On the nature of calcium magnesium acetate road deicer

Article abstract of DOI:10.1016/j.jssc.2018.10.041

CMA (calcium magnesium acetate) road deicers have gained popularity in recent years as an environmentally friendly alternative to traditional rock salt, and as an industrial adsorbent for removing H₂S and other odorous acid gases from gas streams. Despite its increasing commercial use, its exact composition and crystalline structure remained unknown, with subsequent problems in evaluating properties of commercial CMA. Various synthetic routes towards formation of crystalline calcium magnesium acetates were investigated. From aqueous solutions preferential formation of calcium monohydrates or calcium monohydrate acetic acid solvates is observed. Crystals of genuine mixed metal calcium-magnesium acetate were obtained from hot glacial acetic acid. Material suitable for analysis by single crystal X-ray diffraction, SC-XRD, was obtained by slight reduction of solvent volume at 80 °C for several hours. CMA crystallizes in the orthorhombic space group Pnma with a formula of Mg₂Ca(OAc)₆ (OAc = acetate anion), with a magnesium to calcium ratio of two to one. Under the same conditions, but in the absence of magnesium, the acetic acid solvate of calcium acetate, Ca(OAc)₂(HOAc), is obtained, which is also described. Multicrystalline XRD and EDS analysis data of ground CMA samples match those of commercial CMA. Single crystal structural analysis finds an unusually large unoccupied void space of 9.4% of the unit cell volume. Thermal gravimetric analysis, TGA, gives an upper limit of one water molecule per formula unit of CMA, leaving the void space at least partially unoccupied. This helps to better understand CMAs unusually low density, which had been an issue when used as a commercial road deicer, having been described as being easily blown off road surfaces when applied in crystalline or powder form.

Full text of DOI:10.1016/j.jssc.2018.10.041

Journal of Solid State Chemistry 270 (2019) 1–10
Contents lists available at ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
On the nature of calcium magnesium acetate road deicer
b
c
d
c
c
Jennifer R. Miller , Matthew J. LaLama , Rachel L. Kusnic , Darian E. Wilson , Paije M. Kiraly ,
c
a,c,⁎
Samuel W. Dickson , Matthias Zeller
a
Purdue University, Department of Chemistry, 560 Oval Drive, West Lafayette, IN 47907-2084, United States
Pennsylvania State University, Department of Chemistry, University Park, PA 16802, United States
Youngstown State University, Department of Chemistry, 1 University Plaza, Youngstown, OH 44555, United States
Materials Research Laboratories, 290 North Bridge Street, Struthers, OH 44471, United States
b
c
d
A R T I C L E I N F O
A B S T R A C T
Keywords:
CMA (calcium magnesium acetate) road deicers have gained popularity in recent years as an environmentally
Calcium magnesium acetate
Calcium acetate acetic acid
Road deicer
Crystal structure
Acetates
friendly alternative to traditional rock salt, and as an industrial adsorbent for removing H S and other odorous
acid gases from gas streams. Despite its increasing commercial use, its exact composition and crystalline
2
structure remained unknown, with subsequent problems in evaluating properties of commercial CMA. Various
synthetic routes towards formation of crystalline calcium magnesium acetates were investigated. From aqueous
solutions preferential formation of calcium monohydrates or calcium monohydrate acetic acid solvates is
observed. Crystals of genuine mixed metal calcium-magnesium acetate were obtained from hot glacial acetic
acid. Material suitable for analysis by single crystal X-ray diffraction, SC-XRD, was obtained by slight reduction
of solvent volume at 80 °C for several hours. CMA crystallizes in the orthorhombic space group Pnma with a
formula of Mg
2
Ca(OAc)
6
(OAc = acetate anion), with a magnesium to calcium ratio of two to one. Under the
(HOAc),
same conditions, but in the absence of magnesium, the acetic acid solvate of calcium acetate, Ca(OAc)
2
is obtained, which is also described. Multicrystalline XRD and EDS analysis data of ground CMA samples match
those of commercial CMA. Single crystal structural analysis finds an unusually large unoccupied void space of
9.4% of the unit cell volume. Thermal gravimetric analysis, TGA, gives an upper limit of one water molecule per
formula unit of CMA, leaving the void space at least partially unoccupied. This helps to better understand CMAs
unusually low density, which had been an issue when used as a commercial road deicer, having been described
as being easily blown off road surfaces when applied in crystalline or powder form.
1. Introduction
containing chlorides are well known to be highly corrosive towards
roadways, bridges, and vehicles, CMA has no known deleterious effects
Calcium magnesium acetate, commonly abbreviated as CMA, is an
[6]. Pure CMA is non-staining on aluminum surfaces [7].
environmentally-friendly alternative to rock salt (sodium chloride) and
other commonly used ice melts such as calcium chloride, magnesium
chloride, and potassium chloride [1,2]. Unlike traditional rock salt,
CMA is neither toxic nor caustic to humans and animals. It has no
known adverse effects on plants, therefore eliminating roadside
vegetation destruction and soil degradation [3,4]. Its poor mobility in
soil reduces issues with groundwater contamination, which is a concern
with its more mobile counterparts. CMA has high efficacy at lower
temperatures due to its eutectic temperature of − 27 °C which is nearly
twenty degrees lower than that of rock salt [5]. Lastly, while ice melts
A drawback of CMA is its price, with costs being about twenty times
that of rock salt [8,9]. Despite this financial disadvantage, CMA is
commonly used as an ice melt on bridges, roadways, parking garages,
airport runways, and other structures upon which corrosion inhibition
is a primary goal. Residentially, it is used as an environmentally- and
pet-friendly alternative to traditional ice melts. Industrially, it is also
tested as a material for removal of hydrogen sulfide in integrated
gasification combined cycle (IGCC) systems that are used to convert
coal into clean, usable fuel gas [10,11]. An essential step in this
conversion process is the elimination of the environmentally harmful
Abbreviations: OAc, acetate; CA, calcium acetate; CCDC, Cambridge Crystallographic Data Centre; CMA, calcium magnesium acetate; ICDD, International Centre for Diffraction Data;
MA, magnesium acetate; SC-XRD, Single Crystal X-ray diffraction; SEM-EDS, Scanning electron microscopy paired with energy dispersive X-ray spectroscopy; P-XRD, Powder Crystal
X-ray diffraction
⁎
Corresponding author at: Purdue University, Department of Chemistry, 560 Oval Drive, West Lafayette, IN 47907-2084, United States.
E-mail addresses: jrm575@psu.edu (J.R. Miller), mjlalama01@student.ysu.edu (M.J. LaLama), rlkusnic@mrllab.com (R.L. Kusnic), dewilson01@student.ysu.edu (D.E. Wilson),
pmkiraly@student.ysu.edu (P.M. Kiraly), swdickson@student.ysu.edu (S.W. Dickson), zeller4@purdue.edu (M. Zeller).
https://doi.org/10.1016/j.jssc.2018.10.041
Received 30 July 2018; Received in revised form 25 October 2018; Accepted 26 October 2018
Available online 29 October 2018
0022-4596/ © 2018 Elsevier Inc. All rights reserved.

J.R. Miller et al.
Journal of Solid State Chemistry 270 (2019) 1–10
sulfur containing gases such as H
2
S formed from sulfur contained in
observed that any large amount of water present inhibits the formation
of crystals of CMA, and instead preferentially various calcium acetate
derivatives are formed, such as one of the two known forms of calcium
the coal. CMA, and its magnesium free counterpart calcium acetate,
CA, have the ability to form lightweight hollow oxide particles (referred
to as cenospheric) when heated to high temperatures, which can
capture hydrogen sulfide by formation of calcium sulfide, CaS, which
can be oxidized to calcium sulfate for disposal or reuse. Effectiveness is
up to 90% at temperatures from 700 to 1100 °C. Essential for this
process to be effective is the shape and surface area of the cenospheric
particles, with thin and porous walls, which so far have been exclusively
observed for CMA and CA.
