A humidity-controlled precipitation technique enabling discovery of Rb3(H1.5PO4)2

Article abstract of DOI:10.1016/j.jssc.2020.121951

The previously unknown compound Rb₃(H_1.5PO₄)₂ is successfully synthesized here using a newly developed variant of aqueous precipitation crystal growth. The approach exploits the phenomenon of boiling point elevation in concentrated solutions. Crystals of the title compound were obtained upon heating a stoichiometric solution from 110 to 150 ?°C under a high steam partial pressure of 0.83 ?atm. Single crystal X-ray diffraction studies revealed Rb₃(H_1.5PO₄)₂ crystallizes in space group C2/m and is isostructural to Cs₃(H_1.5PO₄)₂. As evidenced by thermal analysis, Rb₃(H_1.5PO₄)₂ does not undergo a phase transition to a trigonal superprotonic phase upon heating. Even under a steam partial pressure of 0.82 ?atm, under which dehydration is suppressed to a temperature of 263 ?°C, no polymorphic transition is detected. The behavior parallels that of Cs₃(H_1.5PO₄)₂ and contrasts that of several structurally and chemically similar selenate compounds. The crystal growth approach developed here may prove particularly useful for obtaining water soluble compounds which are thermodynamically or kinetically disfavored at temperatures close to ambient.

Full text of DOI:10.1016/j.jssc.2020.121951

Journal of Solid State Chemistry 296 (2021) 121951
Contents lists available at ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
A humidity-controlled precipitation technique enabling discovery
of Rb3(H1.5PO4)2
,b,c,*
Sheel Sanghvi a, Sossina M. Haile a
a Department of Material Science and Engineering, Northwestern University, Evanston, IL, USA
b Applied Physics, Northwestern University, Evanston, IL, USA
c Department of Chemistry, Northwestern University, Evanston, IL, USA
A R T I C L E I N F O
A B S T R A C T
Keywords:
The previously unknown compound Rb3(H1.5PO4)2 is successfully synthesized here using a newly developed
variant of aqueous precipitation crystal growth. The approach exploits the phenomenon of boiling point elevation
in concentrated solutions. Crystals of the title compound were obtained upon heating a stoichiometric solution
from 110 to 150 ꢀC under a high steam partial pressure of 0.83 atm. Single crystal X-ray diffraction studies
revealed Rb3(H1.5PO4)2 crystallizes in space group C2/m and is isostructural to Cs3(H1.5PO4)2. As evidenced by
thermal analysis, Rb3(H1.5PO4)2 does not undergo a phase transition to a trigonal superprotonic phase upon
heating. Even under a steam partial pressure of 0.82 atm, under which dehydration is suppressed to a temperature
of 263 ꢀC, no polymorphic transition is detected. The behavior parallels that of Cs3(H1.5PO4)2 and contrasts that of
several structurally and chemically similar selenate compounds. The crystal growth approach developed here may
prove particularly useful for obtaining water soluble compounds which are thermodynamically or kinetically
disfavored at temperatures close to ambient.
Crystal Growth
Precipitation
Humidity
Solid Acid
Cs3H(SeO4)2
Cs3(H1.5PO4)2
1. Introduction
and 180 ꢀC, depending on composition. Surprisingly, the phosphate does
not undergo a superprotonic transition prior to decomposition, even
Solid acids with superprotonic conductivity continue to garner in-
terest as electrolytes in electrochemical cells for clean energy technolo-
gies, in particular fuel cells for converting hydrogen or other fuels into
electricity. Within this class, CsH2PO4 has been pursued for technological
implementation because it displays chemical stability against reaction
with either hydrogen or oxygen along with the requisite high conduc-
tivity (10ꢁ2 S/cm at 250 ꢀC) [1,2]. All other known superprotonic solid
acids bear sulfate or selenate polyanion groups, e.g., CsHSO4,
Cs3H(SeO4)2, and are unstable with respect to formation of H2S or H2Se
on exposure to H2, a reaction that is moreover accelerated in the presence
of Pt, typically utilized as a fuel cell electrocatalyst [3,4]. While the
properties of CsH2PO4 are sufficient to warrant the development of
commercial products, this electrolyte suffers from thermally induced
decomposition unless high levels of humidity are provided to suppress
the dehydration reaction [5–7]. The new solid acid phosphate
Cs3(H1.5PO4)2 [8,9] is essentially isostructural to M3H(XO4)2 (M ¼ Cs,
Rb, K, NH4; X ¼ S, Se) compounds that undergo a superprotonic transi-
tion from their respective room temperature monoclinic phases to a high
conductivity trigonal phase, with a transition temperature between ~ 60
under an atmosphere of full steam (water partial pressure, pH2O ~ 1 atm)
[8,9]. The discovery of Cs3(H1.5PO4)2 and its unexpected phase behavior
motivates efforts to synthesize analogous compounds of the form
M3(H1.5PO4)2 (M ¼ Li, Na, K, Rb, Cs). The results reported here describe
the new compound Rb3(H1.5PO4)2.
