Selective incorporation of functional dicarboxylates into zinc metal-organic frameworks

Article abstract of DOI:10.1039/c1cc10143a

Zinc(ii) nitrate reacts with different ratios of 1,4-benzenedicarboxylic acid (H₂bdc) and 2-halo-1,4-benzenedicarboxylic acid (H ₂bdc-X, X = Br or I) to give [Zn₄O(bdc) _3-x(bdc-X)_x], in which preferen

Full text of DOI:10.1039/c1cc10143a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 3380–3382
www.rsc.org/chemcomm
COMMUNICATION
Selective incorporation of functional dicarboxylates into zinc
metal–organic frameworksw
Andrew D. Burrows,*a Laura C. Fisher,ab Christopher Richardsonac and Sean P. Rigbybd
Received 7th January 2011, Accepted 31st January 2011
DOI: 10.1039/c1cc10143a
Zinc(II) nitrate reacts with different ratios of 1,4-benzenedi-
carboxylic acid (H2bdc) and 2-halo-1,4-benzenedicarboxylic
acid (H2bdc-X, X = Br or I) to give [Zn4O(bdc)3ꢀx(bdc-X)x],
in which preferential incorporation of bdc is observed. The
selective incorporation is related to crystal growth rates, and
the proportion of incorporated bdc-X rises with increasing
reaction time.
relevant for post-synthetic modification,13,14 since doping a
small amount of a particular functional group in a MOF is
essential for applications involving a controlled catalyst
loading.
The first report of a mixed-ligand MC-MOF was by
Kim and co-workers, who reacted zinc(II) with dabco and a
combination of 1,4-benzenedicarboxylic acid (H2bdc) and
2,3,5,6-tetramethyl-1,4-benzenedicarboxylic acid (H2tmbdc)
to form [Zn2(bdc)(tmbdc)(dabco)].15 Subsequently, MC-
MOFs have been prepared in systems based on [Zn4O(bdc)3]
(MOF-5),11,16,17 [Zn4O(4,40-biphenyldicarboxylate)3] (IR-
MOF-9),18 [Al(OH)(bdc)] (MIL-53),19 [Fe3OCl(H2O)2(bdc)3]
(MIL-101)20 and [Zn(1,3-benzenedicarboxylate)(4,40-bpy)].21
Several of these studies have shown that there is preferential
inclusion of one of the ligands, with equimolar mixtures of
dicarboxylic acids leading to one of the dicarboxylates being
incorporated to a greater extent into the structure. For
example, we showed that the reaction between zinc(II) and a
1 : 1 mixture of 2-formyl-4,40-biphenyldicarboxylic acid
(H2bpdc-CHO) and 2-methoxy-4,40-biphenyldicarboxylic acid
(H2bpdc-OMe) gave [Zn4O(bpdc-CHO)0.3(bpdc-OMe)2.7],
with a 1 : 9 ratio of included ligands.18 Yaghi and co-workers
recently published an account of mixed-ligand MOFs based on
[Zn4O(bdc)3] (MOF-5) but containing a range of bdc linkers
substituted at the 2- and/or 5-positions, which they describe as
multivariate MOFs (MTV-MOFs). They showed that up to
eight different dicarboxylates could be introduced into a single
MOF, and also demonstrated that the degree of incorporation
of the different functional groups varied considerably.16
In other systems, the degree of incorporation of the ligands
mirrors the ratios used in the syntheses. For example, Baiker
and co-workers showed that reactions of zinc(II) with
mixtures of H2bdc and 2-amino-1,4-benzenedicarboxylic acid
(H2bdc-NH2) gave [Zn4O(bdc)3ꢀx(bdc-NH2)x] in which x
could be controlled.17 They term the mixed-ligand products
as MIXMOFs.
Over the past ten years, the study of materials in which
bridging ligands connect metal centres or aggregates into
extended coordination networks—commonly referred to as
metal–organic frameworks (MOFs)—has become a major
area of chemistry.1–5 Many of these materials exhibit porosity,
and applications such as hydrogen storage,6 separations,7 and
catalysis8 are attracting considerable attention.
