The Role of Acidity in the Synthesis of Novel Uranyl Selenate and Selenite Compounds and Their Structures

: Herein, the novel uranyl selenate and selenite compounds Rb 2 [(UO 2 ) 2 (SeO 4 ) 3 with one H 2 O and four SeO 42 − ligands, creating inﬁnite chains. (UO 2 ) 2 (HSeO 3 ) 2 (H 2 SeO 3 ) 2 Se 2 O 5 synthesized under slow evaporation conditions adopts a 0D orthorhombic structure in space group Cmc 2 1 with a = 28.4752(12) Å, b = 6.3410(3) Å, c = 10.8575(6) Å, and V = 1960.45(16) Å 3 , consisting of discrete rings of [(UO 2 ) 2 (HSeO 3 ) 2 (H 2 SeO 3 ) 2 Se 2 O 5 ] 2 . (UO 2 ) 2 (HSeO 3 ) 2 (H 2 SeO 3 ) 2 Se 2 O 5 is apparently only the second example of a uranyl diselenite compound to be reported. A combination of single crystal X-ray diffraction and bond valance sums calculations are used to characterise all samples obtained in this investigation. The structures uncovered in this investigation are discussed together with the broader family of uranyl selenates and selenites, particularly in the context of the role acidity plays during synthesis in coercing speciﬁc structure, functional group, and topology formations.


Introduction
The acquisition of novel non-centrosymmetric (NCS) structural phases has formed a topical focus of inorganic chemical synthesis development due to their prominent application in a variety of non-linear optic (NLO) [1,2], piezoelectric [3], and ferroelectric [4] based materials. In the case of NLO materials, stereo active lone pair possessing cations Se 4+ and I 5+ are commonly utilized as their possession of periodic polarizable free lone pair electrons within NSC lattices allow materials with superior second harmonic generation (SHG) properties to be obtained. A growing theme among recent NLO and SHG Se and I related material development studies has been the extension of these to uranium and actinide compounds, as it has been shown that their related properties can often exceed those of non-actinide variants [5][6][7][8]. The properties of these materials can be further improved through the increased polymerization of IO 3 − and SeO 3 2− units into polyiodate and polyselenite units [9,10]. However, although often similarly chemically regarded due to their comparable ionic radii as oxoanions (IO 3 − and SeO 3 2− ), the dissimilar chemistry of the chalcogen group belonging Se cation to that of the Halogen I cation results in somewhat of a disparity in chemical insight, particularly regarding polymerisation of groups as oxoanions. For instance, at present a larger proportion of polyiodate units have been observed in the case of I 2 O 5 , I 3 O 8 − , I 4 O 11 2− , and I 5 O 14 3− whereas only Se 2 O 5 2− is reported for polyselenites [11][12][13]. Accordingly, understanding the synthesis conditions and design which pertain to the polymerization of SeO 3 2− units is a pertinent pursuit in the context of next generation NLO and SHG material generation in addition to fundamental inorganic synthesis development.
It was recently shown that the first uranium and actinide polyiodate, UPI-1 [14], with the polyiodate unit I 4 O 11 2− could be obtained under conditions of extreme acid addition where it was noted similar synthesis experiments with more moderate acid addition failed. It was highlighted in that investigation that, in many related studies [11,15,16], heightened acid addition often results in iodate products consisting of polymerized units and this could be used as a potential means of synthesis control [14]. This theme of utilizing acidic media in synthesis control, although a relatively recent innovation, has seen further promising applications in a variety of other uranyl/actinyl structural chemical investigations. Examples can be seen in uranium and neptunium nitrate studies, including the identification acid controlled geometric isomerism in neptunium and uranium nitrates [17], and also uncovering the crystallization of uranium nitrates as an indicator of reaction competition [18]. The salient effect solution acidity has upon controlling structure formation in uranyl systems has also been recently observed in uranyl chromates and chromate-nitrates [19][20][21]. A further pertinent study by Nazarchuk et al. [22] demonstrated that, under decreasing pH conditions (2.5 → 1.5), successive organically templated uranyl sulfates would form where their structural chemistry was dependent and proportional to the specific pH at crystallisation. The effect of acidic conditions on redox processes was also highlighted by Langer et al. [23], in examining neptunium selenates and selenites, where acid driven disproportion was found to play a key role in determining the final Np and Se redox states in crystallized products. Consequently, the utilisation of acidic conditions and pH control, although a sensitive component, can be viewed as a route to novel and/or difficult to obtain structure formation.