To date, no evidence of the existence of Rb3(H1.5PO4)2 has been
presented. On the contrary, Gaydamaka et al. [10] have shown that solid
state reaction at 55 ꢀC between RbH2PO4 and Rb2HPO4 in equimolar
concentration, corresponding to Rb3(H1.5PO4)2, produces a mixture of
Rb5H7(PO4)4 and Rb2HPO4. Attempts here to obtain Rb3(H1.5PO4)2 by
precipitation at room temperature from an aqueous solution with a
stoichiometric 3:2 M ratio of Rbþ:PO34ꢁ, again corresponding to the
stoichiometry of Rb3(H1.5PO4)2, produced crystals of Rb5H7(PO4)4. The
failure of these efforts to yield Rb3(H1.5PO4)2 does not preclude its
possible existence as a stable phase. For example, the compound
Rb5H3(SO4)4 does not form at ambient temperatures, but is readily
synthesized at temperatures above ~ 150 ꢀC by high temperature solid
state reaction [11]. Furthermore, even the most favorable thermody-
namic phases may be difficult to obtain due to kinetic competition from
* Corresponding author. Department of Material Science and Engineering, Northwestern University, Evanston, IL, USA.
E-mail addresses: sheel@u.northwestern.edu (S. Sanghvi), sossina.haile@northwestern.edu (S.M. Haile).
https://doi.org/10.1016/j.jssc.2020.121951
Received 25 October 2020; Received in revised form 25 December 2020; Accepted 28 December 2020
Available online 5 January 2021
0022-4596/© 2021 Elsevier Inc. All rights reserved.

S. Sanghvi, S.M. Haile
Journal of Solid State Chemistry 296 (2021) 121951
other phases. Here we report the growth of crystals of the new compound
Rb3(H1.5PO4)2 realized using high-temperature, humidity-controlled
precipitation. The material is isostructural to Cs3(H1.5PO4)2. Like the Cs
analog, Rb3(H1.5PO4)2 does not transform upon heating to a trigonal
phase prior to dehydration, although the dehydration may be preceded
by a subtle polymorphic transformation.
referred to as boiling point elevation [12]. We exploited this behavior to
access temperatures above 100 ꢀC for aqueous crystal growth.
The overall growth strategy can be understood in terms of the steps
indicated in Fig. 1(a), a hypothetical, but representative, phase diagram;
a schematic of the apparatus is presented in Fig. 1(b). The salt is first
allowed to equilibrate with steam under elevated temperature (T) and
water partial pressure (pH2O), by bringing the material from state A to
state B. The solid absorbs H2O from the gas atmosphere, undergoing
deliquescence, until reaching a saturation concentration, c0. The con-
centration in the equilibrated solution corresponds to that which has its
liquid-vapor equilibrium line passing through condition B, shown as the
left-most dashed line. The temperature of the system is then raised, while
leaving the inlet water partial pressure unchanged, bringing the envi-
ronmental conditions to C, at which the apparatus is held for a prolonged
period. Because the solution composition does not respond instanta-
neously to the temperature change, the vapor pressure of the solution
rises until it reaches 1 atm, state B0. From this point onwards, evaporation
proceeds quickly and the solution becomes increasing concentrated. The
state of the solution moves from B0 towards C00, with the salt
2. Crystal growth
Growth of single crystals of solid acid compounds is readily achieved
by controlled precipitation from aqueous solution. The principle is
straight-forward. A solution is held at a temperature at which evapora-
tion of H2O occurs such that the concentration of solute gradually ex-
ceeds the solubility limit. Implementing this strategy at temperatures
above 100 ꢀC, at which we speculated Rb3(H1.5PO4)2 is favored, appears
at first glance, to be impossible in the absence of high total pressure
because H2O boils under such conditions. However, a high concentration
of salt in the aqueous solution will shift the vapor-liquid phase boundary
to high temperatures relative to pure H2O, a phenomenon commonly
Fig. 1. Synthesis strategy for achieving crystal growth from high temperature aqueous solution: (a) Hypothetical phase diagram describing crystal growth pathway at
temperatures above 100 ꢀC; and (b) schematic of experimental apparatus. In (a) the solid black line is the liquid-vapor phase boundary of pure water, the dashed black
lines are hypothetical liquid-vapor phase boundaries for salt solutions of varying concentrations, ci, with c0 < c1 < c2 < c3 < c4, and the solid grey line is the pre-
cipitation boundary mapped from the hypothetical liquidus line of the associated temperature-concentration phase diagram. The crystal growth is achieved by
