L-Tryptophan Adsorbed on Au and Ag Nanostructured Substrates: A SERS Study

Abstract

The objective of this study was to determine the most stable conformation of L-tryptophan (L-Tryp) on gold and silver nanoparticles. Additionally, this work investigated how these parameters were influenced by analyte concentration, nanoparticle size, and pH. The purpose of this study was to establish whether L-Tryp molecules interact with the nanoparticles through the carboxylate end, the amino group end, or both. This research has diverse applications in biophysics and medical diagnostics, potentially opening up new avenues in these fields. Moreover, it may enrich the disciplines of chemistry and nanotechnology by offering innovative approaches for future research. These findings represent a significant advancement in understanding the interactions between L-Tryp and nanoparticles, making a meaningful contribution to biophysics and medical diagnostics. Surface-Enhanced Raman Scattering (SERS) spectra of L-Tryp in the 100–4000 cm⁻¹ spectral range were obtained using a 785 nm laser for excitation. Gold (Au) and silver (Ag) nanoparticles (NPs) were synthesized using the citrate reduction method. The experimental procedure involved the use of electrolytes (such as NaCl) for colloid activation, which resulted in very high SERS signals. Modification of nanoparticle surface charge was achieved by adjusting the pH of Au and Ag colloidal suspensions between 2 and 11. The SERS spectra indicate that small-sized nanoparticles require high concentrations of L-Tryp to achieve high sensitivity, whereas larger nanoparticles perform effectively at lower concentrations. The pronounced enhancement of stretching vibrations in the COO⁻ group in the SERS spectra strongly suggests that the carboxylate group attaches to silver nanoparticles (AgNPs). Conversely, for gold nanoparticles (AuNPs), a new band at approximately 2136 cm⁻¹ was observed, indicating that the amino group of L-Tryp interacts with Au in its neutral form. These analyses were complemented by theoretical modeling, employing Density Functional Theory (DFT) calculations run using the Gaussian program to study molecular models in which L-Tryp interacted with AgNP and AuNP substrates in neutral, cationic, and anionic forms.

1. Introduction

L-Tryptophan (L-Tryp) is a large neutral amino acid notable for its unique 3-methyl indole side chain, which makes it an important subject of study [1,2,3,4]. This side chain, despite its high hydrophobicity, can be found on the surface and within protein molecules [5]. The aromatic π electrons in L-Tryp facilitate strong interactions with other molecules that also contain π electrons. While L-Tryp lacks ionizable side chain groups, its relatively unreactive indole -N-H bond (with a pKa of 9.5) can function as a hydrogen bond donor. L-Tryp is significant for its role as a precursor to serotonin, which is subsequently converted into melatonin. It plays a crucial role in regulating mood and sleep patterns and is involved in the synthesis of various neurotransmitters in the brain, as well as the production of niacin [6,7].

Studies on human subjects suggest that L-Tryp may have potential benefits in treating conditions such as Down syndrome and aggressive behavior [5,6,7,8,9,10]. However, challenges arise when L-Tryp levels are altered in cancer patients, leading to significantly reduced serotonin levels, which are associated with depressed mood and lower survival rates [11]. Despite these challenges, the potential applications of L-Tryp in medicine remain promising, highlighting opportunities for further research and development [12].

Recent advancements in nanoscale science have led to the development of novel nanomaterials with exceptional physical, chemical, and biological properties [13,14,15,16]. This progress has led to the development of sensitive and selective detection methods that address some limitations of conventional technologies [17]. Within this context, gold (Au) and silver (Ag) Nanoparticles (NP) have emerged as powerful tools in sensing and imaging applications due to their unique optical properties [18,19,20,21,22,23]. The distinct characteristics of gold and silver nanoparticles, such as their high extinction coefficients, sharp extinction bands, and significant field enhancements, have made them prominent in nanotechnology [18,19,20,21,22,23]. Although Ag presents many advantages over Au, including higher extinction coefficients, sharper extinction bands, and a greater ratio of scattering to extinction, it is used less frequently in sensor development, aside from applications involving Surface-Enhanced Raman Scattering (SERS) [17]. This limited use is primarily due to the lower chemical stability of AgNP compared to AuNP [24,25,26]. While considerable excitement surrounds the potential applications of gold nanoparticles in medical diagnostics and other biological functions, researchers are becoming increasingly aware of the need to investigate potential nanoparticle toxicity before advancing to in vivo applications [27].

Typically, Au and Ag colloids absorb in the visible region from 400 to 700 nm, which allows them to be used as SERS substrates, a light-scattering phenomenon [28]. SERS has become an excellent tool for determining the orientation of molecules absorbed on their metal surfaces, which holds significant potential for biophysical studies [29]. Moreover, SERS can enhance the RS intensities by as much as 10²⁰-fold, enabling the acquisition of vibrational spectra at very low sample concentrations [28]. A previous SERS study of L-Tryp concluded that the amino acid binds to the silver surface through both the carboxylate (-COO⁻) and amino (-NH₂) groups [29]. The aliphatic moiety was found to be close to the nanoparticle surface. At the same time, the nitrogen atom of the indole ring only binds when L-Tryp is the C-terminal residue of a peptide [1,29,30]. In addition, research conducted by Kim and colleagues [31] revealed strong vibrational enhancements of the symmetric stretching mode (-COO-) and the stretching mode -C-COO-, suggesting that the adsorption of the carboxylate and amino groups of L-Tryp on the silver surface was responsible for these enhancements [32]. However, our research presents a more detailed SERS investigation of L-Tryp, requiring consideration of nanoparticle size and colloidal pH. It is anticipated that the surface charges of nanoparticles influence both the observed SERS phenomenon and the interaction between the analyte and the metal surface [33]. Furthermore, calculations were optimized using Density Functional Theory (DFT) on a molecular model, employing Gaussian v16, in which L-Tryp interacts with large gold and silver surfaces in its neutral, cationic, and anionic forms.

