[2 × 2] metallo-supramolecular grids based on 4,6-bis((1H-1,2,3-triazol-4-yl)-pyridin-2-yl)-2-phenylpyrimidine ligands: From discrete [2 × 2] grid structures to star-shaped supramolecular polymeric architectures

Article abstract of DOI:10.1039/d1dt01373d

The self-assembly of bis-tridentate ligands leads to the spontaneous formation of [2 × 2] grid-like metal complexes. However, the synthesis of such ligands is rather cumbersome. In the work, we demonstrate a straightforward synthesis route to prepare bis-tridentate 4,6-bis((1H-1,2,3-triazol-4-yl)-pyridin-2-yl)-2-phenylpyrimidine ligands through double CuAAC click chemistry with 4,6-bis(6-ethynylpyridin-2-yl)-2-phenylpyrimidine as well as their self-assembly into [2 × 2] grid-like metal complexes. In addition, four macromolecular ligands were synthesized starting from azido-end-functionalized poly(2-ethyl-2-oxazoline) (PEtOx) or poly(ethylene glycol) (PEG). These macromolecular ligands were used in the construction of star-shaped supramolecular polymers through complexation with transition metal ions (e.g., Fe²⁺or Zn²⁺). The successful fabrication of complexes and star-shaped polymers was confirmed by UV-vis titration measurements and MALDI-TOF mass spectrometry. However, the chemical structure of the polymer was found to have a strong influence on the [2 × 2] grid formation, which was successful with the PEG-ligands but not with the PEtOx-ligands, while the molecular weight of the PEG did not interfere with grid formation.

Full text of DOI:10.1039/d1dt01373d

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
[2 × 2] metallo-supramolecular grids based on
4,6-bis((1H-1,2,3-triazol-4-yl)-pyridin-2-yl)-2-
phenylpyrimidine ligands: from discrete [2 × 2]
grid structures to star-shaped supramolecular
polymeric architectures†
Cite this: Dalton Trans., 2021, 50,
8746
Xiaowen Xu, a Anne K. Seiffert,a Ruben Lenaerts,a Kanykei Ryskulova,a
Somdeb Jana,a Kristof Van Hecke, b Valentin Victor Jerca c and
a
Richard Hoogenboom
*
The self-assembly of bis-tridentate ligands leads to the spontaneous formation of [2 × 2] grid-like metal
complexes. However, the synthesis of such ligands is rather cumbersome. In the work, we demonstrate a
straightforward synthesis route to prepare bis-tridentate 4,6-bis((1H-1,2,3-triazol-4-yl)-pyridin-2-yl)-2-
phenylpyrimidine ligands through double CuAAC click chemistry with 4,6-bis(6-ethynylpyridin-2-yl)-2-
phenylpyrimidine as well as their self-assembly into [2 × 2] grid-like metal complexes. In addition, four
macromolecular ligands were synthesized starting from azido-end-functionalized poly(2-ethyl-2-oxazo-
line) (PEtOx) or poly(ethylene glycol) (PEG). These macromolecular ligands were used in the construction
of star-shaped supramolecular polymers through complexation with transition metal ions (e.g., Fe2+ or
Zn2+). The successful fabrication of complexes and star-shaped polymers was confirmed by UV-vis titra-
tion measurements and MALDI-TOF mass spectrometry. However, the chemical structure of the polymer
was found to have a strong influence on the [2 × 2] grid formation, which was successful with the PEG-
ligands but not with the PEtOx-ligands, while the molecular weight of the PEG did not interfere with grid
formation.