Traditional interest in uranium and actinide iodate and selenite/selenate compound synthesis stems from understanding potential alteration phases that may occur in spent nuclear fuel (SNF) and in repository environments due to the prominent occurrence of I-129 and Se-79 as fission daughters [24,25]. Many of these studies were performed on what can be described as an explorative basis which left little understanding regarding chemical conditions which dictate synthesis control and outcome. However, more contemporary investigations have utilized systematic approaches, which better enable improved synthesis control and increased understanding and prediction of potential reaction outcomes [17,[26][27][28][29][30][31][32][33]. Saliently, many of these more recent studies have highlighted the unique ability and promising application of uranium derived compounds in societally important materials such as radiation detectors, scintillators to proton conductors [34][35][36].
A key component driving these promising applications is the combination of traditional functional chemical components such as stereo-active lone pair possessing IO 3 − and SeO 3 2− units with that of the uranium's unique chemical character as a 5f element which, together, drives the properties of obtained structures to often rival or excede those of non-uranium/actinide analogues [14,36]. This has been exemplified by UPI-1 and other monoiodate and monoselenite/monoselenate possessing uranyl compounds [5,6,14]. Compared to uranyl iodates, there is relatively less known about selenites and selenates in respect of synthesis conditions and control, despite the interest these receive in context of NLO and SHG materials development. Indeed, it has been more than 30 years since the first and only uranyl polyselenite was reported, UO 2 Se 2 O 5 , possessing the Se 2 O 5 2− moiety [12]. Consequently, there is precedence to further understand the synthesis conditions of uranyl selenites and selenates both in the context of potential SNF alteration phase and for advanced functional material development.
We have recently embarked on a systematic investigation into the structure-property relations and effect of synthesis on novel SNF related actinide phase formation under variable conditions [14,17,[29][30][31]37,38]. In light of the identification of the effect of acid on controlling the polymerization of iodate units in uranyl iodates [14], we have applied similar methods in the following work to investigate the effect acid has on polymerizing selenite and selenate units. In the course of our investigation we have uncovered four novel uranyl selenites/selenates including Rb 2 [(UO 2 ) 2 (SeO 4 ) 3  O, and placed in Teflon lined stainless steel autoclaves. Concentrated selenic acid (60 wt.%) was added to the reaction mixtures to reduce the pH to below 1 confirmed using pH paper. The Teflon lined stainless steel autoclaves were then secured and transferred into a box furnace. The furnace was heated to 220 • C, held there for 36 h, and then cooled to room temperature at a rate of 3 • C/h. Fine greenish single crystals of the three described compounds were recovered from their respective reaction vessels and used for single crystal X-ray diffraction measurements. The crystal size was found to be approximately 30-40 µm in length.
(b) Slow evaporation; (UO 2 ) 2 (HSeO 3 ) 2 (H 2 SeO 3 ) 2 Se 2 O 5 UO 2 (NO 3 ) 2 ·6H 2 O (0.02 g, 0.04 mmol International Bioanalytical Industries Inc.) and H 2 SeO 3 (0.01 g, 0.08 mmol, Alfa-Aesar 99.0% min) were dissolved in 10 mL of deionised water and placed in a 20 mL jar. Concentrated selenic acid (60 wt.%) was added to the jar to reduce the pH to below 1 confirmed using pH paper. The jar containing the solution was placed in a fume cupboard and the solution was left to evaporate to dryness, where the formation of green fine crystals approximately 30 µm in length was observed. Subsequently, they were separated for single crystal X-ray diffraction.

Single Crystal X-ray Diffraction
Suitable quality crystals were selected for analysis and structural refinement using single crystal X-Ray diffraction (SC-XRD). For all title compounds data were recorded with CrysAlisPro software (v.1.171.38.43) on an Agilent Oxford Diffraction Super Nova diffractometer with a Mo Kα tube at 296 K. Absorption corrections for the raw data were performed using the multiscan method. The unit cell was determined, and background effects were processed by the CrysAlisPro software. The initial structures of the title compounds were refined through using SHELXL-2018 within the WinGX (v1.80.05) soft-ware [39], and the ADDSYM algorithm of the PLATON program [40] was used for checking possible higher symmetries. Localisation of H atoms positions was not attempted due to unreliability as only RT data was collected. Data and crystallographic information for all compounds investigated are presented in Table 1

Bond Valence Sums
Bond valence sums (BVS) calculations were undertaken for the cations in  [42] were used for Se and O.

