Effect of linker and metal on photoreduction and cascade reactions of nitroaromatics by M-UiO-66 metal organic frameworks

Article abstract of DOI:10.1016/j.ica.2019.119076

The use of metal organic frameworks (MOFs) as photocatalysts is a promising and growing area of research. Given the diverse architectures, linkers, and metals, it is important to understand their effects on catalysis. Herein we compare six MOFs of the UiO-66 family towards photocatalytic reduction of nitro-aromatics to anilines. These MOFs differ in metal identity (Hf, Zr) and linker, and hence this systematic study provides insights to developing next generation MOFs. We found that Hf-based MOFs are superior to the more commonly studied Zr-analogues. Moreover, the linker identity also impact the photocatalysis, with pyridine-based linkers out-performing aniline based linkers and those that lack an embedded basic site. The MOFs studied have unique selectivities for the photoreduction and also allow for the one-pot synthesis of imines from aromatic aldehydes and nitro-aromatics.

Full text of DOI:10.1016/j.ica.2019.119076

InorganicaChimicaActa497(2019)119076
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Research paper
Effect of linker and metal on photoreduction and cascade reactions of
nitroaromatics by M-UiO-66 metal organic frameworks
T. Elkin, C.T. Saouma⁎
University of Utah, Department of Chemistry, 315 1400 E, Salt Lake City, UT 84112, USA
A R T I C L E I N F O
A B S T R A C T
Keywords:
MOFs
Photocatalysis
Nitroaromatics
Photoreduction
The use of metal organic frameworks (MOFs) as photocatalysts is a promising and growing area of research.
Given the diverse architectures, linkers, and metals, it is important to understand their effects on catalysis.
Herein we compare six MOFs of the UiO-66 family towards photocatalytic reduction of nitro-aromatics to ani-
lines. These MOFs differ in metal identity (Hf, Zr) and linker, and hence this systematic study provides insights to
developing next generation MOFs. We found that Hf-based MOFs are superior to the more commonly studied Zr-
analogues. Moreover, the linker identity also impact the photocatalysis, with pyridine-based linkers out-per-
forming aniline based linkers and those that lack an embedded basic site. The MOFs studied have unique se-
lectivities for the photoreduction and also allow for the one-pot synthesis of imines from aromatic aldehydes and
nitro-aromatics.
1. Introduction
making the UiO-66 MOFs stable under a variety of conditions [18].
Finally, the carboxylate linker can be varied, for example, extended the
Metal organic frameworks (MOFs) are gaining traction in the field of
catalysis due to their unique properties that combine those of hetero-
geneous materials and soluble homogeneous complexes [1–5]. In some
instances, the MOF architecture itself serves as the catalyst [6,7], with
reactivity occurring at the metal node or an organic linker. In others,
established homogenous catalysts are incorporated into the MOF [8,9]
or metal nano-particles are placed in the pores [10,11]. Regardless of
what role the MOF plays, it is important to understand how modifica-
tion of the MOF impacts the overall catalysis; rate, substrate scope, and
catalyst stability are all factors that can be impacted by MOF mod-
ification.
The group IV based UiO-66 family of MOFs has emerged as pro-
mising catalysts or catalyst platforms [2,3]. These MOFs are made up of
M6(μ3-O)4(μ3-OH)4(BDC)6, where BDC represents para-benzene di-
carboxylate. While the original UiO-66 has M = Zr (termed Zr-UiO-66
in this manuscript) [12], analogues with partial or complete substitu-
tion with Hf [13], Th [14], Ti [15], U [16], and Ce [17] have since been
reported (termed M-UiO-66 or MM’-UiO-66 for pure and mixed metal
analogues, respectively). Moreover, the metal nodes can have variable
number of dicarboxylate linkers, varying the connection number from
12 to 6. This provides much flexibility in pore size and electronic
properties without compromising stability [3]. The stability arises from
high-valent hard metals forming strong bonds to the hard carboxylates,
length of the linker, or adding substitution to the aromatic ring, ren-
dering the MOFs chromophores [19].
As a result of these properties, UiO-66 MOFs have been widely
studied for catalytic transformations. For example, the Lewis-acidic Zr-
UiO-66 MOFs catalyze the cyclization of citronellal [20]. Moreover,
variation of the benzene dicarboxylate linker (and hence Lewis acidity)
shows good correlation of rate to Hammett parameter, which serves as
direct evidence that the MOF itself is the catalyst. Similar reactivity
trends were also observed in the acetalization of benzaldehyde [21].
Not all Lewis-acid catalysed UiO-66 reactions follow a linear relation-
ship with Hammett constants, which can be explained by dual roles for
the linkers: inductive modulation of the Lewis acidity at the metal and
variation of the Lewis basicity at the linker [22].