2. Materials and Methods

2.1. Reagents

Chemicals of the highest available purity were used for nanoparticle synthesis. Tetrachloroaurate (III) trihydrate (HAuCl₄·3H₂O, 99.9985% Au; 49% Au, (Strem Chemicals) and silver nitrate (AgNO₃, 99.9995% Ag, Strem Chemicals) were obtained from STREM Chemicals (Newburyport, MA, USA). Sodium citrate dihydrate (C₆H₅Na₃O₇·2H₂O, ≥99%, FG), sodium hydroxide (NaOH, reagent grade, ≥98%, anhydrous pellets), and L-tryptophan (C₁₁H₁₂N₂O₂, 99%) were purchased from Sigma-Aldrich Chemical Co. (Milwaukee, WI, USA) and used without additional purification. L-tryptophan (pKa₁ = 2.4 and pKa₂ = 9.4) exists in its zwitterionic form, with the amino group protonated as –NH₃⁺, both in the solid state and in aqueous solutions at pH < 9.0. Hydrochloric acid (HCl, 36–38%) was supplied by VWR International (Suwanee, GA, USA), while nitrogen gas (N₂, ultra-high purity, UHP) was purchased from Linde Gas Co. (Guayanilla, PR, USA). All solutions were prepared using Milli-Q water. Glassware used for AgNP and AuNP preparation was thoroughly cleaned with aqua regia and rinsed with ultrapure water (18.2 MΩ) before use. Aluminum and gold slide substrates (1 cm × 1 cm) were initially washed with Alconox laboratory detergent (White Plains, NY, USA) and Micro-90 Cleaning Solution (Cole-Parmer, Vernon Hills, IL, USA), then rinsed with absolute ethanol, followed by distilled and deionized UHP water.

2.2. Preparation of Ag and Au Colloids

For the synthesis of AgNP, the procedure reported by Lee and Meisel was adjusted [34]. After preparation, colloids were stored at 8 °C for future use. The obtained AgNP suspensions were characterized using UV-Vis spectrophotometry to ensure that the reaction was complete for the redox synthesis, with an absorption maximum at λ ≈ 410 nm. The synthesized AgNPs had a diameter of 68 nm, a zeta potential of −43.6 mV, and a pH of 7.35. The AuNPs were synthesized using the citrate reduction method. A 1.0 mM chloroauric acid solution was prepared at a rate of 200 mL per 50 mL dilution of HAuCl₄·3H₂O (0.01%) in UHP water. A 1% sodium citrate reducing solution was also prepared. The metal ion and reducing agent solutions were mixed after they had been previously degasified with N₂, and both solutions were placed in the incubator for 15 min to produce the growth solution for aqueous Au ions required to synthesize the desired NPs. A typical synthesis consisted of dissolving 50.0 mL of HAuCl₄·3H₂O (0.01%) in a flask. Then, 300 µL to 500 µL of 1% sodium citrate was added to the flask. The solutions were put in a Barnstead Lab-Line General Incubator and left undisturbed for 12 h until the color changed from colorless to violet and the solutions had a pH of 4.10. Obtained colloids were also stored at 8 °C for future use.

2.3. Instrumentation

Electronic absorption spectra of AgNP and AuNP colloids were obtained using a Duetta Fluorescence and Absorbance Spectrometer, Horiba Scientific, NC, USA. Spectra were recorded in the range of 200 to 800 nm. Quartz cells with a 1.0 cm path length were used in the experiments, and data were collected in absorbance mode. Measurements of the mean hydrodynamic radius and zeta potential of colloidal suspensions at different pH values were conducted at 25 °C employing a Zetasizer Nano Series instrument (Malvern Instruments Ltd., Malvern, Worcestershire, UK). The SEM images for AgNP and AuNP morphology were obtained using a JSM-IT500HR InTouchScope^TM Scanning Electron Microscope, Peabody, MA, USA. These samples for scanning electron microscopy (SEM) were prepared on 400-mesh copper grids with ultrathin carbon/perforated carbon films by depositing 5 μL of the metal nanoparticle suspensions (Ted Pella, Inc., Redding, CA, USA).

RS of solid (crystalline), aqueous L-Tryp solutions, and SERS experiments were performed with 785 nm laser excitation line (Excelsior Newport-Spectra Physics, Irvine, CA, USA) in a RS Microspectrometer model RM-2000 (Renishaw, Inc., Hoffman Estates, IL, USA) equipped with a CCD camera. The acquisition parameters were a 30 s integration time and 10 accumulations for the bulk analyte, and a 30 s integration time with 1 accumulation for SERS samples of L-Tryp adsorbed on AgNP. For L-Tryp-AuNP SERS samples, 10 s integration and 3 accumulations were used. Spectra for RS and SERS were collected within the RS shift range of 100–3500 cm⁻¹.

