Small-angle X-ray scattering documents the growth of metal-organic frameworks

Article abstract of DOI:10.1016/j.cattod.2012.08.044

We present a combined in situ small- and wide-angle scattering (SAXS/WAXS) study on the crystallization of two topical metal-organic frameworks synthesized from similar metal and organic precursors: NH₂-MIL-53(Al) and NH₂-MIL-101(Al). A thorough analysis of SAXS data reveals the most important phenomena occurring during crystallization and unravels the effect of the solvent. NH₂-MIL-53(Al) growth follows two routes: (i) through direct hydrolysis of AlCl₃·6H₂O in water, and (ii) via the intermediate NH₂-MOF-235(Al), which forms in pure DMF or DMF/H₂O mixtures. In the case of pure H₂O as solvent, formation of NH₂-MIL-53(Al) crystals proceeds through steady growth in all three dimensions. The addition of DMF to the synthesis mixture results in amorphous scattering entities forming very rapidly and subsequently arranging into the intermediate phase, NH₂-MOF-235(Al). In DMF/H₂O mixtures, amorphous precursors develop in rapid fashion with fractal character dominating, followed by densification, crystallization of NH₂-MOF- 235(Al) and slow transformation into NH₂-MIL-53(Al). Formation of NH₂-MIL-101(Al) only occurs when pure DMF is used as solvent, and it always proceeds through the formation of the intermediate NH₂-MOF- 235(Al). In this case a smooth scatterer surface is observed, with morphology and size constant in time.

Full text of DOI:10.1016/j.cattod.2012.08.044

Catalysis Today 205 (2013) 120–127
Contents lists available at SciVerse ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Small-angle X-ray scattering documents the growth of metal-organic frameworks
Maarten G. Goestena,∗, Eli Stavitskib, Jana Juan-Alcan˜iza, Alberto Martin˜ez-Joaristia,
Andrei V. Petukhovc, Freek Kapteijna, Jorge Gascona,∗
a Catalysis Engineering, ChemE, TU Delft, Julianalaan 136, 2628BL, Delft, The Netherlands
b National Synchrotron Light Source, Brookhaven National Laboratory, Upton, NY 11973, USA
c Van’t Hoff Laboratory for Physical and Colloid Chemistry, Debye Institute for Nanomaterial Science, Utrecht University, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 15 June 2012
Received in revised form 14 August 2012
Accepted 19 August 2012
Available online 18 October 2012
We present a combined in situ small- and wide-angle scattering (SAXS/WAXS) study on the crystalliza-
tion of two topical metal-organic frameworks synthesized from similar metal and organic precursors:
NH2-MIL-53(Al) and NH2-MIL-101(Al). A thorough analysis of SAXS data reveals the most important
phenomena occurring during crystallization and unravels the effect of the solvent. NH2-MIL-53(Al)
growth follows two routes: (i) through direct hydrolysis of AlCl3·6H2O in water, and (ii) via the inter-
mediate NH2-MOF-235(Al), which forms in pure DMF or DMF/H2O mixtures. In the case of pure H2O
as solvent, formation of NH2-MIL-53(Al) crystals proceeds through steady growth in all three dimen-
sions. The addition of DMF to the synthesis mixture results in amorphous scattering entities forming
very rapidly and subsequently arranging into the intermediate phase, NH2-MOF-235(Al). In DMF/H2O
mixtures, amorphous precursors develop in rapid fashion with fractal character dominating, followed by
densification, crystallization of NH2-MOF-235(Al) and slow transformation into NH2-MIL-53(Al). Forma-
tion of NH2-MIL-101(Al) only occurs when pure DMF is used as solvent, and it always proceeds through
the formation of the intermediate NH2-MOF-235(Al). In this case a smooth scatterer surface is observed,
with morphology and size constant in time.
Keywords:
Metal organic framework
Crystallization
SAXS/WAXS
In situ
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
discovered by fortune rather than by design [36,40,41]. As the
mechanism of crystallization is of utmost importance to crystal
Metal organic frameworks (MOFs) are porous coordination
polymers built up by bridging ligands that coordinate to metal ions
or clusters. The high stability of some MOFs combined with a large
degree of structural and chemical versatility makes them attractive
for a number of applications, such as separation [1–5], gas stor-
age [6–10], catalysis [11–15], drug-delivery [16–20], luminescence
[21–25] and magnetism [26–30].
Although reports on coordination polymers go back to the
1960s, it was only after Robson and Hoskins re-discovered and
boosted the field in 1989 [31] when coordination polymers rapidly
developed, with reports on MOFs with very large surface areas,
uncoordinated metal sites, and sometimes exceptionally high
chemical and thermal stabilities [32–39].
engineering and to the application of these materials in tools and
devices, understanding crystallization becomes a very important
matter.
Scattering techniques are capable of providing detailed informa-
tion about in situ crystal growth and nucleation, and information
from nano- to microscale about size and morphology. The most
powerful scattering techniques are small-angle X-ray scattering
(SAXS) and small-angle neutron scattering (SANS). SAXS provides
high-resolution scattering profiles for systems and events occur-
ring at a wide scale range (1–200 nm). Moreover, simultaneous
recording of scattering at small and wide angles (SAXS/WAXS)
is possible by using multiple detectors. The formation and mor-
phology of particles or crystals, i.e. scattering entities, can be
investigated by SAXS, and their crystalline properties by WAXS,
both in a time-resolved fashion.
Although the number of SAXS based studies on MOFs is small,
interesting investigations have been published in recent years. Tsao
et al. were the first to apply SAXS for the study of MOF systems,
more specifically, MOF-5, mainly providing detailed nanostructural
information about the material [42,43]. Topological information, in
line with the work of Tsao was presented by Klimakow et al. on
mechanosynthesis; they extended structural analysis with quali-
tative use of SAXS [44].
