A chromatography-free total synthesis of a ferrocene-containing dendrimer exhibiting the property of recognizing 9,10-diphenylanthracene

Article abstract of DOI:10.1039/d0dt04261g

Molecules comprising several ferrocene residues constitute an intriguing group of compounds for various applications. Here, the total synthesis of a new example of a ferrocene-containing dendrimer is presented. The target compound was obtained in excellent combined yield (65%) employing facile, chromatography-free methods at each step. Interesting findings, meeting the dynamic covalent chemistry concept, are reported. Cyclic voltammetry analyses revealed one pair of current signals for the ferrocene moieties. Ultimately, the synthesized ferrocene-containing dendrimer has been used as an innovative recognition material for 9,10-diphenylanthracene, a polycyclic aromatic hydrocarbon, with the limit of detection value equal to 0.06 μM. This journal is

Full text of DOI:10.1039/d0dt04261g

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
A chromatography-free total synthesis of a
ferrocene-containing dendrimer exhibiting the
property of recognizing 9,10-diphenylanthracene†
Cite this: Dalton Trans., 2021, 50,
2483
a
a,b
b
Artur Kasprzak, * Monika K. Nisiewicz and Anna M. Nowicka
Molecules comprising several ferrocene residues constitute an intriguing group of compounds for various
applications. Here, the total synthesis of a new example of a ferrocene-containing dendrimer is presented.
The target compound was obtained in excellent combined yield (65%) employing facile, chromato-
graphy-free methods at each step. Interesting findings, meeting the dynamic covalent chemistry concept,
are reported. Cyclic voltammetry analyses revealed one pair of current signals for the ferrocene moieties.
Ultimately, the synthesized ferrocene-containing dendrimer has been used as an innovative recognition
material for 9,10-diphenylanthracene, a polycyclic aromatic hydrocarbon, with the limit of detection
value equal to 0.06 μM.
Received 15th December 2020,
Accepted 8th January 2021
DOI: 10.1039/d0dt04261g
rsc.li/dalton
applied sciences, e.g., medicinal chemistry or organic elec-
tronics. The literature reports on Fc-containing dendrimers
Introduction
1
1–13
14a
Over the years, ferrocene (Fc) has been a coordination com- include the preparation of silicon-,
tetraphenylmethane-,
1
,2
14b
15,16a
pound of wide interest in organic and materials chemistry.
pyridylphenylene-,
or
poly(amidoamine)-based
1
6b
This metallocene has various prospective applications in a structures as well as metallodendrimers,
hydrogen-
3
16c
16d–f
number of important fields, such as medicinal chemistry,
bonded, or coordination-type
assemblies. These dendri-
4
5
6
17
self-healing materials, catalysis , or ion receptors. Even mers showed interesting features, such as water solubility,
though the discovery of Fc was accomplished 80 years ago, in catalytic activity, ion recognition
recent years its chemistry has developed remarkably, and Fc- form inclusion complexes with cyclodextrins. The chemistry
containing molecules have become important synthetic of metallocene-tethered dendrimers has intensively developed
1
3
14b,16c
or the property to
15
targets.
The synthesis of compounds bearing several Fc residues have been summarized in recent reviews.
has been one of the major goals of modern chemistry, since Importantly, the dendrimer chemistry has been commonly
such molecules exhibit intriguing chemical structures, as well merged with the π-conjugation concept, since π-conjugated
as unprecedented properties and applications. Some of the aromatic compounds bearing number of delocalized
over the years, and a number of these interesting molecules
1
8a–e
a
milestones achieved in recent years include, e.g. the reports by π-electrons were reported as interesting frameworks and ideal
7
18f–h,19a
Albrecht & Long on the preparation of a Fc cyclic oligomer or materials for various applications.
In general, the ana-
8
Fc-bearing macrocycles. Special interest has been given to the lysis of the literature examples of Fc-containing dendrimers
synthesis of Fc-containing dendrimers. Dendrimers have brings a conclusion that the previous reports focused mostly
advantages in comparison to their linear or branched macro- on the installation of ferrocene units in the peripheral parts of
molecular analogs, such as well-defined monodispersed struc- the π-conjugated dendrimer cores. The reports dealing with
tures, encouraging bio-physical features and the possibility of aromatic dendrimers with Fc units located both in the periph-
the control of the position and number of introduced eries and the core are sparse, as well as the synthesis of these
1
,9,10
substituents.
These beneficial parameters of dendrimers molecules is challenging; in consequence there is limited
are extremely important also from the viewpoint of their number of reports on the successful synthesis of such orga-
nized compounds (see the graphical representation in Fig. 1).
Even though the examples of such dendrimers can be found
in the literature, one may conclude that the synthetic protocols
are time-, cost- and solvent-consuming, such as in the case of
a
Department Faculty of Chemistry, Warsaw University of Technology, Noakowskiego
Str. 3, 00-664 Warsaw, Poland. E-mail: akasprzak@ch.pw.edu.pl
b
Faculty of Chemistry, University of Warsaw, Pasteura Str. 1, 02-093 Warsaw, Poland
the interesting phosphorous-containing ferrocene-tethered
†
Electronic supplementary information (ESI) available: Compound characteriz-
1
9b,c
ation data and spectra. See DOI: 10.1039/d0dt04261g
dendrimers.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 2483–2492 | 2483

View Article Online
Paper
Dalton Transactions
triphenylbenzene framework (bearing multiple bonds) with
2
2
imine linkages (bearing C(sp ) and N(sp ) atoms) and cyclo-
pentadienyl rings bound on the opposite sides of the iron
atom (Fc units of Fc-den).
The herein presented synthesis of Fc-den bearing nine Fc
residues lies in a convergent synthesis strategy, in which the
key to success was the preparation of two π-conjugated build-
ing blocks 1 and 2 (Fig. 2). In general, we hypothesized that
sequential imine-bond formation reactions shall provide the
target dendrimer and intermediate compounds. Importantly,
the success of this strategy would be also beneficial in terms
of the green synthesis of Fc-den, since the previous studies on
5
,21,22
Fig. 1 Graphical summary of the reported Fc-decorated dendrimers imine-containing triphenylbenzenes
suggested that no
and the aim of this work.
chromatographic purification shall be required.
