16136829, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202203861 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2022].

See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Research Article
www.small-journal.com

Nanoscale Synergetic Effects on Ag–TiO2 Hybrid Substrate

for Photoinduced Enhanced Raman Spectroscopy (PIERS)
with Ultra-Sensitivity and Reusability
Josiah Shondo, Salih Veziroglu, Tim Tjardts, Tamim Bin Sarwar, Yogendra Kumar Mishra,
Franz Faupel,* and Oral Cenk Aktas*

and biomolecules,[1] monitoring the envi-

Here, a 4N-in-1 hybrid substrate concept (nanocolumnar structures, nano­ ronmental pollution and toxicity,[2] iden-
crack network, nanoscale mixed oxide phases, and nanometallic structures) tification of narcotics and explosives,
for ultra-sensitive and reliable photo-induced-enhanced Raman spectroscopy and detection of chemical and biological
(PIERS), is proposed. The use of the 4N-in-1 hybrid substrate leads to an hazards within the forensic context[3,4]
due to its ease of use and integration,
≈50-fold enhancement over the normal surface-enhanced Raman spectros-
chemical-specific detection capability,
copy, which is recorded as the highest PIERS enhancement to date. In addi- nondestructiveness, and label-free mode
tion to an improved Raman signal, the 4N-in-1 hybrid substrate provides a of operation.[5] In comparison to ordi-
high detection sensitivity which may be attributed to the activation possibility nary Raman spectroscopy, SERS pro-
at extremely low UV irradiation dosage and prolonged relaxation time (long vides a superior Raman intensity with
measurement time). Moreover, the 4N-in-1 hybrid substrate exhibits a supe- an enhancement factor (EF) of several
orders of magnitude, thanks to both the
rior photocatalytic degradation performance of analytes, allowing its reuse
electromagnetic mechanism (EM) and
at least 18 times without any loss of PIERS activity. The use of the 4N-in-1 chemical mechanism (CM) involved at
concept can be adapted to biomedicine, forensic, and security fields easily. the substrate surface.[6] EM arises basically
from the localized surface plasmon reso-
nance (LSPR) and CM results from the
1. Introduction chemical interactions (charge transfer and charge resonance)
between the analyte molecules and the substrate surface.[7]
Surface-enhanced Raman spectroscopy (SERS) has become a EM is accepted as the primary SERS mechanism by providing
widely used analytical technique for the detection of biocides an EF of 107–1012 through plasmonic metallic nanostructures,
while CM leads to an EF of 10–103 and it acts as the secondary
J. Shondo, S. Veziroglu, T. Tjardts, T. B. Sarwar, F. Faupel, O. C. Aktas mechanism for SERS.[8,9]
Chair for Multicomponent Materials The substrate plays the most critical role in the performance
Institute of Materials Science of SERS-based analytic techniques. The shape, orientation, and
Faculty of Engineering interspacing of plasmonic nanostructures on the SERS sub-
Kiel University
Kaiserstr. 2 24143, Kiel, Germany strate greatly affect the SERS intensity and sensitivity since all
E-mail: [email protected]; [email protected] these play a role on the interaction of light with the surface.[10,11]
S. Veziroglu, F. Faupel It has been shown that periodic metallic nanostructures pro-
Kiel Nano, Surface and Interface Science KiNSIS vide more reproducible, uniform, and high-density hotspots on
Kiel University the substrate surface.[12] Although various advanced bottom-up
Christian Albrechts-Platz 4 24118, Kiel, Germany
and top-down fabrication methods and material combinations
Y. K. Mishra
(1D, 2D, and 3D nanomaterials) have been demonstrated to
Mads Clausen Institute
NanoSYD prepare SERS active substrates, it is still challenging to achieve
University of Southern Denmark substrates with uniformly distributed hotspots and to maintain
Alsion 2, Sønderborg 6400, Denmark their stability against the analytes, solvents, and aging by the
O. C. Aktas time.[13,14] In addition, time-consuming fabrication methods
Additive Manufacturing Excellence Centre – URTEMM as well as single use nature of SERS substrates cause a high
Kahramankazan, Ankara 06980, Turkey
operational cost.
The ORCID identification number(s) for the author(s) of this article
Apart from intense research going on the development of
can be found under https://doi.org/10.1002/smll.202203861.
SERS substrates with high EFs and low-cost fabrication methods
© 2022 The Authors. Small published by Wiley-VCH GmbH. This is an
open access article under the terms of the Creative Commons Attribution- to produce such substrates, Parkin et al. reported totally a dif-
NonCommercial-NoDerivs License, which permits use and distribution ferent approach which yields a significant enhancement of the
in any medium, provided the original work is properly cited, the use is Raman signal of analyte molecules on a metal–semiconductor
non-commercial and no modifications or adaptations are made. hybrid substrate through preirradiating it by ultraviolet (UV)
DOI: 10.1002/smll.202203861 light (which is termed as photo-induced enhanced Raman