The UiO-66 family of MOFs also serve as photocatalysts, with ap-
plications that range from CO2 reduction to oxidation of organic mo-
lecules [7,15,23-26]. Theoretical studies indicate that initial light ab-
sorption excites an electron from the highest occupied crystal orbital
(HOCO) to the lowest unoccupied crystal orbital (LUCO), both of which
primarily reside on the linkers of UiO-66 MOFs [19,27,28]. Subsequent
metal-to-ligand charge transfer (MLCT) from the linker to node then
allows for charge separation, which increases the lifetime of the excited
state and hence photocatalytic activity, and also reduces the catalyst
active-site. Modulation of the linker shifts the absorption wavelength
⁎ Corresponding author.
E-mail address: caroline.saouma@utah.edu (C.T. Saouma).
https://doi.org/10.1016/j.ica.2019.119076
Received 15 July 2019; Received in revised form 13 August 2019; Accepted 13 August 2019
Availableonline19August2019
0020-1693/©2019ElsevierB.V.Allrightsreserved.

T. Elkin and C.T. Saouma
InorganicaChimicaActa497(2019)119076
from UV to visible, allowing for facile tuning of the excitation energy
[19]; it was found that linker defects also impact these energies [27].
This may explain why different groups see different reactivities, for
example in the photocatalytic reduction of CO2 to formate [7,15]. Fi-
nally, changing the metal identity also impacts photocatalysis, pri-
marily by altering the MLCT energy [28]. These studies underscore the
need to study the effect of both linker and metal on catalytic processes,
and to determine whether the effects are due to differing electronics or
differing morphologies.
crystalline nature of the MOFs and are in agreement of previously re-
ported in the literature for the related UiO-66 MOFs [26,35,36,39].
Thermogravimetric analysis (TGA) of newly prepared MOFs Hf-py and
Zr-py (Fig. 1b) indicate that the MOFs are thermally robust, and consist
of initial mass loss while heating to 100 °C due to the presence of the
remaining solvents, followed by loss of the linker above 500 °C [39].
Elemental analysis of Hf-py is consistent with a structure that has two
missing linkers, while Zr-py has three linkers missing (see Table S1).
Our syntheses yield one more missing linker than that reported by Stock
for each metal [37].
Many photocatalytic studies of Zr-UiO-66 compare Zr-UiO-66 to
related MOFs whereby the BDC linker has an amino substitution, ty-
pically Zr-UiO-66-NH2 (Zr-UiO-66-NH2 = Zr6(μ3-O)4(μ3-OH)4(NH2-
BDC)6 [7,15,26,29]. The latter has heightened photocatalytic activity,
which is ascribed to it’s ability to absorb in the visible region. We hy-
pothesized that the enhanced reactivity of the amino-substituted MOF
may also be in part due to its ability to conduct protons, which is central
to proton-coupled electron transfer (PCET) redox transformations. (In
this manuscript, PCET refers to a reaction where an electron and proton
is transferred, regardless of whether it occurs in a stepwise or concerted
manner). To test this, we sought to prepare and evaluate the photo-
catalytic activity of Zr-UiO-66-py, whereby the BDC linker is replaced
by 2,5-dicarboxylate pyridine. Pyridine is slightly more basic than
aniline in MeCN (12.53 versus 10.62) [30], and thus may offer insight
into the acid-base properties. Moreover, several reports suggest that
pyridine is essential to the transfer of proton and electron equivalents in
a variety of transformations, including the direct reduction of CO2 to
MeOH (both electrochemically [31] and photocatalytically [32,33]),
and its role in Schrock’s N2 reduction catalyst that yields NH3 [34].
Herein we disclose the results of our study that looks at the effect of
linker and metal identities on catalytic reductions. The Zr and Hf ana-
logues of M-UiO-66-NH2, M-UiO-66-py and M-UiO-66 were prepared,
and their photocatalytic reactivity towards the reduction of nitroaro-
matics to anilines compared. Moreover, we describe a one-pot photo-
catalytic condensation of nitroaromatics and aldehydes to imines.
We observe that the particle size of Hf-py is smaller (35–125 nm)
than that Zr-py (450–550 nm) (Fig. 2 and ESI, Fig. S2). Both Hf-py and
Zr-py have semi-spherical morphologies, which may be due to the fast
nucleation of the MOF crystals under the aqueous reflux conditions in
the presence of the acetic acid modulator [43]. BET analysis (ESI, Table
S3) [40,41] reveals surface areas of 279 m2/g for Hf-py and 480 m2/g
for Zr-py. While these values are low for UiO-66 MOFs [42], lower
values [35] were previously observed for the hydrothermally prepared
UiO-66 MOFs. We were unable to estimate the accurate pore size for Hf-
py based on N2 adsorption, and we find that for Zr-py the average pore
size is 15.66 Å. One of the reasons for not being able to measure the
pore size properly may arise from inability to completely remove water
from the pores, as evident from the TGA analysis as well (Fig. 1b).
However, we did not subject MOF Hf-py to high temperatures for
longer periods of time than was stated in the experimental procedure,
since it caused the material to turn completely into black powder ma-
terial, indicating full degradation of Hf-py. Given that all other MOFs
were prepared following literature protocol [12,35,36], we assume si-
milar surface area and porosity to what is published.