In pursuit of new dendrimers consisting of several Fc units
and π-conjugated building blocks, herein a chromatography-
free total synthesis of the novel Fc-containing dendrimer Fc-
Experimental section
Materials and methods
den is presented (Fig. 2). Our goal was to design a new, chrom- Chemical reagents and solvents were commercially purchased
atography-free and efficient way for the synthesis of a Fc-teth- and purified according to the standard methods, if necessary.
ered dendrimer of potential interest to various communities. NMR experiments were carried out using a Varian VNMRS
1
13
The structure of Fc-den includes 1,3,5-triphenylbenzene cores 500 MHz spectrometer ( H NMR at 500 MHz or C NMR at
and imine linkages. These structural motifs have been widely 125 MHz) equipped with a multinuclear z-gradient inverse
employed for the construction of organized π-conjugated mole- probe head. Unless otherwise stated, the spectra were recorded
1
13
cules, including several examples of aromatic dendrimers with at 25 °C. Standard 5 mm NMR tubes were used. H and
Fc units located in the peripheries of the dendrimer.
C
5
,20,21a,b
chemical shifts (δ) were reported in parts per million (ppm)
We expect π-conjugation within a Fc-den framework because of relative to the solvent signal: CDCl , δH (residual CHCl3)
3
enhancing the number of delocalized p-electrons of the 1,3,5- 7.26 ppm, δC 77.2 ppm. NMR spectra were analyzed with the
Fig. 2 Structure of the target Fc-containing dendrimer (Fc-den) and its retrosynthetic analysis.
2484 | Dalton Trans., 2021, 50, 2483–2492
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
1
13
1
MestReNova v12.0 software (Mestrelab Research S.L). H DOSY (m, 6H), 4.96–4.88 (m, 12H), 4.58–4.52 (m, 12H); C{ H} NMR
Diffusion Ordered SpectroscopY) NMR experiments were per- (CDCl , 125 MHz, ppm), δ 193.0 (3C), 160.6 (3C), 141.3 (3C),
(
3
C
formed using a stimulated echo sequence incorporating 138.2 (3C), 127.5 (3C), 123.8 (3C), 121.4 (3C), 120.6 (6C), 84.7
bipolar gradient pulses and with convection compensation. (3C), 82.4 (3C), 72.6 (6C), 71.4 (6C), 70.5 (6C), 70.1 (6C); FT-IR
The gradient strength was logarithmically incremented in 15 (ATR), v 3082, 3014, 2868, 2830, 1682, 1614, 1588, 1508, 1364,
steps from 25% up to 95% of the maximum gradient strength. 1242, 1178, 1038, 820, 744 cm− ; UV-Vis, λmax (CHCl
1
) 264,
3
+
The DOSY Toolbox software was used for DOSY NMR spectral 333 nm; TOF-HRMS (ESI): calcd for C60
processing (The DOSY Toolbox – version 2.5, 2014, Mathias 1024.1582, found: m/z 1024.1561; Anal. calcd for
Nilsson, School of Chemistry, University of Manchester, UK). : C, 70.41; H, 4.43; N, 5.65. Found: C, 70.50; H,
3 3 3
H46Fe N O [M + H] =
60 3 3 3
C H45Fe N O
Fourier-transform infrared (FT-IR) spectra were recorded in 4.44; N, 4.10.
the attenuated total reflectance (ATR) mode with a Thermo
Synthesis of 2
Nicolet Avatar 370 spectrometer with a spectral resolution of
−
1
2
cm
(100 scans). The wavenumbers for the absorption A solution of 3 (1054.2 mg, 3.00 mmol, 1.0 equiv.) and ferroce-
−
1
bands ν were reported in cm
.
necarboxaldehyde (5) (1605.3 mg, 7.50 mmol, 2.5 equiv.) in
UV-vis measurements were performed with the PerkinElmer MeOH (300 mL) was stirred for 5 minutes. Glacial acetic acid
spectrometer Lambda 25, at room temperature in a quartz (1800 μL) was added, and the mixture turned turbid within
cuvette with a 1 cm long optical window. For the UV-Vis several minutes. The reaction mixture was stirred at room
measurements, the wavelengths for the absorption maxima temperature for 24 hours. The solid was filtered off and
λ
max were reported in nm.
washed with methanol (15 mL). The solid residue was sus-
TOF-HRMS (ESI) measurements were performed with a pended in methanol (200 mL), sonicated for 20 minutes at
Q-Exactive ThermoScientific spectrometer. Melting points were room temperature, filtered off, washed with methanol (10 mL)
determined on Standford Research Systems MPA 100 and were and dried at room temperature for 24 hours to give compound
uncorrected.
2 (2052.0 mg; 92% yield) as an orange solid.
1
Elemental analyses were performed using CHNS Elementar
3 H
Mp: 225–226 °C; H NMR (CDCl , 500 MHz, ppm), δ 8.44
Vario EL III apparatus. Each elemental composition was (s, 2H), 7.82 (s, 3H), 7.75–7.71 (m, 6H), 7.31–7.28 (m, 4H),
reported as an average of two analyses. 6.82–6.79 (m, 2H), 4.86 (bs, 2H), 4.80–4.79 (m, 4H), 4.61–4.60
Electrochemical measurements were performed using an (m, 4H), 4.28 (s, 10H); C{ H} NMR (CDCl
Autolab Eco Chemie potentiostat, model PGSTAT 12 controlled 161.5 (2C), 146.0 (2C), 142.3 (1C), 142.1 (1C), 132.1 (2C),
1
3
1
3
, 125 MHz, ppm),
δ
C
via software on a personal computer. All experiments were 128.4 (4C), 128.2 (2C), 123.4 (2C), 123.1 (2C), 121.2 (2C), 115.6
carried out in a three-electrode system. A disc glassy carbon (4C), 115.5 (2C), 73.4 (2C), 71.6 (4C), 69.8 (10C), 69.5 (4C);
electrode (GC, ∅ = 3 mm, BAS Instruments) was used as the FT-IR (ATR), v 3080, 3026, 2890, 1634, 1579, 1496, 1160, 1096,
working electrode, an Ag/AgCl/3 M KCl as the reference elec- 954, 800, 740 cm− ; UV-Vis, λmax (CHCl ) 269, 340 nm;
1
3
+
trode, and a platinum wire as the auxiliary electrode. Before TOF-HRMS (ESI): calcd for C46
H
38Fe
2
N
3
[M + H] = 744.1759,
each experiment, the GC electrode was polished with 1 μm alu- found: m/z 744.1760; Anal. calcd for C46
H
2 3
37Fe N : C, 74.31; H,
minium oxide powder on a wet pad, rinsed with water and 5.02; N, 5.65. Found: C, 74.62; H, 5.00; N, 5.67.
then dried with argon. During all measurements, the electro-
Synthesis of Fc-den
chemical cell was kept in a Faraday cage to minimize the elec-
trical noise. The measurements were carried out with deoxyge- A solution of 1 (460.6 mg, 0.45 mmol, 1.0 equiv.) and 2
nated solutions.