Small 2022, 2203861 2203861 (1 of 8) © 2022 The Authors. Small published by Wiley-VCH GmbH
 16136829, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202203861 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.small-journal.com

spectroscopy—PIERS).[15] Basically, the preirradiation with UV dosage, but this increases the risk of photo-induced (UV driven
light generates oxygen vacancies (VO) (through catalytic reac- photochemical reactions) and/or photocatalytic (catalytic reac-
tions with adsorbed O2 and H2O), and the VO density increases tions induced by photogenerated electrons) bleaching of ana-
with the irradiation time.[15–17] Various studies showed that the lytes.[26] On the other hand, if the photocatalytic bleaching of
metal–oxide semiconductor interface can locate and stabilize analyte molecules can be controlled, this may allow the reus-
VO. Basically, VO leads to the enhancement of the Raman signal ability of PIERS substrates. But for such an approach, the
through facilitating the charge transfer between the photo- metal–semiconductor hybrid substrate should be highly photo-
irradiated oxide semiconductor film (decorated with metallic catalytically active to facilitate the generation of oxygen vacan-
nanostructures) and analyte molecules via metallic nanostruc- cies at extremely low UV irradiation dosage (otherwise, the
tures deposited on the surface (a semiconductor composed of use of high intensity irradiation will be needed, which is detri-
Vo exhibits a valence state below the conduction band and this mental for organic analytes). Additionally, at such a low irradia-
facilitates the charge transfer from VO to the metallic nano- tion dosage, the relaxation time should still be long enough to
structures and then to the analyte molecule).[16] Li et al. showed conduct multiple measurements. Such a PIERS substrate can
that on a Ag–TiO2 hybrid surface photogenerated electrons in be simply cleaned up by increasing the UV irradiation slightly.
TiO2 and the hot electrons injected (LSPR) from Ag nanostruc- For reusability of the PIERS substrate, one should also ensure
tures simultaneously promote the photocatalytic reactions and the stability in terms of PIERS signal (position, intensity, and
charge transfer to analyte molecules.[18] Therefore in metal– sensitivity) after several measurement cycles (analyte loading,
semiconductor hybrid substrates, the increase in the electron measurement, and cleaning) and ideally, stability should be
density (photogenerated) as well as the plasmonic nature of proven by employing different types of analytes for consecutive
metallic nanostructures (LSPR) enhance the Raman scattering cycles.[27]
of analyte molecules.[19] In the current study, we report on a significant enhancement
Following the pioneering work of Parkin's research group, of PIERS signal for Ag–TiO2 hybrid substrate at extremely
various substrates including sole semiconductors and their low UV irradiation dosage (with an irradiation intensity of 4.5
combination with metallic nanostructures, have been studied mW cm−2 and irradiation period of 8 min). We term this novel
to reveal their applicability as PIERS substrates.[15] The ultrafast hybrid substrate, which leads to the enhancement of PIERS
recombination time ranging from picoseconds to microsec- signal by a factor of around 50, as “4N-in-1 PIERS substrate.”
onds hinders the use of semiconductors solely as PIERS sub- In addition to the long relaxation time exceeding 240 min,
strate.[17] Semiconductors and even insulators decorated with the 4N-in-1 PIERS substrate exhibited an excellent reusability
metallic nanostructures have been shown as alternative PIERS without any loss of PIERS activity after 18 cycles.
substrates, but a large bandgap limits the PIERS performance
since incident photon energy becomes insufficient to create
adequate amount of oxygen vacancies.[20] 2. Results and Discussion
Combining metallic nanostructures with semiconductors
such as TiO2 and ZnO, has been shown to trigger the forma- 2.1. 4N-in-1 Hybrid Substrate and Characterization
tion of VO with high density and stability.[12,21] Irradiation wave-
length and duration also play an important role in the VO for- Here 4N refers to i) nanocolumnar structures, ii) nanocrack
mation.[22] The energy of the incident photons must be at least network, iii) nanoscale mixed oxide phases, and iv) nanome-
equal or even larger than the bandgap energy of the semicon- tallic structures (as depicted schematically in Figure 1a. Reac-
ductor substrate to induce VO formation. VO act as additional tive sputtering and subsequently conducted heat treatment
valence states below the conduction band edge of the semi- (heating and air quenching) led to the formation of columnar
conductor, and this facilitates the charge migration from the nanostructures and dense network of nanocracks, respectively.
semiconductor to the metallic nanostructures.[23] Here, VO is A critical temperature has been revealed to maximize the crack
proportional to the irradiation time and the intensity, but very generation and promote the coexistence of rutile and anatase
high irradiation dosages (irradiation intensity and/or time) phases (which foster charge transfer and migration due to
can lead to vacancy healing and photo-induced bleaching of slight differences in their band gaps.[28] High photocatalytic
the analyte molecule, which decrease overall Raman detection nature of mentioned TiO2 film allows the controlled deposition
performance.[24] of metallic Ag nanostructures under low-intensity UV irradia-
The performance of PIERS substrate also depends on the tion, which prevents any agglomeration and excessive growth.
relaxation time; the intensity of the Raman signals decays grad- The 4N-in-1 hybrid substrate was prepared by a two-step
ually to the original level after the removal of light irradiation process. First, TiO2 thin film with a thickness of 500 nm
due to the recombination of electrons and holes.[17] Obviously, (composed of columnar nanostructures and nanocrack net-
a longer relaxation time allows for efficient collection of Raman works as shown in Figures 1b and 1c (Figure S1, Supporting
signals and reproducible measurements. Parkin et al. achieved Information), respectively) was prepared by reactive DC mag-
an EF of 34.79 after exposing TiO2 surface decorated with Au netron sputtering under relatively high oxygen partial pressure
nanostructures to UV light for extremely a long time interval as reported earlier.[29,30] As we showed previously, subsequent
(exceeding 240 min).[15] Zhang et al. reached EF of 27.8 by pre- heat treatment just after the sputtering led to the formation of
irradiating Ag–TiO2 surface with a mercury lamp operating networks of dense nanocracks and mixed rutile and anatase
at extremely high power (300 W) for 1–3 h.[25] In both cases, phases (Figure 1d and Figure S2, Supporting Information). We
high EF was achieved by applying an extremely high irradiation have shown that our sputter deposited TiO2 layer outperforms

