Mechanochemical synthesis and characterization of alkaline earth metal terephthalates: M(C8H4O4)·nH 2O (M = Ca, Sr, Ba)

Article abstract of DOI:10.1002/zaac.201300537

The mechanochemical synthesis offers an easy access to obtain alkaline earth metal terephthalates M(C₈H₄O₄) ·nH₂O (M = Ca, Sr, Ba). In the presented study we describe for the first time the mechanochemical synthesis of powders of Ca(C ₈H₄O₄)·3H₂O, Ca(C ₈H₄O₄), Sr(C₈H₄O ₄)·H₂O, and Ba(C₈H₄O ₄), which so far were only synthesized as single crystals from aqueous solutions or by reactions in an autoclave. Furthermore, a new hydrate Ba(C₈H₄O₄)·2(1.5)H₂O, not described so far in the literature, was prepared. All compounds were characterized by X-ray powder diffraction, thermal analysis, elemental analysis, FT-IR, and MAS NMR spectroscopic measurements. Copyright

Full text of DOI:10.1002/zaac.201300537

ARTICLE
DOI: 10.1002/zaac.201300537
Mechanochemical Synthesis and Characterization of Alkaline Earth Metal
Terephthalates: M(C8H4O4)·nH2O (M = Ca, Sr, Ba)
Gudrun Scholz,*[a] Abdullah Abdulkader,[a] and Erhard Kemnitz*[a]
Keywords: Mechanochemical synthesis; Alkaline earth metal terephthalates; MAS NMR spectroscopy; IR spectroscopy
Abstract. The mechanochemical synthesis offers an easy access to
obtain alkaline earth metal terephthalates M(C8H4O4)·nH2O (M = Ca,
Sr, Ba). In the presented study we describe for the first time the mecha-
as single crystals from aqueous solutions or by reactions in an auto-
clave. Furthermore, a new hydrate Ba(C8H4O4)·2(1.5)H2O, not de-
scribed so far in the literature, was prepared. All compounds were
nochemical synthesis of powders of Ca(C8H4O4)·3H2O, Ca(C8H4O4), characterized by X-ray powder diffraction, thermal analysis, elemental
Sr(C8H4O4)·H2O, and Ba(C8H4O4), which so far were only synthesized
analysis, FT-IR, and MAS NMR spectroscopic measurements.
Mazaj et al. performed a detailed study of structural dynamics
of Ca(BDC)(DMF)(H2O) induced by heating and hydration
and followed changes by different spectroscopic techniques
like FT-IR, MAS NMR, EXAFS, XANES.[7] They found that
a thermal treatment of Ca(BDC)(DMF)H2O) at 400 °C results
in a water-free calcium terephthalate Ca(C8H4O4), which is
transformed to Ca(C8H4O4)·3H2O after exposure to a humid
environment.[7]
In general, a mechanochemical synthesis of alkaline earth
coordination polymers is very rarely described in the literature.
Usually the authors start from metal oxides, carbonates or acet-
ates, and explore the possibilities of liquid assisted grinding
(LAG).[8–11]
In Ref. [5] we showed that phase pure Sr(C8H4O4)·3H2O
can be prepared very easily by milling powders of strontium
hydroxide octahydrate Sr(OH)2·8H2O with terephthalic acid
(1,4-benzenedicarboxylic acid, H2BDC) in a planetary mill.
Based on such an easy access of strontium terephthalate
trihydrate by milling[5] it is the intention of the presented study
to synthesize further alkaline earth terephthalates also on this
mechanochemical way. Herein, the alkaline earth hydroxides
and hydroxide hydrates Ca(OH)2, Sr(OH)2, Ba(OH)2,
Ba(OH)2·H2O, and Ba(OH)2·8H2O were taken as alkaline earth
metal sources. The second component used for the mechano-
chemical reaction is, as in case of Ref. [5], terephthalic acid
C6H4(COOH)2 (H2BDC).
Introduction
Seven terephthalate hydrates or terephthalates of alkaline
earth metals without further chemical components or substitu-
tions were described in the literature so far. Five of them,
namely Ca(C8H4O4)·3H2O,[1,2] Sr(C8H4O4)·4H2O,[2] and
Sr(C8H4O4)·H2O,[3] as well as Ba(C8H4O4)·4H2O[2] and
Ba(C8H4O4)[4] were synthesized from aqueous solutions or by
reactions in an autoclave for several days. As for Ca-coordina-
tion polymers introduced below, the traditional synthesis stra-
tegies for alkaline earth metal coordination polymers are either
precipitations or solvothermal methods.
