Mechanical Activation of Terpyridine Metal Complexes in Polymers

  • 266 Accesses

  • 1 Citations


The mechanical addressability of specific chemical bonds holds a high potential for the improvement of polymeric materials. While in many cases, mechanical forces applied to polymers lead to bond scissoring and materials failure, including mechanophores into the polymer structure can lead to stimuli-responsive materials reacting to an applied force in a predefined manner. In this contribution, the mechanical addressability of bis-terpyridine metal complexes embedded into a polymer structure is investigated. The activation of the transition metal complexes in the metallopolymer is monitored by adding a fluorescent sensor molecule to the metallopolymer solution during ultrasonication. Upon sonication, the activation of the complexes leads to fluorescence-quenching of the sensor. The dependency of the metal ion and the type of polymer as well as their molar mass is investigated in detail, showing that this concept could possibly be used in further application of stimuli-responsive or self-healing materials.


The influence of mechanical force is often visible for polymeric materials utilized in everyday life. Here, material fatigue as a result of bond scissoring by mechanical elongation of bonds in the polymer backbone can be observed [1]. In contrast to this unfavored mechanical activation, the chains could act as a handle to lead the force towards a functional group, which will react in a predefined manner [2]. Such functional groups, called mechanophores, can show various reactions upon mechanical treatment [2].

One major class of functional groups investigated are spiropyranes as well as cyclopropanes and -butanes, which undergo ring opening under mechanical stress [3,4,5]. Furthermore, supramolecular bonds can also be applied as labile bonds in polymers to be addressable via mechanical force [6]. A significant benefit of this non-covalent bond is the tunability of the binding strength [7]. One particular interesting functional group in this context are metal complexes [7]. Thus, the interaction between the metal center and the corresponding ligands can be addressed by several different stimuli, e.g., mechanical force [7], temperature [8] or light [9], resulting in a debonding/decomplexation. This behavior can be utilized for several different applications, in particular catalysis [10]. The latent catalyst is integrated into a polymeric structure and can be activated by application of mechanical force [11]. This approach was already utilized for the catalysis of a range of different reactions such as ring-opening metathesis polymerization (ROMP) [11, 12] or alkyne-azide click reactions [13].

Generally, there are only a few examples in literature showing the potential utilization of metal complexes for the activation via mechanical forces [7]. Thus, carbenes, as well as 2,6-bis(1′-methylbenzimidazolyl)-pyridines were only applied as ligand structures for the mechanical activation in polymers [7, 10, 12, 13]. Furthermore, there is no real correlation between the choice of metal–ligand interaction and the addressability via mechanical force. For this purpose, the current study will focus on the utilization of terpyridine–metal complexes to elucidate a more general understanding of the influence of the chosen metal complex and the addressability via mechanical force.

Terpyridine–metal complexes were chosen due to their large variety of different binding strengths [14]. Additionally, this ligand entity was already utilized in polymer science to design metallopolymers [15,16,17], which are further able to be addressed by different stimuli, e.g., temperature [8]. This behavior could also be applied for different applications like self-healing behavior [18]. Consequently, the utilization of terpyridine–metal interactions for the investigation of addressability by mechanical forces offers the potential to correlate the binding strength with the stimuli-responsiveness. In future projects, embedding these responsive functional groups into polymers can be interesting for creating sensors or stimuli-responsive catalysts. Furthermore, intrinsically healable polymers can be designed in such a fashion that the healing process is initiated by the damage itself, due to the mechanical activation of the complex.

Experimental Part

Materials and Instruments

All chemicals were used as received from TCI, Sigma Aldrich, Acros organics, Spectrum chemical, Molekula, HetCat or Alfa Aesar if not otherwise stated. All solvents were dried over molecular sieve under nitrogen-atmosphere.

Size exclusion chromatography measurement were performed on the following setup: Shimadzu with CBM-20A (system controller), DGU-14A (degasser), LC-10AD vp (pump), SIL-10AD vp (auto sampler), Techlab (oven), SPD-10AD vp (UV detector), RID-10A (RI detector), PSS SDV guard/linear S (5 μm particle size), DMAc + 0.21 wt% LiCl with 1 mL/min at 40 °C, poly(methyl methacrylate) (standard), as well as Shimadzu with CBM-20A (system controller), DGU-14A (degasser), LC-20AD (pump), SIL-20AHT (auto sampler), CTO-10AC vp (oven), SPD-20A (UV detector), RID-10A (RI detector), PSS SDV guard/1000 Å/1,000,000 Å (5 μm particle size) chloroform/isopropanol/triethyl-amine [94/2/4] with 1 mL/min at 40 °C, poly(methyl methacrylate) (standard), or Shimadzu with CMB-20A (system controller), DGU-14A (degasser), LC-10AD vp (pump), SIL-10AD vp (auto sampler), CTO-10A vp (oven), SPD-10AD VP (UV detector), RID-10A (RI detector), PSS SDV guard/linear M THF with 1 mL/min at 30 °C, poly(ethylene glycol) (standard).

All titrations were performed using a standard volume Nano ITC (TA Instruments) at 303 K. Solutions were always prepared prior to use in dry ligand using vacuum dried ligand and metal salt. Blank titrations in dry ligand were performed and subtracted from the corresponding titrations to remove the effect of dilution. The fitting of the measured data was performed with the NanoAnalyze program from TA instruments.

UV–Vis measurements were performed using a Lambda 950 spectrometer, scanning speed: 500 nm/min, slit width: 1 nm. Emission measurements were performed with a JASCO FP 6500 Fluorometer, band width (ex.): 3 nm, band width (em.): 3 nm, scanning speed: 500 nm/min, response: 0.1 s, medium sensitivity.

Differential scanning calorimetry (DSC) was measured on a Netzsch DSC 204 F1 Phoenix instrument under a nitrogen atmosphere with a heating rate of 10 and 20 K/min.

Nuclear magnetic resonance spectra were measured using a Bruker AC 300 (300 MHz), Bruker AC 250 (250 MHz) and a Bruker AC 400 (400 MHz) spectrometers at 298 K if not stated differently. The chemical shift is given in parts per million (ppm on δ-Scale) related to deuterated solvent.

Elemental analysis was performed with a Vaio El III (Elementar).

Ultrasonication was performed with a Hielscher UP200St, UP200St-T (electro-acoustic transducer), UP200St-G (generator), S26d7 titan sonotrode at a working-frequency of 26 kHz, 100% Amplitude, 120 W and 0.8 s pulsation time.

