TY - JOUR
AU - He,, Lili
AB - Abstract Background Surface-enhanced Raman scattering (SERS) has been deployed in the analysis of food at solid and aqueous states. However, its capability has not been fully explored in headspace profiling. Objective To develop an innovative SERS method for analyzing headspace volatile compounds in foods. Methods A volatile-capture device was developed by depositing a film of silver nanoparticles in a vial cap to capture the volatiles released from a model flavor compound (garlic). Results SERS peaks at 1632, 1400, 1291, 1191, 731, and 577 cm−1 were identified in the headspace of the garlic sample, which was representative of an organosulfur compound (diallyl disulfide), and its concentration was determined at 135 ppm, which was comparable to the value determined using GC. Preparation and analysis could be carried out in <10 min for the SERS method. The sensitivity of the SERS method (10 ppm), however, was slightly less than that of the GC method (5 pm). Conclusions The SERS method was able to quantify the concentration of diallyl disulfide in the headspace of a raw garlic ethanolic extract. Compared to GC, the SERS method had a much shorter analysis time and simpler sample preparation procedure than GC when analyzing large numbers of samples. Highlights The innovative “mirror-in-a-cap” substrate was simpler and faster than other reported SERS substrates used for this purpose. Additionally, SERS has much better portability and the potential for real-time monitoring of changes in the garlic headspace concentration during manufacturing and processing. Garlic is widely used in foods, nutrient supplements, and herbal medicines for its characteristic flavor and potential health benefits (1, 2). Garlic has been reported to contain a range of bioactive compounds, including allicin, which is chemically unstable and decomposes into a variety of organosulfur compounds, such as diallyl sulfide (DAS), diallyl disulfide (DADS), and diallyl trisulfide (DATS) (3). The organosulfur compounds in garlic have been reported to exhibit antimicrobial, anticarcinogenic, and anti-cardiovascular disease properties, which has led to their incorporation into dietary supplements and herbal medicines in the form of extracted oils or powders (2–7). Many of these organosulfur compounds are not chemically stable, and so it is important to measure changes in the composition of garlic extract for quality assurance purposes (1). The characterization can be conducted using HPLC for products such as garlic oil extract. However, since these organosulfur compounds are mostly aromatic and volatile, it is often better to identify and quantify them in the headspace of garlic extracts using GC, often coupled with MS (2). Chromatographic methods are currently the gold standard for providing accurate analysis of garlic composition, however, they have a number of drawbacks, including complicated instrument operations and/or sample pretreatment, the requirement for trained personnel, and the lack of portability. Therefore, alternative fast, simple, sensitive, and portable techniques are gaining more attention. Surface enhanced Raman scattering (SERS) is a powerful analytical tool for chemical analysis because it is rapid, sensitive, simple, and portable (8). Previous research has shown that SERS can detect volatile compounds in a gas phase. For instance, a gold nanoparticle-coated SERS fiber was recently developed for the solid-phase extraction and detection of pesticides within the headspace gasses of apple juice (9). SERS has also been used to analyze the presence of sulfur dioxide in red wine samples that were either dropped on a gold nanoparticle-coated thin film or filtered from the headspace onto a graphene oxide/gold nanorod hybrid filter paper (10, 11). Raman spectroscopy has also been used to directly measure flavor and aroma profiles in essential oils placed in quartz cuvettes (12). In addition, SERS has been used to determine the headspace gas profiles of garlic, leeks, and Chinese chives by collecting the volatiles using a syringe and then injecting them into a suspension of silver nanoparticle (AgNP) colloids (13). With the exception of SERS