Ravikiran
Nagarjuna
a,
Anindita
Thakur
b,
Aniket
Balapure
a,
Mohammad S. M.
Saifullah
c,
Jayati
Ray Dutta
*b and
Ramakrishnan
Ganesan
*a
aDepartment of Chemistry, Birla Institute of Technology and Science (BITS), Pilani, Hyderabad Campus, Jawahar Nagar, Kapra Mandal, Medchal District, Hyderabad, Telangana – 500 078, India. E-mail: ram.ganesan@hyderabad.bits-pilani.ac.in
bDepartment of Biological Sciences, Birla Institute of Technology and Science (BITS), Pilani, Hyderabad Campus, Jawahar Nagar, Kapra Mandal, Medchal District, Hyderabad, Telangana – 500 078, India. E-mail: jayati@hyderabad.bits-pilani.ac.in
cLaboratory for Micro and Nanotechnology, Paul Scherrer Institut, Forschungsstrasse 111, 5232 Villigen PSI, Switzerland
First published on 5th December 2023
Although the olefin-based polymerizable sol–gel (PSG) approach has brought about significant advancements in the domain of direct and scalable nanoimprinting of oxides, the high feature size shrinkage (>75–80%) due to low metal content remains a challenge. To address this, our study introduces tailor-made bifunctional monomers containing acetoacetyl and epoxy/oxetane moieties, enabling the formulation of imprintable single-source molecular precursor-based resins with tunable metal content. The introduction of epoxy/oxetane functionality serves a multi-purpose, enabling ring-opening polymerization in a chemically amplified manner while offering additional advantages such as oxygen insensitivity, good adhesion and higher metal content by alleviating the requirement of reactive diluents. The proof-of-concept imprinting studies using resins of candidate metals (Ti, Zr and Nb) have shown a reduced feature size shrinkage of the oxides to as low as ∼50%. The functionality of patterned metal oxide nanostructures is exemplified by TiO2/AgBr nanocomposites—fabricated via two different approaches—displaying high antibacterial efficacy against Escherichia coli under dark-light dual-mode conditions. Our study opens up new avenues in the realms of direct nanoimprinting of functional inorganic materials and their biological applications.
While polymeric materials have predominantly been used in these techniques, there has been a recent surge in the direct fabrication of functional inorganic nanomaterials.19–25 In particular, direct nanoimprinting of metal oxides has emerged as an attractive avenue owing to their wide-ranging applications.13,26–29 Notably, a unique characteristic of metal oxide nanoimprinting is the shrinkage of feature size in the final patterned structures, primarily influenced by the total metal content in the resin or ink.30,31 While this feature size reduction can be advantageously employed to achieve nanoscale features using molds containing sub-micron patterns, the tunability of the metal content becomes a crucial criterion in determining the overall ceramic content, which subsequently governs the resulting pattern density. Recent developments in utilizing nanoparticle suspensions in direct nanoimprinting of oxides have shown significant promise in alleviating the limitations associated with the conventional sol–gel approach, such as low yield and poor mechanical strength. Although this method has been augmented successfully for fabricating three-dimensional structures of functional oxides32 with a significant minimization of the pattern size shrinkage, the reliance of the resin on a solvent-permeable poly(dimethylsiloxane) (PDMS) mold may limit its potential in high throughput fabrication. In contrast, the polymerizable sol–gel approach (PSG), which utilizes a polymerizable liquid precursor, has been shown to be promising in high throughput techniques like step-and-flash imprint lithography.12,30,33 Though the initial works in this approach employed moisture-sensitive resins obtained by reacting metal alkoxides with polymerizable carboxylate derivatives like methacrylic acid or vinyl acetic acid, later works overcame this moisture-sensitiveness and enhanced the shelf-life of the methacrylate precursors by employing acetylacetone-based chelating monomers.31,34,35 Nevertheless, the methacrylate-based monomers require the purging of inert gas to keep off the dissolved oxygen, which otherwise scavenges the free radicals and thereby impacts the degree of polymerization. This constraint was overcome by employing thiol–ene click chemistry, which is known to be oxygen insensitive.36 Despite all these advancements, the lower metal content in the imprintable resin remains a challenge, leading to significant pattern size shrinkage to the tune of 70–80%.