For Sr(C8H4O4)·3H2O we reported for the first time a
successful mechanochemical synthesis, and water-free
Sr(C8H4O4) was synthesized after thermal post-treatment of
the trihydrate obtained by milling.[5]
In addition, six new calcium metal-organic frameworks were
described by Lin et al.[6] all of them have 3D structures built
of polycarboxylate ligands and Ca2+ ions. These carboxylate
ligands are BDC [1,4-benzenedicarboxylate anion (C8H4O4)],
ABDC (2-aminobenzene-1,4-dicarboxylate anion), BTC
(1,3,5-benzenetricarboxylate anion), and H2dhtp (2,5-dihy-
droxyterephthalate anion). These compounds were synthesized
by solvothermal and microwave assisted solvothermal reac-
tions. N,N-Dimethylformamide (DMF) is a terminal ligand in
all compounds except one. Whereas the paper of Lin et al.[6]
is mainly devoted to the characterization of crystal structures,
All milling products are characterized by elemental analysis,
X-ray diffraction (XRD), FT IR, 1H- and 1H-13C CP MAS
NMR spectroscopy, and thermal analysis.
* Dr. G. Scholz
Fax: 030-2093-7277
E-Mail: Gudrun.Scholz@rz.hu-berlin.de
* Prof. Dr. E. Kemnitz
E-Mail: Erhard.Kemnitz@chemie.hu-berlin.de
[a] Department of Chemistry
Results and Discussion
Humboldt-Universität zu Berlin
Brook-Taylor-Straße 2
12489 Berlin, Germany
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/zaac.201300537 or from the au-
thor.
Calcium Terephthalates Ca(C8H4O4)·nH2O
X-ray powder diffractograms of products obtained by mill-
ing of Ca(OH)2 with terephthalic acid applying different mill-
ing impacts are shown in Figure 1a–c.
Z. Anorg. Allg. Chem. 2014, 640, (2), 317–324
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
317

G. Scholz, A. Abdulkader, E. Kemnitz
ARTICLE
Figure 1. X-ray powder diffractograms of products obtained by milling of Ca(OH)2 with terephthalic acid (H2BDC) in a molar ratio of 1:1
applying different milling impacts: (a) 1 h milling, ο: reflections of Ca(OH)2, (b) 4 h milling, (c) 8 h milling; (d) 1 h milling and thermally
post-treated at 300 °C, (e) 4 h milling and thermally post-treated at 300 °C, (f) 8 h milling and thermally post-treated at 300 °C.
It becomes obvious that milling time has a distinct influence of H2O are depicted in Figure 2. From these powder patterns
on the observed reflections. After 1 h of milling the powder it is evident that a well crystalline product is formed and longer
pattern (Figure 1a) consists only of a superposition of reflec- milling times are not necessary (see for comparison Figure 2a,
tions of the precursor compounds, i.e. those of terephthalic b). Moreover, the obtained powder pattern is identical to the
acid (H2BDC) and Ca(OH)2 are visible (the latter are indi- powder pattern given for Ca(C8H4O4)·3H2O[7] and is almost in
cated). A milling time of 4 h (Figure 1b) leads to the issue of agreement with the powder pattern recalculated from single
new reflections beside reflections of still visible residues of crystal data of Ca(C8H4O4)·3H2O (Figure 2d), although the rel-
Ca(OH)2 and H2BDC. Applying the longest milling time of 8 ative intensities of reflections are slightly changed.
h reflections of the precursors disappeared and only reflections
of a new compound left over, however with a bad yield (poor
s/n ratio, Figure 1c). Using pre-milled starting materials did
not improve this situation.
A thermal post-treatment of the milled products for 2 h at
300 °C shows that the original short milling (1 h) did not in-
duce any chemical reaction. Here, well crystalline precursor
compounds can be identified (Figure 1d). This situation
changes for the samples treated before with higher milling im-
pacts. New, but for both samples identical reflections, appeared
after the thermal post-treatment (Figure 1e, f). These powder
diffractograms (Figure 1e, f) are completely in agreement with
the powder pattern given for Ca-BDC(400) in the literature,[7]
which corresponds to the water-free calcium terephthalate
Ca(C8H4O4). It means, the thermal post-treatment of the prod-
ucts obtained by milling with higher impact results in the for-
mation of the water-free calcium terephthalate Ca(C8H4O4).