While many MOFs contain just one type of linking
ligand, MOFs containing two or more types of linker are
becoming increasingly prevalent. In many of these MOFs, the
two linking ligands play different structural roles. In
[Zn2(bdc)2(dabco)] (bdc = 1,4-benzenedicarboxylate, dabco =
1,4-diazabicyclo[2.2.2]octane),9 for example, anionic bdc ligands
connect the Zn2 centres into sheets, whereas neutral dabco
ligands act as pillars, connecting the sheets into a three-
dimensional network. There is, however, another class of
mixed-ligand MOF in which the linkers adopt similar structural
roles, and are disordered within the crystal structures giving
rise to solid solutions. For these materials, it may be possible
to vary the ratio of the ligands within the mixed-ligand MOF,
and even control it through variation of the reaction stoichio-
metry. We regard these materials as examples of mixed-
component metal–organic frameworks (MC-MOFs),10 a class
of MOF which includes analogous systems with mixed metals
as well as core–shell MOFs.11,12 MC-MOFs are particularly
a Department of Chemistry, University of Bath, Claverton Down,
Bath BA2 7AY, UK. E-mail: a.d.burrows@bath.ac.uk;
Fax: +44 (0)1225 386231; Tel: +44 (0)1225 386529
b Department of Chemical Engineering, University of Bath, Claverton
Down, Bath BA2 7AY, UK
Of course, these observations are not necessarily in conflict, as
different systems may behave in different ways, and the degree of
incorporation of a particular ligand into a MOF may depend on
the reaction conditions. It is clear, however, that controlled
incorporation of a particular functionality into a MOF is not
yet universal, and more work is required in this field.
c School of Chemistry, University of Wollongong, Wollongong, NSW
2522, Australia
d Department of Chemical and Environmental Engineering, University
of Nottingham, University Park, Nottingham, NG7 2RD, UK
w Electronic supplementary information (ESI) available: Syntheses,
powder X-ray diffraction data, 1H NMR spectra and ESI mass
spectra. See DOI: 10.1039/c1cc10143a
With the different observations based on MOF-5 and
analogues with substituted bdc ligands in mind, we set out
c
3380 Chem. Commun., 2011, 47, 3380–3382
This journal is The Royal Society of Chemistry 2011

View Article Online
to investigate how changing the relative proportions of
H2bdc and either H2bdc-Br (2-bromo-1,4-benzenedicarboxylic
acid) or H2bdc-I (2-iodo-1,4-benzenedicarboxylic acid) would
affect the composition of the resultant zinc MOFs
[Zn4O(bdc)3ꢀx(bdc-X)x] (X = Br, I). [Zn4O(bdc-Br)3] (IR-
MOF-2) has been previously reported,22 but [Zn4O(bdc-I)3]
has not.
The reaction of Zn(NO3)2ꢁ6H2O with H2bdc-I in DEF pro-
duces [Zn4O(bdc-I)3]ꢁnDEF in good yield. Single crystals of
[Zn4O(bdc-I)3]ꢁnDEF did not diffract strongly enough for a
structural determination, but the powder X-ray diffraction pattern
for this compound confirms that it is isostructural with IRMOF-2.
Reactions were then carried out using mixtures of H2bdc
and H2bdc-X (X = Br, I) in varying ratios to give
[Zn4O(bdc)3ꢀx(bdc-X)x]. These reactions were carried out in
DEF at 95 1C for 170 h, giving crystalline materials in all
cases. The powder X-ray diffraction patterns revealed that the
MC-MOFs are isostructural with each other, and with MOF-5.
Portions of each of the products were digested fully in DCl/
D2O and d6-DMSO and the resultant solutions analysed by 1H
NMR spectroscopy. The integrals for the aromatic protons
were used to give the ratio of the different dicarboxylate
ligands in the products. Individual single crystals were also
digested in HCl (aq), and the observation of both monoanions
in the negative mode ESI mass spectra showed that the
products are truly MC-MOFs as opposed to physical mixtures
of homoleptic MOFs.
Fig.
2
The variation of BET surface areas with proportion of
bdc-X incorporated into [Zn4O(bdc)3ꢀx(bdc-X)x]. For , X = I, and
for , X = Br.
that the BET surface areas of the materials decrease with
increasing proportion of bdc-X (Fig. 2), as expected on
steric grounds. In addition, the larger size of iodine than
bromine is reflected in the lower values observed for the
iodo-containing MOFs.
10 reactions were also undertaken using mixtures of H2bdc-I
and H2bdc-Br of different compositions. In these cases, the
proportions of the 2-halo-1,4-benzenedicarboxylates observed
in the NMR spectra are identical, within experimental error,
to the proportions used in the synthesis. This is shown
graphically in Fig. 3.
