TRƯ ỜNG ĐẠI HỌC S Ư PH ẠM TP HỒ CHÍ MINH T ẠP CHÍ KHOA HỌC HO CHI MINH CITY UNIVERSITY OF EDUCATION JOURNAL OF SCIENCE ISSN: 1859 - 3100 KHOA H ỌC TỰ NHI ÊN VÀ CÔNG NGH Ệ T ập 16 , S ố 6 ( 2019): 16 - 28 NATURAL SCIENCE S AND TECHNOLOGY Vol 16 , No 6 ( 2019 ) : 16 - 28 Email: tapchikhoahoc@hcmue edu vn ; Website: http://tckh hcmue edu vn 16 DEVELOPMENT AND VALIDATION OF HPLC - PDA METHOD FOR THE SIMULTANEOUS DETERMINATION OF AURAMINE O AND RHODAMINE B IN FOODSTUFFS Phan Thi Ngoc Trinh 1 , Nguyen Thanh Thoi 1 , Huynh Thi Nhan 1 , Nguyen Ngoc Hung 1 , Nguyen Thi Tuyet Nhung 1 * , Le Thi Hong Van 2 , Le Van Huan 2 1 Faculty of chemistry – Ho Chi Minh City University of Education 2 Saigon Pharmaceutical Sci ences and Technologies Center, University of Medicine and Pharmacy at Ho Chi Minh City * Corresponding author: Nguyen Thi Tuyet Nhung – Email: nhungntt@hcmue edu vn Received: 18 / 3 /201 9; Revised: 11 / 4 /201 9; Accepted: 15 / 6 /2019 ABSTRACT HPLC - PDA method was used to determine two basic dyes, Auramine O and Rhodamine B in food samples The limit s of detection and quantification of Auramine O (AO) and Rhodamine B (RB) were both 0 02 mg/L and 0 05 mg/L, respectively The recoveries of AO and RB in matrices ranged from 83 00% to 105 91% with the c oncentration 0 5 μg/g, 1 0 μg/g and 1 5 μg/g This is a simple and accurate method which can be applied to quantify of dyes in foodstuffs Keywords: Auramine O, Rhodamine B, HPLC - PDA, foodstuffs 1 Introduction Auramine O (C 17 H 22 N 3 Cl, 303 84 gram/mol, yellow powder, pK a 9 8, 10 7 ( Sabnis R W, 2010) ) and Rhodamine B ( C 28 H 31 N 2 O 3 Cl, 479 01 gram/mol, green powder, pK a 3 7 (Peng Wang, Mingming Cheng, Zhonghai Zhang, 2014) ) are synthetic azo – cationic dyes with the chemical structures shown in Fig 1 RB f unctions as a water tracer fluorescent and is commonly used as a colorant in textiles, paper industry ( Adejumoke A Inyinbor, Folahan A Adekolab and Gabriel A Olatunji, 2015 ) while AO is commonly used to dye wood products, leather, ink , and textile fabrics ( Sabnis R W, 2010 ) A ccording to the International Agency of Research on Cancer, RB is associated with eye, skin , and respiratory tract irritation and AO is possibly carcino genic to humans ( Agents classified by the IARC Monographs, 2019 ) For that reason , the presence of AO and RB in foodstuffs may cause adverse effects on human’s health Ingestion of these dyes over time causes respiratory tract ailments as well as acute and chronic poisoning Therefore, both of them are not listed as food additives allowed for use in Vietnam and other countries ( Vietnamese Ministry of Health, 2001, 2018 ) However, because of their easy coloring, high stability , and low cost, AO and RB are still being used illegally in animal feed, food , and cosmetics as beauty color as well as to ensure the uniform appearance of foodstuffs T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM Phan Thi Ngoc Trinh et al 17 AO RB Fig 1 Chemical structures of AO and RB The color products have received the attention of many researchers in many countries in the world The study of Dong - yang Chen, Hao Zhang, Jia - li Feng, Dong Zeng, Li Ding, Xian - jun Liu, Bang - rui Li (2017) has presented high performance liquid chromatograph y coupled with diode array detector to detect 10 industrial dyes (basic orange 2, basic orange 21, basic orange 22, acid orange II, auramine, basic rhodamine B , and Sudan I – IV) in a great variety of matrices such as meat products (smoked sausages, preserve d ham, fresh meat, yellow - fin tuna), chili powder , and chili oil in China While the limit of detection has been in the range of 0 007 – 0 01 mg/kg, high recoveries (80 6 – 104%) and good reproducibility (1 1 – 5 7%) have been obtained Shruti Singh, Himani Shah , Ritika Shah , and Krishna Shah (2017) have separated and determined S udan and Rhodamine B dye in c hili and curry powder matrices by thin layer chromatography and spectrophotometry method Brazeau J (2018) has detected 27 color compounds (including RB) in food by liquid chromatography/ultraviolet visible Using poly(sodium 4 – styrenesulfonate) modified MIL – 101(Cr) – NH 2 to enable effective adsorption and separation of AO, RB , and pararosaniline from foodstuffs prior to high performance liquid chromatography a nalysis, under optimized conditions, the recoveries of the three dyes in shrimp powder, chili powder, tofu sheets, and tomato sauce have been in the range of 86 8 – 119 3% ( Xue Wang, Hui - Ling Duan, Shi - Yao Ma, Jun Wang, Han - Ying Zhan , & Zhi - Qi Zhang, 2018) Serum samples have been deproteinized with acetonitrile and separated by UHPLC on a reverse - phase C18, using a triple quadrupole tandem mass spectrometer in the