Elongated and substituted triazine-based tricarboxylic acid linkers for MOFs

Article abstract of DOI:10.3762/bjoc.12.219

New triazine-based tricarboxylic acid linkers were prepared as elongated relatives of triazinetribenzoic acid (TATB). Additionally, functional groups (NO₂, NH₂, OMe, OH) were introduced for potential post-synthetic modification (PSM) of MOFs. Functionalized tris(4-bromoaryl)triazine "cores" (3a,3b) were obtained by unsymmetric trimerization mixing one equivalent of an acid chloride (OMe or NO₂ substituted) with two equivalents of an unsubstituted nitrile. Triple Suzuki coupling of the cores 3 with suitable phenyl- and biphenylboronic acid derivatives provided elongated tricarboxylic acid linkers as carboxylic acids 17 and 20 or their esters 16 and 19. Reduction of the nitro group and cleavage of the methoxy group gave the respective amino and hydroxy-substituted triazine linkers.

Full text of DOI:10.3762/bjoc.12.219

Elongated and substituted triazine-based tricarboxylic acid
linkers for MOFs
Arne Klinkebiel, Ole Beyer, Barbara Malawko and Ulrich Lüning*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2016, 12, 2267–2273.
Otto-Diels-Institut für Organische Chemie,
Christian-Albrechts-Universität zu Kiel, Olshausenstr. 40, D-24098
Kiel, Germany
doi:10.3762/bjoc.12.219
Received: 15 June 2016
Accepted: 13 October 2016
Published: 27 October 2016
Email:
Ulrich Lüning* - luening@oc.uni-kiel.de
Associate Editor: P. J. Skabara
*
Corresponding author
©
2016 Klinkebiel et al.; licensee Beilstein-Institut.
Keywords:
License and terms: see end of document.
isoreticular; linker; MOF; Suzuki coupling; triazine
Abstract
New triazine-based tricarboxylic acid linkers were prepared as elongated relatives of triazinetribenzoic acid (TATB). Additionally,
functional groups (NO2, NH2, OMe, OH) were introduced for potential post-synthetic modification (PSM) of MOFs. Functionali-
zed tris(4-bromoaryl)triazine “cores” (3a,3b) were obtained by unsymmetric trimerization mixing one equivalent of an acid chlo-
ride (OMe or NO2 substituted) with two equivalents of an unsubstituted nitrile. Triple Suzuki coupling of the cores 3 with suitable
phenyl- and biphenylboronic acid derivatives provided elongated tricarboxylic acid linkers as carboxylic acids 17 and 20 or their
esters 16 and 19. Reduction of the nitro group and cleavage of the methoxy group gave the respective amino and hydroxy-substi-
tuted triazine linkers.
Introduction
A typical building block for many metal–organic frameworks In the case of the tritopic linker 1,3,5-benzenetricarboxylic acid
(
MOFs) [1-4] carries two functional groups such as carboxylic (BTC, forming HKUST-1 [6] in the presence of copper ions),
acids or heterocycles resulting in a bridging ligand for the an obvious elongated building block is BTB (1,3,5-benzenetri-
metal-containing „edges“ of the porous structures. Linkers with benzoate, Figure 1, left) forming for instance MOF-14 [7]. But
more than two ligating sites (carboxylic acids or heterocycles) this type of elongation has a drawback: the direct vicinity of the
have also been used successfully. Tri- to hexatopic linkers are aryl groups does not allow planarity due to the repulsion of the
also known. The prototype for a trivalent linker is 1,3,5- ortho-hydrogen atoms in the biphenyl subunits. An exchange of
benzenetricarboxylic acid. By elongating the „arms“ of the the central benzene ring by a 1,3,5-triazine (1,3,5-triazine-2,4,6-
linkers, larger pores can be achieved, and often the MOF struc- tribenzoate, TATB, Figure 1, right), however, permits planarity.
tures containing extended linkers resemble those obtained with The respective MOF from TATB and copper is PCN-6 [8].
the smaller parent linker (isoreticular structures, see for instance While the orientation of the carboxylate plane relative to the
[
5]).
central six-membered ring is in the same plane in the TATB
2267