There are a number of potential factors that could influence
the incorporation of the different dicarboxylates into the
growing MOF structure. The presence of a bulky substituent
might limit incorporation of the functionalised ligand, either
through the potential for unfavourable steric interactions
between anions and the framework or by lowering the diffu-
sivity of the anions. This is not consistent with the observed
results which show greater incorporation of the larger bdc-I
into [Zn4O(bdc)3ꢀx(bdc-X)x] than the smaller bdc-Br. It is also
notable that in Yaghi’s study 2,5-dimethyl-1,4-benzenedi-
carboxylate is included into the [Zn4OL3] structure in preference
to bdc, despite the greater bulk.16
20 reactions involving mixtures of H2bdc and H2bdc-I and
20 reactions involving mixtures of H2bdc and H2bdc-Br were
undertaken. The results are summarised graphically in Fig. 1.
In all cases, the proportion of bdc-I or bdc-Br observed in the
product is less than that present in the reaction mixture,
consistent with the report from Yaghi and co-workers that
bdc is incorporated in preference to bdc-Br.16 Notably, the
incorporation of bdc-I is greater than that of bdc-Br. Nitrogen
adsorption measurements on [Zn4O(bdc)3ꢀx(bdc-X)x] showed
Difference in pKa between the acids is also a potential
factor, though this is likely to give rise to two conflicting
effects. The pKa values for 2-halodicarboxylic acids would be
Fig. 1 The relationship between the percentage of bdc-X incorpo-
rated into [Zn4O(bdc)3ꢀx(bdc-X)x] and the percentage of H2bdc-X in
the reaction mixture. For , X = I, and for , X = Br. The green line
represents what would be expected if the incorporation was not
selective.
Fig. 3 The relationship between the percentage of bdc-I incorporated
into [Zn4O(bdc-Br)3ꢀx(bdc-I)x] and the percentage of H2bdc-I in the
reaction mixture.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 3380–3382 3381

View Article Online
proceeds, the mother liquor will become enriched in bdc-X,
which contributes to the growing proportion of bdc-X
incorporated.
In conclusion, we have demonstrated that the preferential
incorporatation of bdc over bdc-I and bdc-Br occurs in MOFs
of the general formula Zn4OL3, and that this selectivity is
related to the relative rates of crystallisation. An important
implication of these results is that the composition of MC-MOFs
is not necessarily uniform. In the system under study here, at
least, the proportion of linkers that bear a halo group is lower
towards the centre of the crystals than it is on the outside.
This anisotropy will have an impact on the properties of these
MC-MOFs, including post-synthetic modifications.
This work shows that the formation of MC-MOFs through
control of the incorporation of ligands that have the same
structural role provides opportunities to tailor the composi-
tion and function of MOFs. Current work is engaged in
expanding the scope of MC-MOFs and assessing the generality
of the results reported in this work.
Fig. 4 The relationship between the percentage of bdc-X incorporated
into [Zn4O(bdc)3ꢀx(bdc-X)x] and the reaction time. For , X = I, and
for , X = Br.
expected to be lower than those for H2bdc (pKa1 3.51, pKa2
4.82) given the inductive effect of the halo group. A similar
effect is notable for the benzoic acids: compare benzoic acid
(pKa 4.20) with 2-iodobenzoic acid (pKa 2.85) and 3-iodobenzoic
acid (pKa 3.86). This means that the concentration of the
2-halo-1,4-benzenedicarboxylates (and the [Hbdc-X]ꢀ mono-
anions) would be higher than that of bdc (and [Hbdc]ꢀ). Given
the greater inclusion of bdc over bdc-I and bdc-Br, the relative
concentrations of the ions cannot be the major factor in their
incorporation into the MOF crystals. However, the greater
basicity of the anions derived from H2bdc would be expected
to make these anions better ligands, so it is conceivable that
the greater ligating ability of the bdc ions wins out over the
lower concentration.
The EPSRC are thanked for financial support.
Notes and references
1 J. L. C. Rowsell and O. M. Yaghi, Microporous Mesoporous
Mater., 2004, 73, 3.
2 G. Ferey, Chem. Soc. Rev., 2008, 37, 191.
´
3 R. Robson, Dalton Trans., 2008, 5113.
4 S. Horike, S. Shimomura and S. Kitagawa, Nat. Chem., 2009, 1,
695.