fiber research, few studies have analyzed the volatiles collected directly from the headspace of a sample. Instead, they have involved some kind of liquid extraction of the sample prior to SERS analysis. Consequently, the development of an effective SERS method that can measure the concentration of volatiles in the actual headspace of a sample is still required. The main objective of this study was to develop a SERS method coupled with a mirror-in-a-cap substrate to analyze the volatile compounds in the headspace of a garlic extract based on solid phase extraction. Furthermore, the potential for quantifying the concentration of the volatile compounds in the headspace of the sample was investigated by utilizing a calibration curve approach. Finally, the quantitative performance of the SERS method developed in this study was compared to a conventional GC method. Methods Materials Fresh garlic was purchased from the local grocery store. DAS, DADS, DATS, 1-propanethiol, 2,5-dimethylthiophene, diallyl sulfide, dimethyl disulfide, dipropyl trisulfide, and s-allyl 2-propene-1-sulfinothioate were purchased from the Sigma Aldrich Company (St. Louis, MO). Silver nitrate, sodium citrate, and ethanol were purchased from Thermo Fisher Scientific (Waltham, MA). Fabrication of AgNP Mirror-in-a-Cap Detector An AgNP mirror-in-a-cap detector was fabricated to enhance the SERS signal from the samples. AgNPs were synthesized according to a well-established protocol (14). Briefly, 100 mL of a 1 mM silver nitrate solution was heated on a hot plate with vigorous stirring at 350°C, until boiling. Immediately after, 1 mL of 100 mM sodium citrate solution was added and the solution was boiled for 25 min until a greenish brown color was observed, which indicated the formation of AgNPs. After cooling to room temperature, approximately 70 nm AgNPs were obtained. The nanoparticle suspension was then diluted to 100 mL using distilled water to reach a final concentration of 1 mM AgNPs. Acetonitrile and hexane were mixed at a mass ratio of 1:1 using a vortex unit. After mixing, the non-polar and polar layers formed were separated. The polar layer was collected and stored as the mediating solvent. Commercial AgNPs were first concentrated by centrifugation to 0.2 mg/mL. Then, 50 μL of AgNPs were slowly added dropwise into 100 μL of mediating solvent. After about 1 min, a mirror-like sediment was formed at the bottom of the sample, which was collected, and then carefully dropped and air dried on an aluminum covered glass slide. The resulting aluminum film, with the dried AgNP mirror on top, was then cut to fit within the caps of 2 mL glass vials. The mirror-in-a-cap SERS active substrate was then fabricated by attaching the segment of film into the inner side of the vial cap using double side tape. Preparation of Minced Garlic, Garlic Extract, and Chemical Standard Ten grams of fresh garlic was minced and then stored in a 20 mL glass vial. For the garlic extract, 10 g of fresh garlic was minced in 10 mL of ethanol in a blender, and then the resulting mixture was stored in a 20 mL glass vial. Ethanol was used to avoid chemical degradation of the compounds in the garlic extract prior to analysis. Three organosulfur compounds that are known to be present in garlic were analyzed in their pure form to establish their characteristic SERS peaks. To keep the same analytical conditions as for the garlic extract, DADS was prepared in ethanolic solutions in concentrations ranging from 5 to 500 ppm. For the reference standards, the pure standards were mixed with AgNPs at a volume ratio of 1:1, and 10 μL of mixture was dried on a gold covered glass slide to form a “coffee ring” for SERS measurements. Analysis of Garlic Headspace Using SERS Aliquots containing 25 μL pure organosulfur standards, DAS, DADS, DATS, and other standard references, were transferred into 2 mL glass vials individually. Mirror-in-a-cap sensors were then placed on glass vials containing the samples for different times and the SERS spectra were acquired. All these measurements were carried out at ambient temperature. To speed up the release of volatile compounds, as well as to mimic the heating conditions used in the GC