Small 2022, 2203861 2203861 (2 of 8) © 2022 The Authors. Small published by Wiley-VCH GmbH
 16136829, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202203861 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.small-journal.com

Figure 1. a) Schematic depicting 4N-in-1 PIERS substrate concept, b) cross-sectional SEM image of TiO2 nanocolumnar structures, c) SEM image of
nanocracks networks on TiO2 thin film;(arrows indicate a crack size of 30–35 nm), d) HR-TEM images of mixed TiO2 phases (Anatase + Rutile phases
are revealed by corresponding lattice spacings), and e) SEM images of TiO2 thin film decorated with Ag nanostructures (corresponding surface cov-
erage is ≈44.9%) via photocatalytic deposition.

well-known P25 TiO2 nanoparticles, which are highly crystal- reported a positive shift in Ti 2p which may be attributed to
line and have a high surface area of 35–65 m2 g−1 (as accepted defective Ti3+ structure.[35] In principle, VO transfer their extra
as the golden standard to evaluate the photocatalytic activity two electrons to the adjacent two Ti4+ atoms to form Ti3+.[36,37]
of materials), by a factor of 29.5%.[29] As a second step, Ag Here the increase in the amount of VO with the increase in
nanostructures were reduced (from an aqueous solution of Ag% on the surface can be explained by the growth of Ag–TiO2
AgNO3) preferentially on TiO2 thin film surface via the photo- interface.[38]
catalytic deposition process developed earlier.[31] Following the Furthermore, the O 1s spectrum originated from adsorbed
irradiation of TiO2 thin film surface with UV light operating oxygen, hydroxyl groups, and lattice oxygen are also shown
at a wavelength of 365 nm and a power of 4.5 mW cm−2 for in Figure S4, Supporting Information.[39] This clearly shows
20 min, photogenerated electrons reduced Ag+ ions to stable that increasing of Ag content on TiO2 surface promotes the
Ag0 (metallic silver) on the TiO2 thin film surface. The prepared adsorption of higher amounts of oxygen and hydroxyl groups
4N-in-1 hybrid surface morphology is visualized by scanning which certainly also plays a role in the formation and stabili-
electron microscopy (SEM) analysis as shown in Figure 1e. One zation of VO.[40] On the other hand, we did not see any oxida-
can alter the surface coverage of Ag% on the TiO2 thin film sur- tion on Ag particles since satellite peaks at 366 eV (Figure S4,
face simply by increasing the photocatalytic deposition time as Supporting Information). This also explains that the increase in
shown in Figure S3, Supporting Information. The possibility adsorbed oxygen amount (O 1s spectra) should be related to the
of using low UV irradiation (thanks to the high photocatalytic change in the state of Ti4+ (increase in the amount of VO and
activity of TiO2 thin film) seems to be effective in hindering the transformation of Ti4+ state to Ti3+).
growth within the solution and uncontrolled agglomeration of
Ag particles on TiO2 surface. We conducted X-ray photoelectron
spectroscopy (XPS) to determine the surface chemistry of the 2.2. PIERS Analysis and Mechanism
prepared 4N-in-1 hybrid substrates with different Ag% surface
coverage. For all samples, we observed that Ag 3d spectra con- To demonstrate the PIERS performance of the 4N-in-1 hybrid
sists of two peaks corresponding to Ag 3d3/2 and Ag 3d5/2 with substrate, we conducted two consecutive SERS measurements
a binding energy difference of ≈6.0 eV, which clearly proves the with and without applying a UV pre-irradiation step. For
formation of stable metallic Ag (Figure S4, Supporting Infor- PIERS, we exposed the 4N-in-1 hybrid substrate to the UV-LED
mation).[32] In Figure S4, Supporting Information, one can see light source operating at a wavelength of 365 nm for 8 min.
that Ti 2p spectra are composed of two prominent characteristic Afterward, 10 µL of ethanolic solution of Rhodamine 6G ana-
peaks known as Ti 2p1/2 and Ti 2p3/2, respectively.[31] The dis- lyte (R6G ≈ 10−7 m) was drop-casted on 4N-in-1 hybrid substrate,
tance between these two peaks is ≈5.6 eV in case of bare TiO2, and the sample was dried in the dark for 10 min. Similarly, we
which can be attributed to the Ti4+ state indicating the forma- loaded the same analyte on another 4N-in-1 hybrid substrate
tion of stable TiO2.[33,34] On the other hand, it can be clearly and allowed it to dry in the dark for also 10 min. The intensity
seen from high-resolution XPS spectra of Ti 2p (Figure S4, Sup- difference of PIERS and SERS spectra recorded at an integra-
porting Information), that increasing Ag content on the surface tion time of 5 s (using a laser source operating at a wavelength
causes a positive shift of the two-distinct Ti 2p peaks. Pei et al. of 532 nm and an intensity of 15 µW) is shown in Figure 2a;

