The influence of temperature (20-1000 °C) on binary mixtures of solid solutions of CH3COOLi·2H2O-MgHPO4·3H2O

Article abstract of DOI:10.1016/j.physb.2009.06.127

Thermally induced phase transitions (20-1000 °C) in the substrates and binary mixtures of CH₃COOLi·2H₂O(1)-MgHPO₄·3H₂O(11) have been analysed. Changes taking place on dehydration and thermal dissociation of bina

Full text of DOI:10.1016/j.physb.2009.06.127

ARTICLE IN PRESS
Physica B 404 (2009) 3620–3636
Contents lists available at ScienceDirect
Physica B
journal homepage: www.elsevier.com/locate/physb
The influence of temperature (20–1000 1C) on binary mixtures of solid
solutions of CH3COOLi ꢀ 2H2O–MgHPO4 ꢀ 3H2O
Ã
Micha" H. Umbreit a, , Dominik Paukszta b
a
ꢀ
University of Medical Sciences, Department of Inorganic and Analytical Chemistry, Poznan, Poland
b
ꢀ
ꢀ
Poznan University of Technology, Faculty of Chemical Technology, Poznan, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Thermally induced phase transitions (20–1000 1C) in the substrates and binary mixtures of
CH3COOLi ꢀ 2H2O(1)–MgHPO4 ꢀ 3H2O(11) have been analysed. Changes taking place on dehydration
and thermal dissociation of binary mixtures prepared with percent molar ratios of 90–10% were studied
by differential thermal analysis (TG, DTG, DTA), IR-spectroscopy and WAXS.
Received 19 January 2009
Received in revised form
14 May 2009
Accepted 5 June 2009
The above-mentioned substrates changed their structure when heated for 1 h at 500 or 1000 1C.
CH3COOLi ꢀ 2H2O(1) (ID: 23-1171) changed the structure at 500 1C to that of Li2CO3 (ID: 22-1141), while
at 1000 1C the structure was impossible to analyse as the compound reacted both with porcelain and
with platinum (crucible materials). MgHPO4 ꢀ 3H2O(11) (Newberyite, ID: 35-780, 19-762) changed its
structure at 500 1C to amorphous phase and at 1000 1C to Mg2P2O7 (ID: 32-626).
Keywords:
CH3COOLi ꢀ 2H2O
MgHPO4 ꢀ 3H2O
Binary mixtures
The following compounds were assayed in the respective binary mixtures heated at 500 1C for 1 h:
70% (1)–30%(11): LiMgPO4 (ID: 18-735), MgO (ID: 4-829); 50%(1)–50%(11): LiMgPO4 (ID: 18-735), Li3PO4
(ID: 25-1030); 30%(1)–70%(11): LiMgPO4 (ID: 32-574); binary mixtures heated at 1000 1C contained the
following compounds: 70%(1)–30%(11): LiMgPO4 (ID: 32-574,18-735), Li3PO4 (ID: 15-760,25-1030), MgO
(ID: 4-829); 50%(1)–50%(11): LiMgPO4 (ID: 32-574, 18-735), MgO (ID: 4-829); 30%(1)–70%(11): LiMgPO4
(ID: 18-735, 32-574), Mg2P2O7 (ID: 22-1152, 8-38), Li4SiO4 (37-1472).
TG
DTA
DTG
IR
WAXS
& 2009 Elsevier B.V. All rights reserved.
1. Introduction
presented in Ref. [10]. Lithium salts, such as the acetate, aspartate
and carbonate, sulphate, glutamate have been the subjects of
thermoanalysis in Ref. [11]. Other lithium salts have been studied
in numerous analyses aimed to determine their behaviour in
binary mixtures. Analyses of binary systems related to their
usefulness in glass preparation have been reported in Ref. [12].
Mixtures containing acetates of Li or K or Cd have been system-
atically studied with many acetates of Na, K, NH4, Ca, Ba, Sr, Ca, Zn,
In inorganic chemistry, binary systems are often studied on
heating. Our study was aimed to determine the behaviour of a
binary system made of lithium acetate and magnesium hydropho-
sphate subjected to heating at temperatures of 500 and 1000 1C.
Literature does not provide much data on this system, in
particular no data are available on chemical transformations as
well as structural transitions. Determination of phase transitions
and the possibility of mutual interactions between two inorganic
substrates, at least one of them being a phosphate, forming solid
solutions at high temperatures has been the subject of numerous
studies [1–4]. Mg3(PO4)2 ꢀ 8H2O (Bobierrite, ID: 16-330) and
MgHPO4 ꢀ 3H2O (Newberyite, ID: 35-780) turned out to be of
particular interest to us. They were considered model substrates
and mixed with a number of inorganic salts [5–7], including
lithium salts. Our first paper in this domain concerned a binary
system of Li3PO4–Mg3(PO4)2 ꢀ 8H2O [8], while the results of
analysis of a solid phase binary system Li2CO3–MgHPO4 ꢀ 3H2O
are described in Ref. [9] and those on the system with BaCO3 are
Table 1
Percentage molar ratios of substrates in binary mixtures.
Sample number
CH3COOLi ꢀ 2H2O
MgHPO4 ꢀ 3H2O
1
100
90
80
70
60
50
40
30
20
10
0
0
10
20
30
40
50
60
70
80
90
100
2
3
4
5
6
7
8
9
10
11
Ã
Corresponding author. Tel.: +48 618 546603; fax: +48 618 546609.
E-mail address: umbreitm@ump.edu.pl (M.H. Umbreit).
0921-4526/$ - see front matter & 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.physb.2009.06.127