Received 25th April 2021,
Accepted 26th May 2021
DOI: 10.1039/d1dt01373d
rsc.li/dalton
arrangement of the metal ions, which has been reported to
provide interesting magnetic, electrochemical and optical pro-
Introduction
The self-assembly of complex and multi-component metallo- perties,3 which has been considered for information storage.4
supramolecular architectures leads to interesting materials These well-defined supramolecular grid-like metal complexes
with a variety of tunable and potentially useful characteristics.1 can be straightforwardly fabricated by the self-assembly of
Both polymeric and discrete metallo-supramolecular struc- well-designed linear organic ligands with multiple metal
tures have been reported to show some promising properties, ion binding sites to assemble into [2 × 2] or larger [n × n]
with reported potential applications in catalysis, magnetism, level (n = 3, 4…) grid-like metal complexes, which exhibit intri-
electrochemistry, and photoactivity, as well as gas separation guing physicochemical behaviors and aesthetically appealing
and absorption.2 Grid-like metal complexes provide a unique achitectures.5,6
In the design and synthesis of such grid-like metal com-
plexes, the design of the ligand is crucial as it directs the self-
assembly towards discrete grid-like complexes. In this respect,
various types of organic ligands such as bis(pyridyl)pyrida-
aSupramolecular Chemistry Group, Centre of Macromolecular Chemistry (CMaC),
Department of Organic and Macromolecular Chemistry, Ghent University,
Krijgslaan 281-S4, B-9000 Ghent, Belgium. E-mail: richard.hoogenboom@ugent.be
zine,7 bis(hydrazone),8,9 pyrazolate-bridged binucleating
ligand,10 bis-tridentate (1,2,3-triazol-4-yl)-picolinamide,11 and
bis(bipyridine)pyrimidine,12 have been successfully developed
to prepare [2 × 2] grid-like assemblies, primarily with octa-
hedral coordination geometry, in the presence of transition
metal ions and/or lanthanide ions. Compared to other ligands,
considerable attention has been given to the 4,6-bis(2,2-bipyri-
bXStruct, Department of Chemistry, Ghent University, Krijgslaan 281-S3,
B-9000 Ghent, Belgium
cCentre of Organic Chemistry “Costin D. Nenitzescu” Romanian Academy,
Spl. Independentei 202B, 060023 Bucharest, Romania
†Electronic supplementary information (ESI) available: Full experimental details
and 1H and 13C NMR spectra. Details of the X-ray analysis of compound 3. CCDC
1945081. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/d1dt01373d
8746 | Dalton Trans., 2021, 50, 8746–8751
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
dyl-6-yl)-2-phenylpyrimidine ligands due to the remarkable functionalized poly(2-ethyl-2-oxazoline) (PEtOx) or poly(ethyl-
magnetic,13 optical14 and electronic properties15 of the corres- ene glycol) (PEG). These polymers were chosen due to their
ponding [2 × 2] grid-like complexes in the presence of specific water solubility, non-ionic characteristic, and high flexibility of
metal ions. However, the synthesis of functionalized 4,6-bis the main chain.23 Finally, the self-assembly of the macroli-
(2,2-bipyridyl-6-yl)-2-phenylpyrimidine ligands is rather cum- gands into star-shaped polymer [2 × 2] grid-like metal com-
bersome and the formation of grid like complexes has only plexes in presence of Fe2+ or Zn2+ metal ions will be discussed
been reported with small organic ligands. Ligands functiona- (Scheme 1).
lized with macromolecules to prepare self-assembled star-
shaped metallo-supramolecular polymers have not been
explored to date with such octahedral coordinating grid-
forming ligands, although grid-like metal complexes appear to
Results and discussion
Synthesis and characterization of the 4,6-bis((1H-1,2,3-triazol-
be ideally suited as self-assembled core for the preparation of
4-yl)-pyridin-2-yl)-2-phenylpyrimidine ligands
star-shaped polymers.9,16–19
The synthesis procedure for the 4,6-bis((1H-1,2,3-triazol-4-yl)-
pyridin-2-yl)-2-phenylpyrimidine ligands is outlined in
Scheme 2. The ligands were synthesized in four steps from
commercially available starting compounds (full details are
included in the ESI†). In the first step, compound 1 was
obtained by palladium-catalyzed Sonogashira cross-coupling
of 2,6-dibromopyridine with triisopropylsilyl (TIPS) acety-
The recent advances in the copper-catalyzed azide–alkyne
cycloaddition (CuAAc) click reaction promoted the use of 1,4-
disubstituted-1,2,3-triazoles in coordination chemistry, which
has led to the development of 1,2,3-triazole-containing ana-
logues of bipyridine, terpyridine and 3,6-di(2-pyridiyl)-pyrida-
zine metal coordinating ligands.20–22 The major advantage of
replacing one or two pyridine rings with a 1,2,3-triazole ring is
the much easier synthesis of functional ligands through
CuAAC.
