Rough Fe₃O₄/gold nanostructure enhanced fluorescence of quantum dots for Salmonella typhimurium detection in cabbage

2.1 Materials

Tetrachloroauric(III) acid trihydrate (HAuCl₄·3H₂O, 99.9 %), sodium malate, trisodium citrate (Na₃C₆H₅O₇), triethanolamine, FeCl₃·6H₂O, FeCl₂·4H₂O, ammonia solution (25 %), Staphylococcus aureus, Escherichia coli, shawnelle oneidensis, enterobacter aerogenes, proteus mirabilis, citrobacter freundii, citrobacter diversus, enterobacter cloacae and bacillus, triethanolamine (98 %) were received from Sigma-Aldrich (Oakville, Canada). S. typhimurium aptamers (5′–NH2–AGT AAT GCC CGG TAG TTA TTC AAA GAT GAG TAG GAA A AGA-3) were purchased from Creative Biolabs (NY, USA).

2.2 Synthesis of Fe₃O₄@citrate NPs

The citrate-capped Fe₃O₄ NPs were prepared through a coprecipitation technique (Ahghari et al., 2023). Shortly, FeCl₃·6H₂O (1.622 g) and FeCl₂·4H₂O (0.994 g) were mixed in a beaker of water (40 mL) and stirred gently for 5 min. Then, the ammonia solution (5 mL) was mixed with the reaction mixture, which was then left alone for 15 min. Afterward, 4.4 g of sodium citrate was mixed with the above reaction solution and the temperature was set up to 90 °C for 30 min. After cooling down the reaction solution to room temperature, the black-colored precipitate was obtained. The sample was centrifuged (6500 rpm, 20 min) and washed several times with water and dried under vacuum conditions.

2.3 Preparation of Fe₃O₄@Au composite NPs

At first, HAuCl₄ solution (20 mL, 0.5 mmol/L) was heated to boiling temperature alongside vigorous stirring. Then, different volumes of Fe₃O₄@citrate NPs were added, spontaneously changing the solution color from brown to burgundy. The Fe₃O₄@Au NPs precipitates were separated through centrifugation (6500 rpm, 20 min) and redispersed in water for the following experiments.

2.4 Preparation of urchin-like Fe₃O₄@Au NPs

Urchin-like Fe₃O₄@Au NPs were prepared based on a reported article with some modifications (Han et al., 2023). To do so, various volumes of Fe₃O₄@Au seeds (100–500 μL), 22 μL of 1 % sodium citrate (1 %, 22 μL), and triethanolamine (1000 μL, 30 mmolL^-1) were added into the aqueous solution of HAuCl₄ (10 mL, 0.25 mmol/L) and strongly stirred at room temperature. The change in color (bluish) will indicate the successful synthesis of urchin-like Fe₃O₄@Au NPs. The Fe₃O₄@Au NPs precipitates were separated through centrifugation (6500 rpm, 20 min) and redispersed in water for the following experiments.

2.5 Preparation of Au QDs

To prepare Au QDs, Au NPs were first synthesized using sodium malate. Briefly, 1 mL of sodium malate solution (1 mmol/L) was added to 20 mM of HAuCl₄ (20 mL) solution and heated at 100 °C for 30 min. The initial light-yellow color of the solution turned reddish indicating the formation of Au NPs. After centrifuging three times (5000 rpm, 30 min), the purified Au NPs were dispersed in MilliQ water. Then, 10 mL of Au NPs solution was autoclaved at 200 °C for 2 h. At this stage, the initial reddish color of the solution turned to deep yellow indicating the formation of Au QDs. At first, the synthesized Au QDs was purified through centrifugation (15000 rpm, 30), then again purified by dialysis. The fluorescence emission of the purified Au QDs was measured using fluorescence spectroscopy and the sample was stored at 4 °C before further uses.