Small 2022, 2203861 2203861 (3 of 8) © 2022 The Authors. Small published by Wiley-VCH GmbH
 16136829, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202203861 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.small-journal.com

Figure 2. a) Comparison of PIERS and SERS performance of 4N-in1 hybrid substrate for the detection of trace amount of RG6 analyte (≈10−7 molar),
b) PIERS mechanism for 4N-in-1 hybrid substrate, c) comparison of UV–vis absorbance spectra of 4N-in-1 hybrid substrate before and after UV irradia-
tion (inset shows the red-shift), and d) relaxation curve (decay of the PIERS signal) after removing UV irradiation.

there is a significant enhancement in terms of Raman signal to commercial SERS substrates (Figure S7, Supporting Infor-
intensity through UV irradiation. The Raman intensity of the mation). Detection limit of 4N-in-1 hybrid structure can be
peak located at 611 cm−1 increases from 894 to 44 098 cps, which extended to very low concentrations around 10−14 m as demon-
is around 50-fold (depending on the distribution density of Ag strated at Figure S8, Supporting Information.
nanostructures, slight differences can be seen in enhancement While in classical SERS, Raman signal enhancement is
in Raman signals recorded from different position on 4N-in-1 mainly determined by the LSPR of metallic nanostructures on
PIERS substrate as shown in Figure S5, Supporting Informa- the surface, in PIERS, an additional enhancement comes from
tion). Our systematic study showed that EF is proportional to the UV induced charge migration from the semiconductor sub-
preirradiation time (Figure S6, Supporting Information) but we strate to the metallic nanostructures as schematically depicted
achieved the highest EF (which corresponds to the highest pho- in Figure 2b. Various studies showed that the metal–oxide
toinduced charge generation) between 5 and 10 min. In gen- interface can locate and stabilize VO.[16,42] Here, the produc-
eral, the rate of a photocatalytic reaction is proportional to the tion of VO permits enhancing the Raman signal of chemical
irradiation flux (high irradiation flux generates high number of molecules through charge transfer processes between the
photo-induced electrical charges). On the other hand, above a photo-irradiated TiO2 film and analyte molecules via metallic
certain value (estimated to be ≈25 mW cm−2) the reaction rate Ag nanostructures.[16] High-resolution XPS spectra of Ti 2p
reaches a saturation;[41] similarly we observed that longer irra- (at Figure S4, Supporting Information) shows that increasing
diation period (>10 min) does not lead to further enhancement Ag content on the surface causes a positive shift of two-distinct
in Raman signal (Figure S6, Supporting Information). Ti 2p peaks, which may indicate indirectly the presence of
Similar like commercial SERS substrates, 4N-in-1 hybrid sub- VO.[35] The increase in the amount of VO with the increase in
strate also exhibits a significant enhancement in Raman signal Ag% on the surface can be explained by the growth of Ag–TiO2
without any preirradiation by UV (which can be referred as interface. Figure S9, Supporting Information, shows initial
classical SERS enhancement). However, after UV preirradiation increasing of the surface coverage (≈6.3%) has significant effect
4N-in-1 hybrid substrate (in PIERS mode) exhibits a superior on the PIERS enhancement which may be correlated with
detection capability (R6G prepared at ≈10−7 m) in comparison the formation and as well as the stabilization of VO. However,

Small 2022, 2203861 2203861 (4 of 8) © 2022 The Authors. Small published by Wiley-VCH GmbH
 16136829, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202203861 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.small-journal.com