Beilstein J. Org. Chem. 2016, 12, 2267–2273.
Figure 1: Steric repulsion between ortho-hydrogen atoms in benzene-1,3,5-tribenzoic acid (BTB) leads to a non-planar structure (left). The exchange
of CH units by nitrogen atoms in the central six-membered ring allows a planar structure: 1,3,5-triazine-2,4,6-tribenzoic acid (TATB, right). The two
structures in the lower half have been calculated with B3LYP//6-31G*.
based MOF PCN-6, the BTB analogue MOF-14 shows a tilt were possible. Structures which are isoreticular to PCN-6 could
angle of ca. 35°. B3LYP//6-31G* calculations of the isolated be synthesized with nitro and amino-substituted TATBs 1b–d
linkers in the gas phase result in dihedral angles of 38.5° for [10] (Figure 2).
BTB and 0.006° for TATB, respectively (for further informa-
tion see Supporting Information File 1). A detailed comparison
of triarylbenzene and triaryl-1,3,5-triazine based MOFs and a
discussion on the differences in sterical hindrance in benzene
and 1,3,5-triazine based structures has been undertaken for
tetrazol terminated linkers [9].
The properties of MOFs are determined by the nature of the
linker, the nature of the metal-containing secondary building
unit (SBU) and of course by the pore size. Additional proper-
ties are expected if other functional groups were present in the
Figure 2: Mono-substituted TATB linkers 1b–d were successfully em-
ployed in the isoreticular syntheses of PCN-6 MOFs [10].
pores. There are two general strategies to introduce additional
functional groups into a MOF: (i) by post-synthetic modifica-
tion, i.e., a reaction of an assembled MOF with some added
reagent, or (ii) by the use of modified linkers in the synthesis of How can this strategy be extended to even larger monofunction-
the MOF hoping that functionalized linkers lead to isoreticular alized triazine linkers? Obviously by the introduction of an ad-
structures.
ditional benzene ring into each arm. Since some decades, palla-
dium-catalyzed cross coupling is the method of choice when
It is obvious that the chance for isoreticular structures decreases aryl–aryl bonds should be constructed, for instance by applying
with the extend with which the linker is altered. Therefore, in the Suzuki–Miyaura coupling [11]. A retrosynthetic analysis
the case of tritopic linkers, it makes a difference whether one (Figure 3) of extended mono-functionalized triazinetricar-
additional functional group is introduced per linker or one per boxylic acids 2 calls for tris(4-bromoaryl)-1,3,5-triazines 3 and
„
arm“. Also the issue of planarity discussed above is related to boronic acids 4 with an additional carboxylic acid functionality.
the amount of additional substituents in the linker.
The respective methyl carboxylate of boronic acid 4 with n = 1,
i.e. 15 (see below, Figure 7), is commercially available. In case
Indeed, when TATB 1a was substituted with only one substitu- of the elongated methyl carboxylate of 4 with n = 2, the pinacol
ent per linker, syntheses of isoreticular analogues of MOFs boronate 18 (Figure 8) has been described in the literature [12].
2268

Beilstein J. Org. Chem. 2016, 12, 2267–2273.
Figure 3: Retrosynthetic analysis for extended TATBs 2: triple Suzuki coupling between tribromotriazines 3 and boronic acids 4.
Results and Discussion
The syntheses of larger linkers 2 therefore need: (i) substituted
Symmetric 1,3,5-triazines such as 3a are usually synthesized by 4-bromobenzoic acid chlorides 5, 4-bromobenzonitrile (6) and
trimerization of respective nitriles [13]. Unsymmetric 1,3,5-tri- the methoxycarbonyl-substituted phenylboronic acid 15 (n = 1)
azines 3 can be made by combining one equivalent of an acid or pinacol biphenylboronate 18 (n = 2).
chloride with two equivalents of a nitrile [14-16] (for a recently
described alternative access to unsymmetrical 1,3,5-triazines; Syntheses
see [17]). In the presence of a suitable Lewis acid such as anti- Four functional groups were chosen as additional substituents
mony(V) chloride, the acid chloride condenses with the nitriles in the extended triazine linkers 2: nitro, methoxy, amino and
to form an oxadiazinium salt from which the triazine can be ob- hydroxy groups. An amino group can be obtained from a nitro
tained by reaction with ammonia (Figure 4).
group by reduction, and a hydroxy group by cleavage of a me-
thoxy group. Therefore, it was sufficient to synthesize the nitro
The respective syntheses have successfully been carried out and methoxy-substituted acid chlorides 5 for the unsymmetric
with nitro and methoxy-substituted benzoyl chlorides 5b and trimerization of the desired tribromotriazines 3.
5
c. The resulting triazines 3b and 3c can then be coupled to
boronic acids or boronates to give substituted and elongated The synthesis of 4-bromo-3-nitrobenzoyl chloride (5b) was
triazine-based linkers. Furthermore, the reduction of the nitro straightforward. Commercially available 4-bromobenzoic acid
group or cleavage of the methoxy group will give additional (8) was nitrated according to a known procedure [18]. The reac-
substituted linkers, amino and hydroxy-substituted ones (see tion with thionyl chloride provided then acid chloride 5b
below).
(Figure 5).
Figure 4: Synthesis of unsymmetrically substituted 2,4,6-tris(bromoaryl)-1,3,5-triazines 3 from one equivalent of a substituted benzoyl chloride 5 and
two equivalents of benzonitrile 6. As intermediates, oxadiazinium salts 7 are formed. The reaction with ammonia yields the desired triazines 3 (overall
yields: 53% (3b), 61% (3c).
2269