In this work, we hypothesized that the outer pyridine rings
of 4,6-bis(2,2-bipyridyl-6-yl)-2-phenylpyrimidine could be
replaced by 1,2,3-triazole rings providing a much easier syn-
thetic access to bis-tridentate [2 × 2] grid-like metal complex
forming ligands functionalized with polymer chains as basis
for supramolecular star-shaped polymers as shown in
Scheme 1. Therefore, the clickable ligand precursor 4,6-bis(6-
ethynylpyridin-2-yl)-2-phenylpyrimidine was first synthesized,
followed by CuAAC with two low molar mass azido compounds
to demonstrate the synthesis as well as to evaluate whether the
novel bis-tridentate 4,6-bis((1H-1,2,3-triazol-4-yl)-pyridin-2-yl)-
2-phenylpyrimidine ligand can self-assemble into [2 × 2] grid-
like metal complexes in presence of Fe2+ or Zn2+ metal ions.
Next, macromolecular ligands were prepared by CuAAC of 4,6-
bis(6-ethynylpyridin-2-yl)-2-phenylpyrimidine with azido end-
Scheme 2 The synthesis procedure of organic (L1–L2), and macromol-
ecular ligands (L3–L6) used in this study.
Scheme 1 Structure of the 4,6-bis((1H-1,2,3-triazol-4-yl)-pyridin-2-yl)-2-phenylpyrimidine ligands and their complexation into [2 × 2] grid-like
metal complexes with Fe2+ or Zn2+ salts leading to star-shaped polymers in case of macromolecular ligands.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 8746–8751 | 8747

View Article Online
Paper
Dalton Transactions
lene.24 Next, the product 1 was reacted with n-BuLi, and, sub-
sequently, reacted with tributyltin chloride to yield compound
2.24 In the third step, a palladium-catalyzed Stille coupling of
compound 2 was performed with 4,6-dichloro-2-phenylpyrimi-
dine (fenclorim) to obtain compound 3, from which a single
crystal X-ray structure was obtained (ESI; Fig. S5†). The ligands
L1–L6 were prepared in a single final step by in situ de-
protection of the TIPS-groups of 3 with tetrabutylammonium
fluoride (TBAF) and, subsequent, CuAAC click reaction with
the azide derivatives. Intermediate purification after the clea-
vage step was not necessary due to the high yield of the reac-
tion. The PEtOx-functionalized macroligands L3–L4 were puri-
fied using preparative size exclusion chromatography
(Biobeads SX-1), while organic ligands L1–L2 and PEG-functio-
nalized macroligands L5–L6, were purified by silica column
chromatography. The removal of copper from the final pro-
ducts was done by dissolving the ligands in dichloromethane
and washing with a 5 wt% EDTA aqueous solution. The azido-
functionalized PEG and PEtOx were synthesized following pre-
viously reported literature procedures (synthesis details and
full characterization are included in the ESI;† molar mass data
in Table 1).25,26
Fig. 1 (a) 1H-NMR spectrum of ligand L1 (middle, CD3CN), and its
corresponding Fe(BF4)2 complex (top, CD3CN), as well as Zn(BF4)2
complex (bottom, CD3CN); (b) the zoom of the aromatic area.
The structures of the obtained ligands were confirmed by
NMR spectroscopy, and high-resolution mass spectrometry
1
(HRMS) for L1 and L2 respectively. The H-NMR spectra of L1
displayed the characteristic signals in the aromatic region
(peak a, c, e) corresponding to the protons of the pyrimidine,
pyridine rings, and triazole rings, respectively (Fig. 1a middle,
and Fig. S12†). Furthermore, the appearance of two triplets
(peak f and g) assigned to the methylene groups connected to
the triazole ring at 4.71 ppm and 3.96 ppm were observed,
detected by the SEC measurement. Moreover, MALDI-TOF MS
also supported the successful preparation of the macroligands
L3–L6, by a shift of the molar mass distribution to higher,
roughly double molar mass (Fig. S23†). The obtained data
from 1H-NMR spectroscopy, SEC, and MALDI-TOF MS for all
the macroligands and their precursors is summarized in
Table 1. The Mn obtained from 1H-NMR spectroscopy, and
MALDI-TOF MS correspond well with the theoretical Mn, while
the Mn obtained from SEC deviates due to the use of a relative
calibration of the SEC against PMMA standards.