further increase in Ag surface coverage (>44.9%) enhances compared to commercial SERS substrate due to its highly active
the reflection therefore reduces the absorption of the light by photocatalytic nature (Figure S10, Supporting Information).
TiO2[26] and this seems to hinder further PIERS enhancement Therefore, by simply exposing the analyte deposited 4N-in-1
(Figure S9, Supporting Information). hybrid substrate to UV light source (operating at 365 nm) just
To further reveal the mechanism of PIERS effects, we meas- after collecting Raman spectra, we decomposed the analyte
ured absorption spectra of 4N-in-1 hybrid substrate before and fully, and this allows the reuse of the 4N-in-1 hybrid substrate.
after the UV irradiation. We observed an apparent increase
in the overall absorption and a clear red-shift of LSPR band
from 405 to 425 nm after the UV irradiation (Figure 2c). Here 2.3. Reusability for PIERS Analysis
red-shift in the 4N-in-1 hybrid substrate can be attributed to
the increased electron density on Ag nanostructures covering To demonstrate the reusability of the proposed hybrid sub-
TiO2.[40] Therefore definitely hot electrons play also a major role strate first, after the UV preirradiation we contaminated the
for achieving high EFs in PIERS. We believe that there is a syn- surface by drop casting 10 µL of dilute (molarity ≈10−7 m) R6G
ergetic effect of the i) plasmonic nature of Ag nanostructures solution. Then we collected the Raman signal from the sur-
(hot electrons), ii) Schottky barrier at the Ag–TiO2 interface, face, and finally, we exposed the substrate to a high intensity
and iii) VO stabilization at the Ag–TiO2 interface. (10 mW cm−2) UV light to clean up the surface totally (clean-
Since PIERS leads to diverse EFs for different analytes and up time depends on the amount of the analyte adsorbed on
even for different Raman bands for the same analytes, one the surface as shown in Figure S11, Supporting Information).
needs to take this into account to conduct reliable comparison The UV preirradiation can be termed as the activation step,
between different measurements (Table S1, Supporting Infor- and other consecutive steps such as loading of the analyte, col-
mation). For instance, while Parkin et al. showed an EF of 2.76 lecting the Raman spectra and the photocatalytic decomposition
(PIERS to SERS signal ratio at a peak position of 610 cm−1) for of analyte (this can be given as a single cycle as shown sche-
rhodamine 6G, we achieved an EF of 49.58 by using the same matically in Figure 3a). We showed the self-cleaning process by
analyte prepared at identical molarity.[15] The decay of PIERS the UV irradiation and reversible Raman activity in Figure 3b.
intensity to its original state lasted longer than 240 min, as PIERS activity could be restored after the simple UV cleaning
shown in Figure 2d, which clearly indicates a very slow vacancy step, and the same substrate could be reused at least 18 times
healing in air. Such a long relaxation time is ideal to get repro- without any loss of PIERS activity, which clearly indicates an
ducible and highly sensitive measurements (4N-in-1 hybrid excellent reusability. Even after some cycles, we observed
substrate exhibited around four to five times longer relaxation that the PIERS signal exceeded its original value, which can
time in comparison to reported values to the date.[15,21,43] Here be easily explained by consecutively repeated UV irradiation
extremely high EF difference between ours and those reported (cumulative activation) for every cycle. When we compare the
so far can be attributed to high photocatalytic activity of 4N-in-1 reusability of 4N-in-1 hybrid surface with those proposed by
hybrid substrate. Parkin et al. and Zhang et al. (surfaces fail after five and 15
In addition to the PIERS effect, the proposed 4N-in-1 hybrid times use, respectively) it is clearly seen that our hybrid sub-
substrate can decompose organic molecules by applying UV- strate exhibits extremely high stability after multiple measure-
irradiation (either by using high intensity UV light or con- ments.[15,21] Therefore, our hybrid surface may act as reliable
ducting a prolonged UV-irradiation time). 4N-in-1 hybrid sub- (see comparison Table S1, Supporting Information) PIERS plat-
strate shows superior decomposition performance of analyte form for multiple detection.

Figure 3. a) Schematic depiction of initial PIERS activation and PIERS analysis cycle, and b) reusability of 4N-in-1hybrid substrate for multiple PIERS
analysis.

Small 2022, 2203861 2203861 (5 of 8) © 2022 The Authors. Small published by Wiley-VCH GmbH
 16136829, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202203861 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.small-journal.com

Figure 4. Illustration of PIERS initial activation and consecutive six PIERS analysis cycles for six different analytes with corresponding PIERS spectra
(Toluene, Allura red, Erythrosine, Indigo tine, Brown HT, and Rhodamine 6G). (A: PIERS Activation, L: PIERS Analyte Loading, M: PIERS Measurement,
and C: PIERS Cleaning)

To further demonstrate the reusability of 4N-in-1 hybrid sub- substrate was held at 5 cm above the prepared solution, as
strate for multiple PIERS analysis, the same substrate was sub- schematically illustrated in Figure 5a. Afterward, the PIERS
jected to six consecutive loading-measurement-clean up cycles signal was acquired, and similarly, the SERS signal was
(following initial activation step) and for every cycle, a new type recorded without the UV preirradiation step for comparison.
of analyte (first cycle: Toluene, second cycle: Allura red, third As shown in Figure 5b, 4N-in-1 hybrid substrate exhibited a
cycle: Erythrosine, fourth cycle: Indigotine, fifth cycle: Brown PIERS signal which is 20 times higher than that recorded
HT, and sixth cycle: Rhodamine 6G) was loaded. As presented by SERS (all characteristic peaks of vinyl chloride recorded
in Figure 4, we were able to detect all characteristics peaks for by PIERS and SERS are given in Table S3, Supporting
each analyte without loss of any analyte signal and position Information).
(as shown in detail in Table S2, Supporting Information). This In conclusion, we proposed an ultrasensitive and reliable
proves that the 4N-in-1 hybrid substrate can preserve its speci- PIERS active substrate termed as “4N-in-1 PIERS substrate”
ficity and reusability after the photocatalytic clean-up. based on Ag–TiO2 hybrid structures. Upon UV irradiation, we
achieved a PIERS enhancement up to 50 times in comparison
to normal SERS intensity. In addition to improved Raman
2.4. Vapor Phase PIERS Analysis signal, 4N-in-1 hybrid substrate provided a high detection sensi-
tivity which may be attributed to the activation possibility with
To reveal how the sensitivity was improved by our PIERS an extremely low incident photon energy and prolonged relaxa-
approach in comparison to normal SERS, we compared the tion time. Moreover, 4N-in-1 hybrid substrate showed a superior
detection capabilities of both methods to monitor the sur- photocatalytic degradation performance of analytes, allowing its
face contamination induced also by a vapor phase analyte. In reuse at least for 18 cycles without any loss of PIERS activity.
a typical experiment, 10 µL of vinyl chloride (≈10 ng mL−1) Our 4N-in-1 hybrid substrate concept can be combined with any
was diluted in a methanol solution. The UV preirradiated type of Raman spectroscopy by simply integrating a UV light