Usually, during a thermal post-treatment up to 300 °C only
a water release can be expected (for comparison see also Fig-
ure S2, Supporting Information). Following that, the powder
pattern of the milling product shown in Figure 1c, i.e. the pre-
Figure 2. X-ray powder diffractograms of products obtained by mill-
ing of Ca(OH)2 with H2BDC (1:1) and addition of (a) 130 μL H2O
(1 h milling), (b) 130 μL H2O (4 h milling), (c) 170 μL H2O (1 h
milling), (d) powder diffractogram of Ca(C8H4O4)·3H2O, recalculated
from single crystal data.
cursor before thermal annealing, should be a calcium
terephthalate hydrate. However, the water content of Ca(OH)2
used as starting material for milling is obviously too low to
give a well crystalline product already after milling. In such a
case adding small amounts of water, i.e. performing liquid as-
sisted grinding, should result in a desired hydrate.
Therefore we can conclude that milling of Ca(OH)2 with
X-ray powder diffractograms of products obtained by mill- H2BDC for only 1 h under addition of small portions of water
ing of Ca(OH)2 with H2BDC and addition of few microliters (LAG) results in the formation of Ca(C8H4O4)·3H2O, which
318
www.zaac.wiley-vch.de
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2014, 317–324

Alkaline Earth Metal Terephthalates: M(C8H4O4)·nH2O
transforms into Ca(C8H4O4) after calcinations at 300 °C (the This can be unambiguously observed comparing the relative
same powder pattern as shown in Figure 1f is obtained).
intensity ratio of 13C signals of carboxyl carbon atoms with
Both FT-IR spectra (Figure S1, Supporting Information) as those of aromatic carbon atoms (cf. Figure S3, Supporting In-
well as DTA-TG measurements (Figure S2, Supporting Infor- formation).
mation), along with elemental analysis (Table S1, Supporting
Attempts to grow single crystals from powders obtained
Information) clearly show the formation of trihydrated and de- after milling [Ca(C8H4O4)·3H2O] and thermal post-treatment
hydrated phases of Ca-terephthalate, respectively. For all meth- [Ca(C8H4O4)], respectively, were so far not successful.
ods mentioned an almost theoretical ideal behavior is ob-
served. For instance, the TG curve of Ca(C8H4O4)·3H2O (Fig-
ure S2) shows – due to the liberation of water (m18) – a mass
Strontium Terephthalates Sr(C8H4O4)·nH2O
loss of 19.8% in the first step up to 200 °C, which is quite
close to the theoretically expected loss of 20.9%. As expected,
the water-free compound shows nearly no mass loss in that
temperature region. The decomposition of the compounds
starts above 400 °C evidenced by the release of CO2 in a two-
step process (m44, Figure S2).
1H-13C CP MAS NMR spectra of Ca(C8H4O4)·3H2O and
Ca(C8H4O4) prepared by milling (LAG) and consecutive ther-
mal treatment are shown in Figure 3. In both cases, the MAS
NMR spectra are in agreement with those from the literature.[7]
The spectrum shown in Figure 3b for Ca(C8H4O4) is identical
to that of Ca-BDC(400).[7] It shows only three signals: one
signal at δ = 173.6 ppm for carboxyl carbon atoms, and two
signals (δ =137 ppm and 131.6 ppm) belonging to aromatic
carbon atoms.
Four strontium terephthalates (hydrates) are described in the
literature: we prepared Sr(C8H4O4)·3H2O by milling of
Sr(OH)2·8H2O with terephthalic acid for only 1 h and obtained
in addition Sr(C8H4O4) after thermal post-treatment.[5]
Furthermore, the monohydrate Sr(C8H4O4)·H2O is known,
which was prepared by solvothermal reaction for 7 d, pub-
lished in 2011.[3] With Sr(C8H4O4)·4H2O[2] the list of stron-
tium terephthalates is completed.
In the presented study we elaborated the conditions for the
successful mechanochemical synthesis of Sr(C8H4O4)·3H2O
starting from Sr(OH)2·8H2O. In addition, the water free hy-
droxide Sr(OH)2 was used as a new precursor compound.
Evaluation of Successful Conditions for Mechanochemical
Synthesis of Sr(C8H4O4)·3H2O
Whereas in Ref. [5] the mechanochemical synthesis and
structure of Sr(C8H4O4)·3H2O were described, the conditions
for a successful synthesis by milling in a planetary mill were
studied in more detail and will be summarized very briefly.
This includes the variation of molar ratios of starting sub-
stances, the variation of milling impacts down to the use of a
mortar, or liquid assisted grinding experiments with CH3OH
to eliminate water.