1
respectively. The H-NMR spectra of the other ligands showed
similar chemical shifts indicating the successful ligand for-
mation through CuAAC (ESI†). Further evidence for the quanti-
tative triazole ring formation was obtained from the FTIR spec-
troscopy. As shown in Fig. S20,† the signal corresponding to
the azido moieties at 2101 cm−1 disappeared after the click
reaction.18,27,28 SEC analysis of the purified macromolecular
Complexation of ligands L1 and L2 with Fe2+ and Zn2+ ions
ligands L3, L4, and L6 revealed a monomodal distribution The organic ligand L1 was selected as a model compound to
with a narrow dispersity and a distinctive shift in the retention investigate the complexation behavior with Fe2+ and Zn2+ ions
time compared to the corresponding PEtOx-N3 or PEG-N3 in CD3CN by 1H-NMR spectroscopy. Addition of 0.5 equiva-
(Fig. S21 and S22†). It should be noted that the molecular lents of Fe(BF4)2 to a solution of L1 resulted in an instant color
weight of L5 and its PEG17-N3 precursor was too small to be change from colorless to red. Further addition of the Fe2+ ions
up to 1 equivalent of Fe2+ led to a second color change from
red to blue indicative of the formation of [2 × 2] grid-like iron
(II) complexes.29 Complexation of L1 with an equimolar
Table 1 Number average molecular weight (Mn) and dispersity of the
macromolecular ligands and the polymeric precursors. Note that the
PEG17 derivatives had a too low molar mass to be analysed by SEC
amount of Zn2+ in CD3CN resulted in a yellowish color change.
The signals in the 1H-NMR spectrum of Fe2+–L1 [2 × 2]
complex (Fig. 1a, top) are broader and very small due to the
Polymer
Mn, NMR
Mn, SEC, RI
ĐSEC, RI
Mn, MALDI
paramagnetism of the iron(II) ions, being consistent with the
reported literature for related complexes.18 Nonetheless, a
clear chemical shift of the signals corresponding to the ligand
can be observed supporting the successful formation of octa-
hedral bis(tridentate) complexes with a grid-like structure.
Complexation of L1 with Zn2+ revealed the characteristic
upfield shift of the a, n, and o protons and a downfield shift of
PEtOx20-N3
PEtOx50-N3
PEG17-N3
PEG45-N3
L3
L4
L5
2300
4900
—
3000
8800
—
3900
8800
17 000
—
1.13
1.16
—
1.05
1.17
1.08
—
2208
4675
805
2039
4796
9891
1622
4162
—
4300
9700
1800
5000
L6
6500
1.08
8748 | Dalton Trans., 2021, 50, 8746–8751
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
1
the b, c, d, and m protons in the H-NMR spectrum (Fig. 1a, 10−4 mol L−1 clearly revealed the appearance of the MLCT
bottom and Fig. 1b).30 The elucidation of the Zn2+–L1 [2 × 2] band at 528 nm (Fig. 2a), corresponding to the color change
complex in organic solution started with a 1H NMR spectro- from colorless to red. Further increasing the Fe2+ content to 1
scopic titration with 0.5 and 1.0 equivalents of Zn2+ ions. eq. led to a shift of the MLCT to 658 nm indicating a transition
1
Compared to the relatively simple H NMR spectrum of L1 in from complexes with two ligands and one Fe2+ up to 0.5
CD3CN, the 1H NMR spectrum revealed the presence of two equivalents of Fe2+ into more stable [2 × 2] grid-like complexes
species consisting of ligand L1 and the Zn2+ grid complex from 0.5 to 1 equivalent of Fe2+ ions, accompanied by a color
upon addition of 0.5 equivalent of Zn2+ (Fig. S24,† middle), change from red to blue. Moreover, no significant change of
indicative of high cooperativity for the [2 × 2] grid complex for- the absorption band was observed after adding more than 1
mation and a slow exchange of the formed grid-like complexes equivalent of Fe2+ ions, suggesting that all of the ligands were
on the NMR time-scale. Chemical shift changes of the proton fully complexed and that the formed grids are stable (Fig. 2a,
signals a (Δδ = −0.3 ppm) and n, o (Δδ = −0.8 ppm) as well as inset). The observation that the shift in the MLCT band only
the appearance of a new signal at ∼4.3 ppm suggested the occurs after addition of 0.5 equivalents of Fe2+ indicates that
complete formation of octahedral bis(tridentate) complexes in there is no strong cooperativity for the formation of the [2 × 2]
line with a grid structure upon the increase of Zn2+ to 1.0 equi- complexes as these are only formed after all ligands have been
valent (Fig. S24,† top). To get more insight into the behavior of consumed in complexes with 2 ligands and 1 Fe2+ ion.
