄)₃ and [Co(NH₃)₆]₂(SiF₆)₃ crystallize in space group Fm3m (No. 225) [11]. The packing diagram for (A) is shown in Figure 1, below.

If, from the above plot, one retains a single [Co(NH₃)₆]³⁺, and on the sides opposite the cobalt cation, a single BF₄^- and a single SiF₆^2-, one obtains Figure 2, below. Such a procedure is easily carried out using a command in DIAMOND [12].

Note that, while the Co and N atoms of the [Co(NH₃)₆]³⁺ cation can orderly sit at a site of -43m symmetry, the hydrogen atoms cannot; thus, they are disordered as shown. Another interesting view of the packing in the lattice of (A) is shown in Figure 3. Here, [Co(NH₃)₆]³⁺ cations are shown linked to one another by the agency of F-H₃N hydrogen bonds, approximately along the b-axis. Along the c-axis, the BF₄^- links the [Co(NH₃)₆]³⁺ cations as well. Additional [Co(NH₃)₆]³⁺ cations are not shown because the component species are diminished in size excessively.

Figure 1 The packing of (A) in projection down the c-axis, slightly canted so as to clearly reveal the presence of both the BF₄^- and SiF₆^2- ions, both of which are ordered. All figures in this report were generated using the program DIAMOND [12].

Figure 2 All three ions lie at positions with symmetry -43m and the BF₄^- and SiF₆^2- anions have only partial, but orderly, occupancies at the sites shown above. One fluorine from the BF₄^- at the lower left-hand was selected, and the entire anion removed; likewise, we then destroyed the SiF₆^2- anion at upper right. The procedure can be reversed with analogous results.

Figure 3 Another view of the contents of the lattice of [Co(NH₃)₆](BF₄)(SiF₆). The cations sit at an inversion center; therefore, the BF₄^- and SiF₆^2- anions extend the columns of the cobalt cations in all directions, shown here along the b- and c-axes, only. However, the overall (3D) packing diagram is so complex that additional components would merely clutter the figure.

3.2 The Structure of [Co(NH₃)₆](F₃) ^. 2H₂O, (B)

In Figure 4, we describe the environment around the unique cation in the asymmetric unit of the orthorhombic space group Pnma, at 100K.

Figure 4 The cation sits at an inversion center as is obvious from the labeling system. It is heavily hydrogen bonded by the fluoride counteranions, which also hydrogen bond the waters of hydration. Many bonds to the amine hydrogen atoms were omitted to avoid unnecessary cluttering. The complexity of the packing in these crystals is shown in Figure 5, below.

Figure 5 F1, F2, F3 and the waters (O1) link rows of cations along the b-axis, while the F3 anions link adjacent rows in this projection down the a-axis as well; note also that there is an inversion center located halfway between the pair of waters at the center of the diagram. Therefore, the structure consists of layers such as the two depicted above.

3.3 Structure of (NH₄)[Co(NH₃)₆](BF₄)(2Cl), (C)

This structure has the BF₄^- anions linked together through the agency of the ammonium cation, which in turn is H-bonded to one of the Cl ions in the lattice joining alternating rows, as displayed below in Figure 6.

Figure 6 Packing in the structure of (C), showing the multiple H bonds in this network. Note that the function of the ammonium cations (N3) is linking strings of tetrafluoroborate anions, which in turn stitch together strings of [cobalt cations + Cl^- ] fragments.

3.4 Structure of (NH₄)[Co(NH₃)₆]₂(SiF₆)₃(Cl) ^. 3H₂O, (D)

Figure 7 In this case, the packing differs from that of the BF₄^- because of the additional waters of hydration that act as bridges between ammonium cations (N14) and –NH₃ ligands to the cobalts. Additionally, three-water clusters link fluosilicates to the cobalt cations via ammonium cations.

The complexity of the hydrogen-bonded network in both (C) and (D) is such that an effort in displaying all those interactions in a single picture simply leads to figures so cluttered that they defeat the purpose of conveying a description of the overall network. Thus, the figures were designed to show examples of all types of hydrogen-bonded interactions present in each. Hopefully, we succeeded in so doing.

Table 1 X-Ray structural parameters for (A), (B), (C), and (D).

4. Conclusions

We are still surprised that the simple trifluoride was not isolated before, since in reality, the use of Teflon-ware for chemistry of fluorides is very old and some chemistry was done earlier using glassware coated with waxes. It simply took the current procedure (see Experimental) for the crystals to form from a stoichiometric solution of the relevant ions by starting with the iodide salt of the cobalt hexamine and adding AgF to precipitate the AgI salt. Starting with the trichloride salt of cobalt hexamine is not advantageous because of the greater solubility of AgCl. Note the desirability of darkness while working with silver!

The unexpectedly isolated double salt of SiF₆^2- and BF₄^- with [Co(NH₃)]³⁺ was described in the section on X-ray crystallography. An interesting and important result of our analysis is that the ratio of Si:B, obtained by crystallographically refining the occupancies of those anionic species produced a number very close to that present in Pyrex glass; namely, the composition of both Corning 7740 and Schott 8830 is given as 80.6% SiO₂; 12.6% B₂O₃; 4.2% Na₂O, 2.2% Al₂O₃; other traces, constitute only ca. 0.4%. To us, it is remarkable that we were able to determine its composition, by X-ray diffraction analysis of (A), especially since we were unaware of the composition of those crystals. The resulting co-crystal is unusual since the Si:B ratio is approximately 4:1. Thus, a study of the phase diagram of such a system would be interesting because of the possibility that there may be other well-defined crystalline phases with Si:B ratios such as 3:1, 2:1, etc. This would be especially so if those phases resulted in well-behaved, orderly co-crystals, given the importance of co-crystallization to the pharmaceutical industry [13].

We have prepared and examined the structures of (NH₄)[Co(NH₃)₆](BF₄)(2Cl), (C) and (NH₄)[Co(NH₃)₆]₂(SiF₆)₃(Cl) · 3H₂O, (D) while attempting to prepare, initially [Co(NH₃)₆]F₃ [B], and preparing instead the double salt, (A). In both cases we failed and, instead, isolated the double salts (C) and (D) - an unexpected bonus because it opens up an interesting area of phase diagrams and double salts analyses, which may be useful in a clearer understanding of that aspect of crystallization phenomena, which is still largely empirical in nature, as perusal of Bernstein’s [13] monograph illustrates.

Acknowledgments

We acknowledge the National Science Foundation for NSF-CRIF Grant No. 0443538 for part of the purchase of the X-ray diffractometer.

Additional Materials

The following additional materials are uploaded at the page of this paper.

1. Figure S1: ORTEP for [Co(NH₃)₆](BF₄)(SiF₆), (A): Ellipsoids are drawn at 40% probability level.
2. Figure S2: ORTEP for [Co(NH₃)₆](F₃) · 2H₂O, (B): Ellipsoids are drawn at 40% probability level.
3. Figure S3: ORTEP of (NH₄)[Co(NH₃)₆](BF₄)(2Cl), (C): Ellipsoids are drawn at 40% probability level.
4. Figure S4: ORTEP of (NH₄)[Co(NH₃)₆]₂(SiF₆)₃(Cl) · 3H₂O, (D): Ellipsoids are drawn at 30% probability level.

Author Contributions

Sandra Mikhael, Maria Shawky, Gulraiz Hashmi are undergraduate students who prepared the complexes under direct supervision of RAL. IB and RAL wrote the manuscript.

Competing Interests

The authors have declared that no competing interests exist.