With the six MOFs prepared, we sought to establish the effect of
linker and metal on the photocatalytic reduction of nitrobenzenes to
aniline derivatives. This reaction was chosen as it is pertinent to pol-
lution remediation (due to adverse health effects) [43,44] and synth-
esis, the aniline derivatives being precursors to fine chemicals [45].
While the photocatalysis of Zr-NH2 and the effect of linker on photo-
catalysis is better established [7,15,23-26], very few studies have been
done on Hf-derived MOFs [46,47].
2. Results and discussion
i
We sought to initiate or studies in neat PrOH, as this serves as the
Zr-UiO-66 (Zr-H), Zr-UiO-66-NH2 (Zr-NH2), Hf-UiO-66-H (Hf-H),
and Hf-UiO-66-NH2 (Hf-NH2) were prepared following standard lit-
erature methods [12,35,36]. The synthesis of Zr-UiO-66-py (Zr-py) and
Hf-UiO-66-py (Hf-py) are described below. While this manuscript was
being prepared, a report by Stock and co-workers [37] described the
syntheses of these MOFs. We nonetheless describe our synthetic pro-
tocol to highlight similarities and differences.
The syntheses of UiO-66-py MOFs that incorporate a pyridine di-
carboxylate linker via solvothermal method did not lead to desired
analogues, unlike the ease of access to the UiO-67 bi-pyridine di-
carboxylate MOFs [38]. For example, our attempts to prepare Zr-py via
standard solvothermal methods using DMF as a solvent led to the for-
mation of amorphous solids (ESI, Fig. S1). However, we found that both
Hf-py and Zr-py can be prepared following the recently developed
hydrothermal synthesis (Scheme 1) [37], which uses water as a solvent
and acetic acid as a modulator.
solvent and has desirable viscosity and solubility properties. It has also
been shown to serve as a sacrificial reductant in related studies [48],
presumably being converted to acetone. Visible light (λ > 400 nm)
irradiation of Hf-py in neat iPrOH with nitrobenzene results in 89%
conversion to aniline after 48 h. Repeating the experiment using UV–vis
light (Hg/Xe 200 W lamp) gave the same level of reduction to aniline
after 6 h. For this reason, all subsequent screening reactions were set up
using the full UV–vis spectrum (Scheme 2).
When the photolysis experiment is repeated with Zr-py, only 11% of
aniline is produced (Table 1) after 18 h of irradiation with UV-light. The
discrepancy may be due to changing the metal or may be due to
changes in the surface area to volume ratio of the MOFs.
As crystals of Hf-py are significantly smaller than those of Zr-py, the
surface area exposed to bulk solution to volume ratio of Hf-py is ~5
times greater than that of Zr-py (assuming all particles are spheres and
taking average radius into the calculation, see Table S2, ESI). The
greater surface area results in more efficient exposure of the MOF
particles to light and thereby results in generation of more active-sites
on the surface, which then should increase turnover frequency of ni-
trobenzene reduction, as mass transport issues are minimized (from
substrate/product entering/leaving the pores, and/or protons and
electrons migrating to the surface from the middle of the MOF).
Alternatively, based on standard reduction potentials, the reduced
Hf MOFs should be more reducing than the Zr counter-parts [49].
Consistent with this, supported Zr-NH2 has a less negative flat-band
potential than Hf-NH2 [47]. Indeed, we observe better reactivity with
Hf-py than Zr-py. It should be noted that others have observed the
reverse reactivity trend with a different linker [46], which likely is due
The PXRD patterns (Fig. 1a) of Hf-py and Zr-py point to the
Scheme 1. Synthesis of Hf-py and Zr-py.
2

T. Elkin and C.T. Saouma
InorganicaChimicaActa497(2019)119076
Fig. 1. Characterization of Hf-py (red) and Zr-py (black): a) PXRD patterns; b) TGA curves. (For interpretation of the references to colour in this figure legend, the
reader is referred to the web version of this article.)
Fig. 2. SEM images of Hf-py and Zr-py: a) Hf-py, magnification ×100 000, HV 1.00 kV, current 6.3 pA, det TLD. b) Zr-py, magnification ×100 000, HV 1.00 kV,
current 25 pA, det TLD.
was surprising, though over-interpretation should be avoided, as dif-
ferent sacrificial reductants were used. By contrast, Hf-H and Hf-NH2
gave 4% and 80% yield of aniline after 24 h, respectively. The reactivity
trend does not correlate to crystal size (Table 1), suggesting that other
parameters must be considered; both the linker and metal identity may
contribute to catalytic reactivity. Defects could also be a factor [27],
given their effect on band gap and ability to aid in mass transfer.
Scheme 2. Photoreduction of nitroaromatic compounds with MOFs as catalysts
The Hf MOFs outperform the Zr analogues, suggesting that they may
and iPrOH as sacrificial reductant and solvent. Catalyst: Hf, Hf-py, Hf-NH2, Zr,
be better suited for catalytic reductions than the more commonly stu-
Zr-py, or Zr-NH2. See Table 2 footnotes for experimental details.
died Zr MOFs. When considering reductions that occur via net PCET,
both the reduction potential and pKa contribute [50]. Hupp, Farha and
co-workers measured the pKas of the bridging hydroxide and aquo li-
gands in a series of Hf and Zr MOFs, including M-H. They found that the
pKas were very similar as the metal was varied [51]. Assuming this also
Table 1
Photoreduction of nitrobenzene with various MOFs in iPrOH under UV–Vis ir-
radiation.