changing the global system state from B to C, causing the solution to follow the example trajectory indicated by the red arrows. See text for further discussion. (For
interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
2

S. Sanghvi, S.M. Haile
Journal of Solid State Chemistry 296 (2021) 121951
concentrations {ci} having values c0 < c1 < c2 < c3 < c4. At some con-
centration along this journey, the maximum solubility is reached, and the
solution enters a regime in which the condensed phase is a solid-liquid
mixture. That is, the solution concentration crosses the precipitation
boundary (solid grey line), a boundary fixed by the locus of points cor-
responding to the temperature-dependent liquidus line of the associated
temperature-concentration phase diagram. Depending on the details of
the liquidus boundary, the equilibrium vapor pressure over such a
condensed phase might now be expected to exceed 1 atm, following, for
example, the trajectory P to C0. So long as the maximum solubility lies
below the concentration c4 that would display a hypothetical liquid-
vapor equilibrium boundary passing through position C, where condi-
tion C corresponding to the T and pH2O of the external parameters during
the crystal growth period, H2O evaporation occurs under the hold con-
ditions, and a dry, crystalline solid is ultimately produced. To avoid
possible deliquescence on cooling, the pH2O is lowered (to attain con-
dition D) before reducing the temperature.
The procedure was implemented here according to the following
specific steps. First, RbH2PO4 was synthesized as one of the reagents. This
material was obtained by methanol-induced precipitation from an
aqueous solution of 1:2 M Rb2CO3 (Alfa Aesar, 99.9% metals basis) and
H3PO4 (Sigma Aldrich, 85 wt%). The product was vacuum filtered and
confirmed to be phase pure RbH2PO4 by X-ray powder diffraction (not
shown). A stoichiometric mixture of this RbH2PO4 and Rb2CO3 was
placed in a quartz boat and heated in a furnace to 110 ꢀC. An inlet gas
stream of water vapor with pH2O ¼ 0.83 atm, achieved by flowing O2
through a water bubbler at 95 ꢀC, was then supplied and the sample held
at this condition for approximately 30 min. The precursors quickly un-
derwent deliquescence to form an aqueous solution (condition B).
Following the initial equilibration and while maintaining the high water
vapor, the furnace temperature was raised to 150 ꢀC at a heating rate of 2
ꢀC/min and held for 24 h (condition C), eventually generating the crys-
talline product. In the final step, the supply of water vapor was turned off,
and the apparatus was allowed to cool in flowing, nominally dry O2 to
room temperature. The product was completely dry and formed of sub-
millimeter transparent crystals in a bed of white powder, identified in
subsequent experiments to be material of the same phase. The material
was stored in a dry 150 ꢀC oven until subsequent characterization.
environments. For measurements under dry conditions, samples were
directly heated to 500 ꢀC at 0.5 ꢀC/min. For measurements under hu-
midity, the desired level of humidification (pH2O ¼ 0.22, 0.43, or 0.82
atm, balance Ar) was introduced at 150 ꢀC, and the samples were
equilibrated for 1 h before further heating to 500 ꢀC at 0.5 ꢀC/min. The
recorded data were smoothed to remove artifacts due to periodic H2O
condensation.
4. Results and discussion
The crystals obtained from the high-temperature, high-humidity
crystallization process were found to be Rb3(H1.5PO4)2. Crystal data and
structure refinement information are listed in Table 1. Atomic parameters
and anisotropic displacement parameters are provided in Tables 2 and 3,
respectively. Rb3(H1.5PO4)2 adopts the C2/m space group, identical to
Cs3(H1.5PO4) [9]. Excluding proton positions, both compounds are iso-
structural to Cs3H(SeO4)2 [17–19]. The X-ray powder diffraction pattern
(collected at 150 ꢀC) is fully accounted for by the structure obtained from
the low temperature single crystal structure determination, Fig. 2(a). Cell
parameters from the powder data refined to a ¼ 10.668(2), b ¼
6.1326(3), c ¼ 7.5420(8) Å, and β ¼ 108.068(3) ꢀ, indicating a 2.6% cell
volume contraction between 150 ꢀC and ꢁ173 ꢀC. Powder diffraction
measurements at room temperature, Fig. 2(b), revealed a gradual trans-
formation to an unidentified crystalline hydrate phase. Exposure to 150
ꢀC recovered the Rb3(H1.5PO4)2 phase. It is this hydration behavior that
required storage (or treatment) of the material at 150 ꢀC prior to
characterization.
