Gold nanoparticle enriched by Q sepharose spheres for chemical reaction tandem SERS detection of malondialdehyde

Gold nanoparticle enriched by Q sepharose spheres for chemical reaction tandem SERS detection of malondialdehyde

Accepted Manuscript Title: Gold Nanoparticle Enriched by Q Sepharose Spheres for Chemical Reaction tandem SERS Detection of Malondialdehyde Authors: R...

NAN Sizes 0 Downloads 0 Views

Accepted Manuscript Title: Gold Nanoparticle Enriched by Q Sepharose Spheres for Chemical Reaction tandem SERS Detection of Malondialdehyde Authors: Ruxia Yan, Zewei Wang, Jie Zhou, Ruyu Gao, Shaowei Liao, Haifeng Yang, Feng Wang PII: DOI: Reference:

S0925-4005(18)31847-1 https://doi.org/10.1016/j.snb.2018.10.078 SNB 25512

To appear in:

Sensors and Actuators B

Received date: Revised date: Accepted date:

29-4-2018 13-9-2018 14-10-2018

Please cite this article as: Yan R, Wang Z, Zhou J, Gao R, Liao S, Yang H, Wang F, Gold Nanoparticle Enriched by Q Sepharose Spheres for Chemical Reaction tandem SERS Detection of Malondialdehyde, Sensors and amp; Actuators: B. Chemical (2018), https://doi.org/10.1016/j.snb.2018.10.078 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Gold Nanoparticle Enriched by Q Sepharose Spheres

SC RI PT

for Chemical Reaction tandem SERS Detection of

U

Malondialdehyde



M

A

N

Ruxia Yan†, Zewei Wang†, Jie Zhou†, Ruyu Gao†, Shaowei Liao†, Haifeng Yang†,*, Feng Wang†,*

The Education Ministry Key Lab of Resource Chemistry, Shanghai Key Laboratory of Rare

D

Earth Functional Materials, Shanghai Municipal Education Committee Key Laboratory of

TE

Molecular Imaging Probes and Sensors, and Department of Chemistry, Shanghai Normal

CC

EP

University, Shanghai, 200234, PR China

Corresponding Author

A

* E-mail: [email protected] (Dr, Wang, F.) Tel: +86-13402154463 * E-mail: [email protected] (Dr, Yang, H. F.)

Notes

1

The authors declare no competing financial interest.

EP

TE

D

M

A

N

U

SC RI PT

Graphical Abstract:

Highlights:

CC

 Gold nanoparticles enriched on the surface of Q sepharose spheres ([email protected]) is prepared.

A

 Chemical reaction tandem SERS method is used to detect trace malondialdehyde in real oil by [email protected]  QSS can adsorb AuNPs, be carrier against AuNPs aggregation in acid environment and products enricher.

2

 The stable SERS signals is expected to do on-site analysis of food quality.

Abstract

SC RI PT

Malondialdehyde (MDA) is a biomarker of lipid peroxidation and intracellular oxidative stress and can also be used to monitor food freshness. However, fast, accurate and ultrasensitive detection of MDA remains a challenge. In this work, a chemical reaction tandem surface-

enhanced Raman spectroscopy (SERS) method was applied for trace MDA detection by gold

nanoparticles (AuNPs) enriched on the surface of Q sepharose spheres ([email protected]). After

U

derivatization of MDA with 2-thiobarbituric acid (TBA), TBA-MDA adduct could be directly

N

detected by SERS in the reaction media utilizing [email protected] as SERS substrate. QSS play

A

multiple roles as AuNPs adsorbent, AuNPs carrier against aggregation in acid environment, and

M

TBA-MDA adduct enricher. SERS detection in solution and the micro-scale homogeneous

D

AuNPs distribution on QSS surface contribute to the improved reproducibility of the substrate.

TE

This method has a wide linear range (0.33 μM to 3.3 mM in a liquid environment), low limit of detection (~10-10 M at dry state) and has already been applied in the direct determination of

CC

EP

MDA in oil samples without standard addition.

A

Keywords: malondialdehyde; chemical reaction; SERS; Q sepharose spheres; oil samples

1. Introduction

3

Malondialdehyde (MDA) is a biomarker of lipid peroxidation and intracellular oxidative stress. In food, MDA is produced by the peroxidation of polyunsaturated fatty acids [1-4] as an indicator to monitor the degree of food rancidity. After absorbed from the gastrointestinal tract

SC RI PT

[5], it can react with DNA to form MDA-DNA adduct and cause DNA damage and mutations [6]. MDA is also produced in biological process of human being. MDA formed during

intracellular oxidative stress can lead to various diseases, including depression [7, 8], chronic heart failure [9], diagnosed gastric adenocarcinoma [10], complex regional pain syndrome,

glaucoma, hypertension, diabetes and atherosclerosis [11]. Therefore, it is important to develop a

U

sensitive and simple method to detect MDA in the food industry and clinical area.

N

To date, many methods have been explored for the detection of MDA. But due to its small

A

relative molecular mass and volatility, MDA is difficult to be detected directly by instrumental

M

methods. Nowadays, some reaction methods relying on the derivatization between MDA and

D

different reagents have been investigated to increase the analytical sensitivity, such as

TE

thiobarbituric acid (TBA) [12-14], 1-methyl-2-phenylindole [15], 2-aminoacridone [16], 9fluorenylmethoxycarbonyl hydrazine [17], 2,4-dinitrophenylhydrazine [18,19] and automated

EP

solid phase analytical derivatization [20]. MDA derivatives could be determined by using spectroscopic method, liquid chromatographic (LC) method, gas chromatographic (GC) method,

CC

or capillary electrophoretic method [21]. The limit of detection (LOD) for MDA in National

A

Standard is 0.03 mg/kg which is detected by HPLC [GB (5009.181-2016)]. S. Zelzer et al. [22] have reported that the LOD for total MDA and free MDA are 0.20 μmol/L and 5 nmol/L, respectively using gas chromatography-mass spectrometry (GC-MS) method. However, such methods are either time consuming, or could not meet the requirement for ultralow level or onsite MDA detection.