The current work focuses on judicious control over the metal content in the imprintable resin while at the same time exploring a non-alkene-based monomer. The goal is to tackle the excessive feature size shrinkage while maintaining the essential requirements for direct nanoimprinting. Two bifunctional monomers containing an acetoacetyl group at one end and a photogenerated acid-triggered ring-openable epoxy or oxetane moiety at the other end were designed and synthesized.37 These monomers, upon chelating with the metal ions, enabled the formation of molecular precursor-based chemically amplified resins (CARs), using which the pattern size reduction has been brought down significantly when compared to the methacrylate-based PSG route. To the best of our knowledge, this is the first report on single-source metal-containing molecular precursor-based CARs for the direct nanoimprinting of metal oxides. By employing CARs comprising complexes of respective metals, here we demonstrate proof-of-concept studies on the fabrication of representative oxides like TiO2, ZrO2 and Nb2O5. It can be noted that previous reports on oxide nanoimprinting via the PSG approach utilized commercially procured monomers. Whereas the current work introduces custom-designed and synthesized CAR-based monomers that fulfill the essential requirements for imprintable resin. Additionally, CARs are known to exhibit oxygen insensitivity, providing the advantage of alleviating any requirement to eliminate oxygen during the imprinting process. Moreover, the inherent high adhesion property of the cyclic ethers facilitates spin-coating of the formulations, yielding high-quality films without the need for any reactive diluents, thereby offering better control over the metal content.
Shifting gears, materials and technologies that mitigate the spread of microbial pathogenesis are the need of the hour, particularly in the backdrop of antimicrobial resistance leading to the formation of multidrug-resistant strains.38–41 Several photocatalytic materials based on oxides have been developed for bactericidal and environmental remediation purposes to mitigate antimicrobial resistance.42,43 A biomimetic array of TiO2 patterns with shark-skin architecture was found to have enhanced antibacterial activity under UV light irradiation.44 Though several materials are being developed for this purpose, the majority of them are active either under dark or light conditions.45–47 Recently, a few literature have reported photocatalytically-active materials that exhibit antibacterial activity under dark-light dual mode.48–52 Nevertheless, such materials fall under the category of powders and hence, the development of dual mode-active coatable compositions could be highly beneficial and attractive for rendering numerous frequently touched surfaces with antibacterial properties, especially in high-risk areas like hospitals and public places.53,54 Therefore, as a potential application of the oxide nanoimprinting route employed here, we chose the popular photocatalyst TiO2 to further create heterojunction with AgBr.55,56 The obtained nanostructures have been studied for their efficacy towards dark-light dual-mode disinfection characteristics.
Scheme 1 Top panel: Comparison of the metal content in various titanium-containing resins employed in PSG-based nanoimprinting approaches. The molecular mass of the resins was estimated from the overall dry weight of the metal-containing monomer and crosslinker (if present), as per the molar ratio reported in the literature, from which the wt% of the metal is determined and presented alongside the formulations. For easy visualization, in most of the metal-containing monomers, one arm having reactive functionality has been depicted, while components of the other three arms have been shown by the abbreviations as mentioned in the original reference of the monomer. The abbreviations ‘OEt’ and ‘OBu’ represent ‘ethoxy’ and ‘butoxy’ moieties, respectively. The resins represented in (a), (b) and (c) are taken from the respective ref. 30, 31 and 36. Bottom panel: Schematic representation of the overall process involving ring opening polymerization chemistry for the direct nanoimprinting of metal oxides. |