2.6 Preparation of nanocomposite film for plasmon enhancement

At first, glass slides (2 × 10 cm) were cleaned using piranha solution for 1 h and washed with deionized water. Then, 1 wt% aqueous solution of positively charged poly(dimethyldiallylammonium chloride) (PDDA) and a 1 mg/mL aqueous solution of negatively charged poly(acrylic acid) (PAA) were prepared separately. The glass slides were first immersed into the PDDA solution for 10 min. Then, the slides were washed with water to remove any unbound polymers and dried with N₂ gas. Next, the slides were immersed in the PAA solution for 10 min and then washed and dried as before. This method was performed to build the (PDDA/PAA)₂/PDDA base layers. These slides were then immersed in spherical-shaped Fe₃O₄@Au NPs solution for 10 min and, subsequently, in the urchin-like Fe₃O₄@Au NPs solution for 10 min. After washing and drying, the slides were immersed into a PDDA/PAA/PDDA solution to make a spacer layer between the metallic NPs surface and QDs. Lastly, the slides were immersed in Au QDs (2 mmol/L) solution for 10 min.

2.7 Conjugation of aptamer with nanocomposite film

The rough film of Fe₃O₄@Au nanostructure prepared above was mixed with an EDC (10 mM) and NHS (4 mmol/L) solution to activate the carboxyl group of the film. Then, the aptamer (10 mmol/L) was added to the film and kept at 37 °C for 1 h. The aptamer-modified film was then washed three times with PBS buffer to remove loosely bound or unbound aptamers and kept at 4 °C before further experiments.

2.8 Conjugation of aptamer with Au QDs

2 mL of Au QDs solution was added to the EDC (10 mmol/L) and NHS (4 mmol/L) solution and kept under dark conditions for 30 min. Then, the aptamer (10 mmol/L) was added to the Au QDs solution and kept at 37 °C for 1 h

2.9 Detection of S. Typhimurium using aptamer modified nano-complex

In this study, aptamer-modified Fe₃O₄@Au and Au QDs form a nano-complex in the presence of S. typhimurium and help to quantify S. typhimurium concentration. Shortly, aptamer-conjugated Fe₃O₄@Au (100 μL) containing various concentrations of S. typhimurium in PBS buffer were added in the different wells of 96-well plates. Then, aptamer-conjugated Au QDs (100 μL) were added to each well, and the fluorescence response within different wells was measured using fluorescence spectroscopy. For the real-life application of the proposed detection method, a fresh cabbage was purchased from the local market and soaked in PBS buffer (10 g) for 1 h. Then, the cabbage was crushed using a blender and filtered to remove impurities. The differently concentrated solution was prepared by mixing S. typhimurium, aptamer conjugated Fe₃O₄@Au (100 μL) & Au QDs, and the changes in fluorescence emission of the mixtures were monitored using fluorescence spectroscopy.

2.10 Characterization

The absorbance and fluorescence properties of NPs were monitored using BioTek spectrophotometer (Synergy H1, USA). The shape and sizes of the NPs were examined by high-resolution transmission electron microscope (HR-TEM) (80–300 LB, Hillsboro, USA) and atomic force microscope (Veeco, diInnova). Zetasizer (Malvern Instruments, Nano ZS) was used for surface charge measurement. The particles were characterized by an X-ray powder diffractometer (XRD; D8 FOCUS 2.2 kW, Bruker, Germany). The lifetime decay profile of QDs was checked using Horiba Lifetime Fluorometer (DeltaFlex TCSPC, London, ON, Canada).

The statistical differences among the datas were computed using one way analysis of variance (ANOVA) and the post hoc Tukey test at the p < 0.05 significance level. The graphs were created using OriginPro-2020.