2.4. RS Experiments

Small quantities of the solid samples were placed on a stainless-steel microscope slide, and RS spectra were acquired within the RS shift range of 100–3800 cm⁻¹. The laser power at the sample was maintained at 11 mW, with the beam set to 100% on the neutral-density filter wheel using the Wire spectral acquisition and analysis software (v. 4.0, Renishaw). Stock solutions of L-Trypt at concentrations ranging from 4.9 × 10⁻³ to 4.9 × 10⁻¹³ M were prepared using UHP water. A droplet of 10 µL L-Tryp 4.9 × 10⁻³ M solution was transferred onto the surface of the aluminum slide substrate, and the Raman signal for this sample with an optimum pH was recorded.

For the SERS measurements, 200 µL of Ag and Au colloids, 100 µL of a stock solution of L-Tryp, and 70 µL of 0.1 M NaCl for colloid activation were transferred to a microcentrifuge vial and mixed by vigorous manual agitation, resulting in a final pH of 7.35 (Ag) and 4.52 (Au). SERS spectra were acquired immediately and at 10, 20, and 30 min after mixing, and then at 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, and 24 h to ensure that the analyte was adsorbed on the CRS surface [35]. A small volume of 10 µL of this sample was transferred onto an aluminum microscope slide for Ag and gold-coated slide for Au, and spectra were acquired at laser powers of 11.2, 15.6, and 20.0 mW. To confirm the results, RS and SERS measurements were repeated several times for each sample.

Nanoparticles have a tendency to aggregate, depending on the pH of the media, the ionic strength, and the analyte with which they interact [15]. To adjust the pH of the media, solutions of NaOH and HCl, 0.10 M and 0.01 M, respectively, were used. These solutions were used to adjust the pH of the colloidal gold and silver suspensions. In one vial, the pH of approximately 1.0 mL of the colloidal suspension of gold or silver was adjusted through controlled addition of NaOH or HCl until the desired pH was obtained. Samples of 200 µL of the colloid at the desired pH were mixed with 100 µL of L-Tryp solutions at concentrations of 4.9 × 10⁻¹² M for Ag NPs and 4.9 × 10⁻⁸ M for AuNP. SERS spectra for each sample were collected every 20 min after mixing with the nanoparticles at a specified pH. An aliquot of 10 µL L-Tryp solution and AgNP and AuNP colloid was transferred onto aluminum or gold slides, and the RS signal for each mixture was recorded and acquired. Fixed laser powers of 15.6 and 20.0 mW range were used to acquire the SERS spectra for L-Tryp solutions in contact with AgNP and AuNP, respectively.

2.5. Computational Method

In this study, the neutral or zwitterion (⁺H₃NCH(R)COO⁻), the cationic species (⁺H₃NCH(R)COOH), and the anionic species (NH₂CH(R)COO⁻) of the molecular structure L-Tryp were optimized using DFT at the B3LYP (Becke, three-parameters, Lee–Yang–Parr) level of theory with 6-311G basis set, using Gaussian™ v.-16 program [36]. For the optimized structures of L-Tryp, vibrational RS frequencies were computed at the same level of theory and compared with experimental spectra. To gain an insight into the interaction of L-Tryp with the metal NP surface, a calculation with a surface of ten (10) Ag and Au atoms was performed. Initially, the energy and vibrational Raman frequencies were calculated for the surface using DFT at the B3LYP level with the LANL2DZ (Los Alamos effective core potential with double zeta function). To evaluate the interaction of different forms of L-Tryp with Ag or Au surfaces, two methods for optimization and frequency calculations were employed. First, calculations were performed with the possible molecular forms of L-Tryp (neutral, cationic, and anionic) and the Ag and Au surfaces. These calculations were performed using pseudo potentials for Ag and Au with the LANL2DZ basis set, and for L-Tryp structures with the 6-31G basis set. This combination was used to save computational costs. The second method of calculation was performed using the LANL2DZ basis set for both Ag and Au atoms, as well as the L-Tryp structures. These methods involved freezing the Ag and Au atoms.

3. Results and Discussion

3.1. Nanoparticles Characterization

AgNP and AuNP were synthesized as described in the Experimental Section. Briefly, AgNP were also reduced chemically using sodium borohydride [34]. However, the NPs obtained did not perform as well as the citrate-reduced NPs at low L-Tryp concentrations. This has been attributed to the optical properties of metallic nanostructures depending more markedly on shape than on size [37]. After synthesizing the nanoparticles, they are characterized using ultraviolet–visible (UV-Vis) spectroscopy, a simple and sensitive technique for confirming the synthesis. It was used to characterize the wavelength max and the width of the absorption band of the electronic surface plasmon. The presence of only one plasmon component (transverse) indicates that the colloids consist mostly spherical for both AgNPs and AuNPs [38].

The Scherrer equation [39,40,41,42] in X-ray diffraction and crystallography is a formula that relates the size of sub-micrometer crystallites in a solid to the broadening of a peak in a diffraction pattern. It is often referred to, incorrectly, as a formula for particle size measurement or analysis. It is named after Paul Scherrer. It is used in the determination of the size of crystals in the form of powder. The Scherrer equation can be written as follows:

τ = (Κλ)/(β cos (θ))

where,

• τ is the mean size of the ordered (crystalline) domains, which may be smaller or equal to the grain size.
• Κ is a dimensionless shape factor, with a value close to unity. The shape factor has a typical value of about 0.9, but varies with the actual shape of the crystallite.
• λ is the X-ray wavelength.
• β is the line broadening at half the maximum intensity (FWHM), after subtracting the instrumental line broadening, in radians. This quantity is also sometimes denoted as Δ(2θ).
• θ = is the Bragg angle.

