The dynamic assembly of covalent organic polygons: Finding the optimal balance of solubility, functionality, and stability

Article abstract of DOI:10.1002/ejoc.201500170

Six variably functionalized phenanthrene-based bis(catechol) derivatives have been synthesized and their ability to undergo dynamic covalent assembly with 1,4-benzenediboronic acid (BDBA) to give discrete, shape-persistent [2+2] assemblies has been invest

Full text of DOI:10.1002/ejoc.201500170

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
FULL PAPER
DOI: 10.1002/ejoc.201500170
The Dynamic Assembly of Covalent Organic Polygons: Finding the Optimal
Balance of Solubility, Functionality, and Stability
Merry K. Smith,[a] Alexander R. Goldberg,[a] and Brian H. Northrop*[a]
Keywords: Self-assembly / Macrocycles / Borates
Six variably functionalized phenanthrene-based bis(cate-
chol) derivatives have been synthesized and their ability to
sulting assemblies in the presence of protic solvents. A model
is proposed to describe how solvent choice and starting ma-
undergo dynamic covalent assembly with 1,4-benzenedi- terial functionality must be carefully balanced in order to
boronic acid (BDBA) to give discrete, shape-persistent [2+2]
assemblies has been investigated. Systematically varying the
functionality of the bis(catechol) derivatives is found to influ-
ence both the reaction conditions necessary to promote their
shift many competing equilibria toward the formation of dis-
crete, soluble covalent organic polygons. The results are ex-
pected to provide additional insight into optimizing the syn-
thesis of other boronate ester polygons/polyhedra as well as
self-assembly with BDBA as well as the stability of the re- related, “infinite” covalent organic frameworks (COFs).
acids and oligo-catechols have been used to produce the
Introduction
majority of COFs to date.[4a–4c,13]
The thermodynamically driven dynamic assembly[1] of
Despite the increasing interest in the synthesis and char-
extended framework materials has been the focus of in- acterization of COFs derived from boronic acids, discrete
creasing interest in the materials science community over assemblies formed from boronic acids are less common. Ex-
the past two decades with the emergence of metal organic amples of porous covalent organic polygons or polyhedra
frameworks[2] (MOFs) in the mid-1990s[3] and the discovery (COPs) prepared from boronic acid and oligo-catechol
of covalent organic frameworks[4] (COFs) in 2005.[5] monomers include porous boronate-ester cages, macro-
Formed through the reversible covalent linkage of organic cycles, capsules, and cavitands, and boroxine cyclo-
subunits, COFs are particularly attractive synthetic targets, phanes.[14–19] Related examples of discrete boronic acid
in large part because of their many advantageous physical based assemblies have also been reported as complexes with
properties. These lightweight, highly ordered, porous poly- sp2 hybridized nitrogen compounds.[12,20] Generally absent
mers exhibit high thermal stabilities, low densities, tunable from this array of macromolecules are discrete assemblies
pore sizes and high internal surface areas,[4a–4c,5,6] leading that are structurally analogous to COFs and prepared from
to potential applications in optoelectronics,[7] electron monomers that are similar or identical to those used to pre-
transport,[8] heterogeneous catalysis,[9] and gas sequestra- pare extended frameworks. If soluble, these “COF ana-
tion and storage.[10] According to the principles of reticular logue” COPs can aid in elucidating the dynamic and kinetic
chemistry[11] coupled with the selection of appropriately re- details of COF formation and may provide insight regard-
versible dynamic covalent reactions, specific combinations ing the optimization of solvent systems and reaction condi-
of organic subunits can allow for the rational design of tions used in COF synthesis. Furthermore, discrete COPs
COFs with predictable framework geometries, pore sizes, are porous, shape-persistent[21] structures with potential ap-
and chemical properties. Boronic acids have proven espe- plications that may extend beyond those of COFs, as they
cially useful in this regard as they are able to reversibly self- are likely to form aggregates and higher-order assemblies in
condense into boroxine anhydrides and can also undergo solution and on solid substrates. With their potential to
dynamic assembly with diols, such as catechol derivatives, form higher-order assemblies in solution, COPs may find
to produce boronate esters.[12] Both systems are well suited applications as liquid crystalline materials,[22,23] in self-as-
for the modular assembly of periodic 2- and 3-dimentional sembled monolayer formation,[24] and in solution-pro-
COFs; indeed, structurally diverse polyfunctional boronic cessable organic photovoltaics.[7c,7d] Additionally, soluble
boronate ester assemblies offer opportunities to tune their
solid-state suprastructures[25] through coordination[20] with
Lewis basic groups such as bipyridyls.
In spite of their potential, there remain very few exam-
ples of discrete, shape-persistent assemblies derived from
boronic acids, likely because their synthesis and study is
[a] Department of Chemistry, Wesleyan University,
52 Lawn Avenue, Middletown, CT 06459, USA
E-mail: bnorthrop@wesleyan.edu
http://www.bnorthrop.faculty.wesleyan.edu
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201500170.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
M. K. Smith, A. R. Goldberg, B. H. Northrop
FULL PAPER
complicated by the same difficulties that have made the both the calixarene and the bis(catechol) starting materials,
characterization and study of COF materials quite chal- tetra-functionalization of the calixarene with alkyl groups
lenging. In particular, the inherent insolubility of boronic (heptyl or phenylethyl) was necessary to promote assembly
acid derived COFs, most COPs, and their boronic acid sec- formation and to ensure the solubility of the resulting cages.
ondary building units (SBUs) has presented several chal- In the related realm of boroxine anhydrides, the triply ferro-
lenges to their preparation and characterization. In the cene-bridged boroxine cyclophane reported by Miljanic´ and
roughly twenty years since the first reports of MOFs, over co-workers[19] could be obtained upon crystallization out of
6,000 distinct MOF structures[2c] have been synthesized and a solution of mesitylene and dioxane (1:1), but was other-
characterized. In a dramatic contrast, over the nine years wise poorly soluble. In a previous work, we reported[29] the
since the first report of a COF fewer than 50 distinct COFs synthesis and characterization of soluble covalent organic
have been reported in the literature.[4b] Similarly, in contrast boronate ester rectangles composed of rigid triphenyl bis-
to the many examples of single crystalline MOFs, no single (catechol) derivatives and benzenediboronic acid. Hexyloxy
crystal of a boronic acid derived COF has been reported. functionalization was found to be necessary for solution
To date, and to the best of our knowledge, only four exam- phase self-assembly and characterization. Very recently,
ples of single crystalline COFs have been reported in the Mastalerz and co-workers synthesized[18a] an impressive
literature.[26,27] The insolublity of COFs has also been a sig- mesoporous cuboctahedral boronate ester cage by refluxing
nificant barrier to detailed mechanistic and kinetic analysis a tetrahydroxy triptycene derivative and 1,3,5-benzene
of COF formation, however Dichtel and co-workers have triboronic acid in chloroform. The introduction of solubi-
recently demonstrated that turbidity analysis can be used to lizing ethyl-groups on the triptycene monomer was found
quantify the kinetics of early stage COF formation from to be essential for cage assembly and analysis. The solubility
within homogeneous reaction solutions,[28] a considerable of the boronate ester cuboctahedron was found to be so
advance in the study of COFs and related materials. Still, precarious that, upon desolvation, the resulting solid could
the insolubility of such frameworks remains a persistent not be redissolved to any extent. Collectively, these results
challenge to their characterization, investigation, and syn- highlight the sensitivity of dynamic covalent assembly to
thetic optimization.
Those groups that have reported soluble, discrete assem- precursors and the desired assemblies themselves.
blies based on boronic acid (Scheme 1) have been forced to The preparation of assemblies derived from boronic acid
the solubility of both the secondary building unit (SBU)
confront issues related to poor solubility. Kobayashi and is further complicated by their often-observed sensitivity to
co-workers reported[16] soluble boronate ester cavitands the choice of reaction solvent or solvents. The choice of an
based on tetra-boronic acid calixarenes and 1,2-bis(3,4-di- appropriate solvent system that promotes the high-yielding
hydroxyphenyl)ethane. Despite the inherent flexibility of formation of desired assemblies from within complex, dy-
Scheme 1. Representative examples of discrete, shape-persistent assemblies that incorporate boronic acid secondary building units (SBUs):
(a) calixerene-based assemblies investigated by Kobayashi,[16] (b) a ferrocene-bridged boroxine cyclophane reported by Miljanic´,[19] (c)
boronate ester rectangles,[29] and (d) triptycene-based boronate ester cuboctahedra reported by Mastalerz.[18a]
2
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
Dynamic Assembly of Covalent Organic Polygons
namically exchanging mixtures of oligomers and secondary Lackinger[24d,24f] have observed similar phenomena during
building units often requires combinatorial screening of the formation of COFs on solid substrates. It is believed
many different solvent mixtures in varying ratios. There is that these small quantities of protic solvents likely catalyze
considerable literature precedent[15a,16a,16d,17b,17c,18,29] for the dynamic exchange of boronic acid and catechol deriva-
the dynamic assembly of boronate ester materials in hydro- tives. Iwasawa has even observed[17c] that the molar ratio
phobic solvents such as dichloromethane, chloroform, tolu- and concentration of mixtures involving benzene, chloro-
ene, and benzene. Boronate ester COFs, by contrast, are form, and methanol can select for different diastereomeric
typically synthesized under solvothermal conditions in mix- boronate ester macrocycles from within complex mixtures.
tures of mesitylene and dioxane.[4–7] Some examples of dis- Large equivalents of methanol or water, however, have been
crete assemblies derived from boronic acid have also been shown by Kobayashi,[16a] Lavigne,[30] and us[29] to promote
prepared in mesitylene:dioxane (1:1), such as the ferrocene- the disassembly of boronate ester assemblies and frame-
based boroxine cyclophane prepared by Miljanic´ and co- works. To complicate matters further, the ferrocene-based
workers[19] (Scheme 1, b). With respect to the other discrete, boroxine anhydride assembly shown in Scheme 1 (b) can
shape-persistent boronate ester assemblies presented in be soaked in pure ethanol without any disruption of the
Scheme 1, the mesoporous cubeoctahedron prepared by assembly,[19] however the assembly hydrolyzes within min-
Mastalerz and co-workers (Scheme 1, d) was assembled in utes in the presence of water.
