Near-UV photo-induced modification in isoreticular metal-organic frameworks

Article abstract of DOI:10.1039/c2jm15183a

A series of isoreticular metal-organic frameworks (IRMOFs) have been prepared using two different ligands protected with photolabile groups: 2-((2-nitrobenzyl)oxy)terephthalic acid (L1) and 2-((4,5-dimethoxy-2- nitrobenzyl)oxy)terephthalic acid (L2). Irradiation at either 365 or 400 nm results in postsynthetic deprotection (PSD), removing the nitrobenzyl protecting groups from these ligands and generating phenolic groups in the pores of the MOF (55-83% yield). The photochemical behaviour of the ligands in the IRMOFs was not wholly predicted by their reactivity in solution (i.e. free ligand). A mixed ligand approach was used, by combining L1 or L2 with NH₂-BDC to produce mixed ligand IRMOFs with 30-40% incorporation of the NH ₂-BDC. These mixed-ligand IRMOFs were then subjected to both postsynthetic modification (PSM) and PSD. Two routes to achieve both PSM and PSD were explored: PSM followed by PSD (route 1) and PSD followed by PSM (route 2). When using 365 nm light for the PSD reaction, route 1 was superior due to absorption of NH₂-BDC at 365 nm. Irradiation at 400 nm gave fewer differences between route 1 and route 2, but the PSD reaction was less efficient (30-40%) for all systems. By combining PSD and PSM, MOFs with highly functionalized pores can be obtained through a combination of pre- and postsynthetic methods.

Full text of DOI:10.1039/c2jm15183a

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 10188
www.rsc.org/materials
PAPER
Near-UV photo-induced modification in isoreticular metal–organic
frameworks†
*
Corinne A. Allen and Seth M. Cohen
Received 13th October 2011, Accepted 25th November 2011
DOI: 10.1039/c2jm15183a
A series of isoreticular metal–organic frameworks (IRMOFs) have been prepared using two different
ligands protected with photolabile groups: 2-((2-nitrobenzyl)oxy)terephthalic acid (L1) and 2-((4,5-
dimethoxy-2-nitrobenzyl)oxy)terephthalic acid (L2). Irradiation at either 365 or 400 nm results in
postsynthetic deprotection (PSD), removing the nitrobenzyl protecting groups from these ligands and
generating phenolic groups in the pores of the MOF (55–83% yield). The photochemical behaviour of
the ligands in the IRMOFs was not wholly predicted by their reactivity in solution (i.e. free ligand). A
mixed ligand approach was used, by combining L1 or L2 with NH2-BDC to produce mixed ligand
IRMOFs with 30–40% incorporation of the NH2-BDC. These mixed-ligand IRMOFs were then
subjected to both postsynthetic modification (PSM) and PSD. Two routes to achieve both PSM and
PSD were explored: PSM followed by PSD (route 1) and PSD followed by PSM (route 2). When using
365 nm light for the PSD reaction, route 1 was superior due to absorption of NH2-BDC at 365 nm.
Irradiation at 400 nm gave fewer differences between route 1 and route 2, but the PSD reaction was less
efficient (30–40%) for all systems. By combining PSD and PSM, MOFs with highly functionalized
pores can be obtained through a combination of pre- and postsynthetic methods.
Telfer,10,11 and Cohen12 introduced a related strategy that has
Introduction
been coined ‘postsynthetic deprotection’ (PSD). PSD involves
synthesizing the framework with a protected functional group on
the linker, then using suitable stimuli (e.g. heat, light, chemical,
etc.) to remove the protecting group, revealing the desired
functionality.11
Multifunctional or ‘multivariate’ materials13 have been real-
ized using mixed-ligand strategies and various combinations of
pre- and postsynthetic approaches.13–18 Using a combination of
these methods can lead to highly functionalized materials
without complicating the synthetic conditions required to obtain
the desired MOF. Within this theme of multiple modification
methods, here we report the use of a light-activated PSD reaction
in combination with a PSM reaction to generate functionalized
isoreticular metal–organic frameworks (IRMOFs).