ARTICLE IN PRESS
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
3621
Pb, Ba, Sr. The use of Li acetate in preparation of crystalline Li2ZrO3
is reported in Ref. [13]. Li acetate has been also used to obtain
sintered LiMn1.5Cu0.5O electrodes [14]. As the reactivity of Li
acetate in thermal binary systems has been analysed in a number
of papers by the above-mentioned authors, we decided to include
this compound in the present work. The other substrate used was
MgHPO4 ꢀ 3H2O whose structure is described in Refs. [15,16].
According to the analyses, it not only crystallizes with three water
Fig. 1. TG, DTG, DTA (200–10 000 1C) of substrates: CH3COOLi ꢀ 2H2O (1); MgHPO4 ꢀ 3H2O (11) and binary mixtures of molar ratio of 90–10% (1) and 10–90% (11) ( see Table 1).

ARTICLE IN PRESS
3622
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
Fig. 1. (Continued)
molecules, but may also occur in stable structures containing 1.2
[17]. Magnesium hydrogen phosphate (MgHPO4 ꢀ 3H2O) trans-
forms at high temperatures into magnesium pyrophosphate with
well-identified structure [18,19].
apparatus. WAXS radiography analysis was performed as de-
scribed in Ref. [5].
3. Results and discussion
To find out the influence of temperature (20–1000 1C) on
CH3COOLi ꢀ 2H2O, MgHPO4 ꢀ 3H2O and solid solution binary mix-
tures obtained with varying quantitative proportions of these
components, thermal differential analysis (TG, DTG, DTA) as well
as IR and WAXS-X-ray diffraction methods were applied.
3.1. Thermoanalysis
To study chemical reactions and associated phase transitions
taking place under the influence of temperature in the substrates
(CH3COOLi ꢀ 2H2O–MgHPO4 ꢀ 3H2O) and their binary mixtures,
they were subjected to thermal differential analyses (Fig. 1(1–11)).
The analysis of CH3COOLi ꢀ 2H2O thermogram (Fig. 11) shows that
in the temperature range of 20–110 1C a small loss of mass (2.5%)
occurs, equivalent to 0.05 water molecule mass as observed on the
TG curve. This effect is attributed to the desorption of
crystallization water condensed on the sample surface. Also, on
the DTA curve at 40 1C an endothermic process starts caused by
the melting of the solid under the influence of crystallization
water. The process ends at 120 1C having its maximum at 56 1C.
Next, on the TG curve between 110 and 180 1C a significant mass
loss of the sample (28.5%) is seen, related to the dehydration of
2.95 crystallization water molecules. This corroborated by a peak
on the DTA curve between 120 and 280 1C with a maximum at
150 1C. The dehydration process yields amorphous CH3COOLi.
Further temperature increase (200–360 1C) does not result in any
mass loss of the sample (TG curve), while on the DTA curve
between 280 and 320 1C an endothermic peak with a maximum at
295 1C appears , attributed to a phase transition. In the further
2. Materials
(1) Chemicals—substrates:
CH3COOLi ꢀ 2H2O(1); Aldrich cat: 51,799;
MgHPO4 ꢀ 3H2O(11); Aldrich cat: 34,075-8.
(2) Substrates and preparation of their binary mixtures:
Substrates (1, 11) were used as supplied (Table 1). Each
compound was grounded, sieved and binary mixtures were
prepared at the percent molar ratios of 90–10% as presented in
Ref. [5]. The substrates and binary mixtures were heated in
melting pots at 500 and 1000 1C always for 1 h as described
in Ref. [5].
(3) Methodology: Apparatus and analytical procedure. Differ-
ential thermal analysis was performed using Derivatograph
Q-1500 D as reported in paper [5]. IR analysis was carried out
using tablets of potassium bromide in Brucker–Vector 22
range of 380–450 1C
a mass loss caused by the thermal
decomposition of amorphous CH3COOLi is observed. It is
corroborated by the presence of two peaks on the DTA curve