chloroform.[18a] Chloroform has also been used by Kobaya-
From these examples alone it is clear that the successful
shi and co-workers[16] (Scheme 1, a) to prepare boronate es- formation of assemblies based on boronic acid, whether in-
ter cavitands. Protic solvents such as methanol, ethanol, finite (COFs) or discrete (COPs), is significantly influenced
and water have played varying roles in the synthesis of by both SBU structure and the choice of solvent(s). A care-
boronate ester and boroxine anhydride assemblies. The for- ful balance must be struck between providing (i) sufficient
mation of boronate ester rectangles (Scheme 1, c), for exam- solubility of the starting material SBUs to initiate assembly,
ple, could not be achieved in pure chloroform,[29] however a (ii) adequate reversibility of initial oligomers and kinetic
mixture of chloroform and methanol (10:1) gave the desired intermediates to allow for error correction, and (iii) a robust
rectangles in high yields. Dichtel[28] and Lavigne[13a,13d] have tolerance of thermodynamically stable product assemblies
demonstrated that the addition of small amounts of meth- within the solvent mixture such that they do not revert back
anol or water can increase the rate of boronate ester as- to starting materials. Ultimately, the structures and solubili-
sembly and COF formation in solution, while Wan[24e] and ties of both starting materials and products in a given sol-
Figure 1. Chemical structures of phenanthrene-based bis(catechol) derivatives 1–6 functionalized with hexyloxy (hexyl), diethylene glycol
monomethyl ether (DEG-CH3), or 1,3,5-tris(hexyloxy) benzyl (THB) groups to vary their solution phase characteristics. Biaryl-linked
bis(catechol) derivatives are shown in (a) while π-extended, acetylene-linked bis(catechol) derivatives are shown in (b). The bis(catechol)s
are designed to self-assembly with 1,4-benzenediboronic acid (BDBA) to give shape-persistent boronate ester ovals 7–12.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
M. K. Smith, A. R. Goldberg, B. H. Northrop
FULL PAPER
vent mixture will determine both what species are present (350–460 nm), and the increased conjugation of phenan-
at equilibrium, how stable they are, and what analytical threne may enable fluorescence to be shifted further toward
methods can be used to investigate them.
longer wavelengths. Furthermore, multiple routes are avail-
Herein we expand upon earlier work in the area of solu- able for the addition of solubilizing groups at positions 9
ble boronate ester rectangles to explore, in greater depth, and 10 of phenanthrene as well as subsequent functionali-
the roles that starting material and product solubilities play zation of positions 3 and 6 with catechol derivatives
in the synthesis and analysis of discrete boronate ester (Scheme 2). Three different functionalities were added at
assemblies. In the current study, we report the synthesis of positions 9 and 10 in an effort to vary the solubility of both
six variably functionalized, phenanthrene-based bis(cate- the starting bis(catechol) phenanthrene derivatives and their
chol) derivatives and evidence for their dynamic self-as- resulting assemblies: hexyloxy substituents (compounds 1
sembly with benzene-1,4-diboronic acid (BDBA) to form and 2); diethylene glycol monomethyl ether substituents
shape-persistent boronate ester ovals (Figure 1). Insight (DEG-CH3, compounds 3 and 4); and 3,4,5-tri(hexyloxy)-
gained from examining the influence of different solubiliz- benzyl substituents (THB, compounds 5 and 6). This selec-
ing groups attached to the phenanthrene bis(catechol)s will tion of substituents was chosen to compare the relatively
contribute more generally to our overall understanding of hydrophobic nature of hexyloxy substituents vs. the more
assembly formation and help pave the way toward a greater hydrophilic nature of DEG-CH3 substituents, along with
variety of discrete, soluble boronate ester assemblies.
the influence of a significantly larger solubilizing group
(THB) that has the potential to promote aggregation[31] in
solution. The catechol moieties were introduced such that
they can be oriented syn-periplanar and thus the bis(cate-
chol)s can function as 180° secondary building units. Ad-
ditionally, the catechol units at positions 3 and 6 of phenan-
threne were either linked directly as biaryls (compounds 1,
3, and 5) or linked via an acetylene spacer (compounds 2,
4, and 6). The acetylene-linked compounds (2, 4, 6, and 24–
26) were notably less soluble than their biaryl counterparts,
both complicating their purification and resulting in lower
product yields. Preparing bis(catechol)s with the two dif-
ferent linkages was valuable as they result in boronate ester
assemblies with larger or smaller shape-persistent cores,
thus allowing the relationship between increased or de-
Results and Discussion
In previous investigations of discrete boronate ester rect-
angles (Scheme 1, c), solubility of linear bis(catechol)s was
imparted by functionalization of their central phenyl rings
with hexyloxy chains. Upon the successful use of these lin-
ear bis(catechol) SBUs in the self-assembly of discrete
boronate ester rectangles it became of interest to explore
the scope of this synthetic approach. Phenanthrene-based
subunits were chosen for the current study in part because
of the increased π-conjugation of their polycyclic aromatic
core. Boronate ester rectangle assemblies were found to
fluoresce in the violet-blue region of the visible spectrum
Scheme 2. Synthetic routes to (a) biaryl- and acetylene-linked bis(catechol)s 1–6 and (b) 1,3,5-tris(hexyloxy) benzyl derivative 28.
4
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
Dynamic Assembly of Covalent Organic Polygons
creased π-surface area and the solubility of resulting assem- conjugated bis(catechol) 2 with BDBA in CDCl3/CD3OD
blies to be investigated. Overall the design is expected to (10:1) and heating to 50 °C resulted in the formation of an
1
promote the [2+2] condensation of bis(catechol)s 1–6 with insoluble precipitate that precluded analysis by H NMR
BDBA (Figure 1) to give discrete, oval-shaped boronate es- spectroscopy. Attempts to obtain NMR spectra in alterna-
ter assemblies.
tive solvents (C6D6, CD3CN, CD2Cl2, CD3SOCD3, and
C2D2Cl4) gave similar results.
The assembly of DEG-CH3-substituted biaryl- and
acetylene-linked bis(catechol)s 3 and 4, respectively, with
BDBA under the same conditions gave similar results as
related bis(catechol)s 1 and 2. Stirring and heating an equi-
molar mixture of BDBA with biaryl-linked bis(catechol) 3
resulted in the formation of a sparingly soluble assembly.
Analysis by 1H NMR spectroscopy (CD3COCD3) again re-
vealed the disappearance of catechol –OH signals between
8.0–8.1 ppm and the concomitant appearance of a singlet
at δ = 7.83 ppm consistent with the formation of boronate
ester species. The assembly product formed upon mixing
BDBA with acetylene-linked bis(catechol) 4, on the other
hand, was again found to be insoluble in all available sol-
vents.
Sparingly Soluble Assemblies
Initial investigations focused on hexyloxy-functionalized
bis(catechol)s 1 and 2. Both phenanthrene-based bis(cate-
chol) derivatives were found to be poorly soluble in chloro-
form, despite the presence of two hexyloxy substituents.
The di-boronic acid linker BDBA is insoluble in chloro-
form. Combining equimolar amounts of BDBA with either
1 or 2 in chloroform leads to heterogeneous mixtures, with
no evidence of assembly formation at ambient temperature
or upon heating to 50 °C. The addition of 10% methanol,
however, aided in the dissolution of both starting materials
and promoted their dynamic assembly (Scheme 3).
NMR spectroscopic analysis consistent with the forma-
tion of discrete, shape-persistent boronate ester assemblies
7 and 9 cannot rule out the potential formation of highly
symmetric boronate-ester oligomers. Furthermore, NMR
analysis of assemblies 8 and 10 provided no insight into
their structure as they were insoluble in all available NMR
solvents. Accurate mass MALDI mass spectrometry was
therefore employed to more conclusively establish the mol-
ecularity of assemblies 7–10. The MALDI mass spectra of
ovals 7 and 9 reveals peaks of m/z = 1376.6329 [M]+ and
1448.5331 [M]+ respectively (Figure 2). Both these values
are in agreement with the calculated values of 1376.6330 for
boronate ester assembly 7 and 1448.5297 for boronate ester
assembly 9, quantitatively supporting the formation of the
Scheme 3. Assembly of biaryl-linked bis(catechol)s 1 and 3 with
BDBA in CDCl3/CD3OD (10:1) gives sparingly soluble discrete
assemblies 7 and 9. Assembly of acetylene-linked bis(catechol)s 2
and 4 with BDBA under the same conditions gives insoluble
boronate ester assemblies. Assembly in CDCl3 results in hetero-
geneous mixtures and no boronate ester formation.
The formation of boronate ester assemblies was evalu-
1
ated by H NMR spectroscopy (CD3COCD3). Preliminary
evidence supportive of the dynamic assembly of hexyloxy-
substituted bis(catechol) 1 and BDBA to give boronate ester
species was observed by the disappearance of catechol –OH
signals at δ = 8.08 and 8.04 ppm along with the appearance
of a new singlet at δ = 7.84 ppm corresponding to the incor-
poration of BDBA (see Supporting Information). The over-
all symmetry of signals observed by 1H NMR spectroscopy
suggested the formation of symmetric assemblies, likely cor-
responding to boronate ester oval 7. While promising, the
low solubility of the assembly prevented conclusive analysis
Figure 2. Accurate mass MALDI mass spectra of 7 and 9 reveals
based on 1H NMR spectra alone. Combining acetylene- that both species are discrete [2+2] boronate ester ovals.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
M. K. Smith, A. R. Goldberg, B. H. Northrop
FULL PAPER
discrete boronate ester ovals. It is interesting to note that are tilted 144° with respect to the central plane of the as-
the MALDI-MS of 9 reveals a second, weaker signal at sembly. All atoms within the shape-persistent core of the
m/z = 1471.5298 corresponding to the [M + Na]+ species, acetylene-linked assemblies, by contrast, are predicted to be
potentially resulting from chelation of Na+ by the DEG- coplanar.
CH3 substituents of 7. Quantitative mass spectra of acetyl-
ene-conjugated boronate ester assemblies 8 and 10 could supportive of the formation of boronate ester ovals 7–10.
not be obtained, likely due to their insolubility. In particular, MALDI mass spectrometric analysis of
Taken together the collective spectroscopic evidence is
Infrared (IR) spectroscopy provided more definitive assemblies 7 and 9 indicates that they are discrete boronate
analysis of the chemical structures of boronate ester assem- ester [2+2] assemblies. However both of these discrete
blies 7–10. All four phenanthrene-based bis(catechol) deriv- boronate ester ovals are poorly soluble in common organic
atives displayed broad peaks centered at 3300 cm–1 corre- solvents. In the case of assemblies involving acetylene-
sponding catechol O–H stretching modes. These peaks dis- linked bis(catechol)s 2 and 4, neither hexyloxy nor DEG-
appear upon assembly with BDBA, indicating the absence CH3 substituents are capable of providing sufficient solubil-
of OH functionalities in the resulting assemblies and, by ity to balance the higher π-surface area of the polycyclic
extension, the complete consumption of both bis(catechol) aromatic core of each assembly. The assemblies were found
and BDBA starting materials. Evidence specific to boronate to be thermally robust as all four compounds display melt-
ester formation is observed primarily in the fingerprint re- ing points above 200 °C.