Although a certain extent of control over MOF formation is
claimed and the terminology of ‘building units design’ and ‘reticular
synthesis’ were coined at the beginning of this century, unpre-
dictability in synthesis does play a role, and it would not be
overly controversial to state that several new topologies have been
∗
Corresponding authors.
E-mail addresses: m.g.goesten@tudelft.nl (M.G. Goesten), j.gascon@tudelft.nl
(J. Gascon).
0920-5861/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2012.08.044

M.G. Goesten et al. / Catalysis Today 205 (2013) 120–127
121
Fig. 1. MOF structures under study: in NH2-MIL-53(Al) Al ions are linked by ␮2-OH bridges that yield 1D chains. In NH2-MIL-101, ␮3-O bridges form a trimeric secondary
building unit (SBU). NH2-MOF-235(Al) is built up from the same trimeric ␮3-O bridged clusters that form NH2-MIL-101(Al).
In attempt at resolving the mechanism of MOF growth, which
is also the scope of this work, three papers using in situ SAXS
have been recently published. Cravillon et al. studied the forma-
tion of ZIF-8 to obtain time-resolved structural information and
demonstrated that nucleation governs the growth of this material
[45]. In similar fashion, our group modelled the kinetics for the
competitive growth between two frameworks built up by identical
precursors: NH2-MIL-53(Al) and NH2-MIL-101(Al) (see Fig. 1) [46],
in which the former is the thermodynamic product and the lat-
ter the kinetic product. In addition, the templation effect on MOF
growth by phosphotungstic acid, one of the best-known Keggin-
type polyoxometalates, were revealed [47]. Molecular-scale effects,
events and growth of MOFs were also revealed through use of other
synchrotron-based techniques, energy dispersive X-ray diffraction
(EDXRD) and X-ray absorption spectroscopy, as published by Wal-
ton et al. [48–53].
ions at once, forming a trimeric secondary building unit (SBU) (see
Fig. 1) [38].
In this work, we present a thorough analysis of previous
and newly collected in situ SAXS/WAXS data collected during
the crystallization of NH2-MIL-101(Al) and NH2-MIL-53(Al). The
formation of MOF precursors and their transformation to small
MOF crystals is studied in detail at a length scale of the order
1–70 nm. Several parameters are determined describing particle
morphology, i.e. radius of Gyration, Porod volume and surface-to-
volume as a function of time, demonstrating that crystallization
of two phases follows very different chemical and morphological
pathways.
2. Experimental
NH2-MIL-53(Al) and NH2-MIL-101(Al) are two MOFs with out-
standing adsorptive, catalytic and optical properties [54–63]. NH2-
MIL-53(Al) with chemical formula Al(OH){O2C C6H3 NH2 CO2},
similar to the unfunctionalized MIL-53 structure, is built up
by from diamond-shaped channels and is the only product
when pure water is the solvent of synthesis. NH2-MIL-101(Al)
is a mesoporous framework with a MTN topology and formula
Al3(O)[DMF]3{O2C C6H3 NH2 CO2}3. Although both MOFs grow
from the same precursors, the aluminium coordination is different
[38,61]. In the case of MIL-53 Al ions are linked by ␮2-OH bridges
that yield 1D chains, whereas for MIL-101 ␮3-O bridges links three
Aluminium chloride (AlCl3·6H2O, Fluka, purum p.a.), 2-
aminoterephthalic acid (Aldrich, 99%) and DMF (Sigma, 99.8%
anhydrous) were used as received. Metal and organic precursors
were first dissolved in DMF or water and then the solutions were
mixed. Synthesis solutions were prepared with the ratios depicted
in Table 1. After vigorous stirring for 5 min, 0.1 ml aliquot was trans-
ferred to the in situ cell.
A specially designed synthesis cell was used for the experi-
ments (Fig. S1). The synthesis solution was loaded between two
mica windows (10 mm diameter, 50–100 ␮m thickness) separated
by PTFE spacers (1.5 mm thickness). Heating was provided by four
Table 1
Synthesis conditions.
Reaction products/solvent
AlCl3·6H2O mass
2-Aminoterephthalic
acid mass (g/solvent)
(g/solvent)
MIL-101 synthesis in DMF
MIL-53 synthesis in H2O
MIL-53 in DMF:water 0.70:0.30
MIL-53 in DMF:water 0.90:0.10
MIL-53 in DMF:water 0.95:0.05
MIL-53/MIL-101 in DMF
0.507/15 ml DMFa
1.97/10 ml H2Ob
1.97/6 ml H2Ob
1.97/2 ml H2Ob
1.97/1 ml H2Ob
1.97/10 ml DMFb
0.564/15 ml DMFa
1.5/10 ml H2Ob
1.5/14 ml DMFb
1.5/18 ml DMFb
1.5/19 ml DMFb
1.5/10 ml DMFb
a
“Low concentration of precursor” in the main text.
“High precursor concentration”.
b

122
M.G. Goesten et al. / Catalysis Today 205 (2013) 120–127
chemically resistant cartridges and temperature was controlled at
the external wall of the PTFE inserts.
I (0)
Time resolved SAXS–WAXS experiments were performed at
the BM16 beamline of the European Synchrotron Radiation Facil-
ity, ESRF (Grenoble, France), and X9 beamline at the NSLS in
Brookhaven (NY, USA). A bending magnet BM16b at ESRF was
equipped with a Roper Scientific CCD camera PI SCX90-1300
mounted on a Newport RV240PE goniometer, and acquisition
occurred with an X-ray research marCCD detector for WAXS and
SAXS measurements, respectively. Indulator-based X9 at NSLS
utilizes the marCCD SAXS detector and Photonic Science WAXS
detector.