4

Surface-enhanced Raman spectroscopy (SERS) is a simple, fast and ultrasensitive technique. In principle, when analytes are adsorbed on the surface of noble metal (especially Au or Ag) nanostructures [23-25], the Raman signals of the analytes can be significantly enhanced by the

SC RI PT

strong electromagnetic field near the Au or Ag nanostructures. SERS spectrum provides abundant fingerprint information of molecular structures, which can differentiate compounds

with similar functional groups [26]. Furthermore, within nano-junctions or gaps (smaller than

~10 nm) of the noble nanoparticles / nanostructures, electromagnetic field can be significantly

elevated by such “hot spot”, which is beneficial for the detection of trace analytes. LOD can be

U

as low as single-molecule level in SERS detection [27]. SERS as a fast signal integration

N

technique needs a low sample consumption. However, analytes without obvious Raman signal,

M

SERS [28, 29], limiting its wide application.

A

analytes with volatility or low affinities towards noble metals are difficult to be detected by

D

Derivatization reactions can be applied to increase the affinity of analytes to the SERS

TE

substrates by introducing functional groups like -SH or -NH2 [30]. What’s more, some derivatization reactions also increase the Raman scattering cross-sectional area to obtain visible

EP

signals [31, 32]. Additionally, specific chemical reactions can further improve the selectivity of the Raman method. As an expectation, chemical-reaction-SERS method would expand the

CC

application fields of SERS analytical technique in a highly selective and sensitive way. As well

A

known, MDA molecule has small Raman scattering cross-sectional area and poor affinity to the substrates and is hard to be detected by SERS technology. Therefore, the chemical-reactionSERS method for detection of MDA attracts more research interests. Zhang et al. [33] treated MDA with 2-thiobarbituric acid (TBA) as the reaction reagents and then successfully performed

5

Ag-nanoparticle-based SERS measurement, but it presented a great relative standard deviation (RSD) up to ~20%. In this work, the chemical reaction between TBA and MDA was also adopted to transform the

SC RI PT

MDA into a Π-Π conjugated product under the acidic condition. A novel SERS substrate was designed and prepared by attaching Au nanoparticles onto the surface of Q sepharose spheres

(QSS). QSS play multiple roles as AuNPs adsorbent, stable AuNPs carrier against aggregation in acid environment, and efficient enricher for capturing TBA-MDA adduct. As the result, the [email protected] SERS detection can be performed in the acidic media without any

U

separation. [email protected] SERS has been successfully applied to determine of trace MDA

A

TE

2. Materials and Methods

D

M

promising clinical application in future.

N

in oil samples which is a simple, sensitive and selective method. The novel SERS substrate has a

EP

2.1 Chemicals and Materials.

1,1,3,3-tetraethoxypropane (TEP, 97%), 2-thiobarbituric acid (TBA, 98%), phosphoric acid

CC

(H3PO4, AR, ≥85%), glutaraldehyde (AR, 50%), formaldehyde (HCHO, AR, 37%~40%), benzaldehyde (AR, 98.5%), chloroauric acid (HAuCl4·4H2O, 99.9%) and rhodamine 6G (R6G,

A

AR) were purchased from Aladdin (Shanghai, China). Sulfuric acid (H2SO4, AR, 95%~98%), trichloroacetic acid (TCA, AR, ≥99%) and ethylenediaminetetraacetic acid disodium salt (EDTA, AR, ≥99%) was purchased from Richjoint Chemical (Shanghai, China). Trisodium citrate dihydrate (AR, >99%) were purchased from Sinopharm Chemical (Shanghai, China). Q

6

sepharose sphere was purchased from Yuanye Chemical (Shanghai, China). All reagents were used as received without further purification. All deionized water (18 MΩcm) used in this work was acquired by a Millipore Direct-Q system. All glassware was cleaned with aqua regia (HCl:

SC RI PT

HNO3 = 3:1) and then thoroughly rinsed with deionized water. 2.2 Preparation of AuNPs.

The 15 nm AuNPs were synthesized following previous protocol [34]. Briefly, 100 mL of 1.0 mM chloroauric acid was added to 100 mL of ultrapure water in an Erlenmeyer flask and then

U

heated to boiling under magnetic stirring. After that, 10 mL of 38.8 mM trisodium citrate was

N

quickly injected. The mixture solution was kept refluxing for 20 min and the solution color

A

changed from colorless to purple and finally to wine red. AuNPs colloid was finally obtained

M

until the solution cooled to room temperature with a final volume of 200 mL. UV-vis absorption

D

spectrum of AuNPs was shown in Figure S2.

TE

2.3 Preparation of TBA-MDA Adduct. Fresh MDA was synthesized by hydrolyzation of 1,1,3,3-tetraethoxypropane (TEP) under

EP

acidic environment [35]. Briefly, 20 mM TEP solution was hydrolyzed in 1% H2SO4 solution for

CC

2 h at room temperature. The MDA solution was diluted with deionized water to desirable concentrations for meeting the experimental need. The MDA solutions were stored at 4 °C and

A

were used within 2 days to avoid possible MDA dimerization (which can be observed by color changes from colorless to yellow) [36, 37]. 0.3025 g of TBA was dissolved in 50 mL of ultrapure water and heated at 55 °C for 45 min and was cooled to room temperature. All the TBA solutions used in this work were freshly prepared to avoid sample contamination. Reaction of TBA with MDA or other aldehydes was carried out [33]. Briefly, 500 μL of MDA, benzaldehyde,

7

glutaraldehyde or formaldehyde with proper concentrations was added into the mixture solution of 750 μL of 440 mM H3PO4 and 250 μL of 42 mM TBA, respectively. The resultant mixtures kept heating at ~90 °C for 1 h. After that, the mixture was cooled to room temperature and stored

SC RI PT

in dark. SERS analysis was performed in less than 2 days. 2.4 Preparation of Real Samples and SERS Detection of MDA.

Real samples were prepared according to the National standard of China [GB (5009.181-

2016)]. 37.5 g TCA and 0.5 g EDTA-2Na were dissolved in ultrapure water and diluted to 500

U

mL. 5 g lard, sunflower seed oil or blend oil was dispersed in 50 mL TCA-EDTA solutions

N

(7.5% TCA and 0.1% EDTA, v/v), respectively. Oil samples were placed in a thermostatic

A

shaker and kept shaking for 30 min at 50 °C, then cooled to room temperature and filtered with

M

double quantitative slow filter paper. The initial filtrate was discarded and following filtrate was collected. 5 mL 20 mM TBA was added into 5 mL of filtrate and reacted in 90 °C water bath for

D

30 min. After the chemical reaction between TBA and MDA, 100 μL of the reaction mixture was

TE

added into 300 μL of AuNPs colloid which had been enriched by Q sepharose spheres. The final

EP

mixtures enriched on Q sepharose spheres were sucked into the capillary glass tubes and then detected by SERS.