(1107.9 mg, 1.49 mmol, 3.3 equiv.) in THF : MeOH 1 : 1 v/v
80 mL) was stirred for 5 minutes. Glacial acetic acid (300 μL)
was added, and the mixture turned turbid within several
(
Synthesis of 1
To a stirred solution of 1,3,5-tris(4-aminophenyl)benzene (3) minutes. The reaction mixture was stirred at room temperature
(
(
140.6 mg, 0.40 mmol, 1.0 equiv.) and 9,10-diphenylantracene for 48 hours. The solid was filtered off and washed with
528.6 mg, 1.60 mmol, 4.0 equiv.) in MeOH (15 mL), a solution THF : MeOH 1 : 1 v/v (10 mL). The solid residue was suspended
of 1,1′-diformylferrocene (4) (871.4 mg, 3.60 mmol, 9.0 equiv.) in THF : MeOH 1 : 1 v/v (100 mL), sonicated for 20 minutes at
in MeOH (13 mL) was added in one portion. Glacial acetic acid room temperature, filtered off, washed with THF : MeOH 1 : 1
(250 μL) was added, and the mixture turned turbid within v/v (10 mL) and dried at room temperature for 24 hours to give
several minutes. The reaction mixture was stirred at room Fc-den (1281.6 mg; 89% yield) as a red solid.
1
temperature for 24 hours. The solid was filtered off and
3 H
Mp: >300 °C; H NMR (CDCl , 500 MHz, ppm), δ 8.45 (s,
washed with methanol (15 mL). The solid residue was sus- 6H), 8.14 (s, 6H), 7.80 (s, 9H), 7.75–7.71 (m, 18H), 7.41 (s, 3H),
pended in methanol (80 mL), sonicated for 20 minutes at 7.31–7.29 (m, 18H), 7.24–7.21 (m, 6H), 6.82–6.79 (m, 6H),
room temperature, filtered off, washed with methanol (10 mL) 5.10–5.09 (m, 12H), 4.85–4.84 (m, 12H), 4.53–4.52 (m, 24H),
1
3
1
and dried at room temperature for 24 hours to give compound 4.28 (s, 30H); C{ H} NMR (CDCl , 125 MHz, ppm), δ 161.6
3
C
1
(393.1 mg; 96% yield) as an orange-red solid.
(6C), 161.5 (6C), 152.4 (3C), 151.2 (3C), 146.2 (3C), 146.1 (3C),
9.95 142.1 (3C), 141.3 (3C), 138.5 (6C), 138.3 (6C), 138.2 (6C), 131.9
1
Mp: 262–263 °C; H NMR (CDCl
3
, 500 MHz, ppm), δ
H
(s, 3H), 8.13 (s, 3H), 7.41 (s, 3H), 7.24–7.22 (m, 6H), 6.82–6.81 (3C), 128.3 (6C), 128.4 (6C), 128.2 (6C), 128.1 (6C), 124.6 (3C),
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 2483–2492 | 2485

View Article Online
Paper
Dalton Transactions
1
24.2 (3C), 123.9 (3C), 123.7 (3C), 121.4 (6C), 121.3 (3C), 115.6 cage 8 and enabled the selective synthesis of 1. Importantly,
(
6C), 115.5 (6C), 82.4 (6C), 80.6 (6C), 73.3 (6C), 71.6 (6C), 71.5 9,10-diphenylanthracene, and unreacted 3 and 4 can be easily
1
(12C), 70.0 (6C), 69.8 (6C), 69.5 (30C), 69.3 (12C); H DOSY removed from the reaction mixture by washing with methanol
−
10
3
NMR (500 MHz, CDCl , 0.7 mM), D 0.901 × 10 ; FT-IR (ATR), to yield pure 1 in an excellent yield of 96% (Table 1a, entry 7).
v 3086, 3024, 2884, 1620, 1584, 1490, 1456, 1168, 1102, 954, Elemental analysis confirmed the synthesis of pure 1. Finally,
18, 746 cm− ; UV-Vis, λmax (CHCl
1
) 284, 295, 363, 492 nm; to confirm that 1 can interact with 9,10-diphenylanthracene
8
3
+
1
TOF-HRMS (ESI): calcd for C198
200.6318, found: m/z 3200.6330; Anal. calcd for NMR analysis was performed (see the ESI† for the data). H
H151Fe
9
N
12 [M + H]
=
(template) by means of non-covalent forces, at first H DOSY
1
3
C
198 9
H150Fe N12: C, 74.32; H, 4.72; N, 5.25. Found: C, 74.40; H, DOSY NMR is a powerful method to track non-covalent inter-
2
4a–c
4
.74; N, 5.26.
actions, such as host–guest complexation or π–π stacking.
The non-covalent phenomena occurring between two mole-
cules in the sample lead to a change in their diffusion coeffi-
1
cients in H DOSY NMR. The diffusion coefficient for 9,10-
Results and discussion
−
10
2
diphenylanthracene in the presence of 1 (D = 7.112 × 10
m
−
1
The synthetic studies were initiated by synthesizing 1,3,5-(4-
s ) was found to be lower in comparison to the sample of
aminophenyl)triphenylbenzene (3; Fig. 3a) in two steps follow- native 9,10-diphenylanthracene in the absence of 1 (D = 9.621
5
−10
2
−1
ing a literature procedure. With 3 in hand, the preparations of × 10
1
m
s
; see the ESI† for the data). This change was
ascribed to the non-covalent π–π stacking interactions between
1 and 9,10-diphenylanthracene. Although the apparent
and 2, the key building blocks of Fc-den, were investigated.