In this work, we designed two low molecular weight monomers that contain an epoxy or an oxetane group at one end that could undergo cationic ring-opening polymerization and an acetoacetate group on the other end to facilitate chelation with the metal of interest. Among the two monomers, epoxy-functionalized acetoacetate (EAA) was synthesized through a trans-esterification reaction between tert-butyl acetoacetate (t-BAA) and glycidol, as reported in the literature.57 Though there is no literature report on the synthesis of oxetane-functionalized acetoacetate (OAA), the transacetoacetylation reaction between t-BAA and 3-methyl-3-oxetane methanol was found to be successful in yielding OAA. The synthesized monomers were structurally characterized using nuclear magnetic resonance (NMR) spectroscopy. The 1H NMR spectrum of EAA in Fig. 1(a) shows the characteristic peaks at δ values of 2.28 (s, 3 Ha), 2.67 (dd, 1 Hf), 2.86 (dd, 1 Hg), 3.23 (m, 1 He), 3.53 (s, 2 Hb), 4.0 (dd, 1 Hc) and 4.42 (dd, 1 Hd). The peak assignments of the characteristic signals have been depicted using the chemical structure, as shown in the figure. Further, the 13C NMR spectrum shows peaks with δ values of 30.12 (C7), 44.53 (C1), 49.06 (C5), 49.57 (C2), 65.59 (C3), 166.80 (C4), 200.27 (C6) in Fig. S1 (ESI†). Both the spectra were found to be matching with the reported values and thus confirmed the successful synthesis of EAA.57Fig. 1(b) shows the 1H NMR of OAA with the distinctive peaks at δ values of 1.34 (s, 3Hd), 2.28 (s, 3 Ha), 3.54 (s, 2 Hb), 4.24 (d, 2 He), 4.39 (d, 2Hf) and 4.52 (d, 2 Hc). The 13C NMR spectrum of OAA shows the respective peaks with δ values at 21.09 (C3), 30.18 (C8), 39.02 (C2), 49.84 (C6), 69.37 (C4), 77.46 (C1), 167.15 (C5) and 200.32 (C7) (Fig. S1, ESI†). The chemical shift, splitting pattern and integral values confirm the successful synthesis of OAA. It can be noted that both the overall percentage yield of the purified product and shelf-life were found to be higher with OAA, which could be attributed to the fact that the 3-membered epoxy ring is highly strained and hence more reactive than the 4-membered oxetane ring.
To formulate the imprintable resin, the synthesized monomers, such as EAA and OAA, were treated with metal alkoxides to obtain the corresponding epoxy- and oxetane-functionalized metal complexes. Fig. S2 (ESI†) shows the plausible chemical reactions between EAA/OAA and a metal alkoxide by chosing titanium ethoxide as a model substance. The ethoxy group is reported to exhibit two 1H NMR signals at 1.25 and 4.35 ppm corresponding to (CH3) and (CH2) protons, respectively.58 To probe the metal–monomer chelation, titanium ethoxide was reacted with the synthesized monomers in a 1:4 ratio and the 1H NMR spectra of the obtained products, Ti(EAA)4 and Ti(OAA)4, were recorded (Fig. 1). After chelation, three characteristic peaks of protons in free ethanol were observed at δ values 1.22 (m, CH3), 3.71 (m, CH2) and 1.98 (s, OH), which confirmed the successful displacement of ethoxy groups by the acetoacetate moieties in a stoichiometric manner.58 Besides, the spectrum of Ti(EAA)4 was largely found to be similar to that of EAA, except for a small peak δ = 5.05 corresponding to the proton in the enolate form of the complex (COCHCO), which is suggestive of the chelation through keto–enol tautomerism.34 A similar pattern in the 1H NMR was observed with Ti(OAA)4 as well. It is worth noting that the characteristic peaks of epoxy and oxetane groups remained intact during the reaction and thus were found to be stable.