It is well known that exact chemical composition and molecular
structure are integrally tied to material properties and performance.
The knowledge of both is essential for understanding the underlying
processes that give a compound its unique characteristics. Without that
knowledge, any attempt to optimize a material's properties, improve its
manufacturing process, or verify and guarantee consistent quality
production are based on trial and error, and as such, are very
ineffectual and associated with unnecessary expense. CMA was first
proposed and tested as an alternative to rock salt as early as the mid-
monohydrate Ca(OAc)
acetic acid solvate, Ca(OAc)
2
(H
2
O)1+x [16,17], and calcium monohydrate
(H O)(HOAc) [18,19,20]. No crystalline
2
2
material containing any magnesium cations formed, with the exception
of magnesium acetate tetrahydrate [21,22,23] from experimental
setups that used no calcium ions. The inhibition of formation of
CMA in the presence of water is in conflict with several earlier reports
which claimed water to be an essential part of CMA, but it reproduces
observations from several later publications and patents, which claim
CMA to be essentially water free, and crystallizes most easily from
anhydrous solvents [12] and at high temperatures [5]. Following one of
these procedures [7], we were able to obtain crystalline CMA from
solutions in glacial acetic acid after removal of all residual water and
slight reduction of solvent volume by heating solutions to ca. 80 °C for
several hours. Sizes of obtained crystals varied, with some attempts
yielding crystals tens of microns in size and intergrown into conglom-
erates, but others yielding crystals up to one millimeter and larger. All
crystals had distinct blocky shapes and were clear and transparent.
Samples obtained from hot extracted commercial CMA and of samples
prepared from mixtures of calcium acetate and magnesium acetate (in
the form of Ca(OAc) ·H O and Mg(OAc) ·4H O) were indistinguish-
2 2 2 2
able, with similar blocky crystal habits, matching powder XRD
patterns, and identical unit cell sizes by single crystal XRD.
Analysis by single crystal XRD proved straightforward, and the
structure of CMA was established at both 100 K and room temperature.
No qualitative difference is observed between the structures at the two
temperatures, indicating the absence of any structural phase transfor-
mation in this temperature range. Data for 100 K is given in Table 1.
Data for 298 K are given in the Supporting material. CMA crystallizes
in an orthorhombic setting in space group Pnma, with four
1
980s. Since then, a large number of reports have investigated its
synthesis and production, properties, and performance for its various
uses. Despite its heavy commercial use and market penetration, CMA's
exact chemical makeup and structure have been unknown prior to this
investigation.
Previously reported descriptions of CMA vary in the ratio of calcium
to magnesium, with more reliable sources giving a 7:3 mol ratio for
magnesium to calcium [12], or a ratio of 2:1 in favor of magnesium.
Hoenke and Rynbrandt give the Mg:Ca to be between 8:2 and 6:4 [13].
Publications and patents also report differing water contents, ranging
from no water present to 3–4 mol water per mole magnesium [1,7,12].
The powder XRD pattern of a putative monohydrate, 2Mg(CH
3 2
COO) .
Ca(CH COO) ·2H O, was published in 1993, but was never verified in
3
2
2
any later publication [14]. Even the nature of CMA as a material
distinct from other calcium and magnesium acetates has not comple-
tely permeated into all publications and reports. As late as 1996, the US
Federal Highway Administration reported the composition of CMA as
2 6
Mg Ca(OAc) formula units per unit cell, Table 1. The ratio of
magnesium to calcium is two to one, in agreement with several of the
more recent reports. No water molecules are present in the material,
neither metal coordinated nor within interstitial voids, confirming the
emerging knowledge of recent years.
“
primarily a mixture of calcium and magnesium acetates, produced
with a 3/7 Ca/Mg ratio” [15]. This statement is in clear conflict with
thermogravimetric analysis and reported distinct powder XRD patterns
for pure CMA samples in earlier reports, such as in several patents
Calcium and magnesium cations are coordinated to acetate oxygen
atoms, forming an infinite three-dimensional lattice (Fig. 1).
Magnesium ions are six coordinated, with only a very slight distortion
from ideal octahedral. Two crystallographically independent magne-
sium ions are present in the crystal lattice. One, labelled Mg1, is located
on a crystallographic inversion centre and bisected by a twofold screw
axis (Fig. 2). The other, Mg2, is embedded in a crystallographic mirror
plane. Also located within this plane is one crystallographically unique
calcium ion. The calcium cation, being substantially larger than the
magnesium ions, features coordination to seven oxygen atoms (Fig. 2).
The coordination geometry does not match any of the standard
descriptors. It might be best described as “expanded trigonal antipris-
matic” (or “expanded octahedral”), with one set of three neighboring
oxygen atoms augmented by a forth oxygen atom to form a distorted
square, parallel opposite to the plane of the other three oxygen atoms of
[
7,12]. Over the last years a clearer picture has emerged, and it is now
accepted knowledge that CMA is a distinct and well defined phase that
stands separately from the other acetates of magnesium and calcium.
Dionysiou and coworkers [5], for example, have argued CMA to be an
anhydrous double salt with a 2:1 magnesium to calcium ratio that
preferentially forms at elevated temperatures and in the absence of
water (e.g. from hot anhydrous acetic acid). However, as in other
previous reports and patent application documents, phase identifica-
tion was limited to comparisons of powder X-ray patterns. Prompted
by the lack of exact structural data for CMA, we report herein on our
investigations and the elucidation of the exact solid state structure of
CMA by means of single crystal X-ray diffraction (SC-XRD).
the octahedron. The “not distorted side” is made up of the three oxygen
i
2
. Results and discussion
atoms O3, O7 and O3 of a MgO
6
octahedron. The four atoms at the
i
i
opposite side are O1 and O1 of a chelating acetate ion, and O5 and O5
of two monodentate acetate ions. The geometry as a whole is mirror
symmetric, with the calcium ion located on the mirror plane (the
symmetry operator i is +x, 1/2-y, +z).
Metal-oxygen bond distances in CMA are in the normal ranges, and
both Mg-O and Ca-O bond lengths are tightly clustered and range from
2.0317(9) to 2.1002(9) Å for Mg-O, and 2.3842(8) to 2.460(6) Å for
Ca-O bonds (Table 2). Bond angles around magnesium, reflecting the
close to octahedral symmetry, are around the expected 90° and 180°.
Deviations are largest for Mg2 to allow for bridging to calcium, with
distortions of up to 8.25° (for O3—Mg2—O7) (Table 2).
For structural analysis using single crystal X-ray diffraction, crystal-
lites ranging from 0.5 to 0.1 mm are commonly needed, although much
smaller crystals can be sufficient when using modern high brilliance
laboratory instruments or synchrotron X-ray sources. Reports on the
best conditions for the synthesis of pure samples of CMA varied, but
several reports and patents indicated that crystals of the required size
could be obtained. We approached the production of crystalline CMA
materials via two routes: independent synthesis from reaction of
magnesium and calcium acetates, and extraction and recrystallization
of commercial samples of CMA. During crystallization attempts (see
Materials and methods section and the Supporting material) we
The coordination modes of the acetate anions vary. Four acetate
2

J.R. Miller et al.
Journal of Solid State Chemistry 270 (2019) 1–10
Table 1
Single Crystal Structural Data for CaMg
2
(OAc)
3
and Ca(OAc)
2
(HOAc).