The molar ratios used for milling Sr(OH)2·8H2O with
H2BDC were varied from 1:0.25 up to 1:2. We found that the
two precursors react stoichiometrically with each other. Re-
flections of Sr(C8H4O4)·3H2O are even visible with the lowest
supply of terephthalic acid, accompanied by still visible reflec-
tions of excess Sr(OH)2·8H2O. The powder pattern indicates
phase pure Sr(C8H4O4)·3H2O for the ratio 1:1, and in combi-
nation with elemental analysis an excess of H2BDC can be
proven for higher proportions in the mixture.
As shown in Ref. [5] within 1 h of milling Sr(C8H4O4)·
3H2O had been formed. However, further tests demonstrated
that even with a milling time of only 15 min the reaction was
complete. But the impact of a planetary mill (see Experimental
Figure 3. 1H-13C CP MAS NMR spectra of (a) Ca(C8H4O4)·3 H2O,
and (b) Ca(C8H4O4) prepared by milling Ca(OH)2 with H2BDC and
170 μL H2O for 1 h (a) and subsequent thermal treatment at 300 °C
(b).(νrot = 10 kHz, p15 = 6 ms).
For Ca(C8H4O4)·3H2O signals for two different carboxylate Section) was still necessary. Attempts by simply grinding the
groups appear at δ = 174.3 ppm and 172.7 ppm, accompanied two precursors in a mortar were not successful.
by an additional signal for an aromatic carbon at δ = 130 ppm
Our previous experiments clearly indicated the formation of
(Figure 3a). This is a consequence of the 3D structure of Sr(C8H4O4) after thermal treatment of mechanochemically
Ca(C8H4O4)·3H2O with CaO8 chains, which was discussed in synthesized Sr(C8H4O4)·3H2O. Attempts to remove the water
detail in the literature.[2,7] 1H-13C CP MAS NMR spectra of directly by milling to obtain Sr(C8H4O4) by adding only few
both compounds show a distinct dependence of relative signal droplets of methanol (LAG) failed. The presence of the solvent
intensities on the contact time (cf. Figure S3). The magnetiza- manifests only in a worse crystallinity of Sr(C8H4O4)·3H2O,
1
tion transfer from H to 13C is generally in favor of more dis- i.e. Sr(C8H4O4) was not formed. On the other hand the mecha-
tant carbon atoms if larger contact times (here p15) are applied. nochemically synthesized Sr(C8H4O4)·3H2O was successfully
Z. Anorg. Allg. Chem. 2014, 317–324
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
319

G. Scholz, A. Abdulkader, E. Kemnitz
ARTICLE
recrystallized from water and cubic colorless crystals were
The thermoanalytical curves (cf. Figure S4, Supporting In-
formed. X-ray structure analysis of these single crystals proved formation) demonstrate a two-step water release below 200 °C,
the formation of Sr(C8H4O4)·4H2O, which is obviously the accompanied by mass loss of 7.6%, which is close to the ex-
most stable phase of strontium terephthalate hydrates, instead pected theoretical mass loss of 6.7% of Sr(C8H4O4)·H2O.
of the expected Sr(C8H4O4)·3H2O. All structure data are iden- These findings are assisted by the carbon and hydrogen con-
tical to previously published results for Sr(C8H4O4)·4H2O.[2] tents determined by elemental analysis (Table S1, Supporting
It means on the other hand that the mechanochemical synthesis Information). The decomposition of the monohydrate starts
is the only possibility to obtain Sr(C8H4O4)·3H2O with locally around 500 °C in one step (release of CO2, m44).
1
formed (SrO8) polyhedra. After recrystallization the local coor-
The H-13C CP MAS NMR spectrum of Sr(C8H4O4)·H2O
dination changes from (SrO8) to (SrO10) (for comparison see (Figure 5) is completely embedded in the spectral patterns re-
Refs. [2,5]).
corded for Sr(C8H4O4)·3H2O and Sr(C8H4O4) (Figure 5). The
spectra of the latter two compounds were also shown in Ref.
[5] for a different contact time.
Mechanochemical Experiments with Sr(OH)2
So far, all of our milling experiments were performed with
Sr(OH)2·8H2O as source for Sr2+ cations. Sr(C8H4O4)·3H2O
and Sr(C8H4O4) were accessible on this way, but the monohy-
drate, also described in Ref. [3] was never formed during a
mechanochemical reaction.
