Tunable magnetic circular dichroism via
electrochemically controlled charge-transfer transition in Ru(bpy) 3 21 aqueous solution
Cite as: Appl. Phys. Lett. 118, 032408 (2021);doi: 10.1063/5.0038347 Submitted: 21 November 2020
.
Accepted: 3 January 2021.
Published Online: 22 January 2021
Hai-PingPan,1,2Ya-HueiHuang,3Hua-ShuHsu,3,a) and Minn-TsongLin1,4,5,a)
AFFILIATIONS
1Department of Physics, National Taiwan University, Taipei 10617, Taiwan
2School of Physics and Optoelectronic Engineering, Foshan University, Foshan, Guangdong 528000, China
3Department of Applied Physics, National Pingtung University, Pingtung 90003, Taiwan
4Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 10671, Taiwan
5Research Center for Applied Science, Academia Sinica, Taipei 11529, Taiwan
a)Authors to whom correspondence should be addressed:[email protected]and[email protected]
ABSTRACT
Seeking potential aqueous solution with tunable magnetic properties by applied electrode potentials is an emerging research topic for the integration of aqueous solution into devices. In this work, Ru(bpy)32þ, which is widely used for high efficiency electroluminescent and photovoltaic devices, has been demonstrated as a potential liquid electrolyte with tunable magnetic properties by applied electrode potentials evidenced by energy resolved magnetic circular dichroism (MCD) spectroscopy. The MCD signal at 450 nm transforms from paramagnetic to nonmagnetic behavior when the applied electrode potentials are >1.3 V. The transition of the MCD signal from paramagnetic to nonmagnetic behavior is ascribed to the disappearance of metal to ligand charge transfer transition during the electrochemical oxidation process. This work might provide a valuable insight into exploration of functional liquid electrolyte with tunable opto-magnetic properties.
Published under license by AIP Publishing.https://doi.org/10.1063/5.0038347
To manipulate the magnetic-related phenomena of a material with the help of electric fields is an important and rapidly developing research area in modern magnetism, in both fundamental science and emerging applications.1–3Voltage (V) control of magneto-resistance, magnetic anisotropy, and magnetization in a solid state device with a ferromagnetic layer exchange-coupled to a non-magnetic solid dielec- tric layer has been demonstrated in the last few decades.2,4,5However, in solid state devices, the dielectric breakdown due to pinholes or defects is very difficult to avoid.
On the other hand, in liquid systems, the charge accumulation induced by an applied electric field at the material-liquid interface can also modulate the magnetic properties similar to that of the material- solid interface.6–8In addition, while a ferromagnetic material is used as a magnetic electrode in an electrochemical system containing freely moving ions, cation intercalation due to applied electrode potentials could also affect the magnetic properties of the ferromagnetic mate- rial.9The operation mechanism is similar to lithium-ion batteries or other energy storage systems.9,10The small ions, such as Naþor Liþ, are responsible for cation intercalation. Because these material-liquid
electrolyte systems do not undergo as much dielectric breakdown, it attracts a lot of interest to seek suitable liquid electrolytes, which can be integrated into solid devices.11–14
In general, variations of magnetism in liquid electrolyte-gated configurations may be related to charge accumulation or cation inter- calation, as mentioned above. However, the disadvantage of charge accumulation is that it is only sensitive on the accessible surface area of ferromagnetic materials. Furthermore, cation intercalation is a rela- tively slow process and the insertion and removal of ions from the fer- romagnetic materials can result in the mechanical failure over time after a few cycle operations due to the degradation. Therefore, a funda- mental question raises: could we find a functional aqueous solution whose magnetism can be directly controlled during the electrochemi- cal process without integration of ferromagnetic materials?
Here, Ruthenium(II)-tris-2,20-bipyridine [Ru(bpy)32þ] could serve as a potential aqueous solution with tunable magnetic properties by applied electrode potentials, without integration of any solid mag- netic materials. Ru(bpy)32þexemplifies a model system of intramolec- ular electron transfer reactions.15–17 Ru(bpy)32þ has been studied
intensively because of possible applications of its metal to ligand charge transfer (MLCT) excited states in solar energy conversion and storage processes.18–21For example, in dye-sensitized solar cells, the efficient photoexcited electron transfer from the MLCT manifold to the conduction band of a semiconductor electrode is allowed by its low-lying excited states.21Although Ru(bpy)32þis one of the deepest studied transition-metal complexes, whose electrochemistry and pho- tochemistry have obtained a complete understanding, the studies of its magnetic properties during electrochemical and photochemical processes are rare. Furthermore, such kinds of magnetic properties of aqueous solution are very difficult to measure by the typical magnetic method, such as a superconducting quantum interference device (SQUID), or magneto-transport measurements. Here, magneto-optical spectroscopy, namely, energy resolved magnetic cir- cular dichroism (MCD), which is the absorption difference between right circular polarized light and left circular polarized light, could be used as a suitable technique to study this issue.