Structural Studies (a) Rb 2 [(UO 2 ) 2 (SeO 4 ) 3 ]
Rb 2 [(UO 2 ) 2 (SeO 4 ) 3 ] synthesized using a hydrothermal method was found to adopt a tetragonal structure in space group P4 2 /ncm with a = 9.8312(4) Å, c = 15.4924(9) Å and V = 1497.38(15) Å 3 using SC-XRD measurements. The structure consists of seven coordinate U atoms which form UO 7 pentagonal bipyramid polyhedra with two short axial and collinear oxo U=O bonds of 1.741(11) Å, consistent with the uranyl bond in the hexavalent state. Surrounding the uranyl groups equatorially in the UO 7 polyhedra are five U-O bonds ranging in length from 2.372(11) to 2.400(10) Å which are provided by coordinating SeO 4 2− ligands fivefold (Figure 1). The Se-O bond lengths were found to range from 1.59(5) to 1.656(11) Å. The UO 2 (SeO 4 ) 5 8− units link with others in a corner sharing motif to form [(UO 2 ) 4 (SeO 4 ) 6 ] 4− sheets in the [110] plane ( Figure 1). The sheets are observed to adopt a corrugated motif viewing from the [100] direction ( Figure 1a) in which Rb + cations are located in the interlayer space. The SC-XRD structural solution indicated that the O(6) was positionally disordered and this is illustrated in Figure 1b where the disordering occurs parallel to the a-b plane. BVS values of 6.11, 6.16, and 6.09 were determined for the U(1), Se(1), and Se (2) positions, supporting the occurrence of hexavalent states for U and Se in the structure of Rb 2 [(UO 2 ) 2 (SeO 4 ) 3 ]. Full BVS values are given in Table 2. A notable observation of Rb 2 [(UO 2 ) 2 (SeO 4 ) 3 ] is the corrugation of the [(UO 2 ) 4 (SeO 4 ) 6 ] 4− sheets. It is typical of uranyl bearing compounds to form 2D layered or 1D chain structures due to the oxo-bond nature of the uranyl group [43]. Deformation of the structure or occurrence of corrugation in layers often occurs through factors, such as the application of pressure during synthesis [30,37,44], non-spherical interlayer ions [45], or the unideal coordination of layer ions [30,46], where the structures or layers deform accordingly to accommodate these. In the case of Rb 2 [(UO 2 ) 2 (SeO 4 ) 3 ], the relatively simplistic formula would suggest a less distorted structure, however the occurrence of UO 7 and SeO 4 polyhedra results in the latter of the described examples. The Se(1)O 4 polyhedra is able to satisfactorily fourfold coordinate to uranyl centres, whereas the Se(2)O 4 polyhedra can only threefold coordinate, which leaves a vacant Se-O ligand that is positionally disordered. Accordingly, it is suspected that this unideal coordination of SeO 4 polyhedra due to the occurrence UO 7 will lead to corrugation of the layers and the occurrence of disorder. Indeed, the same sheet corrugation is observed in Cs 2 [(UO 2 ) 2 (CrO 4 ) 3 ] [  Full BVS values are given in Table 3.  [53]. In all these, chains are constructed through edge-sharing UO 6 and UO 7 polyhedra that link to form sheets via SeO 3 units through a phosphuranylite anion topology [52]. It was noted by Gurzhiy et al. [52], when comparing these natural and synthetic selenites, that they are contrastingly described by different orientation matrices, in that the latter example possess layers which are described by the c2/m space group, whereas for the former structures, layers demonstrate higher symmetry with a pmmn description. Furthermore, when observing Rb 2 [(UO 2 ) 3 (SeO 3 ) 2 O 2 ], this follows where its layer topology is better described in the lower symmetry subset.