analysis, 800 μL aliquots of DADS ethanolic solution were also incubated in the mirror-in-a-cap sensors in 2 mL glass vials for 5 min in a boiling water bath for quantitative analysis. For freshly minced garlic and garlic extracts, 1 g minced garlic or 1 mL ethanol extract was incubated in the mirror-in-a-cap sensor in a glass vial to allow the volatile compounds in the headspace to interact with the substrate for 2 h (room temperature) or for 5 min (boiling) before carrying out the SERS analysis. SERS analysis of the captured volatile compounds on the mirror-in-a-cap substrate were performed using a Raman spectroscopy microscope (DXR, Thermo Scientific) with a 780 nm laser source under the following conditions for headspace analysis: 10× objective; 3.1 μm spot diameter; 5 mW laser power; 2 s exposure time; and 50 μm slit width. Ten spots (Raman spectra) were collected for each sample in Raman analysis and three replicates were measured for each concentration. Analysis of Garlic Headspace using GC Headspace analysis of the volatiles in the headspace above the garlic samples was carried out using a previously described method (15). Briefly, the headspace profile was determined using a gas chromatography instrument (GC2010, Shimadzu, Columbia, MD) equipped with an auto sampler (AOC-20i, Shimadzu). Since the GC is coupled with an injection-based sampler, only ethanolic garlic extract or DADS solutions were analyzed. One milliliter samples were stored in GC vials and 1 μL of sample solution was injected by an auto sampler and vaporized at 250°C in the injector, and then separated within an SH-Rxi-5 ms capillary column (15 m  × 0.25 mm inner diameter × 0.25 μm). The GC column was programmed with the following settings: hold at 80°C; heat from 80 to 100°C at 5 °C/min; hold for 2 min. The overall analysis time for each sample was 7 min. Helium was used as the carrier gas and the flow rate was set at 0.91 mL/min. The headspace profile of garlic was determined by matching the measured retention times with those of the sulfide standards. Concentrations of the dominant sulfides were determined based on the peak area and a calibration curve prepared with sulfide standards. Statistical Analysis In this study, all samples were measured within a Raman shift range of 400–2000 cm−1. Raman spectra were analyzed using Thermo Scientific TQ analyst 8.0 software. The spectra similarity was analyzed by searching the similarity of garlic headspace spectra to the library of organosulfur compounds (i.e., DAS, DADS, DATS) in the matching function of TQ analyst. The Raman readings at the signature peak 1632 cm−1 or the GC peak area of DADS at 4.85 min of the headspace of DADS solutions were plotted as a function of DADS concentration. Linear regression analysis and nonlinear fitting analysis were both conducted using the GraphPad Prism software to determine the best fitting model for quantification. Results and Discussion Headspace Characterization of Minced Garlic A schematic diagram of the mirror-in-a-cap substrate is shown in Figure 1A. An AgNP mirror deposited onto aluminum foil was cut into shape and then fitted into the cap of a 2 mL glass vial and fixed with double-sided tape. The headspace profile characterization of the garlic using the mirror-in-a-cap substrate is illustrated in Figure 1B and the spectra of the garlic headspace is shown in Figure 1C. Distinct peaks were observed in the Raman shift at 1632, 1400, 1291, 1191, 731, and 577 cm−1. Garlic is known to be rich in various organosulfur compounds (16). Therefore, to determine the volatile compounds in the garlic headspace, a series of standard organosulfur compounds known to be present in garlic were tested using the SERS method (seesupplemental information). After band assignment, it was found that DAS exhibited the most similar peak assignments to the distinct bands acquired from the garlic headspace. Nevertheless, some extra peaks and shifted peaks were still observed in the SERS spectra. These results suggest that the dominant compound in the garlic sample could be DAS, but that some other sulfides may also have been present. Interestingly, the headspace SERS spectra showed a much higher resolution