Small 2022, 2203861 2203861 (6 of 8) © 2022 The Authors. Small published by Wiley-VCH GmbH
 16136829, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202203861 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.small-journal.com

Figure 5. a) Schematic illustration of the detection of vapor phase analyte and b) a comparative analysis of vapor phase analyte by PIERS and SERS.

source, and this may trigger new applications for biomedi- X-ray Photoelectron Spectroscopy (XPS): XPS (using Omicron Nano-
cine, forensic, and security fields (a possible case is shown in Technology GmbH, Al anode, 240 W) was employed to determine the
chemical state on the surface. CasaXPS software (2.3.13PR1.0) was used
Figure S12, Supporting Information)
for quantitative analysis and detailed peak investigation. All binding
energies (BE) were referenced according to the C 1s line at 285 eV
(originates from adventitious carbon on the sample surface).
UV–Vis Measurement: For UV–vis analysis, 500 nm thick TiO2 film was
3. Experimental Section deposited on quartz pieces (10 mm × 10 mm), and these samples were
Materials and Methods: Silver nitrate (99.999%), rhodamine 6 g annealed at 650 °C for an hour. Then, prepared samples were fixed onto
(99.8%), methyl alcohol (99.99%), ethyl alcohol (99.8%), 1-methyl-4- a custom-designed optical holder for UV–vis analysis. UV–vis spectra
nitrobenzene (99%), vinyl chloride solution (2000 µg mL−1 in methanol), were collected within the 200–1000 nm wavelength range at a scanning
and methylbenzene (anhydrous 99.8%), were obtained from Sigma- rate of 2 nm s−1 using a spectrophotometer (PerkinElmer LAMBDA-900,
Aldrich, Germany. Allura red, Erythrosine, Brown HT, and Indigotin were Made in Germany). The quartz glass spectrum was used for the baseline
obtained from FastColours LLP, UK. correction.
Preparation of TiO2 Thin film: The deposition of 500 nm thick TiO2 Raman Measurement: Raman Spectroscopy analysis was conducted
thin film was carried out using direct current magnetron sputtering using a commercially available confocal Raman microscope (WITec
from a titanium (Ti) target in a custom-built vacuum deposition alpha300 RA, Ulm, Germany) equipped with an Nd: YAG 532 nm laser
chamber as previously reported.[30,44,45] Silicon wafer pieces source. Intensity calibration was performed using a silicon wafer via the
(10 mm × 10 mm) were used as substrate. Before the deposition most significant silicon mode at 520 cm−1. All SERS and PIERS spectra
of TiO2 thin film, the chamber was conditioned to reach a base were acquired using 1200 g mm−1 BLZ 300NM grating at the excitation
pressure of 10−5 Pa using a rotary pump (Agilent Technologies, source power of 15 µW.
SH-110) in combination with a molecular turbopump (Pfeiffer To acquire the Raman signal of the individual analytes, a fixed volume
Vacuum, HiPace 400). First, the Ti target (Goodfellow, 99.99% of 10 µL of analyte was drop-cast onto the Ag decorated TiO2 and allowed
of 50 mm diameter) was cleaned in a pure Ar plasma for 10 min. to dry in air for 10 min. Typically, SERS spectra were acquired without
Afterward, the deposition was carried out (for 65 min) under a gas UV preirradiation step. On the other hand, in PIERS data acquisition,
mixture of Ar and O2 (at a ratio of 250:5 maintained by using two 10 µL of the analyte was deposited on the UV preirradiated samples and
separate mass flow controllers-MKS Multi-Gas-Controller 647C) allowed to dry in air for 10 min and kept in the dark.
and at a magnetron output power of 120 W. A homogeneous TiO2
film was achieved by rotating an automated sample holder during
the deposition. Afterward, the prepared thin films were annealed
for an hour at 650 °C in an oven (Nabertherm, LE 4/11/R6) and Supporting Information
subsequently air quenched to induce a nano-cracks network (leading Supporting Information is available from the Wiley Online Library or
to high surface area), as was reported previously.[29] from the author.
Photocatalytic Deposition of Ag Nanostructures on TiO2 Thin Film: In a
typical experiment, 6.5 mL of silver nitrate (≈AgNO3 = 1 × 10−3 m) was
filled in methanol, a UV-transparent quartz cuvette as was described in
detail previously.[45] Afterward, prepared TiO2 thin film samples were Acknowledgements
dipped into the quartz cuvette and exposed to low-intensity (operating
at a wavelength of 365 nm and a power of 4.5 mW cm−2) UV light (UV J.S. and S.V. contributed equally to this work. J.S. thanks the
LED StellarNet EPP2000C-SR-50) for 20 min. Then, samples were rinsed Petroleum Technology Development Fund (PTDF) and the Deutscher
with deionized water to remove excess solution and left to dry at room Akademischer Austauschdienst (DAAD) for providing funding for his
temperature. doctoral studies. This work is financially supported by the Deutsche
Scanning Electron Microscopy (SEM): The surface morphology of Forschungsgemeinschaft (DFG, German Research Foundation)- project
prepared samples was analyzed with SEM (Supra55VP-Carl Zeiss) at an number 448935424 i.e., project FA234/35-1. The authors thank NanoBMT
accelerating voltage of 3 kV (with 3 mm working distance). Co. Ltd. for preparing of some analytes.