In this work, transition from the paramagnetic to non-magnetic state of Ru(bpy)32þ-based aqueous solution can be reversed by con- trolling the applied electrode potentials in the electrochemical oxida- tion process at room temperature. The mechanism can be associated with the disappearance of MLCT transition in electrochemical gener- ated Ru(bpy)33þradical ions evidenced by using the operando energy resolved MCD technique. The significant changes in magnetism of Ru(bpy)32þ-based aqueous solution provide strong design-on-demand properties for electrical manipulation of magneto-optical characteris- tics in liquid electrolyte devices without integration of any solid mag- netic materials.
Figure 1(a)shows the planar molecular structure of Ru(bpy)3Cl2, stereostructure of Ru(bpy)32þ, and its corresponding orbital diagram.
With respect to both theoretical predictions and experimental analy- ses, the orbital diagram of Ru(bpy)32þ has been studied in detail.22 Figures 1(b)–1(e) reveal the operando absorption spectra of the Ru(bpy)32þ complex dissolved in aqueous solution during electro- chemical oxidation at 0 V, 1.1 V, and 1.2 V (vs Ag/AgCl) applied electrode potentials. There are three main absorption bands, which peak at 243 nm, 285 nm, and 450 nm, respectively. The absorption peak at 285 nm is easily assigned to ligand-centered p!ptransitions by comparison with the absorption spectrum of the protonated bipyri- dine ligand.22The other two peaks at 243 nm and 450 nm could be assigned to ligand-to-metal charge-transfer (LMCT) and metal-to- ligand charge-transfer (MLCT) transitions, respectively.22–25 It is noted that the MLCT transition involves a metal-centered valence electron transferring from its singlet ground state to the lowest-energy absorption band and localization of the electron on one of the bipyri- dine ligands.15The charge transfer process leads to both electronic and geometric changes in the Ru(bpy)32þcomplex, which may be related to changes in magnetic properties. Here, we will show how we can control the magnetic properties of Ru(bpy)32þ-based aqueous solution through MLCT transitions by using the operando optical absorption and energy resolved MCD spectroscopy.
When the electrode potentials are applied to the electrodes of the spectroelectrochemical cell, the oxidation process [Ru(bpy)32þ -e ! Ru(bpy)33þ] could occur in Ru(bpy)32þ-based aqueous solution.26 The metal ion Ru in the Ru(bpy)32þcomplex loses an electron in the d orbital to the working electrode and generates a RuIII(bpy)33þradical ion.23,27 When the applied electrode potential is low, for instance,
1.1 V and 1.2 V, the energy level of the working electrode is higher than the d orbital in the Ru(bpy)32þcomplex.28The resulting electro- chemical oxidation rate, which is related to the transferring electrons from the d orbital to the working electrode, is low (Fig. 2). Thus, the main component of the solution is still in Ru(bpy)32þground states.
Compared with the absorption spectrum of 0 V, no significant change in the optical absorption spectra at 1.1 V and 1.2 V confirms this result [as shown inFig. 1(b)]. However, as the applied electrode potential further increases [approaching or exceeding the oxidation potential of þ1.3 V (vs Ag/AgCl)29,30], the energy level of the working electrode becomes lower than the d orbital in the Ru(bpy)32þ complex.
Therefore, the electrons transferred from the d orbital to the working electrode increase significantly and high oxidation current is generated (Fig. 2). As a result, the electrochemical oxidation rate of Ru(bpy)32þ to generate a Ru(bpy)33þradical ion increases rapidly. Consequently, the concentration of electrochemical generated Ru(bpy)33þ radical FIG. 1. (a) The planar molecular structure of the Ru(bpy)3Cl2, stereostructure of Ru(bpy)32þand its corresponding orbital diagram. (b)–(e) are optical absorption spectra of Ru(bpy)32þ-based aqueous solution collected during electrolysis at differ- ent applied electrode potentials. The aqueous solution contained 0.4 mM Ru(bpy)3Cl2and 0.1 M NaH2PO4. The absorption peak at 285 nm (䊉) could be assigned to ligand-centered p!ptransitions. The absorption peaks red-shifted (䊊) and the absorption band broadening indicates the electronic and geometric changes in the RuIII(bpy)33þradical ion. The other two remaining intense absorp- tion peaks, at 243 nm () and 450 nm (䉲), could be assigned to metal-to-ligand (or vice versa) charge-transfer (MLCT or LMCT) transitions.
ions in the electrochemical cell rises accordingly at high applied elec- trode potentials (for instance, 1.3 V, 1.4 V, and 1.5 V), which leads to a significant change on the absorption spectra compared with that of low applied electrode potentials [Fig. 1(c)–1(e)]. The phenomenon comes from the oxidation potential of þ1.3 V (vs Ag/AgCl)29,30and is consistent with our cyclic voltammogram of Ru(bpy)32þ-based aque- ous solution inFig. 2.