With the high-flux synchrotron radiation we were able to collect
SAXS and WAXS patterns simultaneously with an excellent signal-
to-noise ratio with a 20 s time resolution. The data were normalized
for the intensity of the X-ray beam and corrected for detector sen-
sitivity prior to background correction. The background scattering
was subtracted.
Treated dataset
Experimental dataset
’x2’
0.01
0.1
1
q / nm-1
Fig. 2. Experimental and treated SAXS dataset of amino MIL-101, in an I(q) vs. log (q)
plot. The treated dataset (gray) is the same as the experimental dataset, but has
additional extrapolated values for q going to zero, and q going to infinity, background
values are subtracted. Values for I(0) and background parameter x2 are indicated.
3. Theory
3.1. SAXS analysis
functions and parameters that can thus be calculated in this time-
resolved experiment are: p(r), Rg, V, and S/V. An example of this
data treatment is given in Fig. 2.
The inverse Fourier transform of the intensity profile yields ꢀ(r),
which is directly related to electron pair distance distribution func-
tion p(r), a distinct characteristic for the morphology of a scattering
entity (see SI). The p(r) function is used to obtain information about
the morphology of the scatterer as a function of time, during the
growth process.
When dealing with crystals that do not possess any high aspect
ratios, deviation from the fourth power law of Porod might occur
when the surface of the scatterer is not smooth. In that case I(q)
obeys behaviour of the form q6−D with Ds representing the dimen-
s
sionality of the surface fractal of the scatterer. Power law decay of
the form q−˛ with ˛ = 4 indicates a scatterer with a smooth surface.
When ˛ < 4, one deals with rougher surfaces. As ˛ approaches 3,
one deals with surfaces of dominant fractal character. Fractal sur-
face development is an important concept for MOFs, which are built
up from smaller building blocks (e.g. trinuclear clusters) that have
their influence on full crystal scale.
ꢀ
∞
r2
V
1
2ꢁ2
sin (qr)
qr
p(r) = r2ꢀ(r) =
q2dq · I(q)
(1)
0
To perform FT analysis, a smooth dataset spanning from 0 to ∞
is needed, but experimental SAXS data have a cut-off at the beam
stop at small q-value, plus (often) a nonzero background value at
large q, resulting in infinite values for the Fourier integral. The first
problem can be solved by extrapolating SAXS data to q = 0 by using
an approximation derived by Guinier in 1939 for singe particle
scattering.
3.2. Kinetic analysis
Analysis of the kinetic profiles was performed using the model
developed by Gualtieri [65] and applied by Millange et al. [66] for
the formation of several prototypical MOFs. This model is based on
decoupling the nucleation and crystal growth processes. In such a
way, the extent of crystallization (C) with time (t) can be simulated
according to the following equation:
ꢁ
ꢂ
−(qRg)2
I1(q) = ꢂn2exc exp
(2)
3
Here, ꢂnexc represents a contrast parameter, the excess of electrons
of the scatterer relative to its surroundings, and Rg is the radius of
gyration, which is indicative for the size of the scatterer.
Further, the Porod volume V is determined, an indicator for the
volume of the scatterer:
ꢃ
ꢄ
1
C =
(1 − exp(−(kgt)n)
(5)
1 + exp(−((t − a)/b))
Fitting a kinetic curve with this expression yields the parameter-
ization of a, b and kg (growth rate constant) and indirectly the
nucleation rate constant, kn = 1/a. In this work, as inferred from
SAXS data evaluation, framework topology and the bi-pyramidal
(MIL-101) and rhombohedral (MIL-53) shape of the final crystals
and as suggested earlier by Millange et al. [48], it is assumed that
MOF crystals grow in three dimensions (n = 3) and that nucleation
is heterogeneous (b = 15).
2ꢁ2 · I1(0)
V =
(3)
Q1
Finally, the scattering intensity is inversely proportional to the
fourth power of q, as deduced by Porod in 1951 [64], which can
be used to determine the specific surface area of the scatterer, i.e.
S/V.
S
I(q)q4
Q
= ꢁ × lim
(4)
V
q→∞
4. Experimental results
Detailed information on the equations and the role of ‘invariant’ Q
in Eqs. (3) and (4) is available in the supplementary information
(SI).
The power law in Eq. (4) is used to extrapolate the q-space to
infinity, using I(q) = x1q−4 + x2. Here, x1 represents SQ/ꢁV while x2
corrects for incomplete subtraction of the background from exper-
imental data. This data treatment yields a smooth dataset without
Bragg peaks suitable for the application of Eqs. (2) and (4). The
Fig. 3 presents the time-resolved in situ WAXS and SAXS pat-
terns recorded at 130 ◦C during the crystallization using high
precursor concentrations in pure water (see Table 1). Under these
conditions, only NH2-MIL-53(Al) is formed, as inferred from the
Bragg peak at q = 6.2 nm−1 which corresponds to the (2 0 0) plane
of NH2-MIL-53(Al) in its narrow pore form [55]. First evidence
of the MIL-53 phase formation can be observed at ∼700 s. The

M.G. Goesten et al. / Catalysis Today 205 (2013) 120–127
123
Fig. 3. Time evolution of the WAXS pattern during the in situ synthesis of NH2-MIL-
53(Al) in H2O @130 ◦C, with corresponding SAXS decays in log–log form (inset).
