Rational Design of Pore Size and Functionality in a Series of Isoreticular Zwitterionic Metal-Organic Frameworks

Article abstract of DOI:10.1021/acs.chemmater.8b03885

The isoreticular expansion and functionalization of charged-polarized porosity has been systematically explored by the rational design of 11 isostructural zwitterionic metal-organic frameworks (ZW-MOFs). This extended series of general structural composition {[M₃F(L1)₃(L2)_1.5]·guests}_n was prepared by employing the solvothermal reaction of Co and Ni tetrafluoroborates with a binary ligand system composed of zwitterionic pyridinium derivatives and traditional functionalized ditopic carboxylate auxiliary ligands (HL1·Cl = 1-(4-carboxyphenyl)-4,4′-bipyridinium chloride, Hcpb·Cl; or 1-(4-carboxyphenyl-3-hydroxyphenyl)-4,4′-bipyridinium chloride, Hchpb·Cl; and H₂L2 = benzene-1,4-dicarboxylic acid, H₂bdc; 2-aminobenzene-1,4-dicarboxylic acid, H₂abdc; 2,5-dihydroxy-1,4-benzenedicarboxylic acid, H₂dhbdc; biphenyl-4,4′-dicarboxylic acid, H₂bpdc; or stilbene-4,4′-dicarboxylic acid, H₂sdc). Single-crystal structure analyses revealed cubic crystal symmetry (I-43m, a = 31-36 ?) with a 3D pore system of significant void space (73-81%). The pore system features three types of pores being systematically tunable in size ranging from 17.4 to 18.8 ? (pore I), 8.2 to 12.8 ? (pore II), and 4.8 to 10.4 ? (pore III) by the choice of auxiliary ligands. All members of this series have noninterpenetrating structures and exhibit robust architectures, as evidenced by their permanent porosity and high thermal stability (up to 300 °C). The structural integrity and specific surface areas could be systematically optimized using supercritical CO₂ exchange methods for framework activation resulting in BET surface areas ranging from 1250 to 2250 m²/g. Most interestingly, as a structural landmark, we found the pore surfaces lined with charge gradients employed by the pyridinium ligands. This key feature results in significant adsorption of carbon dioxide and methane which is attributed to polarization effects. With this contribution we pioneer the reticulation of pyridinium building blocks into extended zwitterionic networks in which specific properties can be targeted.

Full text of DOI:10.1021/acs.chemmater.8b03885

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE
Article
Rational Design of Pore Size and Functionality in a Series
of Isoreticular Zwitterionic Metal-Organic Frameworks
Darpandeep Aulakh, Timur Islamoglu, Veronica F. Bagundes, Juby R. Varghese,
Kyle Duell, Monu Joy, Simon J. Teat, Omar K. Farha, and Mario Wriedt
Chem. Mater., Just Accepted Manuscript • DOI: 10.1021/acs.chemmater.8b03885 • Publication Date (Web): 25 Oct 2018
Downloaded from http://pubs.acs.org on October 26, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 35
Chemistry of Materials
1
1
2
3
4
5
6
Rational Design of Pore Size and Functionality in a Series of Isoreticular
Zwitterionic Metal-Organic Frameworks
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Darpandeep Aulakh,†,|| Timur Islamoglu,‡,|| Veronica F. Bagundes,† Juby R. Varghese,† Kyle
Duell,† Monu Joy,† Simon J. Teat,§ Omar K. Farha‡,ǂ and Mario Wriedt*,†
†Department of Chemistry & Biomolecular Science, Clarkson University, Potsdam, New York
13699, United States of America
‡Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois
60208, United States of America
§Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, California
94720, United States of America
ǂDepartment of Chemistry, Faculty of Science, King Abdulaziz University, Jeddah 21589, Saudi
Arabia
||D.A. and T.I. contributed equally to this work.
* Tel.: +1(315) 268-2355
Fax: +1(315) 268-6610
Email: mwriedt@clarkson.edu
ACS Paragon Plus Environment

Chemistry of Materials
Page 2 of 35
2
1
2
3
4
Abstract
5
6
7
8
9
The isoreticular expansion and functionalization of charged-polarized porosity has been
systematically explored by the rational design of eleven isostructural zwitterionic metal-organic
frameworks (ZW-MOFs). This extended series of general structural composition
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
{[M3F(L1)3(L2)1.5]·guests}n was prepared by employing the solvothermal reaction of Co and Ni
tetrafluoroborates with a binary ligand system composed of zwitterionic pyridinium derivatives
and traditional functionalized ditopic carboxylate auxiliary ligands (HL1·Cl = 1-(4-
carboxyphenyl)-4,4′-bipyridinium chloride, Hcpb·Cl; or 1-(4-carboxyphenyl-3-hydroxyphenyl)-
4,4′-bipyridinium chloride, Hchpb·Cl; and H2L2 = benzene-1,4-dicarboxylic acid, H2bdc; 2-
aminobenzene-1,4-dicarboxylic acid, H2abdc; 2,5-dihydroxy-1,4-benzenedicarboxylic acid,
H2dhbdc; biphenyl-4,4’-dicarboxylic acid, H2bpdc; or stilbene-4,4’-dicarboxylic acid, H2sdc).
Single-crystal structure analyses revealed cubic crystal symmetry (I-43m, a = 31-36 Å) with a
3D pore system of significant void space (73-81%). The pore system features three types of
pores being systematically tunable in size ranging from 17.4-18.8 Å (pore I), 8.2-12.8 Å (pore
II), and 4.8-10.4 Å (pore III) respectively by the choice of auxiliary ligands. All members of this
series have non-interpenetrating structures and exhibit robust architectures, as evidenced by their
permanent porosity and high thermal stability (up to 300°C). The structural integrity and specific
surface areas could be systematically optimized using supercritical CO2 exchange methods for
framework activation resulting in BET surface areas ranging from 1250 to 2250 m2/g. Most
interestingly, as structural landmark, we found the pore surfaces lined with charge gradients
employed by the pyridinium-ligands. This key feature results in significant adsorption of carbon
dioxide and methane which is attributed to polarization effects. With this contribution we pioneer
ACS Paragon Plus Environment

Page 3 of 35
Chemistry of Materials
3
1
2
3
4
5
6
the reticulation of pyridinium building blocks into extended zwitterionic networks in which
specific properties can be targeted.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Introduction
With a high degree of structural, chemical and functional diversity, porous metal-organic
frameworks (MOFs) have attracted considerable attention in recent years because of the potential
they present in a myriad of applications such as chemical separation,1-5 gas storage,6, 7 sensing,8
data storage9-11 and catalysis12, 13, with a focus on CO2/N2 separation in our laboratory.14-17 MOFs
are extended crystalline structures composed of metal clusters or ions connected by multitopic
organic ligands to build one-, two-, or three-dimensional infinite networks with fascinating
structures and properties such as high surface area and high pore volume, defined pore size
distribution, and good thermostability. In addition, the high degree of variability in both organic
and inorganic components allows for selective tailoring of framework structures, pore
environment and functionalities.18, 19
In contrast to current trends in the design of selective small molecule sorbents, we
emphasize small molecule-interactions with incorporated zwitterionic (ZW) moieties of the
framework. Zwitterions are classified as molecules bearing both permanent cationic and
permanent anionic functionalities which develop an electric field gradient on their molecular
surface due to the well separated intramolecular charges.20 Their incorporation in MOFs can thus
create charged organic surfaces (COSs) within the pore environment, leading to polarization
effects on guest molecules, which in turn can be utilized to separate small gases based on their
polarizability. For example, CO2 has a significant quadrupole moment and polarizability21 which
enables its polarization in an electric field, thus providing the potential to form additional
ACS Paragon Plus Environment