4′-(Anthracen-9-yl)-2,2′:6′,2″-terpyridine (1) [19], 5-([2,2′:6′,2″-terpyridin]-4′-yloxy)pentan-1-amine (2) [20] and 6-(2,2′:6′,2″-terpyridin-4′-yloxy)-hexan-1-ol (3) [18] were synthesized according to literature and the synthetic procedures can be found in the supporting information.

The viscosity measurements were performed with the AMVn viscometer (Anton Paar, Graz, Austria) in analogy to that reported previously [21, 22]. The partial specific volume (\( \upsilon \)) of the polymer was determined with the DMA4100 densimeter (Anton Paar, Graz, Austria) at \( T = 20 \,^\circ {\text{C}} \) similar to the procedure utilized recently [21, 22]. Sedimentation velocity experiments were performed using a ProteomeLab XL-I analytical ultracentrifuge (Beckman Coulter Instruments, Brea, CA) with an An-50 Ti eight-hole rotor, at a the rotor speed of 42,000 rpm, and at a temperature of \( T = 20 \,^\circ {\text{C}} \). Ultracentrifuge cells were equipped with double-sector aluminum centerpieces with a 12 mm optical path length. The sedimentation velocity data were analyzed with SEDFIT and the \( c\,(s) \) model with a maximum entropy regularization procedure. Further experimental details are reported in the Supporting Information.

RAFT-Polymerization of Methyl Methacrylate

In a one-neck round bottom flask, the respective amount of monomer, 4-cyano-4-(phenylcarbono-thioylthio)pentanoic acid N-succinimidyl ester (RAFT-agent) and 2,2′-azobis(2-methylpropionitrile) (AIBN) were dissolved in dry N,N-dimethylformamide (DMF) and toluene, respectively, resulting in a 2.0 M solution. The mixture was purged with nitrogen for 45 min. Afterwards, the mixture was stirred and heated to 70 °C for 17 h. The mixture was allowed to cool to room temperature and the polymer was precipitated in cold diethyl ether. P1 and P2 were purified by dissolving and dialysis in THF (MWCO: 3500 g/mol). The THF was replaced twice a day and dialysis was stopped after 2 days. Both polymers were dried in vacuo after removing the solvent under reduced pressure at 40 °C. All masses and details are listed in Table 1.

Table 1 Details of the RAFT-polymerization of P1 and P2


1H NMR (300 MHz, CDCl3): δ = 0.85–1.02 (s, CH2 polymer-backbone), 1.82–2.03 (CH3 polymer-backbone), 3.60 (O=C–O–CH3) ppm

SEC (chloroform/iso-propanol/triethylamine [94/2/4], PMMA-standard): Mn = 60,400 g/mol; Mw = 70,900 g/mol; Ð = 1.17

$$ \begin{array}{*{20}l} {\text{Elemental analysis}}& {\text{calculated:}} &{\text{C:}} 59.89\%,&{\text{H:}} 8.05\%, \\ {} & {{\text{found:}}} & {\text{C:}} 58.59\%, &{\text{H:}} 7.84\% \end{array} $$

DSC: Tg = 125 °C


1H NMR (300 MHz, CDCl3): δ = 0.85–1.02 (s, CH2 polymer-backbone), 1.82–1.98 (CH3 polymer-backbone), 3.60 (O=C–O–CH3) ppm

SEC (DMAc + 0.21% LiCl, PMMA standard):

Mn = 86,500 g/mol; Mw = 104,600 g/mol; Ð = 1.20

$$ \begin{array}{*{20}l} {\text{Elemental analysis}}& {\text{calculated:}} &{\text{C:}} 59.96\%,&{\text{H:}} 8.04\%, \\ {} & {{\text{found:}}} & {\text{C:}} 52.62\%, &{\text{H:}} 7.03\% \end{array} $$

DSC: Tg = 126 °C

End Group Modification of P1 and P2

The respective amount of polymer was dissolved in either dimethylsulfoxide (DMSO) (P3) or N,N-dimethylformamide (DMF) (P4) in a 250 mL round bottom flask and purged with nitrogen for 30 min (details are listed in Table 2). To the dissolved polymer, two equivalents based on the molar mass of the polymer of 5-([2,2′:6′,2″-terpyridin]-4′-yloxy)pentan-1-amine (2), dissolved in DMF or DMSO, were added via syringe. The mixture was stirred at 50 °C overnight, DMF was removed under reduced pressure and THF was added before purification via dialysis in THF (MWCO: 3500 g/mol). The DMSO was removed by dialysis in THF. Dialysis was performed for 2 days, the THF was changed twice a day. After 2 days, the membrane was opened and the solvent was removed under reduced pressure before drying the polymer in vacuo.

Table 2 Details of the end-group modification of P1 and P2 to P3 and P4


1H NMR (300 MHz, CDCl3): δ = 0.86–1.03 (s, CH2 polymer-backbone), 1.82–1.98 (CH3 polymer-backbone), 3.61 (O=C–O–CH3) ppm

SEC (chloroform/iso-propanol/triethylamine [94/2/4], PMMA-standard): Mn = 77,500 g/mol; Mw = 83,500 g/mol; Ð = 1.08

$$ \begin{array}{*{20}l} {\text{Elemental analysis}}& {\text{calculated:}} &{\text{C:}} 59.98\%,&{\text{H:}} 8.08\%, & {}\\ {} & {{\text{found:}}} & {\text{C:}} 59.60\%, &{\text{H:}} 8.37\%, & {\text{S:}} 0.49\%\end{array} $$

DSC: Tg = 120 °C


1H NMR (300 MHz, CDCl3): δ = 0.85–1.02 (s, CH2 polymer-backbone), 1.82–1.98 (CH3 polymer-backbone), 3.60 (O=C–O–CH3) ppm

SEC (DMAc + 0.21% LiCl, PMMA standard)

Mn = 108,400 g/mol; Mw = 124,100 g/mol; Ð = 1.14

$$ \begin{array}{*{20}l} {\text{Elemental analysis}}& {\text{calculated:}} &{\text{C:}} 59.98\%,&{\text{H:}} 8.08\%, & {}\\ {} & {{\text{found:}}} & {\text{C:}} 59.50\%, &{\text{H:}} 8.42\%, & {\text{N:}} 0.32\%\end{array} $$

DSC: Tg = 126 °C

Synthesis of Terpyridine Functionalized PTHF (P5)

PTHF was synthesized by adapting a procedure from Jakobs et al. [23].