than the spectra of the chemical standards obtained from coffee rings. This finding is in agreement with a previous headspace analysis study, where the improved resolution was attributed to less interreference from the environment (9). Figure 1. Open in new tabDownload slide (A) Assembly of the “mirror-in-a-cap” SERS substrate. (B) Illustration of the headspace analysis of minced garlic. (C) SERS spectra of the headspace of minced garlic. Figure 1. Open in new tabDownload slide (A) Assembly of the “mirror-in-a-cap” SERS substrate. (B) Illustration of the headspace analysis of minced garlic. (C) SERS spectra of the headspace of minced garlic. Previous studies have reported that allicin is the dominant flavor compound in garlic (16). Due to its chemical instability, allicin can easily be oxidized into a group of organosulfur compounds that are responsible for the characteristic flavor of garlic. The major compounds in this “garlic oil” are DAS, DADS, and DATS (16–19). To further investigate the corresponding organosulfur compounds, pure DAS, DADS, and DATS were tested with the mirror-in-a-cap substrate at room temperature until distinct Raman spectra were obtained. Their SERS spectra and chemical structures are shown in Figure 2. Due to their similar chemical structures, DAS, DADS, and DATS all showed distinct peaks at 1632, 1400, 1291, 1191, and 577 cm−1, however, additional peaks were also observed at 1606 cm−1 for DAS and at 1550 cm−1 for DATS. Therefore, DADS seems to be the only compound among three which showed the exact same peak assignments as the garlic headspace spectra. A spectra library was later established for DAS, DADS, and DATS and the similarity of garlic headspace spectra was searched within the region from 500 cm−1 to 2000 cm−1, and the similarity percentage is shown. As shown in Figure 2, DADS showed the highest percentage of spectral similarity, which suggested that DADS could be the dominant compound among the garlic headspace profile, which therefore dominates the garlic headspace in the SERS measurements. This result is consistent with previous studies, which have reported that DADS is one of the most abundant compounds found in garlic oil, making up between 22–40% (13, 17). Figure 2. Open in new tabDownload slide Three major sulfides in garlic oil and their corresponding SERS spectra and their spectral similarity to the spectra of headspace of garlic. Figure 2. Open in new tabDownload slide Three major sulfides in garlic oil and their corresponding SERS spectra and their spectral similarity to the spectra of headspace of garlic. Interestingly, distinct SERS peaks took significantly longer (2 h) to appear for the pure standard of DATS than for the other two compounds (< 1 h) at room temperature. The delay of the appearance of the SERS peaks for DATS could be due to its relatively high boiling point. According to the Food and Agriculture Organization of the United Nations, at a pressure of 760 mmHg, the boiling point of DAS is 141°C, DADS is 185°C, and DATS is 229°C. Therefore, at room temperature, both DAS and DADS can be quickly detected by the mirror-in-a-cap substrate, but DATS cannot. Even though distinct SERS spectra were obtained from the DADS and garlic samples, the incubation time is still too long for a fast detection method, especially for lower aroma concentrations. In the standard GC method, a heating process is required for headspace analysis. Therefore, to mimic the GC analyzing conditions and to speed up the release of DADS from the samples into the headspace, a boiling water bath incubator was used to heat the DADS samples in a reproducible fashion. Initially, 500 ppm of DADS in ethanol were placed in a glass vial sealed with a mirror-in-a-cap substrate. This sample was then heated in a boiling water bath and the SERS spectra obtained were compared to the DADS sample incubated at room temperature. As shown in Figure 3A, distinct DADS Raman spectra were observed in only 5 min for the heated sample compared to 2 h for the sample incubated at room temperature, see (Figure 3B). Additionally, as shown in Figure 3B, only AgNPs mirror spectra can be observed after 5 min incubation at room temperature. This