Chemistry of Materials
Page 4 of 35
4
1
2
3
4
5
6
7
8
9
electrostatic interactions within the framework resulting in enhanced adsorption enthalpies.14 In
this context, most of the work has been focused on new materials originating from 1,1’-
disubstituted-4,4’-bipyridinium-based ligands (viologens) exhibiting interesting photochromic
behavior.22-30 Recently, we developed an innovative design strategy to access ZW-MOFs from
anionic viologen derivates. In this process, additional counter ions are not required for charge
compensation of a charged-balanced neutral ZW-MOF.31 Also, we performed systematic
investigations on their synthesis and derivatisation.32 The methylene groups linking pyridinium
and carboxy-phenyl units, in these derivates, impart structural flexibility into the ligands which
results in dynamic and flexible frameworks that are known to exhibit interesting adsorption,
luminescent, magnetic and photochromic behaviors as a function of host–guest interactions
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
triggered by diverse physical and chemical stimuli.29, 30 However, since high framework stability
is imperative for many practical applications, the quest for metal-organic materials with rigid
frameworks has simultaneously been a subject of intense research. With this in mind, we initially
synthesized two isostructural microporous MOFs with general composition of
{[M3F(cpb)3(bdc)1.5]·guest}n with M = Ni (1-Ni) and Co (1-Co) by the rational design of the
rigid and neutral mono-substituted 4,4'-bipyridinium derivate, 1-(4-carboxyphenyl)-4,4′-
bipyridinium chloride (Hcpb·Cl) and its self-assembly with metal (II) tetrafluoroborates and the
auxiliary ligand 1,4-benzenedicarboxylic acid (H2bdc) in DMF/EtOH mixture under
solvothermal conditions. Analysis of the resulting framework topology implied potential pore
size design with distinct functionalities through isoreticulation by systematic variations of the
binary ligand system (HL1·Cl and H2L2). Auxiliary ligands (H2L2) of ditopic nature selected for
this isoreticulation study include 2-aminobenzene-1,4-dicarboxylic acid (H2abdc, 2); 2,5-
dihydroxybenzene-1,4-dicarboxylic acid (H2dhbdc, 3); biphenyl-4,4’-dicarboxylic acid (H2bpdc,
ACS Paragon Plus Environment

Page 5 of 35
Chemistry of Materials
5
1
2
3
4
5
6
7
8
9
4); and stilbene-4,4’-dicarboxylic acid (H2sdc, 5), which resulted in seven additional isostructural
materials of composition {[M3F(cpb)3(L2)1.5]·guest}n (2-5-M). Also, the ZW ligand was
exchanged with 1-(4-carboxyphenyl-3-hydroxyphenyl)-4,4′-bipyridinium chloride (Hchpb·Cl) in
order to extend the general feasibility of this isoreticulation approach, which resulted in two
representative examples of composition {[M3F(chpb)3(bdc)1.5]·guests}n with M = Ni (1'-Ni) and
Co (1'-Co).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Mono-substituted viologen derivates are not well explored and, to the best of our
knowledge, only two studies have been reported with the ZW ligand Hcpb·Cl.33, 34 While
Bezverkhyy et al. reported a series of MOFs which exhibited interesting eye-detectable thermo-
and photochromic behavior, such as regular discoloration of crystals from “edge” to “core” upon
O2 exposure,33 Kitagawa et al. showed that the ligand's pyridinium cationic surface participates
in the strong adsorption of methanol.34 Other explored mono-substituted viologen-based ligands
are shown in Scheme 1, however no study can be found reporting on the systematic
isoreticulation and functionalization of their respective MOFs.
III
IV
VI
I
II
V
Scheme 1. Structural formulas of reported mono-substituted viologen-based ligands. Ligands I,
II, III, IV are of neutral nature, whereas V and VI carry one negative charge.29
ACS Paragon Plus Environment

Chemistry of Materials
Page 6 of 35
6
1
2
3
4
5
6
7
8
9
With this fundamental study we attempt to close this gap by presenting on the synthesis,
structures and physical properties of an isoreticular series of eleven crystalline microporous
MOF materials (1-5-M and 1'-M). Isoreticulation enables the rational design of distinct isolated
cages coexisting with interconnected 3D pore space while showing pyridinium-based charged
pore linings as structural key features (Figure 1). These unique pore environments were
systematically explored for their fundamental structure-property relationships of small molecule
adsorption (N2, CO2, and CH4).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 7 of 35
Chemistry of Materials
7
1
2
3
4
5
6
7
8
9
Figure 1. Schematic representation of functionalization and isoreticulation in a series of
zwitterionic MOFs of composition [M3F(L1)3(L2)1.5]n. MOFs 1-5-M are shown with Lewis
structures of their respective auxiliary ligands (L2) and 1'-M with its respective zwitterionic
ligand (L1). Color code: Co/Ni, turquoise; F, green; O, red and C, grey.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental
General Information
Commercially available reagents were used as received without further purification. Viologen-
based ligands Hcpb·Cl and Hchpb·Cl were synthesized using a modified literature procedure33
(Scheme S1).
Synthesis of ligands 1-(4-carboxyphenyl)-4,4′-bipyridinium chloride (Hcpb·Cl) and 1-(4-
carboxyphenyl-3-hydroxyphenyl)-4,4′-bipyridinium chloride (Hchpb·Cl)
A solution of 4,4′-bipyridine (3.5 g, 22.5 mmol) and 2,4-dinitrochlorobenzene (3.0 g, 15 mmol)
were dissolved in 25 mL of acetone and refluxed for 13 h at 56 °C. After the mixture was cooled
to room temperature, the resulting precipitates were filtered off, washed several times with
dichloromethane and dried in vacuum to give 1-(2,4-dinitrophenyl)-4,4′-bipyridinium chloride as
a grey powder (3.57 g, yield 74%, Scheme S1). 1H-NMR (400 MHz, D2O): δ = 9.46 (d, 1H);
9.30 (d, 2H); 9.01 (d, 1H); 8.89 (d, 2H); 8.74 (d, 2H), 8.32 (d, 1H); 8.08 (d, 2H) (Figure S1). IR
(KBr pellet, cm-1): 휈̃ = 3505 (w), 3105 (s), 2987 (s), 1636 (s), 1613 (s), 1535 (s), 1482 (s), 1452
(m), 1410 (s), 1346 (m), 1278 (s), 1260 (m), 1239 (m), 1223 (m), 1075 (m), 1011 (s), 992 (m),
919 (s), 848 (s), 838 (w), 791 (s) (Figure S13).
ACS Paragon Plus Environment