It was found that the major obstacle for the successful syn-
thesis of the aldehyde-terminated triferrocenyl derivative 1 was binding constant (Kapp) was found to be ca. 1.3 × 10
2
−1
M
not only the possible formation of some oligomers but also a (because this interaction is a dynamic process) the 9,10-diphe-
cage compound 8 (Fig. 3b). In other words, one shall control nylanthracene:1 system has been clearly detected by cold-spray
the supramolecular arrangement formed, especially the ESI-MS analysis (see the ESI† for data). This further confirms
rotation between Fc’s cyclopentadienyl rings of 4 during the that 9,10-diphenylanthracene can interact with 1 via non-
reaction course, in order to selectively obtain 1 instead of 8. covalent π–π stacking interactions and supports the role of
Although the developed method was facile and chromato- 9,10-diphenylanthracene as a template in the synthesis of 1.
graphy-free (the solid product was filtered and washed off with
With this milestone achieved, the synthesis of 2, a diferro-
methanol), numerous attempts had to be made to selectively cenyl derivative of 3, was investigated (Fig. 3b). The method
obtain target compound 1 in good yield (Table 1a). In general, was based on an acid-catalyzed imine-bond formation reaction
better results were found for higher molar excesses of 1,1′- between 3 and ferrocenecarboxaldehyde (5). As expected,
diformylferrocene (4) and its higher concentrations; however, synthesizing 2 without traces of monoferrocenyl or triferroce-
the yields did not exceed 32% (Table 1a, entries 1–4). A break- nyl triphenylbenzenes has been the biggest challenge. Various
through within this extensive screening of reaction parameters reaction parameters were tested (Table 1b). Gratifyingly, the
occurred when the reaction was conducted with an excess of 4, selective and efficient formation of 2 has been achieved by
in its high concentration and in the presence of 9,10-dipheny- tuning the molar equivalent of 5 and its concentration.
lanthracene as a template (Table 1a, entries 5–7). The basics of Similarly to the synthesis of 1, pure 2 has been obtained in
2
3a,b
template’s action were as follows.
Firstly, from a previous very high yield (90–92%; Table 1b, entries 6 and 7) employing
5
report it is known that the confined space of cage 8 is too an easy-to-perform, chromatography-free method (the solid
small to incorporate an aromatic molecule (guest), such as product was filtered and washed off with methanol).
9
,10-diphenylanthracene, between stacked 1,3,5-triphenylben- Ultimately, 2 can also be prepared on a gram scale (Table 1b,
zene moieties. Secondly, the π–π stacking interaction, a entry 7). Elemental analysis confirmed the synthesis of pure 2.
dynamic process, between 9,10-diphenylanthracene and 3 and All that remained to construct Fc-den was merging 1 and 2.
were expected during the reaction course. For the effective In the light of the successful, facile syntheses of 1 and 2, the
1
formation of 8, the CHO moieties of 4 must be in the syn con- ultimate goal was to design a facile, chromatography-free
formation; for the effective synthesis of 8, there should be no methodology to obtain Fc-den. This would make the developed
steric hindrance toward the formation of the stacked cage-type total synthesis chromatography-free in all stages. However, the
product 8. In other words, the anti-conformation of the substi- major issue in this synthetic step was the similar solubility
tuents in the cyclopendatienyl rings of 4 limits the efficient profile for Fc-den to those for 1 and 2, i.e. similarly to 1 and 2,
2 2 3
formation of 8. Thus, we supposed that a molecule (9,10- Fc-den is only soluble in chloroalkanes (CH Cl , CHCl ) and
diphenylanthracene) that interacts via dynamic π–π stacking THF. Seeking a possible reaction parameter, the reaction in
non-covalent interactions with 1 or 3 (or their intermediate the THF/MeOH system was exposed (Fig. 3c; Table 1c). This
products with 4) might therefore increase this steric hin- trial originated from the prediction that 1 and 2 shall remain
drance, which results in the inefficient formation of 8 because dissolved when their THF solution is diluted with MeOH,
of facilitation of the discussed anti-conformation of CHO func- whilst Fc-den shall precipitate from such a reaction mixture.
tionalities. In the light of these facts, this stacking phenom- Through control experiments (Table 1c, entries 1–6), it was
enon with 9,10-diphenylanthracene prevented the formation of indeed found that on the contrary to 1 and 2, Fc-den is not
2486 | Dalton Trans., 2021, 50, 2483–2492
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
Fig. 3 Total synthesis of Fc-den: (a) synthesis of the starting material 3; (b) synthesis of the key building blocks 1 and 2; (c) assembly of 1 and 2 to
obtain Fc-den. Reagents and conditions: (i) 6 (1.0 equiv.), CF SO H (cat.), PhMe, reflux, 48 h; (ii) 7 (1.0 equiv.), N monohydrate (12.0 equiv.), Pd/C,
EtOH, reflux, 16 h; (iii) 3 (1.0 equiv.), 4 (9.0 equiv.), 9,10-diphenylanthracene (4.0 equiv.), CH COOH (cat.), MeOH, rt, 24 h; (iv) 3 (1.0 equiv.), 5 (2.5
equiv.), CH COOH (cat.), MeOH, rt, 24 h; (v) 1 (1.0 equiv.), 2 (3.3 equiv.), CH COOH (cat.), THF : MeOH 1 : 1 v/v, rt, 48 h.
3
3
2 4
H
3
3
3
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 2483–2492 | 2487

View Article Online
Paper
Dalton Transactions
Table 1 Selected optimization of the reaction conditions for the key building blocks 1 (a) and 2 (b) and target Fc-den (c)
b
9
,10-Diphenylanthracene
Yield of 1 (yield
of 8 )
c
Entry
3 (equiv.)
4 (equiv.)
Concentration of 4 (M)
(equiv.)
a
(a)
d
1
2
3
4
5
6
7
1.0
1.0
1.0
1.0
1.0
1.0
1.0
5.0
6.0
9.0
9.0
9.0
9.0
9.0
0.02
0.04
0.06
0.12
0.12
0.12
0.12
—
—
—
—
1.0
4.0
4.0
5 (65)
d
9 (62)
d
18 (60)
32 (64)
d
d
65 (15)
d
95 (0)e
96 (0)
b
Entry
3 (equiv.)
5 (equiv.)
Concentration of 5 (M)
Time (hours)
Yield of 2
f
(b)
d
1
2
3
4
5
6
7
1.0
1.0
1.0
1.0
1.0
1.0
1.0
2.0
2.0
2.0
1.5
2.5
2.5
2.5
0.050
0.050
0.025
0.025
0.040
0.025
0.025
24
12
24
24
24
24
24
64
d
55
d
68
d
41
d
87
d
90
g
92
b
Entry
1 (equiv.)
