Kavisha A.
Sarma
a,
Isis A.
Middleton
a,
Man Him
Chak
a,
Jake P.
Violi
a,
Mohan
Bhadbhade
b,
William A.
Donald
a and
Jonathon E.
Beves
*a
aSchool of Chemistry, UNSW Sydney, Sydney, NSW 2052, Australia. E-mail: j.beves@unsw.edu.au
bMark Wainwright Analytical Centre, UNSW Sydney, Kensington, NSW, Australia
First published on 10th March 2025
A ruthenium(II) complex with a photoswitchable arylazopyrazole ligand is reported. Under irradiation a combination of photoisomerisation and photoejection of the ligand occurs.
Our study uses the photoswitchable pyridyl-azoarylpyrazole ligand 1,8 pyridyl-pyrazole ligand 2,40 and 2,2′-bipyridine (bpy) to form the ruthenium(II) complexes [Ru(bpy)2(E-1)](PF6)2 and [Ru(bpy)2(2)](PF6)238,40 (Fig. 1). Photoswitchable ligand 1 was prepared using a different method than that reported,8 using commercially available aniline, acetyl acetone and 2-hydrazinopyridine to give 1 in 74% yield (S1.2†). Spectroscopic data (S1.2.2†) and a single-crystal X-ray structure (CCDC 2417092, S6.1†) confirmed the structure. The non-photoswitchable ligand 2 was prepared using an adaptation of literature procedures,48 from commercially available acetylacetone and 2-hydrazinopyridine to give 2 in a 40% yield (S1.3†).
The photoswitching properties of 1 were investigated using NMR and UV-vis absorption spectroscopies (S2†). Photostationary states (PSS) of 1 were generated by irradiating a sample of E-1 in acetonitrile-d3 for 15 minutes with 365 nm, 425 nm, 470 nm, or 530 nm light. Irradiation with 530 nm light gave the highest distribution of the E isomer (89% E-1), and irradiation with 365 nm light gave the highest distribution of the metastable Z isomer (98% Z-1) (Table S2†). These values are similar to literature values for ligand 1 in DMSO.8 The thermal half-life of the metastable isomer, Z-1, is 10 days at room temperature in acetonitrile (Table S5†).
Complexes [Ru(bpy)2(E-1)](PF6)2 and [Ru(bpy)2(2)](PF6)2 were synthesised according to previously established methods for related complexes,49 using Ru(bpy)2Cl250 and ligand 1 or 2 in ethylene glycol in a microwave reactor. Complex [Ru(bpy)2(E-1)](PF6)2 was isolated in 23% yield and characterised using NMR spectroscopy, electrospray ionisation mass spectrometry (ESI-MS) and UV-vis spectroscopy (S1.5†). A single-crystal X-ray structure (CCDC 2417124, S6.2†) confirmed that ligand 1 was in an E isomer configuration. Complex [Ru(bpy)2(2)](PF6)2 was isolated in 93% yield and similarly characterised (S1.6†). A single-crystal X-ray structure of the complex (CCDC 2417094, S6.3†) showed that the ligand 2 is equally disordered over all three ligand positions, similar to some other [Ru(bpy)2(L)]2+ complexes.51,52 Complex [Ru(bpy)2(MeCN)2](PF6)2 was synthesised using a modification of literature procedures53,54 in 46% yield, and was similarly characterised (S1.6†).
The photoswitching and photoejection behaviours of [Ru(bpy)2(E-1)](PF6)2 and [Ru(bpy)2(2)](PF6)2 in acetonitrile were studied by UV-vis absorption spectroscopy (Fig. 2). Each complex was irradiated with 365 nm, 445 nm, or 530 nm light. Irradiation with 365 nm light produced the greatest change in the UV-vis spectrum for each complex (Fig. S27 and S28†).
A sample of [Ru(bpy)2(2)](PF6)2 (10 μM, acetonitrile, 298 K) was irradiated with 365 nm light and the absorbance was monitored over 75 minutes (Fig. 2a). The MLCT band (λmax = 449 nm) is blue-shifted by 23 nm to 426 nm, characteristic of [Ru(bpy)2(MeCN)2](PF6)2. The absorption spectrum from 300–500 nm after irradiation is the same as that of [Ru(bpy)2(MeCN)2](PF6)2. This data is consistent with the formation of [Ru(bpy)2(MeCN)2](PF6)2 and free ligand 2 which has no absorption over 300 nm. ESI-MS (Fig. S42†) data showed that signals characteristic of [Ru(bpy)2(2)]2+ disappeared after irradiation, and peaks characteristic of [Ru(bpy)2(MeCN)2]2+ and [2 + H]+ were formed. A solution of [Ru(bpy)2(2)](PF6)2 in acetonitrile-d3 (Fig. S35†) was irradiated with 365 nm light for 30 min, and the 1H NMR spectrum was also consistent with the quantitative formation of [Ru(bpy)2(MeCN)2](PF6) and 2. Our studies show that 2 is an effective photoejectable ligand as the complex quantitatively undergoes photoejection, in line with previous reports.38