Chemistry of Materials
Page 8 of 35
8
1
2
3
4
5
6
7
8
9
To a suspension of 1-(2,4-dinitrophenyl)-4,4′-bipyridinium chloride (0.65 g, 2.0 mmol)
and 2.5 mmol of triethylamine dissolved in 80% ethanol (50 mL), 4-aminobenzoic acid (0.41 g,
3.0 mmol) was added. The reaction mixture was stirred for 24 hours at 90 °C. The resulting
precipitates were filtered off and recrystallized from ethanol and dried in vacuum producing
Hcpb·Cl as beige powder (0.48 g, yield 89%, Scheme S1). 1H-NMR (400 MHz, D2O): δ = 9.51
(d, 2H); 8.90 (d, 2H); 8.78 (d, 2H); 8.12 (d, 2H); 8.09 (d, 2H); 7.80 (d, 2H) (Figure S2). IR data
(KBr, cm−1): 휈̃ = 3330 (w), 3166 (w), 2503 (w), 1909 (w), 1678 (s), 1636 (s), 1604 (s), 1573 (s),
1557 (s), 1536 (s), 1519 (s), 1482 (s), 1452 (m), 1411 (s), 1377 (m), 1315 (m), 1274 (s), 1224
(m), 1121 (m), 1054 (m), 1013 (s), 861 (s), 814 (w), 779 (s) (Figure S13).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Hchpb·Cl was prepared similarly to Hcpb·Cl, but instead of using 4-aminobenzoic acid,
4-aminosalicylic acid (0.46 g, 3.0 mmol) was used and recrystallized from ethanol as yellow
powder (0.37 g, yield 64%, Scheme S1). 1H-NMR (400 MHz, D2O): δ = 9.48 (d, 2H); 8.90 (d,
2H); 8.73 (d, 2H); 8.15 (d, 2H); 7.90 (d, 1H); 7.16 (s, 1H), 7.03 (d, 1H) (Figure S3). IR data
(KBr, cm−1): 휈̃ = 3411 (w), 3123 (w), 3069 (w), 1960 (w), 1656 (s), 1612 (s), 1497 (s), 1438 (s),
1416 (s), 1383 (s), 1282 (s), 1256 (s), 1212 (m), 1083 (s), 1062 (m), 917 (m), 864 (s), 822 (m)
(Figure S13).
Synthesis of MOFs {[M3F(cpb)3(bdc)1.5]·guests}n with M = Ni (1-Ni) and Co (1-Co)
Polycrystalline bulk material of 1-Ni was prepared by the reaction of Ni(BF4)2·6H2O (34.0 mg,
0.1 mmol), Hcpb·Cl (31.1 mg, 0.1 mmol) and H2bdc (8.3 mg, 0.05 mmol) in 2 mL of a 1:1
DMF/EtOH solvent mixture. This mixture was stirred in a closed snap vial at 85 °C for 4 days,
the resultant green precipitates were filtered, washed with EtOH, and Et2O, and dried in air (yield
31 mg, 55% based on Hcpb·Cl). Pink bulk material of 1-Co was prepared in a similar fashion as
ACS Paragon Plus Environment

Page 9 of 35
Chemistry of Materials
9
1
2
3
4
5
6
7
8
9
1-Ni using Co(BF4)2·6H2O (34.0 mg, 0.1 mmol) instead of Ni(BF4)2·6H2O (yield 28 mg, 52%
based on Hcpb·Cl). The purity of bulk materials was confirmed by powder X-ray diffraction
(PXRD) (Figure S15). Green cube-shaped crystals of 1-Ni for single-crystal X-ray diffraction
(SCXRD) studies were isolated from the reaction of a mixture containing Ni(BF4)2·6H2O (34.0
mg, 0.1 mmol), Hcpb·Cl (31.1 mg, 0.1 mmol) and H2bdc (8.3 mg, 0.05 mmol) in 2 mL of a 1:1
DMF/EtOH solvent mixture in a closed snap vial at 85 °C without stirring for 12 hours. Pink
single-crystals of 1-Co were prepared in a similar fashion as 1-Ni but using Co(BF4)2·6H2O
(34.0 mg, 0.1 mmol) instead of Ni(BF4)2·6H2O. Complementing NMR (Figure S4, S5), TGA
(Figure S20) and CHNF analyses resulted in compositions of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
{[Ni3F(cpb)3(bdc)1.5]·(H2bdc)·(BF4)·3(NH4)·4(OH)·DMF·9H2O}n for 1-Ni and
{[Co3F(cpb)3(bdc)1.5]·(H2bdc)·(BF4)·5(NH4)·6(OH)·DMF·3H2O}n for 1-Co. Elemental analysis
calculated for 1-Ni, Ni3BF5N10O30C74H89 (1880.43): C 47.27, H 4.77, N 7.45, F 5.05; found: C
47.84, H 4.43, N 7.60, F 5.35; for 1-Co, Co3BF5N12O26C74H87 (1843.15): C 48.22, H 4.76, N
9.12, F 5.15; found: C 48.67, H 4.59, N 9.00, F 5.43. IR (KBr pellet, cm-1) for 1-Ni: 휈̃ = 3424
(w), 3113 (w), 3062 (w), 2666 (w), 1654 (m), 1636 (s), 1536 (s), 1491 (s), 1451 (s), 1389 (w),
1257 (s), 1083 (s), 857 (s), 824 (m); for 1-Co: 휈̃ = 3425 (w), 2928 (w), 1658 (s), 1606 (s), 1535
(m), 1491 (s), 1390 (w), 1255 (s), 1096 (s), 1061 (s), 862 (m), 822 (m), 783 (m) (Figure S14).
Synthesis of MOFs {[M3F(cpb)3(abdc)1.5]·guests}n with M = Ni (2-Ni) and Co (2-Co)
Polycrystalline bulk materials and single crystals were prepared in a similar fashion as 1-Ni and
1-Co but using H2abdc (18.1 mg, 0.1 mmol) instead of H2bdc (yield for 2-Ni: 46 mg, 72%; for 2-
Co: 42 mg, 80%, both based on Hcpb·Cl). The purity of bulk materials was confirmed by PXRD
(Figure S16). Complementing NMR (Figure S6, S7), TGA (Figure S20) and CHNF analyses
ACS Paragon Plus Environment