18CaMg
442.96
Sum formula, chemical formula
C
12
H
2
O
12, CaMg
2
(OAc)
3
6
C H
10CaO
6 2
, Ca(OAc) (HOAc)
M
r
218.22
Crystal system, space group
Temperature (K)
a, b, c (Å)
Orthorhombic, Pnma
100 )
10.8706 (5), 12.4318 (6), 15.6141 (7)
Monoclinic, C2/c
100
21.4782(11), 6.8537(3), 13.7260(8)
a
β (°)
V (Å )
90
115.452(2)
1824.43(16)
3
2110.11 (17)
Z
4
8
Radiation type
Mo Kα
0.41
Cu Kα
5.994
−1
μ (mm
)
Crystal size (mm)
0.28 × 0.25 × 0.13
0.24 × 0.13 × 0.03
Crystal shape, colour
Diffractometer
Radiation source, monochromator
plate, colourless
Bruker AXS D8 Quest CMOS diffractometer
I-μ-S microsource X-ray tube, Laterally graded multilayer
plate, colourless
Bruker AXS X8 Prospector CCD diffractometer
I-μ-S microsource X-ray tube, Laterally graded multilayer
(Goebel) mirror
(
Goebel) mirror
Absorption correction
Multi-scan, Apex2 v2014.1–1 (Bruker, 2014)
0.701, 0.749
Multi-scan, TWINABS (Sheldrick, 2012)
0.512, 0.753
Tmin, Tmax
No. of measured, independent and observed [I
2σ(I)] reflections
115110, 6877, 4979
15409, 2870, 2674
>
R
int
0.043
0.051
θ values (deg.)
θ
max = 45.3, θmin = 2.6
θmax = 66.8, θmin = 4.6
0.596
−
1
(sin θ/λ)max (Å
)
1.000
Range of h, k, l
R[F > 2σ(F )], wR(F ), S
No. of parameters
h = −17→20, k = −19→19, l = −31→24
0.039, 0.104, 1.04
140
h = −25→22, k = 0→8, l = 0→16
0.034, 0.097, 1.06
123
2
2
2
H-atom treatment
Δρmax, Δρmin (e Å
H-atom parameters constrained
0.88, − 0.91
H-atom parameters constrained
0.51, − 0.31
−3
)
CCDC
1845188
1845190
Fig. 1. Packing views of CMA along the three unit cell axes. The top left view is along the b-axis. Views at top right and bottom left are created by rotations around a and c, creating views
along c (top right) and along a (bottom left). Anisotropic displacement parameters are at 50% probability level. Colour coding: Ca (emerald green), Mg (lime green), O (red), C (grey), H
(white). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
3

J.R. Miller et al.
Journal of Solid State Chemistry 270 (2019) 1–10
Fig. 2. Calcium coordination environment, view along the mirror plane bisecting the Ca
atom. The symmetry operator i of the mirror plane is +x, 1/2-y, +z. The symmetry
ii
operator creating Mg2 (in the back) is −1/2+x, 1/2-y, 1/2-z. 50% probability level for
anisotropically shown atoms.
Table 2
Selected geometric parameters (Å, °). Symmetry equivalent distances and angles omitted.
O1—Mg1
O2—Mg1
O4—Mg1
O3—Mg2
2.0377 (6)
2.0600 (6)
2.0596 (7)
2.0930 (7)
2.0582 (7)
2.0317 (9)
180.0
86.67 (3)
93.33 (3)
180.0
88.89 (3)
89.66 (3)
91.11 (3)
90.34 (3)
180.0
93.24 (3)
175.80 (3)
98.04 (2)
81.11 (4)
O7—Mg2
O1—Ca1
O3—Ca1
O5—Ca1
O7—Ca1
2.1002 (9)
2.4260 (6)
2.4036 (6)
2.4119 (6)
2.3842 (8)
v
O5 —Mg2
vii
O6 —Mg2
iv
v
O1—Mg1—O1
O1—Mg1—O4
O1—Mg1—O4
O4—Mg1—O4
O1—Mg1—O2
O4—Mg1—O2
O1—Mg1—O2
O4—Mg1—O2
O2—Mg1—O2
O3—Mg2— O6
O3—Mg2— O5
O3—Mg2— O5
O7—Mg2— O6
O7—Mg2— O5
O3—Mg2—O7
O7—Ca1—O3
O7—Ca1—O5
O3—Ca1—O5
O3—Ca1—O5
O5—Ca1—O5
O3—Ca1—O1
O5—Ca1—O1
173.38 (4)
94.07 (3)
81.75 (3)
69.93 (2)
85.79 (2)
105.87 (2)
155.57 (2)
68.48 (3)
117.60 (2)
114.80 (2)
152.344 (17)
87.54 (2)
85.21 (2)
vi
iv
iv
iv
iv
3
Fig. 3. Depiction of the hydrophobic pockets surrounding four 84 Å sized voids (probe
3
radius 1.2 Å). The total void volume is 198.78 Å per unit cell, or 9.4% of the unit cell
i
volume.
i
iv
i
acetate anions is three-dimensional, but not equally dense throughout
the whole structure. Indeed, relatively large voids are left unoccupied
v
i
iv
v
O7—Ca1—O1
O3—Ca1—O1
O5—Ca1—O1
3
within the structure, 84 Å in size each, and thus easily large enough to
i
O3—Mg2—O3
house a water molecule, but too small for acetic acid. The voids are
however located in hydrophobic pockets, surrounded by acetate methyl
groups, and thus do not provide a suitable environment for water
molecules. The large unoccupied space within the structure of CMA
Symmetry code(s): (i) x, -y + 1/2, z; (iii) x-1/2, y, -z + 1/2; (iv) -x + 1, -y + 1, -z + 1; (v) x
1/2, y, -z + 1/2; (vi) x + 1/2, -y + 1/2, -z + 1/2; (vii) x + 1/2, -y + 1/2, -z + 1/2.
+
3
explains its quite low intrinsic density of only 1.394 g/cm . Densely
ions are crystallographically unique. One, that of O1, is bisected by the
packed magnesium acetate tetrahydrate, for example, features a
i
mirror plane and coordinates in a chelating manner via O1 and O1 to
3
density of 1.503 g/cm , despite missing the heavier calcium ions of
the calcium ion (symmetry operator i: +x, 1/2-y, +z). Both symmetry
equivalent oxygen atoms also coordinate in a bridging fashion to two
CMA. The lightweight nature of CMA has often been pointed out as a
problem when being used a road deicer, and it is often used in a
mixture with sand or in liquid form to avoid it to being blown off too
easily [24].
The voids divide the structure of CMA in two distinct sections,
taking up about half the unit cell volume each (Fig. 3). Rows stretching
along a line at (x, 1/4, 1/4) and (x, 3/4, 3/4) are densely packed and
dominated by calcium and magnesium cations and acetate oxygen
atoms coordinating to the metal ions. Parallel rows at (x, 3/4, 1/4) and
3
copies of Mg2, giving this acetate an overall chelating bridging μ :
1
2
1
η
Mg-η Ca-η Mg coordination mode. The acetate associated with O6 and
O7 is located within the same mirror plane, coordinated to the calcium
i
ion via O7 opposite of O1 and O1 . Both O6 and O7 are also
3
coordinated to each a copy of Mg2, giving this anion a bridging μ :
2
1
η
MgCa-η Mg coordination mode. The remaining two acetate ions,
2
1
3
located in general positions, have the same bridging μ : η MgCa-η Mg
coordination mode, with O3 and O5 being the bidentate oxygen atoms,
and O2 and O4 the monodentate donors. The carboxylate π-electrons
are not completely delocalized, with the mono and bidentate oxygen
atoms roughly corresponding to localized single and double bonds of
the acetate carboxylate groups, with bond distances of 1.2443(10),
(
3
x, 1/4, 3/4) are dominated by hydrophobic CH groups and surround
the voids, two along each row and four per unit cell. Individual voids
are not quite connecting with each other, but are only separated by
single methyl groups, nearly converting individual spherical voids into
continuous channels.