Beilstein J. Org. Chem. 2016, 12, 2267–2273.
tion of the desired triazines 3b and 3c in 53 and 61% yield, re-
spectively.
Following the synthetic plan of Figure 3, the tribromides 3b and
3
c were used in the Suzuki couplings. In this and respective
following synthetic steps, optimization of the reaction condi-
tions was very important because each of the reactions had to be
performed three times with each triazine. Thus the total yield of
the final product depends on the yield of the single step to
the power of three. In a triple Suzuki coupling, 4-(methoxy-
carbonyl)phenylboronic acid (15) as the coupling partner
yielded elongated and monofunctionalized 1,3,5-triazines 16b
Figure 5: Synthesis of 4-bromo-3-nitrobenzoyl chloride (5b). Condi-
tions: a) HNO3/H2SO4, 3 h 0 °C, 2 h, room temperature, 98%,
b) SOCl2, 2 h reflux, not isolated.
In order to synthesize the respective methoxy compound 5c, and 16c carrying methyl ester groups at the end of each „arm“
commercially available 3-hydroxybenzoic acid (10) was used as (Figure 7). The yields of 90% (16b) and 73% (16c) correspond
starting material (Figure 6). Bromination introduced a bromine to 97% and 90% yield per coupling step, respectively.
atom into the 4 position (11) [19], then the phenol was turned
into its methyl ether. This methylation can be performed with By selective reduction, the nitro-substituted triester 16b could
the free carboxylic acid 11 [20] or with ester 12 [21,22]. In be transferred into its amino derivative. The reaction time and
order to generate the methoxy-substituted benzoyl chloride 5c, the hydrogen pressure had to be optimized. By heterogeneous
methoxy ester 13 was cleaved first [23] and the resulting acid catalytic hydrogenation at 5 bar with a Pd/C catalyst, aminotri-
1
4 was reacted with thionyl chloride. All steps were known azine 16d was obtained in 77% yield after 5 days. Hydrolyses
from the literature and have been carried out on a multigram of all three methyl esters 16b–d provided the tricarboxylic acids
scale.
17b–d in 96% to quantitative yields. Also the methoxy group
could be cleaved to yield a hydroxy-substituted elongated tri-
azine. In this case, four methoxy, three ester and one ether
group had to be cleaved leading to increasingly less soluble
products (carboxylic acids are less soluble than the respective
esters). In order to improve the yield, a methoxy cleaving proce-
dure which has been found to be effective in other demanding
demethylations [24] was used. In liquid pyridinium chloride
(
which melts at 144 °C), apparently all intermediates were suffi-
ciently soluble and consequently all four methoxy groups of 16c
could be cleaved in a one-pot reaction. The tetrafold cleavage
yielded hydroxytriazine 17e in 52% yield. All tricarboxylic
acids 17 are considerably less soluble in most solvents when
compared to the respective methyl esters 16. Nevertheless, they
are sufficiently soluble in, for instance, DMSO or in base to
allow proper analyses and future use in solvothermal syntheses
of MOFs.
Figure 6: Syntheses of 4-bromo-3-methoxybenzoyl chloride (5c).
Tribromotriazines 3 are not only good starting materials for the
syntheses of biphenyl-substituted triazine linkers 16 and 17.
Also longer terphenyl-substituted linkers can be obtained if the
same approach is chosen but a biphenylboronic acid derivative
such as boronate 18 [12] is used instead of the phenyl com-
pound 15.
Conditions: a) Br2, EtOH/HOAc, 30 min, room temperature, 52%,
b) MeOH/H2SO4, MeOH, 16 h, reflux, 94%, c) Me2SO4, K2CO3,
acetone, 3 h, reflux, quant., d) MeI, K2CO3, DMF, 16 h, room tempera-
ture, 83%, e) NaOH, H2O, MeOH, 6 h, room temperature, 66%,
f) SOCl2, DMF (cat.), 2 h, reflux, not isolated.
Cyclotrimerization of one equivalent of an acid chloride 5b
or 5c with two equivalents of 4-bromobenzonitrile (6) was As in the case of the biphenyl derivatives 16 and 17, the nitro
achieved using the Lewis acid antimony(V) chloride as depicted and the methoxy-substituted triazines 3b and 3c were used to
in Figure 4. The resulting colourful oxadiazinium salts 7 were obtain the terphenyl-based nitro and methoxy-substituted
mixed with aqueous ammonia solution resulting in the forma- triesters 19b and 19c (Figure 8). But also the non-substituted
2270