Structural Studies (c) Rb 2 [UO 2 (SeO 4 ) 2 (H 2 O)]·2H 2 O
Rb 2 [UO 2 (SeO 4 ) 2 (H 2 O)]·2H 2 O synthesized using a hydrothermal method was found to adopt an orthorhombic structure in space group Pmn2 1 where with a = 13.041(3) Å, b = 8.579(2) Å, c = 11.583(2) Å and V = 1295.9(5) Å 3 using SC-XRD measurements. The structure consists of seven coordinate U atoms which form UO 7 pentagonal bipyramid polyhedra that contain two short axial and collinear oxo U=O bonds of 1.765(10) Å consistent with the uranyl group for hexavalent U in this coordination geometry. Surrounding the uranyl group equatorially in the UO 7 polyhedra are five oxygen atoms corner-sharing provided by four SeO 4 2− ligands and one oxygen atom. BVS calculation of the lone oxygen ligand donor (O(2)) produces a low value of 0.60 consistent with an aqua H 2 O coordinating ligand (Figure 3c). The SeO 4 2− ligands act to link UO 7 polyhedra in the [001] direction forming infinite chains (Figure 3). Between the chains parallel to the [010] direction, Rb cations are observed with H 2 O molecules where, between the chains parallel to the [001] direction, the inter-space is free (Figure 3). BVS values of 6.19, 6.17, and 6.09 were determined for the U(1), Se(1), and Se (2) [54,55]. Both structures contain the same layer-chain configuration with aqua ligands extending off uranyl centres alternatively along the layer direction (Figure 3b). However the increase in the degree of hydration of Rb 2 [UO 2 (SeO 4 ) 2 (H 2 O)]·2H 2 O compared to Rb 2 [UO 2 (SeO 4 ) 2 (H 2 O)]·H 2 O results in a change of symmetry as well as an increase in unit cell volume (1295.9(5) vs. 1240.4(3) Å 3 respectively [54]). The influence of the degree of hydration has been demonstrated to play a subtle but significant role in the formation of thorium nitrates [56] as well as uranium and neptunium nitrates [17] [59] that such a configuration can lead to geometric isomerism occurring in the orientation of the sheets in these structures. It is noteworthy then that, through a variety of different synthesis/formation means, e.g., natural mineral formation, hydrothermal methods, or organic templating, that similar topological products can be obtained. BVS values of 6.05, 3.96, 4.08, 4.05, and 4.09 were determined for the U(1), Se(1), Se(2), Se (3) and Se(4) positions supporting the occurrence of hexavalent U and tetravalent Se in the structure of (UO 2 ) 2 (SeO 3 ) 2 (H 2 SeO 3 ) 2 Se 2 O 5 . Full BVS values are given in Table 5.   [12]. UO 2 Se 2 O 5 was originally synthesized using high temperature solid state methods involving excess SeO 2 within a sealed tube. This is contrast to (UO 2 ) 2 (HSeO 3 ) 2 (H 2 SeO 3 ) 2 Se 2 O 5 which was synthesized under slow evaporation conditions with excess acid. As previously described efforts were made to synthesis (UO 2 ) 2 (HSeO 3 ) 2 (H 2 SeO 3 ) 2 Se 2 O 5 and/or diselenite variants using conditions of more moderate pH but this subsequently failed. Such an observation was consistent with the uranyl polyiodate UPI-1 where it was noted the synthesis was only possible under conditions of heighted acidity [14]. The occurrence of HSeO 3 − and H 2 SeO 3 ligands in the structure of (UO 2 ) 2 (HSeO 3 ) 2 (H 2 SeO 3 ) 2 Se 2 O 5 are indicative of the acidic synthesis conditions used and it is strongly suggested that they play a key role in structure formation. Extending to non-actinide variants, the occurrence of the diselenite group has also been observed in a number notable examples, including Bi 3 (SeO 3 ) 3 (Se 2 O 5 )F, Al 2 (Se 2 O 5 ) 3 , Ga 2 (Se 2 O 5 ) 3 , and In 2 (Se 2 O 5 ) 3 [13,60]. These structures were found to exhibit superior SHG properties and were synthesized either using a sealed tube high temperature method (Al 2 (Se 2 O 5 ) 3 , Ga 2 (Se 2 O 5 ) 3 and In 2 (Se 2 O 5 ) 3 ) or via a hydrothermal method involving highly acidic conditions (Bi 3 (SeO 3 ) 3 (Se 2 O 5 )F). Another notable synthesis was that of Ca 2 (HSeO 3 ) 2 (Se 2 O 5 ) which was achieved using a slow evaporation method utilizing selenic acid [61]. Consistent with the previous deductions of this investigations, it follows that either high temperature mildly reducing or acidic conditions are necessary to drive the polymerisation of selenite units for the formation of diselenite moieties.