selected reaction monitoring mode at [M] + ion m/z 443 39 → 399 28 for R B Linear has resulted over a concentration range of 0 5 – 100 ng/mL, with a lower limit of quantitation of 0 5 ng/mL ( Yung - Yi Cheng, Tung - Hu Tsai, 2016) In Vietnam, some analytical studies have determined only AO or RB in matrices, such as cosmetics ( Le Th i Huong Hoa, Nguyen Thi Hoang Lien, Thai Nguyen Hung Thu, Trinh Van Lau, 2011) , chicken meat ( Nguyen Thi Kim Thuong, Pham Tuan Anh, 2017) D ifferential pulse adsorptive stripping voltammetry, UV - Vis, HPLC , and UPLC are methods which have been commonly used The standard TCVN 8670:2011 of the Vietnamese Ministry of Science and Technology (2011) mentions the method of T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM T ập 16 , S ố 6 ( 2019): 16 - 28 18 quantifying only RB in foodstuffs Some analytical laboratories have identified AO and RB in food samples by liquid chromatography coupled mass spectrometry LC - MS/MS , which is a sensitive method but requires expensive analytical instruments and a complex sample preparation process Therefore, the main purpose of this work is the simultaneous analysis of two synthetic colors in foods by HPLC - PDA , which is a suitable method for small and medium laboratories 2 Materials and methods 2 1 Chemicals and reagents AO (861030, purity ≥ 85%) and RB (R6626, purity ≥ 95%) were bought from Sigma Methanol (99 9%, Lapscan , Thailand), formic acid (98 - 100%; Scharlau, Scharlab S L - Spain), a cetic acid (100%, Merck, Germany) , and phosphoric acid (85%; Scharlau, Scharlab S L - Spain) were used for mobile phase HPLC Ammonia solution in water (25%, Merck, Germany ), acetonitrile (99 9%, Merck, Germany) , and sodium chloride (99 99%, Suprapur) were used to extract AO and RB from foodstuffs 2 2 Instruments Perkin - Elmer Lambda 25 was used for scanning the absorption spectra of AO and RB Waters 2996 HPLC - PDA, a column C18 (Zorbax Eclipse XDB - C18, 5 m x 4 6 mm x 250 mm, Agilent) , and a precolumn (Eclipse XDB - C18, Agilent) were used for chromatographic analysis This system is a manual accessory for high - sensitivity determination of synthetic food colorants The detection wavelength s of AO and RB were 435 nm and 555 nm, respectiv ely Water Pro PS (Labconco) was used for preparing the aqueous mobile phase The mobile phase comprised of (A) methanol and (B) 0 03% phosphoric acid in water The g radient elution was conducted as follows : solvent A started at 30% and kept the balanced ratio in 5 minutes, then increased to 80% in 11 minutes, finally decreased to 30% in 4 minutes with the flow rate of 1 0 mL/min at 35 o C The injection volume was 20 L An ultrasonic assistance ( Elmasonic S 60 (H) - Elma Schmidbauer) and a centrifuge machine (DNA concentrator - miVac, Genevac Ltd Ipswich) were used for extraction of AO and RB in sampl es 2 3 Sample collection and preparation Nine different samples of three kinds of food: pickled collard greens, chicken meat, and bamboo shoots were purchased from Lotte Mart supermarket ( D istrict 10) and three local markets, namely Tan Phuoc ( D istrict 11), Do Dac ( D istrict 2) and Nhat Tao ( D istrict 10) in Ho Chi Minh City, Vietnam All samples were minced and homogenized, then kept in clean, dry containers and stored in the refrigerator before being analyzed T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM Phan Thi Ngoc Trinh et al 19 2 4 Preparation of standard solu tions Stock standard solutions of AO and RB were prepared by dissolving each standard compound in methanol in a volumetric flask at concentration of 1000 mg/L The stock standard solutions were further diluted with methanol to give 100 mg/L standard solut ions for the recovery tests Calibration standards for AO and RB were prepared each day from the certified standard stock solution in the range from 0 05 to 2 0 mg/L All the standard solutions were prepared in methanol 2 5 Extraction procedure Sample (0 5 g) was accurately weighed, added into a 50 mL centrifuge tube, then mixed with acetonitrile:NH 3 (0 5% v/v in water) at the ratio of 7:3 The sample was thoroughly mixed well by a vortex shaker for 10 minutes and extracted under ultrasonic assis tance for 10 minutes at 40°C, then the mixture was centrifuged for 10 minutes and the solution was transferred into a volumetric flask The same treatment was performed three times on the residual precipitates All extraction solutions were combined and di luted to 25 mL with water, then filtered through a 0 22 μm filter of PTFE membrane before transferred into glass vials for HPLC analysis 3 Results and discussion 3 1 Optimization of HPLC - PDA condition Visible spectrums of AO and RB in two solvents, methanol and acetonitrile, were recorded in