grid complex in organic solution, Zn2+ complex L1 was further
Upon titration of Zn2+ ions to a solution of L1 in aceto-
characterized by 2D (1H-NOESY) NMR spectroscopy. The nitrile, bathochromic shifts from 238 nm to 260 nm and from
1H-NOESY spectrum of the Zn2+–L1 complex and free ligand 314 nm to 354 nm were observed, respectively. The titration
L1 in deuterated acetonitrile are shown in Fig. S25.† A corre- shows a clear isosbestic point indicating that a clean transition
lation between the triazole ring proton e and the phenyl ring from ligands to full [2 × 2] grid complexes occur even at low
proton o was only observed for the complex, indicating the suc- equivalents of Zn2+ ions, again indicating high cooperativity
cessful complexation of [2 × 2] grids with Zn2+ ions (Fig. S25† for the [2 × 2] grid formation, which is consistent with the
bottom). No further correlation was found from the spectrum, result of the 1H NMR titration (Fig. S24†). This higher coopera-
indicating that only one single Zn2+ complex species was tivity compared to the complexation of L1 with Fe2+ ions may
present in the system.
result from the lower stability of the Zn2+ complex with one
Furthermore, UV-vis titrations were performed to study the metal ion and two ligands allowing dynamic exchange towards
complexation of L1 with Fe2+ and Zn2+ ions in acetonitrile the thermodynamically most stable grid-like structures.31 The
(Fig. 2 and S26†). The metal ions were added as metal tetra- maximum absorption was reached at a 1 : 1 ratio of Zn2+ ions
fluoroborate salts due to their good solubility in acetonitrile. to L1, which remained constant upon addition of more Zn2+
Upon addition of Fe2+ ions, the absorption maxima of L1 ions, demonstrating the successful formation of stable [2 × 2]
shifted from 238 to 266 nm and from 315 to 358 nm, respect- grid-like complexes (Fig. 2b).
ively. In addition, the metal ligand charge transfer (MLCT)
Next, MALDI-TOF mass spectroscopy was used to investi-
band appeared at 528 nm (Fig. S26†). This MLCT absorbance gate the [2 × 2] grid-like complex formation. The mass peaks
reached a plateau after addition of 1 eq. of Fe2+ ions, indicat- corresponding to {[Fe4L14](BF4)n}(8−n)+ (n = 4–7) structures were
ing the expected stoichiometry for a [2 × 2] grid-like complex. indeed observed in the MALDI-TOF mass spectrum, further
It should be noted that the intensity of the MLCT band is confirming the successful formation of the [2 × 2] grid-like
rather weak due to the low concentration of the solution (e.g., iron complex (Fig. 3a). However, MALDI-TOF-MS of the Zn2+–
2.3 × 10−5 mol L−1). Increasing the concentration of L1 to 3.4 × L1 grid revealed only the molar mass of the complex contain-
ing two ligands with one Zn2+ ion (i.e., [ZnL12]), which might
be caused by the high laser intensity leading to the fragmenta-
tion of the grid (Fig. 3b), suggesting that the grid-like complex
Fig. 2 UV-vis spectroscopy titration experiments: (a) stepwise addition
of 1 µL of 3.4 × 10−2 mol L−1 of solution of Fe(BF4)2 to a 3.4 × 10−4 mol
L−1 of L1 in CH3CN, (b) stepwise addition of 1.9 µL of 1.8 × 10−3 mol L−1
of solution of Zn(BF4)2 to a 2.3 × 10−5 mol L−1 of L1 in CH3CN. The inset
graphs show the changes in the absorption with the increase of the
amount of transition metal ions.
Fig. 3 Partial MALDI-TOF mass spectra of the grid-like complexes of L1
with (a) Fe2+, and (b) Zn2+, respectively.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 8746–8751 | 8749

View Article Online
Paper
Dalton Transactions
of L1 with Fe2+ ions is more stable than the grid-like complex
of ligand L1 with Zn2+ ions as both NMR and UV-vis spec-
troscopy confirmed full complexation of L1 with Zn2+ ions.