Synthesis and Structural Trends in Uranyl Selenites/Selenates
A comprehension study on the crystal chemistry of natural and synthetic uranyl selenites and some selenates was recently completed by Gurzhiy et al. [52]. In light of this, we will only briefly discuss the crystal chemistry of uranyl selenites and selenates in the context of the novel phases uncovered in this work and direct the reader to Gurzhiy et al. [52] for broader background information. It was highlighted by Gurzhiy et al. [52] that the structural complexity of natural vs. synthetic uranyl selenites is comparatively higher for the former case of structures. Gurzhiy et al. [52] rationalised these observations against the preference for natural-born architectures to exist rather than the synthetic counterparts. Although, as described by Frost et al. [62] in examining the natural uranyl sulfate mineral zippeite, K 2 [(UO 2 ) 6 (SO 4 ) 3 O(OH) 6 ]·8H 2 O, natural conditions often involve a higher temperature gradient compared to that of the synthetic regarding compounds synthesized using hydrothermal methods. Accordingly, high temperature synthesis conditions typically result in higher symmetry structure formation, which as a method for uranyl selenite and selenate synthesis is underutilized typically due to the volatility of selenium and the associated difficulty. However, such synthesis conditions were used to synthesize, prior to this investigation, the only reported diselenite UO 2 Se 2 O 5 [12] as well as notable non-actinide diselenite variants [13,60,61], as previously described. Although the mechanism for selenite polymerisation in UO 2 Se 2 O 5 is not clear it is speculated to be related to the combination of high temperature and the somewhat reducing conditions of a sealed tube. Nevertheless, as highlighted and argued in this investigation, pH control can act as promoter of selenite polymerisation in the case of slow evaporation acid assisted (UO 2 ) 2 (HSeO 3 ) 2 (H 2 SeO 3 ) 2 Se 2 O 5 synthesis.
The role of acidity during synthesis, as described previously, has been noted in several recent studies to act as a key determinant in controlling such chemical properties as structural motif, isomerism and also redox processes [17][18][19][20]22,23]. Nazarchuk et al. [22] described that, during the evaporation of acidic solutions, the successive reduction in pH accompanied by a loss of solution resulted in distinctive organically templated uranyl sulfate structures forming which were structural dependent and proportional to the pH at which crystallisation occurs. In the present study, initial solution pH conditions were targeted at <1, such that, at crystallisation and near complete evaporation of solution, the structures formed were at the extremity of acidity content. Similar extreme pH/acidity conditions and methodologies were utilized in the studies of variable geometric isomerism in uranyl and neptunyl nitrates [17] and also in UPI-1, the first actinide polyiodate [14]. Contrasting the systematic study of Nazarchuk et al. [22] with that of the aforementioned and present studies suggests greater control of structure formation can be obtained through more incremental pH steps and this may allow other structure derivatives to be obtained, e.g., via pH controlled addition or exclusion of acid or water units in structures.
A common feature of the present study has been the structural similarities of the uranyl selenates  [58]. The ability for the exchangeability of the different structural-chemical analogues arises due to the similar sizes of the selenate, sulfate, and chromate ions. Consequently, utilising synthetic techniques, such as acidic hydrothermal/slow evaporation conditions used in the present study, developed for a specific system, such as selenate, these can be readily transferred to sulfate or chromate with either similar or unexpected success in the synthesis. Naturally, variability in redox chemistry can limit transferability, e.g., the ion polymerisation process. However, other structural effects, such as degree of hydration and distortion mechanisms, can be readily examined, controlled, and contrasted against the three different systems.
In the present study and supported by observations in recent investigations [17][18][19][20]22,23], the use of low pH conditions acts as a facile path towards novel structure generation and obtaining topology architectures that would otherwise require more intricate synthesis methods, such as the use of organic templating methods. Consequently, it is suggested then, from the observations in this investigation, that the use and presence of acid during synthesis provides a facile pathway to acquiring uranyl selenite/selenate structures with functional groups or topologies that are either novel or require more intricate methods.

Conclusions
The novel uranyl selenites and selenates Rb 2 [(UO 2 ) 2 (SeO 4 ) 3 ], Rb 2 [(UO 2 ) 3 (SeO 3 ) 2 O 2 ], Rb 2 [UO 2 (SeO 4 ) 2 (H 2 O)]·2H 2 O, and (UO 2 ) 2 (HSeO 3 ) 2 (H 2 SeO 3 ) 2 Se 2 O 5 have been synthesized using either slow evaporation or hydrothermal methods, both involving extensive acid addition, and their structures characterized using single crystal X-ray diffraction. Prior to this investigation, only one uranyl diselenite was known [12]. Accordingly, the synthesis herein of (UO 2 ) 2 (HSeO 3 ) 2 (H 2 SeO 3 ) 2 Se 2 O 5 utilizing acidic conditions provides only the second such example and the first from slow evaporation synthesis. Key to this and the other novel compounds described is the application of extensive acid during synthesis, which is argued to act as a novel and facile method towards acquiring structures and topologies that would otherwise require more intricate methods such as organic templating. With the consideration of recent similar studies [14,17], this investigation points towards the pertinence of acid and pH control in the synthesis of novel uranyl selenite and selenate compositions and structures, particularly those that that may possess salient SHG properties.