SAXS decay, as shown in Fig. 3, follows power law behaviour q−˛
with ˛ increasing from 3.55 to 3.95 with time. These values are
typical for growth of porous phases, as described for porous zeo-
lites, indicating two processes, i.e. smoothening of the surface of
the scattering particles and densification of the internal structure
[67–69].
Fig. 5. p(r) function during the synthesis of NH2-MIL-53. The blue arrow indicates
the increase of intensity and radius of gyration Rg. (For interpretation of the refer-
ences to color in this figure legend, the reader is referred to the web version of the
article.)
Porod volume development (Eq. (3)), is displayed in Fig. 4. The
initial (first 200 s) drop represents dissolution of residual linker
crystallites during temperature stabilization in the crystallization
cell. Linker dissolution is followed by steady particle growth occur-
ring in parallel with the development of the (2 0 0) reflection (Fig. 3).
Crystallite morphology can be deduced from the calculated electron
distance distribution function p(r) (Fig. 5). The particle size growth
is clearly reflected by the increasing intensity, while the maximum
electron pair distance gives information about the shape of the par-
ticles. NH2-MIL-53(Al) crystals possess an elongated character, in
good agreement with the asymmetric curve shape of the p(r) func-
tion (cf. Figs. 5 and 12 in SI) [64,70]. The intensity grows in time
as new crystals are created in a process of continuous nucleation.
The maximum of the curve at r = Rg corresponds with the radius of
gyration. The highest value of r corresponds to the maximum length
of the scatterer, i.e. the particle size.
The increase in scatterer volume can be associated to nucleative
crystal growth, in which the formation of the first primary units is
the rate-limiting step [46]. Time-resolved plots for Rg and S/V are
given in the SI, with trends similar to the Porod volume in Fig. 4. The
constant shape of the p(r) function with time (SI) is attributed to the
fact that on average the particle morphology does not change, as
expected for growth in three dimensions. It is emphasized however,
that the average size of crystals observed during our experiments
is always smaller than 90 nm because of the detection limits of the
detector (the maximum detectable particle size follows 2ꢁ/q).
Contrary to aqueous media, synthesis in pure DMF with low pre-
cursor concentrations (see Table 1) yields NH2-MIL-101(Al) phase.
Fig. 4. Development of the Porod volume of scatterers during synthesis of NH2-MIL-53(Al) in H2O.

124
M.G. Goesten et al. / Catalysis Today 205 (2013) 120–127
Fig. 8. p(r) function shows constant scatterer intensity and morphology during
formation of NH2-MIL-101(Al) in DMF, suggesting a slow solid-to-solid conversion.
Fig. 6. WAXS pattern with the formation of NH2-MIL-101(Al) in DMF and corre-
sponding SAXS decay (inset).
Fig. 9 depicts the evolution in the WAXS and SAXS patterns for
syntheses performed in DMF:H2O = 9:1 mixtures at 130 ◦C at high
precursor concentrations. The Bragg peak of NH2-MOF-235(Al) is
clearly visible right from the beginning of the synthesis (Fig. 9), and
its fast appearance is ascribed to the higher precursor concentra-
tions used (see SI).
Decay parameter values initially increase, but then decrease
from ˛ = 3.55 to 3.00, pointing towards development of a surface
with fractal (rougher) character. The scattering intensities do how-
ever develop in time, pointing at an increasing number of scattering
entities.
The time evolutions of the WAXS pattern and SAXS decay, recorded
at 130 ◦C, are shown in Fig. 6.
Before the MIL-101 crystalline planes, with dominant contrib-
utions by (1 1 1) and (1 1 3) planes [61], start to develop around
500 s, evidence for the early formation of the NH2-MOF-235(Al)
phase is clearly observed. MOF-235(Fe) was firstly described by
Yaghi et al. [71] and identified as a precursor in the formation of
MIL-53(Fe) by Walton et al. [66]. Later, we reported the forma-
tion of the isostructural NH2-MOF-235(Al) at early times during
the formation of NH2-MIL-101(Al) and NH2-MIL-53(Al) in DMF and
DMF/water mixtures [46].
The evolution of the Porod volume (Fig. 10) provides perspec-
tive again. Dissolution of the linker is followed by rapid growth
of volume of scattering entities (this is also seen for the Radius of
Gyration, see SI), which in turn precedes gentler decline in volume.
Comparing this with Fig. 9, the Bragg peak of NH2-MOF-235(Al)
really intensifies at the moment the Porod volume decay starts
levelling off (∼1500 s) and gets more stable in time. Striking is the
‘step’ in intensity drop over the complete width of the q-length scale
which accompanies this sudden crystallization; this is explained
as an effect of the sudden formation of a porous crystalline struc-
ture from the non-crystalline scatterer. Further, the calculated S/V
Strikingly, SAXS decay does not vary in time exhibiting a con-
stant slope with ˛ = 3.8. This value is indicative of scattering
particles with a smooth surface. The Porod volume development
(Fig. 7) initially displays some dissolution of the linker, but after that
the volume of the scatterers remains constant, in agreement with
calculated, invariant in time, values for Rg and S/V (see SI). The p(r)
function–also in strong contrast to MIL-53–shows constant inten-
sity and shape. This implies that scattering entities with ∼25 nm
radius are formed rapidly before the measurements begin, with
the diffraction peaks of NH2-MOF-235(Al) and NH2-MIL-101(Al)
developing at about 500 s and 1500 s, respectively (Fig. 8).
Fig. 7. Porod volume of scatterers during synthesis of amino MIL-101(Al) in DMF.
Fig. 9. Time evolution of the WAXS pattern with the formation of NH2-MOF-235(Al)
No volumetric change is observed during growth of the crystalline phase.
in DMF:H2O = 9:1 and corresponding SAXS decay (inset).