Chemistry of Materials
Page 10 of 35
10
1
2
3
4
resulted in the composition of
5
6
7
8
{[Ni3F(cpb)3(abdc)1.5]·(H2abdc)·3.5(BF4)·1.5(NH4)·DMF·14H2O}n for 2-Ni and
{[Co3F(cpb)3(abdc)1.5]·(H2abdc)·(BF4)·2(NH4)·3(OH)·6H2O}n for 2-Co. Elemental analysis
calculated for 2-Ni, Ni3B3.5F15N11O31C74H92.5 (2130.97): C 41.71, H 4.38, N 7.23, F 13.37;
found: C 41.39, H 3.86, N 7.34, F 13.01; for 2-Co, Co3BF5N10.5O25C71H73.5
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(1756.49): C 48.55, H 4.22 N 8.37, F 5.41; found: C 48.46, H 4.20, N 8.96, F 6.04. IR (KBr
pellet, cm-1) for 2-Ni: 휈̃ = 3439 (w), 3064 (m), 2930 (m), 1656 (s), 1606 (s), 1537 (s), 1394 (w),
1257 (s), 1219 (w), 1062 (w), 860 (s), 783 (s), 702 (m); for 2-Co: 휈̃ = 3431 (w), 3061 (m), 1708
(m), 1636 (s), 1606 (s), 1536 (s), 1390 (w), 1219 (s), 1178 (w), 1058 (s), 820 (s), 784 (s), 701
(m) (Figure S14).
Synthesis of MOF {[Co3F(cpb)3(dhbdc)1.5]·guests}n (3-Co)
Single crystals of SCXRD quality were prepared in a similar fashion as 1-Co, but using H2dhbdc
(19.8 mg, 0.1 mmol) instead of H2bdc. Preparations of phase pure bulk material was not
successful; crystals were obtained in a heterogeneous mixture of unknown by-products which
could not be separated.
Synthesis of MOFs {[M3F(cpb)3(bpdc)1.5]·guests}n with M = Ni (4-Ni) and Co (4-Co)
Polycrystalline bulk materials and single crystals were prepared in a similar fashion as 1-Ni and
1-Co but using H2bpdc (24.2 mg, 0.1 mmol) instead of H2bdc (yield for 4-Ni: 43 mg, 80%; for 4-
Co: 41 mg, 78%, both based on Hcpb·Cl). The purity of bulk materials was confirmed by PXRD
(Figure S17). Complementing NMR (Figure S8, S9), TGA (Figure S20) and CHNF analyses
resulted in the composition of {[Ni3F(cpb)3(bpdc)1.5]·(BF4)·(NH4)·2(OH) ·DMF·11H2O}n for 4-
ACS Paragon Plus Environment

Page 11 of 35
Chemistry of Materials
11
1
2
3
4
5
6
7
8
9
Ni and {[Co3F(cpb)3(bpdc)1.5]·(BF4)·(NH4)·2(OH)·DMF·8H2O}n for 4-Co. Elemental analysis
calculated for 4-Ni, Ni3BF5N8O26C75H83 (1794.37): C 50.2, H 4.66, N 6.24, F 5.29; found: C
50.23, H 4.27, N 6.40, F 5.95; for 4-Co, Co3BF5N8O23C75H77 (1741.05): C 51.74, H 4.46, N
6.44, F 5.46; found: C 51.27, H 4.25, N 6.95, F 6.41. IR (KBr pellet, cm-1) for 4-Ni: 휈̃ = 3420
(w), 3301 (w), 3123 (w), 3123 (w), 2512 (w), 1942 (s), 1637 (s), 1543 (m), 1490 (s), 1414 (w),
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1401 (s), 1325 (s), 1218 (s), 1175 (s), 1083 (s), 956 (s), 862 (s), 783 (m); for 4-Co: 휈̃ = 3418 (w),
3304 (m), 3122 (w), 2931 (w), 1658 (s), 1589 (s), 1541 (m), 1490(s), 1390 (w), 1303 (s), 1256
(s), 1083 (s), 824 (m), 758 (m) (Figure S14).
Synthesis of MOFs {[M3F(cpb)3(sdc)1.5]·guests}n with M = Ni (5-Ni) and Co (5-Co)
Polycrystalline bulk materials of 5-M and single crystals of 5-Ni only were prepared in a similar
fashion as 1-Ni and 1-Co but using H2sdc (26.8 mg, 0.1 mmol) instead of H2bdc (yield for 5-Ni:
22 mg, 42%; for 5-Co: 19 mg, 35%, both based on Hcpb·Cl). Single crystal growth of SCXRD
quality for 5-Co was not successful. The purity of bulk materials was confirmed by PXRD
(Figure S18). Complementing NMR (Figure S10, S11), TGA (Figure S20) and CHNF analyses
resulted in the composition of {[Ni3F(cpb)3(sdc)1.5]·(BF4)·2(NH4)·3(OH)·DMF·8H2O}n for 5-Ni
and {[Co3F(cpb)3(sdc)1.5]·1.5(BF4)·1.5(NH4)·2(OH)·DMF·8H2O}n for 5-Co. Elemental analysis
calculated for 5-Ni, Ni3BF5N9.2O24C78H85 (1814.43): C 51.63, H 4.72, N 6.95, F 5.24; found: C
51.32, H 4.57, N 7.36, F 5.91; for 5-Co, Co3B1.5F7N8.5O23C78H82 (1832.53): C 51.12, H 4.51, N
6.5, F 7.26; found: C 51.52, H 4.22, N 6.56, F 6.95. IR (KBr pellet, cm-1) for 5-Ni: 휈̃ = 3439 (w),
3122 (m), 3067 (w), 2933 (w), 1657 (s), 1639 (s), 1605 (m), 1537 (s), 1490 (s), 1389 (w), 1256
(s), 1062 (s), 822 (s), 784 (m); for 5-Co: 휈̃ = 3427 (w), 3117 (m), 3063 (w), 2927 (w), 1659 (s),
ACS Paragon Plus Environment

Chemistry of Materials
Page 12 of 35
12
1
2
3
4
5
6
1605 (s), 1491 (m), 1389 (w), 1256 (s), 1083 (s), 1062 (m), 823 (s), 785 (s), 704 (m) (Figure
S14).
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of MOF {[Ni3F(chpb)3(bdc)1.5]·guests}n with M = Ni (1’-Ni) and Co (1’-Co)
Polycrystalline bulk material for 1’-Ni and single crystals for 1’-M were prepared in a similar
fashion as 1-Ni and 1-Co but using Hchpb·Cl (32.7 mg, 0.1 mmol) instead of Hcpb·Cl (yield 46
mg, 67% based on Hchpb·Cl). Preparation of phase pure bulk material of 1’-Co was not
successful; crystals were obtained in a heterogeneous mixture of unknown by-products which
could not be separated. The purity of bulk material for 1'-Ni was confirmed by PXRD (Figure
S19). Complementing NMR (Figure S12) and CHNF analyses resulted in the composition of
{[Ni3F(chpb)3(bdc)1.5]·(H2bdc)·3(BF4)·2(NH4)·(OH)·DMF·9H2O}n for 1'-Ni. Elemental analysis
calculated for Ni3B3F13N9O30C74H82 (2032.97): C 43.72, H 4.07, N 6.20, F 12.15; found: C
43.67, H 3.48, N 6.21, F 11.63. IR (KBr pellet, cm-1): 휈̃ = 3429(w), 3103 (w), 3072 (w), 2698
(w), 1684 (m), 1612 (s), 1542 (s), 1467 (s), 1451 (s), 1378 (w), 1243 (s), 1056 (s), 834 (s), 819
(m) (Figure S14).
Single-Crystal Structure Analysis
Single crystals were selected and mounted on MiTeGen® loops in Paratone oil. Crystal data of
compounds 1-Co, 3-Co, 5-Ni and 1'-Ni/Co were collected on a Bruker Kappa Apex II X-ray
diffractometer outfitted with a Mo X-ray source (sealed tube,  = 0.71073 Å) and an APEX II
CCD detector equipped with an Oxford Cryosystems Desktop Cooler low temperature device.
The APEX3 software suite was used for data collection, cell refinement, and reduction.35
Absorption corrections were applied using SADABS.36 Space group assignments were
ACS Paragon Plus Environment

