Structure-regulated enhanced Raman scattering on a semiconductor to study temperature-influenced enantioselective identification

Surface-enhanced Raman scattering (SERS) spectroscopy is an effective technique that can reveal molecular structure and molecular interaction details. Semiconductor-based SERS platforms exhibit multifaceted tunability and unique selectivity to target molecules as well as high spectral reproducibility. However, the detection sensitivity of semiconductors is impeded by inferior SERS enhancement. Herein, a surface and interference co-enhanced Raman scattering (SICERS) platform based on corrugated TiO2 nanotube arrays (c-TiO2 NTs) was developed, and the coupling of structural regulation and photo-induced charge transfer (PICT) effectively optimized the SERS performance of the semiconductor substrate. Due to the regularly oscillating optical properties of the c-TiO2 NTs, well-defined interference patterns were generated and the local electric field was significantly increased, which greatly promoted both the electromagnetic mechanism and PICT processes. The c-TiO2 NTs were subsequently applied as a highly sensitive SICERS substrate to investigate the mechanism of temperature influence on enantioselective identification. This identification process is related to the existence of temperature-sensitive hydrogen bonds and π–π interaction. This work demonstrates a simply prepared, low-cost, and sensitive SERS substrate that enables better investigation into molecular identification.

water with a resistivity of 18.0 M•cm.

Apparatus and characterization.
Interference reflectance spectra of c-TiO 2 NTs were collected by using an USB 2000+ fiber optic spectrometer (Ocean Optics, USA) coupled with a bifurcated fiber optic cable (Ocean Optics, USA).Raman measurements were conducted using a Raman microscopy spectrometer (LabRAM HR, HORIBA Scientific, France).Morphological characterization was carried out using a field-emission scanning electron microscope (SEM, Hitachi SU8000, Japan).
Crystal structures were identified by XRD acquired using an X'Pert XRD spectrometer (Philips, USA) using a CuKα X-ray source.Zeta potentials were measured on a Zetasizer Nano ZS90 analyzer (Malvern, USA).The chiral optical properties were recorded by circular dichroism (CD) spectra using a MOS-450 CD Spectrometer (Bio-Logic, France).X-ray photoelectron spectra (XPS) were recorded on a Perkin-Elmer Physical Electronics 5600 spectrometer using AlKα radiation at 13 kV as excitation source.Fourier transform infrared (FTIR) spectroscopy was performed using a Nicolet 6700 instrument (Thermo Fisher, USA).

Calculation of EFs.
The EFs of c-TiO 2 NTs were determined according to the following equation.Modifying the molecule on the substrate surface caused a change in the effective refractive index of the medium. 4Thus, the interference peaks of the interferometric reflectance spectra were red-shifted after the O-Phos and L-Phe modifications.The characteristic peaks at ~1087 cm -1 (-CN), 1562 cm -1 (-CO), and 2971 cm -1 (-NH 2 ) in the FTIR spectra were attributed to O-Phos and L/D-Phe.

Fig. S1
Fig. S1 and S2 show the SEM images of TiO 2 NTs prepared with different anodization times.The nanotube length and the number of interference corrugations were mainly controlled by the anodization time.The nanotube length influenced the propagation path of incident light inside the nanotube arrays, which in turn affected the interference effect of the substrate.An appropriate nanotube length was conducive to the generation of interference peaks with regular interference corrugations.Fig. S3A and B display the interferometric reflectance spectra and Raman spectra of TiO 2 NTs with varying lengths.

EXPERIMENTAL SECTION Materials and reagents.
Ti sheets (0.1 mm thickness, 99.6% purity) were purchased from Baosheng Hardware (Bao ji).O-phosphorylethanolamine (C 2 H 8 NO 4 P), phenylalanine (C 9 H 11 NO 2 ), tryptophan (C 11 H 12 N 2 O 2 ), carnitine (C 7 H 15 NO 3 ), tyrosine N-(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC) were obtained from Sigma-Aldrich (St. Louis, USA).Ammonium fluoride (NH 4 F), ethylene glycol, glutamic acid (C 5 H 9 NO 4 ), isopropanol (C 3 H 8 O), and ethanol (C 2 H 5 OH) were purchased from Sinopharm Chemical Reagent Co. Ltd. and used as received without further purification.All aqueous solutions were prepared using Millipore Milli-Q 1,2   SICERSand I bulk are the intensities of the selected Raman peak of N719 on the SICERS substrate and pristine silicon wafer, N bulk and N SICERS are the average number of N719 molecules in the scattering area used for Raman and SICERS measurements, respectively.The spectrum of N719 (1 mM) on the pristine silicon wafer were used as the normal Raman reference.The number of probe molecules within the scattering area was estimated using Equation2based on the assumption that the probe molecules were uniformly distributed on the substrates.C represents the molar concentration of the N719 solution, V is the volume of the droplet, N A is Avogadro's constant, and S sub is the effective area of the substrate.