Single crystal structures of the Fe2+–L1 grid-like complexes
could unfortunately not be obtained due to the flexible struc-
ture of the ligand. Consequently, we synthesized L2 by repla-
cing the TEG groups with more rigid phenyl rings. However,
also with the more rigid ligand L2 no single crystals could be
obtained, because of the precipitation of the complex from
CH3CN. Nevertheless, 1H-NMR spectroscopy revealed the for-
mation of the [2 × 2] grid-like structure upon addition of 1
equivalent of Fe2+ ions or Zn2+ ions (Fig. S27†). The L2-M2+
complexes showed similar characteristic shifts as previously
discussed for the L1-M2+ complexes (Fig. 1).
After establishing the synthesis of the 4,6-bis((1H-1,2,3-
triazol-4-yl)-pyridin-2-yl)-2-phenylpyrimidine ligands and their
successful self-assembly in grid-like metal complexes with
Zn2+ ions and Fe2+ ions, we focused our attention to the self-
assembly of the polymeric macroligands L3–L6 to investigate
the formation of supramolecular star-polymers through [2 × 2]
grid-like metal complex formation. The complexation of the
PEtOx macroligands L3 and L4 was initially studied using UV-
vis spectroscopy titrations in CH3CN. Upon addition of the
Fe2+ ions the ligand absorption band was red shifted from
300 nm to 348 nm, indicative of metal–ligand complex for-
mation, as previously observed for the complexation of ligand
L1. However, no distinct MLCT band nor a red or blue color-
ation of the solution was observed, suggesting that no [2 × 2]
grid-like complexes were formed (Fig. S28a and c†). Similar
results were found for Zn2+ complexes (Fig. S28b and d†). It
may be speculated that the PEtOx chains provide a too large
steric bulk, thereby obstructing the formation of the star-
shaped polymeric [2 × 2] grid-like complexes.
Next, the self-assembly of the PEG macroligands was inves-
tigated by UV-vis titration of L5, and L6 with Fe2+, and Zn2+
ions as shown in Fig. 4. During the titration with Fe2+, the
absorption maxima were red shifted as well. Furthermore, the
appearance of the characteristic MLCT band of the Fe2+
complex with L5 at 532 nm indicated the formation of poly-
meric [2 × 2] grids, albeit the MLCT band at 658 nm could not
be observed at this lower concentration (Fig. 4a), which was
also the case for the lower molar mass ligand L1 (Fig. S26†).
The absorption at 349 nm increased linearly upon addition of
Fe2+ and the maximum absorption was reached after approxi-
mately 1 equivalent of Fe2+ corresponding to [2 × 2] grid for-
mation. One may conclude that the [2 × 2] grid formation
formed by the complexes of L5 with Fe2+ indeed occurs despite
of the disappearance of MLCT band at 658 nm due to the low
concentration of L5 solution, which is consistent with the
above discussed result of L1 with Fe2+ at low concentration.
Fig. 4 UV-vis spectroscopy titration experiments: (a) stepwise addition
of 7.8 µL of 1.43 × 10−3 mol L−1 of solution of Fe(BF4)2 to a 3.7 × 10−5
mol L−1 of L5 in CH3CN, (b) stepwise addition of 6.0 µL of 1.8 × 10−3 mol
L−1 of solution of Zn(BF4)2 to a 3.7 × 10−5 mol L−1 of L5 in CH3CN, (c)
stepwise addition of 4.8 µL of 1.3 × 10−3 mol L−1 mol L−1 of solution of
Fe(BF4)2 to a 4.17 × 10−5 mol L−1 of L6 in CH3CN, and (d) stepwise
addition of 3.4 µL of 1.8 × 10−3 mol L−1 of solution of Zn(BF4)2 to a 4.17
× 10−5 mol L−1 of L6 in CH3CN. The inset graphs show the changes in
the absorption with the increase of the amount of transition metal ions.
than L5, shifted from 525 to 658 nm around 0.5 equivalents of
Fe2+ ions, suggesting a transition from complexes with two
ligands and one Fe2+ ion into grid-like complexes. Clearly, the
chain length of the polymer did not influence the formation of
the grids as the maximum of the L1, and L6 also shifted
around 0.5 equiv. of Fe2+ (Fig. S26† and 4c). The titration of
the PEG macroligands with Zn2+ ions revealed similar absorp-
tion band as the lower molar mass [2 × 2] grid-like complex of
L1 at 356 nm (Fig. 4b and d). These results clearly indicate
that the PEG macroligands can be used for the formation of
supramolecular star-shaped PEG. Further MALDI-TOF mass
spectroscopy on the formation of grids based on L5 or L6 with
metal ions could unfortunately not be obtained, presumably
because of insufficient stability of the polymeric grids that
are presumably fragmented at the high laser intensity.