To further validate the chelation of the acetoacetate group of the monomers with the metal and also explore the ring-opening polymerization, Fourier transform infrared (FT-IR) spectroscopic studies of Ti(EAA)4 and Ti(OAA)4 were performed. It was observed from Fig. 2(a) that EAA showed prominent characteristics peaks at 1746 cm−1 and 1712 cm−1 corresponding to the ester and keto carbonyl groups, respectively, besides a characteristic epoxy signature peak at 911 cm−1.59,60 While the FT-IR spectrum of Ti(EAA)4 resin largely resembled that of as-synthesized EAA, two new peaks were also observed at 1624 cm−1 and 1532 cm−1, which could be attributed to the chelation of the carbonyl group in the enol form of EAA with the metal center. The two peaks observed represent the asymmetric and symmetric stretching vibrations of the carbonyl group in the chelated keto–enol form, which binds to the metal center in a bidentate manner.31 After UV exposure and a subsequent post-exposure bake (PEB) step to induce crosslinking, the spectrum of the resin remained almost similar to that of the as-prepared one, except for the epoxy peak. The signature epoxy functional group peak at 911 cm−1 was found to completely disappear after PEB, indicating the participation of the epoxy functional group in the ring-opening polymerization reaction. The intensity of peaks corresponding to the chelation was found to remain unchanged even after the polymerization, indicating the stability of the metal–monomer chelation. A similar trend was also observed with the Ti-containing OAA resin (Fig. 2(b)). In this case, the characteristic oxetane ring breathing peak at 979 cm−1 was found to disappear after curing, which confirmed the occurrence of cationic ring-opening polymerization in this system as well.61
Fig. 2 Probing of metal chelation and ring opening polymerization in (a) Ti(EAA)4 and (b) Ti(OAA)4 resin formulations through FT-IR analyses. |
As mentioned earlier, the study aims to minimize the organic content while fulfilling other requirements such as good film formability, fluidity to fill into the mold cavities, polymerizability for resin solidification during imprinting and adequate mechanical strength to withstand demolding pressures. Comparing the molar mass of ethoxide and EAA, the latter is 3.5 times heavier than the former. Consequently, to verify the lowest possible organic content in an imprintable resin, we formulated compositions such as Ti(OEt)3(EAA/OAA), Ti(OEt)2(EAA/OAA)2, Ti(OEt)(EAA/OAA)3 and Ti(EAA/OAA)4 by progressively increasing the stoichiometry of EAA/OAA in place of each ethoxide arm in Ti(OEt)4. The thermogravimetric analyses (TGA) of all such resin formulations—incorporated with 2 wt% of photoacid generator (PAG) and subjected to UV irradiation—were performed under air atmosphere to probe the thermal degradation profile and the residual metal oxide content (Fig. 3). It can be seen from the figure that the first step of mass loss occurred between 100 and 350 °C, corresponding to the weakly-bound monomers and/or low molecular weight oligomers. The second step of mass loss between 350 and 550 °C could be attributed to the thermal decomposition of the crosslinked polymer. Above 550 °C, a stable plateau at a certain residual mass was observed, which confirmed the formation of the metal oxide. As expected, the mass loss was increased as the content of the monomer in the resin formulation was increased. For instance, The residual mass of TiO2 from Ti(EAA)4 was 11.5% as opposed to 29.4% from Ti(OEt)3(EAA). A similar trend was observed with OAA-based formulations as well. Table 1 summarizes the comparison between the estimated residual metal oxide content in various formulations with the values obtained from the TGA measurements, which revealed a reasonable match between the two. The slightly higher experimental values observed in some compositions may be attributed to the potential evaporation of a small amount of the monomer/alkoxy moiety during UV irradiation and sample loading for TGA measurements.
Fig. 3 TGA profiles of the resin formulations having different ratios of titanium(IV) ethoxide to EAA/OAA: (a) EAA-based and (b) OAA-based resins. |
Sample code | Residual mass of TiO2 (%) | |
---|---|---|
Calculated | Experimental | |
The calculated residual mass was based on the dry mass of the metal–monomer complex. The displaced ethanol and 2 wt% of PAG were not accounted. | ||
Ti(OEt)3(EAA) | 23.5 | 29.4 |
Ti(OEt)2(EAA)2 | 17.6 | 19.4 |
Ti(OEt)(EAA)3 | 14.1 | 15.0 |
Ti(EAA)4 | 11.8 | 11.5 |
Ti(OEt)3(OAA) | 21.7 | 29.8 |
Ti(OEt)2(OAA)2 | 15.7 | 19.4 |
Ti(OEt)(OAA)3 | 11.8 | 15.6 |
Ti(OAA)4 | 10.1 | 10.8 |