Therefore, further milling experiments were performed with
the commercially available water-free hydroxide Sr(OH)2 and
terephthalic acid (H2BDC). With a chosen molar ratio of 1:1
[Sr(OH)2:H2BDC] and milling times of one and four hours,
respectively, we instantly obtained new powder patterns. Re-
flections of the starting materials disappeared and the new re-
flections, depicted in Figure 4a, completely agree with the
powder pattern of Sr(C8H4O4)·H2O, recalculated from single
crystal data given in Ref. [3] (Figure 4b). The product obtained
after 1 h of milling shows a worse crystallinity and is not im-
plemented in Figure 4. Attempts to induce a reaction manually
in a mortar were not favorable. Choosing a molar ratio of 1:2
[Sr(OH)2:H2BDC] the Sr-terephthalate monohydrate is also
formed, but expectedly, reflections belonging to un-consumed
H2BDC are still visible in the powder diffractogram (not
shown here).
1
Figure 5. H-13C CP MAS NMR spectrum of Sr(C8H4O4)·H2O, pre-
pared by milling Sr(OH)2 with H2BDC (molar ratio: 1:1) for 4 h. For
comparison, the spectra of Sr(C8H4O4)·3H2O and Sr(C8H4O4), already
shown in Ref. [5] for another contact time, are added. (νrot = 10 kHz,
p15 = 6 ms).
Two carbon signals with quite different intensity belong to
carboxyl groups (δ =175.8 ppm, 172.8 ppm), and three sites
were resolved for aromatic carbon atoms (δ =137.8 ppm,
1
131.6 ppm, 128.7 ppm) for Sr(C8H4O4)·H2O. Comparing H-
13C CP MAS NMR spectra of Sr(C8H4O4)·H2O with those of
the trihydrate and the water-free compound (Figure 5), the sig-
nals at δ = 175.8 ppm and 128.7 ppm can be assigned to more
strongly hydrogen bridged carboxyl carbon and aromatic car-
bon atoms, respectively. 13C signal intensities recorded for
Sr(C8H4O4)·H2O show a similar dependence on contact times
as shown for the calcium terephthalate compounds in
Figure S3.
The FT IR spectra too are quite similar to those shown in
the supporting information in Ref. [5] for Sr(C8H4O4)·3H2O
and Sr(C8H4O4). The fraction of broad vibration bands above
3000 cm–1 lies, as expected, for the monohydrate, between
Figure 4. X-ray powder diffractograms of the product obtained by
milling of Sr(OH)2 with H2BDC (molar ratio: 1:1) for 4 h (a) and of
Sr(C8H4O4)·H2O (b), recalculated from single crystal data.
320
www.zaac.wiley-vch.de
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2014, 317–324

Alkaline Earth Metal Terephthalates: M(C8H4O4)·nH2O
those of the trihydrate and the water-free compound. For com-
pleteness, the FT IR spectrum of Sr(C8H4O4)·H2O is shown in
the Supporting Information (Figure S5).
Whereas the mechanochemical synthesis of Sr(C8H4O4)·
H2O just by milling for 4 h performs well starting from
Sr(OH)2 and H2BDC, the synthesis of a Ca-terephthalate was
unsuccessful under the same conditions. From this, for Barium,
as an even larger alkaline earth metal cation, an easy and fast
access to Ba-terephthalates applying even shorter milling times
should be expected.
Barium Terephthalates Ba(C8H4O4)·nH2O
From the literature two Ba-terephthalate compounds are
known. These are the water-free Ba(C8H4O4),[4] and
Ba(C8H4O4)·4H2O.[2] In both cases the single crystals were
grown from aqueous solution, and their mechanochemical syn-
thesis was not described so far. Our mechanochemical
experiments aimed at the formation of Ba(C8H4O4) or
Ba(C8H4O4)·nH2O compounds.
We started from Ba(OH)2, Ba(OH)2·H2O, and Ba(OH)2·
8H2O as sources for Ba2+ ions, which all are commercially
available, and milled them with terephthalic acid. XRD powder
patterns of products obtained by milling of barium hydroxides
or hydroxide hydrates with terephthalic acid are collected in
Figure 6.
As deduced above, using Ba(OH)2 1 h of milling is already
sufficient to obtain reflections, which are different to those of
the starting materials (Figure 6a). The same set of reflections
was obtained using Ba(OH)2·H2O as reactant and a milling
time of 4 h (Figure 6b). The reaction with Ba(OH)2·H2O is
also successful after 1 h of milling, however the longer milling
time results here in a better crystallinity. A comparison of
these powder patterns with the powder diffractogram of
Ba(C8H4O4), recalculated from single crystal data from
Ref. [4] (Figure 6c), gives unambiguous evidence for the for-
mation of the water-free barium terephthalate Ba(C8H4O4).