The absorption maxima were located at 395 and 410 nm for Ag and 530 nm for Au. Figure 1 shows the results of the characterization of the colloidal suspensions of AgNPs and AuNPs. A Scanning Electron Microscope (SEM) was used further to confirm the nearly spherical nature of the NPs (Figure 2). The approximate average size of the Au NPs was 60 nm (spherical) and 52 and 55 nm for Ag (spheroids: mixture of spherical and ellipsoidal).

3.2. Optimization of Geometries of L-Tryp on NP Substrate Models

Figure 3a–c shows the geometries of inputs for the calculations of L-Tryp with the AgNP and AuNP surfaces. These were based on optimized structures of the neutral, cationic, and anionic forms of L-Tryp and confirmed that all optimized structures were minimum energy conformations. The optimized structure of the L-Tryp molecule and the model NP surfaces consisting of 10 Ag (1, 0, 0) and 10 Au (1, 0, 0) are shown in Figure 3a Ag/L-Tryp and Figure 3b Au/L-Tryp.

3.3. RS Spectra

RS spectra of dilute solutions of L-Tryp were measured to determine the SERS efficiency of the NPs for the desired application. RS solid L-Tryp and SERS spectra at a low concentration (10⁻¹⁸ M for Ag and 10⁻⁶ M for Au) compared with L-Tryp in aqueous solution (10⁻³ M) are shown in Figure 4. Peak positions are listed in

Table 1. L-Tryp adsorbed on AgNPs and AuNPs.

RS L-Tryp Neutral DFT Calculations		RS L-Tryp Solid	RS L-Tryp Aqueous Solution	SERS DFT Calculations CCSDE		SERS Experimental CCSDE		Assignment (cm−1) DFT Calculations LANL2DZ CCSDE, and Compared with Refs. [2,43]
Ag	Au			Ag	Au	Ag	Au
260		243	245			232		nAg-O
359		352	380	323		329	359	ω-NH2
		408	395	385		385	386	def. R.
		430	412			396	398	def. R, r
476		472		467				β-CC-R, r, γ-CH2
		526	514	579		523	520	β-N-H (r) ν R, r
		537	533	590		529	536	δ(r), β i.p.
	595	598	551	599		574	574	b (R)oop; γ-CH2 (high intensities in SERS)
		630	607	620	640	604	607	γ-CH2; α-NH2; β C-O
		689	679					ν R, r
		710	722			718	711	Def. R, r, n-CN
748	748	749		724	744	740	721	ω-H(R); γ-CH2; β-COO1−
771		757	759	771	758			θ (R), θ (r)
778	778	768		793	775			Def. R, r, α-COO−
787	788			789	790			β (R)-CC-; β (r)
		808						γ-CH2, β-COO1−
845	845	846		869	843		824	β-CH(r)
		858	859			858	864	δ-H(R), β-CH(-NH) (r)
878	878	877		891	889	868		β-H(R), α -H(r)
902	902	933	929	893			942	n-CC-(r); n-CNC-, (r)
959		974		975	980			δ-CH(r); γ-CH2; β-CH; γ-NH2
969	969	993		976	984			δ-CH2; ν-CN
1001		1010	1011	1010	1015	1011	1009	θ (R), θ (r); β H(R)
1031	1031			1028	1024			γ-NH3+, β H(C)
	1037	1066	1060		1036			α(R)s in plane, β-H (R)
	1081	1079		1065	1085	1090	1083	αH(r)
1103	1103	1112		1125	1100			ω-NH3 +, β-H(C)
	1124				1109	1137	1132	ω-NH3 +, β-H(C)
1154	1154	1143	1140	1148	1153			α-H(R), ω-NH3 +, β-CH
1174	1174	1167	1170	1169	1174	1170	1168	γ-CH2, α-H(R);
	1204	1213			1206			υ-(r), υ-C-COO −
1236	1236	1235	1243	1237	1241	1242	1243	α-H(R), γ-H(r)
	1259	1264	1264	1267	1250		1251	γ-H(R), γ-H(r), β-CH
1296	1296	1300	1276		1277	1284	1289	β-H(-CH2)
1318	1318	1318	1319	1291	1296			υ (R), υ (r)
1326	1326	1337		1324	1316	1331	1334	ν-CN, β-CH; ω-CH2
		1342	1345			1343		β-CH, β-H (-CH2)
1360	1360	1361		1361	1365			ω-CH2, β-CH
1376	1376		1374	1375	1374			υ (r), υ (R)
1380	1380			1391	1398			γ-CH(R); n (r); β-CH (-NH); δ-CH2
1398	1398	1427	1399	1404	1405	1396	1387	υs -COO1−
1450	1450	1457		1416	1416	1463	1457	α-CH2, δ-NH2
1459	1459	1461	1461	1453	1441		1464	υ (r), υ (R), ns-NH3 +
1492	1492	1490		1489	1479			β -CH(R)
1517	1517		1504	1513	1506			δ-CH2; SERS: n(R)-CC-, β-CH, β-CH (r) (-NH)
		1560	1557		1541	1560	1541	n(r), n(R)
1587	1587	1581	1603	1598		1595	1578	α-NH3+
1616	1616	1628	1638	1618	1618		1594	Φ (ringst)-CC-st, β-CC-bend, β -CHalk; β-CH (r) (-NH), δ-CH2
					1653			nas-COO−, α-CCO
1659	1659		1666	1661	1667			n (R)-CC-, β-CC-, β CHalk; β-CH (r) (-NH), α-NH3+
							2136	nas-CN
							2254	ns-CN
	2952	2950					2942	nas-H(R)
	3065	3059					3083	ns-NH3+