A 500 mL three-neck round bottom flask with a stirring bar was flame dried with three vacuum-nitrogen-cycles. 100 mL (1.233 mol) of dry THF were transferred via a syringe into the flask and cooled to 0 °C. 0.04 mL (0.178 mmol) of di-tert-butyl pyridine was added via a syringe and the mixture was stirred at 0 °C for 10 min. To the mixture, 0.1 mL (0.883 mmol) of methyl triflate was added and the mixture was further stirred at 0 °C. After 8 h, the reaction was quenched by adding 0.345 g (1.0 mmol) of 6-(2,2′:6′,2″-terpyridin-4′-yloxy)-hexan-1-ol (3) dissolved in 2 mL of THF. The polymer was concentrated under reduced pressure at 40 °C and the polymer was purified by dialysis in THF with a molecular weight cut-off (MWCO) of 1000 g/mol, THF was changed twice a day. After 2 days, dialysis was stopped and the polymer was dried in vacuo.


1H NMR (300 MHz, CDCl3): δ = 1.61 (s, –O–CH2–CH2–CH2–CH2–), 3.41 (s, –O–CH2 CH2–CH2–CH2–) ppm

SEC (THF, PEG standard): Mn = 17,600 g/mol; Mw = 22,300 g/mol; Ð = 1.20

$$ \begin{array}{*{20}l} {\text{Elemental analysis}}& {\text{calculated:}} &{\text{C:}} 66.63\%,&{\text{H:}} 11.18\%, & {}\\ {} & {{\text{found:}}} & {\text{C:}} 65.50\%, &{\text{H:}} 10.83\%, & {\text{N:}} 0.25\%\end{array} $$

DSC: Tg not detected, Tm = 28 °C

Synthesis of Metallopolymers

Synthesis of P3 and P4 Based Metallopolymers and Ultrasound Investigations

The respective amount of polymer of P3 or P4 was dissolved in THF/methanol mixture to give a 4.5 × 10−5 M solution and 0.5 equivalents of the respective metal salt stock solution (8 mM in methanol) was added (for details see Table 3). The round bottom flask was sealed with a septa. Then, the mixture was stirred for an hour and flushed with nitrogen in the last 15 min. The reaction mixtures were cooled to 6 to 8 °C. Afterwards, 4′-(anthracen-9-yl)-2,2′:6′,2″-terpyridine (1) was added under nitrogen flow from a stock solution (1.3 mM, in THF/methanol) in an equimolar amount to the respective polymer, a fluorescence sample was taken and the flask was attached on the sonotrode of the ultrasonicator. The mixture was cooled with ice during the pulsed ultrasonication (0.8 s ultrasound, 0.2 s break) of 120 W and an amplitude of 100%. Sonication was performed for 1 h and fluorescence samples were taken from the metallopolymer-mixture every 15 min.

Table 3 Details on the metallopolymer formation based on the polymers P3 and P4 and the solutions for sonication

To all metallopolymers, control experiments with the same setup were performed. The metallopolymers were formed as described for the sonication experiments. After addition of 1 from a stock solution (1.3 mM in THF/methanol) under nitrogen flow the metallopolymer solutions were cooled to 6 to 8 °C and emission spectra were measured immediately (details see supporting information, Table S2). The metallopolymer solutions were then stirred for 1 h and emission spectra were measured again after 60 min.

The polymers P3 and P4 were also tested towards sonication in presence of 1. For this purpose, the polymer was dissolved in THF/methanol to give a 4.5 × 10−5 M solution without the addition of metal salts, flushed with nitrogen for 15 min, followed by cooling to 6 to 8 °C before adding 1 from a 1.3 mM stock solution (details see supporting information, Table S4). Fluorescence spectrum was measured immediately and sonication was started for 1 h. After sonication stopped, a sample was taken and the fluorescence spectrum was measured again.

Synthesis of P5 Based Metallopolymers and Ultrasound Investigation

The metallopolymers based on P5 were obtained by dissolving the polymer in 75 mL chloroform and dividing the solution into five equal parts. To each solution the respective amount of metal salt (8 mM solution in methanol) was added and stirred overnight (details see Table 4). Afterwards, the metallopolymer solutions were subject to dialysis for 2 days (MWCO: 25,000 g/mol), THF was changed twice a day. The solvent was removed under reduced pressure and the metallopolymers was dried in vacuo.

Table 4 Details on the metallopolymer synthesis starting from P5

For sonication, the metallopolymers were dissolved in a THF/methanol mixture (1:1) to result in a 4.5 × 10−5 M solution and flushed with nitrogen for 15 min followed by cooling to 6 to 8 °C and addition of 1 (1.3 mM, in THF/methanol) in an equimolar amount compared to the metallopolymers (details see Table 5). Emission spectra were measured before sonication and every 30 min during sonication. The mixture was cooled with ice during the pulsed ultrasonication (0.8 s ultrasound, 0.2 s break, output: 120 W, amplitude: 100%) and sonication was performed for 1 h.

Table 5 Details on the composition of the metallopolymer solutions for sonication based on the P5 metallopolmyers

To all metallopolymers, control experiments with the same setup were performed. The metallopolymers were formed as described for the sonication experiment (details see supporting information, Table S3). After addition of 1 from a stock solution (1.3 mM in THF/methanol) under nitrogen flow, the metallopolymer solutions were cooled to 6 to 8 °C and emission spectra were measured immediately. The metallopolymer solutions were then stirred for 1 h and emission spectra were measured again every 30 min.

Furthermore, the polymer P5 was tested via sonication in the presence of 1. For this purpose, the polymer was dissolved in THF/methanol to give a 4.5 × 10−5 M solution without addition of metal salts, flushed with nitrogen for 15 min, followed by cooling to 6 to 8 °C before adding 1 from a 1.3 mM stock solution (details see supporting information, Table S4). Fluorescence spectrum was measured immediately and sonication was started for 1 h. After sonication stopped, a sample was taken and the fluorescence spectrum was measured again.

Results and Discussion

Synthesis and Characterization of 4′-(Anthracen-9-yl)-2,2′:6′,2′′-terpyridine (1)

For the mechanical activation in solution, it is required to measure the amount of activated transition metal ions from the metallopolymer by sonication. For this purpose, a fluorescent sensor molecule should be added to the solution enabling a binding of to the activated metal ions resulting in a quenching of its fluorescence upon complex formation. The ligand motif of the sensor should be similar to the ligand attached to the polymer. Thus, a terpyridine-substituted anthracene was utilized for this purpose: 4′-(anthracen-9-yl)-2,2′:6′,2′′-terpyridine (1, see Fig. 1) [19]. The synthesis of the sensor was performed according to Maheshwaran et al. and is described in the supporting information [19]. The similar structure should provide a similar binding efficiency compared to the terpyridine-functionalized polymer chains. Furthermore, this similar complexation behavior would not shift the complex equilibrium of the metallopolymers.