Despite the hydrophobic nature of the void space, one might
speculate that – under the right experimental circumstances – some
of the voids might be filled with water molecules, or other solvate
molecules, depending on the type of solvent used and the crystal-
1
.2450(10) and 1.2499(14) Å for C3-O2, C5-O4 and C7-O6 for the
shorter and 1.2780(10), 1.2747(10) and 1.2715(13) Å for the longer C-
O bonds. For the symmetric chelating acetate of O1, both C-O bonds
are of equal length, with an intermediate bond length of 1.2614(8) Å.
The overall structure that results from the bridging and chelating
4

J.R. Miller et al.
Journal of Solid State Chemistry 270 (2019) 1–10
lization conditions employed. If this is the case – while not observed
here – there might be indeed CMA batches with substantial water
content, as reported in some of the earlier reports on CMA.
similar. In a first step, variable amounts of loosely bound solvate
molecules are liberated at temperatures between 65 and slightly above
200 °C. This step is followed by conversion of CMA to CaCO
under liberation of one molecule of acetone, starting at between 275
and 350 °C, and finally decomposition of CaCO to CaO and CO occurs
3
and MgO
The presence of the void space not only provides the possibility to
host water or other solvate molecules, but also provides the structure
with a flexibility that would not be expected for a densely packed three-
dimensional metal-acetate framework. Many reports on CMA give a
range for the magnesium to calcium ratios rather than a hard fixed
number [13]. Under thermodynamic control, CMA will be expected to
crystallize in the ideal structure with a 2:1 Mg:Ca ratio as found here by
SC-XRD. All sites occupied by metal ions are clearly either calcium, or
magnesium, with no substitution of either cation in the position of the
other. I.e., there is no indication of formation of a solid solution for the
crystals analyzed, which had been obtained under thermodynamic
conditions as required for formation of large high quality single crystals
3
2
in the range between 580 and 800 °C. Variations between samples of
different origins are largest for the weight % loss below 200 °C,
indicating varying amounts of loosely bound solvate molecules, which
could be both physisorbed as well as potentially enclosed in the void
space of the crystal lattice. Samples prepared by Lu and co-workers
[30] lost up to 10 wt% in this temperature range, all below ca. 150 °C.
Han and Sohn [31] report a 4–5% weight loss before 150 °C, and an
additional similar weight loss just before the onset of acetate decom-
position. For samples prepared here (Fig. 4), the weight loss before
350 °C is minimal. Less than 1.05% by weight was lost for samples
predried for one hour at 80 °C in a slight nitrogen stream. Two minor
weight loss steps are observed. A first loss of about 0.6% between onset
of heating and 75 °C, and a second weight loss between 300 and 345 °C,
of ca. 0.5%. Between 75 and 300 °C the weight remains essentially
constant (0.11% weight loss). We attribute the first weight loss, below
75 °C, to desorption of residual physisorbed solvent (acetic acid or
water). The second step immediately precedes the onset of conversion
(
high temperatures and slow cooling). Batches grown under more
kinetic conditions, on the other hand, might have substantial numbers
of magnesium positions being occupied by calcium, especially when the
crystallization happens in the presence of an excess of calcium as is the
case when producing CMA by reaction of acetic acid with dolomite as
the metal ion source, which features a 1:1 ratio of Mg to Ca. Thanks to
the large void spaces, the crystal lattice should be able to absorb the
addition of an excess of voluminous calcium with relative ease.
Evidence for the flexible nature of the CMA framework is provided
by the acetonitrile solvate polymorph of its manganese derivative,
3
of CMA into CaCO and MgO, and thus might indicate liberation of
solvate molecules from within the weakening crystal lattice. If due to
loss of water the weight loss would translate to about one mole of water
3
Mn(OAc)
2
. Solvent free manganese acetate crystallizes in various
per formula unit of CMA, liberated from either from the 84 Å voids, or
tetragonal [25], orthorhombic [26] or monoclinic [27] settings differ-
ent from that of CMA. Its acetonitrile solvate [28], however, is isotypic
from larger physical voids and defects inside individual crystals. Based
on the single crystal data, which show no residual electron density
within the voids we believe that the latter appears to be more likely.
This is also supported by the hydrophobic nature of these voids,
making them unlikely hosts for polar water molecules.
The largest weight loss is observed between 345 and 500 °C, which
gradually transitions into a second major weight loss above 500 °C that
ends just below 700 °C. No exact plateau is observed between these two
major decomposition steps, indicating that decomposition of acetate
into magnesium oxide and calcium carbonate, and decomposition of
carbonate to oxide overlap to a certain degree. The remaining weight at
500 °C is slightly higher than what would be observed for clean
to CMA, with the CH
unoccupied in CMA. Lattice parameters of the Mn(OAc)
3
-CN molecules hosted in the void space left
solvate are
0.788(2), 12.702(2) and 15.784(3) and its unit cell volume is
2
1
2
3
3
3
162.869 Å at 193 K, vs 2110.11 (17) Å at 100 K and 2148.7(4) Å
at 298 K for solvent free CMA. In the manganese salt, all metal sites are
of course taken up by only one kind of metal, Mn(II), which features an
ion radius about halfway in between that of Mg and Ca. (ionic radii: Mg
0
.72, Ca 1.0 (six fold coordination), Mn(II)hs: 0.83) [29] This indicates
that the CMA framework may easily accommodate swapping of some
magnesium with larger calcium, or vice versa.
The presence of substantial void space within the crystal structure
of CMA prompted us to investigate if solvate molecules might be
trapped within the crystal lattice. Single crystal diffraction did not
indicate the presence of any substantial electron density within the
voids (shown in Fig. 3), with the largest difference electron density
found within the void space less than 0.43 electrons per cubic
Ångström. A Squeeze analysis, using the program Platon, revealed
3
decomposition to CaCO and 2 MgO (found: 43.0%, expected 40.8%),
indicating complete decomposition of all acetate is kinetically delayed.
The final weight at above 700 °C is in agreement with formation of
calcium and magnesium oxide (found 32.0%, expected 30.9%).
3
four electrons within each of the 84 Å voids.
Data from just a small number of well-developed single crystals
might however not be representative of the bulk of the material. To
obtain more reliable data the crystals were also analyzed by thermal
gravimetric analysis, TGA. CMA had been analyzed by TGA previously
[
30,31,32]. However, analysis of the investigated samples had been
lacking in most cases and materials were only poorly characterized.
CMA samples prepared by Lu and co-workers [30] were based on a
calcium to magnesium molar ratio of 2.333:1 (the invers of the ratio we
found by XRD and suggested by most other reports), and samples were
analyzed only by thermal decomposition techniques. Han and Sohn
[
31] used a commercial CMA sample (a Chevron ICE-B-GON Deicer) of
ca. 96% purity (the remainder being stated to be mostly inert silicate
impurities). They determined the atomic ratio of Ca to Mg using direct
current plasma techniques to be 0.3:0.7 in favor of magnesium, in
agreement with most literature data. Adánez and coworkers used CMA
of 95% purity, obtained from Cryotech Deicing Technology. They did
not report any analysis of the material prior to TGA measurements.
Observations made by all three groups and by us (Fig. 4) are
Fig. 4. Thermal Gravimetric analysis data for CMA with selected weight loss numbers.
Crystals grown from hot acetic acid were dried in vacuum, ground into a fine powder and
pre-dried at 80 °C for one hour prior to measurement. Green curve: actual weight (in % of
starting weight, 4.744 mg). Blue curve: derivative weight (% change / temperature). (For
interpretation of the references to color in this figure legend, the reader is referred to the
web version of this article).
5

J.R. Miller et al.