2 (equiv.)
Solvent
Time (hours)
Yield of Fc-den
h
(c)
i, j
1
2
3
4
5
6
7
1.0
1.0
1.0
1.0
1.0
1.0
1.0
3.3
3.3
3.3
3.3
3.0
3.3
3.3
THF
48
48
48
48
48
72
48
18
30
i, j
THF : MeOH 4 : 1 v/v
THF : MeOH 2 : 1 v/v
THF : MeOH 1 : 1 v/v
THF : MeOH 1 : 1 v/v
THF : MeOH 1 : 1 v/v
THF : MeOH 1 : 1 v/v
i, j
71i
89i
85i
84
k
89
a
f
b
c
1
d
e
CH
CH
were found in the product (based on H NMR). Gram scale reaction (1.49 mmol of 2).
3
COOH (cat.), MeOH, rt, 24 h. Isolated yields. Based on H NMR. Reaction scale: 0.02 mmol of 3. Reaction scale: 0.40 mmol of 3.
g
h
i
j
3
COOH (cat.), MeOH. Gram scale reaction (3.00 mmol of 3). CH COOH (cat.), rt. Reaction scale: 0.015 mmol of 1. Impurities of 1 and/or
3
1 k
2
soluble in the THF : MeOH solvent system. Through the appli-
cation of these conditions, Fc-den has been successfully pre-
pared in a chromatography-free way in an excellent yield of
8
9% under mild conditions (Table 1c, entry 4). Importantly,
the large-scale access to 2 provided the possibility of the
chromatography-free gram-scale preparation of Fc-den (89%;
1
Table 1c, entry 7). The presence of one set of signals in
H
NMR (12 groups of signals in total; see Fig. S5†) reveals that,
in solution, all building subunits of Fc-den, i.e. monosubsti-
tuted ferrocenes, 1,1′-disubstituted ferrocenes, terminal 1,3,5-
triphenylbenzene cores, and imine linkages (two sets), are
equivalent on the NMR timescale. The selective formation of
1
one system has been further confirmed by H DOSY NMR ana-
lysis (see Fig. S7†). Finally, elemental analysis confirmed that
1
pure Fc-den has been synthesized. It is worth noting that H
3
Fig. 4 UV-Vis spectra (CHCl ) of 1, 2 and Fc-den.
NMR analyses revealed that Fc-den is stable in air (the solid
1
sample was kept for three months in air and the H NMR spec-
trum was acquired, see Fig. S19†), as well as in water (see ascribed to the π–π* transition of 1,3,5-triphenylbenzene cores,
Fig. S20†).
whilst the latter originated from the presence of Fc and imine
The UV-Vis spectra of 1, 2 and Fc-den in CHCl
3
are pre- moieties. The significant red shifts for these bands for Fc-den
sented in Fig. 4. Fc-den exhibited incomparably higher absor- (294 nm, 362 nm) in comparison to the starting materials 1
bance values in comparison to the starting reactants 1 and 2 (262 nm, 330 nm) and 2 (266 nm, 338 nm) were ascribed to
(see Fig. S15†). The spectra featured two major absorption the π-conjugation within the dendrimer framework. A signifi-
bands at 262–294 nm and 330–362 nm. The first band was cantly smaller peak at 488 nm for Fc-den (observed only for
2488 | Dalton Trans., 2021, 50, 2483–2492
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
this compound) could be attributed to the dynamic E–Z iso- distance between them. The closer the Fc units are to the
2
5
merization behavior within imine-type linkages or some central benzene core, and therefore closer to each other, the
2
6
metal-to-triphenylbenzene charge transfers.
more significant is the electrostatic factor (interaction of the
Fc and its derivatives easily undergo one electron oxidation ferrocene cation with the ferrocene cation of the other mole-
2
7–29a
29b–e
to form a ferrocenium cation in a reversible manner.
cule); therefore multiple CV waves were observed.
Taking
Thus, we investigated the electrochemical behavior of Fc-den into account the structure of the Fc-den molecule presented in
in dichloromethane (DCM) by cyclic voltammetry (CV) using a Fig. 3 the distance between Fc–Fc units as well as the Fc–
conventional glassy carbon electrode. The cyclic voltammo- central benzene core is large enough for the electrostatic inter-
gram of 1 mM Fc-den recorded in DCM with 100 mM tetra- actions to not affect the shape and number of current signals.
butylammonium hexafluorophosphate (TBAHFP) as a support- Moreover, the separation of CV signals in the case of com-
−
1
ing electrolyte for 100 mV s scan rate is presented in the pounds with few identical redox centers also depends on the
2
9b–e
bottom inset of Fig. 5. The recorded voltammogram shows type of electrolyte anions.
only one pair of current signals. This behavior suggests that ferricinium cations such as PF
The anions strongly bonded to
−
−
6
4
or BF reduce the ability of
the redox (Fc) units of Fc-den are electrochemically equivalent CV current signal separation, as well as slow down the electron
and are oxidized and reduced at the same potential. The exchange process. Fc-den contains two types of Fc units in the
number of current signals observed in cyclic voltammograms structure (inner and peripheral one). The presence of two
depends not only on the number of Fc units, but also on the types of redox centres should be visible on the CV curve in the
form of two pairs of currents signals, but it only occurs when
the Fc–Fc distance is very short. Probably in the case of Fc-
den, only one type of Fc unit undergoes an electron exchange
reaction with the electrode surface. This hypothesis was con-
firmed by the comparison of the charge values under cathodic
or anodic peaks obtained for Fc-den and pure Fc. This ratio
(QFc-den : QFc) is 2 : 1. Moreover, the difference in the position
p
of anodic and cathodic peaks (ΔE ) is 350 mV; the value was
determined for scan rates ≤100 mV s− , which suggests rela-
1
tively slow electron transfer and solution resistance. It should
be stressed that the value of ΔE
ditions is equal to 160 mV; in the case of aqueous solution,
the ΔE for Fc and its derivatives in aqueous solutions is
p
for Fc under the same con-
p
roughly equal to 88 mV. This significant difference in the sep-
aration of the anode and cathode current signals in relation to
aqueous solutions is most likely related to the solution
resistance.
In order to determine the value of the diffusion coefficient
of Fc-den, CVs were recorded in an 1 mM solution of Fc-den at
different scan rates in the range of 2–1000 mV s− . The linear
dependence of the current signal intensity versus square root
of the scan rate (see the top inset in Fig. 5) confirmed the
diffusional character of the electrochemical reaction.