Fig. 1

Fluorescence intensity of 1 at 420 nm during titration with different metal salts (FeSO4 × 7 H2O, NiCl2, CoCl2, MnCl2 × 4 H2O and ZnCl2). X: 2 Cl, SO42−

By UV–Vis titration, the obtained spectra gave additional information about complexation processes (see supporting information, Figs. S17 to S21). For titration with iron(II)-sulfate solution, the spectrum showed the characteristic MLCT band at λmax ~ 562 nm [24]. UV–Vis titration with MnCl2 showed that in fact a complexation must take place, since the UV–Vis spectrum showed a change in the ligand centered absorption at λLigand ~ 330 nm [19]. This change indicates a conformational change from trans–trans to cis–cis conformation of the terpyridine upon interaction with the manganese(II)-ions [25].

More importantly, fluorescence spectra were measured in the presence of different metal salt contents. The amount of transition metal ion is plotted against the emission intensity in Fig. 1. In all cases, the intensity of the fluorescence is decreasing with increasing amount of the respective transition metal salt. Maheshwaran et al. postulated that the electronic conjugation of the π-electrons is disrupted upon complex formation with copper(II)-ions and the terpyridine-unit [19]. Since for all of our investigated metal salts, complex formation was indicated by UV–Vis titration, it is likely that the emission quenching is a result of the interaction and complex formation with the transition metal ions.

The complexation behavior was furthermore investigated via isothermal titration calorimetry (ITC) measurements and the results are summarized in Table 6. For all experiments, the ligand was added dropwise to the metal salts in the measurement cell. The results show that approximately for all of the metal salts two ligand molecules interact with one metal ion. Furthermore, the binding strength indicated that manganese(II) forms the weakest metal complexes, which is in good agreement with reports in literature [18]. In contrast, iron(II), cobalt(II) and nickel(II) forms rather stable complexes with a binding strength of ca. 1010 (see Table 6). Remarkably, compared to 2,2′:6′,2″-terpyridine the sensor always revealed lower values for the binding constant to all metal ions investigated indicating that the anthracene-moiety reduces the complex-formation with the metal ions. Remarkably, the difference in the values for the binding constant Ka of 2,2′:6′,2″-terpyridine was minor for iron-, nickel- and cobalt metal salts.

Table 6 Results of the ITC investigation of the sensor molecule 1 with different metal salts

Polymer-Synthesis, End Group Modification and Metallopolymer-Synthesis

In order to prepare metallopolymers, which are suitable for the investigation of the stimuli-responsive behavior, it is required to design polymers featuring a ligand entity, i.e. terpyridine, at the chain end. As a consequence, controlled or living polymerization techniques have to be utilized. The first approach focusses on the reversible addition fragmentation chain transfer (RAFT) polymerization as a controlled radical polymerization technique [26]. This technique enables to obtain polymers with narrow molar mass distributions and defined end-groups, suppressing termination-reactions leading to undefined end groups [27]. As monomer, methyl methacrylate (MMA) was utilized and the chain transfer agent (CTA) carried a N-hydroxy-succinimide-moiety (NHS-moiety) resulting in polymers P1 and P2. The NHS-ester moiety was extensively investigated and was shown to be very sensitive towards the substitution with primary amines [28]. This behavior was utilized for the substitution of the NHS-moiety with a terpyridine moiety by adding 5-(2,2′:6′,2′′-terpyridin-4′-yloxy)pentylamine (2) (see Fig. 2). Afterwards, the addition of metal salts results in the formation of polymer–dimers/metallopolymer).

Fig. 2

Schematic representation of the performed polymer synthesis and post-polymerization modification

After polymerization and end group modification, purification was performed by dialysis in tetrahydrofuran (THF) to remove low molar mass impurities. The polymers were analyzed using size exclusion chromatography (SEC). Table 7 compares the SEC results before and after end group modification. The apparent molar mass of P3 as well as P4 has increased after end group modification. This can possibly be explained by the end group modification, since the attached terpyridine-moiety consists of a higher molar mass compared to the removed NHS-moiety, and also by further purification steps after modification, leading to a possible loss of shorter polymer chains. In addition, due to the polymer chain modification the hydrodynamic volume may change, which might affect the elution of the polymers in the SEC. In the 1H NMR spectrum of P3 as well as of P4, it was difficult to proof the presence of the terpyridine-moiety (see supporting information, Figs. S10 and S11). Unfortunately, by zooming into the region between 7 and 9 ppm in the spectra of P3 and P4, no peaks can be found that could be assigned to the terpyridine-end group. Based on the SEC results of the polymers, there is approximately one terpyridine unit to 774 methyl methacrylate repeating units for P3 and 1082 repeating units on one terpyridine for P4. These ratios make a 1H NMR based end group investigation very difficult.

Table 7 Molar masses by SEC, UV–Vis titration, and analytical ultracentrifugation of P1 to P5

Poly(tetrahydrofuran) (pTHF, in the following P5) was synthesized according to literature by ring-opening polymerization, initiated with methyl triflate and quenched by addition of 6-(2,2′:6′,2′′-terpyridin-4′-yloxy)-hexan-1-ol (3) [23]. This polymer was purified differently but also investigated with the same methods compared to P3 and P4. SEC was performed, the results in Table 7 show its lower molar mass compared to the other polymers. Synthesis of a higher molar mass pTHF was not successful with the required control of the end-group. The ring opening polymerization is not a controlled polymerization technique, therefore it was challenging to obtain a high molar mass polymer with defined end groups. Increasing the reaction time and further cooling of the polymerization mixture did not lead to any other results than gel formation. The 1H NMR of P5 reveal signals for the terpyridine-moiety in the expected region between 7 and 9 ppm (see supporting information, Fig. S13). The signals can only be observed with weak intensity resulting from the small amount of terpyridine compared to the repeating unit of the polymer (approximately 244 repeating units to one terpyridine-moiety). Integration of the respective terpyridine peaks did not prove that all polymer chains were functionalized with a terpyridine moiety.

To investigate the content of terpyridine moieties attached to polymer end groups, UV–Vis titration with FeSO4-solution was performed for all polymers. The formation of an iron(II)-bis-terpyridine complex is displayed by a color change from an almost colorless polymer solution to a deep purple [24]. This characteristic color change resulting from the metal–ligand charge-transfer (MLCT) interaction upon complex formation is easy to detect and allows a simple and fast analysis regarding the functionalization of the polymers with terpyridine moieties [24]. Upon intensity saturation of the MLCT band in the UV–Vis spectra, it is possible to re-evaluate the molar mass of the polymers premising that each polymer is functionalized with one terpyridine-moiety and two terpyridine-units interact with one iron(II)-ion. However, it is well-known that by adding iron(II)-ions only bis-terpyridine complexes will form, the formation of mono-terpyridine complexes is not reported under such conditions [15, 20, 29]. Under this premise, the amount of iron(II)-ions added to the polymer sample is plotted against the intensity (see Fig. S22). The equivalents of iron(II)-ions is based on the molar mass obtained by SEC. This addition leads to a purple polymer solution with λmax ~ 557 nm for P3 and P4 and λmax ~ 558 nm for P5. For P3 and P4, the saturation point and the corresponding molar masses obtained from this experiment are in good agreement with the values obtained via SEC with only small discrepancies (see Table 7). This underlines that almost every polymer-chain was functionalized with a terpyridine-moiety.