gion of the IR spectra of assemblies 7–10. Strong bands
Given the significant structural similarity between target
centered between 658–660 cm–1 were observed for each of boronate ester assemblies 7–10 and previously reported
the four assemblies, corresponding to out-of-plane displace- boronate ester rectangles (Scheme 1, c) it is both surprising
ments of boron and oxygen atoms specific[32] to boronate and interesting to observe such a dramatic difference in the
ester functionalities. Boroxine anhydrides, by contrast, are solubility of this new series of compounds relative to those
characterized by out-of-plane boron displacements above previously studied. Within this series it appears that the in-
700 cm–1. Furthermore, assemblies 7–10 each displayed crease in π-surface is the dominant influence limiting the
sharp B–O and C–O stretching modes between 1050– solubility of proposed boronate ester assemblies 7–10. Each
1070 cm–1 and between 1230–1250 cm–1, respectively. of the target assemblies 7–10 possesses greater π-surface
Stretches in these ranges are also diagnostic of boronate area at their core than the notably more soluble boronate
ester functionalities. Overall, IR spectroscopic analysis of ester rectangles. Only assemblies 7 and 9, i.e. those with less
assemblies 7–10 is supportive of the conclusion that all overall π-surface area within a smaller shape-persistent
starting materials have been consumed (absence of O–H core, could be definitively characterized by mass spectro-
stretching modes) and boronate esters have been formed. metric analysis as being discrete boronate ester oval assem-
No evidence of the formation of boroxine anhydride species blies. The larger acetylene-linked assemblies (8 and 10) with
could be found, indicating consumption of BDBA occurred greater π-surface area proved to be too insoluble for charac-
through its reaction with starting bis(catechol)s 1–4 rather terization by NMR or mass spectrometry. It is of interest to
than through self-condensation.
investigate whether functionalization of the phenanthrene
Boronate ester assemblies 7–10 each display characteris- moieties of biaryl- and acetylene-linked bis(catechol)s with
tically high extinction coefficients, and therefore UV/Vis larger solubilizing groups would be sufficient to overcome
and fluorescence spectra of dilute solutions (10–5 m in acet- the observed insolubility imposed by their relatively large
one) of each assembly could be obtained even though the π-surface areas. Toward this aim, THB-substituted bis(cate-
same solutions had proven too dilute for the acquisition of chol)s 5 and 6 (Figure 1) were synthesized and their as-
1
well-resolved H and 13C NMR spectra: 7 (λabs = 328 nm, sembly with BDBA was investigated.
ε
=
2.1ϫ104 m–1 cm–1),
8
(λabs
=
348 nm,
ε
=
6.8ϫ104 m–1 cm–1), 9 (λabs = 327 nm, ε = 3.5ϫ104 m–1 cm–1),
and 10 (λabs = 348 nm, ε = 4.3ϫ104 m–1 cm–1). All four
assemblies showed significant fluorescence intensity, with 7
Soluble Boronate Ester Assemblies
and 9 emitting in the violet region of the spectra (λem
=
The synthetic route used to prepare phenanthrene bis-
413 nm for both), and 8 and 10 emitting in the blue range (catechol)s 1–4 (Scheme 2, a) is easily adaptable to allow a
at λem = 474 and 475 nm, respectively. The absorbance and variety of solubilizing groups to be incorporated at posi-
fluorescence spectra of minimally soluble acetylene-linked tions 9 and 10 of their phenanthrene moiety. In an attempt
assemblies (proposed structures 8 and 10) are both red- to prepare notably more soluble analogues of boronate ester
shifted relative to biaryl assemblies 7 and 9 on account of assemblies 7–10, 3,4,5-tris(hexyloxy) benzyl bromide 28, a
the greater π-conjugation within their structures. Molecular derivative of gallic acid (Scheme 2, b), was introduced.
modeling of the shape-persistent core of biaryl-linked THB-functionalized bis(catechol)s 5 and 6 were synthesized
boronate ester ovals 7 and 9 (see Figure S5 of the Support- upon reacting 3,6-dibromophenanthrene-9,10-diol (15) with
ing Information) predicts that the catechol and phenan- benzyl bromide 28.
threne moieties twist ca 36.2° relative to each other. This
Initial attempts to assemble THB-substituted boronate
biaryl twist causes the ovals to adopt a bowl-shaped confor- ester ovals 11 and 12 (Scheme 4, bottom) relied upon the
mation wherein the phenanthrene moieties of the assembly same reaction conditions used to prepare assemblies of
6
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
Dynamic Assembly of Covalent Organic Polygons
BDBA with bis(catechol)s 1–4, namely stirring an equi- OHα and –OHβ signals, which appear at δ = 6.1 and
molar amount of BDBA and bis(catechol) derivative 5 or 6 5.8 ppm for bis(catechol) 6, disappear upon assembly. The
at 50 °C in a 10:1 mixture of CDCl3 and CD3OD. 1H NMR disappearance of catechol –OH signals is coincident with
spectra recorded shortly after mixing (ca. 15 min) revealed the emergence of singlet Hγ at δ = 7.96 ppm, which corre-
significantly increased solubility of the species in solution. sponds to the proton of BDBA in assembly 12. Free BDBA
After 3 h of mixing the 1H NMR spectrum remained highly and its corresponding anhydride are insoluble in chloro-
complex and it was clear that multiple species, including a form, therefore appearance of an aromatic boronate ester
significant percentage of starting materials, were present in signal around 7.96 ppm is strongly indicative of the as-
solution. It is a common characteristic of dynamic covalent sembly of BDBA with bis(catechol) 6. Assembly is further
self-assembly[1] that initial 1H NMR spectra reveal multiple supported by 0.6–0.3 ppm downfield shifts of aromatic cat-
1
species as different oligomers and kinetic intermediates echol unit peaks. As can be seen in Figure 3 the H NMR
equilibrate toward a final, thermodynamically stable prod- spectrum of assembly 12 is not cleanly resolved, showing
uct. In the case of target boronate ester assemblies 11 and evidence of baseline noise and the existence of trace signals
1
12, however, signals in each assemblies’ H NMR spectra not attributable to 12. It is likely that minor quantities of
did not sharpen as a function of time. These results were alternative assemblies are present at equilibrium, as is often
somewhat surprising given the overall structural similarities the case in dynamic combinatorial libraries. The intensity
boronate ester ovals 11 and 12 share with assemblies 7 and of signals corresponding to 12, however, suggests that it is
9, which did collapse to give one predominant structure at the dominant species at equilibrium. Analogous changes
equilibrium as indicated by 1H NMR spectroscopy and were observed for the formation of boronate ester oval as-
MALDI mass spectrometry.
sembly 11 (see Figure S1 of the Supporting Information).
Scheme 4. Dynamic covalent assembly of THB-functionalized
boronate ester ovals 11 and 12 from bis(catechol)s 5 and 6, respec-
tively, is achieved in CDCl3. In a mixture of CDCl3/CD3OD (10:1)
gives only starting materials and oligomers.
The primary difference between assemblies 7 and 9 vs.
11 and 12 is the introduction of hexyloxy-substituted gallic
acid derivatives. It was hypothesized that these larger solu-
bilizing groups may have changed the solubility of bis(cate-
chol) starting materials 5 and 6 so significantly that in a
10:1 mixture of CDCl3/CD3OD the dynamic equilibrium is
shifted toward starting materials and small oligomers bear-
ing terminal –OH functionalities. To investigate this hy-
!
Figure 3. Partial H NMR spectra (CDCl3, 400 MHz, 298 K) of
THB-functionalized bis(catechol) 6 (a) and boronate ester assembly
12 (b) indicating shifts of diagnostic proton signals observed upon
self-assembly.
The sensitivity of boronate ester ovals 11 and 12 to protic
pothesis further the self-assembly of boronate ester ovals 11 solvents was further investigated by adding increasing
and 12 was attempted in pure CDCl3. Returning to purely equivalents of CD3OD to a CDCl3 solution of each as-
1
hydrophobic conditions and forgoing the catalytic methanol sembly. The resulting H NMR spectra became disordered
proved successful, and provided assemblies 11 and 12 in upon the addition of 10 equiv. of CD3OD, and continued
good yields (Scheme 4, top). The formation of boronate es- to increase in disorder until the dynamically equilibrating
ter ovals 11 and 12 as dominant, though not necessarily species had fully reverted to starting materials by 50 equiv.
exclusive, species in solution is supported by key shifts in of CD3OD (see Figure S2 of the Supporting Information).
1
their H NMR spectra (Figure 3). Most notably, catechol – For comparison, initial attempts to self-assemble boronate
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
M. K. Smith, A. R. Goldberg, B. H. Northrop
FULL PAPER
ester ovals 11 and 12 in a mixture of 10:1 CDCl3/CD3OD would have near identical IR spectra as boronate ester ovals
corresponded to approximately 1000 equiv. of CD3OD, far 11 and 12. However, it is believed that the proposed discrete
exceeding the 50 equiv. found sufficient to favor disas- assemblies 11 and 12 are the principle, thermodynamic spe-
sembly. This observation is in direct contrast to sparingly cies in solution given that initially complex mixtures of dy-
soluble boronate ester assemblies 7 and 9, which showed no namically equilibrating species collapse to form a domi-
indication of product formation in pure CDCl3 and pro- nant, well-defined final structure observed by 1H NMR
longed stability in 10% CD3OD. For additional compari- spectroscopy.
son, the closely related boronate ester rectangle assemblies
(Scheme 1, c) could only be reverted to starting materials
by refluxing in a mixture of 1:1 D2O/CD3OD.[29] These ob-
servations provided additional evidence that the 3,4,5-tris-
Model of Boronate Ester Assembly Formation
(hexyloxy)benzyl substituents of bis(catechol) derivatives 5
Analysis of THB-functionalized boronate ester assem-
and 6 both increase the solubility of boronate ester assem- blies 11 and 12 indicates that sufficiently large solubilizing
blies 11 and 12 while also decreasing their stability in the groups are able to overcome the general insolubility ob-
presence of protic solvents.
served for assemblies 7–10. Within this series of com-
Further support for the formation of boronate ester ovals pounds, however, increased solubility comes at the cost of
was obtained by IR spectroscopy. Strong intensity bands increased lability in the presence of protic solvents. Collec-
were observed at 1225, 1063, 638 cm–1 (oval 11) and at tively, the results presented herein help build a better under-
1222, 1050, and 659 cm–1 (oval 12). These three IR bands standing the varying roles solubility, functionality, and reac-
are characteristic[32] of boronate esters and correspond to tion conditions can play in the self-assembly, isolation, and
symmetric C–O stretching, B–O stretching, and out-of- characterization of shape-persistent boronate ester assem-
plane displacement of boron and oxygen atoms within the blies. Shown in Figure 4 is a model that summarizes these
boronate ester C2O2B ring. These IR results are consistent roles within the context of the current research.
with the observed results for minimally soluble ovals 7–10.
In Figure 4 different pathways potentially leading to the
The UV/Vis and fluorescence spectra (10–5 m in CHCl3) formation of discrete, shape-persistent boronate ester
were also found to be comparable with the results obtained assemblies are indicated by different equilibrium arrows:
for assemblies 7–10, with high extinction coefficients for dashed arrows represent unproductive pathways while solid
assemblies 11 (332 nm, ε = 4.2ϫ104 m–1 cm–1) and 12 (λabs arrows indicate pathways that may lead to assembly forma-
= 353 nm, ε = 3.7ϫ104 m–1 cm–1), and emission in the blue tion. Furthermore, bold arrows indicate precipitation,
region of the visible spectrum (11, λem = 394 nm and 12, which removes species from dynamic covalent libraries
λem = 409 nm).
