Decorating the Edges of a 2D Polymer with a Fluorescence Label

Article abstract of DOI:10.1021/jacs.6b05456

This work proves the existence and chemical addressability of defined edge groups of a 2D polymer. Pseudohexagonally prismatic single crystals consisting of layered stacks of a 2D polymer are used. They should expose anthracene-based edge groups at the six (100) but not at the two pseudohexagonal (001) and (001) faces. The crystals are reacted with the isotopically enriched dienophiles maleic anhydride and a C18-alkyl chain-modified maleimide. In both cases the corresponding Diels-Alder adducts between these reagents and the edge groups are formed as confirmed by solid state NMR spectroscopy. The same applies to a maleimide derivative carrying a BODIPY dye which was chosen for its fluorescence to be out of the range of the self-fluorescence of the 2D polymer crystals stemming from contained template molecules. If the crystals are excited at λ = 633 nm, their (100) faces and thus their rims fluoresce brightly, while the pseudohexagonal faces remain silent. This is visible when the crystals lie on a pseudohexagonal face. Lambda-mode laser scanning microscopy confirms this fluorescence to originate from the BODIPY dye. Micromechanical exfoliation of the dye-modified crystals results in thinner sheet packages which still exhibit BODIPY fluorescence right at the rim of these packages. This work establishes the chemical nature of the edge groups of a 2D polymer and is also the first implementation of an edge group modification similar to end group modifications of linear polymers.

Full text of DOI:10.1021/jacs.6b05456

Subscriber access provided by The Library | University of Bath
Article
Decorating the Edges of a 2D Polymer with a Fluorescence Label
Yingjie Zhao, Richard Heinrich Maria Bernitzky, Max J.
Kory, Gregor Hofer, Johan Hofkens, and A. Dieter Schlüter
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b05456 • Publication Date (Web): 27 Jun 2016
Downloaded from http://pubs.acs.org on July 4, 2016
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 free 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 accessible to all
readers and 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.
Journal of the American Chemical Society 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 7
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Decorating the Edges of a 2D Polymer with a Fluorescence
Label
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
Yingjie Zhao,† Richard Bernitzky,† Max J. Kory,† Gregor Hofer,† Johan Hofkens,‡ A. Dieter Schlüter*,†
†Institute of Polymers, Department of Materials, Swiss Federal Institute of Technology, ETH Zürich, Vladimir Prelog-Weg
5, 8093 Zürich, Switzerland
‡Department of Chemistry, Divison Molecular Imaging and Photonics, KU Leuven, Celestijnenlaan 200 F, 3001 Heverlee
Belgium
ABSTRACT: This work proves the existence and chemical addressability of defined edge groups of a 2D polymer. Pseudohexago-
nally prismatic single crystals consisting of layered stacks of a 2D polymer are used. They should expose anthracene-based edge
groups at the six (100) but not at the two pseudohexagonal (001) and (00-1) faces. The crystals are reacted with the isotopically
enriched dienophiles maleic anhydride and a C18-alkyl chain-modified maleimide. In both cases the corresponding Diels-Alder
adducts between these reagents and the edge groups are formed as confirmed by solid state NMR spectroscopy. The same applies to
a maleimide derivative carrying a BODIPY dye which was chosen for its fluorescence to be out of the range of the self-
fluorescence of the 2D polymer crystals stemming from contained template molecules. If the crystals are excited at λ= 633 nm their
(100) faces and thus their rims fluoresce brightly while the pseudohexagonal faces remain silent. This is visible when the crystals
lie on a pseudohexagonal face. Lambda mode laser scanning microscopy confirms this fluorescence to originate from the BODIPY
dye. Micromechanical exfoliation of the dye-modified crystals results in thinner sheet packages which still exhibit BODIPY fluo-
rescence right at the rim of these packages. This work establishes the chemical nature of the edge groups of a 2D polymer and is
also the first implementation of an edge group modification similar to end group modifications of linear polymers.