M.G. Goesten et al. / Catalysis Today 205 (2013) 120–127
125
Fig. 10. Time evolution of calculated Porod volume of scatterers during formation of NH2-MOF-235(Al) in DMF:H2O = 9:1.
values (SI) increase during the full period of time, this again points
5. Discussion
at a more complex, fractal nature of the surface, especially during
the initial period.
The experimental results reported above demonstrate that
important information can be extracted from SAXS profiles. It has
to be considered that we deal with a poorly defined system, from
the perspective of scattering entities, and therefore only part of
the information obtained from the SAXS pattern is quantitative.
That said, SAXS certainly adds value to our previous kinetic analysis
[46] and allows chemical interpretation of the results, as discussed
below.
The maximum of the p(r) function (Fig. 11) increases while
showing a shift to larger sizes, and then a shift to smaller sizes,
pointing again at growth and shrinkage, without undergoing any
significant morphologic change. The intensity, i.e. number of scat-
tering entities keeps increasing during this process.
To complete the SAXS analysis, we fitted all kinetic profiles
obtained from integration of the WAXS data using the above men-
tioned Gualtieri equation. The fitting yielded nucleation and growth
rate constants, kn and kg, which are given in Tables 2 and 3. In case
of NH2-MIL-101(Al), from the Arrhenius plot activation energies for
nucleation and growth were found to be 82 4 and 94 6 kJ/mol
respectively, which is in the range of values reported previously for
HKUST-1 and MOF-14 [48].
We propose that NH2-MIL-53(Al) forms through hydrolysis
of AlCl3·6H2O. The formation of ␮2-OH bridged Al polymorphs
through hydrolysis of the hexaaqua aluminium complex in water
Table 2
Crystal growth and nucleation rate constants for NH2-MIL-101(Al) crystallization
extracted from the fitting of experimental profiles recorded at different tempera-
tures. Syntheses were carried out in DMF under low precursors’ concentration (see
Table 1).
a
a
T (K)
kg (×104 s−1
)
kn (×104 s−1
)
393
403
413
2.3
6.3
9.3
5.8
14
20
a
All rate constants are determined with 10% accuracy.
3000 s
2000 s
1000 s
500 s
Table 3
Crystal growth and nucleation rate constants for NH2-MIL-53(Al) and NH2-MIL-
101(Al) crystallization in different solvent mixtures at 403 K.a
300 s
Solvent composition (H2O: DMF)a
kg/104 s−1b
kn/104 s−1b
NH2-MIL-53(Al)
1:0
0.30:0.70
0.10:0.90
0.05:0.95
0:1c
1.2
14.0
7.0
4.1
2.7
45
43
23
14
11
NH2-MIL-101(Al)
0:1c
0
10
20
30
40
50
60
70
80
13.0
15
r
/ nm
a
High concentration of the precursors.
All rate constants are determined with 10% accuracy.
b
Fig. 11. Time evolution of the p(r) function during formation of NH2-MOF-235(Al)
in DMF:H2O = 9:1.
c
Note that two phases are formed simultaneously in these conditions.

126
M.G. Goesten et al. / Catalysis Today 205 (2013) 120–127
Fig. 12. Chemical representations of the MIL-53 (a), MOF-235 (b) and MIL-101 (c) topologies. S denotes solvent and L denotes ligand.
has been thoroughly studied [72–75]. It has been claimed that
equilibrium exists between the monomeric complex and the poly-
mers it forms, favouring the former in non-basic solutions. This
explains the relatively low yield and slow kinetics when pure water
is used as solvent [46]. As expected for this mechanism, in which
6. Conclusions
In this work, we have used synchrotron-based SAXS tech-
nique backed by WAXS, for the in situ study of MOF growth.
It was possible to monitor the temporal development of size,
morphology, and crystallinity of two of the important MOF topolo-
gies, MIL-53 and MIL-101. Based on the information obtained,
description of the formation of these frameworks nears comple-
tion. Although the macroscopic developments on full crystal scale
are now documented for different solvent systems, the chem-
istry behind MOF-235 and its transformation into MIL-53 and/or
MIL-101 should be further studied using complementary in situ
techniques, i.e. EXAFS and NMR.
Al
O Al chains steadily grow with 2-aminoterephthalic acid act-
ing as crosslinker, one observes growth dominated by nucleation,
densification of the porous phase and smoothening of the crystallite
surface in SAXS. A similar nucleation control was earlier reported
for ZIF-8 [45] and also observed during the electrochemical synthe-
sis of NH2-MIL-53(Al) [63].
Whenever DMF is present, the intermediate phase NH2-MOF-
253(Al) appears. This framework consists of ␮3-O bridged, trigonal
SBUs. This cluster is well known for d-block metals, but was only
recently also reported for p-block metals [76–78]. DMF seems to
stabilize this cluster and an immediate result is a change in crystal
growth and kinetics in comparison with the synthesis in water.
It was already shown by Yaghi et al. that DMF acts as terminal
ligand for Fe-based MOF-235 [71]. Moreover, it is well known that
DMF can undergo decomposition at high temperatures, produc-
ing NMe2 that would eventually provide a more basic medium,
increasing in such case the crystallization kinetics, as observed in
Table 3. The rationale behind solvent-induced stabilization of this
phase is at this point unknown, and is currently investigated by our
group. In SAXS, we observe unusually fast growth of scattering par-
ticles of an amorphous phase. But as these then slowly shrink, this
phase changes into crystalline NH2-MOF-235. We propose that the
amorphous phase consists of ␲-stacked SBUs which arrange into
Acknowledgements
Technology Foundation STW, The Netherlands and Hyflux
CEPAration Technologies Europe BV are gratefully acknowledged
for financial support through the partnership programme, STW-
Hyflux CEPAration Inorganic and Hybrid membranes. J.G. gratefully
acknowledges the Dutch National Science foundation (NWO-VENI)
for financial support.