(DCLs) and can contribute to driving equilibria in a given
1
As discussed previously, H NMR and IR spectroscopy direction. As can be seen in the Figure different reaction
alone cannot rule out the formation of highly symmetric conditions allow entrance into the dynamically equilibrat-
boronate ester oligomers as opposed to discrete assemblies. ing libraries of species depending on the functionality of
Mass spectrometry was attempted to quantitatively estab- bis(catechol) SBUs. Less soluble hexyloxy and DEG-CH3-
lish the molecularity of proposed boronate ester ovals 11 substituted species 1–4 are prevented from entering into a
and 12. Unfortunately, despite considerable effort, mass DCL when mixed with BDBA in pure chloroform, even at
spectral analysis of both assemblies by MALDI and electro- elevated temperatures. The addition of 10% CD3OD, how-
spray ionization, both considered generally “soft” ioniza- ever, provides the sufficient solubility. In the case of THB-
tion techniques, revealed only starting materials and frag- substituted bis(catechol)s 5 and 6, however, the presence of
ments. In general, all investigations of proposed assemblies CD3OD shifts the equilibrium too far toward starting mate-
11 and 12 have shown that they are significantly more labile rials thus preventing productive dynamic covalent exchange.
than related boronate ester assemblies 7–10 and previously In the absence of CD3OD, the thermodynamics (and there-
studied boronate ester rectangles. Differential scanning fore equilibrium) shift back toward the assembly of 5 and
calorimetry (DSC) and melting point analyses of all six 6 with BDBA.
boronate ester assemblies also indicate that THB-substi-
Finding productive pathways out of a dynamically con-
tuted assemblies 11 and 12 are less thermally stable than verting mixture of intermediates is of equal importance to
hexyloxy and DEG-CH3 functionalized assemblies 7–10. successful assembly formation. When and how readily a
DSC traces of 11 revealed sudden decomposition, without given assembly is removed from of a DCL, potentially
a coherent melt, at 195 °C. Similarly, assembly 12 was ob- driven by precipitation or aggregation, is especially impor-
served to decompose at 185 °C. By comparison, the less sol- tant. Precipitation rates and pathways influence crystal
uble assemblies 7–10 were each stable well above 200 °C. It growth and quality, the ability to form co-assemblies and
is believed that the general lability of highly functionalized co-crystals, and underlie the (unproductive) precipitation of
assemblies 11 and 12 is the primary factor underlying in the undesired kinetic intermediates. Assemblies involving
difficulty of obtaining mass spectra of the discrete assem- acetylene-linked bis(catechol)s 2 and 4, for example, are so
blies. It is of course possible that the proposed boronate poorly soluble that they readily precipitate out of solution.
ester assemblies 11 and 12 are not in fact discrete and are It is believed that a primary driving force for precipitation
instead mixtures of various oligomers. Such oligomers is the formation of closed, discrete macrocycles that (i) lack
8
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
Dynamic Assembly of Covalent Organic Polygons
Figure 4. Generalized model of the varying pathways that favour the dynamic assembly of boronate ester assemblies and how differences
in SBU structure and reaction conditions influence self-assembly.
any hydroxy functionalities and (ii) are able to interact and that similar models can be applied to other discrete covalent
aggregate[33] through aromatic stacking interactions. The organic polygons and, potentially, the synthesis and isola-
planar, acetylene-conjugated assemblies formed from 2 and tion of covalent organic frameworks.
4 are therefore observed to be the least soluble of the series
investigated. Assemblies formed between biaryl bis(cate-
chol)s 1 and 3 with BDBA display a more productive bal-
ance wherein their low solubility aids in pulling discrete
Conclusions
species out of solution while biaryl twisting disrupts intra-
The results presented herein suggest that the balance be-
molecular π-stacking and aggregation. The resulting assem- tween SBU solubility, assembly formation, and assembly
blies 7 and 9 are therefore stable enough to withstand stability can be highly sensitive. This research builds upon
hydrolysis (equilibrium favoring products) yet still soluble recent related work where seemingly subtle differences in
enough to characterize.
functionality can dramatically influence the success or fail-
The greatly increased solubility of THB-substituted bis- ure of obtaining discrete, shape-persistent boronate ester
(catechol)s 5 and 6, however, renders their resulting assem- (or boroxine anhydride) assemblies. In order to fully de-
blies with BDBA highly labile to disassembly in protic sol- velop the chemistry and applications of discrete boronate
vents. The equilibrium between dynamically interconverting ester assemblies it is necessary to obtain a better under-
species and discrete assemblies 11 and 12 is tenuous and standing of the many factors that determine (i) whether dis-
readily favors starting materials and small oligomers in the crete assemblies will form, (ii) to what extent such assem-
presence of trace water or methanol. It is therefore observed blies will form, and (iii) under what conditions such assem-
that significantly increasing solubility, while helpful in pro- blies will remain stable. Understanding the roles that func-
viding pathways into DCLs, can actually limit the ability tionality, solubility, and stability play in the assembly of
to form and characterize discrete boronate ester assemblies. discrete boronate ester COPs can be expected to also con-
Such highly soluble and readily labile assemblies are less tribute to the optimization of reaction conditions used to
likely to drive dynamically converting equilibria toward a synthesize related COFs. Without a thorough understand-
desired product by precipitation or aggregation processes. ing of the experimental and structural influences that
This subtle balance between the solubility and stability of underlie the assembly of discrete boronate ester polygons/
discrete assemblies directly influences some of their most polyhedra, and infinite boronate ester COFs, it can be ex-
valuable attributes, namely their ability to form stable, pected that the discovery of new frameworks and the opti-
higher-order aggregates in solution, on surfaces, and within mization of their synthesis will largely be an effort of trial
materials applications. While Figure 4 is drawn specifically and error. The results presented herein hope to chart some
in relation to results gathered in the current study it is likely small, yet important, steps toward a better understanding
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
9

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
M. K. Smith, A. R. Goldberg, B. H. Northrop
FULL PAPER
(s, 2 H), 8.26 (d, J = 9.4 Hz, 2 H), 7.71 (d, J = 11.4 Hz, 2 H), 4.43–
4.39 (m, 4 H), 3.88–3.84 (m, 4 H), 3.72–3.69 (m, 4 H), 3.61–3.58
(m, 4 H), 3.42 (s, 6 H) ppm. 13C NMR (CDCl3, 75 MHz): δ =
142.9, 130.5, 128.9, 128.5, 125.2, 124.7, 120.5, 72.4, 72.0, 70.5, 70.4,
59.1 ppm.
of boronate ester self-assembly with the ultimate goal of
more judicious, de novo design and synthesis of new shape-
persistent boronate ester framework assemblies, both dis-
crete and infinite.
Compound 18: The crude product was extracted with hexanes, and
eluted from the column with 1:1 dichloromethane/hexanes, afford-
ing the pure product as yellow solid. Reaction scale: 15 (135 mg,
0.368 mmol), (261 mg, 62%), m.p. 73–75 °C. ESI/APCI (m/z)
[MNH4]+ calculated for C64H96NO8Br2, 1164.5497, found
1164.5459. 1H NMR (CDCl3, 300 MHz): δ = 8.65 (s, 2 H), 8.06 (d,
J = 8.8 Hz, 2 H), 7.66 (d, J = 9.7 Hz, 2 H), 6.6.4 (s, 4 H), 5.20 (s,
4 H), 3.94 (t, J = 6.0 Hz, 4 H), 3.88 (t, J = 6.7 Hz, 8 H), 1.79–1.70
(m, 12 H), 1.51–1.40 (m, 12 H), 1.38–1.29 (m, 24 H), 0.92 (t, J =
6.3 Hz, 18 H) ppm. 13C NMR (CDCl3, 75 MHz): δ = 153.2, 143.1,
138.2, 131.9, 130.4, 128.8, 128.5, 125.3, 120.6, 106.8, 75.8, 73.4,
69.1, 31.7, 31.5, 30.2, 25.7, 22.6, 14.0 ppm.
Experimental Section
General: Chemicals were obtained from commercial sources and
used as purchased. Reagent-grade solvents were used as obtained
from commercial sources. Anhydrous solvents were dried using an
Innovative Technologies SPS-400-5 solvent purification system. 1H
and 13C NMR spectra were recorded with a Varian Mercury
(300 MHz and 75 MHz, respectively) spectrometer using residual
solvent as the internal standard. All chemical shifts are quoted
using the δ scale and all coupling constants are expressed in Hertz
[Hz]. Infrared spectroscopic analysis was performed on a Perkin–
Elmer Spectrum BX FT-IR system. UV/Vis spectroscopy was re-
corded on a Varian Cary 100 Bio UV/Vis spectrophotometer.
Differential scanning calorimetry (DSC) was performed on a TA
Instruments DSC Q20 equipped with an RCS90 cooling system
and were acquired at rates of 10 °C min–1 (heating) and 5 °C min–1
(cooling). ESI/APCI-MS and MALDI-MS analyses were carried
out at the University of California, Riverside, Mass Spectrometry
Facility.
General Procedure for Preparing Diaryl-Substituted Phenanthrenes
21–23: To a heavy walled glass vessel was added phenanthrene di-
halide, 14 (2.5 equiv.), and potassium phosphate (4 equiv.). The
flask was flushed with nitrogen, and palladium(II) acetate (2 mol-
%) and SPhos ligand (4 mol-%) were added. The flask was purged
and backfilled with nitrogen (3ϫ), and degassed toluene and water
(10:1, 0.5 m with respect to phenanthrene) were added. The flask
was sealed with a Teflon® screw-cap and stirred at 100 °C over-
night. The reaction suspension was cooled and filtered through Ce-
lite, eluting with dichloromethane. The filtrate was concentrated
under reduced pressure and purified via column chromatography.
Compound 28: To
a solution of triphenylphosphine (6.3 g,
24.10 mmol) in dichloromethane (80 mL) was added carbon tetra-
bromide (4.0 g, 12.05 mmol). A solution was prepared of 27[34]
(4.9 g, 12.05 mmol) in dichloromethane (24 mL), which was added
to the stirring triphenylenephosphine and carbon tetrabromide
solution at 0 °C. The reaction was stirred at 0 °C for 1.5 h, at which
point water was added (150 mL), and the crude product extracted
with chloroform (3ϫ100 mL). The combined organic layers were
washed with brine (150 mL), dried with MgSO4, and concentrated
under reduced pressure. The crude residue was purified using a pad
of silica, eluting with 1:1 hexanes/dichloromethane, affording pure
28 as a colourless oil (4.7 g, 83%). ESI/APCI (m/z) [MH]+ calcu-
lated for C25H44O3Br, 471.2468, found 471.2464. 1H NMR (CDCl3,
300 MHz): δ = 6.58 (s, 2 H), 4.45 (s, 2 H), 4.00–3.92 (m, 6 H),
1.85–1.74 (m, 6 H), 1.50–1.46 (m, 6 H), 1.35–1.33 (m, 12 H), 0.96–
0.89 (m, 9 H) ppm. 13C NMR (CDCl3, 75 MHz): δ = 153.2, 132.3,
107.0, 73.4, 69.1, 47.0, 31.7, 31.5, 30.2, 29.3, 25.7, 22.6, 14.0 ppm.