Metal–organic frameworks (MOFs) are an intriguing class of
hybrid materials that have been extensively studied in recent
years.1–3 Also known as porous coordination polymers (PCPs)
they are comprised of metal ions linked by organic molecules and
are of interest due to their structured, porous, and highly tunable
nature. Since the introduction of MOFs, many groups have
sought to make functionalized, and therefore specially tailored,
materials. With thousands of organic transformations available,
the challenge remains of how to apply these reactions to MOFs
while maintaining their structural integrity. It has been estab-
lished that functionality can be directly incorporated into a MOF
by simply using a functionalized organic linker molecule.4
However, this method is limited to functional groups that will
not interfere with MOF formation. Based on the idea that
a chemical ‘handle’ or ‘tag’ can be introduced in a ‘presynthetic’
fashion, the addition of functional groups after framework
formation via these handles has been termed ‘postsynthetic
modification’ (PSM).5–7 The PSM method has produced mate-
rials that would otherwise remain difficult or impossible to
prepare by presynthetic approaches.8 Recently, Hupp,9
Utilizing light as a mild and efficient stimulus for chemical
transformations has been studied on a multitude of platforms,
including MOFs.19–23 Previously, we showed 2-nitrobenzyl pro-
tecting groups could be used to reveal pendant phenol and
catechol groups in a UMCM-1 (University of Michigan Crys-
talline Materials) lattice.12 2-Nitrobenzyl derivatives are photo-
labile protecting groups that are commonly used to protect
alcohols, esters, amides, and amines and are easily released upon
exposure to near-UV light. Despite these and other studies of
photochemistry in MOFs, there are no reports that compare the
behaviour of photolabile groups in a framework versus in solu-
tion. There is precedence for the MOF to modify solution state
Department of Chemistry and Biochemistry, University of California, San
Diego, La Jolla, California, 92093-0358, United States. E-mail: scohen@
ucsd.edu; Fax: +1-858-822-5598; Tel: +1-858-822-5596
† Electronic Supplementary Information (ESI) available: Fig. S1–S9,
Tables S1–S4. See DOI: 10.1039/c12jm15183a
10188 | J. Mater. Chem., 2012, 22, 10188–10194
This journal is ª The Royal Society of Chemistry 2012

View Online
behaviour as seen by Champness and George when monitoring
charge transfer bands.24 Here, two different photolabile pro-
tecting groups were incorporated into the IRMOF lattice and
their photocleavage at two different wavelengths was investi-
gated (Scheme 1). Additionally, these ligands were incorporated
into mixed-ligand IRMOFs that were subjected to tandem PSM/
PSD for functionalization. We find that successful execution of
the light-triggered PSD reaction is dependent on order in which
the PSD and PSM reactions are performed.
were collected on
spectrophotometer.
a
Perkin-Elmer Lambda 25 UV-Vis
Ligands
L1 was synthesized as previously reported.12 L2 was synthesized
following a related procedure (ESI).†
IRMOFs
2-((2-Nitrobenzyl)oxy)terephthalic acid (L1, 317 mg, 1 mmol) or
2-((4,5-dimethoxy-2-nitrobenzyl)oxy)terephthalic acid (L2,
377 mg, 1 mmol) and Zn(NO3)2$6H2O (832 mg, 2.8 mmol) were
dissolved in DMF for L1 or DEF for L2 (50 mL). This was
divided into 10 mL portions and transferred to five scintillation
vials (20 mL capacity). The vials were capped, placed into a sand
Experimental
General
N,N-Diethylformamide (DEF) was purchased from TCI Amer-
ica. Zinc nitrate hexahydrate (98%) and N,N-dimethylformamide
(DMF) was purchased from Sigma-Aldrich Inc. 4,5-Dimethoxy-
2-nitrobenzyl bromide (97%) was purchased from Acros
Organics. All chemicals were used without further purification.