Furthermore, P4 was also analyzed by viscometry, density measurements and sedimentation velocity experiments by analytical ultracentrifugation (AUC) (hydrodynamic characterization, see supporting information). The molar mass determined by AUC can be found in Table 7. The molar mass from such an absolute method compared to the value determined by SEC and UV–Vis titration shows that all values are in good agreement and that the UV–Vis titration can be considered reliable.

For P5 the saturation was already reached at about 0.3 equivalents of added iron(II)-ions also resulting in a higher molar mass calculated from this experiment. Consequently, it can be assumed that polymer chains without terpyridine moieties are present and should be removed before the mechanical addressability is analyzed. This purification was performed after the formation of the metallopolymers. After complexation of the added metal salts (FeSO4 × 7 H2O, NiCl2, CoCl2, MnCl2 × 4 H2O and ZnCl2, respectively), the formed metallopolymers were placed into a dialysis membrane with 25,000 g/mol molar mass cut off (MWCO). This pore size enables a separation of the metallopolymers formed from the polymer containing terpyridines and the p(THF) without any ligand moiety at the chain end. After dialysis, the residual solvent was evaporated and the removed amount of colorless polymer was tested to be terpyridine-free by adding iron(II)-ions after re-dissolving. No color change was observed showing that the terpyridine-carrying chains were still present in the metallopolymers. Finally, the membranes were opened again and the metallopolymers were dried in vacuo.

The metallopolymers derived from P3 and P4, respectively, were obtained in a similar manner to those from P5, except that they were not subject to dialysis. The metallopolymers from P3 and P4 were formed in a THF/methanol mixture directly before sonication, by adding 0.5 equivalents of the respective transition metal salt stock solution. All metallopolymer solutions were directly utilized for the sonication/control experiments and the corresponding metal-containing substances were not isolated for analysis.

Interactions Between Metallopolymers and 1

All solutions, which were utilized in sonication experiments, were flushed with nitrogen, so that sonication-induced reactions with oxygen can be eliminated as a source of error. The sensor was added under nitrogen flow, before the start of the sonication experiments and fluorescence measurements were performed immediately after mixing. All measured spectra were compared in their intensity at their maximum emission at 420 nm, normalized to an equimolar solution of 1 without any metal ions present. The results are shown in Fig. 3 indicating major differences between the different metallopolymers at the beginning of the experiments. The quenching before sonication could either go back to the presence of the metallopolymers interacting with 1 or by directly involving the metal ions from the metallopolymers into complexes with the sensor due to the reversible behavior of the metallopolymers. The first option is unlikely due to the high dilution of the samples, whereas thermal induced ligand exchange before sonication should be suppressed by cooling of the solutions. However, this quenching before sonication can be a result from mixed complexes with sensor and polymer, or by full release of the transition metal ions from the metallopolymer. The investigation of this phenomena and the resulting species was not yet successful.

Fig. 3

Summary of the measured emission intensity at 420 nm of 1 in presence of the metallopolymers before sonication. The intensity is normalized to the emission intensity of an equimolar solution of 1 without any metal ions

From the P3 based metallopolymers the highest emission intensity before sonication is measured for the nickel(II) containing metallopolymer. In literature reports, bis-terpyridine-nickel(II) complexes are often considered to feature the highest complex stability compared to the other metal salts used in the current study [15]. In ITC experiments, this tendency was confirmed, but the tendency of complex formation with 1 was lower compared to unfunctionalized 2,2′;6′,2″-terpyridine, mimicking the polymer end-group. This finding indicates that the metallopolymer-complex is stronger than the complex with 1, resulting in a higher emission intensity since less nickel-ions are bound in a complex with 1.

For the iron(II) and cobalt(II) containing metallopolymers the emission intensity before sonication is relatively low. This was surprising, since the ITC revealed a much higher binding constant to the terpyridine compared to 1 for both metal ions leading to a similar assumption as for the nickel-metallopolymer. However, it is possible that the polymer itself shows some influence on the complex formation, e.g., by coil formation and, thus, changing of the polarity around the bis-terpyridine complex.

Taking a closer look on the metallopolymers based on P5, the highest fluorescence intensity was surprisingly found for the manganese(II)-based metallopolymer, which showed the smallest value for the complexation constant in ITC experiments. The nickel(II) and cobalt(II) containing metallopolymers displayed the lowest intensities measured before sonication, despite ITC experiments revealed a higher stability of these terpyridine-complexes. This finding could underline that the polymer chain itself has an influence on the complex formation and stability. However, the polymer dimerization was previously successfully shown by UV–Vis titration.

Control Experiments

In order to investigate the origin of the varied sensor emission before sonication, a series of control experiments were performed. To allow an easy comparison of the obtained results from all control experiments, all measured emission intensities were normalized to an equimolar sample of the sensor without any further additives. The spread between the normalized values of I60 and I0 is provided in percentage.

For the control experiments, the sensor was mixed with the polymer without any metal salt, the measured emission intensity before and during sonication varied only little in dependency of the respective polymer (see supporting information, Fig. S23). This finding indicates that the equilibrium between the metal ions from the metallopolymer and the sensor are responsible for the quenching of sensor emission before sonication, probably due to a very similar structure of the binding site of the ligand. By cooling of the solution to between 6 and 8 °C, it was expected to reduce this behavior, which was unfortunately not 100% perfect.

Due to the high variation of the sensor fluorescence before sonication, control experiments were performed in which the metallopolymer and the sensor mixture were stirred only without any sonication. This procedure enabled to determine if the reduced emission intensity is in fact caused by releasing metal ions due to sonication or if the quenching occurs by ligand-equilibrium between the metallopolymer and the sensor.