ARTICLE IN PRESS
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
3623
between 360 and 440 1C. The first one is a minor, incompletely
developed endothermic peak with a maximum at approximately
380 1C related to the thermal effect of the thermal dissociation of
CH3COOLi (2CH3COOLi32C+Li2O+H2O). The other one is a major
exothermic peak with a maximum at 450 1C corresponding to
carbon combustion (C+O2-CO2m). The next stage of the phase
transition is related to high reactivity of Li2O, which in the
atmosphere of the thermoanalytical furnace reacts with the
present CO2. This leads to formation of Li2CO3, which in high
temperature (about 800 1C) undergoes thermal dissociation, as
manifested on the TG curve.
On the TG curve between 760 and 1000 1C a slow mass loss
related to Li2O sublimation is seen, and an endothermic peak of
Li2O sublimation appears on the DTA curve between 760 and
800 1C with a maximum at 765 1C. The TG curve of MgHPO4 ꢀ 3H2O
(Fig. 1(11)) in the temperature range of 20–90 1C shows a small
loss of mass (1.25%) related to the desorption of crystallization
water condensed on the surface of the compound. In the range of
90–250 1C further mass loss (24.25%) occurs due to the dehydra-
tion of MgHPO4 ꢀ 3H2O. In the range between 250 and 500 1C
(TG curve) further loss of constitutional water (5.5%) is observed
and Mg2P2O7 forms in the reaction 2MgHPO4-Mg2P2O7+H2Om.
These changes are reflected on the DTA curve by one major
endothermic maximum at 200 1C in the range of 20–630 1C which
corresponds to the dehydration processes. Another endothermic
maximum, very small however, occurs at 720 1C and is related to a
phase transition. It may be attributed to the transformation of the
incompletely formed magnesium pyrophosphate form into the
crystalline state, as indicated in previous studies [6,7,9].
The results of thermal analyses of binary mixtures are shown in
Fig. 1(2–10). On the basis of the TG curves recorded for mixtures
containing from 90% to 40% CH3COOLi ꢀ 2H2O (Fig. 1(2–7)) it was
found that the temperature between 20 and 110 1C causes little
loss of crystallization water bound to the crystal surface, similarly
to the effect observed in the substrates. However, proportionally
to the decreasing lithium acetate content in the samples the end
temperature of the process gradually falls between 110 and 80 1C.
Endothermic peaks on the DTA curve assigned to the phase
transition of CH3COOLi ꢀ 2H2O contained in the binary mixtures
are also reduced. The dehydration process (samples 2–7) observed
Fig. 2A. IR spectra A1–A11, samples non-heated.

ARTICLE IN PRESS
3624
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
Fig. 2B. IR spectra B1–B11, samples heated at 500 1C.
between 110 and 350 1C is distinct for each binary sample and
depends on the percent content of the constituents of the
mixtures. This process is seen in two or three stages of sample
mass loss (TG curves) and one or two peaks observed in DTA
curves. The mass loss marked on the TG curve between 350 and
500 1C is related to the thermal decomposition of CH3COOLi after
the loss of crystallization water and release of constitutional water
by MgHPO4 (samples 2–10). The process is complex in character
as it is different for each sample and depends on the proportion of
binary mixture constituents. Mass changes in the samples are
associated with various chemical reactions and resulting exother-
mic changes manifested on DTA curves. Those peaks differ in the
shape and height of maxima (Fig. 1(2–7)). Starting with sample 8
(30% lithium acetate and 70% magnesium hydrogen phosphate)
exothermic changes are not visible, and DTA curves are increas-
ingly similar to the curve recorded for the substrate, magnesium
hydrogen phosphate (Fig. 1(8–11)). All the aforementioned mass
loss processes and related endo- and exothermic changes which
cause thermal changes in the substrates and binary mixtures are
very much pronounced in the DTG curves (Fig. 1(8–11)). As
follows from literature [20,21] many physicochemical properties
of salts and their binary mixtures depend on the ionic potential. If
a system is composed of two salts, such as that in the present
work (CH3COOLi ꢀ 2H2O and MgHPO4 ꢀ 3H2O), the ionic potential of
the system can be expressed as the ratio (m1/m2) of cations
(1 and 2) making up the system in question. The ionic potential mi
of cation X is defined by the ratio of the cation charge (ei) to its
ionic radius (ri): ui ¼ ei/ri. As calculated on the basis of standard
values, m1 for Li+ is 0.006 and m2 for Mg2+ is 0.0125. Hence, the
value of m1
2Z0.42, that is, typical of systems which may form eutectic
mixtures on the assumption that solid solution transitions fulfil
/m2 is 0.552, which falls in the range of 0.95Zm1/
m
the condition m1/m2E1.
3.2. IR spectra
As the results of thermoanalytical studies of both the
substrates and solid solution binary mixtures as well as the
aforementioned theoretical considerations prove that the binary
mixtures may form new compounds with diverse structures and
spectral characteristics at high temperatures, it was decided to
make IR analyses of the resulting sinters. Three series of 11

