Demonstration of a Gel-Polymer Electrolyte-Based Electrochromic Device Outperforming Its Solution-Type Counterpart in All Merits: Architectural Benefits of CeO2 Quantum Dot and Nanorods

For years, solution-type electrochromic devices (ECDs) have intrigued researchers’ interest and eventually rendered themselves into commercialization. Regrettably, challenges such as electrolyte leakage, high flammability, and complicated edge-encapsulation processes limit their practical utilization, hence necessitating an efficient alternate. In this quest, although the concept of solid/gel-polymer electrolyte (SPE/GPE)-based ECDs settled some issues of solution-type ECDs, an array of problems like high operating voltage, sluggish response time, and poor cycling stability have paralyzed their commercial applicability. Herein, we demonstrate a choreographed-CeO2-nanofiller-doped GPE-based ECD outperforming its solution-type counterpart in all merits. The filler-incorporated polymer electrolyte assembly was meticulously weaved through the electrospinning method, and the resultant host was employed for immobilizing electrochromic viologen species. The filler engineering benefits conceived through the tuned shape of CeO2 nanorod and quantum dots, along with the excellent redox shuttling effect of Ce3+/Ce4+, synchronously yielded an outstanding class of GPE, which upon utilization in ECDs delivered impressive electrochromic properties. A combination of features possessed by a particular device (QD-NR/PVDF-HFP/IL/BzV-Fc ECD) such as exceptionally low driving voltage (0.9 V), high transmittance change (ΔT, ∼69%), fast response time (∼1.8 s), high coloration efficiency (∼339 cm2/C), and remarkable cycling stability (∼90% ΔT-retention after 25,000 cycles) showcased a striking potential in the yet-to-realize market of GPE-based ECDs. This study unveils the untapped potential of choreographed nanofillers that can promisingly drive GPE-based ECDs to the doorstep of commercialization.

Table 2 of the main text.It was realized that low-filler loading (0.5 wt%) did not contribute to ΔT of NR0.5/PVDF-HFP/IL/BzV-Fc ECD as compared to filler-less PVDF-HFP/IL/BzV-Fc ECD as seen in Table S2.However, the benefits were realized in increased filler-loading (1 wt%), which displayed an ΔT of ~60% in NR1.0/PVDF-HFP/IL/BzV-Fc ECD.Further increment in filler-loading demonstrated an ΔT of ~58.7% and ~53.1 % for NR1.5/PVDF-HFP/IL/BzV-Fc ECD and NR2.0/PVDF-HFP/IL/BzV-Fc ECD, respectively, which is an inferior performance compared to NR1.0/PVDF-HFP/IL/BzV-Fc ECD.The optimized performance for 1wt % filler loading lies in the fact that lower filler loading doesn't trigger the required enhancement in GPEs, while the higher loading of filler causes transmittance drop due to light scattering stemming from undesired agglomeration/bed formation in the GPE system.

S1.2 Characteristic redox behavior of electrochromic species utilized in ECDs
The three-electrode system was employed to realize the intrinsic characteristics of utilized viologen species [BzV:1,1′-bis(4-fluorobenzyl)-4,4′-bipyridine-1,1′-diium tetrafluoroborate].Briefly, 0.01 M viologen species was dissolved in PC solution using 0.1 M TBABF4 as a supporting electrolyte, and the mixture was placed in a cell cuvette.Further, an ITO with an active area of 1.0 × 1.0 cm 2 was clamped inside the cell to work as a working electrode, while Pt and a homemade Ag/Ag + were used as counter and reference electrodes, respectively.The cyclic voltammetry at a scan rate of 100 mV/s was performed on an as-prepared three-electrode system using a negative potential bias as viologens are cathodically coloring materials displaying blue color in the reduced state.The obtained CV plot is displayed in Fig. S3 with two redox peaks appearing in the forward cycling of BzV; the first reduction peaks between -0.5 to -0.9 V showcases the reduction of viologen dictation into radical cation (BzV 2+ ⇌ BzV +• ) while the subsequent second peak between -0.9 and -1.2 V drives unstable radical cation into neutral ion state (BzV +• ⇌ BzV 0 ).The reverse cycling takes the viologen species back into its second and first oxidation states, respectively, thus completing the redox loop, as evident from Fig. S3.More often, the second redox peak in viologens causes irreversibility and could be avoided to realize the better cycling life of electrochromic material.