, together with the appropriate tentative assignment of vibrational modes. Based on the position of the peaks and their relative intensities, an estimate can be made regarding how the molecules are attached to the NPs’ surface. This assignment was compared to previously published data on L-Tryp [2,43,44]. The enhancement factor (EF = [I_SERS/N_SERS]/[I_RS/N_RS]) was calculated [45,46]. The AgNPs [42] had an enhancement factor of 10¹², and the AuNP [45,47] had an enhancement factor of 10¹⁰ [47] (see Supplementary Materials).

Similarly, the calculated RS shift spectra using DFT at the B3LYP level with the LANL2DZ basis set, measured SERS, and RS band positions and tentative assignments are shown in Table 1. SERS spectra are different than those of RS spectra. Similar peaks are observed in the spectra of both techniques. Still, they also exhibit significant differences, including peaks with varying relative heights, higher intensities, different shapes, and the presence of new peaks. L-Trypt has three different ionic forms, depending on the pH of the solution: cationic (⁺¹H₃NRCOOH), anionic (NH₂RCOO¹⁻), and zwitterionic (⁺¹H₃NRCOO¹⁻) [2,48]. The zwitterion form is a neutral structure but has internally charged molecules; the structures of different ionic forms of L-Tryp are shown in Figure 5 [2,44,49,50]. The results clearly demonstrate a notable improvement in the strength of RS signals. This fact not only highlights the efficacy of SERS colloidal suspensions but also underscores their ability to enhance interactions in the carboxylate and amine groups of amino acid structure. The enhancement of the RS signals suggests a greater accessibility and resonance of molecular vibrations, implying that SERS colloidal suspensions are enhancing the spectroscopic properties of these functional groups. SERS spectra of L-Tryp on AgNPs and AuNPs were found to appear quite similar (Figure 4c–e). Specifically, in Figure 4e, new bands in the ~2130–2254 cm⁻¹ range for AuNPs SERS spectra were observed, showing in each case a weaker vibrational signal located at ~2254 cm⁻¹, tentatively assigned as overtones of asymmetric stretches. A medium intensity at 2130 cm⁻¹ was tentatively assigned as an overtone of the approximately symmetric stretch, but it disappeared at pH values close to the pKa of L-Tryp. In contrast, it was absent from the AgNPs SERS spectrum.

L-Tryp has a (-COO⁻) group, an amino group (-NH₂/-NH₃⁺), and a side group consisting of 3-methyl indole. These are the possible active sites for binding to NPs’ surfaces. For (-COO¹⁻), group stretching modes are observed at 1427 and 1628 cm⁻¹. The band at 1560 cm⁻¹ corresponds to a deformation of the NH₃⁺ species. For the (-NH₂/-NH₃⁺) group, stretching modes were observed at 3065 and 3246 cm⁻¹, respectively, in the RS spectra for the solid and for aqueous solutions of L-Tryp [51]. However, the SERS spectra do not present enhancements of these signals (-NH₂/-NH₃⁺) with Ag but were observed on Au colloids. Therefore, the aromatic C-H stretching vibrations, ring-stretching vibrations, the ring-breathing mode, in-plane and out-of-plane vibrations, and the SERS surface selection mode allow for the suggested orientation of the molecule on the silver and gold surface [52,53]. Also, the C-H stretching vibrations were observed in SERS on the Au surface at 2942 cm⁻¹.

L-Tryp has a (-COO⁻) group, an amino group (-NH₂/-NH₃⁺), and a side group consisting of 3-methyl indole. These are the possible active sites which may bind to NPs’ surfaces. For (-COO¹⁻), group stretching modes are observed at 1427 and 1628 cm⁻¹. The band at 1560 cm⁻¹ corresponds to a deformation of the NH₃⁺ species. For the (-NH₂/-NH₃⁺) group, stretching modes were observed at 3065 and 3246 cm⁻¹, respectively, in the NR spectra for the solid and for aqueous solutions of L-Tryp [49]. However, the SERS spectra do not present enhancements of these signals (-NH₂/-NH₃⁺) with Ag but are observed on Au colloids. Hence, the orientation of the molecule on the silver and gold surface can be inferred from aromatic ring-stretching vibrations, C-H stretching vibrations, the ring-breathing mode, in-plane and out-of-plane vibrations, and the SERS surface selection rules [18,54]. Also, C-H stretching vibrations were observed in SERS on the Au surface at 2942 cm⁻¹.

The theoretical vibrational wavenumbers predicted by DFT at the B3LYP level with the LANL2DZ calculation, along with the observed experimental data of L-Tryp on Ag and Au surfaces, are tabulated in Table 1. Additionally, direct views of these vibrational animations are provided in Gaussian View™. Also, Cao and Fisher compared the vibrational modes of L-Tryp with the DSD infrared spectrum [43].