CC

2.5 Characterization.

A

Field emission scanning electron microscopy (FE-SEM, Hitachi S-4800) was applied to

characterize the morphology and size of the prepared samples and was operated at 5 kV. SERS spectra were acquired on a Jobin Yvon confocal laser Raman system (SuperLabRam II) with a lens of 50 × objective (8 mm) by using a He-Ne operating laser at 632.8 nm with a power of ca. 5 mW. Each spectrum was obtained by three accumulations, and the integration time in each

8

case was 7 s. We took more than 5 points per sample by detecting different QSS and repeated the experiment more than 3 times to get the RSD. Thermal gravity analysis (TGA) was conducted on SDT Q600 TGA (TA instrument). UV-vis spectra were collected by using a UV-6300PC double-

SC RI PT

beam spectrophotometer (VWR). LC experiment was conducted with an UV-detector at 530 nm (Essentia LC-16C). The analytical column was Shimadzu C18 column, maintained at 30 °C.

Mobile phase was 0.01 M ammonium acetate: methanol = 70: 30 with the flow rate setting as 1

U

mL/min.

A

M

3.1 Mechanism Statement of Detection.

N

3. Results and Discussion

As mentioned in the introduction section, MDA has no obvious signal in Raman test.

D

Therefore, chemical-reaction-SERS method is chosen to realize sensitive and selective

TE

determination of MDA as illustrated in Scheme 1 by the aid of [email protected] When reacting 1 mole MDA with 2 moles TBA (Figure S1), TBA-MDA adduct exhibits larger Raman scattering

EP

cross-sectional area than raw MDA. In our reaction system, TBA is always excessive which

CC

ensures complete MDA reaction and the estimated yield between TBA and MDA is about 98.7%. Besides, TBA-MDA adduct has two thione groups instead of one in TBA, so that the

A

TBA-MDA adduct can increase the binding capability onto Au substrates. Therefore, the derivatization reaction of MDA with TBA provides the detection possibility and QSS in SERS substrate can enrich negatively TBA-MDA adduct, which also enhance the sensitivity and stability of the signals recorded spot by spot in after mentioned SERS experiments. Strong cationic Q sepharose spheres bound negatively charged AuNPs evenly on their surfaces can be

9

visible in Figure 1. In addition, Q sepharose spheres show acid-proof, which promote the stability of the [email protected] under the harsh acid reaction environment. 3.2 Chemical Reaction Between TBA and MDA.

SC RI PT

The chemical reaction between TBA and MDA under acidic condition can produce fuchsia

colored TBA-MDA adduct. UV-vis absorbance of MDA and TBA are located at 245 nm [38]

and 264 nm respectively as seen in Figure 2A. After reaction, these two peaks almost disappear and a new peak at ~530 nm emerges (Figure 2B), meaning the successful synthesis of TBA-

U

MDA adduct [33]. The formation of TBA-MDA adduct can also be confirmed by SERS study.

N

AuNPs has an adsorption band of UV-vis at 518 nm (Figure S2). As shown in Figure S3, size

A

distribution of AuNPs is uniform and such AuNPs used for detecting R6G can be as low as 10-7

M

M (Figure S4).

D

In Figure 3d, SERS signals from MDA are weak, meaning low background for next detection

TE

of TBA-MDA adduct. SERS spectrum of 42 mM TBA in water (Figure 3c) has no effect on the SERS spectrum of TBA-MDA (Figure 3a), and the final concentration of TBA were kept at

EP

7mM in the reaction system which included 500 μL of MDA, 750 μL of H3PO4 and 250 μL of TBA. TBA-MDA adduct not only has more extensively π-conjugated electrons and larger

CC

Raman cross-sectional area both than TBA and MDA, but also has higher binding affinity to the AuNPs substrates due to its possessing the two thiol groups. TBA-MDA adduct can contribute

A

higher SERS activity than that of TBA and MDA (Figure 3). The assignments of typical peaks from TBA-MDA adduct are listed in Table S1. The characteristic peaks at 587, 1180, 1271, 1490 and 1554 cm-1 are from TBA-MDA. The SERS intensity at 1271cm-1 is picked out to do

10

quantitative analysis of MDA, which belongs to the vibration of δ C9-H + δ C10-H + γC9-C10 [33]. 3.3 Optimization Results.

SC RI PT

The derivatization reaction and analysis conditions such as pH, volume ratio and integration time were optimized in detail. pH value of the reaction system was adjusted by H3PO4 to

guarantee acidic condition for TBA-MDA reaction [39]. Figure 4A shows the relationship

between solution acidity and SERS intensity. SERS intensity at 1271cm-1 gradually increases

U

with reducing pH value and the maximum signal reaches at pH=2. However, when pH value is

N

lower than 2, excessive amount of H3PO4 might cause the aggregation of AuNPs and weaken the

A

SERS enhancement. While pH value is set greater than 2, it is not beneficial to promote the

M

derivatization reaction between the TBA and MDA. Eventually, pH =2 was selected as the

D

optimal acidity for the consequent experiments.

TE

As shown in Figure 4B, the volume ratio between TBA-MDA adduct and AuNPs was optimized for SERS detection from 2:1 to 1:4 by keeping the constant volume of adduct and

EP

clearly, the volume ratio directly affects the SERS signal. SERS intensity at 1271cm-1 increases with the ratio from 2:1 to 1:3 but a decline occurs when the amount of AuNPs is too much. The

CC

volume ratio of 1:3 might produce much more SERS “hot spot” and was selected for the further

A

experiments.

Different integration time ranging from 1 to 9 s was investigated in the study. SERS intensity

at 1271cm-1 from the TBA-MDA adduct increases an enough level for quantitative analysis as integration time set up to 7 s (Figure 4C and S5). The integration time of 7 s was fixed to conduct the following SERS experiments.