wavelength range of 400 - 600 nm on Perkin - Elmer Lambda 25 In Fig 2 , maximum absorbance at respective wavelengths 430 440 nm (AO) and 545 555 nm (RB) With the s ame concentration of AO and RB in two solvents, the absorption signal of RB in methanol was higher than the figure in acetonitrile, which was the main reason for using methanol to dilute the stock standards In this study, to achieve the best analytical re sults, PDA operated and monitored the absorbance at 435 nm for AO and 555 nm for RB Fig 2 Molecular absorption spectra of AO and RB in methanol (a) and acetonitrile (b) T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM T ập 16 , S ố 6 ( 2019): 16 - 28 20 The mobile phase was optimized in these factors: pH, mobile phase composition , and elution program because of their significant effects on the chromatographic separation process Table 1 Mobile phase conditions Mobile phase conditions Consequences Organic solvent (1) ACN - CH 3 COOH (2) MeOH - CH 3 COOH (1) Better elution, lower sign al, tailing peak (both AO, RB) (2) Better sen sitivity but AO peak was tailed pH (1) MeOH - H 3 PO 4 0,03% (pH 2 6) (2) MeOH - H 3 PO 4 0,072% (pH 2 3) Good separation peak, resolution (R s ) > 1 5, similarity of retention time, capacity, and peak asymmetry factor Isocratic elution (1) MeOH - HCOOH (pH = 3) 40:60 (2) ACN - HCOOH (pH = 3) 40:60 (3) ACN - acetate buffer (pH = 4,2) 80:20 AO and RB weren’t sep a rated in t hese isocratic elution programs Gradient elution (MeOH - H 3 PO 4 0,03%) - Gradient 1: peak fronting for AO chromatography, rising baseline and low AO intensity - Gradient 2: narrow and symmetrical peaks HPLC chromatogram was shown in Figure 3 Finally, the gradient mobile phase using methanol and 0 03% phosphoric acid aqueous solution was chosen as the preferred mobile phase in the following experiments due to its higher efficiency and reasonable analysis time 20 40 60 80 100 0 5 10 15 20 minute % MeOH Gradient 1 Gradient 2 T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM Phan Thi Ngoc Trinh et al 21 Fig 3 HPLC chromatogram of the standard solution at AO (t R 14 48 min) and RB (t R 17 43 min) in 2 0 mg/L 3 2 Evaluation of range linearity, limit of detection (LOD) and limit of quantification (LOQ) From the data from the paper Ping Qi et al (2016); Shruti Singh, Himani Shah, Ritika Shah , and Krishna Shah (2017) and the prediction of small concentration of AO and RB in food samples, the calibration curve was in range from 0 05 mg/L to 2 00 mg/L of each dye concentration Fig 4 Calibration curves for AO and RB Two calibration curves was linear in this range of concentration The limits of detection (LOD) and quantification (LOQ) of AO and RB were estimated by the signal - to - noise S/N of each peak in the standard solutions LOD and LOQ of these substances gained 0 02 mg/L and 0 05 mg/L, respectively 3 3 Optimization extraction procedure Accurately 5 g bamboo shoots sample was spiked 250 L mixture standard 100 mg/L The sample was kept in the refrigerator in 24 hours for optimizing the n umber of extraction Acetonitrile was used as an extraction solvent because AO and RB dissolved completely in this solvent, which adopted to effectively remove the matrix - matched components such as polar pigments, fatty acids , and protein in samples The extraction process was as follows: added 10 mL acetonitrile: NH 3 (0 5% v/v in water) at the ratio of 7:3 to samples The mixture was continued for 10 minutes in an ultrasonic bath, then was y = 3022 8x - 62 647 R² = 0 9999 0 2,000 4,000 6,000 0 0 5 1 1 5 2 Signal (AU) Concentration (mg/L) AO y = 2949 4x - 20 992 R² = 0 9997 0 2,000 4,000 6,000 0 0 5 1 1 5 2 Signal (AU) Concentration (mg/L) RB T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM T ập 16 , S ố 6 ( 2019): 16 - 28 22 centrifuged in 10 minutes To optimize the sample preparation, the same treatment was re peated three more times by using 5 mL of the solvent acetonitrile: NH 3 each time The each time extract was transfe r red to a volumetric flask, and diluted with water to 25 mL The elution solution was filtered through a 0 22 μm filter of PTFE membrane and transferred into glass vials for HPLC - PDA analysis Fig 5 Ratios of AO and RB in four times of extraction As shown in Fig 5 , the recovery ratio of AO and RB in bamboo shoots sample was from 97 74% to 102 15% after three times of extraction Therefore, the process extracted three times using total 20 mL acetonitrile: NH 3 (0 5% v/v in water), which was selected as an optimum condition of extraction 3 4 Method validation To survey th e repeatability of the measurement, a mixture standard of 2 0 mg/L AO and 2 0 mg/L RB was measured 7 times by HPLC - PDA Table 2 Repeatability of signals N Retention time (min) Peak area AO RB AO RB 1 14 467 17 400 5664 660 5588 101 2 14 450 17 383 5753 625 5646 242 3 14 450 17 383 5787 972 5693 114 4 14 450 17 383 5761 298 5711 542 5 14 450 17 383 5812 427 5704 386 6 14 450 17 383 5802 885 5776 700 7 14 433 17 367 5800 698 5748 406 Mean 14 5 9 1x10 - 3 17 4 8 8x10 - 3 5769 47 5695 58 RSD% 0 63% 0 051% 0 82% 1 0% T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM Phan Thi Ngoc Trinh et al 23 RSD of retention time and peak area were 0 63% and 0 82% (for AO); 0 051% and 1 0 % (for RB), respectively (n = 7) The method gained good precision According to AOAC, with RSD value ≤ 1 8%, the result by HPLC - PDA had good precision Method validation and quality control w ere ensured using blanks and spiked samples Fig 6 HPLC chromatograms of spiked chicken meat sample (a) and chicken meat sample (b) Fig 7 HPLC chromatograms of spiked bamboo shoots sample (a) and bamboo shoots sample (b) T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM T ập 16 , S ố 6 ( 2019): 16 - 28 24 There was a similarity in retention time of noise peaks in spiked samples and samples ( Fig 6a, Fig 6b, Fig 7a , and Fig 7b ) Meanwhile, using the peak purity test function of PDA detector, peaks of AO and RB in spiked samples were 100% purity (Fig 6a and Fig 7a) Consequently, the analytical results showed that AO and RB could be determined exactly in this method without the interferences of the food matrices Recov eries were determined as following : 5 g of a chicken meat sample spiked with mixture standard, then the sample was kept in the refrigerator in 24 hours until an alysis Table 3 Recoveries of AO and RB in the recovery samples Concentration (μg/g) Peak area Recovery (%) AO RB AO RB 0 5 888 191 900 504 868 357 891 151 883 218 885 340 857 136 1954 000 1949 214 1866 296 1982 437 1928 280 1869 000 1861 447 91 52 92 79 89 48 91 83 91 01 91 23 88 32 93 27 93 04 89 08 94 63 92 04 89 21 88 85 Mean 90 9 ± 1 4 91 4 ± 2 2 1 0 1918 208 1894 076 1883 574 1981 334 1926 163 1903 961 1812 305 4345 709 4334 987 4292 576 4432 542 4384 205 4163 081 4251 641 102 54 101 25 100 68 105 91 102 96 101 77 96 87 100 80 100 55 99 57 102 82 101 70 96 57 98 62 Mean 101 7 ± 2 5 100 1 ± 1 9 1 5 2656 685 2650 758 2409 565 2477 072 2496 536 2525 13 2716 479 5997 764 5788 187 5354 113 5661 422 5682 338 5983 245 6169 421 91 51 91 31 83 00 85 33 86 00 86 98 93 57 94 70 91 39 84 54 89 39 89 72 94 47 97 41 Mean 88 2 ± 3 6 91 7 ± 4 0 T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM Phan Thi Ngoc Trinh et al 25 The recoveries of AO and RB were determined by using this method and achieved in range from 83 00% to 105 91% Therefore, this analytical method could be applied to determine AO and RB in foodstuffs 3 5 Some analytical results This given method was appl ied to monitor AO and RB in nine food samples collected from some markets in Ho Chi Minh City Table 4 Information of food samples Samples Name Location Pickled collard greens M1 M2 M3 Nhat Tao market Tan Phuoc market Do Dac market Chicken meat M4 M5 M6 Lotte Mart Tan Phuoc market Do Dac market Bamboo shoots M7 M8 M9 Nhat Tao market Tan Phuoc market Do Dac market The results showed that both AO and RB were absence in nine samples The sample M7 was sent to the analytical laboratory of Hoan Vu Scientific Technologies Company Limited for control analysis by LC - MS/MS method LOD of Hoan Vu laboratory’s method for AO and RB were 0 005 mg/L and 0 05 mg/L, respectively The result of LC - MS/MS method was totally appropriate to th e result in this study T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM T ập 16 , S ố 6 ( 2019): 16 - 28 26 Fig 8 HPLC chromatograms of spiked pickled collard greens sample (a) and pickled collard greens sample (b) 4 Conclusion In this study, HPLC - PDA method was developed and validated for simultaneous analysis of AO and RB in food samples While the LOD and LOQ achieved by this method were 0 02 mg/L and 0 05 mg/L for two dyes, the recoveries ranged from 83 00% to 105 91% The method was applied to determine two dyes in nine samples (pickled collard greens, chicken meat , and bamboo shoots) and the analytical results shows that these dyes were absent from all monitored samples This study showed that HPLC - PDA is a reliable and fast tool for monitoring these dyes in foodstuffs in small and medium analytical laboratories Conflict of Interest: Authors have no conflict of interest to declare REFERENCES Adejumoke , A Inyinbor, Folahan , A , Adekola , b & Gabriel A Olatunji (2015) Adsorption of Rhodamine B Dye from Aqueous Solution on Irvingia gabonensis Biomass: Kinetics and Thermodynamics Studies 68 , 115 - 125 Agents classified by the IARC Monographs (2019) International Agency for Research on Cance r, 1 - 123 