Page 13 of 35
Chemistry of Materials
13
1
2
3
4
5
6
7
8
9
determined by examination of systematic absences, E-statistics, and successive refinement of the
structures. Structure solutions were performed with intrinsic phasing using SHELXT-2014, and
refined by least-squares refinement against |F|2 followed by difference Fourier synthesis using
SHELXL-2014.37-39 If not stated differently, non-hydrogen atoms were refined with anisotropic
displacement parameters. Aromatic C-H atoms were positioned with idealized geometry and
were refined with fixed isotropic displacement parameters [Ueq(H) = -1.2Ueq(C)] using a riding
model with dC-H = 0.95 Å. Crystal data of compounds 1-Ni, 2-Ni/Co, and 4-Ni/Co were collected
on a Bruker D8 diffractometer equipped with a PHOTON100 CMOS detector and Oxford
Cryosystems Cryostream 800 plus, on Beamline 11.3.1 of the Advanced Light Source at
Lawrence Berkeley National Lab, using silicon (111) monochromated synchrotron radiation (λ =
0.7749 Å). Dispersion corrections appropriate for this wavelength were calculated using the
Brennan method in XDISP40 within WinGX41. Structure solutions and refinements were
performed similar as described above.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CCDC 1863492-1863501 contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk. Selected crystal data and details on structure determinations are listed in
Table 1.
ACS Paragon Plus Environment

Chemistry of Materials
Page 14 of 35
14
1
2
3
4
Table 1. Selected crystal data and details on structure determinations from SCXRD analysis.
5
6
7
8
Compound
Formulaa
Compositionc
MW [g∙mol-1]
Crystal system
Space group
a [Å]
1-Ni
1-Co
2-Ni
2-Co
3-Co
b
Ni3BF5N10O30C74H89 Co3BF5N12O26C74H87 Ni3B3.5F15N11O31C74H92.5 Co3BF5N10.5O25C71H73.5 Co3FN6O15C60H12
[Ni3F(cpb)3(bdc)1.5]n [Co3F(cpb)3(bdc)1.5]n [Ni3F(cpb)3(abdc)1.5]n
1880.43
Cubic
I-43m
32.0068(10)
32789(3)
100(2)
8
0.617
0.475
[Co3F(cpb)3(abdc)1.5]n
1756.49
Cubic
[Co3F(cpb)3(dhbdc)1.5]n
1252.55
Cubic
9
1843.15
Cubic
I-43m
32.340(4)
33825(11)
170(2)
8
0.501
0.316
0.552/0.745
25.025
32396
5404
210
0.1786
0.0536
0.1137
0.951
2130.97
Cubic
I-43m
31.9734(9)
32686(3)
100(2)
8
0.634
0.478
0.863/0.967
28.934
161125
6012
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
I-43m
I-43m
32.2073(10)
33409(3)
100(2)
32.3227(11)
33769(3)
170(2)
V [Å3]
T [K]
Z
8
8
Dcalc [g∙cm-3]
0.620
0.416
0.822/0.936
28.567
159332
5929
188
0.0448
0.0906
0.2686
0.493
0.317
0.673/0.746
24.985
32882
5367
176
0.0996
0.0420
µ [mm-1]
Min/max transmission 0.857/0.981
Ɵmax [deg]
Measured reflections
Unique reflections
Parameter
28.563
154167
5822
180
188
Rint
0.0498
0.0834
0.2450
1.180
0.0430
0.0843
0.2516
1.203
R1 [F0 > 4σ(F0)]
wR2 [all data]
GOF
0.0906
0.978
0.231/-0.215
1.163
0.840/-0.650
Δρmax/Δρmin [e∙Å-3]
1.002/-0.655
0.246/-0.259
1.290/-0.627
ACS Paragon Plus Environment

Page 15 of 35
Chemistry of Materials
15
1
2
3
4
5
6
7
8
9
Compound
Formulaa
Compositionc
MW [g∙mol-1]
Crystal system
Space group
a [Å]
4-Ni
4-Co
5-Ni
1’-Ni
1’-Co
b
Ni3BF5N8O26C75H83
[Ni3F(cpb)3(bpdc)1.5]n [Co3F(cpb)3(bpdc)1.5]n [Ni3F(cpb)3(sdc)1.5]n
1794.37
Cubic
I-43m
34.9443(14)
42671(5)
100(2)
8
0.526
0.368
Co3BF5N8O23C75H77
Ni3BF5N9.2O24C78H85 Ni3B3F13N9O30C74H82 Ni3B3FN6O15C64.2H33.3
[Ni3F(chpb)3(bdc)1.5]n [Co3F(chpb)3(bdc)1.5]n
1814.43
Cubic
1741.05
Cubic
I-43m
35.0946(10)
43224(4)
100(2)
8
0.520
0.324
0.863/0.935
28.964
213469
7891
2032.97
Cubic
I-43m
32.141(2)
33204(7)
170(2)
8
0.515
0.363
0.662/0.745
24.990
29177
5289
186
0.0938
0.0508
0.1377
1.018
1324.46
Cubic
I-43m
32.3395(6)
33822(2)
170(2)
8
0.520
0.318
0.55/0.63
24.990
56497
5382
I-43m
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
36.3790(9)
48145(4)
180(2)
8
0.378
0.251
0.720/0.610
25.016
84980
7624
140
V [Å3]
T [K]
Z
Dcalc [g∙cm-3]
µ [mm-1]
Min/max transmission 0.750/0.914
Ɵmax [deg]
Measured reflections
Unique reflections
Parameter
31.082
226269
9428
223
223
207
Rint
0.0574
0.0766
0.2364
1.157
0.0446
0.0888
0.2654
1.208
0.1805
0.0727
0.2223
1.044
0.0891
0.0381
0.0995
1.036
R1 [F0 > 4σ(F0)]
wR2 [all data]
GOF
Δρmax/Δρmin [e∙Å-3]
1.329/-0.916
1.145/-0.527
0.440/-0.602
0.320/-0.295
0.239/-0.261
aFormula reflects framework composition and also molecules occupying the voids as obtained from elemental analysis.
bFormula reflects framework from SCXRD data only, no bulk composition available.
cComposition reflects the core framework only.
ACS Paragon Plus Environment