11

3.4 Selectivity of MDA Detection. Some possible interference compounds were chosen for the study of detection selectivity for MDA. The reactions of aldehydes with TBA were carried at 90 °C and room temperature, while

SC RI PT

the amount of TBA and H3PO4 were kept constant. It can be seen from Figure 5 that at room temperature other aldehydes have little influence except for glutaraldehyde. TBA-glutaraldehyde adduct fabricated at room temperature has relatively high Raman intensity, this might be due to the pyridine like moiety in TBA-glutaraldehyde which produces SERS signal at ~1271 cm-1 [40, 41]. Figure 5 also shows that, for all aldehydes, the SERS intensities of their adduct made at 90

U

°C are weaker than those at room temperature, especially for glutaraldehyde, the SERS signal of

N

TBA-glutaraldehyde fabricated at 90 °C is much lower than that at room temperature. The

A

possible reason is that with increasing temperature the self-polymerization rate of glutaraldehyde

M

rose exponentially [42]. This polymerization can not only decrease the available aldehyde groups

D

but also blocks the active sites (aldehyde groups) of the glutaraldehyde that are responsible for

TE

its bioactivity [43]. Therefore, high temperature was chosen as the reaction temperature for improving the selectivity of MDA SERS analysis. Representative SERS spectra of different

EP

aldehydes-TBA adduct together with raw aldehydes are shown in Figure S6. Consequently, the

CC

detection specificity of the TBA-MDA -SERS method could be validated. 3.5 [email protected] SERS Detection of MDA in Oil Samples.

A

We would like to analyze TBA-MDA adduct by using [email protected] in the capillary tube.

Many researchers use dried noble metal colloids as SERS substrate, but the “coffee ring effect” will cause significant inhomogeneity in nanoparticle distributions, leading to poor reproducibility of the SERS result. In reference 33, the authors used 1% KCl as aggregation agent. Though

12

SERS signal can be increased due to salt induced colloid aggregation, it’s hard to control the degree of aggregation. Meanwhile, the degree of aggregation varies with time, it is therefore not surprising that the SD (+/- 20%) is large.

SC RI PT

SERS detection in solution can avoid uncontrollable colloid aggregation, leading to better reproducibility. In order to increase SERS intensity and AuNPs stability, AuNPs were adsorbed on commercialized strong cationic QSS, the homogeneously distribution of adsorbed AuNPs on QSS surface over micro-scale range is proved by enlarged image of Figure 1B. The theoretical laser spot diameter in our case is calculated to be 1.22*lambda/NA = 1.22*632 / 0.5 = 1.5

U

micron, which is in the micro-scale range rather than nanoscale. Therefore, as long as the micro-

N

scale homogenous can be reached, the nanoscale inhomogeneous can be averaged and offset by

M

A

micro- scale homogeneity.

D

The value of mAu / mQSS was calculated by the following equation:

TE

π*RAu*ρAu / RQSS*ρQSS = π*7.6*10-3*19.3 / 45*1.81 = 0.00566 where ρQSS = 1.81 g/cm3 and ρAu = 19.32 g/cm3 are the densities for the QSS and AuNPs,

EP

respectively. RQSS = 45 μm (D50 is about 90 μm provided by the manufacture) and RAu = 7.6 nm

CC

(shown in Figure S2 and S3) are the radii for QSS and AuNPs, respectively. Thermal gravity analysis (TGA) result (Figure S7) showed a weight retain of 0.035 / 6.1= 0.00574, therefore the

A

coverage of AuNPs on QSS in dry state is calculated to be higher than 100%. Though this dry state value cannot be directly applied in wet state, a closely packed AuNPs layer on QSS can be imagined, which also confirmed the homogeneity of the adsorbed AuNPs. As above mentioned, QSS can adsorb the final adduct of TBA-MDA to increase Raman signal level in the liquid. Figure 6 shows relatively SERS mapping that [email protected] have captured

13

TBA-MDA adduct with RSD ~ ±10%. As AuNPs will only adsorb at the surface of the QSS and will not penetrate inside, strong SERS intensity can only be observed at the edge the crosssectional SERS image. QSS with a mean diameter about 90 μm has low background and no

SC RI PT

interference with TBA-MDA SERS spectra (Figure S8). As seen in Figure 7A, the color of the QSS changes from colorless to wine red and then to fuchsia with the addition of AuNPs and TBA-MDA adduct successively. It’s worth emphasizing that no AuNPs dissociation from [email protected] is observed at pH as low as 2. Though low pH values can dissociate the

electrostatic attraction between AuNPs and QSS, it can enhance the hydrogen bond between the

U

protective citric acid on AuNPs and the hydroxyl group on agarose, which prevent the AuNPs

N

dissociation or aggregation in harsh acid conditions. In addition, the color of [email protected] with

A

adduct adsorption shows no change after carefully rinsed by water, which depicts the strong

M

adsorption capability of [email protected] Optical microscopic images of [email protected] with and without TBA-MDA adduct in the capillary tube are taken as shown in Figure 7B, which saying

TE

D

the same story of tight adsorption ability of QSS. Photos of TBA-MDA adduct in which the adding concentrations of MDA ranging from 10-2 to

EP

10-10 M display the color varying from fuchsia to colorless (Figure 8C). Figure 8A is SERS spectra of TBA-MDA adduct with the final concentrations of MDA (500 μL) in the reaction

CC

system (750 μL of H3PO4 and 250 μL of TBA) ranging from 33 nM to 3.3 mM. A linear

A

relationship spans the concentrations from 0.33 μM to 3.3 mM, plotted by SERS intensity at 1271 cm-1 with correlation (R2=0.93) as shown in the inset of Figure 8B. The enhancement factor (EF) was calculated in accordance with the Raman intensity at 1271 cm−1 by the following equation [44]: EF=ISERSNbulk/IbulkNSERS

14

where ISERS and Ibulk are the Raman intensities for the SERS and the normal Raman, respectively. Nbulk and NSERS are the number of molecules under laser illumination for bulk sample, and the number of molecules in the substrates, respectively. The EF of our SERS substrate could reach

SC RI PT

about 2*105 in strong acid environment. The limit of detection could be reached down to ~10-10 M (Figure S9) as SERS signal is acquired from the dried mixture of TBA-MDA and

[email protected] in ambient condition. What’s more, this SERS method can be employed to

directly detect TBA-MDA in the reaction solution without any separation steps, which is much preferable for on-field detection.

U

[email protected] was applied in the quantification of MDA in different edible oils such as

N

lard, sunflower seed oil and blend oil which had been lipid peroxidation to monitor the degree of

A

food rancidity [45]. As shown in Table 1 and in Figure 9, MDA is clearly found in all oil

M

samples by this reaction and [email protected] method, and the concentrations of MDA in

D

those oils locate from 12.3 to 52.6 μM. For detecting the spiked oil samples, the recoveries are

TE

found to be 89.25-110.25% with RSDs of 7.81-12.64% (n = 5). As comparison with LC results given in Figures S10 and S11, and as clearly seen in Table 2, this chemical-reaction-SERS

EP

method by using [email protected] to detect trace MDA in oil samples is reliable with good

CC

reproducibility.