Compound 21: Note: for this compound, Sphos Palladacycle
(4 mol-%) was used in place of palladium(II) acetate and Sphos
ligand. Similarly, potassium carbonate (4 equiv.) was used in place
of potassium phosphate. The product eluted with 10:1 hexanes/
dichloromethane, and the pure product isolated as an white solid.
Reaction scale: 16 (404 mg, 0.712 mmol), yield (753 mg, 99%
yield). (Note: 21 has also been isolated using the same reaction
conditions as 22, with a lower yield of 50%) M.p. 104.9–105.7 °C.
APCI-MS (m/z) [M + H]+ calculated for C62H99O6Si4, 1051.6513,
1
found 1051.6511. H NMR (CDCl3, 300 MHz): δ = 8.81 (s, 2 H),
8.27 (d, J = 8.5 Hz, 2 H), 7.79 (d, J = 8.5 Hz, 2 H), 7.31–7.23 (m,
4 H), 6.97 (d, J = 7.6 Hz, 2 H), 4.24 (t, J = 12 Hz, 4 H), 1.99–1.89
(m, 4 H), 1.64–1.57 (m, 4 H), 1.44–1.39 (8 H), 1.05 (s, 18 H), 1.04
(s, 18 H), 0.95 (t, J = 6.0 Hz, 6 H), 0.29 (s, 12 H), 0.27 (s, 12 H)
ppm. 13C NMR (CDCl3, 75 MHz): δ = 147.1, 146.6, 143.1, 138.1,
134.8, 128.8, 128.6, 125.8, 122.8, 121.4, 120.4, 120.3, 73.7, 31.7,
30.5, 25.9, 22.7, 18.6, 14.1, –4.0 ppm.
General Method for the Preparation of Dibromophenanthrene Deriv-
atives 16–18: To a mixture of 15,[35] the appropriate alkyl halide
or tosylate (3 equiv.), potassium carbonate (4–6 equiv.), 18-crown-
6 (catalytic), and lithium bromide (catalytic) were added under an
inert atmosphere in a minimum amount of dry dimethylformamide.
The reaction solution was stirred at 80 °C overnight, cooled, and
poured over water. The organic material was extracted (3ϫ) with
an organic solvent, and the combined organic layers washed twice
with water, dried with MgSO4, and concentrated under reduced
pressure. The crude product was purified by flash chromatography.
Compound 22: The product eluted with 12:1 dichloromethane/ethyl
acetate, affording 22 as a pale yellow solid Reaction scale: 17
(365 mg, 0.638 mmol), yield (623 mg, 90% yield), m.p. 89.4–
92.3 °C. APCI-MS (m/z) [M + Na]+ calculated for C60H94O10N-
1
aSi4, 1109.5816, found 1109.5851. H NMR (CDCl3, 300 MHz): δ
= 8.80 (s, 2 H), 8.32 (d, J = 8.5 Hz, 2 H), 7.73 (dd, J = 8.5, 1.5 Hz,
2 H), 7.43–7.09 (m, 8 H), 6.83 (d, J = 8.2 Hz, 2 H), 4.46–4.43 (m,
4 H), 3.90–3.87 (m, 4 H), 3.74–3.71 (m, 4 H), 3.63–3.60 (m, 4 H),
3.43 (s, 6 H), 1.03 (s, 18 H), 1.01 (s, 18 H), 0.27 (s, 12 H), 0.24 (s,
12 H) ppm. 13C NMR (CDCl3, 75 MHz): δ = 147.1, 146.7, 142.8,
138.3, 134.8, 132.1, 129.5, 128.9, 128.4, 125.8, 123.5, 123.2, 121.4,
120.4, 120.3, 72.4, 72.1, 70.6, 59.1, 26.1, 18.6, –4.0 ppm.
Compound 16: The crude product was extracted with hexanes and
eluted from the column with 10% dichloromethane in hexanes. Re-
action scale: 15 (0.8 g, 2.17 mmol), yield (0.9 g, 82%). Spectro-
scopic characterization matched that reported in the literature.[35]
Compound 17: The crude compound was extracted with ethyl acet-
ate and eluted from the column with 3:1 hexanes/ethyl acetate, af-
fording a pale yellow oil that gradually solidified. Reaction scale:
15 (313 mg, 0.85 mmol), yield (410 mg, 88% yield), m.p. 55.7–
56.6 °C. APCI-MS (m/z) [M + H]+ calculated for C24H29O6Br2,
Compound 23: The pure product eluted from the column in pure
dichloromethane. Reaction Scale: 18 (230 mg, 0.200 mmol), yield
(157 mg, 47%), yellow solid, m.p. 55–57 °C. ESI/APCI (m/z) [MNa]
+ calculated for C100H158O12NaSi4, 1686.0723, found 1686.0682.
1H NMR (CDCl3, 300 MHz): δ = 8.83 (s, 2 H), 8.28 (d, J = 8.5 Hz,
1
571.0325, found 571.0334. H NMR (CDCl3, 300 MHz): δ = 8.65
10
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
Dynamic Assembly of Covalent Organic Polygons
2 H), 7.77 (d, J = 8.5 Hz, 2 H), 7.27–7.22 (m, 4 H), 6.97 (d, J = General Procedures for the Deprotection of TBDMS Groups to Af-
7.9 Hz, 2 H), 7.75 (s, 4 H), 5.5 (s, 4 H), 3.99–3.91 (m, 12 H), 1.82– ford Phenanthrene-Bis(catechol)s 1–6. Method I: To a solution of
1.71 (m, 12), 1.54–1.41 (m, 12 H), 1.39–1.27 (m, 24 H), 1.06 (s, 18 TBDMS-protected phenanthrene monomer dissolved in dry di-
H), 1.05 (s, 18 H), 0.95–0.86 (m, 18 H), 0.29 (s, 12 H), 0.28 (s, 12 methylformamide (0.1 m with respect to phenanthrene) was added
H) ppm. 13C NMR (CDCl3, 75 MHz): δ = 153.2, 147.2, 146.7,
143.1, 138.4, 138.0, 134.7, 132.5, 129.0, 128.3, 125.8, 123.0, 121.4,
120.4, 120.3, 106.6, 75.8, 73.4, 69.1, 31.8, 31.6, 30.3, 29.4, 26.0,
25.8, 22.6, 18.5, 14.0, –4.0 ppm.
potassium fluoride (8 equiv.), followed by concentrated hydro-
bromic acid (0.6 equiv.). The reaction was monitored by TLC and
stirred overnight, at which point 2 m HCl was added (excess), and
the product extracted with an organic solvent. The combined or-
ganic layers were washed with brine, dried with MgSO4, and con-
centrated under reduced pressure. The crude product was purified
by column chromatography. Note: yields were not optimized for
phenanthrene deprotections.
General Procedure for Preparing Ethynyl-Spaced Phenanthrenes 24–
26: To a nitrogen-charged Schlenk flask was added 9,10-disubsti-
tuted 3,6-dibromophenanthrene (1 equiv.), 20 (3 equiv.), and tri-
phenylphosphine (0.2 equiv.). To the mixture was added trans-
dichlorobis(triphenylphosphine)palladium (0.1 equiv.), and copper
iodide (0.2 equiv.) in that order. The flask was evacuated, purged
with nitrogen, and wrapped in foil. To the reaction flask was added
dry, degassed piperidine and tetrahydrofuran (0.27 m and 0.125M
with respect to phenanthrene), and the reaction stirred for ca. 65 h,
at which point water was added, and the product extracted with an
organic solvent. The combined organic layers were washed with
brine, dried with MgSO4, and concentrated under reduced pressure.
The crude product was purified by column chromatography.
Method II: To a given TBDMS-protected phenanthrene derivative
was added an excess of tetraethylene-glycol (80 equiv.). Tetra-
hydrofuran was slowly added, with stirring, until reaction became
homogeneous. Potassium fluoride (8 equiv.) was added, provoking
an immediate color change. The reaction was monitored by thin
layer chromatography, and was complete within 30 min. Water was
added, and the product extracted with diethyl ether. The combined
organic layers were washed with water and brine. The combined
aqueous layers were back extracted with diethyl ether, and the com-
bined organic layers dried with MgSO4 and concentrated under
reduced pressure. The residue was purified by column chromatog-
raphy. The low solubility of compounds 2, 3, 4, and 6 made chro-
matographic purification nontrivial, significantly impacting prod-
uct yields. For bis(catechol)s 1–6 carbon spectra were omitted due
to very low solubility resulting in little to no discernible signal
above baseline noise.
Compound 24: The crude product was extracted with dichlorometh-
ane, and eluted from the column in 10% dichloromethane in hex-
anes. Reaction scale: 16 (177 mg, 0.331 mmol), yield 160 mg
(0.145 mmol, 44%). The product was isolated as a yellow oil.
APCI-MS (m/z) [M]+ calculated for C66H98O6Si4, 1096.6435, found
1098.6437. 1H NMR (CDCl3, 300 MHz): δ = 8.80 (s, 2 H), 8.19 (d,
J = 8.2 Hz, 2 H), 7.73 (d, J = 8.2 Hz, 2 H), 7.13–7.08 (m, 4 H),
6.83 (d, J = 8.2 Hz, 2 H), 4.22 (t, J = 7.3 Hz, 4 H), 1.96–1.87 (m,
Compound 1 (isolated using Method I): To quench the reaction,
4 H), 1.63–1.58 (m, 4 H), 1.43–1.36 (m, 8 H), 1.03 (s, 18 H), 1.01 20 mL 2 m HCl was added, and the crude product was extracted
(s, 18 H), 0.94 (t, J = 7.3 Hz, 6 H), 0.27 (s, 12 H), 0.24 (s, 12 H)
ppm. 13C NMR (CDCl3, 75 MHz): δ = 148.0, 146.8, 143.7, 129.8,
129.2, 127.9, 126.0, 125.6, 124.2, 122.4, 121.1, 120.9, 116.1, 90.2,
88.4, 73.8, 31.7, 30.4, 25.9, 22.7, 18.5, 14.1, –4.0 ppm.
with diethyl ether. The pure product eluted from the column with
100% ethyl acetate. Reaction Scale: 21 (570 mg, 0.542 mmol), yield
274 mg (0.461 mmol, 85%). The product was isolated as a lavender
solid, m.p. 148–154 °C. APCI-MS (m/z) [M]+ calculated for
1
C38H42O6, 594.2976, found 594.2995 (ppm error = 3.2). H NMR
Compound 25: The crude product was extracted with ethyl acetate,
and eluted from the column in 5% ethyl acetate in dichlorometh-
ane. Reaction Scale: 17 (1.4 g, 2.45 mmol), yield 695 mg
(0.612 mmol, 25%). The product was isolated as a dark oil. APCI-
MS (m/z) [M + Na]+ calculated for C64H94O10NaSi4, 1157.5816,
(CO(CD3)2, 300 MHz): δ = 9.01 (s, 2 H), 8.28 (d, J = 8.8 Hz, 2 H),
8.08 (s, 2 H), 8.04 (s, 2 H), 7.88 (dd, J = 8.8, 1.8 Hz, 2 H), 7.40 (d,
J = 2.4 Hz, 2 H), 7.27 (dd, J = 8.2, 2.4 Hz, 2 H), 6.99 (d, J =
8.2 Hz, 2 H), 4.26 (t, J = 6.4 Hz, 4 H), 2.00–1.91 (m, 4 H), 1.69–
1.59 (m, 4 H), 1.47–1.36 (m, 8 H), 0.94 (t, J = 7.3 Hz, 6 H) ppm.