The -C-C- ring breathing mode and -C-C-C- trigonal bending are tentatively assigned to the bands at 748–757 cm⁻¹ and 1001–1010 cm⁻¹ for L-Tryp. The modes were observed as a strong band at 1011 cm⁻¹ for Ag and 1009 cm⁻¹ for Au in SERS spectra, respectively. A corresponding RS band has been observed in the same region [41,54]. Neither a considerable red shift nor significant band broadening related to this vibration mode were identified in the SERS spectra of L-Tryp, indicating a low probability of a direct ring π-orbital to metal interaction. The spectral intensities in the RS solid L-Tryp, RS aqueous L-Tryp solution observed at 1318, 1326, 1337, 1342, 1360–1361, 1380, 1416–1427, 1441–1457, 1479–1492, 1557–1560, 1578–1587, 1616–1618 cm⁻¹ are assigned to carbon vibration in indole ring stretching. The degenerate frequency associated with in-plane carbon bending vibrations has been identified at 529 and 536 cm⁻¹ in the SERS spectra of Ag and Au, as well as at 574 cm⁻¹.

A split into two non-symmetric components is observed at 380 and 395 cm⁻¹, assigned by degenerate carbon out-of-plane bending vibrations. The C-H stretching vibrations are associated with the presence of bands at 2950–2952 and 3059–3065 cm⁻¹. Observing the C-H stretching band in the SERS spectra suggests an inclined orientation of the aromatic group of the ring on a metal substrate, a phenomenon that is well-documented in the literature [51,55]. In this work, the aromatic ring C-H stretching band was not observed in SERS spectra on Ag, indicating that the indole ring adsorbed flat on the Ag surface. The orientation of L-Tryp on the AuNPs surface suggests that there is a certain angle between the ring plane and the Au surface for the appearances of both in-plane modes and out-of-plane modes, and that is not shown in the SERS spectra. The frequencies in the RS solid L-Tryp, RS aqueous L-Tryp solution at 1079, 1167, 1235–1243, and 1276 cm⁻¹ are assigned to -C-H in-plane bending vibration. This frequency was observed at 1090/1083, 1170/1168, 1242/1243, and 1284/1289 cm⁻¹ in the SERS spectra of L-Tryp on Ag and Au colloidal suspension, respectively. The C-H out-of-plane bending vibrations in the RS solid L-Tryp and RS aqueous L-Tryp solution have been assigned at 845, 933-929, and 974cm⁻¹. The weak bands at 1463 and 1457 cm⁻¹ in SERS spectra for Ag and Au, respectively, correspond to the CH₂ scissoring mode. In L-Tryp, the wagging vibrations of CH₂ are found at 1318 cm⁻¹, which is justified by the DFT calculation but does not appear in the SERS spectra.

The carbonyl asymmetric stretching appears as a very weak band at 1618 cm⁻¹ in the SERS spectrum for Ag and Au colloids. The carbonyl stretching indicates that C=O does not interact directly with the metal surfaces of Au and Ag, as evidenced by a very weak vibration. It is also revealing to observe the ns (COO¹⁻) bands at 1396 cm⁻¹ for Ag. This indicates that the Ag-COO¹⁻ and Au-COO¹⁻ bonds formed by the adsorption of L-Tryp on Ag and Au are not ionic. The bonds would then have a covalent character. This carboxylate (-COO¹⁻) group binds to the metal surface via either the oxygen lone pair electrons or the carboxylate π system [47]. On the surface, the geometry of the (-COO¹⁻) is stated by steric hindrance between the carboxylate ion and other adjacent substituents, which determines the nature of binding of these molecules on the surface.

The -N-H out-of-plane bending deformation is suggested to 604/607 cm⁻¹ in the SERS by Ag and Au. The wagging vibrations of -NH₃⁺ can suggest that the amino acids present in the NH₂ form twist when lying on the Ag/Au surface, as determined by pH-dependent measurements, for L-Tryp, located at 1137 cm⁻¹ for Ag and 1132 cm⁻¹ for Au. The -NH₂ form, rather than the -NH₃+ form, is regarded as a preferential structure when amino acids attach to the Ag surface in SERS measurements, according to previous SERS studies [29,56]. This blue shift in its frequency, approximately ±15 cm⁻¹, has been previously reported for other amino acids [18]. The band at 1066 and 1060 cm⁻¹ observed in the RS solid L-Tryp and RS aqueous L-Tryp solution, which corresponds to the n(CN) vibration, was not observed in SERS spectra on AgNPs and AuNPs. It probably overlaps with the band at 1011 cm⁻¹ for Ag and 1009 cm⁻¹ for Au, having suffered a larger red shift on the AuNPs surface because the SERS frequencies of these modes are usually very close to those observed in the RS solid of amino acids. The NH₂ scissoring modes show up between 1590 and 1670 cm⁻¹ in primary aromatic amines. The bands at 1598-1603 cm⁻¹ and 1666–1661 cm⁻¹ were assigned to -NH₂ scissoring mode in RS solid L-Tryp and RS aqueous L-Tryp solution.