4. Conclusions

A

In summary, chemical reaction tandem SERS method to determine trace MDA was realized

through constructing novel SERS substrate by AuNPs enriched on the surface of Q sepharose spheres. TBA-MDA adduct exhibits larger Raman scattering cross-sectional area than raw MDA and has stronger binding capability onto SERS substrates both than MDA and TBA.

15

[email protected] could directly detect active TBA-MDA adduct in the reaction media with high sensitivity. QSS can not only adsorb AuNPs and enrich TBA-MDA adduct, but can also improve the stability of the adsorbed AuNPs and maintain their homogeneous

SC RI PT

distribution on QSS surface in acid environments. Such [email protected] method had a linear quantitative range from 0.33 μM to 3.3 mM in a liquid environment and a detection limit of ~10-10 M could be achieved, which were applied in detection of MDA in real oil samples

successfully. This method is reliable, simple, cost effective and convenient, which could be

U

expected to do on-field analysis of food quality by the help with portable Raman system.

N

Acknowledgements

A

This work is supported by the National Natural Science Foundation of China (No.21475088),

M

International Joint Laboratory on Resource Chemistry (IJLRC), Shanghai Key Laboratory of

D

Rare Earth Functional Materials, and Shanghai Municipal Education Committee Key Laboratory

EP

TE

of Molecular Imaging Probes and Sensors.

Authour Biographies

CC

Ruxia Yan, Master candidate in Shanghai Normal University

A

Zewei Wang, Master candidate in Shanghai Normal University Jie Zhou, Master candidate in Shanghai Normal University Ruyu Gao, Bachelor candidate in Shanghai Normal University Shaowei Liao, mMaster candidate in Shanghai Normal University

16

Haifeng Yang, Professor, Shanghai Normal University

SC RI PT

Feng Wang, Associate Professor, Shanghai Normal University

REFERENCES

[1] K.S. Rhee, Minimization of further lipid peroxidation in distillation 2- thiobarbituric acid test

U

of fish and meat, J. Food Sci. 43 (1978) 1776-1778.

N

[2] P. Bergamo, E. Fedele, M. Balestrieri, P. Abrescia, L. Ferrara, Measurement of

A

malondialdehyde levels in food by highperformance liquid chromatography with fluorometric

M

detection, J. Agr. Food Chem. 46 (1998) 2171-2176.

D

[3] R. Guillén-Sans, M. Guzmán-Chozas, Aldehydes in Food and its Relation with the TBA Test

TE

for Rancidity, Eur. J. Lipid Sci. Technol. 97 (2010) 285-286. [4] R. Mateos, E. Lecumberri, S. Ramos, L. Goya, L. Bravo, Determination of malondialdehyde

EP

(MDA) by high-performance liquid chromatography in serum and liver as a biomarker for

CC

oxidative stress: application to a rat model for hypercholesterolemia and evaluation of the effect of diets rich in phenolic antioxidants from fruits, J. Chromatgr. B. 827 (2005) 76-82.

A

[5] J.S. Cohn, Oxidized fat in the diet, postprandial lipaemia and cardiovascular disease, Curr. Opin. Lipidol. 13 (2002) 19-24. [6] Y. Zhang, S.Y. Chen, T. Hsu, R.M. Santella, Immunohistochemical detection of malondialdehyde-DNA adducts in human oral mucosa cells, Carcinogenesis. 23 (2002) 207-211.

17

[7] V.O. Kotan, E. Sarandol, E. Kirhan, G. Ozkaya, S. Kirli, Effects of long-term antidepressant treatment on oxidative status in major depressive disorder: a 24-week follow-up study, Prog. Neuro-Psychopharmacol. Biol. Psychiatry. 35 (2011) 1284-1290.

recurrent depression, J. Affective Disord. 143 (2012) 34-38.

SC RI PT

[8] C. Stefanescu, A. Ciobica, The relevance of oxidative stress status in first episode and

[9] C.A. Michalakeas, J.T. Parissis, A. Douzenis, M. Nikolaou, C. Varounis, I. Andreadou, N. Antonellos, S. Markantonis-Kiroudis, I. Paraskevaidis, I. Ikonomidis, E. Lykouras, D.

U

Kremastinos, Effects of sertraline on circulating markers of oxidative stress in depressed patients

N

with chronic heart failure: a pilot study, J. Card. Failure. 17 (2011) 748-754.

A

[10] Y.C. Wei, F.L. Zhou, D.L. He, J.R. Bai, H. Ding, X.Y. Wang, K.J. Nan, Oxidative stress in

M

depressive patients with gastric adenocarcinoma, Int. J. Neuropsychopharmacol. 12 (2009) 1089-

D

1096.

TE

[11] Z. Singh, I.P. Karthigesu, P. Singh, R. Kaur, Use of malondialdehyde as a biomarker for assessing oxidative stress in different disease pathologies: a review, Iran. J. Public Health. 43

EP

(2014) 7-16.

CC

[12] R. Mateos, E. Lecumberri, S. Ramos, L. Goya, L. Bravo, Determination of malondialdehyde (MDA) by high-performance liquid chromatography in serum and liver as a biomarker for

A

oxidative stress: application to a rat model for hypercholesterolemia and evaluation of the effect of diets rich in phenolic antioxidants from fruits, J. Chromatgr. B. 827 (2005) 76-82.

18

[13] M. Kaykhaii, H. Yahyavi, M. Hashemi, A simple graphene-based pipette tip solid-phase extraction of malondialdehyde from human plasma and its determination by spectrofluorometry, Anal. Bioanal. Chem. 408 (2016) 4907-4915.

SC RI PT

[14] T.P.A. Devasagayam, K.K. Boloor, T. Ramasarma, Methods for estimating lipid peroxidation: an analysis of merits and demerits, Indian J. Biochem. Biophys. 40 (2003) 30 0-308.

[15] I. Erdelmeier, D. Gérardmonnier, J.C. Yadan, Reactions of N-methyl-2-phenylindole with

N

peroxidation, Chem. Res. Toxicol. 11 (1998) 1176-1183.

U

malondialdehyde and 4-hydroxyalkenals. Mechanistic aspects of the colorimetric assay of lipid

A

[16] M. Giera, D.P. Kloos, A. Raaphorst, Mild and selective labeling of malondialdehyde with 2-

M

aminoacridone: assessment of urinary malondialdehyde levels, Analyst. 136 (2011) 2763-2769.