ARTICLE IN PRESS
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
3625
Fig. 2C. IR spectra C1–C11, samples heated at 1000 1C.
samples each were used in the analysis: A1–A11: non-heated;
B1–B11: heated at 500 1C, C1–C11: heated at 1000 1C (Fig. 2(A–C)).
The IR analysis of CH3COOLi ꢀ 2H2O (Fig. 2A, spectrum A1)
showed that between 4000 and 2250 cmꢁ1 a distinct absorption
peak with the maxima at 3374, 3156 and 2361 cmꢁ1 occurs,
attributed to hydroxyl vibrations. The range between 2250 and
1150 cmꢁ1 shows two maxima (1597, 1435 cmꢁ1) also attributed
to the –OH groups in the compound. In the third spectral range
characteristic of CH3COOLi ꢀ 2H2O (1150–600 cmꢁ1) maxima are
seen at 1056, 862 and 690 cmꢁ1. In the range of 550–200 cmꢁ1
between 2250 and 1150 cmꢁ1 shows two small maxima of
different height (1707, 1638 cmꢁ1) also attributed to the –OH
groups in MgHPO4 ꢀ 3H2O. The third range characteristic of
MgHPO4 ꢀ 3H2O is that between 1150 and 800 cmꢁ1 with a broad
band from the phosphate group with the maxima at 1249, 1164,
1012 and 896 cmꢁ1. The spectral analysis of the binary mixtures
shows that in non-heated samples major features typical of the
MgHPO4 ꢀ 3H2O spectrum begin to develop with the gradual
decrease in the molar fraction of CH3COOLi ꢀ 2H2O in the samples
(A2–A10). This effect is particularly pronounced between 4000
and 2250 cmꢁ1 whereby the spectrum of the hydroxyl groups
becomes increasingly strong due to the increased molar fraction
of MgHPO4 ꢀ 3H2O. The same effect related to the vibrations of the
hydroxyl groups in the compounds is observed between 2250 and
1150 cmꢁ1. The maxima assigned to the –OH groups seen in the IR
spectra of lithium acetate are at 1597 and 1435 cmꢁ1. These are
also observed in the IR spectra of binary mixtures with 90–10%
CH3COOLi ꢀ 2H2O content in the sample (Fig. 2, A1–A10). It is noted
that the increasing fraction of MgHPO4 ꢀ 3H2O (samples A2–A9)
does not induce changes in the positions of the maxima of the two
only one maximum appears at 503 cmꢁ1
, attributed to the
bending–twisting vibrations. The maximum corresponds to the
interactions of the Li+ cation with the hydroxyl group of the acetic
acid. The IR analysis of MgHPO4 ꢀ 3H2O (Fig. 2A, spectrum A11)
shows that between 4000 and 2250 cmꢁ1 a broad absorption peak
with four characteristic maxima at 3570, 3264, 2480 and
2365 cmꢁ1 occurs, attributed to the vibrations of hydroxyl groups
in the molecule. With decreasing content of lithium acetate in the
samples studied, intensity of bands originated from hydroxyl
groups originated from crystallization water decreases. The range

ARTICLE IN PRESS
3626
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
Fig. 3A1. X-ray pattern of CH3COOLi ꢀ 2H2O.
Fig. 3A11. X-ray pattern of MgHPO4 ꢀ 3H2O.
peaks (1597 and 1435 cmꢁ1), but contributes to the reduction in
their intensity. For the 90% content of MgHPO4 ꢀ 3H2O in the
binary mixture with 10% CH3COOLi ꢀ 2H2O, the maxima disappear
altogether. The analysis of hydroxyl group vibrations in
MgHPO4 ꢀ 3H2O proves it unambiguously that they occur in
different wave length ranges than in the spectrum of CH3COO-
Li ꢀ 2H2O. They are seen between 2000 and 1625 cmꢁ1 with two
maxima at 1707 and 1638 cmꢁ1. The maxima are also visible,
albeit increasingly less intense in the spectra of samples A11–A7
(Fig. 2A, A11–A7).
Another broad range is located between 1159 and 350 cmꢁ1
with a maximum at 1056 cmꢁ1 typical of the carboxyl group
(Fig. 2A, A1–A7). The other peaks occurring within the range are at
862, 692 and 500 cmꢁ1 being attributed to the bonding of the Li+
cation with the anionic group. It is interesting that the sharp peak
with a maximum at 1056 cmꢁ1 attributed to the acetate group

ARTICLE IN PRESS
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
3627
Fig. 3A4. X-ray pattern of CH3COOLi ꢀ 2H2O in a mixture consisting of CH3COOLi ꢀ 2H2O (70%) and MgHPO4 ꢀ 3H2O (30%).
Fig. 3A6. X-ray pattern of MgHPO4 ꢀ 3H2O in a mixture consisting of CH3COOLi ꢀ 2H2O (50%) and MgHPO4 ꢀ 3H2O (50%).
(Fig. 2A, A1–A6) broadens with increasing molar fraction of
MgHPO4 ꢀ 3H2O in the samples. Starting with sample A7
(Fig. 2A, A7–A10) its shape becomes typical of the maximum of
MgHPO4 ꢀ 3H2O. The maximum becomes increasingly distinct
with the decreasing molar fraction of CH3COOLi ꢀ 2H2O
and decreasing fraction of MgHPO4 ꢀ 3H2O. The IR spectral analysis
of binary mixtures heated for 1 h at 500 1C (Fig. 2B, B1–B11)
shows that heated compounds have different structure from the
unheated ones. The IR spectrum of lithium acetate dehydrate (A1)
heated at 500 1C shows the characteristic bands of carbonates:
ꢂ
a wide band within the range of 1900–1100 cmꢁ1 with a
maximum at 1440 cmꢁ1 which corresponds to the vibrations of
u3 ðCO23ꢁÞ;