As mentioned in Table 1, four resist formulations using each monomer along with 2 wt% of PAG were prepared and studied for their imprintability. The morphology of the imprints before and after calcination was analyzed using field-emission scanning electron microscopy (FE-SEM). The PDMS mold used for the imprinting had the line and space pattern widths as ∼2.15 μm and ∼1.75 μm, respectively. After ring-opening polymerization, the thin films obtained with Ti(OEt)3(EAA) and Ti(OEt)3(OAA) were found to be brittle and peeled off on demolding and hence were found not suitable for imprinting (data not shown). This could be due to the lesser monomer content that could have resulted in a poorer degree of polymerization and, thereby, lacking the mechanical stability to sustain the pattern integrity. Though the imprintability was found to be substantially improved in Ti(OEt)2(EAA)2 and Ti(OEt)2(OAA)2, significant cracks were observed at the macroscopic scale (Fig. S5, ESI†), due to which a substantial flaking off of the films occurred during calcination. When the metal-to-monomer ratio was increased to 1:3 and 1:4, the obtained imprints were found to be smoother and mechanically stable, even after calcination. Fig. 4 shows the aerial and cross-sectional morphologies of the imprints obtained using Ti(OEt)(EAA)3, Ti(EAA)4, Ti(OEt)(OAA)3 and Ti(OAA)4. Comparing Ti(OEt)(EAA)3 and Ti(EAA)4, the respective widths of the as-imprinted line and space were found to be in the range of ∼1.5–1.6 μm and 2.3–2.4 μm. This has shown that the crosslinking and drying off of the resin leads to ∼10% of shrinkage during the imprinting step itself. After calcination, ∼48% shrinkage in the pattern width was observed with Ti(OEt)(EAA)3, while that with Ti(EAA)4 was found to be ∼60%. This higher shrinkage with Ti(EAA)4 is in line with the decreasing metal content with increasing monomer content. The magnitude of pattern width shrinkage in the case of Ti(OEt)(OAA)3 and Ti(OAA)4 was found to be very closer to the respective EAA counterparts. The findings revealed that the most favorable formulation for achieving successful imprinting with minimal shrinkage was the compositions comprising a 1:3 ratio of the metal alkoxide to the monomer. Notably, the shrinkage obtained in this study was considerably lower, in contrast to the ∼75% shrinkage reported in previous studies involving (meth)acrylate and thiol–ene-based direct nanoimprinting techniques.12,31,33,35,36 Additional experiments were conducted using dimple molds with Ti-EAA based resins to produce nanopillars to investigate pattern shrinkage in a circular geometry. In this case, the employment of Ti(OEt)(EAA)3 resulted in a diameter shrinkage of 30% following calcination, whereas with Ti(EAA)4, a slightly higher shrinkage of 34% was observed. It is worth mentioning that the diameter shrinkage of nanopillars after calcination using the methacrylate-based Ti-containing resins was previously reported to be ∼50%.33 It can further be noted that the X-ray diffraction (XRD) analyses of the calcined films obtained using Ti(EAA)4 and Ti(OAA)4 confirmed the formation of the anatase crystalline phase of TiO2 (Fig. S6, ESI†).
Two additional candidates, Zr and Nb, were selected for the imprinting experiments to evaluate the versatility of the CAR-based approach utilized in this study for different metal oxides. To achieve this, Zr(n-OBu)4 and Nb(OEt)5 were reacted with appropriate amounts of EAA, resulting in the generation of Zr(n-OBu)(EAA)3, Zr(EAA)4, Nb(OEt)(EAA)4 and Nb(EAA)5 complexes. The FE-SEM images in Fig. 5 depict the line and space patterns of the as-imprinted resins before and after the calcination process. The results have demonstrated good imprintability of the CAR-based resins for these candidates, in which the overall pattern size shrinkage was slightly lesser than the methacrylate-based approach. In the case of niobium oxide, cracks were detected alongside the line patterns, consistent with previous findings.33 These cracks appear to stem from excessive cross-linking and stress within the residual layer. A notable distinction in pattern distortion and cracking emerged when comparing oxides obtained using Nb(OEt)(EAA)4 and Nb(EAA)5 resin formulations, highlighting the influence of cross-linking intensity on the outcome. Also, the homogeneity and percentage shrinkage of the nanostructures presented in this work could be ascertained and compared with earlier-reported PSG resins using techniques like variable angle spectroscopic ellipsometer (VASE) and grazing incidence small angle X-ray scattering (GI-SAXS). Fine-tuning these parameters could potentially mitigate the observed cracking and enhance the pattern quality. Table 2 summarizes the feature size shrinkage of the line and space patterns of all the resin formulations employed in this work.