finding suggests that the heating process is very important to promote the release of DADS into the headspace, which can effectively shorten the sample incubation time. Figure 3. Open in new tabDownload slide (A) SERS spectra of the headspace of 500 ppm DADS in ethanol incubated with boiling water bath for 5 min. (B) SERS spectra of the headspace of 500 ppm DADS in ethanol incubated under room temperature for 5 min and 2 h. Figure 3. Open in new tabDownload slide (A) SERS spectra of the headspace of 500 ppm DADS in ethanol incubated with boiling water bath for 5 min. (B) SERS spectra of the headspace of 500 ppm DADS in ethanol incubated under room temperature for 5 min and 2 h. Quantification of DADS in Headspace of Garlic Ethanolic Extract After demonstrating the ability of SERS to qualitatively detect the presence of specific volatiles in a garlic headspace, the ability of SERS to quantify the DADS in the headspace above garlic was investigated. For these studies, the garlic extract was prepared by immersing freshly minced garlic in ethanol (10 g garlic in 10 mL ethanol), used to facilitate comparison with the GC method. The headspace profile of the ethanolic garlic extract was compared with the DADS standards in ethanol. All samples were incubated in a boiling water bath for 5 min and the spectra of the headspace of garlic ethanolic extract were measured (Figure 4A). As expected, all the peaks present in the spectra of the headspace volatiles collected from the minced garlic corresponded to DADS (Figure 1B) and there are no peaks corresponding to DAS and DATS. For comparison, 1% DATS in ethanol, which is much higher than its concentration of 0.08% in garlic, was also tested by heating in a water bath for 5 min. However, no signature peaks were observed, which was attributed to the high boiling point of DATS (data not shown). For the quantification of DADS, its distinct peak at a Raman shift of 1632 cm−1 was chosen as the reference band and the intensity was plotted as a function of DADS concentration from 1–500 ppm in ethanol to build a calibration curve (Figure 4B). A positive correlation was observed between the intensity and the concentration of DADS, however, the trendline was not linear. According to a literature review, the release of DADS from a food system into the headspace follows a second-order saturation decay because its release rate decreases gradually when the headspace becomes saturated (20). This phenomenon may explain why the Raman intensity readings were not proportional to the concentration: the headspace might become saturated with DADS during heating, resulting in a deceleration of aroma release. Therefore, a second-order prediction model was used in Figure 4B, which gave a goodness of fit of 99.7%. The equation representing the calibration curve is shown below. Raman Intensity=-0.01×C2+11.5×C+86.9 Figure 4. Open in new tabDownload slide (A) SERS spectra of the headspace of garlic ethanolic extract. (B) Raman intensity at 1632 cm−1 plotted as a function of the DADS concentration. The coefficient of determination and the equation are displayed. Figure 4. Open in new tabDownload slide (A) SERS spectra of the headspace of garlic ethanolic extract. (B) Raman intensity at 1632 cm−1 plotted as a function of the DADS concentration. The coefficient of determination and the equation are displayed. where C is the concentration of the volatiles. In the spectra of the headspace of the garlic extract, the intensity of the peak at 1632 cm−2 was measured as 1465. After plotting back to the prediction model, the concentration of DADS in the garlic extract headspace was determined as 135 ppm. To determine the accuracy of results obtained from the mirror-in-a-cap substrate, a standard gas chromatography method was performed on the same samples following a previously described protocol (15). Figure 5A shows the chromatogram of garlic extract, DADS, and Allyl Disulfide (ADS). According to the retention time of DADS (4.85 min) and ADS (2.1 min), both of them were found to be present in the headspace of the garlic extract, and the DADS dominated the chromatogram with a concentration of 112 ppm determined by the calibration curve. After