1H NMR spectra were recorded on a Varian FT-NMR spec-
trometer (400 MHz). ESI-MS was performed using a Thermo-
Finnigan LCQ-DECA mass spectrometer and analyzed using the
Xcalibur software suite. Thermogravimetric analysis was done
using a TA instrument Q600 SDT instrument. Powder X-ray
diffraction spectra were collected on a Bruker D8 Advance
bath, and heated in a ramping oven to 100 C at 2.5 C minꢁ1
,
ꢂ
ꢂ
incubated at 100 ꢂC for 48 h, and cooled to room temperature at
2.5 ꢂC minꢁ1. The mother liquor was decanted and the block
crystals were rinsed with DMF (3 ꢃ 10 mL), followed by CHCl3
(10 mL), and left to soak overnight. After 24 h, the CHCl3 was
decanted and fresh CHCl3 was added. This was repeated two
more times for a total of 3 days of soaking in CHCl3. Yield:
IRMOF-1-L1 (Zn4O(C15H9NO7)3, FW ¼ 1223.2 g molꢁ1) 45 mg
(0.036 mmol, 11%); IRMOF-1-L2 (Zn4O(C17H13NO9)3, FW ¼
1403.4 g molꢁ1) 32 mg (0.023 mmol, 7%).
ꢀ
diffractometer at 40 kV, 40 mA for Cu Ka (l ¼ 1.5417 A). Gas
sorption measurements were performed on a Micromeritics
ASAP 2020 Adsorption Analyzer. Surface areas were calculated
using the Brunauer-Emmett-Teller (BET) method. All reported
BET values have a standard deviation of approximately ꢀ150 m2
gꢁ1 based on at least three independent experiments. Single
crystal X-ray diffraction data was collected at 200 K on a Bruker
IRMOF-1-(L1)(NH2)
IRMOF-1-(L1)(NH2) was synthesized by employing the same
reaction conditions as stated above and using Zn(NO3)2$6H2O
(832 mg, 2.8 mmol), L1 (602 mg, 1.9 mmol), and 2-amino-
terephthalic acid (NH2-BDC, 90 mg, 0.5 mmol). Yield: (Zn4O
(C15H9NO7)2.1(C8H5NO4)0.9, FW ¼ 1100.7 g molꢁ1) 58 mg
(0.052 mmol, 9%).
ꢀ
Apex diffractometer using Mo ka radiation (l ¼ 0.71073 A) with
the Apex 2010 software. Solid State UV-Vis absorbance spectra
were taken on a StellarNetEPP 2000C spectrometer with
a diffuse reflectance probe. Solution UV-Vis absorbance spectra
IRMOF-1-(L2)(NH2)
IRMOF-1-(L2)(NH2) was synthesized by employing the same
reaction conditions as stated above and using Zn(NO3)2$6H2O
(832 mg, 2.8 mmol), 2-((4,5-dimethoxy-2-nitrobenzyl)oxy)ter-
ephthalic acid (377 mg, 1.0 mmol), 2-aminoterephthalic acid
(NH2-BDC,
90
mg,
0.5
mmol).
Yield:
(Zn4O
(C17H13NO9)1.8(C8H5NO4)1.2, FW ¼ 1168.0 g molꢁ1) 35 mg
(0.029 mmol, 7%).
Postsynthetic deprotection
Each vial of crystals was rinsed with EtOAc (3 ꢃ 10 mL), leaving
the crystals stored in 10 mL of EtOAc. The vial was sealed with
a screw cap and placed into a Rayonet RPR-200 photoreactor
with either 365 nm or 400 nm lamps. Samples were irradiated for
a total of 48 h. During irradiation, the crystals were washed with
EtOAc (2 ꢃ 10 mL) three times a day. Upon completion, the
crystals were rinsed with EtOAc (2 ꢃ 10 mL) and exchanged into
CHCl3 (2 ꢃ 10 mL) until further use.
Postsynthetic modification
IRMOF-1-(L)(AM1), L ¼ L1 or L2. The storage solution from
Scheme 1 Ligand naming system and incorporation of the protected
HO-BDC ligands into an IRMOF framework followed by PSD.
a vial of IRMOF-1-(L1)(NH2) or IRMOF-1-(L2)(NH2) was
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 10188–10194 | 10189

View Online
decanted and CHCl3 (5 mL) was added. Acetic anhydride (30 mL,
0.32 mmol) was added, the vial was capped, and the mixture
placed into an oven at 55 ꢂC for 24 h. This process was repeated
one more time for IRMOF-1-(L2)(NH2) for a total reaction time
of 48 h. The crystals were rinsed with CHCl3 (3 ꢃ 10 mL) and left
to soak overnight. The crystals were soaked in 10 mL of fresh
CHCl3 for the 2 days, replacing the mother liquor once every
day. ꢄ99% yield.