recently reported the rotor-shaped, anthracene-based monomer
1 and the hexaprismatic 2D polymer crystals obtained from it
INTRODUCTION
2D polymers and other 2D materials currently attract consid-
erable attention because they exhibit properties that—in re-
flecting their dimensionality—are different from common 3D
materials.1,2 One way to synthesize 2D polymers rests upon the
so-called single crystal approach.3 In this approach trifunc-
tional monomers are pre-organized in layered crystals such
that the reactive groups of neighboring monomers, typically
anthracenes, form tight pairs [face-to-face (ftf) stacking]. A
photochemical stimulus then brings about the reaction be-
tween them. As irradiation continues, eventually all pairs are
converted into the covalent crosslinks of the two-dimensional
networks that form in each layer of the crystal.3 The final
crystal is then composed of an array of a huge number of 2D
polymer sheets placed on top of one another typically in the
crystallographic c direction. Ideally this geometrically con-
fined polymerization proceeds in a topochemical fashion.4
This means that the internal changes associated with the reac-
tion do not prevent a single crystal to be converted into anoth-
er single crystal. In such a case, structure analysis of the re-
sulting 2D polymer is particularly easy because XRD can be
applied.3c,d
by photochemically triggered dimerization (λ=465 nm) of ftf
stacked anthracene units of neighboring monomers in the
monomer crystals (Figure 1a). The monomers are held in place
by the same compound 1, which also acts as template (mono-
mer:template = 2:1), and by solvent molecules.3d The polymer-
ization can be pushed to full conversion whereby all internal
functional groups of the monomers (not those of the template
molecules though) are converted into crosslinks, while the
groups that happen to reside at the edge of the monomer crys-
tal should remain unchanged, caused by the lack of coupling
partner (Figure 1a, blue circle). For these crystals we use the
abbreviation C. The presentation of anthracene edge groups at
some crystal surfaces in principle offers the opportunity to
prove their presence by selective modification before or after
exfoliation of the 2D polymer single crystal into sheet packag-
es and single sheets.
End groups of linear polymers play a fundamentally im-
portant role. Not only are they of importance for molar mass
determination and the prevention of un-zipping reactions but
also they may provide surface adhesion, compatibility in
blends, and solubility. Furthermore, they are the key units for
the synthesis of blockcopolymers. Thus it is a matter of great
interest to not only create 2D polymers but also to explore
whether these sheet-like entities carry the proposed edge
groups and to show that these edge groups can in fact be
chemically modified in a pre-determined fashion. While this
will widen the impact of 2D polymers as it happened for linear
polymers, it will also provide an opportunity to influence the
exfoliation process, which sometimes can be challenging.
The term “2D polymer” is currently under discussion but it
seems that an increasing number of laboratories accept the
definition which was put forward a couple of years back and
refined recently.1,2 According to it, 2D polymers are monolay-
er covalent sheets with a tiled structure. The tiles are the topo-
logically planar repeat units. At the edges of a sheet are end
groups, which—in differentiation from the well-known end
groups of linear polymers—we refer to as edge groups. We
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 7
However, given the extremely high molar masses expected for Chemical modifications of crystalline materials are not un-
1
2
3
4
5
6
7
8
2D polymers (estimated in the range of billions of Da), the
ratio between edge groups and repeat units will be very low
and analytics are likely to turn complicated. If a goal such as
edge modification is to be pursued, for handling reasons it
seemed easier to work with sheet crystals rather than single
sheets or sheet packages. However, aside from the concentra-
tion issue there is also the additional complication that the
exact molecular-scale structure at the crystal faces is not nor-
mally known. Even highest quality XRD data does not provide
information on holes, protrusions, terraces and caves. One can
therefore not expect to obtain conversion numbers when
chemically addressing sheet edges.
common. They have been routinely performed to modify
faces, channels and pore walls of zeolites6 and of metal-
organic frameworks (MOFs),7 for example. For such purposes
fluorescence dyes have also been used.8 Crystals of synthet-
ic9,10 and natural polymers11,12 have been modified at their
surface both in face-selective and unselective fashion. Modifi-
cations of the surfaces of organic crystals, however, have
practically not been reported, as there are no reported attempts
to modify 2D polymers at their edges. The work by Ciscek et
al. is a noticeable exception regarding the former, who modi-
fied tetracene and rubrene single crystal surfaces by using
Diels-Alder (DA) chemistry with a collection of dienophiles
including maleic acid anhydride (MAH) and N-methyl malei-
mide.13 Another interesting example regards a COF microcrys-
tal the surface of which was decorated with a fluorescence
tag.14 For comprehensiveness, we note that the main planes of
nano-sized covalent sheets by Kim et al. have been decorated
via host-guest interactions and covalent bonds.15 In the present
context it is also worth noting that edge modifications of mon-
odisperse nanographenes have recently been published.16
They, however, involve comparably small entities (few nm
range) and have been carried out in solution. Finally, we note
that whenever surfaces are involved in chemical reactions the
issue of unspecific physisorption versus chemical reaction
arises. The experiments performed have therefore to be select-
ed such that they can differentiate these possibilities and une-
quivocally prove that chemical reaction has taken place.