3.4. Theoretical Discussion

RS shift calculations of optimized neutral or zwitterion (⁺H₃NRCOO⁻), cationic (⁺¹H₃NRCOOH), and anionic (NH₂RCOO¹⁻) forms of L-Tryp structure were compared with the experimental RS spectra of solid L-Tryp and with spectra of aqueous solutions of L-Tryp. The geometries of these optimizations are shown in Figure 3. For these structures, a comparison of experimental and calculated RS spectra is shown in Figure 6. It is possible to observe similarities between the RS spectrum of solid L-Tryp and the calculated neutral form of this molecule. Still, there are differences because the calculations were performed without anharmonic corrections to save computational cost. The optimization of L-Tryp with the AgNP and AuNP surfaces was completed with the convergence criterion accepted for the neutral form. Still, the calculation with pseudopotential produced a deformation of the L-Tryp structure. For the optimization with the LANL2DZ basis set for all the atoms, it was possible to see an orientation of the carboxylate group of the L-Tryp structure. This was oriented in the same form as the calculation with pseudopotential (Figure 3).

For the chemisorption process to be observed on AgNP and AuNP surfaces, a bond must form between the metal surface of Ag and Au and one of the adsorbed molecules; for example, an Ag-N, Ag-O, Ag-S, or Ag-X (X = halogen) bond [56,57,58]. The interactions between the adsorbate and the metal surface allow the influence of chemical enhancement on SERS spectra to be determined. The relatively weak interactions between the solid surface and the adsorbate, resulting from physical adsorption, show an adsorption enthalpy greater than −25 kJ/mol. In chemical adsorption, the adsorbed molecule binds strongly to the metal surface, exhibiting an adsorption enthalpy of less than −40 kJ/mol [53,59]. In the chemisorption process, the chemical structure and symmetry of the adsorbed molecule change due to the formation of bonds with the surface. This mechanism often introduces wave number shifts [47] between the vibrations of the adsorbed molecules compared to the normal RS spectrum of these molecules, as shown in the Au/L-Tryp SERS spectra in Figure 5.

The energy gaps between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) for L-Tryp on AgNP and AuNP surfaces were calculated at the B3LYP/6- 311G level of theory. These are represented in Figure 7. The energy gap reflects the chemical activity of the molecule. The charge transfer (CT) within the molecule will be greater, the smaller the energy gap between the HOMO and LUMO. In Figure 7, it can be observed that the Fermi levels of the AgNP and AuNP surfaces are located at −446.7 kJ mol⁻¹ and -509.9 kJ mol⁻¹, respectively (Figure 7a,b). The two main binding sites of the carboxyl group and the amino group were taken into consideration to perform the calculations. The highest occupied molecular orbital (HOMO) for L-Tryp is located at −141.8 kJ/mol, thus precluding a CT from L-Tryp to the conduction band of the gold and silver surface. In the cationic form (acid) of L-Tryp, it is important to note that the energy of the first low-lying empty molecular orbital (LUMO) falls within the energy range of the valence band of the silver surface, but is too close to the Fermi level of the gold surface. This fact explains the CT from the Ag and Au surfaces to L-Tryp, and the LUMO (225.8 kJ mol⁻¹) is localized on the -COO¹⁻ moiety, which explains why part of this moiety is directly facing the Ag surface [51]. The adsorption energies of the cationic form of L-Tryp on Ag and Au of the amino group were estimated as 107.7 and −12.4 kJ mol⁻¹, respectively. For adsorption energies through the carboxylate group anionic form of L-Tryp, the values estimated for Ag are −611.7 kJ mol⁻¹ and for Au are −731.8 kJ mol⁻¹. These results were in great agreement with the interpretation that L-Tryp adsorbs via the carboxylate group and the amino group (-NH₂) form on Ag, and for the Au surface, adsorbs via the amino group in its protonated NH₃¹⁺ form and the ionized carboxyl group (-COO⁻).

It is also informative that the difference in adsorption energy between the carboxylate and the amino group is much smaller for Au than for Ag. This could be because the carboxylate group was observed weakly in Au but more intensely in Ag.

3.5. Effect of pH on Vibrational Bands of L-Tryp on Ag and Au Colloids

The pH can affect the way a molecule orients itself on the surface of a substrate. In particular, it alters the protonation state and, consequently, the number and nature of the binding sites between molecules and nanoparticles [35]. Numerous reports are available in the literature focusing on the interactions of L-Tryp with Ag colloids [2,35], and a systematic study on the effect of gradual changes in net atomic charge distributions at different terminals of L-Tryp on Ag metal interaction was also recently reported [35]. However, in this work, the most probable conformation of L-Tryp on Ag and Au colloidal solution was assessed by varying the pH of AgNP and AuNP in the pH range from 2.0 to 11.0. It was important to determine if there were differences in the spectral profiles of the biomolecule studied. The AgNP and AuNP colloidal suspensions were then prepared and pelleted at pH 3.00 and 8.00, respectively. Zeta potentials were measured for each sample to identify the main forces that mediate interparticle interactions and to understand dispersion and aggregation processes with regard to the stability of metal colloids. The measurements of “as-prepared” colloids to generate average values of −37.4 mV at pH 7.35 for AgNP and −34.9 mV at pH 4.10 for AuNP confirm the moderate stability of Ag and Au nanoparticles. The zeta potential value indicates the stability of the colloidal suspension; the higher it is, the more stable the system is, as it suggests that the charged particles repel each other, minimizing the natural tendency to aggregate. As expected, when the acid solution was added to the NP’s suspension, the surface charge turned to more positive values, resulting in a decrease in the magnitude of the zeta potential (which became less negative). High zeta potential values were obtained at pH 2 and 11, therefore correlating with the decrease in SERS signal enhancement as a result of aggregation and flocculation of the Ag and Au colloids.