Small 2022, 2203861 2203861 (7 of 8) © 2022 The Authors. Small published by Wiley-VCH GmbH
 16136829, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202203861 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.small-journal.com

Conflict of Interest [17] D. Glass, E. Cortés, S. Ben-Jaber, T. Brick, W. J. Peveler,

C. S. Blackman, C. R. Howle, R. Quesada-Cabrera, I. P. Parkin,
The authors declare no conflict of interest. S. A. Maier, Adv. Sci. 2019, 6, 1901841.
[18] L. Li, L. Song, L. Zhu, Z. Yan, X. Cao, Catal. Sci. Technol. 2018, 8,
1277.
[19] T. Bora, in Noble and Precious Metals - Properties, Nanoscale Effects
Data Availability Statement and Applications (Eds: M. S. Seehra, A. D. Bristow), IntechOpen,
The data that support the findings of this study are available in the London 2018, pp. 118–129.
supplementary material of this article. [20] S. Cong, Y. Yuan, Z. Chen, J. Hou, M. Yang, Y. Su, Y. Zhang, L. Li,
Q. Li, F. Geng, Z. Zhao, Nat. Commun. 2015, 6, 7800.
[21] M. Zhang, H. Sun, X. Chen, J. Yang, L. Shi, T. Chen, Z. Bao, J. Liu,
Y. Wu, ACS Sens. 2019, 4, 1670.
Keywords [22] J. Zhao, Z. Wang, J. Lan, I. Khan, X. Ye, Nanoscale 2021, 13, 8707.
hybrid structures, photocatalysis, reusability, surface-enhanced Raman [23] R. Kavitha, S. G. Kumar, Chem. Pap. 2020, 74, 717.
spectroscopy, ultra-sensitivity [24] H. Yamashita, M. Harada, J. Misaka, M. Takeuchi, K. Ikeue,
M. Anpo, J Photochem. Photobiol. A Chem. 2002, 148, 257.
Received: June 22, 2022 [25] M. Zhang, H. Sun, X. Chen, J. Yang, T. Chen, Z. Bao, J. Liu, Y. Wu,
Revised: August 26, 2022 ACS Sens. 2019, 4, 1670.
Published online: [26] S. Veziroglu, A.-L. Obermann, M. Ullrich, M. Hussain, M. Kamp,
L. Kienle, T. Leißner, H.-G. Rubahn, O. Polonskyi, T. Strunskus,
J. Fiutowski, M. Es-Souni, J. Adam, F. Faupel, O. C. Aktas, ACS Appl.
Mater. Interfaces 2020, 12, 14983.
[1] A. J. M. Fleischmann, P. J. Hendra Mcquillan, Chem. Phys. Lett. [27] X. Jiang, X. Sun, D. Yin, X. Li, M. Yang, X. Han, L. Yang, B. Zhao,
1974, 26, 163. Phys. Chem. Chem. Phys. 2017, 19, 11212.
[2] J. Langer, D. J. de Aberasturi, J. Aizpurua, R. A. Alvarez-Puebla, [28] B. Henkel, T. Neubert, S. Zabel, C. Lamprecht, C. Selhuber-Unkel,
B. Auguié, J. J. Baumberg, G. C. Bazan, S. E. J. Bell, A. Boisen, K. Rätzke, T. Strunskus, M. Vergöhl, F. Faupel, Appl. Catal. B 2016,
A. G. Brolo, J. Choo, D. Cialla-May, V. Deckert, L. Fabris, K. Faulds, 180, 362.
F. J. García de Abajo, R. Goodacre, D. Graham, A. J. Haes, C. [29] M. Z. Ghori, S. Veziroglu, B. Henkel, A. Vahl, O. Polonskyi,
L. Haynes, C. Huck, T. Itoh, M. Käll, J. Kneipp, N. A. Kotov, T. Strunskus, F. Faupel, O. C. Aktas, Sol. Energy Mater. Sol. Cells
H. Kuang, E. C. Le Ru, H. K. Lee, J. F. Li, X. Y. Ling, et al., ACS Nano 2018, 178, 170.
2020, 14, 28. [30] A. Vahl, S. Veziroglu, B. Henkel, T. Strunskus, O. Polonskyi,
[3] C. Andreou, M. R. Hoonejani, M. R. Barmi, M. Moskovits, O. C. Aktas, F. Faupel, Materials 2019, 12, 2840.