D

[17] J. Mao, H. Zhang, J. Luo, New method for HPLC separation and fluorescence detection of

EP

832 (2006) 103-108.

TE

malonaldehyde in normal human plasma, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci.

[18] O. Korchazhkina, C. Exley, S.A. Spencer, Measurement by reversed-phase high-

CC

performance liquid chromatography of malondialdehyde in normal human urine following

A

derivatisation with 2, 4-dinitrophenylhydrazine, J. Chromatogr. B. 794 (2003) 353-362. [19] A.S. Sim, C. Salonikas, D. Naidoo, Improved method for plasma malondialdehyde

measurement by high-performance liquid chromatography using methyl malondialdehyde as an internal standard, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 785 (2003) 337-344.

19

[20] H.L. Lord, J. Rosenfeld, V. Volovich, Determination of malondialdehyde in human plasma by fully automated solid phase analytical derivatization, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 877 (2009) 1292-1298.

SC RI PT

[21] A. Jouyban, K. Ansarin, M. Khoubnasabjafari, Critical Review of Malondialdehyde Analysis in Biological Samples, Curr. Pharm. Anal. 12 (2016) 4-17.

[22] S. Zelzer, R. Oberreither, C. Bernecker, Measurement of total and free malondialdehyde by gas–chromatography mass spectrometry-comparison with high-performance liquid

U

chromatography methology, Free Radical Res. 47 (2013) 651-656.

N

[23] Y. Yang, J. Liu, Z.W. Fu, D. Qin, Galvanic Replacement-Free Deposition of Au on Ag for

A

Core−Shell Nanocubes with Enhanced Chemical Stability and SERS Activity, J. Am. Chem.

M

Soc. 136 (2014) 8153-8156.

[24] T. Bai, J. Sun, R. Che, L. Xu, C. Yin, Z. Guo, N. Gu, Controllable Preparation of

D

Core−Shell Au−Ag Nanoshuttles with Improved Refractive Index Sensitivity and SERS

TE

Activity, ACS Appl. Mater. Interfaces. 6 (2014) 3331-3340.

EP

[25] T. Yang, X. Guo, Y. Wu, Facile and label-free detection of lung cancer biomarker in urine by magnetically assisted surface-enhanced Raman scattering, ACS Appl. Mater. Interfaces. 6

CC

(2014) 20985-20993.

A

[26] D. Sun, G. Qi, F. Cao, A recyclable silver ions-specific surface-enhanced Raman scattering (SERS) sensor, Talanta. 171 (2017) 159-165.

20

[27] X.M. Qian, S.M. Nie, ChemInform Abstract: Single-Molecule and Single-Nanoparticle SERS: From Fundamental Mechanisms to Biomedical Applications, Chem. Soc. Rev. 37 (2008) 912-920.

SC RI PT

[28] X. Gu, J.P. Camden, Surface-Enhanced Raman Spectroscopy-Based Approach for Ultrasensitive and Selective Detection of Hydrazine, Anal. Chem. 87 (2015) 6460-6464.

[29] Z. Zhang, Y. Zhan, Y. Huang, Large-volume constant-concentration sampling technique

coupling with surface-enhanced Raman spectroscopy for rapid on-site gas analysis, Spectrochim.

U

Acta, Part A. 183 (2017) 312-318.

N

[30] F. Wang, S. Cao, R. Yan, Z. Wang, Selectivity/Specificity Improvement Strategies in

A

Surface-Enhanced Raman Spectroscopy Analysis, Sensors, 17 (2017) 2689.

M

[31] Z. Zhang, C. Zhao, Y. Ma, Rapid analysis of trace volatile formaldehyde in aquatic products

D

by derivatization reaction-based surface enhanced Raman spectroscopy, Analyst. 139 (2014)

TE

3614-3621.

[32] Y. Zheng, Z. Chen, C. Zheng, Derivatization reaction-based surface-enhanced Raman

EP

scattering (SERS) for detection of trace acetone, Talanta. 155 (2016) 87-93.

CC

[33] D. Zhang, R. Haputhanthri, S.M. Ansar, Ultrasensitive detection of malondialdehyde with

A

surface-enhanced Raman spectroscopy, Anal. Bioanal. Chem. 398 (2010) 3193-3201. [34] Y.R. Zhang, Y.Z. Xu, Y. Xia, A novel strategy to assemble colloidal gold nanoparticles at the water-air interface by the vapor of formic acid, J. Colloid Interface Sci. 359 (2011) 536-541.

21

[35] N. Candon, N. Tuzmen, Very rapid quantification of malondialdehyde (MDA) in rat brain exposed to lead aluminium and phenolic antioxidants by high performance liquid chromatographyfluorescence detection, Neurotoxicology. 29 (2008) 708-713.

SC RI PT

[36] I.J. Jeon, W.G. Ikins, Analyzing food for nutrition labeling and hazardous contaminants, Food Sci. Technol. 28 (1994) 29.

[37] R. Guillén-Sans, I.M. Vicario, M. Guzmán-Chozas, Further studies and observations on 2thiobarbituric acid assay (fat autoxidtion) and 2-thiobarbituric acid-aldehyde reactions, Mol.

U

Nutr. Food Res. 41 (2010) 162-166.

N

[38] L.L. De Zwart, J. Venhorst, M. Groot, J.N. Commandeur, R.C. Hermanns, J.H. Meerman,

A

B.L. Van Baar, N.P. Vermeulen, Simultaneous determination of eight lipid peroxidation

M

degradation products in urine of rats treated with carbon tetrachloride using gas chromatography

D

with electron-capture detection, J. Chromatogr. B: Biomed. Sci. Appl. 694 (1997) 277-287.

TE

[39] D.D. Rio, A.J. Stewart, N. Pellegrini, A review of recent studies on malondialdehyde as toxic molecule and biological marker of oxidative stress, Nutr. Metab. Cardiovasc. Dis. Nmcd.

EP

15 (2005) 316-328.

CC

[40] P.M. Hardy, G.J. Hughes, H.N. Rydon, Formation of quaternary pyridinium compounds by the action of glutaraldehyde on proteins, J. Chem. Soc., Chem. Commun. 5 (1976) 157-158.

A

[41] R.L. Gieseking, M.A. Ratner, G.C. Schatz, Theoretical modeling of voltage effects and the chemical mechanism in surface-enhanced Raman scattering, Faraday Discuss. 205 (2017) 149171.