3.1 Spectroscopic and microscopic characterization of nanomaterials

The optical and electromagnetic properties of Au NPs are prone to changes depending on their sizes and shapes. Particularly in our experiments, a non-spherical shape with many epitaxial growths was intentionally synthesized, inducing irregular optical properties. Fig. 1 presents the experimental data for two different types of core–shell Fe₃O₄@Au NPs, i.e., spherical and urchin-like NPs. First, an Au shell was formed around the Fe₃O₄ core NPs, which is the common procedure for producing many different types of NPs. The Fe₃O₄ NPs were directly added to a boiling HAuCl₄ solution, where the Au salts in the solution resided. Sodium citrates on the Fe₃O₄ NPs’ surfaces reduced the gold salts in the solution to form an Au shell (this is the final step of spherical NP synthesis). After that, the urchin-like Fe₃O₄@Au core–shell NPs were synthesized in the presence of citrate and triethanolamine through a seed-mediated growth approach. Fig. 1 displays representative TEM images of the Fe₃O₄ NPs (Fig. 1A), spherical Fe₃O₄@Au NPs (Fig. 1B), and urchin-like Fe₃O₄@Au NPs (Fig. 1C). It can be clearly observed in the TEM micrograph that the Fe₃O₄ NPs are almost spherical in shape and have a size of ∼ 10 nm and the solution color is yellow. We added the Fe₃O₄@citrate solution to the HAuCl₄ solution to reach a total volume of 10 mL, which led to the synthesis of Fe₃O₄@Au NPs (Fig. 1B). At this point, the solution color changed to purple. After adding triethanolamine (1000 μL) and Fe₃O₄@Au seeds (1000 μL) into the reaction system, urchin-like Fe₃O₄@Au NPs with many spikes were obtained (Fig. 1C). The branches were carefully counted in the TEM images. The average number of branches per NP was approximately 25–30. The sizes of the urchin-like Fe₃O₄@Au NPs were about 185 nm, and the solution color was blue. The diameters of urchin-like NPs were tunable from 105 to 185 nm by altering the feed ratio of the seeds and triethanolamine. A detailed X-ray study was carried out to characterize the composition of significant features, i.e., core and branch of urchin NP (Fig. S1 and Table S1).

Close

Fig. 1

TEM images of (A) Fe₃O₄ NPs, (B) Fe₃O₄@Au NPs, and (C) urchin-like Fe₃O₄@Au NPs and (D) a 3dimentional (3D) schematic presentation of an urchin-like Fe₃O₄@Au NP.

Export to PPT

In this study, Fe₃O₄@Au seeds with diameters of about 30 nm were used for the synthesis of urchin-like Fe₃O₄@Au NPs. Here, trisodium citrate acted as a surface ligand and both citrate and triethanolamine acted as reductants during the synthesis of the urchin-like Fe₃O₄@Au NPs. A step-by-step reduction process, i.e., Au^Ⅲ→Au^Ⅰ→ Au⁰ was followed during the synthesis reaction. However, the weak reduction ability of trisodium citrate at room temperature allows only the conversion of Au^Ⅲ to Au^Ⅰ instead of Au⁰. However, the high selectivity of triethanolamine to reduce Au^Ⅰ to Au⁰ on the Fe₃O₄@Au seed surface in the presence of gold seeds permits the completion of the stepwise reduction process (Han et al., 2023). Moreover, triethanolamine controlled the formation of an urchin-like structure on the surface of Fe₃O₄@Au NPs. To test this, we examined the ultraviolet–visible (UV–visible) spectra of the solution for triethanolamine-assisted reduction of HAuCl₄ in the presence and absence of particle seeds. We examined the kinetics of the reaction solution; no particle formation was observed until the seeds were added (Fig. S2). This demonstrates the selectivity of triethanolamine and the importance of the seeds.