Thermogravimetric analysis (TGA) of IRMOF-1-L1 displays
a sharp weight loss at 260 ꢂC (ꢄ40%) indicative of thermal
liberation of the nitrobenzyl group (ESI†). IRMOF-1-L2 does
not have the same drastic drop, but has a gradual thermal trace
that shows MOF degradation ꢄ400 ꢂC. Dissolution of IRMOF-
1-L1 and IRMOF-1-L2 crystals in acid allowed for analysis of
1
their chemical composition by H NMR spectroscopy and elec-
trospray ionization mass spectrometry (ESI-MS). Both methods
showed that ligands L1 and L2 were successfully incorporated
into the IRMOFs without degradation or loss of the protecting
group (Fig. 2). BET surface area measurements with dinitrogen
at 77 K gave values of 1407 m2 gꢁ1 and 1164 m2 gꢁ1 for IRMOF-
1-L1 and IRMOF-1-L2, respectively (Table 1). Although these
surface area values are lower than that of many other IRMOFs,4
they are consistent with the introduction of the large nitrobenzyl
groups into the framework pores.
Results and discussion
Synthesis of IRMOFs
We have previously reported the incorporation of the relatively
bulky L1 ligand into the UMCM-1 framework, which is well
accommodated by the large pores in this material (14 and
12,25
ꢀ
32 A).
However, we were somewhat surprised to find that
both L1 and a dimethoxy analogue L2 could be fully incorpo-
rated into an IRMOF lattice with relative ease (11 A).4,26 Bulky
ꢀ
ligands, such as 2,5-bis(benzyloxy)terephthalic acid have been
incorporated into an IRMOF, but the degree of incorporation
(with BDC as a co-ligands) was limited to ꢄ30%, indicative of
a steric restriction for bulky groups in the IRMOF lattice.13
Combining L1 with Zn(NO3)2$6H2O in N,N-dimethylforma-
mide (DMF) or L2 witꢂh Zn(NO3)2$6H2O in N,N-diethy-
lformamide (DEF) at 100 C for 48 h produced cubic colorless
or yellow crystals of IRMOF-1-L1 and IRMOF-1-L2, respec-
tively. If DMF was used for the synthesis for IRMOF-1-L2, no
crystals were obtained. The IRMOF lattice structure of both
materials was directly confirmed by both powder X-ray
diffraction (PXRD) and single crystal X-ray diffraction (XRD)
techniques (Fig. 1, ESI†). XRD structure determination of
IRMOF-1-L1 and IRMOF-1-L2 unambiguously resolved the
expected IRMOF lattice (ESI†). In addition, diffuse electron
density was found in the pores of both materials that was
suggestive of the nitrobenzyl protecting groups, but the
substituents could not be clearly assigned due to severe disorder.
The composition of the ligands in these frameworks was verified
by other methods.
Fig. 2 1H NMR of digested samples of IRMOF before and after PSD at
both 365 nm and 400 nm wavelengths. Starting materials are marked with
black circles and the product is marked by red squares. A 1H NMR for an
authentic sample of HO-BDC is provided as a reference.
Fig. 1 Powder X-ray diffraction of frameworks before and after PSD.
10190 | J. Mater. Chem., 2012, 22, 10188–10194
This journal is ª The Royal Society of Chemistry 2012

View Online
Table 1 BET surface areas for the IRMOFs. Percent deprotection (%
1
framework (based on NMR and ESI-MS analysis of the final
MOF products). All samples showed little to no change in the
PXRD patterns (Fig. 1) after irradiation, confirming the pristine
nature of the crystals. Preservation of cystallinity was also readily
apparent by visual inspection, where only a color change was
observed, but the crystals appeared otherwise undisturbed as
confirmed through single crystal XRD (ESI†).
PSD) based on H NMR spectroscopic analysis of samples digested in
d6-DMSO/DCl/D2O solution after irradiation at either 365 or 400 nm.
The standard deviation for all reported BET values is no more than
ꢀ150 m2 gꢁ1
MOF
BET (m2 gꢁ1
)
% PSD 365 nm % PSD 400 nm
IRMOF-1-L1
IRMOF-1-L2
IRMOF-1-(L1)(NH2) 1584
IRMOF-1-(L2)(NH2) 1615
IRMOF-1-(L1)(AM1) 1427
IRMOF-1-(L2)(AM1) 1344
1407
1164
80 (2344 m2 gꢁ1) 83
Changing the electronics of the nitrobenzyl group is a common
tactic used to shift the absorption wavelength for photocleavage.