MOF
Time [h]
Yield [%]
Crystal Size [nm]
holds for the reduced MOFs, then the reduction potential may be suf-
ficient to predict reactivity towards reductive PCET in the UiO-66 fa-
mily of MOFs; for a given pKa, a more negative reduction potential
would correspond to a weaker bond dissociate free energy and hence a
more reducing H-atom [50].
The reactivity of the Hf MOFs decreases with linker identity in the
following order: py > bdc-NH2 > bdc. This finding may be rationa-
lized by the ability of pyridine to facilitate proton transfer from iPrOH
or the reduced cluster nodes to the nitroaromatic compound in overall
PCET reactions [52]. The protic nature of the bdc-NH2 may likewise
facilitate the transfer of protons from buried nodes to the surface. Al-
ternatively, it may be due to the ease of photo-excitation by the various
linkers, which would point to the smaller gap between the primarily
linker in character LUMO orbital and primarily metal cluster HOMO
orbital [53]. Given the experimental results, we cannot rule out either
Hf-py
Hf-NH2
Hf-H
Zr-py
Zr-NH2
Zr-H
6
89
80
4
11
0
~80
24
24
18
24
24
~10,000
~140
~400
~50
0
~350
to modulation of both the HOCO-LUCO gap and the MLCT energies.
To determine if the linker identity impacts the reactivity, UiO-66
and UiO-66-NH2 analogues of both Hf and Zr were likewise tested in the
same photoreduction of nitrobenzene (Table 1). Both Zr-H and Zr-NH2
showed no catalytic activity towards the reduction of nitrobenzene over
24 h. Given that Zr-NH2 can photocatalytically reduce CO2 to formate
[26,29] and protons to H2 [30,31], the lack of nitrobenzene reduction
3

T. Elkin and C.T. Saouma
InorganicaChimicaActa497(2019)119076
Table 2
photoreduce NO2 functionality selectively, without damaging poten-
tially reducible groups such as CN and ester, similar to that observed in
other photocatalytic systems [55,56]. Moreover, the MOF catalysts
described here do not reduce p-OH and p-NH2 nitrobenzenes as other
systems do [45,55-57], indicating finer control and selectivity to spe-
cies that are reduced.
Placing the substituent ortho to the nitro group shuts down the re-
activity completely for both Hf-py and Zr-py. For example, placing a
phenyl group ortho to the nitro functionality gives no corresponding
aniline, which may be rationalized on steric grounds. The ortho phenol
likewise gives no aniline, which again may be due to deprotonation of
the phenol, changing the reduction potential [50]. While placing the
aldehyde function ortho to the nitro group again results in lack of re-
activity for Zr-py, formation of the corresponding alcohol instead of
aniline was observed when Hf-py was used as a catalyst. This indicates
that reduction of the aldehyde may occur prior to that of the nitro-
functionality, as no imine oligomers are observed as is the case in p-
CHO. This result was unexpected though not unique [45], as alcohols
typically can be used as sacrificial reductants in these systems, giving
the corresponding aldehyde as the oxidized product.
The stability of the catalysts Hf-py and Zr-py was tested in three
consecutive reduction reactions of p-CNC6H4NO2. The MOFs were
shown to be robust, showing minor peak broadenings in their PXRD
patterns (Fig. 3a) and not losing catalytic efficiency (Fig. 3b).
Both Hf-py and Zr-py, facilitate further chemistry with p-CHO ni-
trobenzene to give an imine functionality (vide supra). Our system is the
first, to our knowledge, to demonstrate such photocatalytic behaviour
when an aldehyde is introduced into the nitroaromatic ring. To prove
the hypothesis of the in situ imine generation from photoreduced ni-
troaromatics, we tested MOFs Hf-py and Zr-py in intermolecular cas-
cade catalytic transformation between nitrobenzene and benzaldehyde,
allowing the formed aniline to react and give the corresponding imine.
Reduction scope of nitroaromatic compounds.
Yield of anilines (%)(a)
R
σp [54]
Hf-py
Zr-py
p-NO2
p-CN
p-CF3
p-C(O)Me
p-CHO
p-OC(O)Me
H
p-Me
p-OMe
p-OH
p-NH2
p-NEt2
o-Ph
o-CHO
o-OH
0.78
0.66
0.54
0.50
0.42
0.31
0.0
−0.17
−0.27
−0.37
−0.66
−0.72
–
93
95
55
73
0(c)
21
89
56
20
4
48
87
25
22
0(c)
6
11
5
4
2
0
0
0
5
0
0
–
–
11(b)
0
0
0
In a typical catalytic reaction, 20 mg of MOF, 1 mmol of substrate and 20 mL of
iPrOH were put into the 25 mL Teflon-capped quartz tube with a stir bar. Then
reaction vessel was closed and placed 15 cm from 200 W Hg/Xe lamp irradia-
tion for 20 h, while the stirring rate was set to 900 rpm. After 20 h, an aliquot of
the reaction was transferred through a short Celite plug to remove the solids
and injected into GC (a) GC yield; average of three runs. (b) Product is the
corresponding alcohol. (c) Product speculated to be a mixture of imine oligo-
mers.
of these scenarios.