These structural findings are supported by thermoanalytical
Figure 6. X-ray powder diffractograms of the products obtained by
milling of (a) Ba(OH)2 with H2BDC (molar ratio: 1:1) for 1 h, (b)
Ba(OH)2·H2O with terephthalic acid (molar ratio: 1:1) for 4 h, (c) pow-
der diffractogram of Ba(C8H4O4), recalculated from single crystal
data,[4] (d) powder diffractogram of Ba(C8H4O4)·4H2O, calculated
from own single crystal structure data, (e) Ba(OH)2·8H2O with H2BDC
(molar ratio: 1:1) for 4 h.
measurements. DTA-TG curves of the milling products display sulted in a sample with a completely new powder pattern (Fig-
an only marginal mass loss of about 3% near 300 °C (Figure ure 6e), not identical with that of Ba(C8H4O4) (Figure 6a–c)
S6, Supporting Information) in the case of Ba(OH)2 as precur- and also not in agreement with the powder pattern of
sor. Using Ba(OH)2·H2O as precursor, the mass loss is a bit Ba(C8H4O4)·4H2O, recalculated from own and literature single
higher (≈ 6%) and corresponds to the loss of one water crystal data[2] (Figure 6d). The same new powder pattern (Fig-
molecule assuming formally a formula of Ba(C8H4O4)·H2O ure 6e), but with a worse signal to noise ratio (worse crystal-
(Figure S6b). The release of water does not occur not below linity) and slightly changed relative intensities of the reflec-
300 °C, which is high compared to Ca- and Sr-terephthalates tions, is obtained with shorter milling times (1 h, not shown
(cf. Figure S2a, Figure S4, and Figure 2 in Ref. [5]). However, here). On the other hand, longer milling times, here 8 h, lead
with the knowledge of the crystalline phases identified in both to a further removal of water, and, as a consequence, the new
cases as Ba(C8H4O4) (Figure 6a–c) residues of water or even product decomposes. The final powder pattern indicates the
residues of X-ray amorphous hydroxides might be assumed to formation of Ba(C8H4O4), i.e., the water-free terephthalate.
be present in the sample (Figure 6b, Figure S6b). The latter
The thermal behavior of the new product (Figure 6e) is
may also explain small deviations in the carbon and hydrogen given in Figure 7. The total mass loss of 8.2% up to 400 °C
contents for the mechanochemically synthesized Ba(C8H4O4) does not correspond to a trihydrate, where a loss of about 15%
samples (Table S1), which were both found to be slightly too should be expected. Moreover, for a tetrahydrate, known from
low.
the literature,[2] the mass loss should be with 19.3% even
Milling experiments (milling time: 4 h, molar ratio: 1:1) higher. However, an assumed sum formula of Ba(C8H4O4)·
using Ba(OH)2·8H2O and terephtalic acid as precursors re- 1.5H2O exactly fits to this mass loss, and also Ba(C8H4O4)·
Z. Anorg. Allg. Chem. 2014, 317–324
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
321

G. Scholz, A. Abdulkader, E. Kemnitz
ARTICLE
2H2O, with a mass loss of 10.7% at water release, is nearby.
The same intermediate result is obtained with elemental analy-
sis (Table S1). Bearing in mind that the carbon content is usu-
ally found a bit too low, the values obtained by thermal and
elemental analysis along with the powder pattern (Figure 6e)
can be allocated either to Ba(C8H4O4)·2H2O or to
Ba(C8H4O4)·1.5H2O.
Figure 7. Thermoanalytical curves of the new product prepared by
milling Ba(C8H4O4)·8H2O with H2BDC (molar ratio: 1:1) for 4 h;
(m18: H2O, m44: CO2).
Figure 8. 1H-13C CP MAS NMR spectra of samples prepared by mill-
ing of: (a) Ba(OH)2 with H2BDC (molar ratio: 1:1) for 1 h,
(b) Ba(C8H4O4)·H2O with H2BDC (molar ratio: 1:1) for 4 h,
(c) Ba(C8H4O4)·8H2O with H2BDC (molar ratio: 1:1) for 4 h (νrot
10 kHz, p15 = 6 ms).