Chemistry of Materials
Page 16 of 35
16
1
2
3
4
5
6
Powder X-Ray Diffraction (PXRD)
PXRD data for compounds 1-M, 2-M and 4-M were recorded on the 17BM beamline at the
7
8
9
Advanced Photon Source, Argonne National Laboratory (Argonne, IL, USA). The incident X-ray
wavelength was 0.72768 Å. Data were collected using a Perkin-Elmer flat panel area detector
(XRD 1621 CN3-EHS) over the angular range 1-20° 2-Theta. PXRD data for 5-M and 1'-Ni was
collected on a Bruker D2 Phaser diffractometer equipped with a Cu sealed tube (λ = 1.54178 Å).
Powder samples were dispersed on low-background discs for analyses. Simulation of the PXRD
data was performed by using single crystal data and the powder pattern module of the mercury
CSD software package.42
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Thermogravimetric Analysis (TGA)
TG data was collected with a TGA Q50 from TA instruments. All measurements were performed
using platinum crucibles in a dynamic nitrogen atmosphere (50 mL/min) and a heating rate of 3
°C∙min-1. The instrument was corrected for buoyancy and current effects and was calibrated using
standard reference materials.
Adsorption Analysis
Supercritical CO2 drying was performed using a TousimisTM Samdri® PVT-30 critical point
dryer (Tousimis, Rockville, MD, USA). The as-synthesized samples were activated by soaking in
ethanol for 2 days, during which time ethanol was replenished three times. The supercritical CO2
was bled off at a rate of 0.5 mL/min until the chamber reached ambient pressure (typically
overnight). The chamber was opened, and the sample was quickly capped and transferred into an
Argon filled glovebox where it was transferred to a pre-weighed sample tube. The evacuated
analysis tubes containing the activated samples were then carefully transferred to an electronic
ACS Paragon Plus Environment

Page 17 of 35
Chemistry of Materials
17
1
2
3
4
5
6
7
8
balance and weighed to determine the sample mass. The tubes were then transferred to the
analysis port of the gas adsorption instrument. The samples were kept under vacuum at room
temperature for 2 hours prior to N2 adsorption isotherm collection. N2 isotherms were measured
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
by volumetric gas analyzer using a Micromeritics Tristar II 3020 (Micromeritics, Norcross, GA),
CO2 and CH4 isotherms were collected on ASAP2020. For all isotherms, warm and cold free
space correction measurements were performed using ultra-high purity He gas (UHP grade 5.0,
99.999% purity). All gases used are UHP grade (99.999% purity). N2 isotherms at 77 K were
measured in liquid nitrogen, isotherms at 273, 288 and 298 K were measured using temperature
circulating baths. All temperatures and fill levels were monitored periodically throughout the
measurement.
Elemental Analysis
Elemental data (C, H, N and F) was obtained from Atlantic Microlab, Inc.
Spectroscopy
FT-IR data were recorded on a Nicolet iS10 from Thermo Scientific. 1H-NMR data were
recorded on Avance DMX-400 from Bruker.
Results and Discussion
Crystal Structures and Synthetic Strategies
The preparation of materials was optimized into a consistent synthesis scheme: Ni or Co
tetrafluoroborates reacted under solvothermal conditions with a binary ligand mixture composed
of a pyridnium-carboxyl ZW ligand (HL1·Cl = Hcpb·Cl or Hchpb·Cl) and a traditional carboxyl
ACS Paragon Plus Environment

Chemistry of Materials
Page 18 of 35
18
1
2
3
4
auxiliary ligand (H2L2 = H2bdc, H2abdc, H2dhbdc, H2bpdc or H2sdc). Employing this scheme,
5
6
eleven isoreticular materials (1-2-Ni, 4-5-Ni, 1-5-Co, 1'-Ni and 1'-Co) could be crystallized.
7
8
9
SCXRD analysis revealed a core framework composition of [M3F(L1)3(L2)1.5]n (Figure S21-S30).
Their large void spaces and severe disorder of guest molecules resulted in very weak diffraction
using conventional laboratory X-ray sources. Hence, synchrotron radiation was used to obtain
high-resolution data sets for 1-Ni, 2-Ni/Co, and 4-Ni/Co. Compounds 1-Co, 3-Co, 5-Ni and 1'-
Ni/Co could be refined from conventional data. Identification of 5-Co was confirmed by PXRD
alone as no quality single crystals could be grown (Figure S18). The crystal structure of 1-Ni is
discussed below in detail as a representative example of the isoreticular series. 1-Ni crystallizes in
the non-centrosymmetric cubic space group I-43m with eight formula units per unit cell. The
asymmetric unit consists of one Ni2+ cation (S.O.F. 1/2, site m), one F- anion (S.O.F. 1/6, site
3m), and fractions of the ligands, 1/2 cpb and 1/4 bdc (Figure S21). The crystal structure exhibits
9-connected tricapped trigonal-prismatic metal nodes composed of corner-sharing octahedrally
coordinated Ni2+ cations (Figure 2A). The equatorial positions of these trimeric Ni2+-centred
octahedra are connected by six bidentate carboxyl groups of three L1 cpb ZW ligands and three
L2 bdc2- auxiliary ligands. The inner-facing octahedral axial positions share one central μ3-F-
anion, whereas the outer-facing axial positions feature bridging N-coordinated cpb ligands
(Figure 2A). This structural arrangement leads to three Ni-(μ3-F)-Ni angles of 120°. Interestingly,
only 17 µ3-F-bridged metal trimers of composition [M3F(O2CR)3X1.5] vs. 768 oxo-bridged trimers
are reported in the Cambridge Structure Database (CSD, November 2017), with the herein
presented structures being the first porous fluoro-bridged examples. The 9-connected trimeric
metal nodes result in a ncb-type network topology.43-47 This net features in 1-Ni three type of
pores (pores I, II and III) in a 3:1:4 ratio (Figure 2B-D). Pore I is mainly composed of L1 edges
resulting in a quasi-bisdisphenoidal cage of composition (Ni3F)8(L1)12(L2)2 fitting a sphere of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 19 of 35
Chemistry of Materials
19
1
2
3
4
5
6
7
8
9
18.8 Å in diameter (Figure 2B). Each pore is interconnected by four adjacent pores of same type
through large apertures (Figure 2E) resulting in a nbo-type three-dimensional channel system
(Figure 2H) which is responsible for the largest portion of the 72% void space within the
framework (Table 2). The channel topology has a pseudo-octagonal nature running along the
cube faces featuring cationic pyridnium sites pointing towards the inner core. Three L2 ligands of
each Ni trimer bridge adjacent trimers into a regular tetrahedral cage of composition (Ni3F)4(L2)6
fitting a sphere of 8.2 Å in diameter, denoted as pore II (Figure 3C), whereas pores I and II are
not interconnected. The size of this cavity is only determined by the length of L2. Pore III is a
trigonal-pyramidal cage of composition (Ni3F)4(L1)3(L2)3 fitting a sphere of 5.6 Å in diameter.
Each pore II connects with four pores III through small apertures to form bcu arrays of isolated
triakis tetrahedral cages (Figure 2G). Interestingly, changing the length of L2 can anisotropically
modulate the size and shape of pore III (Table 2). Finally, the isolated bcu cavities [(pore II)(pore
III)4] connect to the nbo channel through small apertures between pore III and pore I, furnishing a
hierarchical channel-cavity biporous system.45
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Chemistry of Materials
Page 20 of 35
20
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Structural features of isoreticular series: (A) 9-connected metal node of composition
M3F(L1)6(L2)3 with respective linear bridging ligands (L1 blue and L2 red). (B-D) Building
blocks of 1-Ni (L1 = cpb, L2 = bdc) as representative example of isoreticular series. (B) pseudo-
octagonal channel-type pore of composition (Ni3F)8(L1)12(L2)2 (pore I, yellow sphere of Ø 18.8
Å). (C) regular tetrahedral cage-type pore of composition (Ni3F)4(L2)6 (pore II, pink sphere of Ø
8.2 Å). (D) trigonal-pyramidal cage-type pore of composition (Ni3F)4(L1)3(L2)3 (pore III, green
sphere of Ø 5.6 Å. (E-H) Crystal packings of 1-Ni. (E) pores I are each interconnected by four
adjacent pores of same type, while pores II are isolated of pores I. (F, G) pores II are
interconnected by four pores III in a tetrahedral fashion while each being surrounded by six
interconnected pores I; (H) 3D arrangement of pores I and II, the interconnection of pores I
ACS Paragon Plus Environment