As shown in Table 2, a variety of nitroaromatics can be reduced
photocatalytically with either Hf-py or Zr-py in iPrOH without any
special precautions in terms of experimental conditions such as oxygen
or water removal. For all nitroaromatics tested, Hf-py is more active
than Zr-py under the same conditions. The reactivity trend regarding
the chemical nature of the nitrocompound remains the same for both
MOFs. We observed that both the position and the electronic nature of
the substituent influence the outcome of the reaction.
With the substituent placed farther away in the para- position, the
electronic properties of the substituent dominate the reactivity, with
Hammett parameters providing an approximation for reaction yield.
More electron withdrawing groups promote the photoreduction of the
nitroaromatic compounds to the corresponding anilines more efficiently
for both Hf-py and Zr-py. While the electronic effects appear to dom-
inate, sterics may also play a role, given that the reactions may proceed
inside the pores. Consistent with this, we did not see a correlation of
crystal size and reactivity (Table 1), indicating that the reactions likely
occur at the surface and within the pores. We have observed photo-
catalytic reactivity of MOFs in related system that must occur at the
surfaces of the crystals [48]. Moreover, the basic pyridine linker may
deprotonate acidic protons, which may account for the drop in yield for
both p-OH and p-NH2 nitrobenzenes, rendering the reduction more
challenging [50]. With p-CHO, no aniline is quantified; the presence of
larger m/z peaks in the MS suggests instead a mixture of imine oligo-
mers form. This suggests that the reduced aniline condenses with the
aldehyde (vide infra).
The influence of the electronic effect has been corroborated further
by using 1,4-dinitrobenzene as a substrate which produced the 4-ni-
troaniline with 93% yield for Hf-py and 48% for Zr-py. Further irra-
diation yielded only trace amounts of the fully reduced diamino-com-
pound when Hf-py was used as catalyst, and no fully reduced product
was observed for Zr-py, supporting the inhibitory nature of para-elec-
tron donating substituents on the reduction of the nitroaromatics to
their corresponding anilines. Additional confirmation of the trend has
been offered by using –NEt2, a strongly electron donating moiety, re-
sulting in no conversion from the starting material. From this panel of
the nitrocompounds, we have observed that both Hf-py and Zr-py
Fig. 3. Stability of MOFs in three consecutive catalytic reductions of p-
CNC6H4NO2: a) PXRD patterns of Hf-py (black and grey) and Zr-py (magenta
and red) before (bottom) and after (top) three consecutive photoreduction re-
actions; b) chemical reactivity of MOFs in three consecutive runs of reduction.
(For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
4

T. Elkin and C.T. Saouma
InorganicaChimicaActa497(2019)119076
combining nitro-aromatics and aromatic aldehydes in one-pot to
cleanly give the resulting mixed imine species.
4. Experimental
4.1. General conditions
HfCl4, ZrCl4 (Strem Chemicals), Zr(NO3)4x5H2O (Ivy Chemicals),
2,5-pyridinedicarboxylic acid, terephthalic acid, aminoterephthalic
acid (Sigma-Aldrich) were used without any further purification.
Following MOFs: Zr-UiO-66,2 Zr-UiO-66-NH2 [35], Hf-UiO-66 and Hf-
UiO-66-Hf [36] were prepared following the literature procedures. Gas
Chromatograms were obtained equipped with FID, using non-chiral
stationary phase capillary column and He as carrier gas. Photocatalytic
reactions were carried out in quartz 25 mL tubes, using Newport 200W
Hg/Xe lamp.
4.2. Synthesis of Hf-UiO-66-py (Hf-py).
The 5.1 g (5.3 mmol) of HfCl4 and 2.4 g of 2,5-pyridinedicarboxylic
(14 mmol) acid were loaded into 150 mL of H2O/Acetic acid (3/2) so-
lution. The resulting suspension was stirred at rate 200 rpm during the
heating to the reflux (105–110 °C) to evenly suspend the two solids in
the suspension. After the reaction reached reflux temperature, the
suspension was kept under static conditions for 24 h. After that time the
suspension was allowed to cool to the room temperature and filtered
through 0.45 µm filter paper. The remaining white solid was soaked in
anhydrous MeOH for 3 days, during that period the solvent phase was
changed to fresh MeOH each day. After the solvent has been removed,
the remaining white solid was activated by heating to 120 °C under
vacuum for 16 h, yielding MOF 1 as reddish powder. Yield 83%, based
on HfCl4.
Scheme 3. Imine condensation scope, blue: Hf-py, red: Zr-py, (number): yield
of alcohol. (For interpretation of the references to colour in this figure legend,
the reader is referred to the web version of this article.)