comparing the results obtained from GC and SERS, it was found that the concentration of DADS obtained using the GC method (112 ppm) is somewhat lower than that determined by the SERS method (135 ppm). In SERS, the peak of 1632 cm−1 represents the vibration of the C = C stretching, which is a functional group not only in the structure of DADS but also in DAS, DATS, and some other garlic constituents. Therefore, the intensity of the signal measured at 1632 cm−1 can be considered to be the sum of all those compounds. As a result, the DADS concentration determined by SERS was slightly higher than that determined by GC. Even so, the SERS method does provide a reasonably good estimate of the DADS concentration because it is the dominant component. Figure 5. Open in new tabDownload slide (A) Chromatogram of the garlic extract in ethanol, 0.05% DADS in ethanol, and allyl disulfide in ethanol. (B) The calibration curve of the DADS, the area of band plotted as a function of concentration of DADS. Figure 5. Open in new tabDownload slide (A) Chromatogram of the garlic extract in ethanol, 0.05% DADS in ethanol, and allyl disulfide in ethanol. (B) The calibration curve of the DADS, the area of band plotted as a function of concentration of DADS. Conclusions Table 1 shows the analytical behavior between GC and SERS methods for the characterization of the garlic headspace. Compared to the GC results, the mirror-in-a-cap SERS method showed a much shorter overall analysis time and a simpler sample preparation procedure. This is because GC often requires a sample cleaning step to protect the column from clogging, which will increase the sample preparation time, whereas SERS does not require any sample cleaning. Although the SERS method also requires a heating step, since multiple samples can be incubated together in a large batch, the sample preparation time for multiple samples is the same as for a single sample (5 min). However, even though the SERS method is promising in the profiling of DADS in the headspace of garlic, the lowest detectable concentration, the coefficient of determination, and the accuracy still need to be improved. Additionally, some volatile compounds are unstable to heat, which will cause a change in the headspace profile during sample preparation (21). This is an important limitation for chromatographic headspace detection methods, since samples might take days to be delivered and analyzed, when the profile has already changed dramatically, and the target might be lost. Conversely, the availability of portable SERS devices enables the rapid on-site measurement of volatiles, which may reduce errors caused by chemical changes during storage. Future studies will focus on the further improvement of the sensitivity and quantitative ability of the SERS method, as well as its extension to other volatile compounds in food products. Funding National Institute of Food and Agriculture-USDA and Department of Food Science, University of Massachusetts, Amherst. The research leading to these results are funded by USDA-NIFA 2015-67021-22993, 2015-67017-23070, 2016-67017-24458, and USDA-NIFA hatch MAS00491. Conflict of Interest: Yanqi Qu and Lili He declare that there is no conflict of interest in this manuscript. Table 1. Analytical behaviors comparison between GC and SERS . Gas chromatography . SERS . Analyzing time 7 min for each sample 1 min for each sample Sample pretreatment Sample might need to be filtered for the protection of columns 5 min boiling water bath for a batch of samples Actual limit of detection, ppm 5 10 Coefficient of determination (r2), % 99.99 99.72 Determined DADS concentration in headspace of garlic extract, ppm 112 135 . Gas chromatography . SERS . Analyzing time 7 min for each sample 1 min for each sample Sample pretreatment Sample might need to be filtered for the protection of columns 5 min boiling water bath for a batch of samples Actual limit of detection, ppm 5 10 Coefficient of determination (r2), % 99.99 99.72 Determined DADS concentration in headspace of garlic extract, ppm 112 135 Open in new tab Table 1. Analytical behaviors comparison between GC and SERS . Gas chromatography . SERS . Analyzing time 7 min for each sample 