Journal of Solid State Chemistry 270 (2019) 1–10
Fig. 5. Powder-XRD patterns of CMA as synthesized (ground single crystals) and of commercial CMA. Peaks at 5.2 and 7.4 degrees indicated with stars * most likely originate from
traces of Ca(OAc) hemihydrate (powder database entries ICDD 00-014-0792 [18] and 00-039-0199 [33]) or Ca(OAc) dihydrate (CCDC entry 00-039-0529 [34]), for which no
2
2
structural data have so far been reported.
To test whether the structure obtained for CMA is indeed equivalent
to samples described previously and sold commercially, we compared
powder-XRD patterns of CMA as synthesized and of commercial CMA,
Fig. 5. Within the margin of experimental error, patterns are identical.
Fig. S2 and S3 in the Supporting information show reproductions of
typical powder XRD pattern of CMA, which also match those of
synthesized CMA. Fig. 6 depicts a powder pattern obtained from a
ground commercial sample of CMA and its fit by Rietveld refinement
against the model from single crystal diffraction. The fit indicates the
Inc., N8336 Deadwood Point Beach Road, Fond du Lac, WI 54937)
yielded a Mg:Ca ratio of 1.82:1.00 (Fig. S4), thus slightly depleted in
magnesium when compared to the expected 2:1 ratio for pure
CaMg
could be admixtures of more calcium rich phases. This is indicated by
the apparent presence of Ca(OAc) hemihydrate or Ca(OAc) dihydrate,
in both commercial as well as synthesized CMA (Figs. 3 and 4). No
structural data are available for either Ca(OAc) hemihydrate (powder
database entries ICDD 00-014-0792 [18] and 00-039-0199 [33]) or
Ca(OAc) dihydrate (ICDD entry 00-039-0529 [34]), hindering quanti-
2 3
(OAc) , as obtained from SC-XRD. One possible explanation
2
2
2
presence of trace amounts of either Ca(OAc)
database entries ICDD 00-014-0792 [18] and 00-039-0199 [33]) or
Ca(OAc) dihydrate (CCDC entry 00-039-0529) [34], but otherwise the
experimental pattern and Rietveld fit do match well. For both Ca(OAc)
0.5H O as well as Ca(OAc) × 2H O no structural data have so far been
reported, preventing phase quantification by Rietveld methods [35].
2
hemihydrate (powder
2
fication by Rietveld analysis. The slight misfit observed by EDS analysis
could also indicate substitution of some magnesium sites by calcium due
to the apparently flexible nature of the CMA crystal lattice, as discussed
above. Thus, while the data obtained by SEM-EDS and by SC-XRD are in
general agreement, there is no absolute proof that there is no variation of
the Ca:Mg ratio possible in CMA.
A detailed analysis of the spread of Mg to Ca ratios in CMA was
beyond the scope of this investigation. In a more limited analysis we
tested whether the structure of CMA can tolerate the complete absence
of magnesium in its lattice by recrystallizing calcium acetate under the
same conditions as used to obtain crystalline CMA, but in the absence
2
2
×
2
2
2
To verify that commercially sold samples of CMA not only have a
similar structural arrangement, but also match the 2:1 ratio of magne-
sium to calcium found in the single crystal samples, we further analyzed
a commercial sample using scanning electron microscopy paired with
energy dispersive X-ray spectroscopy (SEM-EDS). Analysis of a typical
sample (“Happy Paws™ Ice Melt”, The Green Earth Deicer Company,
Fig. 6. Rietveld refinement of a sample of CMA as synthesized (ground single crystals). Blue: experimental pattern. Red: Rietveld fit. Green: tick marks of indexed symmetry allowed
reflections. For peaks indicated with stars *, see Fig. 5. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
6

J.R. Miller et al.
Journal of Solid State Chemistry 270 (2019) 1–10
of magnesium ions. Calcium acetate hydrate was dissolved in acetic
acid and heated to 80 °C for several hours until all solvent was
completely boiled off. The process was repeated to remove all traces
of water, and the residue was recrystallized from hot acetic acid by slow
cooling. The crystals obtained have a distinctly different shape from
CMA: clusters of heavily intergrown thin platelets rather than well-
defined blocks. Single crystal analysis of the crystals revealed a material
structurally and chemically not related to CMA (Fig. 7 and 8). Instead,
a layered structure being composed of calcium ions surrounded by both
2
acetate as well as acetic acid, with formula Ca(OAc) ·HOAc was
formed. Similar to CMA, this compound is also not unknown, a
matching powder X-ray pattern has been reported as far back as
1
958 (ICDD: 00-010-0781) [18], but again the actual structure had
never been determined or published.
A possible reason for the absence of published structural data for
2
Ca(OAc) ·HOAc could be non-merohedric twinning that its crystals
consistently suffer from. The twinning (see SI for details) leads to a
diffraction pattern that can be easily mistaken for a higher symmetry
larger repeat unit, which could have hampered correct indexing of its
diffraction pattern, even in the presence of otherwise well behaved
single crystals.
Fig. 8. View of calcium acetate acetic acid, Ca(OAc) (HOAc), showing the layered nature
2
In its crystal structure, determined from a twinned crystal, one
of the material. View down the b-axis direction. The probability level is 50% for
2
+
anisotropically shown atoms.
single crystallographically unique Ca ion is seven coordinated to
oxygen atoms of acetate ions and acetic acid molecules ( Fig. 7). The
coordination environment of calcium is best described as a pentagonal
bipyramid. Five oxygen atoms of acetate ions and acetic acid are
arranged in an equatorial belt around the metal ion in a nearly planar
arrangement, and two oxygen atoms of a second type of acetate ion are
taking up the two axial positions of the bipyramid.
3
The C-O bond distances of the μ chelating acetate are virtually
identical, with bond lengths of 1.261(3) and 1.262(3) Å. Those of the
second carboxylate are slightly differentiated, due to the presence of the
hydrogen bond originating from the acetic acid molecule, and the C-O
bond of O4, the H-bond acceptor, is with 1.276(2) Å longer than that of
O3 with 1.245(2) Å.
As for CMA, the coordination modes of the acetate anions vary. There
are two crystallographically unique acetate ions. One, that of O1 and O2,
2
Individual layers in the structure of Ca(OAc) ·HOAc are only
2
loosely connected with each other. The singly bound acetic acid
molecules protrude furthest from the layers, and they do interdigitate
with their counterparts in the next layer, preventing slippage of layers
against each other and giving the structure three-dimensional stability
is chelating the calcium ion in an η fashion. Both O1 and O2 also
coordinate to each a second calcium ion, giving this acetate an overall
1
2
1
3
chelating bridging μ : η Ca-η Ca-η Ca coordination mode and connecting
the metal ions into double ribbons that stretch along the c-axis direction.
The oxygen ions of this acetate ion are both located in the equatorial belt
of the pentagonal bipyramid, contributing four of the oxygen atoms, with
the fifth being the keto oxygen atom O5 of the acetic acid molecule. The
oxygen atoms of the second acetate ion, O3 and O4, take up the axial
positions of the bipyramid and form the connection between parallel
(
Fig. 8). Interactions between layers is however limited to van der
Waals interactions and a few weak C-H···O interactions originating
from the acetic acid methyl groups towards the acetic acid keto atoms
in the next neighboring layer.
In contrast to CMA, there are no solvate accessible void spaces in
3
the structure of Ca(OAc)
2
·HOAc, and its density is with 1.589 g/cm in
double ribbons, connecting them into infinite layers. The coordination
1
the range expected for a densely packed organic calcium salt. Its larger
density, compared to CMA, would thus make Ca(OAc) ·HOAc a more
easily applicable material, if used as a road deicer. Ca(OAc) ·HOAc
mode of the second acetate is μ
2
: η bridging. The protonated oxygen
atom of the acetic acid molecule is not metal coordinated, but tightly
2
1
hydrogen bonded to O4 of the η bridging acetate ion.