The formation of the corresponding imine in good yields (81% for Hf-
py, 93% for Zr-py) proves the participation of the reduced ni-
trocompound in the consecutive imine condensation with the aldehyde.
An alternative photochemical mechanism would invoke the participa-
tion of molecular oxygen, as has been shown to occur in Ti MOFs [58].
We rule out this mechanism as it allows for homo-condensation of
amines to imines, and it also proceeds via a carbon-based radical that is
not feasible when aromatic amines are employed.
The scope has been investigated using different aldehydes and ni-
troaromatics (Scheme 3). In general, the yield of imines is higher when
Zr-py is used rather than Hf-py, which contrasts the trend for ni-
troaromatic reduction (Tables 1 and 2). This discrepancy in activity
may be due to 1) slower conversion of the nitrocompound to the amine
in the presence of benzaldehyde, and/or 2) competitive reduction of the
aldehyde to the corresponding alcohol. Indeed, Hf-py is capable of
reducing the aldehydes to the corresponding alcohols under the pho-
tocatalytic conditions, albeit at sluggish rates (SI, Table S6). Both of
these effects point to the possibility of the competition between the
aldehyde and nitro functional groups for the active-site in Hf-py and
point to Hf being more reducing than Zr.
When Zr-py is used as a catalyst for the cascade catalysis, reduction
of the aldehydes is not observed at all. Only a trace amount (3%) of
alcohol is detected when ortho-Br benzaldehyde is used. The yields are
either similar or higher to the single-step photoreduction of the corre-
sponding nitrocompounds. This indicates that the competition between
aldehyde and nitro groups for the Zr-based catalytic site is either weak
or non-existent, contradictory to the Hf analogue.
4.3. Synthesis of Zr-UiO-66-py (Zr-py)
The same procedure described for 1 was applied to get 2, using 6.8 g
(15.6 mmol) of Zr(NO3)4x5H2O as a metal precursor, 2.57 g (15 mmol)
of 2,5-pyridinedicarboxylic acid as a linker and 150 mL of H2O/Acetic
acid (3/2) solution as a solvent and modulator mixture. Yield 81%,
based on Zr(NO3)4x5H2O.
4.4. General photoreduction conditions
In a typical catalytic reaction, 20 mg of MOF, 1 mmol of substrate
and 20 mL of iPrOH were put into the 25 mL J-Young quartz tube. Then
reaction vessel was closed and placed 15 cm from 200 W Hg/Xe lamp
irradiation for 20 h, while the stirring rate was set to 900 rpm. After
20 h, an aliquot of the reaction was transferred through a short Celite
plug to remove the solids, and injected into GC.
4.5. General catalytic photocascade reactions
In a typical catalytic reaction, 20 mg of MOF, 1 mmol of nitro
compound, 1 mmol of aldehyde and 20 mL of iPrOH were put into the
25 mL Teflon-capped quartz tube. Then reaction vessel was closed and
placed 15 cm from 200 W Hg/Xe lamp irradiation for 20 h, while the
stirring rate was set to 900 rpm. After 20 h, an aliquot of the reaction
was transferred through a short Celite plug to remove the solids, and
injected into GC–MS.
3. Conclusions
In sum, we have conducted a comparative study that probed the role
of both the metal and linker identities on photo-reduction reactions. For
the reduction of nitro-aromatics to anilines, we found that Hf-based
MOFs have superior activity to their Zr-analogues. Moreover, the linker
also impacts catalysis, with py > NH2 > H. We emphasize that the
MOFs also differed in their crystal sizes and defects, making it im-
possible to generalize these trends. However, the results indicate that
both metal and linker can contribute to reactivity. We found that Hf-py
has different selectivity for the reduction of nitroaromatics to anilines
than related systems, providing a potential platform to selectively re-
duce mixtures of nitro-aromatics. Finally, we showed the feasibility of
Acknowledgements
Funds for this work were provided by the University of Utah
Department of Chemistry and USTAR start-up funds. We thank Kyle
O’Malley and Prof. Whitty, Combustion and Energy Laboratory of the
Department of Chemical Engineering, University of Utah (BET) and
5

T. Elkin and C.T. Saouma
InorganicaChimicaActa497(2019)119076
Nanofab (SEM), University of Utah. We also thank Moumita
Bhattacharya and Katy Webb for help and fruitful scientific discussions.
48 (2012) 11656–11658.
[25] L. Shen, R. Liang, M. Luo, F. Jing, L. Wu, PCCP 17 (2014) 117–121.
[26] L. Shen, S. Liang, W. Wu, R. Liang, L. Wu, Dalt. Trans. 42 (2013) 13649–13657.
[27] A. De Vos, K. Hendrickx, P. Van Der Voort, V. Van Speybroeck, K. Lejaeghere,
Chem. Mater. 29 (2017) 3006–3019.
[28] X.-P. Wu, L. Gagliardi, D.G. Truhlar, J. Am. Chem. Soc. 140 (2018) 7904–7912.
[29] M. Sun, S. Yan, Y. Sun, X. Yang, Z. Guo, J. Du, D. Chen, P. Chen, H. Xing, Dalt.