We thank ESRF for provision of the beamtime at BM16 beamline,
and we are grateful to Dr. Francois Fauth for his assistance during
the use of BM16 beamline. Use of the National Synchrotron Light
Source, Brookhaven National Laboratory, was supported by the U.S.
Department of Energy, Office of Science, Office of Basic Energy Sci-
ences, under Contract No. DE-AC02-98CH10886. Dr. Lin Yang is
acknowledged for his help during beamtime at NSLS beamline X9.
Dr. Evgeny Pidko and Dr. Joop ter Horst are acknowledged for
their contributions through scientific discussion.
−
NH2-MIL-253(Al) as AlCl4 anions diffuse into the voids, balance
charge of the cluster and create the crystalline phase. During this
process the fractal character of the crystals and their roughness
increase, as indicated by the change in slope in the SAXS decay
(Fig. 10). NH2-MOF-253(Al) is then only slowly transformed into
NH2-MIL-53(Al) over a long period of time (+48 h) [46]. In any case,
yield improves tremendously after an extended period of synthe-
sis. It appears that the equilibrium between MOF-235 and MIL-53
structure is favourable towards the latter.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.cattod.
2012.08.044.
Pure DMF has to be used to obtain NH2-MIL-101(Al). From
chemical perspective, only removing the terminal aqua ligand
is necessary to convert NH2-MOF-235(Al) into NH2-MIL-101(Al)
(see Fig. 12), which might be the reason behind the con-
stant scatterer volume, morphology and surface in time. This
could be rationalized by a crystal rearrangement, rather than a
formation–dissolution–formation mechanism, as proposed earlier
[46]. Previous work by Bras et al. drew similar conclusions from
analogous analysis for the formation of aluminophosphates crys-
talizing from thick gels, were diffusion would be limited [79]. The
synthesisin anhydrousDMF is unique as it is the only reportedroute
towards NH2-MIL-101(Al). It seems that only in pure DMF, reaction
of the aqua ligand and aluminium tetrachloride anion to form a
hydroxyl ligand and hydrochloric acid is favourable. Enhanced sta-
bility of hydrochloric acid in DMF solvent has been demonstrated
in literature before [80].
References
[1] A. Huang, W. Dou, J. Caro, Journal of the American Chemical Society 132 (2010)
15562.
[2] Y.-S. Li, F.-Y. Liang, H. Bux, A. Feldhoff, W.-S. Yang, J. Caro, Angewandte Chemie
International Edition 49 (2010) 548.
[3] H. Bux, W. Yang, J. Caro, Journal of Membrane Science (2010).
[4] J. Gascon, F. Kapteijn, Angewandte Chemie International Edition (in English) 49
(2010) 1530.
[5] B. Zornoza, A. Martinez-Joaristi, P. Serra-Crespo, C. Tellez, J. Coronas, J. Gascon,
F. Kapteijn, Chemical Communications 47 (2011) 9522.
[6] B. Chen, N.W. Ockwig, A.R. Millward, D.S. Contreras, O.M. Yaghi, Angewandte
Chemie International Edition 44 (2005) 4745.
[7] H. Frost, T. Düren, R.Q. Snurr, Journal of Physical Chemistry B 110 (2006) 9565.
[8] B. Xiao, P.S. Wheatley, X. Zhao, A.J. Fletcher, S. Fox, A.G. Rossi, I.L. Megson, S.
Bordiga, L. Regli, K.M. Thomas, R.E. Morris, Journal of the American Chemical
Society 129 (2007) 1203.
[9] M. Dinca, J.R. Long, Angewandte Chemie International Edition 47 (2008) 6766.
˘

M.G. Goesten et al. / Catalysis Today 205 (2013) 120–127
127
[10] P.K. Allan, P.S. Wheatley, D. Aldous, M.I. Mohideen, C. Tang, J.A. Hriljac, I.L. Meg-
son, K.W. Chapman, G. De Weireld, S. Vaesen, R.E. Morris, Dalton Transactions
(Cambridge, England: 2003) 41 (2012) 4060.
[46] E. Stavitski, M. Goesten, J. Juan-Alcan˜iz, A. Martinez-Joaristi, P. Serra-Crespo,
A.V. Petukhov, J. Gascon, F. Kapteijn, Angewandte Chemie International Edition
50 (2011) 9624.
[11] C.-D. Wu, A. Hu, L. Zhang, W. Lin, Journal of the American Chemical Society 127
(2005) 8940.
[12] F.X.L.I. Llabrés i Xamena, A. Abad, A. Corma, H. García, Journal of Catalysis 250
(2007) 294.
[13] J.Y. Lee, O.K. Farha, J. Roberts, K.A. Scheidt, S.B.T. Nguyen, J.T. Hupp, Chemical
Society Reviews 38 (2009) 1450.
[14] A. Corma, H. García, F.X. Llabrés, I. Xamena, Chemical Reviews 110 (2010) 4606.
[15] M.G. Goesten, J. Juan-Alcan˜iz, E.V. Ramos-Fernandez, K.B.S.S. Gupta, E. Stavitski,
H.V. Bekkum, J. Gascon, F. Kapteijn, Journal of Catalysis 281 (2011) 177.
[16] P. Horcajada, C. Serre, M. Vallet-Regi, M. Sebban, F. Taulelle, G. Ferey, Ange-
wandte Chemie International Edition 45 (2006) 5974.