Beilstein J. Org. Chem. 2016, 12, 2267–2273.
Figure 7: Triple Suzuki–Miyaura coupling between tribromotriazines 3 and boronic acid 15 and subsequent hydrolyses of the esters 16. Conditions:
a) Pd(PPh3)4, K3PO4, dioxane, H2O, 48 h, reflux, 16b: 90%, 16c: 73%, b) Pd/C, H2, 5 bar, CH2Cl2, 5 d, room temperature, 16d: 77%. c) 17b: LiOH/
H2O/dioxane, 48 h, room temp, 96%; 17c: LiOH/H2O/THF, 24 h, 60 °C, quant.; 17d: H2O/THF, 24 h, 60 °C, quant., d) pyridinium hydrochloride, 12 h,
200 °C, 17e: 52%.
Figure 8: Triple Suzuki coupling between tribromotriazines 3 and boronate 18. Conditions: a) 19a: Pd(PPh3)4, K3PO4, dioxane/H2O, 2 d, 120 °C,
2
1
4%; 19b: Pd(dppf)Cl2, K3PO4, dioxane/H2O, 3 d, 120 °C, 16%; 19c: Pd(PPh3)4, K3PO4, dioxane/H2O, 2 d, 100 °C, 81%; b) LiOH/H2O/THF, 24 h,
20 °C, 20a: 99%; c) NaOH/H2O/dioxane, 24 h, 120 °C, 20b: 79%; d) pyridinium hydrochloride, 12 h, 200 °C, 20e: 62%.
2271