2
does however lack the other beneficial properties that make CMA
uniquely suited for this purpose, such as its neutral pH due to the
absence of acetic acid molecules in its crystal lattice, and the powerful
freezing point depression properties of CMA.
3. Conclusions and outlook
Calcium magnesium acetate, CMA, has been in use for over thirty
years as a commercial road deicer and as a precursor for industrial
adsorbents for removing H S and other odorous acid gases from gas
2
streams. Despite this extensive use, its exact nature had never been
univocally established, and many of its most basic properties, such as
its relative amounts of calcium and magnesium or the possibility of it
being a hydrated species, had been contentious points of discussion.
Using X-ray diffraction techniques, we have now established composi-
tion and structure of CMA. Samples suitable for analysis by single
crystal X-ray diffraction were obtained from hot glacial acetic acid, and
the authenticity of the samples was confirmed by comparison with
commercial CMA. The formula of CMA was established as
2 6
Mg Ca(OAc) , with an exact 2:1 ratio of magnesium to calcium. No
2
Fig. 7. Packing view of calcium acetate acetic acid, Ca(OAc) (HOAc). View diagonally
down the a-axis direction. Anisotropic displacement parameters are at 50% probability
level.
water was observed in the crystal lattice of the analyzed samples by
single crystal diffraction, but the structure exhibits an unusually low
7

J.R. Miller et al.
Journal of Solid State Chemistry 270 (2019) 1–10
density with substantial void space that could potentially be occupied
by small solvate molecules as was observed for an isotypic manganese
acetate acetonitrile solvate. Analysis of samples by thermal gravimetric
analysis gave an upper limit of one molecule of water per formula unit
of CMA, confirming that for samples prepared from hot glacial acetic
acid the void space remains at least partially if not completely
unoccupied. CMA's unusually low density helps to explain observations
from its commercial use as a road deicer, having been described as
being easily blown off road surfaces when applied in crystalline or
powder form. Attempts to obtain crystals suitable for SC-XRD from
aqueous solutions led to preferential formation of calcium acetate
acetic acid and water solvates. In the absence of both magnesium and
with acetone and collected. The hygroscopic crystals were analyzed by
single crystal X-ray diffraction, yielding the known structure of
magnesium acetate tetrahydrate [21,22] (see SI for details), and stored
in a tightly closed vessel until further use. The yield was essentially
quantitative.
4.2. Single crystal growth
A) Calcium and magnesium acetate hydrate crystals from pure and
mixed calcium/magnesium acetate precursors
Initial trials to grow single crystals of calcium magnesium
acetate, CMA, followed a biphasic extractive procedure as described
by Dionysiou et al. [8] for the growth of single crystalline Mg and
Ca acetates. The method involved recrystallization of mixtures of
calcium acetate monohydrate and magnesium acetate tetrahydrate
utilizing mixtures of aqueous acetic acid and organic phases (a
mixture of trioctyl- and tridecylamines as well as 2-ethyl-1-
hexanol). Various ratios of water and glacial acetic acid, and
calcium and magnesium have been used, but only two types of
crystalline phases formed crystals large enough for analysis by SC-
XRD. From solutions with medium to large mole fractions of
calcium to magnesium acetate and no excess acetic acid fine
needles of calcium acetate monohydrate formed (identified by
SC-XRD). In the presence of acetic acid, with medium to large
mole fractions of calcium to magnesium acetate, large blocky
crystals of calcium acetate monohydrate acetic acid solvate were
isolated (identified by SC-XRD). In the presence of large quantities
of magnesium (> 70 mol%), only amorphous material was ob-
tained.
water, the acetic acid solvate of calcium acetate, Ca(OAc)
obtained.
2
(HOAc), is
4
. Material and methods
.1. Sample preparation
All chemicals were obtained from established vendors and used as
4
received if not specified otherwise. A commercial CMA sample was
purchased from The Green Earth Deicer Company, Inc., N8336
Deadwood Point Beach Road, Fond du Lac, WI 54937 (Happy
Paws™ Ice Melt).
4
.1.1. Calcium acetate monohydrate, Ca(OAc)
2 2
·H O
Calcium acetate was prepared from calcium carbonate and diluted
acetic acid. The solubility of calcium acetate decreases with tempera-
ture, thus no crystalline material can be obtained by cooling of
saturated solutions [36]. Calcium acetate monohydrate was instead
isolated by salting out with acetone from the reaction mixture,
following the procedure described by Panzer [29]. Acetone was slowly
added to the vigorously stirred solution to form a fine precipitate.
When no more solid formed the precipitate was filtered off with
suction. The filter cake was washed with copious amounts of acetone,
isolated and suspended again in acetone with vigorous stirring, and the
solid again filtered off with suction. The process was repeated twice
until the filter cake was essentially odorless (no acetic acid smell). The
solid was dried in air for several days with intermediate breaking up
and grinding, and analyzed by powder X-ray diffraction. Rietveld
refinement yielded one of the triclinic polymorphs of calcium acetate
monohydrate [16] with lattice constants a = 6.755, b = 11.103, c =
B) Calcium magnesium acetate crystals by extraction from commercial
CMA samples
Several grams of crude beige colored pellets of commercial CMA
were ground into a powder heated in copious amounts of water (ca.
500 mL) until boiling and filtered while hot. The aqueous solution
was left standing for several days upon which large amounts of fine
needles of calcium acetate monohydrate formed (identified by
single crystal X-ray diffraction). The remaining solution was
decanted after no more crystalline material formed, and evaporated
to dryness using rotary evaporation. The remaining white solid was
dried in air on filter paper, and split into 0.5 g portions for further
use. Three individual portions were suspended in glacial acetic
acid, 1.5, 1 and 0.5 mL respectively, and heated to ca 80 °C in open
vessels (5 mL scintillation vials with loosely screwed on caps). All
samples initially completely dissolved upon heating. Slow removal
of remaining traces of water and slight reduction of acetic acid
volume (to about one half of the original volume) by boiling at
80 °C over several hours, followed by slow cooling from 80 °C to
room temperature induced formation of single crystals of CMA.
Best formed crystals were obtained from the vial starting with 1 mL
of acetic acid. Crystals were isolated from remaining glacial acetic
acid, washed with acetone, dried in air and analyzed by SC-XRD
and powder X-ray diffraction.
1
1.801 Å, α = 116.41, β = 92.40, γ = 97.52° as the only crystalline
component, with no indication of the other reported polymorph
37,38] being present (see SI for details, Fig. S1). The material was
[
stored in a closed container until further use. The yield was essentially
quantitative.
4
.1.2. Magnesium acetate tetrahydrate, Mg(OAc)
2 2
·4H O
Magnesium carbonate proved to be unreactive under the conditions
used for synthesis of calcium acetate as described above, with most
magnesium carbonate left undissolved even after overnight heating to
reflux off the solvent, and more reactive magnesium oxide was used
instead [39]. 5 g MgO (0.124 mol) were suspended in 19 mL of
deionized water in a 150 mL beaker glass with a stir bar. 14.2 mL
C) Calcium magnesium acetate crystals from mixtures of calcium
acetate monohydrate, Ca(OAc)
hydrate, Mg(OAc) ·4H
The process followed the procedure as outlined above, for CMA
crystals from commercial CMA samples. Ca(OAc) ·H and
Mg(OAc) ·4H O were mixed in a 1:2 mol ratio. 500 mg quantities
2 2
·H O and magnesium acetate tetra-
2
2
O
(
14.9 g, 0.248 mol) of glacial acetic acid was carefully added and the
mixture was slowly heated to slight boiling while stirring. The mixture
was kept at that temperature until all magnesium oxide was completely
dissolved. The solution was allowed to cool to room temperature,
transferred to a round bottom flask, and the volume was reduced using
a rotary evaporator until the residue showed signs of solidification. The
flask was removed from the rotary evaporator and carefully closed to
keep out atmospheric moisture and left standing to crystallize. After
several days, large crystals of magnesium acetate tetrahydrate formed.