For this purpose, immediately after mixing of metallopolymer and sensor, fluorescence measurements were performed and repeated after stirring for 60 min. The results gave a mixed picture, not only depending on the metal ions but also on the polymer itself. The quenching for the P3 control experiments was found to be highest when cobalt(II)-ions were present (14%, normalized to the sensor emission) in the metallopolymer, followed by iron(II) (11%), nickel(II)- (8%) and zinc(II)-ions (1%). The observed tendency should reflect the results found by ITC investigation, instead, the quenching indicates, that nickel(II)-, iron(II)- and cobalt(II)-ions are likely to form either mixed complexes or complexes with 1, instead of the metallopolymer. We expected that the metal-ions would remain in the metallopolymer, since ITC showed that for all cases the complex formation with the sensor is less favored, compared with terpyridine. We assume that the polymer itself must show major influence on the metallopolymer formation and stability. In the manganese(II)-containing metallopolymers, no quenching was observed in 60 min. Interpreting these results in the context with the performed ITC results, they have a good agreement with the findings of the manganese(II)-containing metallopolymer. The ITC revealed that the value for Ka is relatively small (see Table 6), whereas for 1 there was no interaction observed. However, in fluorescence titration of 1 it was shown that with increasing content of manganese(II)-ions, quenching can be expected. Since no quenching was observed here, the manganese(II)-ions possibly seem to remain in the metallopolymer.

The control experiments of P4 revealed higher values for the quenching in general when metal ions were present. We decided to investigate the metal ions, which are considered to form the most stable bis-terpyridine complexes. When P4 is turned into a metallopolymer and stirred in the presence of the sensor 1, higher quenching values are observed than for P3. The quenching of the iron(II) containing metallopolymer of P4 increased to 33%, the cobalt(II) containing metallopolymer showed a 30% quenching, whereas the nickel(II) containing metallopolymer showed the smallest quenching effect with 27%. Since the interactions between polymer and sensor are unlikely responsible for quenching (sensor and polymer, respectively have a concentration of 4.5 × 10−5 M), the metal ions must lead to the emission quenching. The fluorescence intensity increased slightly during sonication of the metal–salt free sample revealing a poor interaction of the sensor and the polymer itself. Polymers P3 and P4 only differ in their molar masses underlining the assumption that the length of the chain has major influence on the metallopolymer formation and stability. For all metallopolymers of P4, a similar value for the quenching was observed, with the smallest value found for the nickel(II)-containing metallopolymer. Compared to the nickel(II) containing metallopolymer of P3, this tendency is confirmed. Here, we also found the smallest value for quenching.

The control experiments of P5 revealed a similar trend to the previous findings. In that case, the quenching was found to be high despite the ITC investigations lead us to assume a more stable metallopolymer and a lower tendency for complexation with 1. Since P5 is not only a different polymer type, but also consists of a much lower mass compared to P3 and P4, both factors cannot be eliminated to influence the metallopolymer stability. Regarding the manganese(II)- and zinc(II)-containing metallopolymers of P5, the fluorescence intensity increased after stirring for 60 min by 4% or 3%, respectively. As mentioned before, the ITC revealed a higher tendency of the manganese(II)-ions for complex formation with terpyridine than with the sensor 1. Furthermore, a similar value for Ka was found for the sensor and the terpyridine in case ZnCl2 was utilized in ITC experiments. However, for both metal ions it is possible that the complex equilibrium is disrupted when the sensor is added and adapts during the control experiment.

Mechanical Activation of the Metallopolymers via Sonication

Sonication was performed while cooling all solutions and keeping them under nitrogen atmosphere. The sonication was performed with 0.8 s pulsation time and the fluorescence intensity was monitored during 60 min by sample taking every 15 min. The concentration of the polymer solution is influencing the activation by increasing solution viscosity leading to lower dissociation rates of polymer chains due to lower strain rates [30]. To eliminate such an influence, we diluted the metallopolymers to a concentration of 4.5 × 10−5 M.

Figure 4 shows the fluorescence intensity of 1 in the cobalt(II)-metallopolymer solution before (dark brown) and after 60 min of sonication (light brown). The intensity at 420 nm is normalized for all investigated metallopolymers to an equimolar sample of 1 without any polymer and metal salts. The relative quenching is determined to the intensity before sonication (results are summarized in Table 9). Remarkably, for P3 the quenching is always found to be higher for sonicated experiment setups than for the only-stirring control counterpart after 60 min. For the manganese(II) and zinc(II) containing metallopolymers the change in emission intensity was again found to be relatively low. For these metallopolymers, a higher quenching value was expected, since the interactions between the metal ions and the terpyridine-ligands were found be low in ITC experiments, the dissociation was expected to be induced easily by applying ultrasound. However, for the mentioned metallopolymers the quenching of the sensor emission is in the same range compared to metal-free sonication control experiments, so that the quenching of the sensor could also go back to interactions between the polymer and 1 by sonication induces processes or side reactions. For the iron(II) and nickel(II) based metallopolymers of P3 the change in emission intensity is significant (15% and 11%, respectively). The quenching was slightly higher compared to the control experiment, where for P3-Fe solution a decreasing of the sensor emission by 11% was observed. For the control experiment with P3-Ni, the sensor emission decreased only by 6% after 60 min, showing that the mechanical addressability compared to the control experiments is better for the nickel(II)-containing metallopolymer.

Fig. 4

Fluorescence spectrum of the sensor in the P3-Co metallopolymer solution. The maximum is reached at 420 nm (λex 350 nm). Concentration of 1: 4.5 × 10−5 M in 1:1 (v:v) tetrahydrofuran/methanol

In the context of our ITC studies, 2,2′:6′,2″-terpyridine revealed a higher binding constant compared to the sensor 1 mimicking the terpyridine-unit at the end of the polymer chain. The binding constant to the sensor 1 is lower compared to the unsubstituted terpyridine, indicating that it shows in general a lower tendency for complex formation with 1 and the nickel(II)-ions prefer interacting with the polymer end-group. This finding reveals that the metal ions from the metallopolymer were in fact activated by sonication and not only by equilibrium processes, since control experiments showed no comparable results regarding the sensor emission quenching: neither sonication of the sensor with the metal ion-free polymer nor the stirring of the metallopolymer with the sensor led to comparable results.

The emission of the sensor was remarkably quenched during sonication for the P3-Co by 26%. Before the sonication was started, the value for the emission intensity is already much lower for all the investigated cobalt-containing metallopolymers, which revealed a poor incorporation of the cobalt(II)-ions into the polymer structure. After 60 min sonication of P3-Co, the emission was remarkably decreased revealing that the cobalt-ions must have been activated in a high amount. Otherwise, the high value for the quenching cannot be explained, since the P3-Co control experiment only showed a quenching of 14%. In fact, it seems that the cobalt based metallopolymer of P3 is the best to be addressed by mechanical forces.

The polymer P4 consists of a higher molar mass compared to P3, and should, thus, show a better addressability for mechanical activation via sonication, as it was already reported in literature several times [23, 31, 32]. To obtain better comparability, both polymers consist of the same repeating unit and the same end-groups.