Beilstein J. Org. Chem. 2016, 12, 2267–2273.
6
.
Chui, S. S.-Y.; Lo, S. M.-F.; Charmant, J. P. H.; Orpen, A. G.;
Williams, I. D. Science 1999, 283, 1148–1150.
precursor 3a was coupled leading to the unsubstituted parent
compound 19a, a triester which has not been described yet.
Direct hydrolyses of triesters 19a and 19b gave the carboxylic
acids 20a and 20b.
doi:10.1126/science.283.5405.1148
7
8
9
1
.
.
.
Chen, B.; Eddaoudi, M.; Hyde, S. T.; O'Keeffe, M.; Yaghi, O. M.
Science 2001, 291, 1021–1023. doi:10.1126/science.1056598
Ma, S.; Sun, D.; Ambrogio, M.; Fillinger, J. A.; Parkin, S.; Zhou, H.-C.
J. Am. Chem. Soc. 2007, 129, 1858–1859. doi:10.1021/ja067435s
Dincă, M.; Dailly, A.; Tsay, C.; Long, J. R. Inorg. Chem. 2008, 47,
Finally, a hydroxy-functionalized terphenyl-based linker was
synthesized from the methoxy-substituted triester 19c. As suc-
cessfully applied for 16c, molten pyridinium chloride was used
as the cleaving reagent and all four methoxy groups of 19c
could be cleaved leading to the hydroxy-substituted triacid 20e
in 62% yield.
11–13. doi:10.1021/ic701917w
0.Mühlbauer, E.; Klinkebiel, A.; Beyer, O.; Auras, F.; Wuttke, S.;
Lüning, U.; Bein, T. Microporous Mesoporous Mater. 2015, 216, 51–55.
doi:10.1016/j.micromeso.2015.06.007
11.Miyaura, N.; Suzuki, A. J. Chem. Soc., Chem. Commun. 1979,
66–867. doi:10.1039/c39790000866
8
1
2.Barnard, R. A.; Dutta, A.; Schnobrich, J. K.; Morrison, C. N.; Ahn, S.;
Matzger, A. J. Chem. – Eur. J. 2015, 21, 5954–5961.
doi:10.1002/chem.201406332
All three triacids 20a, b and e precipitate from water with de-
creasing pH. In DMSO, they are soluble and consequently,
NMR analytics have been carried out in DMSO-d6.
1
3.Cook, A. H.; Jones, D. G. J. Chem. Soc. 1941, 278–282.
doi:10.1039/jr9410000278
Conclusion
14.Eitner, P.; Krafft, F. Ber. Dtsch. Chem. Ges. 1892, 25, 2263–2269.
doi:10.1002/cber.18920250218
In conclusion, 13 new elongated relatives of TATB 16, 17, 19
and 20 have been synthesized in batch sizes up to several
grams, of which 11 were mono-substituted. Syntheses, purifica-
tions and solubilities are satisfying in the case of the biphenyl
derivatives 16 and 17. The handling of the terphenyl deriva-
tives 19, and especially 20, becomes more difficult due to a
lower solubility. If the solublities were not sufficient for the
solvothermal syntheses of MOFs, “discrete” methylation [25]
could find a remedy.
1
5.Meerwein, H.; Laasch, P.; Mersch, R.; Spille, J. Chem. Ber. 1956, 89,
09–224. doi:10.1002/cber.19560890207
2
16.Schmidt, R. R. Chem. Ber. 1965, 98, 334–345.
doi:10.1002/cber.19650980203
17.Tiwari, A. R.; Akash, T.; Bhanage, B. M. Org. Biomol. Chem. 2015, 13,
0973–10976. doi:10.1039/C5OB01835H
1
1
8.Bhat, L.; Mohapatra, P. P.; Bhat, S. R. Compositions, Synthesis, and
Methods of Using Quinolinone Based Atypical Antipsychotic Agents.
U.S. Pat. Appl. US20080293736 A1, Nov 27, 2008.
1
9.Li, X.; Longenecker, K. L.; Pei, Z.; Sham, H. L.; Wiedeman, P. E.
Pyrrolidine-2-carbonitrile derivatives and their use as inhibitors of
dipeptidyl peptidase-iv (dpp-iv). WO Pat. Appl. WO2005023762 A1,
March 17, 2005.
Supporting Information
2
0.Burrows, A. D.; Frost, C. G.; Mahon, M. F.; Richardson, C.
Supporting Information File 1
Angew. Chem., Int. Ed. 2008, 47, 8482–8486.
Experimental details.
doi:10.1002/anie.200802908
[
http://www.beilstein-journals.org/bjoc/content/
Angew. Chem. 2008, 120, 8610–8614. doi:10.1002/ange.200802908
1.Govindachari, T. R.; Viswanathan, N.; Ravindranath, K. R.;
Anjaneyulu, B. Indian J. Chem. 1973, 11, 1081–1083.
2.Failli, A.; Quagliato, D.; Andrae, P.; Heffernan, G.; Coghlan, R.;
Shen, E. Pyrrolobenzodiazepine arylcarboxamides and derivatives
thereof as follicle-stimulating hormone receptor antagonists. U.S. Pat.
Appl. US20060199806 A1, Sept 7, 2006.
2
2
supplementary/1860-5397-12-219-S1.pdf]
Acknowledgements
This work was supported by the Deutsche Forschungsgemein-
schaft (Lu 378/24) as part of the priority program SPP 1362
23.Fukaya, Y.; Mihara, Y.; Morizono, D.; Ohtake, Y.; Oishi, T.; Shoji, T.;
Takashima, Y. 2,3,4,5-tetrahydro-1H-1,5-benzodiazepine derivative
and medicinal composition. WO Pat. Appl. WO2006051851 A1, May
(
porous metal–organic frameworks). We thank Dr. Torsten
Winkler for carrying out the calculations.
18, 2006.
2
4.Dawson, C. R.; Kurtz, A. P. J. Med. Chem. 1971, 14, 729–732.
doi:10.1021/jm00290a015
References
1.
2.
3.
Zhou, H.-C. J.; Kitagawa, S., Eds. Metal–Organic Frameworks (MOFs).
Chem. Soc. Rev. 2014, 43, 5415–6176. doi:10.1039/C4CS90059F
Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. Science
2
5.Köhl, I.; Lüning, U. Synthesis 2014, 46, 2376–2382.
doi:10.1055/s-0033-1339028
2013, 341, 1230444. doi:10.1126/science.1230444
Farusseng, D. Metal-Organic Frameworks; Wiley-VCH: Weinheim,
Germany, 2011. doi:10.1002/9783527635856
4
.
.
Férey, G. Chem. Soc. Rev. 2008, 37, 191–214. doi:10.1039/B618320B
Furukawa, H.; Go, Y. B.; Ko, N.; Park, Y. K.; Uribe-Romo, F. J.; Kim, J.;
O’Keeffe, M.; Yaghi, O. M. Inorg. Chem. 2011, 50, 9147–9152.
doi:10.1021/ic201376t
5
2272

Beilstein J. Org. Chem. 2016, 12, 2267–2273.
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.12.219
2273