As shown in Fig. 3, the Rb and PO4 ions in Rb3(H1.5PO4)2 are arranged
in a pseudo-hexagonal arrangement, a feature typical of M3H(XO4)2
compounds [20]. Each of the crystallographically distinct Rb cations is
surrounded by a coordination polyhedron of ten oxygen atoms, with Rb –
O bond distances ranging from 2.83 to 3.45 Å, Table 4. The single,
crystallographically distinct PO4 tetrahedron is slightly distorted,
Table 5, with bond distances ranging from 1.50 to 1.57 Å and angles
deviating by no more than 3ꢀ from the ideal value of 109.5ꢀ.
The hydrogen atoms, H(1) and H(2), are associated with the oxygen
atoms of the two longer P – O bonds, in agreement with the reported
correlation between hydrogen bond lengths and P – O bonds distances of
H2PO4
ꢁ
3(H1.5PO4) [9], the phosphate groups are linked
anions [21]. As in Cs
3. Characterization
Table 1
Crystal data and structure refinement information for Rb3(H1.5PO4)2.
Single crystal X-ray diffraction (SCXRD) data were collected at ꢁ173
ꢀC (100 K) under flowing nitrogen using a Bruker APEX II diffractometer
equipped with an Incoatec Microfocus Mo Kα source and HELIOS MX
Empirical formula
Rb3H3O8P2
multilayer optics. The selected crystal was found to display non-
merohedral twinning (i.e., was an intergrowth of two crystals with
random orientation with respect to one another). A multi-scan absorption
correction was applied using SADABS and twinning was accounted for
using TWINABS. The crystal structure was solved by direct methods using
SHELXS in the OLEX2 GUI and refined with SHELXL [13–15]. Powder
X-ray diffraction (PXRD) was performed at the growth temperature of
150 ꢀC and at ambient temperature using a Rigaku Ultima IV diffrac-
Formula weight
Temperature
Crystal system
Space group
449.37 amu
100 K
Monoclinic
C2/m
Unit-cell dimensions
a ¼ 10.611(2) Å
b ¼ 6.0390(9) Å
c ¼ 7.5221(11) Å
α
¼ 90ꢀ
ꢀ
β ¼ 108.534(4)
γ ¼ 90ꢀ
tometer (Cu Kα source) fitted with a custom, high-temperature stage [9].
Volume
Z
Density (calculated)
Absorption coefficient
F(000)
456.99(12) Å3
2
In both cases data were collected under ambient laboratory atmosphere
(~25 ꢀC, ~35% relative humidity). Rietveld refinement of the structure
model was performed using the program GSAS-II [16]. Parameters
refined were background, scale factor, sample displacement, Lorentzian
broadening, spherical harmonic preferred orientation, lattice constants,
and isotropic displacements, with a single Uiso for all elements of the
same type.
Simultaneous TGA and DSC of Rb3(H1.5PO4)2 was performed using a
Netzsch STA 449 F3 equipped with a water vapor furnace for controlling
pH2O in the range 0–1 atm. Humidification was obtained by injecting
steam into Ar. Samples ~30 mg in mass were heated to 150 ꢀC and held
for 1 h under flowing dry Ar to remove surface absorbed water. Thermal
profiles were then measured under both dry and humidified
3.266 g/cm3
16.360 mmꢁ1
416.0
Instrument
Radiation
Bruker APEX II
Mo Kα (λ ¼ 0.71073)
2Θ range for data collection
Index ranges
Reflections collected
Independent reflections
Data/restraints/parameters
Goodness-of-fit on F [2]
5.712–74.248ꢀ
ꢁ17 ꢂ h ꢂ 16, 0 ꢂ k ꢂ 10, 0 ꢂ l ꢂ 12
1259
1259 [Rint ¼ 0.0779, Rsigma ¼ 0.0439]
1259/0/42
1.041
R1 ¼ 0.0371, wR2 ¼ 0.0828
R1 ¼ 0.0513, wR2 ¼ 0.0888
2.12/-1.69
Final R indexes [I ꢃ 2
σ
(I)]
Final R indexes [all data]
Largest diff. peak/hole/e Åꢁ3
3

S. Sanghvi, S.M. Haile
Journal of Solid State Chemistry 296 (2021) 121951
Table 2
Atomic coordinates, site, occupancy, and equivalent isotropic displacement (Ueq) parameters for Rb3(H1.5PO4)2 at ꢁ173 ꢀC. (C2/m a ¼ 10.611(2) Å, b ¼ 6.0390(9) Å, c
¼ 7.5221(11) Å, β ¼ 108.534(4) ꢀ) Ueq is defined as one-third of the trace of the orthogonalised Uij tensor.