The photoswitchable complex, [Ru(bpy)2(E-1)](PF6)2 (13 μM, acetonitrile, 298 K), was irradiated at 365 nm and the UV-vis absorbance from 250–550 nm was monitored over 75 minutes (Fig. 2b). After irradiation, the MLCT band does not match that of [Ru(bpy)2(MeCN)2](PF6)2. The spectrum of [Ru(bpy)2(E-1)](PF6)2 after irradiation is also not any combination of the spectra of E-1, Z-1 and [Ru(bpy)2(MeCN)2](PF6)2, suggesting the formation of a new species, presumably [Ru(bpy)2(Z-1)](PF6)2. We used NMR spectroscopy to analyse the new species that was formed. A sample of [Ru(bpy)2(E-1)](PF6)2 in acetonitrile-d3 was irradiated with 365 nm light for 2.5 hours. A series of new 1H NMR peaks were formed. To determine the composition of the mixture we used the signal intensity of the methyl protons associated with E-1 and Z-1 (HB2, see Fig. 1 for labelling, S1.2.2† for data), the methyl protons of [Ru(bpy)2(E-1)](PF6)2 (HB2, S1.5.2†), the protons corresponding to [Ru(bpy)2(MeCN)2](PF6)2 (HD6, S1.7.2†), and the peak at 2.85 ppm which was assigned to the methyl protons of [Ru(bpy)2(Z-1)](PF6)2 (HB2). These peaks were chosen as they are well separated, and their signal intensities were normalised for the number of protons in each case. Using these signals [Ru(bpy)2(MeCN)2](PF6)2 and free ligand 1 make up just 5% of the total mixture, indicating that photoejection occurs for [Ru(bpy)2(E-1)](PF6)2, as discussed above for [Ru(bpy)2(2)](PF6)2. However, the photoejection of ligand 1 does not occur quantitively as was observed for ligand 2, which is surprising as the binding motifs of the pyridyl-pyrazole are the same for both 1 and 2. Other 1H NMR peaks with a relative abundance of 16% of this mixture are also formed (Fig. 3, highlighted in yellow), which we assign to [Ru(bpy)2(Z-1)](PF6)2. The NMR sample was then heated at 75 °C for 5 hours (Fig. 3). The peaks characteristic of [Ru(bpy)2(MeCN)2](PF6)2 and ligand 1 do not change in signal intensity and remain as 5% of the mixture. However, the peaks assigned to [Ru(bpy)2(Z-1)](PF6)2 decrease in intensity to become less than 2% of the mixture and the peaks of [Ru(bpy)2(E-1)](PF6)2 increase in intensity to form 93% of the mixture. This change is consistent with the thermal isomerisation from the metastable isomer [Ru(bpy)2(Z-1)](PF6)2 to the stable [Ru(bpy)2(E-1)](PF6)2 isomer. The new species formed could also be a partially ejected complex with either the pyridyl nitrogen or pyrazolyl nitrogen substituted by solvent to form [Ru(bpy)2(1)(MeCN)](PF6)2. However, no such complex was observed via ESI-MS (S5.1†). The species formed by irradiation could not be separated from [Ru(bpy)2(E-1)](PF6)2via TLC or column chromatography, indicating that this new species has a similar polarity to [Ru(bpy)2(E-1)](PF6)2. The measured diffusion coefficient55 of the new species (Table S6†) is similar to that of [Ru(bpy)2(E-1)](PF6)2, implying that the new species and [Ru(bpy)2(E-1)](PF6)2 are also similar in size.
The evidence from the UV-vis, ESI-MS and NMR data all suggests that the new species formed is [Ru(bpy)2(Z-1)](PF6)2. The mechanism of formation could be that [Ru(bpy)2(E-1)](PF6)2 first undergoes photoejection to form [Ru(bpy)2(MeCN)2](PF6)2 and E-1, then the free ligand photoisomerises to form Z-1, and finally re-coordinates to the ruthenium centre to form [Ru(bpy)2(Z-1)](PF6)2. Heating [Ru(bpy)2(MeCN)2](PF6)2 and E-1 or Z-1 in acetonitrile-d3 at 50 degrees and monitoring by NMR showed that under these conditions the ligand did not coordinate to the ruthenium centre and the solution remains as free 1 and [Ru(bpy)2(MeCN)2](PF6)2 (Fig. S37 and S38†). This suggests that the photoisomerisation of the ligand 1 occurs while coordinated to the ruthenium centre. The relative NMR signals of [Ru(bpy)2(E-1)](PF6)2 after irradiation with 365 nm light for 2.5 h showed 80% of the [Ru(bpy)2(E-1)](PF6)2 remains as [Ru(bpy)2(E-1)](PF6)2 while 5% undergoes photoejection to form [Ru(bpy)2(MeCN)2](PF6)2, and 15% isomerises to form [Ru(bpy)2(Z-1)](PF6)2. The composition of [Ru(bpy)2(1)](PF6)2 in the mixture is 16% of the metastable [Ru(bpy)2(Z-1)](PF6)2 and 84% of the stable [Ru(bpy)2(E-1)](PF6)2 complex. This ratio is significantly lower than the distribution of the metastable isomer of 1 alone under identical irradiation conditions (98% Z-1 and 2% E-1) (Table S2†). The ratio is also lower than the PSS of 1 in a 1:
1 molar solution of 1 and [Ru(bpy)3](PF6)2 (48% Z-1 and 52% E-1) (S2.4†), indicating that the difference is not simply due to the ruthenium complex acting as an inner filter for the available light. This data is consistent with photoisomerisation being inhibited upon coordination to the ruthenium metal centre.
This example shows that three different ruthenium complexes can be accessed when irradiating with the same wavelength of light. Each complex will have different photophysical properties and could influence catalytic performance. While the selectivity of this first example is not ideal, it does demonstrate a pathway for developing photoswitchable complexes with more than two configurations.
Footnote |
† Electronic supplementary information (ESI) available. CCDC 2417092, 2417094 and 2417124. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5dt00320b |
This journal is © The Royal Society of Chemistry 2025 |