Excess water was allowed to rinse off in the round bottom flask and
crystals were washed with small amounts of ice cold ethanol and then
2
2
O
2
2
of the mixture were suspended in 1 mL glacial acetic acid and
heated to ca 80 °C in open vessels (5 mL scintillation vials with
loosely screwed on caps) upon which all material completely
dissolved. Water and acetic acid were completely boiled off. The
process was repeated twice to assure that all water was completely
removed from the sample. Recrystallization from acetic acid (ca. 1
mL) by slow cooling from 80 °C to room temperature yielded single
crystals of CMA. Crystals were isolated from remaining glacial
8

J.R. Miller et al.
Journal of Solid State Chemistry 270 (2019) 1–10
acetic acid, washed with acetone, dried in air and analyzed by single
crystal X-ray diffraction unit cell determination and powder X-ray
diffraction.
scanning electron microscope marketed by Topcon (ISI), Inc. of
Paramus, NJ. EDS data were collected from the sample for elemental
identification using a Si(Li) crystal detector manufactured by Gresham
Scientific Instruments Ltd. of Buckinghamshire, UK coupled to a
multichannel analyzer interface manufactured by 4pi Analysis, Inc.
and resident in an Apple Macintosh® G4 workstation. Data were
collected between 0 and 25 keV in standardless format and quantified
using the Revolution software [48]. X-ray peaks were assigned to their
corresponding element using a table of major x-ray emission energies.
Only elements boron and heavier in the periodic table were included.
Spectra and numerical details are given in the SI.
D) Calcium acetate acetic acid crystals, Ca(OAc)
cium acetate monohydrate
2
(HOAc), from cal-
In a similar procedure as used for CMA, above, several milligrams
of calcium acetate monohydrate were suspended in glacial acetic acid in
a 5 mL scintillation vial. Water and acetic acid were completely boiled
off. The process was repeated twice to assure that all water was
completely removed from the sample. The white residue was taken
up again in glacial acetic acid (ca 1 mL) and heated to ca 80 °C.
Additional glacial acetic acid was added until all solid was completely
dissolved when hot. Once all solid was dissolved, the vial was capped
4.6. Thermal gravimetric analysis
(
to keep out atmospheric moisture), the hot plate switched off and the
TGA data were collected on a TA Instruments TGA Q50 from 40 °C
to 800 °C at a rate of 10 °C per minute. A batch of material was ground
into a fine powder, pre-dried at 80 °C in a slight stream of nitrogen for
one hour to remove any loosely bound surface adsorbed traces of
residual acetic acid or other solvent remnants from the workup
procedure.
solution was allowed to slowly cool to room temperature. Clusters of
crystals were obtained after several hours. Individual platelets were
isolated from the heavily intergrown clusters for SC-XRD.
4
.3. Single crystal X-ray diffraction
Diffraction data were collected on a Bruker AXS ApexII CCD
Supplementary material
diffractometer,
Prospector CCD diffractometer at 298 or 100 K using monochromatic
Mo or Cu K radiation with the omega scan technique. Data were
a Bruker AXS D8 Quest CMOS or Bruker X8
Additional
Ca(OAc) (H
Mg(OAc) (H
details
on
Ca(OAc)
single
(HOAc),
crystal
structures
for
O),
α
2
2
O)1.075
,
2
Ca(OAc)
2
(HOAc)(H
2
collected, unit cells determined, and the data integrated and corrected
for absorption and other systematic errors using the Apex2 suite of
programs [40] and Sadabs or Twinabs [41]. The space groups were
assigned and the structures were solved by direct methods using the
SHELXTL suite of programs [42] and refined by full matrix least
2
2
O)
4
, CMA, all at 100 K, and CMA at RT. Details of
SEM-EDS and P-XRD analysis data of a commercial CMA sample.
Reproductions of literature P-XRD patterns of CMA (for comparison).
A listing, in table form, of all reported calcium, magnesium and calcium
magnesium acetates (including hydrates and acetic acid solvates, data
base entry numbers and unit cell information, where available).
Rietveld refinement figure for calcium acetate monohydrate. SEM-
EDS spectrum of a commercial CMA sample. Original TGA curves with
and without selected weight loss numbers.
2
squares against F with all reflections using Shelxl2014 [43,44], using
the graphical interface Shelxle [45]. H atoms attached to carbon and
acetic acid oxygen atoms were positioned geometrically and con-
strained to ride on their parent atoms, with carbon hydrogen bond
3
distances of 0.98 Å for CH moieties, and 0.84 Å for acetic acid O-H
moieties, which were allowed to rotate but not to tip to best fit the
experimental electron density. Water hydrogen atoms were located in
difference density maps and were refined with a distance restraint for
the O-H bond of 0.84(2) Å. Uiso(H) values were set to 1.5 times the
value of Ueq(C/O). Experimental details for the structure of 100 K CMA
and calcium acetate mono acetic acid solvate are given in Table 1.
Details of the previously reported structures of magnesium acetate
tetrahydrate and of calcium acetate monohydrate at 100 K and of CMA
at 298 K are given in the Supporting material. Complete crystal-
lographic data for all structures, in CIF format, have been deposited
with the Cambridge Crystallographic Data Centre. CCDC 1845188-
Acknowledgments
J. R. M., M. J. L., P. M. K., D. E. W. and S. W. D. thank the Choose
Ohio First Scholarship program of the Ohio Department of Higher
Education for financial assistance and technical support. M. Z. would
like to thank the National Science Foundation for funding for the single
crystal X-Ray diffractometer via NSF DMR-1337296.
This work was supported by the National Science Foundation of the
United States through grants DMR 1337296 and CHE 0087210
(funding for the single crystal X-ray diffractometer), by the Ohio
Board of Regents Grant CAP-491, and by Youngstown State University.
1
845193 contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Appendix A. Supplementary material
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/
cif.
Supplementary data associated with this article can be found in the
online version at doi:10.1016/j.jssc.2018.10.041.
4
.4. Powder X-ray diffraction (P-XRD)
References
Data was collected in Bragg Brentano (cup) mode on a Rigaku
Miniflex powder XRD diffractometer using monochromatic Cu K
α
[1] D.L. Wise, Y.A. Levendis, M. Metghalchi (Eds.), Calcium Magnesium Acetate. An
Emerging Bulk Chemical for Environmental Applications, Elsevier, New York,
radiation. The samples were loaded in aluminum cups and were
collected in Bragg Brentano mode. Search/Match phase identification
was performed using the HighScore software [46] against the ICCD
PDF4 + and ICDD PDF4/Organics database [47]. Rietveld refinements
were performed against the models of the single crystal structure data
sets using the HighScore software of Panalytical. Refinement of
preferred orientation was included using a spherical harmonics model.
1
991.
[
2] Highway Deicing and Calcium Committee on the Comparative Costs of Rock Salt
and Calcium Magnesium Acetate (CMA) for Highway Deicing, Transportation
Research Board Special Report 235, National Research Council Washington, D.C.
1
991.
[
3] There had been speculation of a potential adverse environmental effect of CMA
used as an anti-icing material due to its constituent acetate ions, which in principle
are able to increase the biochemical oxygen demand (BOD) of streams nearby
application site, causing the depletion of dissolved oxygen (J.P. Connolly, P.R.