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
We here report the chemical modification of the fully pol-
ymerized crystals C with two different average sizes (1-2 ꢀm:
Csmall and 10-20 ꢀm: Clarge) with the powerful dienophiles
maleic anhydride (MAH) and N-octadecylmaleimide (OMI)
(Figure 1b). The crystals are pseudohexagonally shaped with
the crystallographic c-direction perpendicular to the hexagonal
faces. 2D polymer sheets are arranged perpendicular to the c-
axis and thus the edge groups are exposed at the six (100)
crystal faces. The different sizes of C were chosen because
they have different volume to surface ratios which allows
cross-checking analytical results obtained on the crystals by
whether or not they reflect this ratio. For the modifications
fully polymerized crystals were used and they were first car-
ried out with MAH and OMI to prove the feasibility of the
anticipated chemical reaction. To tackle the sensitivity issue,
both components were used in 13C-isotopically labelled form
(at C-2 and C-3) (Figure 1b) which should result in a signal
enhancement of a factor of 10’000. In this context it is worth
noting that the ratio between edge groups and internal RUs is
1: 5000 for Clarge (calcd for 20 ꢀm) and 1: 600 for Csmall (calcd
for 2 ꢀm). The NMR signals for the decorated Csmall are there-
fore expected to rise considerably beyond the regular signals
while for Clarge they should appear almost as normal.
Figure 1. (a) Single-crystal approach employed for synthesis of
2D polymer. Note that the voids between the monomers are filled
with template and solvent molecules which are omitted here for
clarity. A third of the monomer molecules serve as unreactive
templates. (b) Synthetic scheme for surface functionalization of
the 2D polymer crystal through Diels–Alder reaction. Note that
the anthracene units shown refer to one of the many sheets the
crystal is composed of. The individual layers of the crystals are
slightly shifted to one another which is meant to emphasize the
stacking of sheets but has no crystallographic resemblance. (c)
Synthetic scheme for surface modification of the 2D polymer
crystal at its six (100) faces by using a maleimide reagent with
BODIPY dye. (d, e) Schematic of the MAH modified 2D polymer
crystal from the top (d) and from the side (e). Color code: green,
bulk RU or template; light red: outermost RU or template at the
crystal (100) face the initially active edge anthracene unit of
which (dark red) has reacted with MAH (blue).
RESULTS AND DISCUSSION
The crystals obtained after modification were analyzed by
13C CP/MAS NMR spectroscopy to confirm the bond for-
mation as required for the proposed DA reaction. After this
initial proof of reactivity, we concentrated on attaching a fluo-
rescence label instead of simple dienophiles (Figure 1c). This
was attempted by the same DA chemistry for which purpose
the near-infrared N-substituted maleimide boron dipyrrome-
thene (BODIPY) dye, 2, was synthesized (Scheme S1). This
dye was chosen because it emits at λ = 660 nm which is out of
ACS Paragon Plus Environment

Page 3 of 7
Journal of the American Chemical Society
MAH were put into a sealed tube under argon and heated to
120 C for 12 h. Working in neat reagent ensures maximum
the fluorescence range of C tailing up to λ = 640 nm. This
long wavelength emission of C is not caused by the 2D poly-
mer within C but rather by the unconsumed template mole-
cules which fill the voids of the 2D polymer inside the crystal.
o
1
2
3
4
5
6
7
8
concentration of the dienophile in this heterogeneous reaction.
Thereafter the crystals were washed five times each with
CH2Cl2, diethylether, and methanol and then dried under vac-
uum. The crystals were examined by optical microscopy be-
fore and after the process and found to remain unchanged
(Figures S2, S3). Thus neither did the MAH serve as exfolia-
tion agent nor did the conditions lead to any destruction of the
crystals. To prove the reaction between MAH and edge an-
thracenes the crystals were investigated by IR spectroscopy.
However, even for Csmall not only was the signal due to the
anhydride moiety of MAH very weak (Figure S4) but also it
could not be differentiated whether it was due to covalently
bonded MAH. MAH could alternatively have been phy-
sisorbed to the crystal surface or taken up into the crystal.
Similar results were obtained with TGA-MS and pyrolysis-
GC/MS (Figures S5-7). Extensive washings did not have an
impact.
The dye-modified crystals were investigated by confocal la-
ser scanning microscopy (CLSM) equipped for recording
lambda scans. A lambda scan records a series of individual
images, each image at a specific wavelength, within a wave-
length range defined by the user. When applied to a particular
point in the image, the emission spectrum at that position can
be recorded if one scans in small enough steps. Here measur-
ing in the lambda mode would allow confirming the expected
exclusive edge modification at the (100) faces by rim-selective
fluorescence emission. In order to differentiate unspecific
adsorption and covalent connection of the BODIPY-
containing maleimide reagent, this reagent was also synthe-
sized and used with 13C-isotopically enriched C-2 and C-3
position. The formation of the expected DA adduct was con-
firmed by solid state NMR spectroscopy (Figures 2g, h).