Based on previous reports,19 it was expected that the nitrobenzyl
group (L1) would cleave with a lmax at 365 nm, while the
dimethoxy derivative (L2) would cleave more effectively at 400
nm; however, in the MOFs the expected trend was not observed.
Also, quantitative deprotection, as was seen in the UMCM
system,12 was not achieved in these IRMOF materials, despite
screening a variety of reaction conditions. The differences
between the expected photochemical behaviour and our obser-
vations here may be due to a number of factors. As will be dis-
cussed below, we believe a combination of light absorption and
scattering by the IRMOF explains, in part, the attenuation of the
photochemical reactions. Both absorption and scattering
diminish the amount of light that penetrates through the
framework and thus impedes PSD.
53
58
42
40
63
77
<10
33
53
70
Postsynthetic deprotection of IRMOFs
IRMOF-1-L1 and IRMOF-1-L2 were irradiated in a photo-
chemical reactor at one of two wavelengths, 365 or 400 nm, for 48
h to effect PSD of the materials and obtain IRMOFs with
phenolic groups in the pores. The percent conversion of the PSD
reaction was determined using 1H NMR after digesting the
materials in acid (Fig. 2). IRMOF-1-L1 gave 80–83% conversion
to IRMOF-1-OH whether irradiated at either 365 nm or 400 nm
(Table 1). A concomitant increase in BET surface area, as
observed in our previous studies,12 was found for the product
IRMOF-1-OH, showing an average BET value of ꢄ2344 m2 gꢁ1
(Fig. 3, Table 1). IRMOF-1-L2 also did not show a substantial
difference in PSD efficiency as a function of wavelength, giving
between 53–58% conversion to IRMOF-1-OH whether irradi-
ated at 365 or 400 nm (Fig. 2). A steric effect may be contributing
to this slow deprotection of IRMOF-1-L2, as the pores of this
IRMOF are more restricted than IRMOF-1-L1 due to the
additional methoxy groups on the protecting group. The dime-
thoxynitrosobenzaldehyde byproduct of the photochemical
reaction may be more strongly retained in the pores and further
interfere with the PSD reaction, due to the large extinction
coefficient of dimethoxynitrosobenzaldehyde.27 However, we
find little evidence that these aldehyde byproducts remain in the
Combining PSD and PSM on IRMOFs
Mixed ligand approaches have been utilized to prepare highly
functionalized MOF materials.14,18 Functional groups that can
be introduced presynthetically can serve as chemical handles for
PSM. In a recent study from our group, we investigated tandem
PSM on a mixed ligand system. Two different BDC ligands,
NH2-BDC and Br-BDC, were incorporated into a UiO-66
framework and modified orthogonally via PSM.15,28 In this case,
the order in which the orthogonal PSM reactions were performed
(i.e. PSM of NH2-BDC before Br-BDC versus PSM of Br-BDC
then NH2-BDC) had no effect on the overall yields. Hence the
PSM reactions could be performed in any order, allowing for the
preparation of several bifunctional MOFs.15
In order to investigate the possibility of performing tandem
PSD-PSM, NH2-BDC was chosen as the co-ligand to make
mixed IRMOFs. Upon mixing L1 and NH2-BDC in a 3.8 : 1
ꢂ
ratio with Zn(NO3)2 in DEF at 100 C for 48 h, amber-colored
crystals of IRMOF-1-(L1)(NH2) were obtained. The 3.8 : 1
ligand ratio gave the highest quality crystals based on visual
inspection and PXRD analysis (data not shown) from
a screening of a large number of ligand ratios and reaction
1
conditions. H NMR spectroscopy of the digested crystals indi-
cate that NH2-BDC comprised ꢄ30% of the ligands of this
IRMOF. PXRD and single crystal XRD confirmed that
IRMOF-1-(L1)(NH2) possessed the correct topology, although
the BDC substituents could not be assigned in the difference
map. The BET surface area of IRMOF-1-(L1)(NH2) was ꢄ1584
m2 gꢁ1, higher than IRMOF-1-L1, which is consistent with the
smaller NH2-BDC ligand being incorporated (Table 1). Using
similar conditions for L2 and NH2-BDC in a 2 : 1 ratio, amber
blocks were obtained with ꢄ40% NH2-BDC incorporation. The
more efficient incorporation of NH2-BDC into IRMOF-1-(L1)
(NH2) and IRMOF-1-(L2)(NH2) is attributed the greater solu-
bility of NH2-BDC when compared to L1 and L2. It is also
Fig. 3 N2 adsorption and desorption isotherms for IRMOF-1-L1 and
IRMOF-1-OH generated by PSD using 365 nm light.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 10188–10194 | 10191

View Online
possible that steric factors, L1 and L2 being much more bulky,
play a role in the differences observed with respect to ligand
incorporation and the overall low yields of IRMOF formation.