Brazeau , J (2018) Identification and Quantitation of Water - Soluble Synthetic Colors in Foods by Liquid Chromatography/Ultraviolet - Visible Method Development and Validation ACS Omega, 3 , 6577 - 6586 Dong - yang Chen, Hao Zhang, Jia - li Feng, Dong Z eng, Li Ding, Xian - jun Liu, & Bang - rui Li (2017) Research on the Determination of 10 Industrial Dyes in Foodstuffs Journal of Chromatographic Science, 55 (10), 1021 - 1025 T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM Phan Thi Ngoc Trinh et al 27 Le Thi Huong Hoa, Nguyen Thi Hoang Lien, Thai Nguyen Hung Thu, & Trinh Van Lau (20 11) Method development of determination of illegal dye Rhodamine B in cosmetics Vietnam Pharmaceutical Journal, 51 (10), 23 - 27 Nguyen Thi Kim Thuong, Pham Tuan Anh (2017, Sep - Oct) Study on determination Auramine O in food samples by differential pulse adsorptive stripping voltammetry Testing today - Journal of Vietnam Asscociation of Testing Laboratories , (1), 6 - 10 Peng Wang, Mingming Cheng, & Zhonghai Zhang (2014) On different photodecomposition behaviors of rhodamine B on laponite and montmorillonite clay under visible light irradiation Journal of Saudi Chemical Society, 18 (4), 308 - 316 Ping Qi, Zhi - an Liang, Yu wang, Jian Xiao, Jia Liu, Qi ng - qiong Zhou, Chun - hao Zheng, Li - ni Luo, Zi - hao Lin, Fang Zhu, & Xue - wu Zhang (2016) Mixed hemimicelles solid - phase extraction based on sodium dodecyl sulfate - coated nano - magnets for selective adsorption and enrichment of illegal cationic dyes in food m atrices prior to high - performance liquid chromatography - diode array detection detec Journal of Chromatography A, 1437 , 25 - 36 Sabnis R W (2010) Synthesis and Industrial Applications In Handbook of biological dyes and stains (pp 27 - 29) John Wiley & So ns, Inc , Publication Shruti Singh, Himani Shah, Ritika Shah , & Krishna Shah (2017) Identification and Estimation of Non - Permitted Food Colours (Sudan and Rhodamine - B Dye) In Chilli and Curry Powder by Rapid Colour Test, Thin Layer Chromatography and Sp ectrophotometry International Journal of Current Microbiology and Applied Sciences, 6 (7), 1970 - 1981 Vietnamese Ministry of Health (n d ) Decision Issuing the "Regulation of the list of additives permitted for use in food " In 3742/2001/QĐ - BYT Vietnam ese Ministry of Heathy (n d ) The Circular issued the list of food, food additives, food processing support substances and tools, packaging materials, containing food to be identified as a commodity code according to the list of exported and imported goo ds to Vietnam In 05/2018/TT - BYT Vietnamese Ministry of Science and Technology (n d ) Foodstuffs - Determination of rhodamine B by high performance liquid chromatography (HPLC) In TCVN 8670 : 2011 Xue Wang, Hui - Ling Duan, Shi - Yao Ma, Jun Wang, Han - Yi ng Zhan & Zhi - Qi Zhang (2018) Selective adsorption and separation of illegal cationic dyes from foodstuffs with anionic polyelectrolyte functionalization of metal - organic frameworks Food Additives & Contaminants: Part A, 35 (10), 1870 - 1880 Yung - Yi Cheng, & Tung - Hu Tsai (2016) A validated LC – MS/MS determination method for the illegal food additive rhodamine B: Applications of a pharmacokinetic study in rats Journal of Pharmaceutical and Biomedical Analysis, 125 , 394 - 399 T ẠP CHÍ KHOA HỌC - Trư ờng ĐHSP TPHCM T ập 16 , S ố 6 ( 2019): 16 - 28 28 XÂY D Ự NG V À Đ Á NH GI Á PHƯƠNG PH Á P HPLC - PDA D Ù NG Đ Ể X Á C Đ Ị NH Đ Ồ NG TH Ờ I AURAMINE O V À RHODAMINE B TRONG TH Ự C PH Ẩ M Phan Th ị Ng ọ c Trinh 1 , Nguy ễ n Thanh Thơi 1 , Hu ỳ nh Th ị Nh à n 1 , Nguy ễ n Ng ọ c Hưng 1 , Nguy ễ n Th ị Tuy ế t Nhung 1* , Lê Th ị H ồ ng Vân 2 , Lê Văn Hu ấ n 2 Khoa Hóa h ọ c – Trư ờ ng Đ ạ i h ọ c Sư ph ạ m Thành ph ố H ồ Ch í Minh Trung tâm Khoa h ọ c v à Công ngh ệ Dư ợ c S à i G ò n – Trư ờ ng Đ ạ i h ọ c Y Dư ợ c Th à nh ph ố H ồ Ch í Minh * Corresponding author: Nguy ễn Thị Tuyết Nhung – Email: nhungntt@hcmue edu vn Ngày nh ận b ài: 18 - 3 - 2019; ngày nh ận b ài s ửa: 11 - 4 - 2019; ngày duy ệt đăng: 15 - 6 - 2019 T Ó M T Ắ T Nghiên c ứ u n à y ph á t tri ể n phương ph á p HPLC - PDA v ớ i m ụ c đ í ch x á c đ ị nh hai thu ố c nhu ộ m Auramine O v à Rhodamine B trong c á c m ẫ u th ự c ph ẩ m Gi ớ i h ạ n ph á t hi ệ n v à gi ớ i h ạ n đ ị nh lư ợ ng c ủ a Auramine O (AO) v à Rhodamine B (RB) c ù ng c ó gi á tr ị l ầ n lư ợ t l à 0,02 mg/L v à 0,05 mg/L H ệ s ố thu h ồ i c ủ a AO v à RB trong n ề n m ẫ u th ự c ph ẩ m c ó gi á tr ị trong kho ả ng 83,00% đ ế n 105,91% ở c á c n ồ ng đ ộ ch ấ t phân t í ch 0,5 μg/g, 1,0 μg/g v à 1,5 μg/g Phương ph á p phân t í ch đư ợ c th ự c hi ệ n đơn gi ả n, c ó k ế t qu ả ch í nh x á c v à ph ù h ợ p đ ể đ ị nh lư ợ ng c á c thu ố c nhu ộ m n à y trong th ự c ph ẩ m T ừ kh ó a: Auramine O, Rhodamine B, HPLC - PDA, th ự c ph ẩ m