=
Attempts to grow single crystals from the new milling prod-
uct were successful. However, their structure can be unambig-
uously identified as Ba(C8H4O4)·4H2O in agreement with the
crystal structure database.[2] The calculated powder pattern
(Figure 6d) shows that the formed single crystal does not
correspond to the powder sample obtained originally by mill-
ing (Figure 6e).
few microliters of water to the mixture of precursor com-
pounds. With liquid assisted grinding the milling time can be
considerably reduced to only 1 h and phase pure Ca(C8H4O4)·
3H2O is obtained. A thermal post-treatment of the trihydrate
results in Ca(C8H4O4).
1
All H-13C CP MAS spectra of the products obtained from
mechanochemical reactions of Ba(OH)2, Ba(OH)2·H2O, and
Ba(OH)2·8H2O with H2BDC are summarized in Figure 8. In
agreement with XRD patterns these spectra are identical for
the samples obtained by mechanochemical reactions with
Ba(OH)2 and Ba(OH)2·H2O (Figure 8a, b) and are the typical
signals of Ba(C8H4O4). Carbon resonances of the third product
(Figure 8c) are different and indicate the formation of a new
Ba(C8H4O4)·nH2O, which is assumed to be Ba(C8H4O4)·2(1.5)
H2O.
In addition to the mechanochemical synthesis described for
Sr(C8H4O4)·3H2O in Ref. [5] with Sr(OH)2·8H2O as starting
material,
a
successful mechanochemical formation of
Sr(C8H4O4)·H2O is possible starting from Sr(OH)2 and tereph-
thalic acid in a molar ratio 1:1. The use of the water free hy-
droxide requires a stronger mechanical impact (4 h of milling)
for a successful reaction. Recrystallization experiments re-
sulted in Sr(C8H4O4)·4H2O and showed that Sr(C8H4O4)·3H2O
is only accessible on the mechanochemical route.
The mechanochemical synthesis of Ba(C8H4O4) is very easy
both with Ba(OH)2 and Ba(OH)2·H2O as starting materials.
Starting with Ba(OH)2·8H2O at milling the formation of a new
Conclusions
With the presented study we have shown that a mechano- barium terephthalate hydrate, so far identified as Ba(C8H4O4)·
chemical synthesis of alkaline earth metal terephthalates 2(1.5)H2O, can be reported.
M(C8H4O4)·nH2O (M = Ca, Sr, Ba) is in general possible. A
Also in the barium system, the single crystals obtained after
successful mechanochemical reaction depends on the cation, recrystallization, possess a structure and composition different
the water supply of the precursor compounds, and the milling from the powder obtained after milling. It seems that chemical
impact.
In general, the necessary milling time decreases with in- pounds not accessible by classical solution chemistry.
creasing cationic radius. Thus, 8 h of milling are necessary for The simplified reaction paths for the successful mechano-
reactions induced in a planetary mill open a way to new com-
the synthesis of Ca(C8H4O4)·3H2O from Ca(OH)2 and tereph- chemical formation of alkaline earth metal terephthalates are
thalic acid, and, the crystallinity of the product is bad. This summarized in Scheme 1. Further milling experiments with or-
situation can be distinctly improved after the addition of only ganic linkers different from terephthalic acid are in progress.
322
www.zaac.wiley-vch.de
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2014, 317–324

Alkaline Earth Metal Terephthalates: M(C8H4O4)·nH2O
Scheme 1. Reaction paths for the formation of M(C8H4O4)·nH2O. Time on the arrows indicates the respective milling time (see Experimental
Section).
Thermal Analysis: The thermal behavior was studied with
a
Experimental Section
NETZSCH thermoanalyzer STA 409 C Skimmer®, being additionally
equipped with a conventional high-temperature SiC oven. A DTA-TG
sample carrier system with platinum crucibles (baker, 0.8 mL) and Pt/
PtRh10 thermocouples were used. Measurements were performed in a
nitrogen atmosphere applying a heating rate of 10 K·min–1.
Preparation: All mechanochemical reactions were performed in a
commercial planetary mill “Pulverisette 7 premium line” (Fritsch, Ger-
many) under access of air. Each silicon nitride grinding bowl was filled
with 1 g of a powder mixture and assembled with five silicon nitride
balls. A ball to powder mass ratio of 14.5 was ensured. All samples
were milled with a rotational speed of 600 rpm and different milling
times, mentioned in the text.
Elemental Analysis: Carbon, nitrogen, and hydrogen contents were
determined with an EURO EA equipment (HEKAtech GmbH).