Trans. 47 (2017) 909–915.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.ica.2019.119076.
[30] I. Kaljurand, A. Kütt, L. Sooväli, T. Rodima, V. Mäemets, I. Leito, I.A. Koppel, J. Org.
Chem. 70 (2005) 1019–1028.
[31] E. Barton Cole, P.S. Lakkaraju, D.M. Rampulla, A.J. Morris, E. Abelev, A.B. Bocarsly,
J. Am. Chem. Soc. 132 (2010) 11539–11551.
References
[32] D.J. Boston, Y.M.F. Pachón, R.O. Lezna, N.R. de Tacconi, F.M. MacDonnell, Inorg.
Chem. 53 (2014) 6544–6553.
[33] D.J. Boston, C. Xu, D.W. Armstrong, F.M. MacDonnell, J. Am. Chem. Soc. 135
(2013) 16252–16255.
[34] D.V. Yandulov, R.R. Schrock, Science 301 (2003) 76–78.
[35] Z. Hu, A. Nalaparaju, Y. Peng, J. Jiang, D. Zhao, Inorg. Chem. 55 (2016)
1134–1141.
[36] G.W. Peterson, J.B. DeCoste, F. Fatollahi-Fard, D.K. Britt, Ind. Eng. Chem. Res. 53
(2014) 701–707.
[37] S. Waitschat, D. Fröhlich, H. Reinsch, H. Terraschke, K.A. Lomachenko, C. Lamberti,
H. Kummer, T. Helling, M. Baumgartner, S. Henninger, N. Stock, Dalt. Trans. 47
(2017) 1062–1070.
[38] H. Fei, S.M. Cohen, Chem. Commun. 50 (2014) 4810–4812.
[39] M.J. Katz, Z.J. Brown, Y.J. Colon, P.W. Siu, K.A. Scheidt, R.Q. Snurr, J.T. Hupp,
O.K. Farha, Chem. Commun. 49 (2013) 9449–9451.
[1] A. Dhakshinamoorthy, A.M. Asiri, H. Garcia, Angew. Chem. Int. Ed. 55 (2016)
5414–5445.
[2] Y. Bai, Y. Dou, L.-H. Xie, W. Rutledge, J.-R. Li, H.-C. Zhou, Chem. Soc. Rev. 45
(2016) 2327–2367.
[3] S. Yuan, J.-S. Qin, C.T. Lollar, H.-C. Zhou, ACS Central Science 4 (2018) 440–450.
[4] C.-C. Wang, J.-R. Li, X.-L. Lv, Y.-Q. Zhang, G. Guo, Energy Environ. Sci. 7 (2014)
2831–2867.
[5] F.X. Llabrés i Xamena, J. Gascon (Eds.), Metal Organic Frameworks as
Heterogeneous Catalysts, The Royal Society of Chemistry, Cambridge, 2013.
[6] J. Yano, V.K. Yachandra, V.T. Voorhis, D.-M. Catalysis, Catalysis (2017).
[7] D. Sun, Y. Fu, W. Liu, L. Ye, D. Wang, L. Yang, X. Fu, Z. Li, Chem. Eur. J. 19 (2013)
14279–14285.
[8] H. Fei, M.D. Sampson, Y. Lee, C.P. Kubiak, S.M. Cohen, Inorg. Chem. 54 (2015)
6821–6828.
[40] T.C. Wang, W. Bury, D.A. Gómez-Gualdrón, N.A. Vermeulen, J.E. Mondloch,
P. Deria, K. Zhang, P.Z. Moghadam, A.A. Sarjeant, R.Q. Snurr, J.F. Stoddart,
J.T. Hupp, O.K. Farha, J. Am. Chem. Soc. 137 (2015) 3585–3591.
[41] S. Brunauer, P.H. Emmett, E. Teller, J. Am. Chem. Soc. 60 (1938) 309–319.
[42] Z. Hu, D. Zhao, Dalton Trans. 44 (2015) 19018–19040.
[43] C. Li, S. Taneda, A.K. Suzuki, C. Furuta, G. Watanabe, K. Taya, Toxicol. Appl.
Pharmacol. 217 (2006) 1–6.
[44] H. Huang, J. Zhou, H. Liu, Y. Zhou, Y. Feng, J. Hazard. Mater. 178 (2010) 994–998.
[45] Ö. Karahan, E. Biçer, A. Taşdemir, A. Yürüm, S. Gürsel, Eur. J. Inorg. Chem. 2018
(2018) 1073–1079.
[9] X. Zhang, N.A. Vermeulen, Z. Huang, Y. Cui, J. Liu, M.D. Krzyaniak, Z. Li, H. Noh,
M.R. Wasielewski, M. Delferro, O.K. Farha, ACS Appl. Mater. Interfaces 10 (2018)
635–641.
[10] A. Crake, K.C. Christoforidis, A. Kafizas, S. Zafeiratos, C. Petit, Appl. Catal. B 210
(2017) 131–140.