The P4 based metallopolymers were synthesized according to the metallopolymer formation of P3 and also flushed with nitrogen before sonication of the cooled solutions. The results are summarized in Table 9 and show the relative quenching values after 60 min of sonication. The change in emission intensity is different for all of the metallopolymers. Before sonication, the cobalt-containing metallopolymer setup already shows the smallest emission intensity before normalization (see supporting information Fig. S23), revealing again a low stability of the metallpolymer as it was found for P3-Co. The highest emission intensity is measured for the iron(II)-containing metallopolymer setup, followed by the nickel(II)-containing metallopolymer. This is in good agreement to the Ka values from the ITC results. Remarkably, after sonication the highest change in emission intensity is measured for the nickel(II)-containining metallopolymer, in which the emission is quenched by 55%, to about half of its original value. The control experiment showed a lower quenching value of 27% and the mechanical activation was achieved in a higher yield compared to the nickel(II)-containing metallopolymer of P3. The effect of better addressability by using a higer molar mass of the polymer was not observed for P4-Fe. The quenching was found to be in the same range compared to the sonication-free control experiment. Additionally, the results of the P4-Co sample imply again that the interactions between the cobalt-ions and the terpyridine on the polymer are weaker compared to the other metal salts. The emission intensity is already lower compared to the P3 counterpart before the start of the sonication experiment (see supporting information, Fig. S23), revealing a low incorporation of the cobalt-ions into the modified polymer end groups and high involvement of the cobalt-ions in complexes with the sensor. Considering the relative quenching values of 13%, not only the P4-Co control experiment was more successful, but also the setup with P3-Co, since the value is almost twice as high compared to the sensor quenching obtained with P4-Co. However, the assumption of the molar mass leading to an easier addressability of the metal complexes in the polymer structure can only be verified for the nickel(II)-containing metallopolymer of P4. For the other metallopolymers P4-Co and P4-Fe, this assumption was not successfully proven.

The third metallopolymer species, which was subject to sonication experiments, was the P5 based metallopolymer. This polymer consists of quite short polymer chains, compared to the other two investigated polymers. Consequently, less sonication-induced cleavage of the metal ions from the metallopolymer can be expected [33]. After flushing with nitrogen, the sensor was added and I0 samples were taken before sonication. The intensity of the emission was highest for the manganese-containing metallopolymer, followed by zinc-, nickel-, iron- and finally cobalt-containing metallopolymers (see supporting information). As previously found, the cobalt-containing metallopolymer seems to be the least stable one indicated by the already low emission intensity of the sensor before sonication. To test the mechanical addressability of the metal complexes in the polymer, the metallopolymer solutions were subject to pulsed sonication as already described for the other metallopolymers of P3 and P4. The emission intensity of the sensor after 60 min of sonication in the metallopolymer solution was compared to the emission intensity of the I0 samples and revealed some interesting facts.

In general, the relative quenching of the sensor emission during sonication was higher than expected and is comparable to the quenching observed in the previous sonication experiments (see Tables 8 and 9). Since P5 consists of a much lower molar mass compared to P3 and P4, it was expected to be poorly addressable via mechanical forces as it has been reported in literature for short polymers [33]. The sample with the metallopolymer P5-Ni showed the highest change in sensor emission intensity, which was not found in a stirring-only control experiment revealing a higher amount of activated nickel-ions during sonication.

Table 8 Results of the relative quenching values of the control experiments for all investigated metallopolymers
Table 9 Results of the sonication experiments of all investigated metallopolymers

The quenching of the cobalt-based metallopolymer was the second highest with 19%, however, the stirring-only control experiment revealed the same result (19% as well). For the sonication of the iron(II)-containing metallopolymer, a higher altering of the sensor emission by 9% was measured by sonication compared to the control experiment without ultrasound (7%). However, sonication only showed slight improvements of the iron(II)-activation in the metallopolymer. The manganese and zinc containing metallopolymer revealed a special behavior in this context. While for sonication the emission was slightly quenched by 3% and 4% respectively, in the control experiment with only stirring of the sensor-metallopolymer solution showed a slightly increased emission of the sensor. In the P3 based sonication experiments these metal ions have also shown small quenching values, probably due to poor interaction between the sensor and these metal salts or by interactions between the polymer itself and the sensor. However, such an increased emission intensity of 1 was only found in the control-experiments with P5-based metallopolymers. This could possibly show that with equilibration of the mixture, the metal-ions reveal in general a higher tendency for metallopolymer-formation compared to the complex-formation with 1. Since the sensor and the metal–salt stock solution utilized for the experiment were the same than for P3, these tendencies should have been observed for these metallopolymers as well. Since this was not the case, the polymer structure itself influences the stability of the metallopolymer.

Conclusion and Outlook

Within the current study, PMMA polymers were synthesized with modified end groups resulting in a terpyridine moiety at the end of each polymer chain. PTHF was synthesized via ring-opening polymerization and quenched with 6-(2,2′:6′,2′′-terpyridin-4′-yloxy)-hexan-1-ol. The terpyridine-functionalized polymers were investigated via SEC and UV–Vis titration with FeSO4-solution showing the successful dimerization of the polymer chains. The mechanical activation of the metallopolymers was tested with ultrasound in solution and monitored by the quenching of a sensor molecule. The sensor emission was quenched faster if ultrasound was applied to the metallopolymer solutions of P3. It was possible to show that the molar mass of the polymer has a significant influence on the activation of the metal ions via sonication, but also the polymer structure itself if of importance. Furthermore, a similar activation of the PTHF-based metallopolymers could be revealed despite a much lower molar mass compared to all PMMA-based metallopolymers. In general, the metal salts used for the metallopolymers seem to have an influence on their stability. Here, further detailed investigations on the metallopolymer stability are required (e.g., by analytical ultracentrifugation for absolute molar masses and isothermal titration calorimetry). At the current point, the ITC experiments only provide some hints for the predication of the mechanical activation. Presumably, kinetic effects (beside the thermodynamic binding parameters) seem to have an influence on the stimuli-responsive behavior.

The easy-to-prepare structure could be further utilized for mechano-responsive materials, which allow a precise debonding upon mechanical stress and re-bonding upon stress-release. For this purpose, further tests are required to investigate the mechanical addressability of the metal complexes in the polymer bulk. By further tailoring of this system, it would be interesting to develop mechano-responsive sensors that show greater fluorescence quenching upon mechanical load, or even a color-change upon metallopolymer activation.


  1. 1.

    J. Li, T. Shiraki, B. Hu, R.A.E. Wright, B. Zhao, J.S. Moore, J. Am. Chem. Soc. 136, 15925–15928 (2014)

  2. 2.