Atom
Site
Occupancy
x
y
z
Ueq (Å2)
Rb(1)
Rb(2)
P
O(1)
O(2)
O(3)
H(1)
H(2)
2a
4i
4i
8j
4i
4i
4h
2b
1
1
1
1
1
1
1
1
0
0.5
0
0
0.2100(3)
0
0
0.5
0.01034(11)
0.01062(10)
0.0096(2)
0.0110(3)
0.0139(5)
0.0118(5)
0.14(5)
0.73815(4)
0.09974(11)
0.0156(2)
0.1096(3)
0.2332(3)
0
0.14084(5)
0.26466(14)
0.1741(3)
0.4749(4)
0.235(4)
0
0.19(2)
0
0
0.5
0.09(5)
Table 3
Anisotropic displacement parameters for Rb3(H1.5PO4)2 at ꢁ173 ꢀC.
Atom
U11 (Å2)
U22 (Å2)
U33 (Å2)
U23 (Å2)
U13 (Å2)
U12 (Å2)
Rb(1)
Rb(2)
P
O(1)
O(2)
O(3)
0.0120(3)
0.0119(2)
0.0093(5)
0.0125(10)
0.0127(15)
0.0096(14)
0.0083(2)
0.00936(14)
0.0109(4)
0.0096(7)
0.0207(13)
0.0140(11)
0.0105(2)
0.0119(2)
0.0084(4)
0.0112(8)
0.0085(11)
0.0133(12)
0
0
0
0.0034(2)
0.00567(13)
0.0026(3)
0.0040(7)
0.0033(10)
0.0058(10)
0
0
0
ꢁ0.0014(6)
0.0027(6)
0
0
0
0
Fig. 2. PXRD patterns collected from Rb3(H1.5PO4)2: (a) at 150 ꢀC and (b) under ambient conditions of ~25 ꢀC and ~35% relative humidity. In (a) the pattern is fully
described by the structure determined for Rb3(H1.5PO4)2 at ꢁ173 ꢀC, as indicated by the agreement between the measured (black) and calculated (red) patterns. In (b)
evolution of the patterns is evident, with the peaks of a secondary phase (*), presumed to be a hydrate, growing after 20 min of exposure to ambient conditions. (For
interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
by hydrogen bonds to form changes that extend along [001]. The chain
linkages alternate between a single P–O(2)-H(2)-O(2)-P bond and a pair
of P–O(1)-H(1)-O(1)-P bonds. Thus, there are three such bonds per
phosphate group and the O(1) and O(2) atoms simultaneously serve as
donor and acceptor in their respective hydrogen bonds. Refinement of
the proton positions suggested these atoms both reside at
single-minimum, high symmetry positions equidistant from the oxygen
atoms, Table 6. While the O(2) … O(2) distance of 2.472(7) Å is
consistent with such a configuration, the O(1) … O(1) distance of
2.533(5) Å is atypically long for a single-minimum hydrogen bond [22,
23], particularly given the low temperature of the measurement. The
result also contrasts what was found in Cs3(H1.5PO4)2, in which a
double-minimum hydrogen bond was resolved at this location [9]. The
structure solution here, which also yields unusually large displacement
parameters for the protons, may be influenced by the twinned nature of
the crystal employed in the measurement, and also points to the diffi-
culties that can be encountered when trying to use x-ray diffraction to
obtain proton positions with high accuracy in compounds containing
heavy elements.