While decreasing in its intensity, the rim fluorescence re-
mained visible even for thin sheet packages obtained from
micromechanical exfoliation.
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
The synthesis of the monomer, its crystallization and its
polymerization (λ = 465 nm, 3 h) have been reported.3d,5 The
crystals Clarge were obtained following the published protocol,
while Csmall were obtained from monomer crystals the size of
which was kept small by pouring a hot 2-cyanopyridine solu-
tion of the monomer directly into methanol. The polymeriza-
tion conversions were proven to be quantitative by XRD
amongst other methods. While the shape of the large crystals
was uniformly pseudohexagonal, the small crystals were less
regular because of the abrupt quenching process.
As mentioned above, the precise molecular structure of the
outermost layer of crystals which form the crystal faces is not
normally known. High resolution AFM imaging can in princi-
ple shed some light onto this. Since one of the intentions of the
present work was to attach a bulky fluorescence dye to 2D
polymer sheet edges that form a crystal face, we were interest-
ed to at least qualitatively assessing the accessibility of the
(anthracene) edge groups in C. We therefore constructed a
molecular model of a reactive (100) face of C from XRD data,
two perspectives of which are shown in Figures 1d and 1e
with the anthracenes having reacted with parent MAH (for
simplicity). This concerns all anthracenes irrespective of
whether they are edge groups of the 2D polymer or rather
belong to the not covalently bound template. As can be seen
from the representation of three consecutive sheets, there is
not much steric crowding around the DA adducts which made
us optimistic regarding the key reaction, which was the at-
tachment of the dye. Note that this model is idealized and it
may well be that the faces contain holes and caves and it is by
no means clear that on a real crystal all edge groups can in fact
be reached by a reagent. In particular it is unknown whether
the template molecules right at the edge are actually present or
“fall out” of the crystal leaving empty niches behind. For
graphical representations of the (100) [and (00-1)] surfaces of
the crystals without template molecules, see Figure S1. Be-
cause of the absence of through-pores only the outermost
template molecules can fall out. The remaining large majority
of templates can only leave the crystal during exfoliation.
Figure 2. 13C CP/MAS spectra of (a) Csmall; (b) Clarge; (c) Csmall
after reaction with MAH-2,3-13C2; (d) Clarge after reaction with
MAH-2,3-13C2; (e) Csmall after reaction with OMI-2,3-13C2; (f)
Clarge after reaction with OMI-2,3-13C2; (g) Csmall after reaction
with 2-2,3-13C2; (h) Clarge after reaction with 2-2,3-13C2.
To clarify this important issue unequivocally, we resorted to
13C CP/MAS NMR spectroscopy using isotopically labeled
MAH and now also OMI reagents (labelled at C-2 and C-3).
Reactions and work-up were performed as in the case of unla-
beled MAH. Figure 2 shows the results. For comparison rea-
sons Figures 2a and b contain the solid state NMR spectra of
the parent crystals Csmall and Clarge. Figures 2c and d show the
spectra of the reaction products with labeled MAH. Two sig-
nals at δ = 135 ppm (α) and at δ = 47 ppm (β) are immediately
apparent. Interestingly for both the large and the small crystals
the low field signal α appears in a constant intensity ratio to all
other signals, while the intensity of the high field signal β
strongly depends on crystal size. This suggests that α is due to
MAH somehow taken up into the crystal, while β is due to the
expected MAH DA adduct. This latter assignment was con-
firmed by model studies (for details, see Chapter S1e) and is in
line with reported shifts of such DA adducts.17 Why and how
MAH is taken up into the crystals is not yet understood.
Besides these two signals there also appeared a broad low
intense signal at δ = 35 ppm (γ) which has not yet been as-
signed. Figures 1e and f present the outcome with labeled
The first modification reactions were performed with MAH.