Contrary to information contained in most SERS reports on AgNP, a 532 nm laser excitation line did not produce highly enhanced SERS spectra or the typical highly resolved traces when exciting the SERS spectra of L-Tryp. This situation is illustrated in Figure 8a. These results also demonstrate that for different diameters of spherical AgNP and AuNP, higher SERS intensities were obtained for average NP sizes of 40–60 nm. This is shown for AuNP in Figure 8b.

AgNP and AuNP were mixed with L-Tryp at different concentrations, and pH measurements were performed prior to measuring RS spectra (see Supplementary Materials). To ensure adsorption equilibrium and pH stabilization, spectral measurements were obtained 20 min after preparing the mixtures. A detailed analysis of the relative variation in intensities of different vibrational bands in Figure 8 and Figure 9 for Ag and Au, along with the results obtained from DFT calculations, shows that with changes in pH, an interesting increase in the metal–molecule interaction is observed for Ag/L-Tryp and Au/L-Tryp. Peak areas of spectra obtained at different pH levels were calculated using OPUS™ software (V. 3.0; Bruker Optics, Billerica, MA, USA). The integration was performed in the regions that showed the most intense bands in the SERS spectra assigned to the bond vibration of the biomolecule.

According to the figures shown, there is a close relationship between the SERS peak areas and the pH for the L-Tryp species under study. At pH values below pK_a and above pK_b, the spectral profile decayed markedly compared to neutral or basic pH values for Ag surfaces and neutral or acidic pH values for Au surfaces, where a high-intensity enhancement of the vibrational modes was evident, reaching maximum intensity at pH ~9.42 and pH ~3.73. Therefore, parameters such as analyte concentration, nanoparticle size, and colloidal pH values have a huge effect on establishing the most stable conformation and orientation of L-Tryp on silver and AuNP surfaces since, according to the results, L-Tryp molecules interact with the Ag surface through the carboxylate group nearest at its pK_b = 9.40, at ~10⁻¹² M concentration, with a particle size of approximately 60 nm, and Au surface L-Tryp molecules interact through the amino group in a neutral form close to its pK_a at pH 3.73, at a concentration of ~10⁻⁸ M and ~40 nm particle size. Good results were also observed at values near its isoelectric point (IP).

3.6. Concentration Profile: L-Tryp on AgNP

Figure 10 shows the obtained concentration profile for L-Tryp adsorbed on AgNP at pH = 7.4. The concentration was linear from picomolar to low femtomolar. A logarithmic (base 10) scale had to be used with the abscissa (“x” value) equal to the L-Tryp concentration and the coordinate (“y” value) the integrated areas of the 1010 cm⁻¹ SERS vibrational band. A limit of detection (LOD) of 10⁻² fM (10⁻¹⁷ M or 1.70 × 10⁻¹³ g) was obtained.

The coefficient of correlation (R²) was 0.988. This is an excellent result when we consider that SERS is mainly a LOD technique, but not a low limit of quantification (LOQ) technique. Typical values for LOQ are ±15–18%. We obtained ±12%. [13,14,15,16,21,22,23,25,26]

4. Conclusions

The AgNP and AuNP colloidal suspensions, with an average particle size ranging from approximately 40 to 60 nm, serve as effective sensing platforms for developing SERS methodologies. These methodologies help establish the most stable orientations of amino acids, such as L-Tryp. The limit of detection for L-Tryp is around 10⁻¹³ M for AgNP and about 10⁻⁸ M for AuNP. The AgNP had an enhancement factor of 10¹⁰, and the AuNP had an enhancement factor of 10¹². RS intensity signals were considerably modified by the pH of the colloidal suspensions, which produced changes in the electrostatic charge on the nanoparticle surfaces. The preparations and surface conditions of the nanoparticles influenced the binding of L-Tryp to the colloids, as well as the electrolytes used to promote the attachment of the molecules to the surface. The position and relative intensities of the peaks in the SERS spectrum of L-Tryp depend on several factors: the excitation laser line, the type of metal substrate, the pH of the medium, ionic strength, the concentration of the solution, the aggregation of the colloid, and the potential chemical interactions between the amino acid and the metal substrate. Specifically, the vibrations υs (-COO⁻) and υs (NH₂) observed in SERS on the Ag surface, and the detection of the νs (-NH₃⁺) at approximately 3030–3130 cm⁻¹ on the Au surface, confirm that L-Tryp was adsorbed in its anionic form on AgNP substrates. The L-Tryp molecule interacts with AgNP through the -COO⁻ and -NH₂ groups, while the indole ring fragment is near the surface and interacts with the Au surface via NH₃⁺, depending on the pH of the colloidal suspension. The pyrrole moiety is located further from the surface than the benzene ring. Therefore, L-Tryp molecules are likely tilted on the AgNP and AuNP surfaces due to the ionized carboxyl group, creating an angle between the indole rings and the surface of the AuNP. Our experiments revealed that the SERS spectra vary according to the different formulations of Ag colloids [2]. The theoretical data closely align with the experimental results.

SERS studies of amino acids and peptides are crucial, as they provide insight into the interactions between proteins and metal substrates. Moreover, they could provide valuable information for much more complex and larger proteins and peptides. L-Tryp and its derivatives are widely distributed and can be converted into several biologically important compounds. Consequently, the results obtained are of significant interest to physicists, neurobiologists, electrochemists, bioengineers, and others.