Interestingly, upon complexation of L1 and L5 there is an
immediate appearance of the absorption band at 356 nm
while for L6 there is a more gradual change in the maximum
absorption wavelength, indicating that the longer PEG chains
suppress the cooperativity for grid formation, most likely due
to the enhanced steric hindrance.
Conclusions
Increasing the concentration would aid the detection of MLCT In conclusion, we have developed a versatile strategy for the
band at 658 nm. This assumption was further proved by the synthesis of bis-tridentate metal coordinating ligands by
successful determination of MLCT band at 658 nm for the CuAAc click reaction of 4,6-bis(6-ethynylpyridin-2-yl)-2-phenyl-
complexes of L6 with Fe2+ ions at a high concentration. pyrimidine with azido derivatives. The resulting 4,6-bis((1H-
Regardless of the high molar mass, the absorption maximum 1,2,3-triazol-4-yl)-pyridin-2-yl)-2-phenylpyrimidine ligands L1
of macroligand L6, measured at slightly higher concentration and L2 were demonstrated to self-assemble into metallosupra-
8750 | Dalton Trans., 2021, 50, 8746–8751
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
molecular grid-like complexes with Fe2+ and Zn2+ ions as were
proven by H-NMR, 2D NMR and UV-vis spectroscopy in com-
bination with MALDI-TOF-MS. The macromolecular ligands
were also successfully synthesized. The formation of the poly-
8 A. R. Stefankiewicz, G. Rogez, J. Harrowfield, M. Drillon
and J. M. Lehn, Dalton Trans., 2009, 5787–5802.
9 A. M. Madalan, X. Y. Cao, G. Rogez and J. M. Lehn, Inorg.
Chem., 2014, 53, 4275–4277.
1
meric grids was found to be dependent on the structure of the 10 B. Schneider, S. Demeshko, S. Neudeck, S. Dechert and
polymer as [2 × 2] grid like supramolecular complexes could F. Meyer, Inorg. Chem., 2013, 52, 13230–13237.
not be formed with the PEtOx macroligands L3 and L4. 11 D. E. Barry, C. S. Hawes, J. P. Byrne, B. la Cour Poulsen,
However, self-assembly of the PEG macroligands (e.g., L5, and
L6) with Fe2+ ions or Zn2+ ions resulted in the formation of
M. Ruether, J. E. O’Brien and T. Gunnlaugsson, Dalton
Trans., 2017, 46, 6464–6472.
star-shaped polymers by [2 × 2] grid-like metal complex for- 12 M. S. Alam, S. Stromsdorfer, V. Dremov, P. Muller,
mation. UV-vis titration experiments have shown that in the
initial stage complexes with two ligands and one Fe2+ are
J. Kortus, M. Ruben and J. M. Lehn, Angew. Chem., Int. Ed.
Engl., 2005, 44, 7896–7900.
formed while at a later stage (i.e., after 0.25 eq. of M2+) a tran- 13 E. Breuning, M. Ruben, J. M. Lehn, F. Renz, Y. Garcia,
sition to grid-like complexes with four ligands and four Fe2+
ions is taking place. Furthermore, the large PEG-chains lower
V. Ksenofontov, P. Gütlich, E. Wegelius and K. Rissanen,
Angew. Chem., 2000, 112, 2563–2566.
the cooperativity for grid-formation, ascribed to enhanced 14 A. Semenov, J. P. Spatz, M. Möller, J.-M. Lehn, B. Sell,
steric hindrance. The reported synthetic route can be applied
to a wide variety of azido end-functionalized polymers, leading
D. Schubert, C. H. Weidl and U. S. Schubert, Angew. Chem.,
Int. Ed., 1999, 38, 2547–2550.
not only to the development of new supramolecular star- 15 D. M. Bassani, J. M. Lehn, S. Serroni, F. Puntoriero and
shaped polymers but also forms the basis for new functional
materials and hydrogels.