50 mg amounts of crystals Csmall and Clarge and 500 mg of
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 7
OMI. In contrast to MAH there is neither an indication for crystals emit from their rims only (e). Overview images are
1
2
3
4
5
6
7
8
incorporation into the crystal nor for unspecific physisorption,
while there is again a clear signature for the anticipated DA
reaction due to a signal at δ = 46 ppm which is again referred
to as β. The high field signal at δ = 29 ppm (δ) is caused by
the C18 alkyl chain. The small peak at δ = 65 ppm (ε) has not
been assigned. The signals from 35 ppm to 5 ppm and at δ =
22 ppm in Figure 2c and d are due to the spinning side bands
and are marked (∗).
available in Figure S8. Figure 3f shows the same section in
transmission mode. While image (e) points in the desired
direction, it was nevertheless necessary to confirm that the
emission spectrum truly belongs to the dye. We therefore
applied the lambda mode of the CLSM for unmodified crystals
(g) and modified crystals at the two different excitation wave-
lengths λ = 488 nm (h) and λ = 633 nm (i). While the excita-
tion at 488 nm already shows strong emission from the rim
together with bulk emission from the crystal, excitation at 633
nm clearly makes only the rims to appear. Figure 3j finally
shows the fluorescence spectra taken at the indicated positions
1-4 in images (g) and (h) together with the solution spectrum
of the dye molecule 2 in CH2Cl2. The spectra taken at position
1 and taken of the dye in solution show virtually the same
maximum emission wavelength with the former showing
broader signals owing to the limited resolution of the lambda
mode recording.
9
There are three main conclusions to be drawn from these
spectra. First, there is clear evidence for edge anthracene units
to be able to react with both dienophiles MAH and OMI in the
expected DA fashion. The signals of the corresponding DA
adducts at δ = 47 ppm and δ = 46 ppm are in fact observed.
Second, the different sizes in starting crystals are reflected in
the signal intensities of the products. Thus, for both MAH and
OMI the intensities of their DA adduct signals are much larger
for the small crystals than for the large ones. Third, the signal
intensities of the OMI adducts to edge anthracenes are lower
than for the corresponding MAH adducts for both crystal
sizes. While this is not in agreement to the observed reactivity
differences between MAH and parent maleimides,18 it may
well reflect the fact that OMI carries a large substituent, the
C18 chain. This may not only reduce the reactivity of this
dienophile in general but also in regard to accessing edge
anthracenes “hidden” in niches. If this interpretation holds true
it would point towards conversion limitations associated with
reagent size, but it is clearly too early to expand on this aspect.
The reason why seemingly only MAH is taken up into the
crystals may have something to do with its small size.
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
Next we turned our attention to the edge-selective modifica-
tion of 2D polymers with a fluorophore. On the basis of the
findings presented above, the near-infrared fluorescent dye 2,
based on BODIPY carrying a maleimide linking unit, was
synthesized (Scheme S1) and reacted with the crystals Clarge
(Figure 1c). The reaction conditions were similar except for
using the dye in toluene solution (60 mM) rather than neat.
The initial fluorescence studies were carried out using the dye-
modified crystals as obtained, followed by studies using mi-
cromechanically cleaved material. Two aspects were first
looked into. Do crystals that happen to lie on the substrate
with their hexagonal faces fluoresce at the rims only and does
the wavelength of the emission allow an unequivocal assign-
ment to the dye rather than a template?
Figure 3 shows typical images obtained using a CLSM with
an option for the lambda mode. The images (a-c) refer to un-
modified crystals. At an excitation wavelength of λ = 488 nm
all crystals emit light at all faces (a). At an excitation wave-
length of λ = 633 nm there is no visible emission which is a
clear indication that the template fluorophores are not excited
at this wavelength (b). Figure 3c represents the same section in
transmission mode. In images (a) and (c) a crystal is highlight-
ed by a white circle that fell over and does not present its
pseudohexagonal face therefore but rather the (100) faces with
all the edge groups. Comparing the image in transmission
mode (c) of this crystal with the fluorescence mode in (a) the
features do not have the same shape. This is a consequence of
working at the focal plane of flat laying crystals.
Figure 3. CLSM images of unmodified crystals excited at 488 nm
(a) and at 633 nm (b). (c) Transmission image of unmodified
crystal. CLSM images of modified crystals excited at 488 nm (d)
and at 633 nm (e). (f) Transmission image of modified crystal. (g)
Lambda scanning images (resolution: 9 nm) of unmodified crys-
tals excited at 488 nm (g), and of BODIPY dye 2-modified crys-
tals excited at 488 nm (h) and at 633 nm (i). (j) Uncorrected fluo-
rescence spectra taken at the positions 1, 2, 3, 4 marked in (g) and
(h) under CLSM conditions and of the BODIPY dye 2 in CH2Cl2
(excited at 488 nm).
The images (d-f) refer to BODIPY-modified crystals. When
excited at λ = 488 nm again the crystals emit from all faces
(d). When excited at λ = 633 however the horizontally lying
In a control experiment to check whether there is non-
specific adsorption of the dye-containing reagent, the unmodi-
ACS Paragon Plus Environment

Page 5 of 7
Journal of the American Chemical Society
fied crystals were stirred in toluene with the reagent under An example may illustrate what we believe should now be
1
conditions (25°C) where the DA reaction takes place either
very slowly or not at all. Interestingly, fluorescence analysis
also showed bright rims albeit in much lower intensity. It was
thus necessary in this case to also unequivocally confirm the
covalent connection that the above experiments had suggested.