[17] K.M.L. Taylor, A. Jin, W. Lin, Angewandte Chemie International Edition 47 (2008)
7722.
[47] J. Juan-Alcan˜iz, M. Goesten, A. Martinez-Joaristi, E. Stavitski, A.V. Petukhov, J.
Gascon, F. Kapteijn, Chemical Communications 47 (2011) 8578.
[48] F. Millange, R. El Osta, M.E. Medina, R.I. Walton, CrystEngComm 13 (2011) 103.
[49] S. Surble, F. Millange, C. Serre, G. Ferey, R.I. Walton, Chemical Communications
(2006) 1518.
[50] G.D. Combarieu, S. Hamelet, F. Millange, M. Morcrette, J.-M. Tarascon, G. Ferey,
R.I. Walton, Electrochemistry Communications 11 (2009) 1881.
[51] F. Millange, C. Serre, N. Guillou, G. Ferey, R.I. Walton, Angewandte Chemie
International Edition 47 (2008) 4100.
[52] R. El Osta, M. Frigoli, J. Marrot, M.E. Medina, R.I. Walton, F. Millange, Crystal
Growth and Design 12 (2012) 1531.
[53] F. Millange, M.I. Medina, N. Guillou, G. Ferey, K.M. Golden, R.I. Walton, Ange-
wandte Chemie International Edition 49 (2010) 763.
[18] P. Horcajada, C. Serre, G. Maurin, N.A. Ramsahye, F. Balas, M. Vallet-Regi, M.
Sebban, F. Taulelle, G. Ferey, Journal of the American Chemical Society 130
(2008) 6774.
[19] P. Horcajada, T. Chalati, C. Serre, B. Gillet, C. Sebrie, T. Baati, J.F. Eubank, D.
Heurtaux, P. Clayette, C. Kreuz, J.-S. Chang, Y.K. Hwang, V. Marsaud, P.-N. Bories,
L. Cynober, S. Gil, G. Ferey, P. Couvreur, R. Gref, Nature Materials 9 (2009) 172.
[20] K.M.L. Taylor-Pashow, J. Della Rocca, Z. Xie, S. Tran, W. Lin, Journal of the Amer-
ican Chemical Society 131 (2009) 14261.
[21] K.-L. Wong, G.-L. Law, Y.-Y. Yang, W.-T. Wong, Advanced Materials 18 (2006)
1051.
[22] B.D. Chandler, D.T. Cramb, G.K.H. Shimizu, Journal of the American Chemical
Society 128 (2012) 10403.
[23] M. Xue, G. Zhu, Y. Li, X. Zhao, Z. Jin, E. Kang, S. Qiu, Crystal Growth and Design
8 (2008) 2478.
[24] C.A. Kent, D. Liu, T.J. Meyer, W. Lin, Journal of the American Chemical Society
134 (2012) 3991.
[25] J. Rocha, L.D. Carlos, F.A.A. Paz, D. Ananias, Chemical Society Reviews 40 (2011)
926.
[26] K. Barthelet, J. Marrot, D. Riou, G. Ferey, Angewandte Chemie International
Edition 41 (2002) 281.
[54] S. Couck, T. Rémy, G.V. Baron, J. Gascon, F. Kapteijn, J.F.M. Denayer, Physical
Chemistry Chemical Physics 12 (2010) 9413.
[55] S. Couck, E. Gobechiya, C.E.A. Kirschhock, P. Serra-Crespo, J. Juan-Alcan˜iz, A.
Martinez-Joaristi, E. Stavitski, J. Gascon, F. Kapteijn, G.V. Baron, J.F.M. Denayer,
ChemSusChem 5 (2012) 740.
[56] S. Couck, J.F.M. Denayer, G.V. Baron, T. Rémy, J. Gascon, F. Kapteijn, Journal of
the American Chemical Society 131 (2009) 6326.
[57] E. Stavitski, E. Pidko, S. Couck, T. Remy, E. Hensen, B. Weckhuysen, J.F.M.
Denayer, J. Gascon, F. Kapteijn, Langmuir 27 (2011) 3970.
[58] A. Boutin, S. Couck, F.-X. Coudert, P. Serra-Crespo, J. Gascon, F. Kapteijn, A.H.
Fuchs, J.F.M. Denayer, Microporous and Mesoporous Materials 140 (2011) 108.
[59] P. Serra-Crespo, E. Stavitski, F. Kapteijn, J. Gascon, RSC Advances 2 (2012) 5051.
[60] P. Serra-Crespo, M.A. van der Veen, E. Gobechiya, K. Houthoofd, Y. Filinchuk,
C.E.A. Kirschhock, J.A. Martens, B.F. Sels, D.E. De Vos, F. Kapteijn, J. Gascon,
Journal of the American Chemical Society 134 (2012) 8314.
[61] P. Serra-Crespo, E.V. Ramos-Fernandez, J. Gascon, F. Kapteijn, Chemistry of
Materials 23 (2011) 2565.
[62] E.V. Ramos-Fernandez, C. Pieters, B. van der Linden, J. Juan-Alcan˜iz, P. Serra-
Crespo, M.W.G.M. Verhoeven, H. Niemantsverdriet, J. Gascon, F. Kapteijn,
Journal of Catalysis 289 (2012) 42.
[27] W. Ouellette, M.H. Yu, C.J. O’Connor, D. Hagrman, J. Zubieta, Angewandte
Chemie International Edition 45 (2006) 3497.
[63] A. Martinez-Joaristi, J. Juan-Alcan˜iz, P. Serra-Crespo, F. Kapteijn, J. Gascon, Crys-
tal Growth and Design 12 (2012) 3489.