    J.N. Brantley, K.M. Wiggins, C.W. Bielawski, Polym. Int. 62, 2–12 (2013)

  3. 3.

    J.M. Lenhardt, A.L. Black, S.L. Craig, J. Am. Chem. Soc. 131, 10818–10819 (2009)

  4. 4.

    G.R. Gossweiler, G.B. Hewage, G. Soriano, Q. Wang, G.W. Welshofer, X. Zhao, S.L. Craig, ACS Macro Lett. 3, 216–219 (2014)

  5. 5.

    D.A. Davis, A. Hamilton, J. Yang, L.D. Cremar, D. Van Gough, S.L. Potisek, M.T. Ong, P.V. Braun, T.J. Martínez, S.R. White, J.S. Moore, N.R. Sottos, Nature 459, 68–72 (2009)

  6. 6.

    A.P. Haehnel, Y. Sagara, Y.C. Simon, C. Weder, in Polymer Mechanochemistry, ed. by R. Boulatov (Springer, Cham, 2015), pp. 345–375

  7. 7.

    D.W.R. Balkenende, S. Coulibaly, S. Balog, Y.C. Simon, G.L. Fiore, C. Weder, J. Am. Chem. Soc. 136, 10493–10498 (2014)

  8. 8.

    S. Bode, L. Zedler, F.H. Schacher, B. Dietzek, M. Schmitt, J. Popp, M.D. Hager, U.S. Schubert, Adv. Mater. 25, 1634–1638 (2013)

  9. 9.

    C. Heinzmann, S. Coulibaly, A. Roulin, G.L. Fiore, C. Weder, ACS Appl. Mater. Interfaces 6, 4713–4719 (2014)

  10. 10.

    R. Groote, R.T.M. Jakobs, R.P. Sijbesma, Polym. Chem. 4, 4846–4859 (2013)

  11. 11.

    A. Piermattei, S. Karthikeyan, R.P. Sijbesma, Nat. Chem. 1, 133 (2009)

  12. 12.

    R.T.M. Jakobs, R.P. Sijbesma, Organometallics 31, 2476–2481 (2012)

  13. 13.

    P. Michael, W.H. Binder, Angew. Chem. Int. Ed. 54, 13918–13922 (2015)

  14. 14.

    A. Wild, A. Winter, F. Schlütter, U.S. Schubert, Chem. Soc. Rev. 40, 1459–1511 (2011)

  15. 15.

    R. Shunmugam, G.J. Gabriel, K.A. Aamer, G.N. Tew, Macromol. Rapid Commun. 31, 784–793 (2010)

  16. 16.

    S. Schmatloch, M.F. González, U.S. Schubert, Macromol. Rapid Commun. 23, 957–961 (2002)

  17. 17.

    S. Schmatloch, A.M.J. van den Berg, A.S. Alexeev, H. Hofmeier, U.S. Schubert, Macromolecules 36, 9943–9949 (2003)

  18. 18.

    S. Bode, M. Enke, R.K. Bose, F.H. Schacher, S.J. Garcia, S. van der Zwaag, M.D. Hager, U.S. Schubert, J. Mater. Chem. A 3, 22145–22153 (2015)

  19. 19.

    D. Maheshwaran, T. Nagendraraj, P. Manimaran, B. Ashokkumar, M. Kumar, R. Mayilmurugan, Eur. J. Inorg. Chem. 2017, 1007–1016 (2017)

  20. 20.

    P.R. Andres, R. Lunkwitz, G.R. Pabst, K. Böhn, D. Wouters, S. Schmatloch, U.S. Schubert, Eur. J. Inorg. Chem. 2003, 3769–3776 (2003)

  21. 21.

    M. Grube, M.N. Leiske, U.S. Schubert, I. Nischang, Macromolecules 51, 1905–1916 (2018)

  22. 22.

    I. Nischang, I. Perevyazko, T. Majdanski, J. Vitz, G. Festag, U.S. Schubert, Anal. Chem. 89, 1185–1193 (2017)

  23. 23.

    R.T.M. Jakobs, S. Ma, R.P. Sijbesma, ACS Macro Lett. 2, 613–616 (2013)

  24. 24.

    H. Hofmeier, U.S. Schubert, Macromol. Chem. Phys. 204, 1391–1397 (2003)

  25. 25.

    E. Belhadj, A. El-Ghayoury, E. Ripaud, L. Zorina, M. Allain, P. Batail, M. Mazari, M. Sallé, New J. Chem. 37, 1427–1436 (2013)

  26. 26.

    S. Perrier, Macromolecules 50, 7433–7447 (2017)

  27. 27.

    H. Willcock, R.K. O’Reilly, Polym. Chem. 1, 149–157 (2010)

  28. 28.

    S.Y. Wong, D. Putnam, Bioconjugate Chem. 18, 970–982 (2007)

  29. 29.

    M.A.R. Meier, B.G.G. Lohmeijer, U.S. Schubert, J. Mass Spectrom. 38, 510–516 (2003)

  30. 30.

    M.W.A. Kuijpers, P.D. Iedema, M.F. Kemmere, J.T.F. Keurentjes, Polymer 45, 6461–6467 (2004)

  31. 31.

    M.B. Gordon, S. Wang, G.A. Knappe, N.J. Wagner, T.H. Epps, C.J. Kloxin, Polym. Chem. 8, 6485–6489 (2017)

  32. 32.

    S.L. Potisek, D.A. Davis, N.R. Sottos, S.R. White, J.S. Moore, J. Am. Chem. Soc. 129, 13808–13809 (2007)

  33. 33.

    K. Wei, Z. Gao, H. Liu, X. Wu, F. Wang, H. Xu, ACS Macro Lett. 6, 1146–1150 (2017)

Download references


The authors would like to thank the Deutsche Forschungsgemeinschaft (DFG, SCHU 1229/26-1). Furthermore, S.Z. is grateful to the Carl-Zeiss foundation for funding. I.N. acknowledges support from the Thüringer Ministerium für Wirtschaft, Wissenschaft und Digitale Gesellschaft (TMWWDG, ProExzellenz II, NanoPolar) for funding the Solution Characterization Group (SCG) at the Jena Center for Soft Matter (JCSM), Friedrich Schiller University Jena.

Author information

Correspondence to Ulrich S. Schubert.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (DOCX 6489 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Hannewald, N., Enke, M., Nischang, I. et al. Mechanical Activation of Terpyridine Metal Complexes in Polymers. J Inorg Organomet Polym 30, 230–242 (2020).

Download citation


  • Metallopolymer
  • Mechanochemistry
  • Stimuli responsive polymer
  • Terpyridine–metal complexes