The peak intensity of absorption spectra at 450 nm decreases rap- idly in solution, which contains mainly Ru(bpy)33þ radical ions at 1.5 V (as shown inFig. 3). This suggests that the absorption band at 450 nm is MLCT (d!p) transition. Due to the metal ion Ru already losing an electron in the d orbital at the electrochemical oxidation pro- cess to generate RuIII(bpy)33þ, which is a d5complex,31stabilized and not easily being photoexcited, the probability of MLCT transition sig- nificantly decreases in RuIII(bpy)33þradical ions. This leads to a rapid decrease in absorption peak intensity at 450 nm. Meanwhile, the absorption peak intensity at 243 nm gradually increases as the applied electrode potential further increases (Fig. 3). The absorption peak intensity at 243 nm increases by about 20% at 1.5 V, which contains mainly RuIII(bpy)33þradical ions in solution. This result suggests that the absorption band at 243 nm could be attributed to LMCT (p ! d) transition. The loss of an electron in the d orbital of Ru will result in more unoccupied states in the RuIII(bpy)33þ radical ion and thus enhance the LMCT (p ! d) transition-related absorption.
Furthermore, the absorption peak at 285 nm (p!ptransition) is red- shifted to 300 nm and the peak intensity drops for RuIII(bpy)33þradi- cal ions compared with RuII(bpy)32þground states [as shown inFig.
1(e)]. This is ascribed to the change in ligand geometry, mainly due to the variation in the angle between the two bipyridine rings owing to the increased charge of the central Ru ion.20,32It leads to weaker inter- action between the pyridine rings, which primarily controls the energy and rate of p!ptransitions.32Thus, the absorption peaks are red- shifted, and the peak intensity decreases for RuIII(bpy)33þradical ions due to the changes of geometric and electronic structures of the bpy ligand in RuIII(bpy)33þradical ions. Both the geometric and electronic changes in the RuIII(bpy)33þ radical ion compared with those of Ru(bpy)32þground states may lead to changes in magnetic properties.
To further confirm that the magnetic properties of Ru(bpy)32þ- based aqueous solution could be manipulated by electrochemical oxidation, we measured the energy resolved MCD intensity of Ru(bpy)32þ-based aqueous solution during electrochemical oxidation at different applied electrode potentials. Figure 4 depicts the MCD intensity monitored at 450 nm over a B ¼ 60.8 T field range at differ- ent applied electrode potentials. When the applied electrode potentials were low (1.2 V), the magnetic field-dependent MCD (MCD-H) curves behave as paramagnetic curves [Figs. 4(a)–4(c)]. Because the MCD signals at 450 nm can be related to MLCT transition (d ! p) as discussed above, the origin of paramagnetic MCD curves should come from the photoexcited electron in pstates. The slopes of the MCD-H curves are similar. The results are consistent with the absorp- tion characterization that the absorption intensities are similar when the applied electrode potentials were low [1.2 V,Fig. 1(b)].
Interestingly, as the applied electrode potentials further increase until it exceeds 1.2 V (1.3 V, 1.4 V, and 1.5 V), the slopes of the MCD- H curves decrease significantly with increasing applied electrode potentials (Fig. 5). For instance, the values of the MCD slope at 1.3 V, 1.4 V, and 1.5 V are 1.61, 0.11, and 0.17, far less than the value at 1.1 V and 1.2 V (4.95 and 4.76). The significant decrease in the slope con- firmed the transformation process from paramagnetic to non- magnetic behavior as applied electrode potentials increase. These results further demonstrated that the central metal core Ru of Ru(bpy)32þcomplexes has lost an electron in the electrochemical oxi- dation process and generated a large amount of RuIII(bpy)33þradical ions, which is a stabilized d5 complex. The probability of the photoexcited electron transfer process in the stabilized d5complex is low. Thus, the transformation from paramagnetic MCD-H to non- magnetic MCD-H behavior, which was observed during the electro- chemical oxidation process [Ru(bpy)32þ-e ! RuIII(bpy)33þ], further indicates that the photoexcited electron could be the origin of para- magnetic properties. Therefore, the lack of photoexcited electrons in the solution while the applied electrode potentials exceed 1.4 V reveals a non-magnetic behavior and does not show the paramagnetic MCD effect.