Paquin, T.J. Mulligan, K.B. Wu, L. Davanzo, Calcium magnesium acetate biode-
gradation and its impact on surface waters: in: C.R. Goldman, G.J. Malyj, (eds.),
The Environmental Impact of Highway Deicing—Proceedings of a symposium held
October 13, 1989 at the University of California, Davis Campus, Institute of
4
.5. Energy dispersive X-ray spectroscopy (EDS)
A ground sample was placed onto the goniometer stage of a
9

J.R. Miller et al.
Journal of Solid State Chemistry 270 (2019) 1–10
Ecology, no. 33, 1990, p. 140–156.). An in depth study (the following reference) at
Council, Washington, D.C, 1985, pp. 8–12.
several creeks in the vicinity of U.S. Highway26 did however show no a discernible
differences in BOD or calcium and magnesium concentration before and after
application of CMA.
4] The Effect of Calcium Magnesium Acetate (CMA) Deicing Material on the Water
Quality of Bear Creek, Clackamas County, Oregon, 1999, U.S. Department of the
Interior U.S. Geological Survey, Prepared in cooperation with the Oregon
Department of Transportation. Water-Resources Investigations Report 00-4092.
Portland, Oregon, 2000.
[25] Hui-Juan Yang, Hui-Zhong Kou, Zhong-Hai Ni, Ai-Li Cui, Ru-Ji Wang, Inorg.
Chem. Commun. 8 (2005) 846–849.
[26] J.D. Martin, R.F. Hess, Chem. Commun. (1996) 2419–2420.
[27] Jian Lin, Lei Yuan, Dong-Wei Guo, Gao Zhang, Shi-Yan Yao, Yun-Qi Tian, Struct.
Chem. 20 (3) (2009) 505–508.
[28] J. Fielden, P. Kögerler, Acta Cryst. C 65 (2009) m224–m227.
[29] R.D. Shannon, Acta Cryst. A32 (1976) 751–767.
[30] S. Niu, K. Han, C. Lu, R. Sun, Appl. Energy 87 (2010) 2237–2242.
[31] D.H. Han, H.Y. Sohn, AIChE J. Environ. Energy Eng. 48 (12) (2002) 2971–2977.
[32] J. Adánez, F. García-Labiano, L.F. de Diego, V. Fierro, Energy Fuels 13 (1999)
440–448.
[
[
[
5] D.D. Dionysiou, M. Tsianou, G.D. Botsaris, Cryst. Res. Technol. 35 (2000)
1035–1049.
6] United States Environmental Protection Agency, Office of Water (4303)
Washington, DC 20460, EPA-821-R-00-016 August 2000, EPA Preliminary Data
Summary, Airport Deicing Operations (Revised).
[33] J. Panzer, J. Chem. Eng. Data 7 (1962) 140.
[34] O. Arternova, V. Orlova, I. Lepeshkov, Russ. J. Inorg. Chem. (Engl. Transl.) 30
(1985) 606.
[
[
7] F. Su, United States Patent 5,430,185, 1995.
8] D.D. Dionysiou, M. Tsianou, G.D. Botsaris, Ind. Eng. Chem. Res. 39 (2000)
2 2
[35] The reported powder XRD patterns for Ca(OAc) × 0.5H O (ICDD 00-014-0792
4192–4202.
and 00-039-0199) and Ca(OAc) × 2H O (ICDD 00-039-0529) are partially
2
2
[
9] A. P. Mathews, US. Pat. No. 5,498,362, 1996.
superimposable, with all but one line reported for the dihydrate having exact
counterparts among the more intense lines of the hemi-hydrates. This might
indicate a structural relationship between the two phases. The similarity of the
patterns might also point towards the presence of mixtures of phases for one or
both of the reported compounds.
[
10] J. Adanez, F. Garcia-Labiano, L.F. De Diego, V. Fierro, Energy Fuels 13 (2) (1999)
40–448.
11] F. Garcia-Labiano, L.F. De Diego, J. Adanez, J. Environ. Sci. Technol. 33 (2) (1999)
88–293.
4
[
2
[
[
[
12] H. E. Todd, Jr., D. L. Walters, United States Patent 4,913,831, 1990.
13] K. A. Hoenke, J. D. Rynbrandt, United States Patent 5,132,035, 1992.
14] V.T. Orlova, A.A. Frolov, O.N. Evstaf’eva, Zhurnal Neorganicheskoi Khimii, 38(12),
[36] C. Saury, R. Boistelle, F. Dalemat, J. Bruggeman, J. Chem. Eng. Data 38 (1993)
56–59.
[37] P. van der Sluis, A. Schouten, A.L. Spek, Acta Cryst. C 43 (1987) 1922–1924.
[38] R. Helems, L.B. Cole, E.M. Holt, Inorg. Chim. Acta 152 (1988) 9–15.
[39] D. Stamires, T. J. Pinnavaia, M. Brady, Mike, PCT International Appl.,
2002072526, 19 Sep 2002.
[40] Apex2 v2014.1-1, SAINT V8.34A (Bruker, 2014) Bruker Advanced X-ray Solutions,
Bruker AXS Inc., Madison, Wisconsin, USA.
[41] L. Krause, R. Herbst-Irmer, G.M. Sheldrick, D. Stalke, J. Appl. Cryst. 48 (2015)
3–10.
1993, pp. 2041–2044 (Russ. J. Inorg. Chem. (Engl. Transl.) 38, 1993, 1914).
[
15] Stephen A. Ketcham, L. David Minsk, Robert R. Blackburn, Edward J. Fleege,
Manual of Practice for an effective anti-icing program, A Guide For Highway Winter
Maintenance Personnel. Prepared for the Federal Highway Administration by US
Army Cold Regions Research and Engineering Laboratory Corps of Engineers
Hanover, New Hampshire, 03755-1290, Report No. FHWA-RD-95-202, June
1996.
[
[
[
[
[
[
16] E.A. Klop, A. Schouten, P. van der Sluis, A.L. Spek, Acta Cryst. C 40 (1984) 51–53.
17] Updated structural data of this phase are given in the SI of the article.
18] P. Walter-Levy, C. R. Seances Acad. Sci. (Paris) 246 (1958) 1553.
19] E.A. Klop, A.L. Spek, Acta Cryst. C 40 (1984) 1817–1819.
20] Updated structural data of this phase are given in the SI of the article.
21] P. de Wolff, Technisch Physische Dienst, Delft, The Netherlands. ICDD Grant-in-
Aid. Sample from M. Walter-Levy c.s., Compt. Rend., 249 1234 (1959), footnote 3.
22] D.E. Irish, J. Semmler, N.J. Taylor, G.E. Toogood, Acta Cryst. C 47 (1991)
[42] SHELXTL (Version 6. 14) (2000-2003) Bruker Advanced X-ray Solutions, Bruker
AXS Inc., Madison, Wisconsin, USA.
[43] G.M. Sheldrick, Acta Cryst. A64 (2008) 112–122.
[44] G.M. Sheldrick, Shelxl2014, University of Göttingen, Germany, 2014.
[45] C.B. Hübschle, G.M. Sheldrick, B. Dittrich, J. Appl. Cryst. 44 (2011) 1281–1284.
[46] HighScore, Version 4.5, PANalytical B.V., Almelo, The Netherlands, 2015.
[47] ICDD. PDF-4+ 2018 and PDF4/Organics (Database), in: Dr. Soorya Kabekkodu
(eds.), International Centre for Diffraction Data, Newtown Square, PA, USA, 2018.
[48] 4pi Spectral Engine Resource Manual and Revolution User Guide, 4pi
Microanalysis and Imaging Technology, 3500 Westgate Drive, Suite 403, Durham
NC 27707, USA.
[
2322–2324.
[
[
23] Updated structural data of this phase are given in the SI of the article.
24] D.D. Ernst, G. Demich, T. Wieman, Calcium magnesium acetate research in
Washington StateTransportation Research Record 1019, TRB, National Research
1
0