22

[42] K.E. Rasmussen, J. Albrechtsen, Glutaraldehyde. The influence of pH, temperature, and buffering on the polymerization rate, Histochem., 38 (1974) 19-26. [43] J.A. Kiernan, Formaldehyde, formalin, paraformaldehyde and glutaraldehyde: What they

SC RI PT

are and what they do, Microscopy Today 00-1 (2000) 8-12. [44] X.M. Lin, Y. Cui, Y.H. Xu, Surface-enhanced Raman spectroscopy: substrate-related issues, Anal. Bioanal. Chem. 394 (2009) 1729-1745.

[45] M. Viau, C. Genot, L. Ribourg, A. Meynier, Amounts of the reactive aldehydes,

malonaldehyde, 4-hydroxy-2-hexenal, and 4-hydroxy-2-nonenal in fresh and oxidized edible oils

U

do not necessary reflect their peroxide and anisidine values, Eur. J. Lipid Sci. Technol. 118

A

CC

EP

TE

D

M

A

N

(2016) 435-444.

23

SC RI PT

A

M

A

N

U

B

D

Figure 1. SEM images of the SERS substrates. A: QSS and its partial enlargement. B:

A

CC

EP

TE

[email protected] which AuNPs adsorbed on the surface of QSS and its partial enlargement.

24

0.8

0.8

0.6

b 0.4

0.4

0.2

0.2

0.0 200

0.6

250

300

350

0.0 350

400

400

SC RI PT

Absorbance (a.u.)

Absorbance (a.u.)

B

A

a

450

500

550

600

650

700

Wavelength (nm)

Wavelength (nm)

Figure 2. UV-vis spectra of the two raw materials and products. A: MDA (a) and TBA (b). B:

A

CC

EP

TE

D

M

A

N

U

TBA-MDA adduct.

25

SC RI PT

Raman Intensity (cps)

1271 1180

587

1490 1554

400

600

800

A

N

U

a b c d e

1000

1200

1400

1600

1800

2000

D

M

Raman Shift (cm-1)

TE

Figure 3. SERS spectra of TBA-MDA (a), TBA (c) and MDA (d) by using AuNPs as SERS

A

CC

EP

substrate. Raman spectra of AuNPs (e) and TBA-MDA (b) obtained as control.

26

1.5

2.0

2.5

3.0

3.5

4.0

(cps)

-1 2:1

pH

1:1

1:2

1:3

1:4

ratio

SC RI PT

1.0

C

Raman Intensity at 1271cm

-1

(cps)

1271 cm-1

B

Raman Intensity at 1271cm

Raman Intensity at 1271cm

-1

(cps)

1271 cm-1

A

0

1

2

3

4

5

6

7

8

9

10

time (s)

Figure 4. Optimization of derivatization reaction and analysis conditions between TBA with

MDA at 90 °C including A: effect of pH on SERS intensity at 1271 cm-1; B: effect of the volume ratio between derivatization product and AuNPs on SERS intensity at 1271 cm-1; C: effect of the

U

integration time on SERS intensity at 1271 cm-1. Error bars represent the standard deviations of

A

CC

EP

TE

D

M

A

N

five measurements.

27

SC RI PT U N A M

D

Figure 5. SERS intensities at 1271 cm-1 recorded from adduct of MDA, glutaraldehyde,

TE

formaldehyde and benzaldehyde reacted with TBA at 90 ℃ for 1h. The reaction temperatures are

A

CC

EP

90 °C and room temperature (r.t.), respectively.

28

b

c

SC RI PT

a

U

Figure 6. SERS mapping of cross-sectional TBA-MDA on the substrate of [email protected] (a)

N

SERS intensity image of TBA-MDA on the substrate of [email protected], (c) optical image of

A

TBA-MDA on the substrate of [email protected] SERS substrate, and (b) is the merge image of (a)

A

CC

EP

TE

D

M

and (c).

29

A

c

a

b

QSS

QSS+AuNPs

B

e

d

a

c+TBA-MDA

b fully adsorbed

d fully adsorbed

SC RI PT

b

Figure 7. A: The visible observation of color changes during adding AuNPs and TBA-MDA

adduct into the system, successively: (a) Q sepharose spheres, (b) adding AuNPs into (a), taken

U

at early stage, (c) fully adsorbing AuNPs onto QSS, (d) adding TBA-MDA adduct into

N

[email protected] solution as seen in (c) and (e) fully adsorbing TBA-MDA adduct by

A

[email protected] B: The image of (a) Q sepharose spheres, (b) TBA-MDA adduct adsorbed on

A

CC

EP

TE

D

M

[email protected] by the microscopy of Raman system.

30

B

8000

1271cm-1

8000

7000

7000

6000 5000

6000

y=1260.44x+8589.13 R2=0.93255

5000 4000 3000 2000 1000

4000

0 -7

-6

3000

-5

-4

lg c

2000 1000

1100

1200

1300

1400

0

Raman Shift (cm-1)

-8

-7

-6

SC RI PT

Raman Intensity (cps)

33 nM

Raman Intensity (cps)

3.3 mM

Intensity (a.u.)

A

-3

-2

-5

-4

-3

-2

lg c

10-3

10-4

10-5

10-6

10-7

10-8

10-9

10-10

N

U

C 10-2

A

Figure 8. (A) SERS spectra of TBA reacted with different concentrations of MDA ranging from

M

0.1 μM to 10 mM at 90 ℃ for 1h. The final concentrations of MDA (500 μL) in the reaction system (750 μL of H3PO4 and 250 μL of TBA) ranging from 33 nM to 3.3 mM. (B) The linear

D

relationship plotted by intensity of TBA-MDA at 1271 cm-1, and (C) the color of TBA-MDA

TE

adduct changing from fuchsia to colorless with different concentrations of MDA ranging from

A

CC

EP

10-2 M to 10-10 M. Error bars represent the standard deviations of five measurements.

31

SC RI PT

Raman Intensity (cps)

a b

600

800

1000

1200

1400

1600

d e f g

1800

A

400

N

U

c

M

Raman Shift (cm-1)

D

Figure 9. SERS spectra of MDA derivatives by TBA in real samples with [email protected] SERS

TE

substrates at 90 ℃, (a) MDA standard solution, (b) lard, (c) sunflower seed oil, (d) blend oil, and

EP

SERS spectra of raw materials with [email protected] SERS substrates, (e) raw lard, (f) raw

A

CC

sunflower seed oil, (g) raw blend oil.