For this purpose the 13C-isotopically labelled form of the
BODIPY dye reagent 2-2,3-13C2 was synthesized (Scheme S1)
and reacted with the crystals under the same conditions as had
been applied for non-isotopically enriched dye-reagent (120
°C in toluene in a sealed tube). The13C CP/MAS spectra of
Csmall shows a signal at δ = 46 ppm which clearly corresponds
to the DA adduct (Figure 2g). As expected, the intensity of
this signal for Clarge is much lower than for Csmall (Figure 2h)
which renders it virtually invisible. This finally proves that the
surface of 2D polymer crystal can be covalently modified by
the BODIPY dye at the expected edge positions. The reason
for the preferred physisorption at the (100) rather than the
pseudohexagonal faces is not known at present. It is interest-
ing to note in this context that the geometry of the holes left
behind by “fallen out” template molecules differs between the
faces. While for the (100) faces the holes are rather niches or
caves, for the hexagonal faces they are more like a well (Fig-
ure S1).
possible: The 2D polymer used in the present study undergoes
depolymerization at about 180 °C. Supposed single sheets or
thin sheet packages of this polymer are treated with a focused
laser of the appropriate wavelength and intensity so as to in-
duce this very depolymerization reaction (retro-[4+4]-
cycloaddition), these objects could be cut into predetermined
shapes and into the obtained objects holes of predetermined
geometry could again be cut. The new rims created by such
processes would expose anthracene units, the products of the
laser-induced retro-cycloaddition reaction. With the chemistry
presented here, these rims could then be decorated at will
employing the tools of organic chemistry. We thus foresee
fascinating opportunities in the combination of 2D polymers
with their defined and addressable molecular structure and
synthetic chemistry. For the fluorescence label decorated
sheets presented here, we further foresee the intriguing oppor-
tunities to investigate the undulation behavior of a 2D polymer
in solution by time-resolved (single molecule) fluorescence
spectroscopy/microscopy once available in single-sheet
form.19 With these sheets one could also contribute to the
visualization of possibly existent grain boundaries (mosaicity)
in the polymer single crystals. Finally, we note that edge mod-
ifications cannot be performed easily with MOFs because of
their being based on equilibrium-type chemistries.20
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
In the next step, the large dye-decorated crystals Clarge dye
,
were subjected to micromechanical exfoliation using Nitto
tape (SPV 224P). The crystals were sprinkled onto this adhe-
sive tape which was then folded onto itself and peeled apart
about 50 times. Subsequent sheet package transfer onto mi-
croscopy cover slips was performed by firmly pressing the
Nitto tape onto a conventional transparent adhesive tape which
then was turned upside down and placed on the cover slid.
This sample was then directly used for CLSM investigations.
As for the crystals, BODIPY fluorescence was observed at the
package rims though with reduced intensity (Figure S9). The
thickness of the packages is difficult to determine in this set-
up. Preliminary estimates by AFM of the thinnest features in
Figure S9 suggest the range of a few 10 nm.
ASSOCIATED CONTENT
Supporting Information
Synthesis; Structure models of crystal faces (100) and (001);
Model studies on crystal stability; Attempts to prove chemical
reaction on polymer single crystal; Overview CLSM images of
covalently dye-labelled polymer single crystal; CLMS images of
micromechanically exfoliated chemically modified 2D polymer
sheet packages. (PDF)
AUTHOR INFORMATION
Corresponding Author
* ads@mat.ethz.ch
CONCLUSION
Notes
The authors declare no competing financial interest
In summary, the dienophiles MAH and OMI can be cova-
lently connected to edge anthracene units presented at the
(100) faces of the 2D polymer single crystals obtained through
photochemically triggered polymerization of single crystals of
monomer 1. The connection proceeds through conventional
DA reaction. This statement is largely based on solid state
NMR spectroscopy using isotopically labeled MAH and OMI
reagents. The same chemistry was further employed to attach a
fluorescence dye to the same faces and thus to the sheet edges.
Accordingly, the crystal rims rather than the bulk of these
crystals emit in the range of the dye used. This finding has
important consequences. First, it proves the proposed chemis-
try (anthracenes) at the edges of the 2D polymer used for
modification. Neither XRD nor analytical techniques such as
IR-, UV/vis-, fluorescence spectroscopy or solid state NMR
spectroscopy allow for identification of unaltered edge groups
because of their extremely low concentration. Second, it
proves that a significant fraction of the edge anthracenes is
actually accessible even to spatially demanding reagents,
although conversions in the classical sense cannot be given.