1
0
.5
Therefore, from the slope of the plot Ipa = f(v ), the diffusion
coefficient for the studied ferrocene derivative according to the
Randles–Sevcik equation was determined and was equal to
.31 × 10− cm ·s . In turn, from the intercept of the depen-
9
2 −1
1
f
dence ln(Ipa) = f(Epa − E ), the electron-transfer rate constant
= 3.01 × 10− cm s ) was determined according to the
4
−1
(k
0
formula:
ꢀ
ꢁ
αnF
RT
*
0
Ipa ¼ 0:227nFaC ko exp ꢀ
ðEpa ꢀ Ef Þ
Fig. 5 (A) Cyclic voltammograms of Fc-den recorded in DCM at various
scan rates. Top inset: Plot of Fc-den oxidation currents versus the
square root of the scan rate. Bottom inset: the CV voltammogram in a
where Ipa is the current intensity of the anodic peak, n is the
number of electron exchange during the electrode process, F is
the Faraday constant, A is the electrode surface area, C* is
wider potential range (v = 100 mV s− ). (B) Cyclic voltammograms of Fc-
1
0
den and Fc recorded in DCM at 50 mV s−1. Experimental conditions:
the concentration of the electroactive species, Epa is the
C
Fc-den = 1.0 mM, CTBAHFP = 100 mM, T = 21 °C.
potential of the anodic peak, E is the formal potential, R is the
f
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 2483–2492 | 2489

View Article Online
Paper
Dalton Transactions
gas constant, T is temperature, and α is the transition to engineer the application of Fc-den as the building unit of
coefficient. an analytical device dedicated to the specific recognition of
The Fc units were attached to the 1,3,5-triphenylbenzene 9,10-DPA, a polycyclic aromatic hydrocarbon (PAH) commonly
skeletons via imine bonds. It is known that by applying too used in, e.g. the technologies of organic light emitting
3
1,32
high a potential >1.1 V such bonds can be broken relatively diodes.
The analytical characteristics of the sensor (GC/Fc-
easily, as representatively shown in our previous report on Fc- den-TBAHFP-Nafion®), which has been fabricated as noted
3
0
functionalized 1,3,5-triphenylbenzene. However, in the case above, were determined electrochemically on the basis of the
of this compound, the decomposition of the imine bond was changes in the Fc-den oxidation current signal. The represen-
not observed even when the applied potential was higher than tative differential pulse voltammograms (DPV) plotted as a
1
.3 V.
function of the 9,10-DPA concentration are presented in Fig. 7.
Due to the presence of phenyl rings in the Fc-den structure, After the addition of 9,10-DPA to the solution, a decrease in
this compound can be an excellent receptor for other aromatic the intensity of the Fc-den current signal was observed. This
compounds. The ability of Fc-den in the detection of aromatic decrease was greater when a greater concentration of the ana-
compounds was checked electrochemically. To form a reco- lysed compound was introduced into the solution. The signal
gnition layer, the glassy carbon electrode surface was modified stabilization after the addition of 9,10-DPA took place after
by placing a 10 μL droplet of 1 mM Fc-den in DCM with three consecutive voltammograms were recorded in succession
addition of 100 mM TBAHFP and 5% Nafion® and left to dry (total time: 4 minutes). The detection limit (LOD) was esti-
in a desiccator. Before the voltammetric experiments, the reco- mated from the calibration plot as the three standard devi-
gnition layer was first cycled between 0 and 1 V in an 100 mM ations of the five independent controls (measurements of the
aqueous solution of tetrabutylammonium perchlorate (TBAP) current in the pure buffer solution). The regression equations
until a stable voltammogram was obtained. In order to prove describing the linear responses of IFc-den = f(C9,10-DPA) as well
the analytical potential of the proposed compound, the as the analytical ranges and LODs are presented in Table 2A,
measurements were carried out using the following aromatic while the calculated inter- and intraday precisions are given in
compounds: 1,4-terphenyl, pyrene, phenanthrene, fluorene Table 2B.
and 9,10-diphenylanthracene (9,10-DPA). In the presence of
A great advantage of the proposed sensor is that 9,10-DPA
2
00 µM analyte in the solution, a drastic decrease in the inten- did not form too stable complexes with Fc-den, so the receptor
sity of the Fc-den oxidation signal was observed only in the layers did not require regeneration. It was enough to immerse
case of 9,10-DPA (Fig. 6). In the case of the other compounds, the sensor (GC/Fc-den-TBAHFP-Nafion®) after the measure-
the registered changes in the intensity of the Fc-den current ment in a pure 100 mM TBAP aqueous solution, and the
signal were insignificant. This high sensitivity and selectivity sensor was ready for re-use, which is a very important factor
of Fc-den to 9,10-DPA are probably due to the steric fit; it is from the viewpoint of the application of the designed method-
possible to form a complex by π–π interactions.
ology. It should be also stressed that analytical studies were
Having known the above discussed specificity and inspired performed in an aqueous environment with small addition of
by the fact that 9,10-diphenylantracene can be used as a tem- DCM (not higher than 3%). Finally, it is worth noting that no
plate during the synthesis steps of Fc-den (Fig. 3b), we began significant interaction between the parent subunit structure of
Fig. 6 DPV voltammograms of GC/Fc-den-TBAHFP-Nafion® recorded
in the presence of various aromatic analytes in an aqueous solution of
00 mM TBAP. Canalyte = 200 μM. The dotted line stands for the spec-
trum of native Fc-den.
Fig. 7 DPV
voltammograms
of
a
receptor
(GC/Fc-den-
1
TBAHFP-Nafion®) recorded at various concentrations of 9,10-DPA in an
aqueous solution of 100 mM TBAP. Inset: Calibration plots.