[28] D.T. Tran, N.A. Chernova, D. Chu, A.G. Oliver, S.R.J. Oliver, Crystal Growth and
Design 10 (2010) 874.
[64] O. Glatter, O. Kratky (Eds.), Small-Angle X-Ray Scattering, Academic Press, Lon-
don, 1982.
[29] M. Kurmoo, Chemical Society Reviews 38 (2009) 1353.
[30] W. Zhuang, H. Sun, H. Xu, Z. Wang, S. Gao, L. Jin, Chemical Communications 46
(2010) 4339.
[31] B.F. Hoskins, R. Robson, Journal of the American Chemical Society 111 (1989)
5962.
[65] A. Gualtieri, Physics and Chemistry of Minerals 28 (2001) 719.
[66] F. Millange, M.I. Medina, N. Guillou, G. Ferey, K.M. Golden, R.I. Walton, Ange-
wandte Chemie International Edition in English 49 (2010) 763.
[67] P.-P.E.A. De Moor, T.P.M. Beelen, R.A. Van Santen, Microporous Materials 9
(1997) 117.
[32] H. Li, M. Eddaoudi, M. O’keeffe, O.M. Yaghi, Nature 402 (1999) 276.
[33] S. Kitagawa, R. Kitaura, S. Noro, Angewandte Chemie International Edition 43
(2004) 2334.
[34] M. Kondo, T. Yoshitomi, K. Seki, H. Matsuzaka, S. Kitagawa, Angewandte Chemie
International Edition 36 (1997) 1725.
[35] M. Eddaoudi, D. Moler, H. Li, B. Chen, T. Reineke, M. O’keeffe, O.M. Yaghi,
Accounts of Chemical Research 34 (2001) 319.
[36] O.M. Yaghi, M. O’keeffe, N. Ockwig, H. Chae, M. Eddaoudi, J. Kim, Nature 423
(2003) 705.
[37] G. Ferey, C. Serre, C. Mellot-Draznieks, F. Millange, S. Surble, J. Dutour, I. Mar-
giolaki, Angewandte Chemie International Edition 43 (2004) 6296.
[38] G. Ferey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surble, I. Mar-
giolaki, Science 309 (2005) 2040.
[39] C. Serre, F. Millange, C. Thouvenot, M. Nogues, G. Marsolier, D. Louer, G. Ferey,
Journal of the American Chemical Society 124 (2002) 13519.
[40] G. Ferey, Journal of Solid State Chemistry 152 (2000) 37.
[41] G. Ferey, Chemical Society Reviews 37 (2008) 191.
[42] C.-S. Tsao, M.-S. Yu, T.-Y. Chung, H.-C. Wu, C.-Y. Wang, K.-S. Chang, H.-L. Chent,
Journal of the American Chemical Society 129 (2007) 15997.
[43] C.-S. Tsao, M.-S. Yu, C.-Y. Wang, P.-Y. Liao, H.-L. Chen, U.-S. Jeng, Y.-R. Tzeng,
T.-Y. Chung, H.-C. Wu, Journal of the American Chemical Society 131 (2009)
1404.
[68] P. de Moor, T. Beelen, R.A. Van Santen, L. Beck, M. Davis, Journal of Physical
Chemistry B 104 (2000) 7600.
[69] T.P.M. Beelen, W.H. Dokter, H.F. van Garderen, A.A. van Santen, E. Pantos, Journal
de Physique IV, Colloque C8 3 (1993) 393.
[70] L.A. Feigin, D.I. Svergun, Structure Analysis by Small-Angle X-Ray and Neutron
Scattering, Plenum Press, New York, 1987.
[71] A.C. Sudik, A.P. Cote, O.M. Yaghi, Inorganic Chemistry 44 (2005) 2998.
[72] A. Violante, P.M. Huang, Clays & Clay Minerals 33 (1985) 181.
[73] A. Violante, P. Huang, Clays & Clay Minerals 41 (1993) 590.
[74] K.P. Prodromou, A.S. Pavlatou-Ve, Clays & Clay Minerals 43 (1995) 111.
[75] M. Digne, P. Sautet, P. Raybaud, H. Toulhoat, E. Artacho, Journal of Physical
Chemistry B 106 (2002) 5155.
[76] C. Volkringer, D. Popov, T. Loiseau, G. Ferey, M. Burghammer, C. Riekel, M.
Haouas, F. Taulelle, Chemistry of Materials 21 (2009) 5695.
[77] H. Hatop, M. Ferbinteanu, H.W. Roesky, F. Cimpoesu, M. Schiefer, H.-G. Schmidt,
M. Noltemeyer, Inorganic Chemistry 41 (2002) 1022.
[78] P. Lemoine, A. Bekaert, J.D. Brion, B. Viossat, Zeitschrift Fur Kristallographie –
New Crystal Structures 221 (2006) 309.
[79] S. Nikitenko, A.M. Beale, A.M.J. van der Eerden, S.D.M. Jacques, O. Leynaud,
M.G. O’Brien, D. Detollenaere, R. Kaptein, B.M. Weckhuysen, W. Bras, Journal of
Synchrotron Radiation 15 (2008) 632–640.
[80] O. Temkin, Homogeneous Catalysis with Metal Complexes: Kinetic Aspects and
Mechanisms, John Wiley & Sons, 2012.
[44] M. Klimakow, P. Klobes, A.F. Thünemann, K. Rademann, F. Emmerling, Chem-
istry of Materials 22 (2010) 5216.
[45] J. Cravillon, C.A. Schröder, R. Nayuk, J. Gummel, K. Huber, M. Wiebcke, Ange-
wandte Chemie International Edition 50 (2011) 8067.