5 A. U. Czaja, N. Trukhan and U. Muller, Chem. Soc. Rev., 2009,
¨
38, 1284.
6 L. J. Murray, M. Dinca and J. R. Long, Chem. Soc. Rev., 2009, 38,
1294.
7 J.-R. Li, R. J. Kuppler and H.-C. Zhou, Chem. Soc. Rev., 2009, 38,
1477.
Observations on the crystal growth rates of [Zn4O(bdc)3]
(MOF-5), [Zn4O(bdc-Br)3] (IRMOF-2) and [Zn4O(bdc-I)3]
suggested another factor to be relevant too. When reactions
to form [Zn4O(bdc)3], [Zn4O(bdc-Br)3] and [Zn4O(bdc-I)3]
were set up side by side using identical conditions, crystals
were always observed first for [Zn4O(bdc)3], with those of the
the halo-containing products taking longer to form. This
suggested that the faster crystal growth rate for [Zn4O(bdc)3]
might be a factor in the observed greater proportion of bdc in
the mixed bdc/bdc-Br and bdc/bdc-I products.
8 L. Ma, C. Abney and W. Lin, Chem. Soc. Rev., 2009, 38, 1248.
9 D. N. Dybtsev, H. Chun and K. Kim, Angew. Chem., Int. Ed.,
2004, 43, 5033.
10 A. D. Burrows, CrystEngComm, 2011, DOI: 10.1039/
C0CE00568A.
11 K. Koh, A. G. Wong-Foy and A. J. Matzger, Chem. Commun.,
2009, 6162.
12 S. Furukawa, K. Hirai, K. Nakagawa, Y. Takashima, R. Matsuda,
T. Tsuruoka, M. Kondo, R. Haruki, D. Tanaka, H. Sakamoto,
S. Shimomura, O. Sakata and S. Kitagawa, Angew. Chem., Int.
Ed., 2009, 48, 1766.
13 Z. Wang and S. M. Cohen, Chem. Soc. Rev., 2009, 38, 1315.
14 S. M. Cohen, Chem. Sci., 2010, 1, 32.
In order to assess this, the reactions between zinc(II) and
50 : 50 mixtures of H2bdc and H2bdc-X (X = Br, I) were
carried out for shorter periods of time, with the products
analysed after 48 h, 72 h, 96 h, 120 h and 168 h. As expected,
shorter reaction times led to lower yields, and indeed no
product was observed for the reaction of zinc(II) with
H2bdc/H2bdc-Br after 48 h. In all other cases, however, there
was enough material to digest and analyse by 1H NMR
spectroscopy. These results are summarised graphically in
Fig. 4. With shorter reaction times there is clearly a lower
incorporation of bdc-Br or bdc-I into the MOF material, with
the degree of incorporation going up with increasing reaction
time. This is consistent with greater preferential inclusion of
bdc at the early stages of crystal growth. As the reaction
15 H. Chun, D. N. Dybtsev, H. Kim and K. Kim, Chem.–Eur. J.,
2005, 11, 3521.
16 H. Deng, C. J. Doonan, H. Furukawa, R. B. Ferreira, J. Towne,
C. B. Knobler, B. Wang and O. M. Yaghi, Science, 2010, 327, 846.
17 W. Kleist, F. Jutz, M. Maciejewski and A. Baiker, Eur. J. Inorg.
Chem., 2009, 3552.
18 A. D. Burrows, C. G. Frost, M. F. Mahon and C. Richardson,
Angew. Chem., Int. Ed., 2008, 47, 8482.
19 S. Marx, W. Kleist, J. Huang, M. Maciejewski and A. Baiker,
Dalton Trans., 2010, 39, 3795.
20 K. M. L. Taylor-Pashow, J. Della Rocca, Z. Xie, S. Tran and
W. Lin, J. Am. Chem. Soc., 2009, 131, 14261.
21 T. Fukushima, S. Horike, Y. Inubushi, K. Nakagawa, Y. Kubota,
M. Takata and S. Kitagawa, Angew. Chem., Int. Ed., 2010, 49,
4820.
22 M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O’Keeffe
and O. M. Yaghi, Science, 2002, 295, 469.
c
3382 Chem. Commun., 2011, 47, 3380–3382
This journal is The Royal Society of Chemistry 2011