32

malondialdehyde

TBA-MDA adduct

SC RI PT

2-thiobarbituric acid

TBA-MDA adduct

Q Sepharose sphere

D

M

A

N

U

gold nanoparticle

A

CC

EP

TE

Scheme 1. Schematic illustration of the process for detection of MDA.

33

Table 1. Real sample detection and recovery test of MDA in different oil samples by spiking different concentrations of MDA ranging from 4 μM to 12 μM in 10 mL reaction system.

9.81

98.88

12.64

110.25

9.37

14.46

4

9.04

8

SC RI PT

12

U

9.14

94.50

7.81

12.64

92.25

8.16

17.56

102.50

9.61

8.63

93.75

8.68

8

12.02

89.25

7.98

12

16.27

94.92

9.00

TE

4

M

A

CC

EP

90.50

8

12

4.88 (48.8)

(n=5)

4.85

seed oil

blend oil

(%)

4

sunflower 5.26 (52.6)

(μM)

Added (μM)

N

1.23 (12.3)

lard

RSD (%)

A

Original (μM)

Recovery

D

Targets

Found

34

Table 2. Comparison SERS method with LC method for detection of MDA spiked 80 μM to 160 μM. The final concentrations of MDA (500 μL) in the reaction system (750 μL of H3PO4 and 250 μL of TBA) ranging from 26.67 μM to 53.33 μM. LC

Added (μM)

Recovery

RSD

Found

Recovery

(%)

(%)

(%)

(%)

11.04

28.48

106.79

U

Sample

SC RI PT

SERS

6.54

39.37

98.43

8.32

49.87

93.52

Found (μM)

26.67 (80)

26.24

98.39

2

40 (120)

40.34

100.85

3

53.33 (160)

54.63

102.44

A

CC

EP

TE

D

M

A

N

1

35

Supporting Information Gold Nanoparticle Enriched by Q Sepharose Spheres for Chemical

SC RI PT

Reaction tandem SERS Detection of Malondialdehyde Ruxia Yan†, Zewei Wang†, Jie Zhou†, Ruyu Gao†, Shaowei Liao†, Haifeng Yang†,*, Feng Wang†,* †

The Education Ministry Key Lab of Resource Chemistry, Shanghai Key Laboratory of Rare

Earth Functional Materials, Shanghai Municipal Education Committee Key Laboratory of

U

Molecular Imaging Probes and Sensors, and Department of Chemistry, Shanghai Normal

EP

TE

D

M

A

N

University, Shanghai, 200234, PR China

CC

Figure S1. Chemical derivatization reaction between 2-thiobarbituric acid and malondialdehyde

A

to form TBA-MDA adduct.

36

0.6

Au NPs

0.4

0.3

0.2

0.1

0.0 400

450

500

550

600

N

A

CC

EP

TE

D

M

A

Figure S2. UV-vis spectrum of AuNPs colloid.

U

Wavelength (nm)

650

SC RI PT

Absorbance (a.u.)

0.5

37

SC RI PT

A

CC

EP

TE

D

M

A

N

U

Figure S3. Size distribution of AuNPs by number.

38

SC RI PT U N A

M

Figure S4. SERS spectra of R6G on AuNPs with different concentrations (ranging from 10-5 M

A

CC

EP

TE

D

to 10-7 M) and blank (blue).

39

Table S1. The vibrational frequencies of TBA-MDA adduct and assignments of the bands [33]. SERS (cm-1)

Assignment

587

νC=O+νC=O

1180

δN-H+δO-H(ring A)+δN-H(ring B)

1271

δC9-H+δC10-H+νC9-C10

1490

νC-OH(ring B)+δC8-H+δC9-H+δC10-H

1554

νOHC=NCS+δO-H(ring B)+δN-H(ring A)

SC RI PT

TBA-MDA

A

CC

EP

TE

D

M

A

N

U

i: ν-vibration, δ-in plane bending.

40

SC RI PT U N

A

Figure S5. SERS spectra of TBA-MDA adduct with different integration time (ranging from 1 s

A

CC

EP

TE

D

M

to 9 s).

41

SC RI PT U N

A

Figure S6. SERS spectra of different aldehydes-TBA adduct prepared at 90 °C by using AuNPs,

A

CC

EP

TE

D

benzaldehyde and glutaraldehyde.

M

involving TBA-MDA, TBA-HCHO, TBA-benzaldehyde, TBA-glutaraldehyde, HCHO,

42

100

SC RI PT

Weight retain(%)

80

60

40

U

20

300

400

500

A

200

N

0

600

700

800

M

Temperature (℃ )

A

CC

EP

TE

D

Figure S7. Thermal gravity curve of [email protected]

43

2000 1800

1400 1200 1000 800

a

600 400

b

200 0 600

800

1000

1200

1400

1600

1800

U

400

SC RI PT

Raman Intensity (cps)

1600

N

Raman Shift (cm-1)

A

Figure S8. The SERS spectrum of [email protected] SERS detection of TBA-MDA (a) and

A

CC

EP

TE

D

M

the Raman spectrum of raw Q sepharose spheres (b).

44

400

300

SC RI PT

Raman Intensity (cps)

500

200

100

0 400

600

800

1000

1200

1400

1600

1800

U

Raman Shift (cm-1)

N

Figure S9. Evaluation of limit of detection of TBA-MDA adduct by [email protected] in dry

A

state. The final concentration of MDA (500 μL) is 0.033 nM in the reaction system (750 μL of

A

CC

EP

TE

D

M

H3PO4 and 250 μL of TBA).

45

6000000

y=173524.99x+12594.20 R2=0.9998

SC RI PT

Area (mV. min)

5000000

4000000

3000000

2000000

0 0

5

10

15

20

N

U

1000000

25

30

35

M

A

MDA final concentration (μM)

Figure S10. The linear relationship of TBA-MDA adduct with concentrations of MDA ranging

D

from 5 to 100 μM. The final concentrations of MDA (500 μL) in the reaction system (750 μL of

A

CC

EP

TE

H3PO4 and 250 μL of TBA) ranging from 1.67 to 33 μM, detected by the LC method.

46

SC RI PT

Figure S11. The chromatogram of the standard TBA-MDA adduct. Mobile phase: 0.01 mol/L

U

ammonium acetate: methanol =70∶30 (v/v), detection wavelength: 530 nm, flow rate: 1.0

A

CC

EP

TE

D

M

A

N

mL/min, column temperature: 30 °C.

47