Page 21 of 35
Chemistry of Materials
21
1
2
3
4
5
6
results into 1D channel-type pores running along the cube faces. Note that pores III have been
omitted for the sake of clarity in packings E, F and H.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Modulation and precise control of MOF structures to yield topologies with high porosity
is crucial for their function. Although some ZW-MOFs have been previously reported, these
materials possess mostly highly dense low porosity structures, thus their effective modulation of
porosity still remains a significant design and synthetic challenge. Herein, we report the first
example of the modulation of porosity in a series of ZW-MOFs via a strategy of selective
elongation of metal-organic cages. Initial structural investigations of 1-Ni and 1-Co set the stage
to design an isoreticular series with various pore sizes, surface areas and functionalities by
replacing bdc2- with structurally related ditopic carboxylate linkers (Figure 3). Indeed, using each
of the linkers 2-aminobenzene-1,4-dicarboxylate (2), 2,5-dihydroxybenzene-1,4-dicarboxylate
(3), biphenyl-4,4'-dicarboxylate (4), stilbene-4,4'-dicarboxylate (5) instead of bdc2 yielded 2-
Ni/Co through 5-Ni/Co. Each member of this ZW-MOF series has been isolated and was
formulated subsequently by chemical microanalysis, TGA and SCXRD studies (Table 1). In
common, the crystal structures of 1-5-M exhibit a 3D cubic porous network differing in the
nature of functional groups decorating the pores and in the metrics of their pore structure (Figure
3 and S31). Thus, by varying the length and nature of the polyaromatic cores of the linkers L2 on
going from 1-M to 5-M, we have rationally extended the length of metal-organic cages and
effectively avoided framework interpenetration. The largest spheres that can fit within these
cages, taking into account the van der Waals radii of surface atoms, have diameters of 17.4-18.8
Å (pore I), 8.2-12.8 Å (pore II), and 4.8-10.4 Å (pore III). We also show that the isoreticular
series 1-5-M can be further expanded by changing the ZW ligand L1. Representatively, we
synthesized 1’-M under conditions identical for 1-M by replacing cpb with chpb.
ACS Paragon Plus Environment

Chemistry of Materials
Page 22 of 35
22
1
2
3
4
Pore I
Pore II
Pore III
1-M
2-M
3-M
4-M
5-M
5
6
7
8
18.8 Å
5.8 Å
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8.2 Å
5.6 Å
5.8 Å
18.4 Å
18.4 Å
18 Å
7.2 Å
7.2 Å
4.8 Å
4.8 Å
5.8 Å
10.1 Å
12.1 Å
11.8 Å
8.8 Å
17.4 Å
12.8 Å
10.4 Å
Figure 3. Schematic representation of crystal structures of isoreticular series 1-5-M along with
their respective pore structures, pore dimensions, and Lewis structures of modulating auxiliary
ligands. All structures shown are composed of the ZW ligand cpb (Fig. 2A).
ACS Paragon Plus Environment

Page 23 of 35
Chemistry of Materials
23
1
2
3
4
Adsorption Properties
5
6
7
8
While single crystal analysis confirmed the formation of low density (high porosity) ZW-MOFs,
maintaining the porosity upon solvent removal is very crucial for potential applications.
Conventional solvent exchange routes followed by thermal activation under vacuum for MOF
activation consistently resulted in framework collapse as indicated by PXRD and BET analysis.
Thus, supercritical CO2 exchange methods were successfully employed as mild alternative routes
for framework activation.48, 49 Subsequently, selected MOFs were characterized for their specific
surface areas and pore volumes as elucidated from N2 isotherms collected at 77 K (Figure 4 and
Table 2). 1-Ni revealed a permanent porosity after solvent removal and BET surface area as high
as 2400 m2ꢀg-1 and total pore volume of 0.929 cm3ꢀg-1. The isotherm up to a relative pressure
(p/p0) of 1.0 revealed a reversible type I behaviour with no significant hysteresis between the
sorption and desorption traces, attributed to the key features of its interconnected 3D channel
pores. The sorption reaches near saturation at low relative pressures (p/p0 < 0.05) and thereafter
increases very slowly up to 1 atm, indicating the microporous nature of the framework. Similar
type I behavior with a high N2 adsorption plateau and the absence of hysteresis indicative of
microporosity was exhibited by 1’-Ni and 2-Co. The steep increase in the N2 isotherm after p/p0
= 0.8 for 1-Co and 4-Ni can be attributed to the condensation of N2 within the intercrystallite void
spaces. It was found that the N2 adsorption capacity decreases in the order 5-Ni < 4-Ni < 1-Ni
(Figures 4A and S32). This is in full agreement with the decreasing length of the pillar linkers
from sdc2- to bpdc2- to bdc2-. Surprisingly, although 1’-Ni and 1-Co have similar pore sizes as 1-
Ni, their N2 uptake under the conditions studied was very low compared to 1-Ni. For 2-Ni and 5-
Ni, even the use of supercritical CO2 drying did not lead to decent surface areas, possibly due to
framework instability in EtOH (Figure S32 and Table S1).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Chemistry of Materials
Page 24 of 35
24
1
2
3
4
The incremental pore size distribution (PSD) curves obtained from N2 adsorption
5
6
7
8
isotherms by using the density functional theory (DFT) method for cylindrical pores is shown in
Figure 4B. Importantly, two different pore sizes were observed for 1-Ni, 1’-Ni and 4-Ni where
the larger pores can be attributed to the interconnected channels (pore I) and the smaller ones to
the cage like pores II and III between the nodes. The pore sizes of the frameworks 1-Co and 2-
Co, as calculated from N2 sorption studies, are found to be ~13.8 Å, which can be attributed to
the incomplete activation of the frameworks resulting from the presence of ZW ligands in the
pores as confirmed by NMR measurements (Figures S5 and S7).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Additionally, the short-term air stability was also assessed by exposing 1-Ni to air for 3
hours followed by activation at 120 °C under dynamic vacuum, confirming the retention of
porosity (Figure S33).
ACS Paragon Plus Environment