IRMOF-1-(L2)(NH2) gave a BET value ꢄ1615 m2 gꢁ1, under-
standably due to the larger L2 ligand, and a higher percentage of
NH2-BDC (Table 1). Analogous to the orthogonal PSM
mentioned previously,15 we explored two routes for tandem PSD-
PSM: PSM followed by PSD (route 1), and PSD followed by
PSM (route 2) (Scheme 2). These two routes were evaluated
based on the best combination of percent PSM, percent PSD,
and crystallinity of the final materials. It is worth noting that, like
our previous study on tandem PSM,15 this methodology poten-
tially leads to two intermediate and a final MOF from each
starting framework.
Using route 1, where PSM is performed before PSD, we found
both IRMOF-1-(L1)(NH2) and IRMOF-1-(L2)(NH2) were
quantitatively transformed to the acetamide IRMOF-1-(L1)
(AM1) and IRMOF-1-(L2)(AM1), respectively. After PSM,
PSD on IRMOF-1-(L1)(AM1) gave ꢄ53–63% PSD at 365 or 400
nm (Table 1, Fig. 4). This is markedly lower than the photo-
cleavage of IRMOF-1-L1 at either wavelength. Crushing the
crystals gave only modest increases in the PSD yields of mixed
IRMOFs (10–15%). However, IRMOF-1-(L2)(AM1) gave ꢄ70–
77% PSD at 365 or 400 nm, which is an improved yield when
compared to IRMOF-1-L2. Unfortunately, we were unable to
identify reaction conditions where PSD in these IRMOFs would
go to completion. Nonetheless, for route 1 it was found that L2
was a better group for PSD at both wavelengths.
Route 2, where PSD was performed prior to PSM, gave less
impressive results than route 1. Photolysis of IRMOF-1-(L1)
(NH2), at 365 nm shows minimal (<10%) PSD (Table 1, Fig. 4).
Fig. 4 1H NMR of digested IRMOF-1-(L1)(AM1) (top, route 1) and
IRMOF-1-(L1)(NH2) (bottom, route 2) before and after PSD at 365 nm
or 400 nm. Starting materials are marked with black circles and the
product is marked by red squares. A 1H NMR for an authentic sample of
HO-BDC is provided as a reference.
The reason for this low yield is discussed in detail below. At 400
nm, 42% PSD was observed, which is much lower than the 83%
PSD observed for IRMOF-1-L1. Interestingly, IRMOF-1-(L2)
(NH2) shows 33% and 40% deprotection at 365 nm and 400 nm,
respectively (ESI†). This correlates better with the 53% and 58%
PSD obtained with IRMOF-1-L2 at the same wavelengths.
However, due to the generally lower than desired PSD results,
route 2 was not further explored. Overall, it appears route 1 gives
better PSD conversions than route 2. Unlike our previous study
on tandem PSM, the order of PSD versus PSM was important for
obtaining the desired materials.15
In order to understand the poor PSD at 365 nm for IRMOF-1-
(L1)(NH2), the absorption spectra of the homogeneous, free
ligand systems were examined. UV-Vis spectroscopy studies on
the methyl diester forms of the ligands provided some insight into
Scheme 2 Tandem PSD-PSM routes for functionalization of mixed
ligand IRMOFs. Yields are based on 1H NMR spectroscopic analysis of
samples digested in d6-DMSO/DCl/D2O solution.