S. Campagna, Chem. – Eur. J., 2003, 9, 5936–5946.
16 R. Hoogenboom, B. C. Moore and U. S. Schubert,
Macromol. Rapid Commun., 2010, 31, 840–845.
17 N. G. White, J. A. Kitchen, J. A. Joule and S. Brooker, Chem
Commun., 2012, 48, 6229–6231.
18 L. H. Uppadine, J. P. Gisselbrecht and J. M. Lehn, Chem
Commun., 2004, 6, 718–719.
Conflicts of interest
There are no conflicts to declare.
19 F. Robert, B. Tinant, R. Clérac, P.-L. Jacquemin and
Y. Garcia, Polyhedron, 2010, 29, 2739–2746.
20 M. Juríček, M. Felici, P. Contreras-Carballada, J. Lauko,
S. R. Bou, P. H. J. Kouwer, A. M. Brouwer and A. E. Rowan,
J. Mater. Chem., 2011, 21, 2104–2111.
Acknowledgements
X. Xu acknowledges the Chinese Scholarship Council (CSC,
File No. 201506780014) and Ghent University (BOF, File No. 21 M. H. Klingele and S. Brooker, Coord. Chem. Rev., 2003,
01SC1717) for the financial support for his Ph.D program; 241, 119–132.
R. Hoogenboom thanks FWO Flanders and Ghent Unviersity 22 P. Thongkam, S. Jindabot, S. Prabpai, P. Kongsaeree,
for continuous financial support. KVH thanks the Research
Foundation – Flanders (FWO) (projects AUGE/11/029 and
G099319N) for funding.
T.
Wititsuwannakul,
P.
Surawatanawong
and
P. Sangtrirutnugul, RSC Adv., 2015, 5, 55847–55855.
23 A. Mero, G. Pasut, L. D. Via, M. W. M. Fijten,
U. S. Schubert, R. Hoogenboom and F. M. Veronese,
J. Controlled Release, 2008, 125, 87–95.
24 A. Joosten, Y. Trolez, J. P. Collin, V. Heitz and J. P. Sauvage,
J. Am. Chem. Soc., 2012, 134, 1802–1809.
Notes and references
1 D. E. Barry, D. F. Caffrey and T. Gunnlaugsson, Chem. Soc. 25 B. N. Gacal, V. Filiz and V. Abetz, Macromol. Chem. Phys.,
Rev., 2016, 45, 3244–3274. 2016, 217, 672–682.
2 A. J. McConnell, C. S. Wood, P. P. Neelakandan and 26 G. Volet, T.-X. Lav, J. Babinot and C. Amiel, Macromol.
J. R. Nitschke, Chem. Rev., 2015, 115, 7729–7793. Chem. Phys., 2011, 212, 118–124.
3 M. Ruben, J. Rojo, F. J. Romero-Salguero, L. H. Uppadine 27 X. Xu, J. F. R. Van Guyse, V. V. Jerca and R. Hoogenboom,
and J. M. Lehn, Angew. Chem., Int. Ed. Engl., 2004, 43,
3644–3662.
4 J.-F. Létard, P. Guionneau and L. Goux-Capes, Top. Curr.
Chem., 2004, 235, 221–249.
5 B. Schneider, S. Demeshko, S. Dechert and F. Meyer,
Angew. Chem., Int. Ed., 2010, 49, 9274–9277.
6 A. R. Stefankiewicz, G. Rogez, J. Harrowfield, M. Drillon
and J.-M. Lehn, Dalton Trans., 2009, 5787–5802.
Macromol. Rapid Commun., 2020, 41, 1900305.
28 X. Xu, V. V. Jerca and R. Hoogenboom, Macromol. Rapid
Commun., 2020, 41, 1900457.
29 R. Hoogenboom, D. Wouters and U. S. Schubert,
Macromolecules, 2003, 36, 4743–4749.
30 V. Aubert, V. Guerchais, E. Ishow, K. Hoang–Thi, I. Ledoux,
K. Nakatani and H. Le Bozec, Angew. Chem., Int. Ed., 2008,
47, 577–580.
7 M. T. Youinou, N. Rahmouni, J. Fischer and J. A. Osborn, 31 X. Xu, F. A. Jerca, V. V. Jerca and R. Hoogenboom, Adv.
Angew. Chem., 1992, 104, 771–773.
Funct. Mater., 2019, 29, 1904886.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 8746–8751 | 8751