presented in Fig. 4(a). Under dry conditions, the onset of Rb3(H1.5PO4)2
dehydration, indicated by mass loss, begins at ~ 215 ꢀC and reaches
completion at ~ 375 ꢀC. If dehydration proceeds in a manner similar to
that of Cs3(H1.5PO4)2, then the expected dehydration reaction is [8]:
2 Rb3(H1.5PO4)2 → 2RbPO3 þ Rb4P2O7 þ 3H2O
The theoretical weight loss of 6.0 wt% matches the measured weight
loss of 5.9 wt %. As humidity is increased, the onset of dehydration (Td) is
shifted to 255, 260, and 263 ꢀC for pH2O levels of 0.22, 0.43, and 0.82
atm, respectively. Moreover, the dehydration under these conditions
occurs via two clear steps, in contrast to the gradual mass loss evident
under dry conditions, with the first of these steps further resolving into a
two-step process at high pH2O. The DSC profiles recorded under active
humidification, Fig. 4(b), show strong thermal anomalies corresponding
to the temperatures of the dehydration events, with a similar upwards
shift in temperature with increasing humidification level and resolution
of the dehydration phenomenon into multiple steps. A close examination
of the DSC profiles just prior to the first dehydration event reveals a very
small thermal anomaly with an onset temperature of ~251 ꢀC. Because of
the weak intensity of this peak, complete characterization was not
The TGA profiles of Rb3(H1.5PO4)2 under various humidities are
4

S. Sanghvi, S.M. Haile
Journal of Solid State Chemistry 296 (2021) 121951
Table 5
Bond distances and angles in the phosphate tetrahedron of Rb3(H1.5PO4)2 at
ꢁ173 ꢀC.
Atom
Atom
Distance (Å)
Atom
Atom
Atom
Angle (ꢀ)
P
P
P
O(1)
O(2)
O(3)
1.575(2)
1.551(3)
1.502(3)
O(1)
O(1)
O(1)
O(2)
P
P
P
P
O(1)
O(2)
O(3)
O(3)
107.2(2)
106.63(10)
111.54(10)
112.9(2)
Table 6
Hydrogen bond environment of Rb3(H1.5PO4)2 at ꢁ173 ꢀC.
Atoms
Distance (Å)
Angle (ꢀ)
Donor(D)
Hydrogen(H)
Acceptor(A)
O
D/A – H
OD/A
OD/A
⋯
O – H – O
O(1)
O(2)
H(1)
H(2)
O(1)
O(2)
1.272(13)
1.236(4)
2.533(5)
2.472(7)
169(11)
180
Rb3(H1.5PO4)2. As discussed in the context of Cs3(H1.5PO4)2 [8,9], the
absence of a superprotonic transition in M3(H1.5PO4)2 compounds and
Rb3H(SO4)2 [5,25] may be connected to the relatively short O(2) … O(2)
bond and, in the case of the phosphates, the overall high number of
hydrogen bonds that stabilize the monoclinic structure.
5. Conclusions
In this work we demonstrate a humidity-controlled precipitation
process that enables the discovery of Rb3(H1.5PO4)2. At ambient tem-
perature and atmosphere (~25 ꢀC, ~35% relative humidity) the com-
pound transforms to an unidentified hydrate phase, presumably
contributing to the difficulties encountered in prior attempts to crystal-
lize Rb3(H1.5PO4)2 at temperatures below 150 ꢀC. While the synthesis
approach can be considered a variant of conventional aqueous precipi-
tation, it differs in that it exploits the chemical characteristics of highly
concentrated solutions. In particular, the liquid-vapor phase boundary of
such solutions is naturally, and potentially very strongly, shifted towards
high temperature relative to pure water. When combined with a high
level of humidity in the flowing gas stream, the precipitation/crystal
growth step can be carried out at temperatures much higher than 100 ꢀC,
providing access to water-soluble phases that may be inaccessible at
lower temperatures. Furthermore, though not demonstrated here, the
approach can, in principle, be expanded such that the oxygen or
hydrogen partial pressure in the humidified gas stream is also actively
regulated, providing additional control over the synthesis outcome. Hy-
drothermal synthesis provides access to similarly high temperatures and
even higher pressures, however, it is restricted to the growth of sparingly
soluble materials and, due to the fully sealed environment, is incom-
patible with independent control of gas-phase chemical potential [26].
The synthesis approach demonstrated here offers the additional advan-
tage of a relatively simple experimental set-up, requiring little more than
a high temperature oven and a heated water bubbler.
Fig. 3. The crystal structure of Rb3(H1.5PO4)2 shown in projection: (a) along the
c* direction, and (b) along the b direction. In (b) the labeled H1 atom occurs at
elevation y ¼ ꢁ0.19; another O1–H1–O1 bond, with the proton at elevation y ¼
0.19 (and identical x and z coordinates), also links the same pair of phosphate
groups. Such linkages, along with the O2–H2–O2 bonds, create a hydrogen
bonded chain of alternating single and double hydrogen bonds that extends
along c.
Table 4
Bond distances in the Rb–O coordination polyhedron of Rb3(H1.5PO4)2 at ꢁ173
ꢀC. Multiplicity of the bond is indicated by ꢄ #.