10192 | J. Mater. Chem., 2012, 22, 10188–10194
This journal is ª The Royal Society of Chemistry 2012

View Online
framework. IRMOF-1-L1 was found to photodeprotect to
IRMOF-1-OH more effectively than IRMOF-1-L2 at both
wavelengths.
Recently, multi-functionalized materials have been heavily
studied because of their distinct and interesting characteristics in
comparison to monofunctional MOFs. To prepare multifunc-
tional MOFs, L1 and L2 were combined with NH2-BDC to
produce IRMOF-1-(L1)(NH2) and IRMOF-1-(L2)(NH2),
respectively. It was found that performing PSM prior to PSD
was preferred, due to the strong absorption of the NH2-BDC
ligand, which interfered with the PSD reaction. Unlike the ester
form of NH2-BDC, which photodegrades in solution, the NH2-
BDC ligand was persistent in the IRMOF, which only served to
further interfere with PSD reaction. These findings indicated that
the reaction order in this PSD-PSM combination was an
important consideration to achieve good conversions to the
desired materials.
In summary, we have studied the heterogeneous PSD of
nitrobenzyl protected MOFs in comparison to the homogeneous
systems. The utility of multiple postsynthetic approaches (PSD
and PSM) on a single, bifunctional MOFs has been demon-
strated. The ability to combine pre- and various post-synthetic
methods is a key step in obtained highly tailorable materials for
technological applications. Related studies, including other
postsynthetic photochemical processes are being investigated
and will be reported in due course.
Fig. 5 UV-Visible absorption spectra of the methyl diester form of
several ligands (50 mM) in EtOAc.
the failure of these PSD reactions (Fig. 5). Photocleavage of 2-
nitrobenzyl groups has been well studied with the near-UV range
(300–400 nm) providing the optimal wavelengths for promotion
of this reaction.19 The lmax for the methyl diesters forms of L1,
NH2-BDC, AM1-BDC, and HO-BDC were 310, 368, 330, and
325 nm, respectively (Fig. 5). This strongly indicates that the
reduced PSD of L1 in IRMOF-1-(L1)(NH2) can be attributed to
the strong absorption by the NH2-BDC ligands compared to L1
at 365 nm. Based on this hypothesis, the solution spectra in Fig. 5
suggest that the addition of the methoxy groups in L2 should
alter the PSD behaviour in IRMOF-1-(L2)(NH2) as observed
(vide supra). Previous reports indicate that the addition of
methoxy substituents allows photocleavage at wavelengths up to
420 nm.23 The broad and very intense absorption of L2 from
300–400 nm drastically increase the efficiency of L2 to absorb at
365 nm thus rendering the observed PSD in the presence of NH2-
BDC much more efficient.
Acknowledgements
We thank Dr Min Kim for helpful discussions, Dr Y. Su (U.C.S.
D.) for performing mass spectrometry experiments, and Dr C.
Moore, Prof. A. Rheingold (U.C.S.D.), and Phuong Dau for
assistance with single-crystal X-ray structures. This research was
supported by a grant from the National Science Foundation
(CHE-0952370). This material is based upon work supported by
the National Science Foundation Graduate Research Fellowship
(C.A.A.) under Grant No. DGE1144086.
The solution UV-Vis spectroscopy also assisted in our
understanding of the more effective PSD achieved when using
route 1. Acetylation of the amine group to give IRMOF-1-(L1)
(AM1) blue-shifts the lmax to 330 nm, thus reducing the inter-
fering absorption at 365 nm and enhancing PSD. Additionally,
IRMOF-1-(L2)(AM1) deprotects most out of all the mixed
MOFs studied presumably due to a combination of the wide
absorbing L2 and the now unobtrusive AM1. This conclusion is
loosely supported by the solid-state UV-Vis spectra of the cor-
responding IRMOFs (ESI†).
Notes and references
1 G. Ferey, Chem. Soc. Rev., 2008, 37, 191–214.
2 J. L. C. Rowsell and O. M. Yaghi, Microporous Mesoporous Mater.,
2004, 73, 3–14.
3 H. Li, M. Eddaoudi, M. O’Keeffe and O. M. Yaghi, Nature, 1999,
402, 276–279.