2490 | Dalton Trans., 2021, 50, 2483–2492
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
Table 2 (A) Calibration equation obtained in aqueous solution and (B) recovery results
2
Regression equation
R
Analytical range [μM]
LOD [μM]
RSD (%)
A: Calibration equation obtained in an aqueous solution of 100 mM TBAP (n = 3)
3
I = (1.83 ± 0.03)·10- C9,10DPA + (0.096 ± 0.003)
0.9995
0.1–200
0.06
4.5
9,10-DPA added [μM]
9,10-DPA found [μM]
RSD (%) intraday
RSD (%) interday
Recovery (%)
B: Recovery results of TBAP solution spiked by 9,1-DPA (n = 3)
5
5
1
.00
0.0
00
4.90
50.5
101
3.3
4.2
4.8
5.2
5.0
4.9
98
99
99
Fc-den, i.e., 1 without formyl functionalities, which was
recently reported by our group, was observed during the
Acknowledgements
3
0
electrochemical assays. This means that the organized, Financial support from Warsaw University of Technology
π-conjugated structure of Fc-den is indeed essential for provid- (WUT) is acknowledged. This work was implemented as a part
ing the studied electrochemical recognition ability.
of Operational Project Knowledge Education Development
2014–2020 co-financed by the European Social Fund, project
no. POWR.03.02.00-00-I007/16-00 (POWER 2014–2020). We
would like to thank Dr Magdalena Poplawska (WUT) for fruit-
ful discussions and for sharing her knowledge in organic syn-
thesis over the years of our scientific collaboration.
Conclusions
In conclusion, herein it is shown that the π-conjugated ferro-
cene-containing dendrimer (Fc-den) can be successfully pre-
pared employing facile, efficient, chromatography-free pro-
cesses that proceed under mild conditions. Noteworthily, the
presented total synthesis can be regarded as time-, solvent-
and cost-saving since there is no need for chromatographic
purification at each synthetic step. While the imine-bond for-
mation reaction has been demonstrated to be a convenient
way for the synthesis of many valuable π-conjugated mole-
cules, the presented total synthesis provides important tools
for chromatography-free methods in total synthesis as well as
for controlling the dynamic covalent chemistry behaviours
with ferrocene-tethered systems. Electrochemical studies with
Fc-den revealed one pair of ferrocenes’ current signals.
Ultimately, the title dendrimer has been employed for the con-
struction of an electrochemical sensor dedicated to the reco-
gnition of 9,10-diphenylanthracene. The limit of detection
value equalled 0.06 μM, and importantly receptor layers con-
sisting of Fc-den did not require regeneration. Taking into
account the easiness of the preparation of the target dendri-
mer under mild conditions, the possibility of its gram-scale
synthesis, as well as its property to recognize selected aromatic
species and to track this process under electrochemical con-
ditions, it is believed that this work would open up new
avenues in the chemistry of ferrocene-bearing molecules. Our
future work in this topic will include, e.g. the preparation of
further generations of Fc-den and new Fc-den-derived highly
ordered molecules, as well as the evaluation of their properties
and applications.
Notes and references
1
2
D. Astruc, Eur. J. Inorg. Chem., 2017, 6–29.
K. Heinze and H. Lang, Organometallics, 2013, 32, 5623–
5625.
3
S. Daum, V. F. Chekhun, I. N. Todor, N. Y. Lukianova,
Y. V. Shvets, L. Sellner, K. Putzker, J. Lewis, T. Zenz,
I. A. M. de Graaf, G. M. M. Groothuis, A. Casini, O. Zozulia,
F. Hampel and A. Mokhir, J. Med. Chem., 2015, 58, 2015–
2024.
4
5
6
7
M. Nakahata, Y. Takashima, H. Yamaguchi and A. Harada,
Nat. Commun., 2011, 2, 511.
A. Kasprzak and P. A. Guńka, Dalton Trans., 2020, 49, 6974–
6979.
T. Romero, A. Caballero, A. Tarraga and P. Molina, Org.
Lett., 2009, 11, 3466–3469.
M. S. Inkpen, S. Scheerer, M. Linseis, A. J. P. White,
R. F. Winter, T. Albrecht and N. J. Long, Nat. Chem., 2016,
8
, 825–830.
8
9
L. E. Wilson, C. Hassenrgck, R. F. Winter, A. J. P. White,
T. Albrecht and N. J. Long, Angew. Chem., Int. Ed., 2017, 56,
6838–6842.
E. Abbasi, S. F. Aval, A. Akbarzadeh, M. Milani,
H. T. Nasrabani, S. W. Joo, Y. Hanifehpour, K. Nejati-
Koshki and R. Pashaei-Asl, Nanoscale Res. Lett., 2014, 9,
247.
1
0 (a) K. Ariga, H. Ito, J. P. Hill and H. Tsukube, Chem. Soc.
Rev., 2012, 41, 5800–5835; (b) W.-J. Li, W. Wang,
X.-Q. Wang, M. Li, Y. Ke, R. Yao, J. Wen, G.-Q. Yin, B. Jiang,
X. Li, P. Yin and H.-B. Yang, J. Am. Chem. Soc., 2020, 142,
Conflicts of interest
There are no conflicts to declare.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 2483–2492 | 2491

View Article Online
Paper
Dalton Transactions
8
473–8482; (c) B. Jiang, L.-J. Chen, Y. Zhang, H.-W. Tan,
E. Manoury, G. Balavoine and J.-P. Majoral, Macromolecules,
2000, 33, 7328–7336; (c) C.-O. Turrin, J. Chiffre, J.-C. Daran,
D. de Montauzon, G. Balavoine, É. Manoury,
A.-M. Caminade and J.-P. Majoral, C. R. Chim., 2002, 5,
309–318.
L. Xu and H.-B. Yang, Chin. Chem. Lett., 2016, 27, 607–612.
1 S. Nlate, J. Ruiz, J.-C. Blais and D. Astruc, Chem. Commun.,
2
2 B. Alonso, I. Cuadrado, M. Moran and J. Losada, J. Chem.
Soc., Chem. Commun., 1994, 2575–2576.
3 (a) C.-O. Turrin, E. Manoury and A.-M. Caminade,
1
1
1
000, 417–418.
20 E. David, K. Thirumoorthy and N. Palanisami, Appl.
Organomet. Chem., 2018, 32, e4522.
Molecules, 2020, 25, 447; (b) C. Ornelas, J. R. Aranzaes, 21 (a) D. Kaleeswaran, P. Vishnoi and R. Murugavel, J. Mater.
E. Cloutet, S. Alves and D. Astruc, Angew. Chem., Int. Ed.,
007, 46, 872–877; (c) J. Camponovo, J. Ruiz, E. Cloutet and
D. Astruc, Chem. – Eur. J., 2009, 15, 2990–3002.
Chem. C, 2015, 3, 7159–7171; (b) J. Palomero, J. A. Mata,
F. Gonzalez and E. Peris, New J. Chem., 2002, 26, 291–297.
22 E. G. Percastegui, J. Mosquera, T. K. Ronson, A. J. Plajer,
M. Kieffer and J. R. Nitschke, Chem. Sci., 2019, 10, 2006–
2018.