TRƯỜNG ĐẠI HỌC SƯ PHẠM TP HỒ CHÍ MINH HO CHI MINH CITY UNIVERSITY OF EDUCATION TẠP CHÍ KHOA HỌC JOURNAL OF SCIENCE ISSN: KHOA HỌC TỰ NHIÊN VÀ CÔNG NGHỆ NATURAL SCIENCES AND TECHNOLOGY 1859-3100 Tập 16, Số (2019): 16-28 Vol 16, No (2019): 16-28 Email: tapchikhoahoc@hcmue.edu.vn; Website: http://tckh.hcmue.edu.vn DEVELOPMENT AND VALIDATION OF HPLC-PDA METHOD FOR THE SIMULTANEOUS DETERMINATION OF AURAMINE O AND RHODAMINE B IN FOODSTUFFS Phan Thi Ngoc Trinh1, Nguyen Thanh Thoi1, Huynh Thi Nhan1, Nguyen Ngoc Hung1, Nguyen Thi Tuyet Nhung1*, Le Thi Hong Van2, Le Van Huan2 Faculty of chemistry – Ho Chi Minh City University of Education Saigon Pharmaceutical Sciences and Technologies Center, University of Medicine and Pharmacy at Ho Chi Minh City * Corresponding author: Nguyen Thi Tuyet Nhung – Email: nhungntt@hcmue.edu.vn Received: 18/3/2019; Revised: 11/4/2019; Accepted:15/6/2019 ABSTRACT HPLC-PDA method was used to determine two basic dyes, Auramine O and Rhodamine B in food samples The limits of detection and quantification of Auramine O (AO) and Rhodamine B (RB) were both 0.02 mg/L and 0.05 mg/L, respectively The recoveries of AO and RB in matrices ranged from 83.00% to 105.91% with the concentration 0.5 µg/g, 1.0 µg/g and 1.5 µg/g This is a simple and accurate method which can be applied to quantify of dyes in foodstuffs Keywords: Auramine O, Rhodamine B, HPLC-PDA, foodstuffs Introduction Auramine O (C17H22N3Cl, 303.84 gram/mol, yellow powder, pKa 9.8, 10.7 (Sabnis R.W, 2010)) and Rhodamine B (C28H31N2O3Cl, 479.01 gram/mol, green powder, pKa 3.7 (Peng Wang, Mingming Cheng, Zhonghai Zhang, 2014)) are synthetic azo–cationic dyes with the chemical structures shown in Fig RB functions as a water tracer fluorescent and is commonly used as a colorant in textiles, paper industry (Adejumoke A Inyinbor, Folahan A Adekolab and Gabriel A Olatunji, 2015) while AO is commonly used to dye wood products, leather, ink, and textile fabrics (Sabnis R.W, 2010) According to the International Agency of Research on Cancer, RB is associated with eye, skin, and respiratory tract irritation and AO is possibly carcinogenic to humans (Agents classified by the IARC Monographs, 2019) For that reason, the presence of AO and RB in foodstuffs may cause adverse effects on human’s health Ingestion of these dyes over time causes respiratory tract ailments as well as acute and chronic poisoning Therefore, both of them are not listed as food additives allowed for use in Vietnam and other countries (Vietnamese Ministry of Health, 2001, 2018) However, because of their easy coloring, high stability, and low cost, AO and RB are still being used illegally in animal feed, food, and cosmetics as beauty color as well as to ensure the uniform appearance of foodstuffs 16 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Phan Thi Ngoc Trinh et al AO RB Fig Chemical structures of AO and RB The color products have received the attention of many researchers in many countries in the world The study of Dong-yang Chen, Hao Zhang, Jia-li Feng, Dong Zeng, Li Ding, Xian-jun Liu, Bang-rui Li (2017) has presented high performance liquid chromatography coupled with diode array detector to detect 10 industrial dyes (basic orange 2, basic orange 21, basic orange 22, acid orange II, auramine, basic rhodamine B, and Sudan I–IV) in a great variety of matrices such as meat products (smoked sausages, preserved ham, fresh meat, yellow-fin tuna), chili powder, and chili oil in China While the limit of detection has been in the range of 0.007–0.01 mg/kg, high recoveries (80.6–104%) and good reproducibility (1.1–5.7%) have been obtained Shruti Singh, Himani Shah, Ritika Shah, and Krishna Shah (2017) have separated and determined Sudan and Rhodamine B dye in chili and curry powder matrices by thin layer chromatography and spectrophotometry method Brazeau J (2018) has detected 27 color compounds (including RB) in food by liquid chromatography/ultraviolet visible Using poly(sodium 4–styrenesulfonate) modified MIL–101(Cr)–NH2 to enable effective adsorption and separation of AO, RB, and pararosaniline from foodstuffs prior to high performance liquid chromatography analysis, under optimized conditions, the recoveries of the three dyes in shrimp powder, chili powder, tofu sheets, and tomato sauce have been in the range of 86.8–119.3% (Xue Wang, Hui-Ling Duan, Shi-Yao Ma, Jun Wang, Han-Ying Zhan, & Zhi-Qi Zhang, 2018) Serum samples have been deproteinized with acetonitrile and separated by UHPLC on a reverse- phase C18, using a triple quadrupole tandem mass spectrometer in the selected reaction monitoring mode at [M]+ ion m/z 443.39 → 399.28 for RB Linear has resulted