ARTICLE IN PRESS
3628
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
Fig. 3A8. X-ray pattern of MgHPO4 ꢀ 3H2O in a mixture consisting of CH3COOLi ꢀ 2H2O (30%) and MgHPO4 ꢀ 3H2O (70%).
Fig. 3B1. X-ray pattern of Li2CO3 formed from CH3COOLi ꢀ 2H2O as a result of heating at 500 1C (sample A1-500 1C).
ꢂ
ꢂ
a narrow band within the range of 1100–1030 cmꢁ1 with a
The samples with the 10% and 20% MgHPO4 ꢀ 3H2O content in
maximum at 1081 cmꢁ1 which corresponds to the u1 vibrations
of the carbonate group;
the binary mixture with CH3COOLi ꢀ 2H2O (Fig. 2B, B2,B3) show
different structures of IR spectra, when compared to both the
substrates and the other samples. The hydroxyl bands disappear
in these spectra and those of the samples heated for 1 h at
500 1C. Moreover, the samples were registered for characteristic
a band within the range of 1030–809 cmꢁ1 with a prominent
maximum at 868 cmꢁ1 which corresponds to the u2 vibrations
ðCO23Þ.

ARTICLE IN PRESS
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
3629
Fig. 3B11. X-ray pattern of MgHPO4 ꢀ 3H2O after heating at 500 1C.
Fig. 3B4. X-ray pattern of LiMgPO4 formed from a mixture of CH3COOLi ꢀ 2H2O (70%) and MgHPO4 ꢀ 3H2O (30%) as a result of heating at 500 1C.
vibrations for u3 ðCO32ꢁÞ in the interval of 1900–1100 cmꢁ1, and u1
and u2 ðCO23ꢁÞ in the interval of 1100–809 cmꢁ1. Starting with
sample B4 which contains 30% MgHPO4 ꢀ 3H2O, the spectra of the
other binary mixtures (Fig. 2B, B4–B9) show maxima at the same
positions and of the same intensity; however, with increasing
fraction of magnesium hydrogen phosphate the spectral range
attributed to the phosphate group with its tendency for transition
into the pyrophosphate (Mg2P2O7) is increasingly broader, similar
to the pure substrate, MgHPO4 ꢀ 3H2O (Fig. 2B, B11).
The IR analysis of samples heated for 1 h at 1000 1C was
impossible as no spectrum could be recorded for CH3COOLi ꢀ 2H2O
at that temperature due to its complete decomposition into
gaseous products, and the remaining Li2O partially melted into
porcelain and sublimed.

ARTICLE IN PRESS
3630
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
Fig. 3B6. X-ray pattern of LiMgPO4 formed from a mixture of CH3COOLi ꢀ 2H2O (50%) and MgHPO4 ꢀ 3H2O (50%) as a result of heating at 500 1C.
Fig. 3B8. X-ray pattern of LiMgPO4 formed from a mixture of CH3COOLi ꢀ 2H2O (30%) and MgHPO4 ꢀ 3H2O (70%) as a result of heating at 500 1C.
The IR spectrum of the other substrate (MgHPO4 ꢀ 3H2O) heated
compound was characterized in our previous work [8] on binary
mixtures of Li3PO4–Mg3(PO4)2 ꢀ 8H2O. Another band (within
a wide spectral range 2500–750 cmꢁ1) observed in the spectra
of the samples (Fig. 2C, C3–C5) is the one attributed to the
phosphate group in the lithium compounds obtained on heating
process binary mixtures. The maxima of bending–stretching
vibrations between 750 and 350 cmꢁ1 in the spectra of samples
C3–C5 (Fig. 2C, C3–C5) are also distinct from those of the other
for 1 h at 1000 1C (Fig. 2C, C11) indicates the formation of
magnesium pyrophosphate. The IR analysis of the binary mixtures
(Fig. 2C, C2–C11) shows that the only sample with the spectrum
comparable to those of the others is C2 with 90% of CH3COO-
Li ꢀ 2H2O and 10% of MgHPO4 ꢀ 3H2O. The spectrum, as for its
maximum, appears to be identical with that of Li3PO4, which was
resolved with the use of WAXS (ID: 25-1030, 15-760). This

ARTICLE IN PRESS
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
3631
Fig. 3C11. X-ray pattern of Mg2P2O7 formed from MgHPO4 ꢀ 3H2O as a result of heating at 1000 1C.
Fig. 3C4a. X-ray pattern of LiMgPO4 formed from a mixture of CH3COOLi ꢀ 2H2O (70%) and MgHPO4 ꢀ 3H2O (30%) as a result of heating at 1000 1C.
binary mixtures. Starting with sample C6 to sample C10
(Fig. 2C, C6–C10) with increasing hydrogen phosphate
fraction in the sample, the pyrophosphate structure begins to
form under the influence of temperature of 1000 1C in all the
binary mixtures.
3.3. X-ray analysis
3.3.1. Substrates and non-heated mixtures
On the basis of the roentgenograms recorded it was
established using X-RAYAN that in the non-heated substrate