S1.3 Investigation of the piezoelectric effect in PVDF-HFP-based ECD
The utilized polymer (PVDF-HFP) in our fabricated ECDs is a typical piezoelectric material.Thus, a representative device (QD-NR/PVDF-HFP/IL/BzV-Fc ECD) was investigated for the effect of piezoelectricity, which might possibly originate during any exposed applied stress or heat (speculated to generate from coloration process).First, the effect of stress was probed by simply monitoring the device's open circuit voltage (OCV) before and after the applied stress, as shown in Fig. 5d,e.The OCV of tested ECD is expected to alter upon a weight press due to the accumulation of charge observed in the piezoelectric materials.A metal standard (~200 g) was used for employing stress over the ECD's active area, as shown in Fig. 5e.As apparent in Fig. s 5d,e, the ECD registered no noticeable change in OCV (only 0.1 mV) upon applied stress, thus discouraging the possibilities of piezoelectricity rendered from the PVDF-HFP.This behavior may be attributed to the presence of a spacer (DuPont 60 μm Surlyn ® frame) used during the fabrication of the ECD cell, which significantly nullifies the effect of applied stress on the gel-polymer electrolyte.Moreover, to examine the probability of piezoelectric effect emerging from continuous coloration/bleaching, the device was subjected to 1000 CV cycles in a voltage range and scan rate of 0 ~ 1.1 V and 100 mV/s, respectively.The OCV was immediately recorded as soon as the 1000 th cycle was completed, as displayed in Fig. 5f.As compared to the pristine state (Fig. 5d), no observable change in OCV was noticed after the 1000 CV cycles of ECD, thus indicating that either the generated heat was too small to trigger any piezoelectric effect or no measurable heat generation occurred in the device during the electrochromic process.This observation was further supported by the IR thermal imaging of the ECD recorded during the bleaching and coloration process, as shown in Fig. S7a,b.A temperature increment of only ~1.1 0 C was observed in the colored state of ECD compared to the bleached state, indicating a weak heating effect.To summarize, no meaningful evidence of piezoelectricity was observed in our fabricated ECD due to an inadequate stress or heat effect.

S1.4 Discussion on systematic incorporation of X (=BzV, BzV-Fc, QD-NR/BzV, QD-NR/BzV-Fc) in the host PVDF-HFP/IL ECD
Four ECD configurations were prepared by systematically incorporating X (=BzV, BzV-Fc, QD-NR/BzV, QD-NR/BzV-Fc) to host PVDF-HFP/IL ECD, as to distinctively understand the effect brought by nanofillers.The fabricated ECDs were shuttled between their bleached and colored state by employing a potential bias of 0.0 and 1.0 V, respectively.The obtained ∆T and stability behavior (1000 cycles) of ECDs are highlighted and quantified in Fig. S6 and Table S4, respectively.It was observed that ECD with bare BzV was able to achieve ΔT of ~56% at 603 nm, though degraded abruptly (Fig. S6e) from both bleached and colored states indicating the poor reaction kinetics without Fc and nanofillers.In contrast, when the host ECD was inducted with the BzV-Fc combo, a considerable improvement was seen in cycling stability, though only a marginal color enhancement was observed in the bleaching side of the ECD, suggesting that Fc positively contributes to the backoxidation reaction of BzV +• ⇌ BzV 2+ .This observation is in good agreement with previous reports [1,2].Moreover, upon fillers' (here QD-NR) utilization with BzV, a significant enhancement was observed in the bleached state of the host device, as seen in Fig. S6c and Table S4, though a drop in colored state was also observed.This phenomenon indicated that while the CeO2 fillers provide enough redox interactive sites towards the cathodic direction (or more negative potential), their functionality still needs support from the counter electrode species (Fc here) at the anodic side.This assumption was validated by further induction of counter electrode species Fc in the filler-BzV duo, which demonstrated excellent coloration improvement at both the cathodic and anodic sides, as seen in Fig. S6d.The switching ability of both bare filler and Fc inducted ECDs, showcased a competent beginning however dropped severely in the later cycles, as evidenced by Fig. S6f,g.This issue was addressed when Fc and QD-NR were inducted together in the ECD assembly, as witnessed in Fig. S6d,h.It is inferred that Fc and CeO2 simultaneously facilitate anodic and cathodic side reactions for BzV chromophores, which in turn render enhanced optical contrast and remarkable electrochemistry.In contrast, the ECDs utilizing bare Fc or CeO2 do not reach their full bleaching and colored states, respectively, which in turn generates traces of irreversible layers on the electrode surface due to the dimerization tendency of BzV radical cation [3,4].Such accumulation/deposition on the electrode surface greatly hinders the device kinetics and mounts higher charge transfer resistance on the electrode surface, thus leading to the gradual degradation of ECDs in the long term.