over a concentration range of 0.5–100 ng/mL, with a lower limit of quantitation of 0.5 ng/mL (Yung-Yi Cheng, Tung-Hu Tsai, 2016) In Vietnam, some analytical studies have determined only AO or RB in matrices, such as cosmetics (Le Thi Huong Hoa, Nguyen Thi Hoang Lien, Thai Nguyen Hung Thu, Trinh Van Lau, 2011), chicken meat (Nguyen Thi Kim Thuong, Pham Tuan Anh, 2017) Differential pulse adsorptive stripping voltammetry, UV-Vis, HPLC, and UPLC are methods which have been commonly used The standard TCVN 8670:2011 of the Vietnamese Ministry of Science and Technology (2011) mentions the method of 17 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Tập 16, Số (2019): 16-28 quantifying only RB in foodstuffs Some analytical laboratories have identified AO and RB in food samples by liquid chromatography coupled mass spectrometry LC-MS/MS, which is a sensitive method but requires expensive analytical instruments and a complex sample preparation process Therefore, the main purpose of this work is the simultaneous analysis of two synthetic colors in foods by HPLC-PDA, which is a suitable method for small and medium laboratories Materials and methods 2.1 Chemicals and reagents AO (861030, purity ≥ 85%) and RB (R6626, purity ≥ 95%) were bought from Sigma Methanol (99.9%, Lapscan, Thailand), formic acid (98-100%; Scharlau, Scharlab S.L - Spain), acetic acid (100%, Merck, Germany), and phosphoric acid (85%; Scharlau, Scharlab S.L - Spain) were used for mobile phase HPLC Ammonia solution in water (25%, Merck, Germany), acetonitrile (99.9%, Merck, Germany), and sodium chloride (99.99%, Suprapur) were used to extract AO and RB from foodstuffs 2.2 Instruments Perkin-Elmer Lambda 25 was used for scanning the absorption spectra of AO and RB Waters 2996 HPLC-PDA, a column C18 (Zorbax Eclipse XDB - C18, m x 4.6 mm x 250 mm, Agilent), and a precolumn (Eclipse XDB - C18, Agilent) were used for chromatographic analysis This system is a manual accessory for high-sensitivity determination of synthetic food colorants The detection wavelengths of AO and RB were 435 nm and 555 nm, respectively Water Pro PS (Labconco) was used for preparing the aqueous mobile phase The mobile phase comprised of (A) methanol and (B) 0.03% phosphoric acid in water The gradient elution was conducted as follows: solvent A started at 30% and kept the balanced ratio in minutes, then increased to 80% in 11 minutes, finally decreased to 30% in minutes with the flow rate of 1.0 mL/min at 35oC The injection volume was 20 L An ultrasonic assistance (Elmasonic S 60 (H) - Elma Schmidbauer) and a centrifuge machine (DNA concentrator - miVac, Genevac Ltd Ipswich) were used for extraction of AO and RB in samples 2.3 Sample collection and preparation Nine different samples of three kinds of food: pickled collard greens, chicken meat, and bamboo shoots were purchased from Lotte Mart supermarket (District 10) and three local markets, namely Tan Phuoc (District 11), Do Dac (District 2) and Nhat Tao (District 10) in Ho Chi Minh City, Vietnam All samples were minced and homogenized, then kept in clean, dry containers and stored in the refrigerator before being analyzed 18 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Phan Thi Ngoc Trinh et al 2.4 Preparation of standard solutions Stock standard solutions of AO and RB were prepared by dissolving each standard compound in methanol in a volumetric flask at concentration of 1000 mg/L The stock standard solutions were further diluted with methanol to give 100 mg/L standard solutions for the recovery tests Calibration standards for AO and RB were prepared each day from the certified standard stock solution in the range from 0.05 to 2.0 mg/L All the standard solutions were prepared in methanol 2.5 Extraction procedure Sample (0.5 g) was accurately weighed, added into a 50 mL centrifuge tube, then mixed with acetonitrile:NH3 (0.5% v/v in water) at the ratio of 7:3 The sample was thoroughly mixed well by a vortex shaker for 10 minutes and extracted under ultrasonic assistance for 10 minutes at 40°C, then the mixture was centrifuged for 10 minutes and the solution was transferred into a volumetric flask The same treatment was performed three times on the residual precipitates All extraction solutions were combined and diluted to 25 mL with water, then filtered through a 0.22 μm filter of PTFE membrane before transferred into glass vials for HPLC analysis Results and discussion 3.1 Optimization of HPLC-PDA condition Visible spectrums of AO and RB in two solvents, methanol and acetonitrile, were recorded in wavelength range of 400-600 nm on Perkin-Elmer Lambda 25 In Fig 2, maximum absorbance at respective wavelengths 430 440 nm (AO) and 545 555 nm (RB) With the same concentration