ARTICLE IN PRESS
3632
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
Fig. 3C4b. X-ray pattern of Li3PO4 formed from a mixture of CH3COOLi ꢀ 2H2O (70%) and MgHPO4 ꢀ 3H2O (30%) as a result of heating at 1000 1C.
Fig. 3C4c. X-ray pattern of MgO (periclase) formed from a mixture of CH3COOLi ꢀ 2H2O (70%) and MgHPO4 ꢀ 3H2O (30%) as a result of heating at 1000 1C.
sample A1 has the structure of CH3COOLi ꢀ 2H2O (ID: 23-1171)
substrates contained in non-heated binary mixtures. The
radiograph of the sample with 70% CH3COOLi ꢀ 2H2O and 30%
MgHPO4 ꢀ 3H2O (sample A6) clearly shows matched peaks from
both mixture components (Fig. 3A4). When mixed in this
proportion, they do not interact, that is, no new compound is
(Fig. 3A1), while sample A11 has the structure of MgHPO4 ꢀ 3H2O,
Newberyite (ID: 35-780, 19-762) (Fig. 3A11). For comparative
purposes X-ray analyses of samples A4, A6 and A8 were
performed to characterize the crystal structures of the

ARTICLE IN PRESS
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
3633
Fig. 3C6. X-ray pattern of LiMgPO4 formed from a mixture of CH3COOLi ꢀ 2H2O (50%) and MgHPO4 ꢀ 3H2O (50%) as a result of heating at 1000 1C.
Fig. 3C8a. X-ray pattern of LiMgPO4 formed from a mixture of CH3COOLi ꢀ 2H2O (30%) and MgHPO4 ꢀ 3H2O (70%) as a result of heating at 1000 1C.
formed. The radiograph of the equimolar mixture of 50%
CH3COOLi ꢀ 2H2O and 50% MgHPO4 ꢀ 3H2O (Fig. 3A6) shows the
maxima characteristic of both substances and confirms their
presence in sample A6. The spectrum shows that the intensity of
magnesium hydrogen phosphate peaks (ID: 35-780, 19-762) is
higher than those of the lithium acetate, which confirms the
Newberyite structure. The other radiograph was recorded of a
non-heated binary mixture with 30% CH3COOLi ꢀ 2H2O and 70%
MgHPO4 ꢀ 3H2O (sample 3A8). The presence of both compounds
was confirmed. Strong peaks from MgHPO4 ꢀ 3H2O indicate that
the compound can be identified by four identification cards (ID:
35-780, 19-762). Minor maxima attributed to CH3COOLi ꢀ 2H2O
show that the fraction of the compound in the mixture is small.
3.3.2. Substrates and binary mixtures heated for 1 h at 500 1C
The substrate, CH3COOLi ꢀ 2H2O (sample B1), after heating
transforms into crystalline Li2CO3 (ID: 22-1141) (Fig. 3B1) distinct

ARTICLE IN PRESS
3634
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
Fig. 3C8b. X-ray pattern of Mg2P2O7 formed from a mixture of CH3COOLi ꢀ 2H2O (30%) and MgHPO4 ꢀ 3H2O (70%) as a result of heating at 1000 1C.
Fig. 3C8c. X-ray pattern of Li4SiO4 from a mixture of CH3COOLi ꢀ 2H2O (30%) and MgHPO4 ꢀ 3H2O (70%) as a result of heating at 1000 1C.
from that observed in the non-heated lithium acetate sample (A1).
compounds present in sample B4. The strongest peaks are
recorded, along with other peaks, at angles of 431 and 36.691,
which confirms the formation of MgO in the thermal reaction
occurring in the binary mixture. The radiograph of the equimolar
mixture of 50% CH3COOLi ꢀ 2H2O and 50% MgHPO4 ꢀ 3H2O
(Fig. 3A6) shows the maxima characteristic of both substances
and confirms their presence in sample A6. The spectrum shows
The other substrate, MgHPO4 ꢀ 3H2O (sample B11) heated at 500 1C
changed its structure from crystalline to amorphous (Fig. 3B11).
The radiograph of the sample with 70% CH3COOLi ꢀ 2H2O and 30%
MgHPO4 ꢀ 3H2O (Fig. 3B4) clearly shows the matched peaks from
LiMgPO4 (ID: 18-735) occurring in the spectra of the crystal forms.
Li3PO4 (ID: 25-1030) and MgO (ID: 4-829) are the other