[11] E. Gutterød, S. Øien-Ødegaard, K. Bossers, A.-E. Nieuwelink, M. Manzoli, L. Braglia,
A. Lazzarini, E. Borfecchia, S. Ahmadigoltapeh, B. Bouchevreau, Ind. Eng. Chem.
Res. 56 (2017) 13206–13218.
[12] J.H. Cavka, S. Jakobsen, U. Olsbye, N. Guillou, C. Lamberti, S. Bordiga,
K.P. Lillerud, J. Am. Chem. Soc. 130 (2008) 13850–13851.
[13] S. Jakobsen, D. Gianolio, D.S. Wragg, M.H. Nilsen, H. Emerich, S. Bordiga,
C. Lamberti, U. Olsbye, M. Tilset, K.P. Lillerud, Phys. Rev. B 86 (2012) 125429.
[14] C. Falaise, J.-S. Charles, C. Volkringer, T. Loiseau, Inorg. Chem. 54 (2015)
2235–2242.
[46] T.L.H. Doan, H.L. Nguyen, H.Q. Pham, N.N. Pham-Tran, T.N. Le, K.E. Cordova,
Chem. Asian J. 10 (2015) 2660–2668.
[47] X.-D. Du, X.-H. Yi, P. Wang, W. Zheng, J. Deng, C.-C. Wang, Chem. Eng. J. (2018).
[48] C.T. Saouma, S. Richard, S. Smolders, M.F. Delley, R. Ameloot, F. Vermoortele,
D.E. De Vos, J.M. Mayer, J. Am. Chem. Soc. 140 (2018) 16184–16189.
[49] M. Weller, T. Ovorton, J. Rourke, F. Armstrong, Inorganic Chemistry, Oxford
University Press, Oxford, 2018.
[15] Y. Lee, S. Kim, J.K. Kang, S.M. Cohen, Chem. Commun. 51 (2015) 5735–5738.
[16] C. Falaise, C. Volkringer, J.-F. Vigier, N. Henry, A. Beaurain, T. Loiseau, Chem. Eur.
J. 19 (2013) 5324–5331.
[50] J.J. Warren, T.A. Tronic, J.M. Mayer, Chem. Rev. 110 (2010) 6961–7001.
[51] R.C. Klet, Y. Liu, T.C. Wang, J.T. Hupp, O.K. Farha, J. Mater. Chem. A 4 (2016)
1479–1485.
[17] M. Lammert, M.T. Wharmby, S. Smolders, B. Bueken, A. Lieb, K.A. Lomachenko,
D.D. Vos, N. Stock, Chem. Commun. 51 (2015) 12578–12581.
[18] A. Polyzoidis, M. Schwarzer, S. Loebbecke, Microporous Mesoporous Mater. 208
(2015) 30–35.
[19] K. Hendrickx, D.E.P. Vanpoucke, K. Leus, K. Lejaeghere, A. Yperen-De Deyne,
V. Speybroeck, P. Voort, K. Hemelsoet, Inorg. Chem. 54 (2015) 10701–10710.
[20] F. Vermoortele, M. Vandichel, B. Van de Voorde, R. Ameloot, M. Waroquier, V. Van
Speybroeck, D.E. De Vos, Angew. Chem. Int. Ed. 51 (2012) 4887–4890.
[21] M.N. Timofeeva, V.N. Panchenko, J.W. Jun, Z. Hasan, M.M. Matrosova, S.H. Jhung,
Appl. Catal. A 471 (2014) 91–97.
[52] J. Bonin, C. Costentin, M. Robert, J.-M. Saveant, Org. Biomol. Chem. 9 (2011)
4064–4069.
[53] K. Hendrickx, D.E.P. Vanpoucke, K. Leus, K. Lejaeghere, A. Van Yperen-De Deyne,
V. Van Speybroeck, P. Van Der Voort, K. Hemelsoet, Inorg. Chem. 54 (2015)
10701–10710.
[54] C. Hansch, A. Leo, R.W. Taft, Chem. Rev. 91 (1991) 165–195.
[55] Z. Zand, F. Kazemi, S. Hosseini, Tetrahedron Lett. 55 (2014) 338–341.
[56] A. Cors, S.M. Bonesi, R. Erra-Balsells, Tetrahedron Lett. 49 (2008) 1555–1558.
[57] A. Hernández-Gordillo, S. Obregón, F. Paraguay-Delgado, V. Rodríguez-González,
RSC Adv. 5 (2015) 15194–15197.
[22] J.F. Blandez, A. Santiago-Portillo, S. Navalón, M. Giménez-Marqués, M. Álvaro,
P. Horcajada, H. García, J. Mol. Catal. A: Chem. 425 (2016) 332–339.
[23] C. Gomes Silva, I. Luz, F.X. Llabrés i Xamena, A. Corma, H. García, Chem. Eur. J. 16
(2010) 11133–11138.
[58] D. Sun, L. Ye, Z. Li, Appl. Catal. B 164 (2015) 428–432.
[24] J. Long, S. Wang, Z. Ding, S. Wang, Y. Zhou, L. Huang, X. Wang, Chem. Commun.
6