Page 25 of 35
Chemistry of Materials
25
1
2
3
4
A
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
B
Figure 4. N2 adsorption isotherms at 77 K (A) and corresponding pore size distribution plots (B)
for MOFs 1-Co, 1-Ni, 1’-Ni, 2-Co and 4-Ni.
Table 2. Selected porosity parameters of MOFs 1-Co, 1-Ni, 1’-Ni, 2-Co and 4-Ni.
MOF
1-Co
1-Ni
Surface area (m2 g-1) N2 uptake (cm3 g-1)* PSD (Å)
Pore volume (cm3 g-1)
1250
2400
1500
1600
2250
407
659
384
478
705
13.7
0.462
0.929
0.494
0.589
0.816
12.6/14.8
12.6/14.8
13.8
1’-Ni
2-Co
4-Ni
12.7/16.8
*at 0.85 p/p0 and 77 K (values 0.85-1.0 p/p0 are in part affected by N2 condensation)
ACS Paragon Plus Environment

Chemistry of Materials
Page 26 of 35
26
1
2
3
4
5
6
7
8
9
The microporous nature and presence of charged pore linings in the studied frameworks
prompted us to explore their CO2 and CH4 adsorption properties at non-cryogenic temperatures,
and to calculate the corresponding binding affinities. The sorption isotherms of CO2 and CH4 at
273 K for the activated frameworks probed up to a relative pressure (p/p0) of 1.0 at STP are
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
reported in Figure 5A, B. We were pleased to discover that the maximal CO2 uptake was reported
for 4-Ni (~3 mmol g-1 at 273 K and 1 bar) which also exhibits the highest surface area among the
other MOFs studied here and double to [Cu(ipq)](BF4) (where H2ipq+ = 1-(3,5-dicarboxyphenyl)-
4-(pyridin-4-yl)pyridinium reported by Shimizu et al.50 The CO2 and CH4 adsorption isotherms at
273, 288 and 298 K were employed to calculate the respective Qst values through the Clausius-
Clapeyron equation, representing the relative magnitude of the adsorbate-adsorbent and
adsorbate-adsorbate interactions (Figure 5C, D and S35-S37). It is worth noting that CO2
isotherms do not show steep uptake at low pressures which confirms that the CO2-framework
interactions are solely physisorption. Additionally, this also confirms that metals are
coordinatively saturated and there are no defects (e.g. missing linkers) around the node which
would result in open Ni2+/Co2+ adsorption sites and therefore strong adsorption of CO2.51 Two
successive CO2 isotherms on 2-Co (which presents primary amine moiety) at 298 K were also
collected without activation after the first measurement and the reasonable overlap between two
measurements confirms the regeneration of the material without applying heat (Figure S34).
Physisorptive type interactions with CO2 were further confirmed by the binding affinity
calculations where zero-coverage binding affinities are calculated to be around 24-32 kJ mol-1.
This is desirable as it translates into less energy input for sorbent regeneration. Logically, binding
affinity of CO2 decreases with loading as the higher affinity sites are saturated. The highest CH4
uptake was observed with 4-Ni, among others, due to its high surface area. Unlike CO2, binding
ACS Paragon Plus Environment

Page 27 of 35
Chemistry of Materials
27
1
2
3
4
5
6
affinities for CH4 did not decrease with increased loading suggesting that CH4 shows a similar
affinity to all sites within the framework.
7
8
9
A
B
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
D
C
Figure 5. (A) CO2 and (B) CH4 adsorption isotherms at 273 K for 1-Co, 1’-Ni, 2-Co and 4-Ni;
and the corresponding isosteric heats of (C) CO2 and (D) CH4 adsorption.
Conclusions
The rational design and synthesis of a series of eleven porous zwitterionic MOFs was explored
via a reticular synthesis approach by employing a variety of ditopic carboxylate linkers with
diverse functional groups. The novel and intrinsic value of this design approach lies in the ability
ACS Paragon Plus Environment

Chemistry of Materials
Page 28 of 35
28
1
2
3
4
5
6
7
8
9
to control and direct the reticulation of pyridinium building blocks into extended networks in
which specific properties can be targeted. Single-crystal structure evaluations revealed that their
unique three-dimensional pore systems could be systematically tuned by varying the linker
groups from 2-aminobenzene-1,4-dicarboxylate (abdc2-) and 2,5-dihydroxybenzene-1,4-
dicarboxylate (dhbdc2-) to biphenyl-4,4’-dicarboxylate (bpdc2-) and stilbene-4,4’-dicarboxylate
(sdc2-). In addition, supercritical CO2 activation methods were employed to optimize specific
surface areas and adsorption capacities. Given the unique structural attributes of the herein
studied isoreticular MOF series, including their thermal stability and adsorption properties, it is
expected that this work will contribute to inspire the development of new structural design
strategies for the discovery of specifically underexplored zwitterionic MOFs.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Acknowledgment
M.W. gratefully acknowledges Clarkson University for generous start-up funding, the Donors of
the American Chemical Society Petroleum Research Fund (Award No. 56295-DNI10), and the
National Science Foundation CAREER Program (Award No. 1752771) for support of this
research. Work carried out at the Advanced Light Source (ALS) is supported by the Director,
Office of Science, Office of Basic Energy Science of the U.S. Department of Energy, under
contract DE-AC02- 05CH11231. Work performed at the Advanced Photon Source at Argonne
National Laboratory was supported by the U.S. Department of Energy, Office of Science, Office
of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357.
Supporting information available
ACS Paragon Plus Environment

Page 29 of 35
Chemistry of Materials
29
1
2
3
4
5
6
7
8
Details of ligand synthesis and characterization, PXRD patterns, TGA curves, NMR spectra, IR
spectra, adsorption isotherms with Langmuir fits, crystal structure packings and CIF files. This
material is available free of charge via the Internet at http://pubs.acs.org
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
References
(1)
Li, J.-R.; Sculley, J.; Zhou, H.-C., Metal–Organic Frameworks for Separations. Chem.
Rev. 2011, 112, 869-932.
(2)
Bobbitt, N. S.; Mendonca, M. L.; Howarth, A. J.; Islamoglu, T.; Hupp, J. T.; Farha, O. K.;
Snurr, R. Q., Metal–Organic Frameworks for the Removal of Toxic Industrial Chemicals
and Chemical Warfare Agents. Chem. Soc. Rev. 2017, 46, 3357-3385.
Denny Jr, M. S.; Moreton, J. C.; Benz, L.; Cohen, S. M., Metal–Organic Frameworks for
Membrane-Based Separations. Nat. Rev. Mater. 2016, 1, 16078-16094.
Bae, Y.-S.; Snurr, R. Q., Development and Evaluation of Porous Materials for Carbon
Dioxide Separation and Capture. Angew. Chem., Int. Ed. 2011, 50, 11586-11596.
Yang, R. T., Adsorbents: Fundamentals and Applications. ed.; John Wiley & Sons Inc.,
Hoboken, 2003.
(3)
(4)
(5)
(6)
(7)
(8)
He, Y.; Zhou, W.; Qian, G.; Chen, B., Methane Storage in Metal-Organic Frameworks.
Chem. Soc. Rev. 2014, 43, 5657-78.
Suh, M. P.; Park, H. J.; Prasad, T. K.; Lim, D.-W., Hydrogen Storage in Metal–Organic
Frameworks. Chem. Rev. 2011, 112, 782-835.
Hu, Z.; Deibert, B. J.; Li, J., Luminescent Metal-Organic Frameworks for Chemical
Sensing and Explosive Detection. Chem. Soc. Rev. 2014, 43, 5815-5840.
ACS Paragon Plus Environment