ARTICLE IN PRESS
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
3635
Table 2
(ID: 4-829) appear. However, the presence of Li4SiO4 (ID: 37-1472)
was confirmed. The compound formed in the reaction of lithium
with the silica from the porcelain of the crucible used in the
heating.
Identification of substrates CH3COOLi ꢀ 2H2O(1) and MgHPO4 ꢀ 3H2O(11) and binary
mixtures by means of X-ray diffraction patterns.
Temperature
Sample number
Chemical formula
Identity card number
All thermal transitions of both substrates and their binary
mixtures, non-heated and heated for 1 h at 500 and 1000 1C, are
shown in Table 2.
20 1C
A1: 100%(1)
CH3COOLiꢀ 2H2O
CH3COOLiꢀ 2H2O
MgHPO4 ꢀ 3H2O
MgHPO4 ꢀ 3H2O
23-1171
23-1171
35-780
35-780
A6: 50%(1)–50%(11)
A11: 100%(11)
500 1C
A1: 500 1C 100%(1)
B6: 50%(1)–50%(11)
Li2CO3
22-1141
25-1030
18-735
4. Conclusions
Li3PO4
LiMgPO4
Amorphous
In this work, the results of TG, DTG, DTA, IR and WAXS
measurements of substrates and of binary mixtures CH3COO-
Li ꢀ 2H2O–MgHPO4 ꢀ 3H2O non-heated or heated at 500 and
1000 1C are presented.
Using the TG, DTG and DTA methods, the boundaries of the
phase transitions in substrates and in the binary mixtures were
found. In the mixtures with lithium acetate content of 30–90%, we
have observed mainly lithium cation containing products, which
probably results from activity of Li+ cation. This points to the
predominant effect of the above cation on the composition of
products obtained after heating.
B11: 100%(11)
1000 1C
100%(1)
Sublimation
LiMgPO4
Li3PO4
C4: 70%(1)–30%(11)
32-574; 18-735
15-760; 25-1030
4-829
MgO
C6: 50%(1)–50%(11)
C8: 30%(1)–70(11)
LiMgPO4
MgO
32-574; 18-735
4-829
LiMgPO4
Mg2P2O7
MgO
18-735;32-574
22-1152; 32-626
4-829
a
Li4SiO4
37-1472
C11: 100%(11)
Mg2P2O7
32-626; 22-1152
The IR analyses completed with WAXS data permitted
identification of the following compounds formed upon exposure
to high temperatures in the substrates: Li2CO3 is formed at 500 1C
from CH3COOLi ꢀ 2H2O, Mg2P2O7 is formed at 1000 1C from
MgHPO4 ꢀ 3H2O, whereas in the binary mixtures at the latter
temperature LiMgPO4, Li3PO4, MgO and Mg2P2O7 are formed.
It was found that LiMgPO4 is formed in all binary mixtures heated
at 500 and 1000 1C from the substrates. The presence of Li3PO4
is confirmed also in the mixture containing 50% of both
substrates, heated at 500 1C for 1 h and also in the mixture
containing 70% CH3COOLi ꢀ 2H2O and 30% MgHPO4 ꢀ 3H2O heated
at 1000 1C for 1 h. The formation of MgO as a result of thermal
reactions was confirmed using the WAXS method in the mixture
70% CH3COOLi ꢀ 2H2O and 30% MgHPO4 ꢀ 3H2O heated at 500 1C
and in all investigated binary mixtures heated for 1 h at 1000 1C.
The results obtained confirmed that transitions and transfor-
mations taking place in the binary systems studied can be
identified only on the basis of experimental data. In this study,
aimed at identification of reactions induced by temperature in the
solid phase between lithium acetate and magnesium hydropho-
sphate, we have proved not only the formation of Mg2P2O7 and
LiMgPO4, but also the formation of MgO, which was not
theoretically predicted. This phenomenon has not been docu-
mented in the literature. Results of our study can be of potential
interest to those working on production technology of glass and
ceramics.
a
This compound is formed in the reaction of Li+ with the silica from the
crucible used for the heating.
that the intensity of magnesium hydrogen phosphate peaks
(ID: 35-780, 19-762) is higher than those of the lithium acetate,
which confirms the Newberyite structure. The other radiograph
was recorded of
a non-heated binary mixture with 30%
CH3COOLi ꢀ 2H2O and 70% MgHPO4 ꢀ 3H2O (sample A8). The
presence of both compounds was confirmed. Strong peaks from
MgHPO4 ꢀ 3H2O indicate that the compound can be identified by
four identification cards (ID: 35-780, 19-762). Minor maxima
attributed to CH3COOLi ꢀ 2H2O show that the fraction of the
compound in the mixture is small. The radiograph of a binary
mixture with 50% CH3COOLi ꢀ 2H2O and 50% MgHPO4 ꢀ 3H2O
(Fig. 3B6) indicates the presence of two crystal forms: of
LiMgPO4 (ID: 18-735) and of Li3PO4 (ID: 25-1030). The
radiograph of a binary mixture with 30% CH3COOLi ꢀ 2H2O and
70% MgHPO4 ꢀ 3H2O (Fig. 3B8) confirms only the presence of
LiMgPO4 (ID: 32-574).
3.3.3. Substrates and binary mixtures heated for 1 h at 1000 1C
As for the sample of one substrate, CH3COOLi ꢀ 2H2O, no
analysis could be performed, because lithium acetate (CH3COO-
Li ꢀ 2H2O) when heated from 20 to 1000 1C first melted, then
released water, CO2, and converted into Li2O which sublimed at
1000 1C. In the other substrate, MgHPO4 ꢀ 3H2O (Fig. 3C11), the
structure changed completely into Mg2P2O7 (ID: 32-626). The
analysis of radiographs of binary mixtures prepared from
substrates heated for 1 h at 1000 1C indicates that different
crystal structures form in respective mixtures. The radiograph of
the sample with 70% CH3COOLi ꢀ 2H2O and 30% MgHPO4 ꢀ 3H2O
(Fig. 3C4a–c) clearly shows LiMgPO4 present in two crystal forms
(ID: 32-574, 18-735) along with two maxima from Li3PO4 (ID: 15-
760, 25-1030), MgO is also observed to give the characteristic
peaks at 431 and 36.391 (ID: 4-829). The analysis of the radiograph
of the sample with 50% CH3COOLi ꢀ 2H2O and 50% MgHPO4 ꢀ 3H2O
(Fig. 3C6) clearly indicates the presence of LiMgPO4 in two crystal
forms (ID: 32-574, 18-735). The maxima characteristic of MgO (ID:
4-829) are also pronounced. The radiograph of the sample with
30% CH3COOLi ꢀ 2H2O and 70% MgHPO4 ꢀ 3H2O (Fig. 3C8a–c)
clearly indicates the presence of LiMgPO4 in the crystal form
(ID: 18-735, 32-574). Also the signals assigned to Mg2P2O7 (ID: 22-
1152) occurring in crystal form and the maxima attributed to MgO
Acknowledgements
The financial support of Polish State Committee for Scientific
Research (KBN), Project no. 501–01–000269, is acknowledged. The
authors want to thank to A. Jedrasiewicz and T. Rudas for their
assistance in laboratory experiments.
References
[1] S.N. Achary, R. Mishra, O.D. Jayakumar, S.K. Kulshreshtha, A.K. Tyagi, J. Solid
State Chem. 180 (2007) 84.
[2] S. Benmokhtar, H. Belmal, A. El Jazouli, J.P. Chaminade, P. Gravereau, S. Pechev,
J.C. Grenier, G. Villeneuve, D. de Waal, J. Solid State Chem. 180 (2007) 772.
[3] S.F. Zhang, G.Z. Liu, S.T. Zheng, G.Y. Yang, J. Solid State Chem. 180 (2007) 1943.
[4] S. Benmokhtar, A. El Jazouli, A. Aatiq, J.P. Chaminade, P. Gravereau, A.
ꢁ
Wattiaux, L. Fournes, J.C. Grenier, J. Solid State Chem. 180 (2007) 2004.
[5] M.H. Umbreit, A. Jedrasiewicz, M. Boudraa, Japan Assoc. Inorg. Phosph. Chem.
Phosphorus Res. Bull. (Japan) 13 (2002) 243.