1
found 1157.5848. H NMR (CDCl3, 300 MHz): δ = 8.80 (s, 2 H),
8.82 (d, J = 8.5 Hz, 2 H), 7.73 (d, J = 8.5 Hz, 2 H), 7.13–7.08 (m, Compound 2 (isolated using Method I): Instead of dimethylform-
4 H), 6.84 (d, J = 8.2 Hz, 2 H), 4.46–4.43 (m, 4 H), 3.90–3.86 (m, amide, the solvent used for this reaction was a 2:1 solution of tetra-
4 H), 3.74–3.71 (m, 4 H), 3.63–3.60 (m, 4 H), 3.43 (s, 6 H), 1.03 (s, hydrofuran:dimethylformamide. To quench the reaction, 10 mL of
18 H), 1.01 (s, 18 H), 0.27 (s, 12 H), 0.24 (s, 12 H) ppm. Low
solubility precluded the collection of a well-resolved 13C NMR
spectrum.
2M HCl was added, and the crude product was extracted with di-
ethyl ether. The pure product eluted from the column with 1:1 hex-
anes/ethyl acetate. Reaction Scale: 24 (158 mg, 0.144 mmol), yield
50 mg (0.078 mmol, 54%). The product was isolated as a purple
solid, m.p. 131–133 °C (note, the dark, oily nature of this material
makes observing the melting point difficult). APCI-MS (m/z) [M +
Compound 26: The reaction was not extracted, but instead was fil-
tered through Celite, eluting with dichloromethane. The filtrate was
concentrated under reduced pressure, and the crude material puri-
fied by column chromatography, eluting with 1:1 hexanes/dichloro-
methane, affording pure 26 as a yellow semisolid. Reaction scale:
18 (120 mg, 0.104 mmol), yield (93 mg, 52%). ESI/APCI (m/z) [M
1
H]+ calculated for C42H43O6, 643.3054, found 643.3063. H NMR
(CO(CD3)2, 300 MHz): δ = 8.94 (s, 2 H), 8.31 (s, 4 H), 8.24 (d, J
= 8.5 Hz, 2 H), 7.74 (d, J = 8.5 Hz, 2 H), 7.12 (d, J = 1.8 Hz, 2
H), 7.04 (dd, J = 8.2, 1.8 Hz, 2 H), 6.88 (d, J = 7.9 Hz, 2 H), 4.23
(t, J = 6.6 Hz, 4 H), 1.97–1.88 (m, 4 H), 1.65–1.55 (m, 4 H), 1.43–
1.38 (m, 8 H), 0.92 (t, J = 6.9 Hz, 6 H) ppm.
+
Na]+ calculated for C104H158O12NaSi4, 1734.0723, found
1734.0746. 1H NMR (CDCl3, 300 MHz): δ = 8.81 (s, 2 H), 8.17 (d,
J = 8.6 Hz, 2 H), 7.70 (d, J = 8.6 Hz, 2 H), 7.16–7.08 (m, 4 H),
6.84 (d, J = 8.2 Hz, 2 H), 6.69 (s, 4 H), 5.23 (s, 4 H), 3.99–3.87 (m, Compound 3 (isolated using Method I): To quench the reaction,
12 H), 1.80–1.74 (m, 12 H), 1.54–1.42 (m, 12 H), 1.36–1.30 (m, 24 20 mL of 2M HCl was added, and the crude product was extracted
H), 1.04 (s, 18 H), 1.01 (s, 18 H), 0.93–0.89 (m, 18 H), 0.27 (s, 12 with ethyl acetate. The pure product eluted from the column with
H), 0.25 (s, 12 H) ppm. 13C NMR (CDCl3, 75 MHz): δ = 153.2, 100% ethyl acetate. Reaction Scale: 22 (620 mg, 0.570 mmol), yield
148.0, 146.8, 138.2, 132.2, 129.8, 129.0, 128.1, 126.0, 125.6, 124.2,
122.6, 121.3, 121.1, 116.0, 106.9, 90.4, 88.3, 75.8, 73.4, 69.1, 31.8,
31.6, 30.3, 29.3, 26.0, 25.8, 22.6, 14.0, –4.1 ppm.
(161 mg, 45%). Product was isolated as dark red oil. APCI-MS
(m/z) [M + Na]+ calculated for C36H38O10Na, 653.2357, found
653.2346 (ppm error = –1.7). 1H NMR (CO(CD3)2, 300 MHz): δ
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
11

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
M. K. Smith, A. R. Goldberg, B. H. Northrop
Assembly 7 (isolated using Method I): Reaction Scale: 1 (17.0 mg,
FULL PAPER
= 9.01 (s, 2 H), 8.45 (d, J = 8.8 Hz, 2 H), 8.10 (s, 2 H), 8.05 (s, 2
H), 7.88 (dd, J = 8.5, 1.5 Hz, 2 H), 7.40 (d, J = 2.1 Hz, 2 H), 7.29 0.029 mmol), yield 18.8 mg (95%). The product was isolated as a
(dd, J = 8.2, 2.1 Hz, 2 H), 6.99 (d, J = 8.2 Hz, 2 H), 4.46–4.43 (m, grey solid that was sparingly soluble in acetone, m.p. Ͼ 200 °C. IR
4 H), 3.92–3.89 (m, 4 H), 3.72–3.69 (m, 4 H), 3.59–3.56 (m, 4 H),
3.35 (s, 6 H) ppm.
(powder, ATR): ν = 1605, 1395, 1325, 1250, 1121, 1018, 804, 699,
˜
660 cm–1. MALDI (m/z) [M]+ calculated for C88H8411B4O12,
1
1376.6330, found 1376.6329. H NMR (CD3COCD3, 300 MHz): δ
Compound 4 (isolated using Method I): To quench the reaction,
30 mL of 2 m HCl was added, and the crude product was extracted
with dichloromethane. The pure product eluted from the column
with 1:1 acetone/hexanes. Reaction Scale: 25 (690 mg, 0.608 mmol),
isolated yield (33 mg, 12%). The product was isolated as a dark
solid, m.p. 93–94 °C (note, the dark, oily nature of this material
= 8.99 (s, 4 H), 8.48 (s, 4 H), 8.28 (d, J = 4.2 Hz, 4 H), 8.04–8.03
(m, 4 H), 7.90–7.83 (m, 12 H), 6.97 (d, J = 4.1 Hz, 4 H), 4.27 (t, J
= 6.6 Hz, 8 H), 2.05–1.93 (m, 8 H), 1.72–1.58 (m, 8 H), 1.51–1.38
(m, 16 H), 0.96 (t, J = 6.6 Hz, 12 H) ppm.
Assembly 8 (isolated using Method I): Reaction Scale: 2 (21.0 mg,
makes observing the melting point difficult). APCI-MS (m/z) [M + 0.033 mmol), yield 23.4 mg (97%). The product was isolated as a
H]+ calculated for C42H37O10, 701.2381, found 701.2370. 1H NMR
(CO(CD3)2, 300 MHz): δ = 8.92 (s, 2 H), 8.40 (d, J = 8.5 Hz, 2 H),
8.31 (s, 4 H), 7.72 (d, J = 8.5 Hz, 2 H), 7.10 (d, J = 1.8 Hz, 2 H),
7.02 (dd, J = 8.2, 1.9 Hz, 2 H), 6.86 (d, J = 7.9 Hz, 2 H), 4.43–4.40
(m, 4 H), 3.86–3.84 (m, 4 H), 3.67–3.64 (m, 4 H), 3.54–3.51 (m, 4
H), 3.31 (s, 6 H) ppm.
black solid, m.p. Ͼ 200 °C. IR (powder, ATR): ν˜ = 1699, 1607,
1489, 1396, 1325, 1234, 1066, 1017, 805, 698, 659, 576 cm–1.
Assembly 9 (isolated using Method I): Reaction Scale: 3 (32.1 mg,
0.052 mmol), yield 26.5 mg (72%). The product was isolated as a
grey solid that was sparingly soluble in acetone, m.p. Ͼ 200 °C. IR
(powder, ATR): ν = 1601, 1476, 1393, 1316, 1230, 1050, 1018, 810,
˜
659 cm–1. MALDI (m/z) [M]+ calculated for C84H7611B4O20,
Compound 5 (isolated using Method II): The pure product eluted
with 25% diethyl ether in dichloromethane as a dark yellow solid.
Reaction scale: 23 (153 mg, 0.092 mmol), yield (89 mg, 80%). De-
composition: 125–135 °C. ESI/APCI (m/z) [M + Na]+ calculated
for C78H102O12Na, 1229.7264, found 1229.7233. 1H NMR (CDCl3,
300 MHz, 0.05 m, chemical shifts of several aromatic signals were
found to vary with concentration): δ = 8.40 (s, 2 H), 8.20 (d, J =
8.5 Hz, 2 H), 7.60 (d, J = 8.1 Hz, 2 H), 7.12–6.86 (m, 6 H), 6.71 (2
H), 6.51 (s, 4 H), 5.50 (2 H), 4.47 (s, 4 H), 4.04 (t, J = 6.6 Hz, 4
H), 3.79 (t, J = 6.4 Hz, 8 H), 1.86–1.78 (m, 12 H), 1.74–1.66 (m,
12 H), 1.41–1.26 (m, 24 H), 0.91 (t, J = 6.7 Hz, 18 H) ppm.
(CO(CD3)2, 300 MHz): δ = 9.01 (s, 2 H), 8.29 (d, J = 8.5 Hz, 2 H),
8.06 (s, 4 H), 7.85 (d, J = 8.5 Hz, 2 H), 7.39 (s, 2 H), 7.26 (dd, J =
7.9, 2.2 Hz, 2 H), 7.0 (d, J = 7.9 Hz, 2 H), 6.83 (s, 4 H), 5.27 (s, 4
H), 3.94–3.88 (m, 12 H), 1.77–1.66 (m, 12 H), 1.56–1.40 (m, 12 H),
1.39–1.27 (m, 24 H), 0.89 (t, J = 6.5 Hz, 18 H) ppm.