C. D. Meinhart, ACS Nano 2013, 7, 7157. [31] S. Veziroglu, M. Z. Ghori, A. Obermann, K. Röder, O. Polonskyi,
[4] H. Dies, J. Raveendran, C. Escobedo, A. Docoslis, Sens. Actuators B T. Strunskus, F. Faupel, O. C. Aktas, Phys. Status Solidi 2019, 216,
Chem. 2018, 257, 382. 1800898.
[5] T. K. Naqvi, A. Bajpai, M. S. S. Bharati, M. M. Kulkarni, [32] A. M. Ferraria, A. P. Carapeto, A. Maria, 2012, 86, 1988.
A. M. Siddiqui, V. R. Soma, P. K. Dwivedi, J. Hazard. Mater. 2021, [33] Z. Cheng, S. Zhao, Z. Han, Y. Zhang, X. Zhao, L. Kang, CrystEng-
407, 124353. Comm 2016, 18, 8756.
[6] Y. S. Yamamoto, Y. Ozaki, T. Itoh, J. Photochem. Photobiol. C 2014, [34] S. Zhao, Z. Cheng, L. Kang, M. Li, Z. Gao, RSC Adv. 2017, 7, 50064.
21, 81. [35] D. N. Pei, L. Gong, A. Y. Zhang, X. Zhang, J. J. Chen, Y. Mu,
[7] K. A. Willets, R. P. Van Duyne, Annu. Rev. Phys. Chem. 2007, 58, H. Q. Yu, Nat. Commun. 2015, 6, 2.
267. [36] Y. Xu, S. Wu, P. Wan, J. Sun, Z. D. Hood, RSC Adv. 2017, 7, 32461.
[8] L. Guerrini, D. Graham, Chem. Soc. Rev. 2012, 41, 7085. [37] M. Batzill, E. H. Morales, U. Diebold, Chem. Phys. 2007, 339, 36.
[9] B. N. J. Persson, K. Zhao, Z. Zhang, Phys. Rev. Lett. 2006, 96, [38] Y. Wang, M. Zhang, S. Lv, X. Li, D. Wang, C. Song, ACS Omega
207401. 2020, 5, 13994.
[10] R. Pilot, R. Signorini, C. Durante, L. Orian, M. Bhamidipati, [39] D. Briggs, in Handbook of Adhesion, 2nd ed. (Ed: D. E. Packham),
L. Fabris, Biosensors 2019, 9, 57. Wiley, New York 2005, pp. 621–622.
[11] S. Gullace, V. Montes-García, V. Martín, D. Larios, V. Girelli Conso- [40] L. Yang, W. Wang, H. Jiang, Q. Zhang, H. Shan, M. Zhang, K. Zhu,
laro, F. Obelleiro, G. Calogero, S. Casalini, P. Samorì, Small 2021, 17, J. Lv, G. He, Z. Sun, Sens. Actuators B Chem. 2017, 242, 932.
2100755. [41] J.-M. Herrmann, Catal. Today 1999, 53, 115.
[12] A. Fularz, S. Almohammed, J. H. Rice, ACS Appl. Nano Mater. 2020, [42] T. Zhou, K. Ioannidou, F. J. Ulm, M. Z. Bazant, R. J. M. Pellenq,
3, 1666. Proc. Natl. Acad. Sci. U. S. A. 2019, 166, 10652.
[13] J. Chen, H. Su, X. You, J. Gao, W. M. Lau, D. Zhang, Mater. Res. [43] K. Abid, N. H. Belkhir, S. B. Jaber, R. Zribi, M. G. Donato,
Bull. 2014, 49, 560. G. Di Marco, P. G. Gucciardi, G. Neri, R. Maâlej, J. Phys. Chem. C
[14] S. Zhao, H. Wang, L. Niu, W. Xiong, Y. Chen, M. Zeng, S. Yuan, 2020, 124, 20350.
L. Fu, Small 2021, 17, 2103442. [44] J. Shondo, S. Veziroglu, D. Stefan, Y. K. Mishra, T. Strunskus,
[15] S. Ben-Jaber, W. J. Peveler, R. Quesada-Cabrera, E. Cortés, F. Faupel, O. C. Aktas, Appl. Surf. Sci. 2020, 537, 147795.
C. Sotelo-Vazquez, N. Abdul-Karim, S. A. Maier, I. P. Parkin, Nat. [45] S. Veziroglu, M. Z. Ghori, M. Kamp, L. Kienle, H. G. Rubahn,
Commun. 2016, 7, 12189. T. Strunskus, J. Fiutowski, J. Adam, F. Faupel, O. C. Aktas, Adv.
[16] G. Barbillon, Materials 2021, 14, 4423. Mater. Interfaces 2018, 5, 1800465.

Small 2022, 2203861 2203861 (8 of 8) © 2022 The Authors. Small published by Wiley-VCH GmbH