Resin | Feature size of the imprint after ring opening polymerization | Metal oxide feature size after the heat-treatment of imprinted structures | Total feature size reduction with respect to mold feature size (%) | Total feature size reduction for the corresponding metal oxide as per ref. 31 | ||
---|---|---|---|---|---|---|
Width of imprint (μm) | Feature size reduction (%) | Width of the metal oxide feature (μm) | Feature size reduction (%) | |||
The dimensions represented here are obtained with a standard deviation of ±5%. | ||||||
Ti(OEt)(EAA)3 | 1.53 | 12 | 0.79 | 48 | 55 | 84 |
Ti(EAA)4 | 1.62 | 8 | 0.67 | 59 | 62 | 84 |
Ti(OEt)(OAA)3 | 1.55 | 11 | 0.79 | 49 | 55 | 84 |
Ti(OAA)4 | 1.68 | 4 | 0.65 | 61 | 63 | 84 |
Zr(n-OBu)(EAA)3 | 1.53 | 12 | 0.72 | 53 | 59 | 74 |
Zr(EAA)4 | 1.58 | 10 | 0.63 | 60 | 64 | 74 |
Nb(OEt)(EAA)4 | 1.54 | 12 | 0.62 | 60 | 65 | 77 |
Nb(EAA)4 | 1.59 | 9 | 0.52 | 67 | 70 | 77 |
To demonstrate the functionality of the imprinted nanostructures, we chose TiO2 as the model compound, known for its high photocatalytic activity, chemical resistance, non-toxicity and optimal valence and conduction band positions for generating reactive oxygen species (ROS).62–65 Nevertheless, its wide band gap (3.2 eV) requires ultraviolet radiation to produce photogenerated electrons and holes. A plethora of literature is available on enhancing the photocatalytic activity of TiO2 by creating heterojunctions with various other visible-light active semiconductors.66–68 On a different note, thin film coatings that exhibit disinfection characteristics under dark and visible-light conditions are the urgent need of the hour, particularly in the wake of antimicrobial resistance. It can be noted that silver-based materials exhibit excellent antibacterial activities under dark conditions, while silver halides are preferred choices in heterojunction photocatalysts.69–71 Keeping these factors in view, we conducted a proof-of-concept study focusing on nanocomposites of TiO2 with AgBr to demonstrate the potential application of thin film metal oxide nanostructures. This combination was chosen due to AgBr's dual functionality as a photosensitizer and a dark-active antimicrobial substance.48 Also, the combination is known to yield Type-II heterojunction, possessing favorable attributes for high photocatalytic activity.72 Our initial approach involved mixing a surfactant-based precursor of AgBr, specifically the silver nitrate–tetraoctylammonium bromide complex (Ag–TOAB), with a Ti-containing resin to achieve a homogeneous nanocomposite. However, this addition of the AgBr precursor led to resin destabilization and subsequent precipitation. When we attempted the same procedure with the methacrylate-based Ti-containing resin reported in the literature, a similar destabilization of the resin was observed.31 In light of these challenges, we employed two alternative methods to establish heterojunctions between TiO2 and AgBr: (i) a one-step method, in which Ag–TOAB complex was spin-coated over the as-imprinted Ti(OEt)(OAA)3 resin, followed by thermolysis; and (ii) a two-step method, wherein the TiO2 nanostructures have been produced in the first step, followed by spin-coating and calcination of Ag–TOAB on top of them. The obtained nanostructures were characterized using FE-SEM, elemental mapping, X-ray photoelectron spectroscopy (XPS) and XPS valence band (VB) measurements. Due to the very thin nature of the samples, XRD measurements did not show any difference with the deposition of AgBr (data not shown).