2
1
4 (a) S. Sengupta and S. K. Sadhukhan, Organometallics,
2
001, 20, 1889–1891; (b) E. S. Serkova, A. A. Chamkin,
K. L. Boldyrev and Z. B. Shifrina, Macromolecules, 2020, 53, 23 (a) Templated Organic Synthesis, ed. F. Diederich and
2
735–2743.
P. J. Stang, WILEY-VCH Verlag GmbH, Weinhein, Germany,
2007, ISBN: 978-3-527-61352-6; (b) Y. Xu, R. Kaur, B. Wang,
M. B. Minameyer, S. Gsanger, B. Meyer, T. Drewello,
D. M. Guldi and M. von Delius, J. Am. Chem. Soc., 2018,
140, 13413–13420.
1
1
5 B. J. Ravoo, Dalton Trans., 2008, 1533–1537.
6 (a) C. M. Cardona and A. E. Kaifer, J. Am. Chem. Soc., 1998,
1
20, 4023–4024; (b) L. Wang, L.-J. Chen, J.-Q. Ma,
C.-H. Wang, H. Tan, J. Huang, F. Xiao and L. Xu,
J. Organomet. Chem., 2016, 823, 1–7; (c) M.-C. Daniel, 24 (a) Y. Cohen, L. Avram and L. Frish, Angew. Chem., Int. Ed.,
J. Ruiz and D. Astruc, J. Am. Chem. Soc., 2003, 125, 1150–
151; (d) Q.-J. Li, G.-Z. Zhao, L.-J. Chen, H. Tan,
C.-H. Wang, D.-X. Wang, D. A. Lehman, D. C. Muddiman
2005, 44, 520–554; (b) F. Parenti, F. Tassinari, E. Libertini,
M. Lanzi and A. Mucci, ACS Omega, 2017, 2, 5775–5784;
(c) L. Avram and Y. Cohen, J. Org. Chem., 2002, 67, 2639–
2644.
1
and H.-B. Yang, Organometallics, 2012, 31, 7241–7247;
(e) Q. Han, Q.-J. Li, J. He, B. Hu, H. Tan, Z. Abliz, 25 T. Berbasova, E. M. Santos, M. Nosrati, C. Vasileiou,
C.-H. Wang, Y. Yu and H.-B. Yang, J. Org. Chem., 2011, 76,
660–9669; (f) G.-Z. Zhao, L.-J. Chen, C.-H. Wang,
J. H. Geiger and B. Borhan, ChemBioChem, 2016, 17, 407–
414.
9
H.-B. Yang, K. Ghosh, Y.-R. Zheng, M. M. Lyndon, 26 A. J. King, Y. V. Zatsikha, T. Blessener, F. Dalbec, P. C. Goff,
D. C. Muddiman and P. J. Stang, Organometallics, 2010, 29,
137–6140.
M. Kayser, D. A. Blank, Y. P. Kovtun and V. N. Nemykin,
J. Organomet. Chem., 2019, 887, 86–97.
6
1
7 A. Salmon and P. Jutzi, J. Organomet. Chem., 2001, 637–639, 27 P. Molina, A. Tárraga, D. Curiel and M. D. Velasco,
95–608. J. Organomet. Chem., 2001, 637–639, 258–265.
8 (a) L. Xu, Y.-X. Wang, L.-J. Chen and H.-B. Yang, Chem. Soc. 28 T. Yamaguchi, K. Takahashi and T. Komura, Electrochim.
Rev., 2015, 44, 2148–2167; (b) S. Ø. Scottwell and Acta, 2001, 46, 2527–2535.
J. D. Crowley, Chem. Commun., 2016, 52, 2451–2464; 29 (a) M. Vilas-Boas, E. M. Pereira, C. Freire and
5
1
(
(
c) E. Perris, Coord. Chem. Rev., 2004, 248, 279–297;
d) L. Xu, L.-J. Chen and H.-B. Yang, Chem. Commun., 2014,
A. R. Hillman, J. Electroanal. Chem., 2002, 538, 47;
(b) R. J. LeSuer and W. E. Geiger, Angew. Chem., Int. Ed.,
2000, 39, 248–250; (c) F. Barrière, N. Camine and
W. E. Geiger, J. Am. Chem. Soc., 2002, 124, 7262–7263;
(d) F. Barrière and W. E. Geiger, J. Am. Chem. Soc., 2006,
128, 3980–3989; (e) F. Barrière and W. E. Geiger, Acc. Chem.
Res., 2010, 43, 1030–1039.
5
0, 5156–5170; (e) G.-Y. Wu, L.-J. Chen, L. Xu, X.-L. Zhao
and H.-B. Yang, Coord. Chem. Rev., 2018, 369, 39–75;
f) M. Saito, H. Shinokubo and H. Sakurai, Mater. Chem.
(
Front., 2018, 2, 635–661; (g) W. Wang, L.-J. Chen,
X.-Q. Wang, B. Sun, X. Li, Y. Zhang, J. Shi, Y. Yu, L. Zhang,
M. Liu and H.-B. Yang, Proc. Natl. Acad. Sci. U. S. A., 2015, 30 A. Kasprzak, P. A. Guńka, A. Kowalczyk and A. M. Nowicka,
1
12, 5597–5601; (h) D. Astruc, C. Ornelas and J. Ruiz, Acc.
Dalton Trans., 2020, 49, 14807–14814.
Chem. Res., 2008, 41, 841–856.
9 (a) C. Schubert, J. T. Margraf, T. Clark and D. M. Guldi,
31 S. B. Lee, S. N. Park, C. Kim, H. W. Lee, H. W. Lee,
Y. K. Kim and S. S. Yoon, Synth. Met., 2015, 203, 174–179.
1
Chem. Soc. Rev., 2015, 44, 988–998; (b) C.-O. Turrin, 32 H. Chen, W. Liang, Y. Chen, G. Tian, Q. Dong, J. Huang
J. Chiffre, D. de Montauzon, J.-C. Daran, A.-M. Caminade, and J. Su, RSC Adv., 2015, 5, 70211–70219.
2492 | Dalton Trans., 2021, 50, 2483–2492
This journal is © The Royal Society of Chemistry 2021