The analysis of HR-TEM and powder XRD data confirmed the existence of highly active facets on the urchin-like Fe₃O₄@Au NPs (Fig. 2). Fig. 2D illustrates the XRD pattern of urchin-like Fe₃O₄@Au NPs. The XRD peaks of urchin-like Fe₃O₄@Au NPs matched perfectly with both its precursor materials: Au and Fe₃O₄ bulk counterparts. The diffraction peaks of urchin-like Fe₃O₄@Au NPs observed at the (1 1 1), (2 0 0), (2 2 0), (3 1 1), and (2 2 2) positions can be assigned to Au planes structure. The diffraction peaks obtained at (2 2 0), (3 1 1), and (4 4 0) positions can be assigned to the Fe₃O₄ plane structure (Adhikari et al., 2022). The finding of XRD peaks revealed the polycrystalline nature of urchin-like Fe₃O₄@Au NPs. In addition, the strongest diffraction peak found at the (1 1 1) position reveals the formation of a branched structure of the Fe₃O₄@Au NPs could be attributed to the rapid buildup of Au⁰ on the (1 1 1) lattice planes. The HR-TEM observation of the branches confirmed this hypothesis (Fig. 2B). For example, the interplanar distances of 0.211 and 0.196 nm were observed and found to be consistent with the (1 1 1) and (2 0 0) planes of Au crystals. These findings revealed that the direction of the Fe₃O₄@Au NPs structure formation was vertical towards the (1 1 1) plane and that the plane has a high-energy facet.

Close

Fig. 2

(A) HR-TEM image of one branch of the urchin-like NPs. (B) Growth direction of the branch. (C) Selected area electron diffraction (SAED) pattern of the urchin-like NPs. (D) XRD pattern of urchin-like NPs.

Export to PPT

Au NPs have strong absorption peaks in the visible region because of their surface plasmon properties. Pointedly, citrate-capped Au NPs in water exhibit a strong absorbance peak at around 520 nm. The absorbance spectra of the Fe₃O₄@Au NPs prepared with Fe₃O₄ NP solution were presented in the Fig. S3A. There were no noticeable absorption peaks of Fe in the visible regions of the spectra of the pure Fe₃O₄ NP solution, however, absorbance peaks in the range of 470–800 nm were observed for the Fe₃O₄@Au NP solutions, revealing the development of an Au shell on the top of Fe₃O₄ NPs.

Close

Fig. 3

TEM images of urchin-like Fe₃O₄@Au NPs for different seed amounts: (A) 100 μL, (B) 200 μL, (C) 300 μL, and (D) 400 μL.

Export to PPT

The peak wavelength changed from 600 to 670 nm in the UV–visible absorption spectra by increasing the reactive volume from 100 to 500 μL (Fig. S3B), indicating that the diameters of the urchin-like NPs can be controlled in the range from 175 nm to 95 nm (Fig. 3). Here, the higher Au⁰/seed ratio of the seed solution helps to increase the growth of the NPs. It is preferable to control the size of the Au nanocrystals by varying the initial concentration of citrate to that of Au solution. In our reaction, the spherical Fe₃O₄@Au NPs were used as the nucleation centers of Au for forming urchin-like Fe₃O₄@Au core–shell NPs. Hence, a faster reduction was observed with increasing the reaction ratio of Fe₃O₄@Au/HAuCl₄. This depleted the solution of precursor atoms available for growth, leading to smaller particles. This mechanism matched well with the results of TEM images, as shown in Fig. 3 (A-D). Furthermore, branched growth was facilitated rather than isotropic growth because of the higher Au⁰/seed ratio, and the type of growth also depended on the amount of triethanolamine.

The triethanolamine concentration-dependant properties of the Fe₃O₄@Au NPs were presented in Fig. 4. The absorption peak of the Fe₃O₄@Au NPs moved from 610 to 650 nm with increasing amount of triethanolamine from 125 to 1000 μL (Fig. 4A). The particle size and number of branches increased (Fig. 4B, 4C, and 4D) as well which indicated that the addition Au⁰ in the reaction system has played a significant role in the construction of urchin-like Fe₃O₄@Au NPs. The increase in the amount of triethanolamine accelerates the conversion of Au^I to Au⁰, forming the urchin-like structure of NPs. Although the absorption peak of the NPs formed with the addition of 1000 μL of triethanolamine generally appears at longer wavelengths than that of NPs formed with 500 μL, the opposite result was occasionally observed (Fig. 4A). This meant that 1000 μL of triethanolamine was adequate for the present reduction process. Most probably, the extra triethanolamine had adsorbed on the surface of the NPs and suppressed its growth. This result implies that there are multiple functions of triethanolamine, which acts both as a reducing and capping agent.