1 min for each sample Sample pretreatment Sample might need to be filtered for the protection of columns 5 min boiling water bath for a batch of samples Actual limit of detection, ppm 5 10 Coefficient of determination (r2), % 99.99 99.72 Determined DADS concentration in headspace of garlic extract, ppm 112 135 . Gas chromatography . SERS . Analyzing time 7 min for each sample 1 min for each sample Sample pretreatment Sample might need to be filtered for the protection of columns 5 min boiling water bath for a batch of samples Actual limit of detection, ppm 5 10 Coefficient of determination (r2), % 99.99 99.72 Determined DADS concentration in headspace of garlic extract, ppm 112 135 Open in new tab Supplemental Information Supplemental information is available on the J. AOAC Int. website. JAOAC does not publish color figures in the print version. Color images are published online only. References 1 Kim S.M. , Wu C.M., Kobayashi A., Kubota K., Okumura J. ( 1995 ) J. Agric. Food Chem. 43 , 2951 – 2955 . doi:10.1021/jf00059a033 Crossref Search ADS 2 Lanzotti V. ( 2006 ) J. Chromatogr. A 1112 , 3 – 22 . doi:10.1016/j.chroma.2005.12.016 Crossref Search ADS PubMed 3 Amagase H. , Petesch B.L., Matsuura H., Kasuga S., Itakura Y. ( 2001 ) J. Nutr . 131 , 951s – 1123s Crossref Search ADS 4 Calvo-Gómez O. , Morales-López J., López M.G. ( 2004 ) J. Chromatogr. A 1036 , 91 – 93 . doi:10.1016/j.chroma.2004.02.072 Crossref Search ADS PubMed 5 Avato P. , Tursi F., Vitali C., Miccolis V., Candido V. ( 2000 ) Phytomedicine 7 , 239 – 243 . doi:10.1016/S0944-7113(00)80010-0 Crossref Search ADS PubMed 6 Nakagawa H. ( 2001 ) Carcinogenesis 22 , 891 – 897 . doi:10.1093/carcin/22.6.891 Crossref Search ADS PubMed 7 Yi L. , Su Q. ( 2013 ) Food Chem. Toxicol . 57 , 362 – 370 . doi:10.1016/j.fct.2013.04.001 Crossref Search ADS PubMed 8 Yang T. , Wang P., Guo H., He L. ( 2017 ) Ref. Modul. Food Sci . doi:10.1016/B978-0-08-100596-5.21090-1 9 Chen H. , Wang C., Zhang Z., He L. ( 2018 ) J. Food Prot . 81 , 1087 – 1092 . doi:10.4315/0362-028X.JFP-17-505 Crossref Search ADS PubMed 10 Deng Z. , Chen X., Wang Y., Fang E., Zhang Z., Chen X. ( 2015 ) Anal. Chem. 87 , 633 – 640 . doi:10.1021/ac503341g Crossref Search ADS PubMed 11 Li D. , Duan H., Ma Y., Deng W. ( 2018 ) Anal. Chem. 90 , 5719 – 5727 . doi:10.1021/acs.analchem.8b00016 Crossref Search ADS PubMed 12 Schulz H. , Baranska M., Belz H.H., Rösch P., Strehle M.A., Popp J. ( 2004 ) Vib. Spectrosc . 35 , 81 – 86 . doi:10.1016/j.vibspec.2003.12.014 Crossref Search ADS 13 Si M. , Li L., Chuanyun Z. ( 2016 ) J. Food Process Technol 6 , 2. doi:10.4172/2157-7110.1000520 14 Gao S. , Glasser J., He L. ( 2016 ) J. Vis. Exp . e53791 .doi:10.3791/53791 15 Doi T. , Wang M., McClements D.J. ( 2019 ) Food Res. Int . 116 , 1 – 11 . doi:10.1016/j.foodres.2018.12.056 Crossref Search ADS PubMed 16 Martins N. , Petropoulos S., Ferreira I.C.F.R. ( 2016 ) Food Chem . 211 , 41 – 50 . doi:10.1016/j.foodchem.2016.05.029 Crossref Search ADS PubMed 17 Wang H.C. , Pao J., Lin S.Y., Sheen L.Y. ( 2012 ) Ann. N. Y. Acad. Sci . 1271 , 44 – 52 . doi:10.1111/j.1749-6632.2012.06743.x Crossref Search ADS PubMed 18 Lee S.-N. , Kim N.-S., Lee D.-S. ( 2003 ) Anal. Bioanal. Chem . 377 , 749 – 756 . doi:10.1007/s00216-003-2163-z Crossref Search ADS PubMed 19 Brodnitz M.H. , Pascale J.V., Van Derslice L. ( 1971 ) J. Agric. Food Chem. 19 , 273 – 275 . doi:10.1021/jf60174a007 Crossref Search ADS 20 Springett M.B. , Rozier V., Bakker J. ( 1999 ) J. Agric. Food Chem. 47 , 1125 – 1131 . doi:10.1021/jf980199n Crossref Search ADS PubMed 21 Voice T.C. , Kolb B. ( 1993 ) Environ. Sci. Technol. 27 , 709 – 713 . doi:10.1021/es00041a014 Crossref Search ADS © AOAC INTERNATIONAL 2020. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Headspace Characterization and Quantification of Aromatic Organosulfur Compounds in Garlic Extracts Using Surface-Enhanced Raman Scattering with a Mirror-in-a-Cap Substrate
JF - Journal of AOAC INTERNATIONAL
DO - 10.1093/jaoacint/qsaa021
DA - 2020-09-01
UR - https://www.deepdyve.com/lp/oxford-university-press/headspace-characterization-and-quantification-of-aromatic-organosulfur-spfU0bG3XQ
SP - 1201
EP - 1207
VL - 103
IS - 5
DP - DeepDyve
ER -