1
1448.5297, found 1448.5331. H NMR (CD3COCD3, 300 MHz): δ
= 8.97 (br. s, 4 H), 8.67 (br. s, 4 H), 8.42 (d, J = 8.7 Hz, 4 H), 7.92
(d, J = 8.4 Hz, 4 H), 7.84 (s, 8 H), 7.38 (d, J = 8.6 Hz, 4 H), 6.97
(d, J = 7.8 Hz, 4 H), 4.50–4.39 (m, 8 H), 3.95–3.82 (m, 8 H), 3.78–
3.62 (m, 8 H), 3.62–3.51 (m, 8 H), 3.35 (br. s, 12 H) ppm.
Assembly 10 (isolated using Method I): Reaction Scale: 4 (17.9 mg,
0.026 mmol), yield 17.9 mg (82%). The product was isolated as an
grey solid, m.p. Ͼ 200 °C. IR (powder, ATR): ν = 1604, 1477, 1396,
˜
1319, 1234, 1060, 997, 813, 660 cm–1.
Assembly 11 (isolated using Method II): Reaction scale: 5 (12.9 mg,
0.0107 mmol), yield orange solid 11 (13.9 mg, 99%). Decomposes
at 199 °C. IR (powder, ATR): ν = 1591, 1378, 1332, 1290, 1225,
˜
1
1110, 1063, 1039, 1000, 638 cm–1. H NMR (CDCl3, 300 MHz): δ
= 8.65 (s, 4 H), 8.28–8.12 (m, 4 H), 8.23 (s, 8 H), 7.83–7.70 (m, 8
H), 7.36–7.30 (m, 8 H), 6.72 (s, 8 H), 5.22 (s, 8 H), 3.98–3.88 (m,
24 H), 1.81–1.72 (m, 24 H), 1.55–1.43 (m, 24 H), 1.40–1.23 (m, 48
H), 0.91 (t, J = 5.0 Hz, 36 H) ppm.
Compound 6 (isolated using Method II): The product was extracted
with ethyl acetate, and eluted from the column with 10% diethyl
ether in dichloromethane, affording 6 as a dark yellow solid Reac-
tion Scale: 26 (130 mg, 0.076 mmol), yield (41 mg, 43%). Decom-
poses at 199 °C. ESI/APCI (m/z) [M + Na]+ calculated for
C78H102O12Na, 1229.7264, found 1229.7233. 1H NMR (CDCl3,
300 MHz, 0.05 m): δ = 8.59 (s, 2 H), 8.05 (d, J = 8.5 Hz, 2 H), 7.56
(d, J = 7.9 Hz, 2 H), 7.08–6.84 (m, 6 H), 6.73 (s, 4 H), 6.11 (s, 2
H), 5.78 (s, 2 H), 5.15 (s, 4 H), 3.99 (t, J = 5.6 Hz, 4 H), 3.90 (t, J
= 6.4 Hz, 8 H), 1.80–1.70 (m, 12 H), 1.53–1.49 (m, 12 H), 1.37–
1.26 (m, 24 H), 0.90 (t, J = 5.9 Hz, 18 H) ppm.
Assembly 12 (isolated using Method II): Reaction scale: 6 (15.7 mg,
0.0125 mmol) yield orange solid 12 (13.9 mg, 82%). Decomposes
at 185 °C. IR (powder, ATR): ν = 1592, 1435, 1353, 1316, 1222,
˜
1
1110, 1078, 1050, 1018, 820 cm–1. H NMR (CDCl3, 300 MHz): δ
= 8.51 (s, 4 H), 8.06 (d, J = 20 Hz, 4 H), 7.96 (s, 8 H), 7.63–7.59
(m, 8 H), 7.20–7.15 (m, 8 H), 6.56 (s, 8 H), 4.80 (s, 8 H), 3.93 (t, J
= 6.6 Hz, 8 H), 3.86 (t, J = 6.2 Hz, 16 H), 1.81–1.72 (m, 24 H),
1.52–1.42 (m, 24 H), 1.40–1.29 (m, 48 H), 0.93 (t, J = 6.6 Hz, 36
H) ppm.
Conditions to Afford Phenanthrene Assemblies 7–12. Method I: To
a mixture of organic tetra-ol (1 equiv.) and 1,4-benzenediboronic
acid (1 equiv.) was added 10:1 chloroform/methanol (chloroform
0.01 m with respect to reagents). The reaction solution was stirred
at 50 °C for 3 h at which point 4 Å molecular sieves were added,
and the reaction stirred overnight at room temperature. The reac-
tion mixture was dried with MgSO4, filtered (rinsing with acetone),
and the solvents removed via rotary evaporation. The solid residue
was subjected to high vacuum at 90 °C for 1 h to afford solvent
free assemblies.
Acknowledgments
This research was supported by the Wesleyan University. The au-
thors also thank the University of California, Riverside, Mass
Spectrometry Facility for mass spectrometric analysis.
[1] For reviews of dynamic covalent assembly, see: a) S. J. Rowan,
S. J. Cantrill, G. R. L. Cousins, J. K. M. Sanders, J. F. Stoddart,
Angew. Chem. Int. Ed. 2002, 41, 898–952; Angew. Chem. 2002,
114, 938–993; b) P. T. Corbett, J. Leclaire, L. Vial, K. R. West,
J.-L. Wietor, J. M. K. Sanders, S. Otto, Chem. Rev. 2006, 106,
3652–3711; c) Y. Jin, C. Yu, R. J. Denman, W. Zhang, Chem.
Soc. Rev. 2012, 42, 6634–6654; d) Y. Jin, Q. Wang, P. Taynton,
W. Zhang, Acc. Chem. Res. 2014, 47, 1575–1586.
Method II: A suspension was prepared of organic tetra-ol (1 equiv.)
and 1,4-benzenediboronic acid (1 equiv.) in CDCl3 (0.002 m) and
refluxed at 90 °C overnight. The resulting solution was cooled,
passed through a Teflon® syringe filter, and concentrated under
reduced pressure to afford the desired assembly. For assemblies 7–
12 carbon spectra could not be obtained because of their insolubil-
ity.
[2] For recent reviews of MOFs, see: a) M. Eddaoudi, D. B. Moler,
H. L. Li, B. Chen, T. M. Reineke, M. O’Keeffe, O. M. Yaghi,
12
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
Dynamic Assembly of Covalent Organic Polygons
Acc. Chem. Res. 2001, 34, 319–330; b) J. Y. Lee, O. K. Farha,
J. Roberts, K. A. Scheidt, S. T. Nguyen, J. T. Hupp, Chem. Soc.
Rev. 2009, 38, 1450–1459; c) H. Furukawa, K. E. Cordova, M.
O’Keeffe, O. M. Yaghi, Science 2013, 341, 974–986; as well as
the 2012 issue of Chem. Rev. devoted to MOFs: d) Furukawa,
K. E. Cordova, M. O’Keeffe, O. M. Yaghi, Chem. Rev. 2012,
112, 673–1268.
The term “Metal-Organic Framework” was introduced by
Yaghi in 1995, see: a) O. M. Yaghi, H. Li, J. Am. Chem. Soc.
1995, 117, 10401–10402; however structures resembling MOFs
had been prepared prior to the introduction of the term, see:
b) B. F. Hosking, R. Robson, J. Am. Chem. Soc. 1990, 112,
1546–1554; c) S. R. Batten, R. Robson, Angew. Chem. Int. Ed.
1998, 37, 1460–1494; Angew. Chem. 1998, 110, 1558–1595, and
references cited therein.
For recent reviews of COFs, see: a) X. Feng, X. Ding, D. Jaing,
Chem. Soc. Rev. 2012, 41, 6010–6022; b) J. W. Colson, W. R.
Dichtel, Nat. Chem. 2013, 5, 453–465; c) S.-Y. Ding, W. Wang,
Chem. Soc. Rev. 2013, 42, 548–568; d) Z. Xiang, D. Cao, J.
Mater. Chem. A 2013, 1, 2691–2718.
A. P. Côté, A. I. Benin, N. W. Ockwig, M. O’Keeffe, A. J.
Matzger, O. M. Yaghi, Science 2005, 310, 1166–1170.
a) E. L. Spitler, B. T. Koo, J. L. Novotney, J. W. Colson, F. J.
Uribe-Romo, G. D. Gutierrez, P. Clancy, W. R. Dichtel, J. Am.
Chem. Soc. 2011, 133, 19416–19421; b) E. L. Spitler, J. W. Col-
son, F. J. Uribe-Romo, A. R. Woll, M. R. Giovino, A. Saldivar,
W. R. Dichtel, Angew. Chem. Int. Ed. 2012, 51, 2623–2627; An-
gew. Chem. 2012, 124, 2677–2681.
a) S. Wan, F. Gándara, A. Asano, H. Furukawa, A. Saeki,
S. K. Dey, L. Liao, M. W. Ambrogio, Y. Y. Botros, X. Duan,
S. Seki, J. F. Stoddart, O. M. Yaghi, Chem. Mater. 2011, 23,
4094–4097; b) X. Ding, J. Guo, X. Feng, Y. Honshihito, J. Guo,
S. Seki, P. Maitarad, A. Saeki, S. Nagase, D. Jiang, Angew.
Chem. Int. Ed. 2011, 50, 1289–1293; Angew. Chem. 2011, 123,
1325–1329; c) X. Feng, L. Chen, Y. Honsho, O. Saengsawang,
L.-L. Liu, L. Wang, A. Saeki, S. Irle, S. Seki, Y. Gond, D.
Jiang, Adv. Mater. 2012, 24, 3026–3031; d) M. Dogru, M.
Handloser, F. Auras, T. Kunz, D. Medina, A. Hartschuh, P.
Knochel, T. Bein, Angew. Chem. Int. Ed. 2013, 52, 2920–2924;
Angew. Chem. 2013, 125, 2992–2996.
Dienstmaier, D. D. Medina, M. Dogru, P. Knochel, T. Bein,
W. M. Heckl, M. Lackinger, ACS Nano 2012, 6, 7324–7242; g)
J.-T. Yu, Z. Chen, J. Sun, Z.-T. Huang, Q.-Y. Zheng, J. Mater.
Chem. 2012, 22, 5369–5373; h) X. Feng, Y. Dong, D. Jiang,
CrystEngComm 2013, 15, 1508–1511.
K. Kataoka, T. D. James, Y. Kubo, J. Am. Chem. Soc. 2007,
129, 15126–15127.
[14]
[15]
[3]
[4]
a) N. Christinat, R. Scopelliti, K. Severin, Angew. Chem. Int.
Ed. 2008, 120, 1874–1878; Angew. Chem. 2008, 47, 1848–1852;
b) B. Icli, N. Christinat, J. Tönnemann, C. Schüttler, R. Scopel-
liti, K. Severin, J. Am. Chem. Soc. 2009, 131, 3154–3155.
a) N. Nishimura, K. Kobayashi, Angew. Chem. Int. Ed. 2008,
47, 6255–6258; Angew. Chem. 2008, 120, 6251–6254; b) N. Ni-
shimura, K. Yoza, K. Kobayashi, J. Am. Chem. Soc. 2010, 132,
777–790; c) N. Nishimura, K. Kobayashi, J. Org. Chem. 2010,
75, 6079–6085; d) M. Mitsui, K. Higashi, R. Takahashi, Y.