Close

Fig. 4

(A) UV–visible absorption spectra of urchin-like Fe₃O₄@Au NPs in different volumes of triethanolamine: 125 μL (a), 250 μL (b), 500 μL (c), and 1000 μL (d). The amounts of seeds and sodium citrate were fixed at 100 and 22 μL, respectively. (B–D) TEM images of urchin-like Fe₃O₄@Au NPs in different volumes of triethanolamine: (B) 125 μL, (C) 500 μL, and (D) 1000 μL.

Export to PPT

3.2 Modifications of fluorescence emission of QDs on rough surface

The coupling of excitonic properties of QDs and plasmonic properties of metallic nanoparticles at a nanoscale distance widens its applications in diverse areas. In this paper, we demonstrated the photoluminescence (PL) changes of QDs attached to metallic films with different controlled surface roughness. The surface roughness of a metallic film can have a unique effect on achieving a strong coupling regime due to the scattering of plasmonic energy. Layer-by-layer (LBL) assembly was used as a bottom-up nanofabrication technique to make nanocomposite films. A schematic illustration of metal-enhanced fluorescence is shown in Scheme S1.

In our experiment, we deposited equal amounts of spherical and urchin-like Au NPs on a glass substrate, which was confirmed by measuring the optical density (OD) of the surface. The OD of each film was fixed at 0.6. The root-mean-square (rms) roughness R_rms was observed using topographic images obtained from AFM measurements in the tapping mode, as shown in Fig. S4A and S4B. The R_rms of the spherical gold NP surface is about 1.2772 ± 0.21 nm, whereas, for urchin-like gold NPs, the Rrms is 2.2286 ± 0.19 nm. A comparison of the R_rms values of the two surfaces indicates that urchin-like NPs produced more defects in the film because of the uneven surface structure in the nanoscale process. It is known that rough films are more advantageous in enhancing optical properties (Ahmed et al., 2014). PL spectra of the samples were recorded using a 380 nm excitation xenon lamp. Fig. S4C shows the PL spectra taken for Au QDs on two different gold surfaces. We found more than twofold emission enhancement of the Au QDs deposited on urchin-like gold compared to that of the spherical gold surface. PL decay was studied with a 380 nm laser excitation source at room temperature. The curves in Fig. S4D show the PL time profiles of the QDs on two different gold surfaces. The decay time of Au nanocrystals on urchin-like gold surfaces (3.23 ± 0.06 ns) is shorter than that for spherical gold surfaces (4.10 ± 0.10 ns). It is well known that the photoluminescence of a semiconductor nanoparticle is significantly enhanced when the local surface plasmon field of metal is nearby. It is also noticeable that the local field effect can be achieved by using rough metal surface instead of metal nanoparticles. Usually, the light field is diffraction-limited in a nano-size dielectric slit. However, the field is dramatically strengthened in the case of metallic slit as the surface plasmon field is concentrated at a local spot. Therefore, a similar effect can be expected in an urchin-like Fe₃O₄@Au structure as the local field near the urchin tips is possibly stronger than the dipole field of a spherical metal nanoparticle. In fact, we observed over twofold enhancement of the photoluminescence in urchin-like Fe₃O₄@Au structures in comparison to spherical Au nanoparticles. The decay time of Au QDs is shortened whereas in the absence of metal structures, there are no changes to the decay time. Regarding the sharp tip of an urchin-like structure, one may be concerned with non-radiative energy transfer or quenching near the short inter-molecular distance. However, the decay time constant barely changes in the case of spherical Au nanoparticles despite the photoluminescence enhancement. These results suggest that the enhancement by the urchin-like Fe₃O₄@Au is possibly associated with the surface plasmonic local field effect in the nanotips of the urchin-like Fe₃O₄@Au structure.