Since the MCD effect indicates the change in optical polarization, it implies that the optical polarization under an applied magnetic field can be tuned and suppressed by electric bias. Our results could offer strong design-on-demand characteristics for electrically manipulated magneto-optical devices based on Ru(bpy)32þ aqueous solution, as shown inFig. 5. It is noted that such variations in magnetic properties of Ru(bpy)32þ-based aqueous solution could be related to charge transfer transition during the electrochemical oxidation process. In FIG. 2. The cyclic voltammogram of Ru(bpy)32þ-based aqueous solution (containing
0.4 mM Ru(bpy)3Cl2and 0.1 M NaH2PO4) at a platinum gauze working electrode (scan rate 100 mV/s).
FIG. 3. The absorption intensity vs electrode potential at 450 nm and 243 nm of Ru(bpy)32þ-based aqueous solution during electrolysis is shown.
comparison with material-liquid electrolyte systems whose magnetism could be manipulated by cation intercalation, there are some advan- tages in our liquid electrolyte-only system whose magnetism is con- trolled by the electrochemical oxidation process. The electrochemical process in the liquid electrolyte does not cause material degradation and mechanical failure over time. On the other hand, the magnetic sig- nal change ratio is generally limited by the accessible surface area in material-liquid electrolyte systems via cation intercalation. Here, a sig- nificant magnetic signal change ratio (100%) due to transformation from paramagnetic to non- magnetic behavior could be observed in our electrolyte-only system. In addition, the reaction time of charge transfer in the electrochemical process is much faster than that of cat- ion intercalation in material-liquid electrolyte systems. Such a
functional liquid electrolyte, in conjunction with electrode potential- tunable MCD signals, can be devised to design magneto-optical and magneto-optoelectronic devices.
In summary, the Ru(bpy)32þ-based aqueous solution has been demonstrated as a potential aqueous solution with tunable magnetic properties by applied electrode potentials evidenced by energy resolved MCD spectra. Furthermore, the accompanying MCD signal at 450 nm transforms from paramagnetic to non-magnetic MCD-H behaviors as applied electrode potentials further increase. The transformation behaviors could be reversed by tuning applied electrode potentials dur- ing the electrochemical oxidation process. These findings pave the way toward further exploration of the functional liquid electrolyte with tunable opto-magnetic properties and lead to the development of mag- netic field-manipulation of photovoltaic and electroluminescent devices.
See thesupplementary materialfor the experimental description and setup (Fig. S1) for optical absorption and energy resolved MCD spectra measurements.
AUTHORS’ CONTRIBUTIONS
H.-P.P. performed the experimental measurements and analysis.
Y.-H.H. helped with the magnetic circular dichroism (MCD) measure- ments. H.-S.H. designed operando-MCD experiments. H.-P.P., H.- S.H., and M.-T.L. wrote the manuscript. M.-T.L. conceived the idea and supervised the whole project.
This work was supported by the Minster of Science and Technology (MOST), Taiwan, and Taiwan Consortium of Emergent Crystalline Materials (TCECM) (Project Nos. 107–2112- M-002–024-MY3 and 108–2923-M-002–002-MY2). Financial FIG. 4. The MCD intensities vs H recorded at 450 nm of Ru(bpy)32þ-based aqueous solution during electrolysis obtained at (a) 0 V, (b) 1.1 V, (c) 1.2 V, (d) 1.3 V, (e) 1.4 V, and (f) 1.5 V.
FIG. 5. The MCD slope vs applied electrode potential of Ru(bpy)32þ-based aqueous solution during electrolysis is shown. The transformation from paramagnetic to non-magnetic behavior can be observed as the applied electrode potentials exceed 1.4 V. It indicates that the Ru(bpy)32þaqueous solution could be used for electrically controlled polarization devices under an applied magnetic field. The suppression of MLCT transition (d!p) at 450 nm should be responsible for the transformation.
support from the Center of Atomic Initiative for New Materials (AI-Mat), National Taiwan University, from the Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education in Taiwan (No. 108L9008) is acknowledged. The authors would like to thank Professor Yang-Fang Chen for the valuable discussion at the Department of Physics, National Taiwan University. The author (H.-P. P.) would like to acknowledge the funding support from the Basic and Applied Basic Research Fund of Guangdong Province of China through Grant No. 2019A1515110711, National Natural Science Foundation of China (Grant No. 12004073), Research Fund of Guangdong-Hong Kong-Macao Joint Laboratory for Intelligent Micro-Nano Optoelectronic Technology (No. 2020B1212030010).