Hirumi, K. Kobayashi, Photochem. Photobiol. Sci. 2014, 13,
1130–1136.
a) N. Iwasawa, H. Takahagi, J. Am. Chem. Soc. 2007, 129,
7754–7755; b) H. Takahagi, S. Fujibe, N. Iwasawa, Chem. Eur.
J. 2009, 15, 13327–13330; c) S. Ito, k. Ono, N. Iwasawa, J. Am.
Chem. Soc. 2012, 134, 13962–13965; d) S. Ito, H. Takata, K.
Ono, N. Iwasawa, Angew. Chem. Int. Ed. 2013, 52, 11045–
11048; Angew. Chem. 2013, 125, 11251–11254.
a) G. Zhang, O. Presly, F. White, I. M. Oppel, M. Mastalerz,
Angew. Chem. Int. Ed. 2014, 126, 1542–1546; Angew. Chem.
2014, 126, 1516–1520; b) G. Zhang, O. Presly, F. White, I. Op-
pel, M. Mastalerz, Angew. Chem. Int. Ed. 2014, 53, 5126–5130;
Angew. Chem. 2014, 126, 5226–5230.
T.-H. Chen, W. Kaveevivitchai, N. Bui, O. S. Miljanic´, Chem.
Commun. 2012, 48, 2855–2857.
For representative examples of the use of B–N coordination
in self-assembled boronate ester materials, see: a) H. Höpfl,
J. Organomet. Chem. 1999, 581, 129–149; b) N. Christinat, R.
Scopelliti, K. Severin, Chem. Commun. 2004, 1158–1159; c) V.
Barba, I. Betanzos, J. Organomet. Chem. 2007, 692, 4903–4908;
d) N. Christinat, R. Scopelliti, K. Severin, J. Org. Chem. 2007,
72, 2192–2200; e) N. Christinat, R. Scopelliti, K. Severin,
Chem. Commun. 2008, 3660–3662; f) M. Hutin, G. Bernardi-
nelli, J. R. Nitschke, Chem. Eur. J. 2008, 14, 4585–4593.
For recent reviews of shape-persistent polygons and polyhedra,
see: a) S. Höger, Chem. Eur. J. 2004, 10, 1320–1329; b) W.
Zhang, J. S. Moore, Angew. Chem. Int. Ed. 2006, 45, 4416–
4439; Angew. Chem. 2006, 118, 4524–4548; c) G. Zhang, M.
Mastalerz, Chem. Soc. Rev. 2014, 43, 1943–1947.
[16]
[17]
[18]
[5]
[6]
[7]
[19]
[20]
[8]
[9]
X. Feng, L. L. Lui, Y. Honsho, A. Saeki, S. Seki, S. Irle, Y. P.
Dong, A. Nagai, D. L. Jaing, Angew. Chem. Int. Ed. 2012, 51,
2618–2622; Angew. Chem. 2012, 124, 2672–2676.
[21]
[22]
a) S.-Y. Ding, J. Gao, Q. Wang, Y. Zhang, W.-G. Song, C.-Y.
Su, W. Wang, J. Am. Chem. Soc. 2011, 133, 19816–19822; b)
Y. Matsushima, R. Nishiyabu, N. Takanashi, M. Haruta, H.
Kimura, Y. Kubo, J. Mater. Chem. 2012, 22, 24124–24131.
a) S. S. Han, H. Furukawa, O. M. Yaghi, W. A. Goddard, J.
Am. Chem. Soc. 2008, 130, 11580–11581; b) D. P. Cao, J. H.
Lan, W. C. Wang, B. Smit, Angew. Chem. Int. Ed. 2009, 48,
4730–4733; Angew. Chem. 2009, 121, 4824–4827; c) H. Furu-
kawa, O. M. Yaghi, J. Am. Chem. Soc. 2009, 131, 8875–8883.
O. M. Yaghi, M. O’ Keefe, N. Ockwig, H. Chae, M. Eddaoudi,
J. Kim, Nature 2003, 423, 705–714.
a) N. Fujita, S. Shinkai, T. D. James, Chem. Asian J. 2008, 3,
1076–1091; b) K. Severin, Dalton Trans. 2009, 5254–5264; c)
R. Nishiyabu, Y. Kubo, T. D. James, J. S. Fossey, Chem. Com-
mun. 2011, 47, 1124–1150; d) S. D. Bull, M. G. Davidson, J. M.
van den Elsen, J. S. Fossey, A. T. A. Jenkins, Y.-B. Jiang, Y.
Kubo, F. Marken, K. Sakurai, J. Zhao, T. D. James, Acc. Chem.
Res. 2013, 46, 312–326.
a) R. W. Tilford, W. R. Gemmill, H.-G. zur Loye, J. J. Lavigne,
Chem. Mater. 2006, 18, 5296–5301; b) A. P. Côté, H. M. El-
Kaderi, H. Furukawa, J. R. Hunt, O. M. Yaghi, J. Am. Chem.
Soc. 2007, 129, 12914–12915; c) H. M. El-Kaderi, J. R. Hunt,
J. L. Mendoza-Cortés, A. P. Côté, R. E. Taylor, M. O’Keeffe,
O. M. Yaghi, Science 2007, 316, 268–272; d) R. W. Tilford, S. J.
Magavero, P. J. Pellechia, J. J. Lavigne, Adv. Mater. 2008, 20,
2741–2746; e) M. Dogru, A. Sonnauer, A. Gavryushin, P. Kno-
chel, T. Bein, Chem. Commun. 2011, 47, 1707–1709; f) J. F.
L. A. Tatum, C. J. Johnson, A. A. P. Fernando, B. C. Ruch,
K. K. Barakoti, M. A. Alpuche-Aviles, B. T. King, Chem. Sci.
2012, 3, 3261–3264.
[10]
[23]
[24]
T. Wöhrle, A. Baro, S. Laschat, Materials 2014, 7, 4045–4056.
Several methods have been reported for the preparation of
COFs on solid surfaces, including ultrahigh vacuum deposi-
tion, see: a) N. A. A. Zwaneveld, R. Pawlak, M. Abel, D. Cata-
lin, D. Gigmes, D. Bertin, L. Porte, J. Am. Chem. Soc. 2008,
130, 6678–6679; b) S. Clair, M. Able, L. Porte, Chem. Commun.
2014, 50, 9627–9635; for Ullmann coupling, see: c) T. Faury, S.
Clair, M. Abel, F. Dumur, D. Gigmes, L. Porte, J. Phys. Chem.
C 2012, 116, 4819–4823; for drop-casting followed by thermal
dehydration, see: d) J. F. Dienstmaier, A. M. Gigler, A. J. Go-
etz, P. Knochel, T. Bein, A. Lyapin, S. Reichlmaier, W. M.
Heckl, M. Lackinger, ACS Nano 2011, 5, 9737–9745; e) C.-Z.
Guan, D. Wang, L.-J. Wan, Chem. Commun. 2012, 48, 2943–
2945; f) J. F. Dienstmaier, D. D. Medina, M. Dogru, P. Kno-
chel, T. Bein, M. Heckl, M. Lackinger, ACS Nano 2012, 6,
7234–7242; and under solvothermal conditions: g) J. W. Col-
son, A. R. Woll, A. Mukherjee, M. P. Levendorf, E. L. Spitler,
V. B. Shields, M. G. Spencer, J. Park, W. R. Dichtel, Science
[11]
[12]
[13]
2011, 332, 228–231; and ref.[6b]
.
[25]
See, for example: a) W. Niu, B. Rambo, M. D. Smith, J. J. La-
vigne, Chem. Commun. 2005, 5166–5168; b) W. Niu, M. D.
Smith, J. J. Lavigne, Cryst. Growth Des. 2006, 6, 1274–1277.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
13

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
M. K. Smith, A. R. Goldberg, B. H. Northrop
FULL PAPER
[26] Three distinct covalent nitroso polymer networks have been
solved by single-crystal X-ray diffraction, see: D. Beaudoin, T.
Maris, J. D. Wuest, Nat. Chem. 2013, 5, 830–834.
[27] Single-crystals of an imine-based framework have been solved
using 3D electron diffraction analysis applying the method of
rotation electron diffraction (RED) for data collection, see: Y.-
B. Zhang, J. Su, H. Furukawa, Y. Yun, F. Gándara, A. Duong,
X. Zou, O. M. Yaghi, J. Am. Chem. Soc. 2013, 135, 16336–
16339.
J. Gebbink, G. van Koten, Eur. J. Org. Chem. 2007, 18, 2931–
2939; c) P. Dechambenoit, S. Ferlay, B. Donnio, D. Guillon,
M. W. Hosseini, Chem. Commun. 2011, 47, 734–736.
[32] M. K. Smith, B. H. Northrop, Chem. Mater. 2014, 26, 3781–
3795.
[33] Recent research by Dichtel et al. has suggested that the forma-
tion of 2D COFs may proceed by a templated polymerization
mechanism largely driven by intermolecular interactions
wherein growing COF oligomers template the addition and re-
[28] B. J. Smith, W. R. Dichtel, J. Am. Chem. Soc. 2014, 136, 8783–
action of additional monomers, see ref.[28]
.
8789.
[34] V. Percec, M. Mihai, Y. Tsuda, B. M. Rosen, S. Uchida, M. R.
[29] M. K. Smith, N. P. Powers-Riggs, B. H. Northrop, Chem. Com-
mun. 2013, 49, 6167–6169.
[30] L. M. Lanni, R. W. Tilford, M. Bharathy, J. J. Lavigne, J. Am.
Chem. Soc. 2011, 133, 13975–13983.
Imam, G. Ungar, P. A. Heiney, Chem. Eur. J. 2009, 15, 8994–
9004.
[35] B. Boden, J. Hui, M. MacLachlan, J. Org. Chem. 2008, 73,
8069–8072.
[31] a) K. Tanabe, T. Yasuda, M. Yoshio, T. Kato, Org. Lett. 2007,
9, 4271–4274; b) R. van de Coevering, P. Bruijnincx, C. Walree,
Received: February 5, 2015
Published Online: ᭿
14
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O50170
/KAP1
Date: 12-03-15 17:52:36
Pages: 15
Dynamic Assembly of Covalent Organic Polygons
Covalent Organic Polygons
Getting in and out of the library: Six phen-
anthrene-based bis(catechol) derivatives
are reported and their ability to undergo
dynamic covalent self-assembly with
benzenediboronic acid is investigated. The
influence of different solubilizing groups at
the 9- and 10-positions of phenanthrene,
the effects of greater or lesser π-surface
area, and the influence of hydrophobic and
protic solvents on the efficacy of self-as-
sembly are each investigated.
M. K. Smith, A. R. Goldberg,
B. H. Northrop* .............................. 1–15
The Dynamic Assembly of Covalent Or-
ganic Polygons: Finding the Optimal Bal-
ance of Solubility, Functionality, and Stab-
ility
Keywords: Self-assembly / Macrocycles /
Borates
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
15