For this, more detailed insights into the exact molecular struc-
ture of crystals at all their faces would be required including
knowledge about the fate of the topmost template molecules.
ACKNOWLEDGMENT
We thank the ETH Zürich for financial support. ScopeM kindly
provided access to the CSLM. We thank Dr. René Verel for his
competent help with solid state NMR spectroscopy and Prof. Paul
Smith and Jan Vermant for access to their optical microscopes
and Dr. Zhiqiang Ma for help with TGA-MS. We thank Daniel
Messmer and Stan van de Poll for critically reading the manu-
script.
REFERENCES
(1) (a) Sakamoto, J.; van Heijst, J.; Lukin, O.; Schlüter, A. D. An-
gew. Chem. Int. Ed. 2009, 48, 1030-1069; (b) Payamyar, P.; King, B.
T.; Öttinger, H. C.; Schlüter, A. D. Chem. Commun. 2016, 52, 18-34.
(2) (a) Colson, J. W.; Dichtel, W. R. Nat. Chem. 2013, 5, 453-465;
(b) Cai, S.-L.; Zhang, W.-G.; Zuckermann, R. N.; Li, Z.-T.; Zhao, X.;
Liu, Y. Adv. Mater. 2015, 27, 5762-5770; (c) Zhuang, X.; Mai, Y.;
Wu, D.; Zhang, F.; Feng, X. Adv. Mater. 2015, 27, 403-427; (d) Zang,
Y.; Aoki, T.; Teraguchi, M.; Kaneko, T.; Ma, L.; Jia, H. Polym. Rev.
2015, 55, 57-89; (e) Baek, K.; Hwang, I.; Roy, I.; Shetty, D.; Kim, K.
Acc. Chem. Res. 2015, 48, 2221-2229; (f) Xiang, Z.; Cao, D.; Dai, L.
Polym. Chem. 2015, 6, 1896-1911; (g) Rodriguez-San-Miguel, D.;
Amo-Ochoa, P.; Zamora, F. Chem. Commun. 2016, 52, 4113-4127;
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 7
(h) Sakamoto, J.; Shinkai, S. KOBUNSHI RONBUNSHU 2015, 73,
42-54.
1
(3) (a) Kissel, P.; Erni, R.; Schweizer, W. B.; Rossell, M. D.; King,
B. T.; Bauer, T.; Götzinger, S.; Schlüter, A. D.; Sakamoto, J. Nat.
Chem. 2012, 4, 287; (b) Bhola, R.; Payamyar, P.; Murray, D. J.;
Kumar, B.; Teator, A. J.; Schmidt, M. U.; Hammer, S. M.; Saha, A.;
Sakamoto, J.; Schlüter, A. D.; King, B. T. J. Am. Chem. Soc. 2013,
135, 14134; (c) Kissel, P.; Murray, D. J.; Wulftange, W. J.; Catalano,
V. J.; King, B. T. Nat. Chem. 2014, 6, 774; (d) Kory, M. J.; Wörle,
M.; Weber, T.; Payamyar, P.; van de Poll, S. W.; Dshemuchadse, J.;
Trapp, N.; Schlüter, A. D. Nat. Chem. 2014, 6, 779.
2
3
4
5
6
7
8
9
(4) (a) Wegner, G. Z. Naturforsch. 1969, 824-832; (b) Schmidt, G.
M. J. Pure Appl. Chem. 1971, 12, 647-678.
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
(5) Kory, M. J.; Bergeler, M.; Reiher, M.; Schlüter, A. D. Chem.
Eur. J. 2014, 20, 6934-6938.
(6) (a) Wu, H.; Zheng, B.; Zheng, X.; Wang, J.; Yuan, W.; Jiang, Z.
J. Power Sources 2007, 173, 842-852; (b) Arora, M.; Eddy, N. K.;
Mumford, K. A.; Baba, Y.; Perera, J. M.; Stevens, G. W. Cold Reg.
Sci. Technol. 2010, 62, 92-97.
(7) (a) Tanabe, K. K.; Wang, Z.; Cohen, S. M. J. Am. Chem. Soc.
2008, 130, 8508-8517; (b) Goto, Y.; Sato, H.; Shinkai, S.; Sada, K. J.
Am. Chem. Soc. 2008, 130, 14354-14355; (c) Gadzikwa, T.; Farha, O.
K.; Malliakas, C. D.; Kanatzidis, M. G.; Hupp, J. T.; Nguyen, S. T. J.