Atom
Atom
Distance (Å)
Rb(1)
Rb(1)
Rb(1)
Rb(2)
Rb(2)
Rb(2)
O(1)
O(2)
O(3)
O(1)
O(2)
O(3)
3.060(2) ꢄ 4
The new compound Rb3(H1.5PO4)2 discovered through this synthetic
process crystallizes in space group C2/m and is isostructural to
Cs3(H1.5PO4)2. In contrast to structurally and chemically similar com-
pounds like Cs3H(SeO4)2 and Rb3H(SeO4)2, Rb3(H1.5PO4)2 does not un-
dergo a phase transition to a trigonal superprotonic phase upon heating,
even under a steam partial pressure of 0.82 atm, under which dehydra-
tion is suppressed to a temperature of 263 ꢀC. The absence of a super-
protonic phase parallels the behavior of Cs3(H1.5PO4)2. A small thermal
anomaly is detected just prior to dehydration and may reflect the partial
exsolution of a minor phase, as has been observed in Cs3(H1.5PO4)2.
3.2629(13) ꢄ 4
2.898(3) ꢄ 2
3.013(2) ꢄ 2, 3.142(2) ꢄ 2, 3.446(2) ꢄ 2
2.828(3) ꢄ 1
2.938(3) ꢄ 1, 3.1058(8) ꢄ 2
possible. However, in contrast to the dehydration processes, it was
generally found (from a study of multiple samples, not shown) to be
invariant with pH2O, and thus not a result of loss of H2O from the
structure. Given the small thermal signature, it is unlikely that this
thermal event corresponds to a transformation to a superprotonic
trigonal phase, as is known in several M3H(SeO4)2 analogs [24]. In
Cs3(H1.5PO4)2, heating induces a partial exsolution of a small amount of a
CsH2PO4-like phase, and it is possible a similar phenomenon occurs in
CRediT authorship contribution statement
Sheel Sanghvi: Conceptualization, Investigation, Writing - original
5

S. Sanghvi, S.M. Haile
Journal of Solid State Chemistry 296 (2021) 121951
Fig. 4. Thermal analysis of Rb3(H1.5PO4)2 under the atmospheres indicated: (a) TGA, and (b) DSC profiles. Inset in (a) highlights the mass loss behavior between 240
and 285 ꢀC, whereas that in (b) highlights the thermal behavior between 245 and 260 ꢀC.
[9] S. Sanghvi, S.M. Haile, Crystal structure, conductivity, and phase stability of
Cs3(H1.5PO4)2 under controlled humidity, Solid State Ionics 349 (2020) 115291.
[10] A.A. Gaydamaka, V.G. Ponomareva, I.N. Bagryantseva, Phase composition, thermal
draft, Writing - review & editing. Sossina M. Haile: Supervision, Writing
- review & editing, Funding acquisition.
and transport properties of the system based on the mono- and dihydrogen
phosphates of rubidium, Solid State Ionics 329 (2019) 124–130.
Declaration of competing interest
[11] C. Panithipongwut, S.M. Haile, High-temperature phase behavior in the
Rb3H(SO4)(2)-RbHSO4 pseudo-binary system and the new compound
Rb5H3(SO4)(4), Solid State Ionics 213 (2012) 53–57.
[12] K.A.B.S. Dill, Molecular Driving Forces: Statistical Thermodynamics in Biology,
Chemistry, Physics, and Nanoscience. Garland Science, vol. 2, Taylor & Francis Group,
2010.
[13] O.V. Dolomanov, L.J. Bourhis, R.J. Gildea, J.A.K. Howard, H. Puschmann, OLEX2: a
complete structure solution, refinement and analysis program, J. Appl. Crystallogr.
42 (2) (2009) 339–341.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgements
[14] G. Sheldrick, Crystal structure refinement with SHELXL, Acta Crystallogr. C 71 (1)
(2015) 3–8.
Financial support has been provided by National Science Foundation
(DMR 1807234). SCXRD experiments made use of the Integrated Mo-
lecular Structure Education and Research Center (IMSERC) at North-
western University, which has received support from SHyNE, the State of
Illinois and International Institute for Nanotechnology (IIN). The authors
thank Charlotte Stern for single crystal data collection.
€
[15] G. Sheldrick, SADABS, Version 2.10, University of Gottingen, Germany, 2003.
[16] B.H. Toby, R.B. Von Dreele, GSAS-II: the genesis of a modern open-source all
purpose crystallography software package, J. Appl. Crystallogr. 46 (2) (2013)
544–549.