Fig. 6(i) and (ii) show the FE-SEM images of the TiO2/AgBr nanocomposites obtained via both methods. The nanocomposite obtained through the one-step method resembled the pristine TiO2, whereas, in the two-step method, a few particulates were visible, likely due to the presence of AgBr nanoparticles. Nevertheless, the elemental mapping analysis revealed the uniform dispersion of silver and bromine across the overall surface. XPS survey scan, narrow scan and VB measurements have been performed over the TiO2/AgBr nanocomposites and compared to those of pristine TiO2. The Ti 2p narrow scan analysis in the pristine oxide sample showed the characteristic signals arising from the levels 2p3/2 and 2p1/2 at 458.6 eV and 464.3 eV, respectively.73 Further, the spin–orbit coupling of 5.7 eV, along with a distinctive satellite peak at 471.9 eV, provided confirmation of titanium being present in its ‘+4’ oxidation state in the oxide form. In the one-step nanocomposite, the binding energy of the Ti 2p3/2 was slightly shifted towards higher energy, measuring 456.0 eV. In the nanocomposite obtained through the two-step method, this binding energy was found at an intermediate value of 458.8 eV. While the spin–orbit coupling value was maintained at 5.7 eV in the nanocomposites, the intensity of the satellite peak was noticeably diminished. In addition, the peak intensity of Ti was found to be relatively lesser in the AgBr deposited samples when compared to the pristine TiO2. These observations could be attributed to the effective deposition and surface coverage of AgBr over the TiO2 surface. The surface coverage was further supported by the O 1s narrow scan spectra, which showed a two-humped feature with peak maxima around 530 eV and 532.3 eV, corresponding to lattice oxygen and surface hydroxyl groups of TiO2, respectively.73 Additionally, the Ag 3d narrow scan in both the nanocomposites showed a spin–orbit coupling of 6 eV originating from two signature peaks of Ag+ at 367.5 eV and 373.5 eV, representing the 3d5/2 and 3d3/2 levels, respectively.48 Similarly, a broad peak at ∼68 eV, resulting from the overlap between Br 3d5/2 and 3d3/2 levels, confirmed the presence of bromide in the nanocomposites.48 Comparing the two methods, the one-step method exhibited nearly three times higher silver content (2.13 at%) than the two-step method (0.66 at%), indicating superior AgBr loading in the former approach. These observations confirm the successful deposition of AgBr onto the imprinted TiO2 nanostructures.
To gain insights into the impact of AgBr deposition on TiO2, XPS VB analyses were conducted on three samples and the findings are depicted in Fig. 6(iii).73 VB spectra are widely recognized as indicators of occupied electronic states in semiconducting materials. In the case of pristine TiO2, the VB spectrum exhibited an onset (VB maxima in the band diagram) at ∼0.25 eV, with the peak intensity reaching a maximum at the binding energy of 7.2 eV. In contrast, the one-step nanocomposite displayed a VB onset at −0.3 eV, with the peak intensity reaching its maximum at an earlier energy level of ∼4.9 eV. Additionally, the overall spectral profile of the one-step nanocomposite was considerably broader than that of pristine TiO2, suggesting a higher population of electronic states closer to the VB maxima due to the incorporation of AgBr. On the other hand, in the two-step method-derived TiO2/AgBr nanocomposite, the VB onset resembled that of pristine TiO2. At the same time, the maximum intensity of the peak was reached at 4.9 eV, similar to the one-step method nanocomposite. Consequently, the VB spectrum of the two-step nanocomposite occupied an intermediate position between the other two samples. These observations highlight the favorable attributes achieved through the one-step method for the photocatalytic process (vide infra) and are consistent with the results obtained from XPS narrow scan analyses.