S1.5 Investigation of the heat effect and role of the safe potential window in the switched ECDs
Some experiments were designed and performed to analyze the extent of heat effects originating due to electrochromic switching for our flagship QD-NR/PVDF-HFP/IL/BzV-Fc ECD.
The temperature distribution over the device's active area was recorded before and after 10,000 switching cycles using a thermal imaging camera (Fluke TiS75+), and the corresponding images are shown in Fig. S7a-c.It was observed from the thermal images that a temperature (°F in image scale) difference of ~1.1 0 C persisted between the bleached and colored states of ECD (Fig. S7a,b), whereas a difference of ~1.6 0 C was recorded between pristine and long switched ECD (Fig. S7a,c).
Considering the adequate thermal stability of gel-polymer electrolyte-based ECDs, the possibility of bubble/liquid formations due to such miniature temperature difference is rare.Further, inheriting the concern of electrolyte/PC escape, the drilled holes were sealed right after the electrolyte injection in our fabricated ECDs.Also, the blank ECD was carefully edge encapsulated through a Surlyn ® frame to avoid side leakage, as mentioned in experimental section 2.4 of the main text.In addition to the above two precautions, the presence of nanofiber assembly (PVDF-HFP) further reduces the possibility of electrolyte escape by facilitating an efficient electrolyte entrapment.Evidently, no bubble/void formation in the ECD's active area was seen after 10,000 cycles when shuttled in a voltage window of 0.9 V (0.0 to 0.9 V), as can be seen in Fig. S7d.
To examine the possibility of bubble formations, a duplicate QD-NR/PVDF-HFP/IL/BzV-Fc ECD was switched in an unsafe potential window of 4.0 V (-2.0 to 2.0 V) for 100 cycles.After some initial cycling, a considerable bubble/void was observed inside the ECD's active area, as shown in Fig. S7e.Regardless of the fact that the utilized species in the ECD were vacuumed-dried before use, the possibility of unremoved moisture/trapped air persists, for example, due to the presence of BMIMBF4 (moisture prone) in the ECD.Such presence of moisture entrapment might lead to undesired reactions (such as water splitting) when subjected to a higher voltage window (here, 4.0 V), thus evolving traces of oxygen/hydrogen, which in turn might generate bubbles/voids in the ECD.Another possibility of such bubble formation results from the breakage of electrolyte/PC in such a higher operating voltage window.In short, our experimentation suggested that miniature temperature differences in ECD assembly certainly occur after numerous cycles.However, the difference is insufficient to trigger any bubble formations or electrolyte escape.On the other hand, operating ECDs in an unsafe potential window can abruptly lead to electrolyte breakage or the creation of bubbles that deters the ECD performance.

S1.6 Discussion regarding interfacial junctions and defects in hybrid nanostructures
Though quantum dots possess interesting electronic and optical properties, their high selfaggregation tendency makes it difficult to exploit their surface area fully.[5,6] Growing QDs in-situ over the NR surfaces was presumed to mitigate the agglomeration issue of QDs, thus retrieving more surfaces for interactions.The improved electrochromic features revealed by QD-NR-based ECD can partially be attributed to this "QD's surface recovery" aspect.In addition, the facet/surface engineering of CeO2 is expected to tune the kinetic parameters responsible for the improved charge transfer and proliferated redox centers.[7,8] It is now in good agreement that distinct morphologies in CeO2 nanostructure deliver dissimilar degrees of surface electronic structures, reaction centers, electric conductivity, and adsorption sites depending on the exposed facets.[9][10][11][12] The interesting feature of 0D/1D hybrid nanostructures (in our study, QD-NR) can be attributed to discrete interfaces and individual components with their own crystal planes of varying orientation.Such properties enable the hybrid nanostructures to possess singly charged and neutral vacancies at the defective junctions (0D/1D).[7] We also observed the presence of several surface vacancies in the synthesized QD-NR filler due to microstrains, as witnessed through XPS and EPR spectra in Fig. 1.The available microstrains in such tuned nanostructures are reported to have low formation energies and facilitate the creation of an anion Frankel defect, which in turn causes numerous single and surface oxygen vacancies.The synergistic shuffling of single and surface oxygen vacancies leads to considerable structural rearrangement in the crystal lattice of CeO2, forming distinct dimeric and trimeric vacancy clusters with different orientations.[8,12] Briefly, it can be deciphered that the availability of reactive surfaces and synergistic improvement brought by defect surfaces and corrugated junctions seems to have a decisive role that allows hybrid nanostructures such as QD-NR in our study to deliver superior electrochromic performance as compared to their bare NR and QD counterparts.

Fig. S1 .
Fig. S1.The TEM image highlights the average length and width of synthesized CeO2 nanorods.

Fig. S6 .
Fig. S6.The dynamic transmittance and stability plots of host PVDF-HFP/IL ECD with (a,e) bare

Table S1 .
Quantification of redox centers for as-synthesized nanofillers

Table S2 .
Optical contrast of ECDs with different filler loadings

Table S4 .
Transmittance and stability analysis of host PVDF-HFP/IL ECD with different configurations

Table S5 .
Partial list of recent reports announcing major developments in the field of SPE/GPE based ECDs