FT IR: IR spectra were recorded in transmission in the range of 4000–
750 cm–1 using KBr pellets. The spectrometer system was a Perkin-
Elmer FT-PE Spectrum System 2000.
Commercially accessible powders of Ca(OH)2, Sr(OH)2, Sr(OH)2·
8H2O, Ba(OH)2, Ba(OH)2·H2O, and Ba(OH)2·8H2O were used as
alkaline earth metal sources. They were milled with terephthalic acid
[1,4-(COOH)2–C6H4, (H2BDC), Aldrich, chemical purity 98%] in a
molar ratio of 1:1 if not otherwise indicated.
Supporting Information (see footnote on the first page of this article):
FT-IR- and MAS NMR spectra, thermoanalytical curves and results of
the elemental analyses for the samples under consideration.
In addition, simple attempts by grinding the two precursors together
in a mortar were done.
Acknowledgements
The products obtained by milling Sr(OH)2·8H2O [or Ba(OH)2·8H2O]
with H2BDC (1:1) were recrystallized from water and cubic colorless
crystals were formed. This was done by adding about 3 mL of water
to a small amount (one small spatula) of the product, shaking with
hand, and waiting until two different layers (aqueous and non aqueous)
were formed. Some white particles precipitated down. The two layers
were removed by a glass pipette into a new vial. Almost 7–10 d later
the crystals were formed.
U. Kätel and Dr. M. Feist are kindly acknowledged for performing
elemental analysis and thermal analysis, respectively. Dr. F. Emmerling
(BAM Berlin) and Dr. A. Dimitrov are kindly acknowledged for struc-
tural discussions.
References
XRD: X-ray diffractograms were recorded with a XRD-3003-TT dif-
fractometer (Seiffert & Co., Freiberg) with Cu-Kα radiation (λ =
1.542 Å; 2 θ range: 5° Յ 2 θ Յ 64°; step scan: 0.05°, step time: 5 s).
Reflections were compared with diffractograms of the JCPDS-PDF
data base.[12]
[1] T. Matsuzaki, Y. Iitaka, Acta Crystallogr., Sect. B 1972, 28, 1977.
[2] R. H. Groenemann, J. L. Atwood, Cryst. Eng. 1999, 2, 241.
[3] L. Yang, D. Zhao, G. Li, Acta Crystallogr. Sect. E 2011, 67, m282.
[4] S. M. F. Lo, S. S. Y. Chui, I. D. Williams, Acta Crystallogr., Sect.
C 1998, 54, 1846.
[5] G. Scholz, F. Emmerling, M. Dreger, E. Kemnitz, Z. Anorg. Allg.
Chem. 2013, 639, 689.
[6] P.-C. Liang, H.-K. Liu, C.-T. Yeh, C.-H. Lin, V. Zima, Cryst.
Growth Des. 2011, 11, 699.
[7] M. Mazaj, G. Mali, M. Rangus, E. Zunkovic, V. Kaucic, N. Z.
Logar, J. Phys. Chem. C 2013, 117, 7552.
MAS NMR Spectroscopy: 1H-13C CP MAS NMR spectra were re-
corded with a Bruker AVANCE 400 spectrometer (Larmor frequency:
ν13C = 100.6 MHz) with a 4 mm MAS probe (Bruker Biospin). 13C
1
chemical shift values are given with respect to TMS. All H-13C CP
MAS NMR spectra were taken with different contact times (p15) rang-
ing from 200 μs to 8.5 ms.
[8] T. Friscic, L. Fabian, CrystEngComm 2009, 11, 743.
Z. Anorg. Allg. Chem. 2014, 317–324
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
323

G. Scholz, A. Abdulkader, E. Kemnitz
ARTICLE
[9] F. C. Strobridge, N. Judas, T. Friscic, CrystEngComm 2010, 12,
2409.
[10] K. Fujii, A. L. Garay, J. R. Hill, E. Sbircea, Z. Pan, M. C. Xu,
D. C. Apperley, S. L. James, K. D. M. Harris, Chem. Commun.
2010, 46, 7572.
[11] P. Wang, G. Li, Y. Chen, S. Chen, S. L. James, W. Yuan,
CrystEngComm 2012, 14, 1994.
[12] JCPDS-ICDD, International Centre for Diffraction Data: PDF-2
Database (Sets 1–51 plus 70–89). PA 19073–3273 USA., Release
2001. – PCPDFWIN Version 2.2.
Received: October 23, 2013
Published Online: January 14, 2014
324
www.zaac.wiley-vch.de
© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2014, 317–324