The disinfection performance of TiO2/AgBr nanocomposites was assessed under dual-mode conditions of dark and light in comparison to AgBr-free TiO2 (Fig. 7). In the absence of light, the pristine TiO2 nanostructures showed no bactericidal activity even after 8 h of incubation. In stark contrast, the nanocomposites obtained through both methods demonstrated complete bactericidal activity above 1 h of incubation, which was maintained throughout the 8 h experimental duration. Notably, the one-step method-derived nanocomposite exhibited significantly higher antibacterial activity, likely due to its higher silver content.48,49,74 It can be noted that the pristine AgBr coating was found to exhibit complete bactericidal activity by 4 h, which was lesser than the nanocomposites. This could be attributed to its poor distribution, as AgBr is known to form discontinuous microflowers.75 Nevertheless, under light conditions, a significant activity was observed by 2 h, demonstrating the visible-light activity of AgBr. Under visible-light exposure conditions, the photodisinfection performance of the nanocomposites was found to be accelerated, while no effect was observed with pristine TiO2 nanostructures. The nanocomposites exhibited significant inhibition of bacterial growth within 30 min, while complete inhibition was observed within 60 min. In this case too, the one-step method-derived nanocomposite outperformed the one obtained through the two-step method. To gain additional insights to this observation, we performed a fluorescence-based assay, which revealed a ∼3-fold increase in the ROS generation both in one-step and two-step nanocomposites, in comparison to pristine TiO2 and AgBr. These results additionally corroborate that the higher antibacterial activity of the one-step nanocomposite arises from the higher silver loading (for dark activity) and equivalent amount of ROS generation under light conditions. Deriving insights from XPS, the enhanced disinfection performance of the nanocomposites under light could be attributed to the higher electron density near the VB edge, as observed in the VB spectra (vide supra). It is generally known that the exposure of light with an energy equal to or greater than the bandgap of a semiconductor would result in the excitation of an electron in the VB to the conduction band (CB). This process generates a hole in the VB and an electron in the CB. Based on the oxidation/reduction potentials of the VB and CB of the semiconductor, these holes and electrons can react with water, producing reactive oxygen species (ROS) such as hydroxyl and superoxide anion radicals. These radicals can disrupt bacterial membranes, leading to the leakage of intracellular components and eventually leading to the cell death.52 To investigate the photodisinfection mechanistic pathway, we conducted an experiment using the two-step method-derived nanocomposite by employing isopropanol (IPA) and 4-hydroxy-2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPOL) as the scavengers of hydroxyl and superoxide anion radicals, respectively. The results demonstrated that the nanocomposite without any radical scavenger exhibited complete photodisinfection performance above 30 min. In contrast, the samples containing IPA and TEMPOL showed decreased photodisinfection efficacy, allowing bacterial growth to persist for up to 1 h. These findings suggest that both types of radicals play a significant role in the photodisinfection process.
The functionality of the imprinted structures was demonstrated by creating TiO2/AgBr nanocomposites using one-step and two-step methods and evaluating their antibacterial efficacy under dark-light dual-mode conditions. The nanocomposites exhibited remarkable disinfection performance, surpassing the inactivity of pristine TiO2 nanostructures. Among the two methods, the one-step method-derived nanocomposite demonstrated superior antibacterial activity, attributed to its higher silver content. In the dark, the one-step method-derived nanocomposites displayed complete bactericidal activity after 1 h of incubation and the antibacterial maintenance was ascertained up to 8 h. In contrast, pristine TiO2 showed no effect throughout the entire experiment. On the other hand, under visible-light exposure, the nanocomposite exhibited accelerated photodisinfection, inhibiting bacterial growth within 30 min and achieving complete inhibition by 45 min. These findings make the nanocomposites promising as coatings for high-risk areas like hospitals and public places due to their efficient dual-mode disinfection properties. Moreover, this study serves as a foundation for exploring the biological applications of imprinted functional nanomaterials, benefiting the healthcare sector, electronic industry and environmental remediation.
Sample code | Titanium(IV) ethoxide (mmol) | EAA/OAA (mmol) | PAG (mmol) |
---|---|---|---|
Ti(OEt)3(EAA) | 1 | 1 | 0.0075 |
Ti(OEt)2(EAA)2 | 1 | 2 | 0.0150 |
Ti(OEt)(EAA)3 | 1 | 3 | 0.0225 |
Ti(EAA)4 | 1 | 4 | 0.0300 |
Ti(OEt)3(OAA) | 1 | 1 | 0.0873 |
Ti(OEt)2(OAA)2 | 1 | 2 | 0.0174 |
Ti(OEt)(OAA)3 | 1 | 3 | 0.0262 |
Ti(OAA)4 | 1 | 4 | 0.0350 |
To derive quantitative information on the ROS generation, DCFH-DA assay was used.82 For this, a stock solution of 3 mM was prepared in 1 mL of 100% ethanol, from which a secondary stock of 50 μM was prepared in 1× PBS. The substrate were taken in a 24 well plate added with 800 μL of autoclaved millipore water, to which 200 μL of 50 μM DCFH-DA was added to get the final working concentration as 10 μM. The plate was then incubated under light till 1 h, following which the fluorescence was recorded at the excitation wavelength of 488 nm and the emission wavelength of 525 nm.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3ma00666b |
This journal is © The Royal Society of Chemistry 2024 |