DATA AVAILABILITY
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
1F. Matsukura, Y. Tokura, and H. Ohno,Nat. Nanotechnol.10, 209 (2015).
2O. O. Brovko, P. Ruiz-Dıaz, T. R. Dasa, and V. S. Stepanyuk,J. Phys.: Condens.
Matter26, 093001 (2014).
3H.-S. Hsu, Y.-C. Chang, J.-Y. Huang, Y.-H. Huang, Y.-F. Liao, J.-S. Lee, S.-J.
Sun, and Y.-D. Yao,Carbon140, 10 (2018).
4C. Song, B. Cui, F. Li, X. Zhou, and F. Pan,Prog. Mater. Sci.87, 33 (2017).
5Y. Shiota, T. Nozaki, F. Bonell, S. Murakami, T. Shinjo, and Y. Suzuki,Nat.
Mater.11, 39 (2012).
6E. Dislaki, S. Robbennolt, M. Campoy-Quiles, J. Nogues, E. Pellicer, and J.
Sort,Adv. Sci.5, 1800499 (2018).
7H. L. Yakel,J. Phys. Chem. Solids41, 1097 (1980).
8T. Traußnig, S. Topolovec, K. Nadeem, D. Vinga Szabo, H. Krenn, and R.
W€urschum,Phys. Status Solidi RRL5, 150 (2011).
9S. Dasgupta, B. Das, M. Knapp, R. A. Brand, H. Ehrenberg, R. Kruk, and H.
Hahn,Adv. Mater.26, 4639 (2014).
10Y. Tang, Y. Zhang, W. Li, B. Ma, and X. Chen,Chem. Soc. Rev.44, 5926 (2015).
11J. Walter, H. Wang, B. Luo, C. D. Frisbie, and C. Leighton,ACS Nano10, 7799 (2016).
12C. Leighton,Nat. Mater.18, 13 (2019).
13S. Zhao, Z. Zhou, C. Li, B. Peng, Z. Hu, and M. Liu,ACS Nano12, 7167 (2018).
14S. Robbennolt, E. Menendez, A. Quintana, A. Gomez, S. Auffret, V. Baltz, E.
Pellicer, and J. Sort,Sci. Rep.9, 10804 (2019).
15W. Gawelda, M. Johnson, F. M. F. De Groot, R. Abela, C. Bressler, and M.
Chergui,J. Am. Chem. Soc.128, 5001 (2006).
16A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, and A. V. Zelewsky, Coordin. Chem. Rev.84, 85 (1988).
17N. H. Damrauer, G. Cerullo, A. Yeh, T. R. Boussie, C. V. Shank, and J. K.
McCusker,Science275, 54 (1997).
18D. W. Thompson, A. Ito, and T. J. Meyer, Pure Appl. Chem.85, 1257 (2013).
19E. M. Kober and T. J. Meyer,Inorg. Chem.21, 3967 (1982).
20D. A. Hoff, R. Silva, and L. G. Rego,J. Phys. Chem. C115, 15617 (2011).
21B. Wenger, M. Gr€atzel, and J.-E. Moser,J. Am. Chem. Soc. 127, 12150 (2005).
22F. E. Lytle and D. M. Hercules,J. Am. Chem. Soc.91, 253 (1969).
23K. Kalyanasundaram,Coordin. Chem. Rev.46, 159 (1982).
24K. W. Lee, J. D. Slinker, A. A. Gorodetsky, S. Flores-Torres, H. D. Abruna, P. L.
Houston, and G. G. Malliaras,Phys. Chem. Chem. Phys.5, 2706 (2003).
25C. Creutz and N. Sutin,J. Am. Chem. Soc.98, 6384 (1976).
26B. Noble and R. D. Peacock,Inorg. Chem.35, 1616 (1996).
27K. W. Hipps and G. A. Crosby,J. Am. Chem. Soc.97, 7042 (1975).
28A. J. Bard and L. R. Faulkner, Electrochemical Methods: Fundamentals and Applications, 2nd ed. (Wiley, New York, 2001).
29W.-Y. Lee,Microchim. Acta127, 19 (1997).
30K. A. F€ahnrich, M. Pravda, and G. G. Guilbault,Talanta54, 531 (2001).
31E. M. Kober and T. J. Meyer,Inorg. Chem.22, 1614 (1983).
32A. Hauser and E. Krausz,Chem. Phys. Lett.138, 355 (1987).