Am. Chem. Soc. 2009, 131, 13613-13615; (d) Jung, S.; Kim, Y.; Kim,
S.-J.; Kwon, T.-H.; Huh, S.; Park, S. Chem. Commun. 2011, 47, 2904-
2906; (e) Ma, M.; Gross, A.; Zacher, D.; Pinto, A.; Noei, H.; Wang,
Y.; Fischer, R. A.; Metzler-Nolte, N. CrystEngComm 2011, 13, 2828-
2832; (f) Nagata, S.; Sato, H.; Sugikawa, K.; Kokado, K.; Sada, K.
CrystEngComm 2012, 14, 4137-4141; (g) Roy, P.; Schaate, A.;
Behrens, P.; Godt, A. Chem. Eur. J. 2012, 18, 6979-6985; (h) Morris,
W.; Briley, W. E.; Auyeung, E.; Cabezas, M. D.; Mirkin, C. A. J. Am.
Chem. Soc. 2014, 136, 7261-7264.
(8) (a) Yaghi, O. M.; O’Keeffe, M.; Ockwig, N. W.; Chae, H. K.;
Eddadoudi, M.; Kim, J. Nature 2003, 423, 705-714. (b) Ockwig, N.
W.; Delgado-Friedrichs, O.; O’Keeffe, M.; Yaghi, O. M. Acc. Chem.
Res. 2005, 38, 176-182. (c) Kondo, M.; Furukawa, S.; Hirai, K.;
Kitagawa, S. Angew. Chem. Int. Ed. 2010, 49, 5327-5330. (d) Zhou,
H.-C.; Long, J. R.; Yaghi, O. M. Chem. Rev. 2012, 673-674.
(9) Keller, A. Rep. Prog. Phys. 1968, 31, 623.
(10) (a) Peterlin, A.; Meinel, G. J. Polym. Sci., Part B: Polym. Phys.
Let. 1965, 3, 1059-1064; (b) Blundell, D. J.; Keller, A.; Connor, T. M.
J. Polym. Sci., Part B: Polym. Phys. 1967, 5, 991-1012; (c) Fischer,
E. W.; Sterzel, H. J.; Wegner, G. Kolloid-Z. u. Z. Polymere 1973, 251,
980-990.
(11) (a) Moon, R. J.; Martini, A.; Nairn, J.; Simonsen, J.;
Youngblood, J. Chem.Soc. Rev. 2011, 40, 3941-3994; (b) Eyley, S.;
Thielemans, W. Nanoscale 2014, 6, 7764-7779.
(12) (a) Dong, S.; Roman, M. J. Am. Chem. Soc. 2007, 129, 13810-
13811; (b) Nielsen, L. J.; Eyley, S.; Thielemans, W.; Aylott, J. W.
Chem. Commun. 2010, 46, 8929-8931.
(13) Qualizza, B. A.; Prasad, S.; Chiarelli, M. P.; Ciszek, J. W.
Chem. Commun. 2013, 49, 4495-4497.
(14) Calik, M. ; Sick, T. ; Dogru, M. ; Döblinger, M. ; Datz, S.
Budde, H. ; Hartschuh, A. ; Auras, F. ; Bein, T. J. Am. Chem. Soc.
2016, 138, 1234-1239.
(15) Baek, K.; Yun, G.; Kim, Y.; Kim, D.; Hota, R.; Hwang, I.; Xu,
D.; Ko, Y. H.; Gu, G. H.; Suh, J. H.; Park, C. G.; Sung, B. J.; Kim, K.
J. Am. Chem. Soc. 2013, 136, 6523-6528.
(16) Tan, Y.-Z.; Yang, B.; Parvez, K.; Narita, A.; Osella, S.;
Beljonne, D.; Feng, X.; Müllen, K. Nat. Commun. 2013, 4, 2646.
(17) Phutdhawong, W.; Buddhasukh, D.; Pyne, S. G.; Rujiwatra, A.;
Pakawatchai, C. Synth. Commun. 2006, 36, 881-883.
(18) Qualizza, B. A.; Ciszek, J. W. J. Phys. Org. Chem. 2015, 28,
629-634.
(19) Woll, D.; Braeken, E.; Deres, A.; De Schryver, F. C.; Uji-i, H.;
Hofkens, J. Chem. Soc. Rev. 2009, 38, 313-328.
(20) (a) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M.
Science 2013, 341, 1230444. (b) Zhou, H.-C.; Kitagawa, S. Chem.
Soc. Rev. 2014, 43, 5415-5418.
ACS Paragon Plus Environment

Page 7 of 7
Journal of the American Chemical Society
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
7
ACS Paragon Plus Environment