4 M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O’Keeffe
and O. M. Yaghi, Science, 2002, 295, 469–472.
5 Z. Wang and S. M. Cohen, Chem. Soc. Rev., 2009, 38, 1315–1329.
6 S. M. Cohen, Chem. Sci., 2010, 1, 32–36.
7 K. K. Tanabe and S. M. Cohen, Chem. Soc. Rev., 2011, 40, 498–519.
8 S. Bernt, V. Guillerm, C. Serre and N. Stock, Chem. Commun., 2011,
47, 2838–2840.
9 T. Gadzikwa, O. K. Farha, C. D. Malliakas, M. G. Kanatzidis,
J. T. Hupp and S. T. Nguyen, J. Am. Chem. Soc., 2009, 131, 13613–
13615.
10 R. K. Deshpande, G. I. N. Waterhouse, G. B. Jameson and
S. G. Telfer, Chem. Commun., 2012, DOI: 10.1039/C1CC12884A.
11 R. K. Deshpande, J. L. Minnaar and S. G. Telfer, Angew. Chem., Int.
Ed., 2010, 49, 4598–4602.
12 K. K. Tanabe, C. A. Allen and S. M. Cohen, Angew. Chem., Int. Ed.,
2010, 49, 9730–9733.
13 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–
850.
Conclusion
We have successfully incorporated two bulky BDC ligands into
an IRMOF lattice. L1 and L2 have the same HO-BDC ligand
protected with a photolabile group; 2-nitrobenzyl (L1) or 4,5-
dimethoxy-2-nitrobenzyl (L2). These groups are known to easily
cleave upon exposure to near-UV light and the differences in
electronics should alter the deprotection as a function of wave-
length used. The IRMOFs prepared from these ligands were
irradiated at either 365 or 400 nm to induce PSD of the
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 10188–10194 | 10193

View Online
14 A. D. Burrows, CrystEngComm, 2011, 13, 3623–3642.
15 M. Kim, J. F. Cahill, K. A. Prather and S. M. Cohen, Chem.
Commun., 2011, 47, 7629–7631.
16 A. D. Burrows, L. C. Fisher, C. Richardson and S. P. Rigby, Chem.
Commun., 2011, 47, 3380–3382.
23 A. Albini and M. Fagnoni, Handbook of synthetic photochemistry,
Wiley-VCH, Weinheim, 2010.
24 A. J. Blake, N. R. Champness, T. L. Easun, D. R. Allan,
H. Nowell, M. W. George, J. Jia and X.-Z. Sun, Nat. Chem.,
2010, 2, 688–694.
17 S. Marx, W. Kleist, J. Huang, M. Maciejewski and A. Baiker, Dalton
Trans., 2010, 39, 3795–3798.
25 K. Koh, A. G. Wong-Foy and A. J. Matzger, Angew. Chem., Int. Ed.,
2008, 47, 677–680.
18 T. Lescouet, E. Kockrick, G. Bergeret, M. Pera-Titus and
D. Farrusseng, Dalton Trans., 2011, 40, 11359–11361.
19 J. A. Mccray and D. R. Trentham, Annu. Rev. Biophys. Biophys.
Chem., 1989, 18, 239–270.
20 C. G. Bochet, Tetrahedron Lett., 2000, 41, 6341–6346.
21 G. Mayer and A. Heckel, Angew. Chem., Int. Ed., 2006, 45, 4900–4921.
22 H. Sato, R. Matsuda, K. Sugimoto, M. Takata and S. Kitagawa, Nat.
Mater., 2010, 9, 661–666.
26 Z. Wang and S. M. Cohen, J. Am. Chem. Soc., 2007, 129, 12368–
12369.
27 M. Gaplovsky, Y. V. Il’ichev, Y. Kamdzhilov, S. V. Kombarova,
M. Mac, M. A. Schworer and J. Wirz, Photochem. Photobiol. Sci.,
2005, 4, 33–42.
28 J. H. Cavka, S. Jakobsen, U. Olsbye, N. Guillou, C. Lamberti,
S. Bordiga and K. P. Lillerud, J. Am. Chem. Soc., 2008, 130,
13850–13851.
10194 | J. Mater. Chem., 2012, 22, 10188–10194
This journal is ª The Royal Society of Chemistry 2012