ARTICLE IN PRESS
3636
M.H. Umbreit, D. Paukszta / Physica B 404 (2009) 3620–3636
[6] M.H. Umbreit, A. Jedrasiewicz, J. Klos, A. Walowska, Japan Assoc. Inorg.
Phosph. Chem. Phosphorus Res. Bull. (Japan) 14 (2002) 57.
[7] M.H. Umbreit, J.P. Auffredic, A. Jedrasiewicz, Japan Assoc. Inorg. Phosph.
Chem. Phosphorus Res. Bull. (Japan) 18 (2005) 99.
[13] L. Montanaro, J.P. Lecompte, J. Mater. Sci. 27 (1992) 3763.
[14] Y. Ein-Eli, W.F. Howard Jr., S.H. Lu, S. Mukerjee, J. McBreen, J.T. Vaughey,
M.M. Thackeray, /www.osti.gov/bridge/servlets/purl/10990RXQMCp/web
viewable/10990.pdfS.
[8] M.H. Umbreit, A. Jedrasiewicz, Japan Assoc. Inorg. Phosph. Chem. Phosphorus
Res. Bull. (Japan) 18 (2005) 15.
[9] M.H. Umbreit, A. Jedrasiewicz, D. Paukszta, N. Szyda, Japan Assoc. Inorg.
Phosph. Chem. Phosphorus Res. Bull. (Japan) 22 (2008) 1.
[15] F. Abbona, R. Boiselle, R. Haser, Acta Cryst. Sect. B 35 (1979) 2514.
[16] J. Chen, S. Natarajan, P.A. Wright, R.H. Jones, J.M. Thomas, C.R.A. Catlow, J.
Solid State Chem. 103 (1993) 519.
[17] A. Bensalem, G. Iyer, S. Amar, J. Solid State Chem. 114 (1995) 598.
ˇ
´
[10] P. Pasierb, R. Gajerski, M. Rokita, M. Rekas, Physica B 304 (2001) 463.
[11] G.B. Tabon-Zapata, E.G. Ferrer, S.B. Etheverry, E.J. Baran, J. Thermal Anal.
Calorimetry 61 (2000) 29.
[18] U.B. Mioc, D.M. Minic, J. Serb. Chem. Soc. 52 (1987) 317.
´
mcˇ, D.M. Minic´, Termochim. Acta 116 (1987) 131.
[19] N. Petranovic, U.B. Mio
[20] G.H. Cartledge, J. Amer. Chem. Soc. 50 (1928) 301.
[21] W. Gawel, A. Gorniak, J. Nuclear Mater. 340 (2005) 64.
[12] R.F. Bartholomew, S.S. Lewek, J. Amer. Ceram. Soc. 53 (1970) 445.