3.3 Detection of S. typhimurium

The enhancement of fluorescence emission on urchin-like Fe₃O₄@Au NPs film was utilized for the detection of sensitive S. typhimurium. The conjugation of the target specific aptamer with urchin-like Fe₃O₄@Au NPs film and Au NPs was confirmed through the ELISA method. As shown in Figure S5, a higher absorbance value was achieved for aptamer-conjugated urchin-like Fe₃O₄@Au NPs film and Au NPs indicating the successful conjugation of aptamer with urchin-like Fe₃O₄@Au NPs film and Au NPs.

Then, different concentrations of S. typhimurium solutions were added to the aptamer with urchin-like Fe₃O₄@Au NPs film and Au QDs, and the changes in fluorescence intensity were observed. As shown in Fig. 5, a linear relationship of fluorescence intensity and S. typhimurium concentration in the range of 50–1000 CFU mL⁻¹ (Fig. 5A) and 100–1000 CFU mL⁻¹ (Fig. 5B) was achieved in PBS buffer and the cabbage sample, respectively, and the calculated LOD value in cabbage sample was 32 CFU mL⁻¹.

Close

Fig. 5

Detection of S. typhimurium in PBS (A)and crushed cabbage sample (B).

Export to PPT

The specificity of biosensor is a common problem in immunological-based assay and in molecular-based identification, which might lead to a false positive result. Thus, the use of a highly specific aptamer biorecognition element could prevent such problems. In general, aptamer as a biorecognition element showed a good selectivity detection towards target analytes. The specificity of this study was examined in the presence of Staphylococcus aureus, Escherichia coli, shawnelle oneidensis, enterobacter aerogenes, proteus mirabilis, citrobacter freundii, citrobacter diversus, enterobacter cloacae and bacillus. As shown in Figure S6, a significant enhancement in fluorescence was observed for S. typhimurium whereas no such enhancement was found for Escherichia coli, shawnelle oneidensis, enterobacter aerogenes, proteus mirabilis, citrobacter freundii, citrobacter diversus, enterobacter cloacae and bacillus indicating the proposed sensing strategy to be highly selective towards S. typhimurium.

To assess the analytical performance of the present study, S. typhimurium was added to a fresh vegetable washing solution and used as samples for testing. We also studied the intra- and inter-day experiment to check the accuracy of the proposed method. Precision and accuracy were assessed by replicate analysis (n = 6) of spiked samples at different concentrations. A unifactorial ANOVA was carried out to evaluate if there were statistically significant differences between the fluorescence intensity values of the spherical and non-spherical surfaces, two tests in a day, tests in different days. The results showed that the fluorescence assay rough morphology presented higher fluorescence intensity values than those of smooth surfaces, with statistically significant differences (p-value of 0.018). Moreover, a one-way ANOVA was performed to assess whether there were statistically significant differences between the intraday assay and intraday fluorescence assay values from the rough surfaces. No differences were observed in this case since there were p-value of 0.37 and 0.41, respectively. Hence, the proposed method is highly reproducible, repeatable, and well-fitted to apply in real-life application in complex media.

As presented in Table 1, the intraday and interday precisions were between 2 % and 5 % and the detection of S. typhimurium in the spiked sample has a satisfactory percent recovery of 98.00 % to 104.22 %, with an RSD value of less than 4.0 %. These results revealed that the proposed detection method is highly precise and trustworthy for the detection of S. typhimurium in real life samples.

The proposed detection method was validated with a commercial kit (CD Creative Diagnostics, USA). The assay protocol and sample preparation were strictly followed as mentioned in the product manual. As shown in Table S2, the detection performance of the proposed method was superior to the commercial one. Moreover, a comparison study with the reported article for S. typhimurium detection has been presented in Table S3 that revealed the present study is comparable to other reported articles. Most importantly, the synthesized nanoparticles were stable upto one year. As shown in Table S4, zeta potential values were unchanged upto one years revealing the stability of nanoparticles. The proposed assay showed 95 % reproducibility with different batches of nanoparticles (Table S5).