[17] R. Sonntag, R. Melzer, T. Wessels, P.G. Radaelli, Cs3H(SeO4)2 at 300K by high-
resolution neutron powder diffraction, Acta Crystallogr. C 53 (11) (1997)
1529–1531.
[18] A.I. Baranov, B.V. Merinov, A.B. Tregubchenko, L.A. Shuvalov, N.M. Shchagina,
Phase transitions, structure, protonic conductivity and dielectric properties of
Cs3H(SeO4)2 and Cs3(H, D)(SeO4)2, Ferroelectrics 81 (1) (1988) 187–191.
[19] B. Merinov, N. Bolotina, A. Baranov, L. Shuvalov, Crystal structure of Cs3H(SeO4)2
(T¼ 295 K) and its changes in phase transformations, Kristallografiya 33 (6) (1988)
1387–1392.
[20] D. Yi, S. Sanghvi, C.P. Kowalski, S.M. Haile, Phase behavior and superionic
transport characteristics of (MxRb1–x)3H(SeO4)2 (M ¼ K or Cs) solid solutions,
Chem. Mater. 31 (23) (2019) 9807–9818.
[21] M. Ichikawa, Dependence of the distortion of the tetrahedra in acid phosphate
groups HnPO4 (n ¼ 1-3) on hydrogen-bond length, Acta Crystallogr. Sect. B-Struct.
Commun. 43 (1987) 23–28.
[22] M. Ichikawa, O-H vs. O...O distance correlation, geometric isotope effect in OHO
bonds, and its application to symmetric bonds, Acta Crystallogr. Sect. B Struct. Sci.
34 (JUL) (1978) 2074–2080.
[23] I.-H. Oh, Y.J. Sohn, M. Meven, G. Heger, Symmetrical hydrogen bond in
SO4–H–SO4 dimer in K3H(SO4)2 by neutron diffraction, J. Phys. Soc. Jpn. 88 (12)
(2019), 124601.
[24] A. Baranov, Crystals with disordered hydrogen-bond networks and superprotonic
conductivity, Review. Crystallography Reports 48 (6) (2003) 1012–1037.
[25] L.A. Cowan, R.M. Morcos, N. Hatada, A. Navrotsky, S.M. Haile, High temperature
properties of Rb3H(SO4)2 at ambient pressure: absence of a polymorphic,
superprotonic transition, Solid State Ionics 179 (9) (2008) 305–313.
[26] A. Rabenau, The role of hydrothermal synthesis in preparative chemistry, Angew
Chem. Int. Ed. Engl. 24 (12) (1985) 1026–1040.
References
[1] D.A. Boysen, S.M. Haile, H. Liu, R.A. Secco, High-temperature behavior of CsH2PO4
under both ambient and high pressure conditions, Chem. Mater. 15 (3) (2003)
727–736.
[2] S.M. Haile, C.R.I. Chisholm, K. Sasaki, D.A. Boysen, T. Uda, Solid acid proton
conductors: from laboratory curiosities to fuel cell electrolytes, Faraday Discuss 134
(2007) 17–39.
[3] R.B. Merle, C.R.I. Chisholm, D.A. Boysen, S.M. Haile, Instability of sulfate and
selenate solid acids in fuel cell environments, Energy Fuels 17 (1) (2003) 210–215.
[4] T. Uda, D.A. Boysen, S.M. Haile, Thermodynamic, thermomechanical, and
electrochemical evaluation of CsHSO4, Solid State Ionics 176 (1) (2005) 127–133.
[5] A. Ikeda, S.M. Haile, The thermodynamics and kinetics of the dehydration of
CsH2PO4 studied in the presence of SiO2, Solid State Ionics 213 (2012) 63–71.
[6] Y.K. Taninouchi, T. Uda, Y. Awakura, A. Ikeda, S.M. Haile, Dehydration behavior of
the superprotonic conductor CsH2PO4 at moderate temperatures: 230 to 260
degrees C, J. Mater. Chem. 17 (30) (2007) 3182–3189.
[7] D.A. Boysen, T. Uda, C.R.I. Chisholm, S.M. Haile, High-performance solid acid fuel
cells through humidity stabilization, Science 303 (5654) (2004) 68.
[8] V. Ponomareva, I. Bagryantseva, B. Zakharov, N. Bulina, G. Lavrova, E. Boldyreva,
Crystal structure and proton conductivity of a new Cs3(H2PO4)(HPO4).2H2O phase
in the caesium di- and monohydrogen orthophosphate system, Acta Crystallogr. C
73 (10) (2017) 773–779.
6