of AO and RB in two solvents, the absorption signal of RB in methanol was higher than the figure in acetonitrile, which was the main reason for using methanol to dilute the stock standards In this study, to achieve the best analytical results, PDA operated and monitored the absorbance at 435 nm for AO and 555 nm for RB Fig Molecular absorption spectra of AO and RB in methanol (a) and acetonitrile (b) 19 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Tập 16, Số (2019): 16-28 The mobile phase was optimized in these factors: pH, mobile phase composition, and elution program because of their significant effects on the chromatographic separation process Table Mobile phase conditions Mobile phase conditions Consequences (1) Better elution, lower signal, tailing Organic (1) ACN - CH3COOH peak (both AO, RB) solvent (2) MeOH - CH3COOH (2) Better sensitivity but AO peak was tailed (1) MeOH - H3PO4 0,03% (pH 2.6) Good separation peak, resolution (Rs) > pH 1.5, similarity of retention time, capacity, and peak asymmetry factor (2) MeOH - H3PO4 0,072% (pH 2.3) Isocratic (1) MeOH - HCOOH (pH = 3) 40:60 (2) elution ACN - HCOOH (pH = 3) 40:60 (3) ACN - acetate buffer (pH = 4,2) 80:20 Gradient 100 % MeOH Gradient AO and RB weren’t separated in these elution 80 Gradient isocratic elution programs (MeOH - 60 H3PO4 - Gradient 1: peak fronting for AO 0,03%) 40 minute chromatography, rising baseline and low 10 15 20 AO intensity 20 - Gradient 2: narrow and symmetrical peaks HPLC chromatogram was shown in Figure Finally, the gradient mobile phase using methanol and 0.03% phosphoric acid aqueous solution was chosen as the preferred mobile phase in the following experiments due to its higher efficiency and reasonable analysis time 20 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Phan Thi Ngoc Trinh et al Fig HPLC chromatogram of the standard solution at AO (tR 14.48 min) and RB (tR 17.43 min) in 2.0 mg/L 3.2 Evaluation of range linearity, limit of detection (LOD) and limit of quantification (LOQ) From the data from the paper Ping Qi et al (2016); Shruti Singh, Himani Shah, Ritika Shah, and Krishna Shah (2017) and the prediction of small concentration of AO and RB in food samples, the calibration curve was in range from 0.05 mg/L to 2.00 mg/L of each dye concentration AO RB y = 3022.8x - 62.647 6,000y = 2949.4x - 20.992 R² = 0.9999 R² = 0.9997 Signal (AU)6,000 Signal (AU)4,0004,000 2,000 2,000 0.5 1.5 0.5 1.5 Concentration (mg/L) Concentration (mg/L) Fig Calibration curves for AO and RB Two calibration curves was linear in this range of concentration The limits of detection (LOD) and quantification (LOQ) of AO and RB were estimated by the signal-to- noise S/N of each peak in the standard solutions LOD and LOQ of these substances gained 0.02 mg/L and 0.05 mg/L, respectively 3.3 Optimization extraction procedure Accurately g bamboo shoots sample was spiked 250 L mixture standard 100 mg/L The sample was kept in the refrigerator in 24 hours for optimizing the number of extraction Acetonitrile was used as an extraction solvent because AO and RB dissolved completely in this solvent, which adopted to effectively remove the matrix-matched components such as polar pigments, fatty acids, and protein in samples The extraction process was as follows: added 10 mL acetonitrile: NH3 (0.5% v/v in water) at the ratio of 7:3 to samples The mixture was continued for 10 minutes in an ultrasonic bath, then was 21 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Tập 16, Số (2019): 16-28 centrifuged in 10 minutes To optimize the sample preparation, the same treatment was repeated three more times by using mL of the solvent acetonitrile: NH3 each time The each time extract was transferred to a volumetric flask, and diluted with water to 25 mL The elution solution was filtered through a 0.22 μm filter of PTFE membrane and transferred into glass vials for HPLC-PDA analysis Fig Ratios of AO and RB in four times of extraction As shown in Fig 5, the recovery ratio of AO and RB in bamboo shoots sample was from 97.74% to 102.15% after three times of extraction Therefore, the process extracted three times using total 20 mL acetonitrile: NH3 (0.5% v/v in water), which was selected as an optimum condition of extraction 3.4 Method validation To survey the repeatability of the measurement, a mixture standard of 2.0 mg/L AO and 2.0 mg/L RB was measured times by HPLC-PDA Table Repeatability of signals N Retention time (min) Peak area AO RB AO RB 14.467 17.400 5664.660 5588.101 14.450 17.383 5753.625 5646.242 14.450 17.383 5787.972 5693.114 14.450 17.383 5761.298 5711.542 14.450 17.383 5812.427 5704.386 14.450 17.383 5802.885 5776.700 14.433 17.367 5800.698 5748.406 Mean 14.5 9.1x10-3 17.4 8.8x10-3 5769 47 5695 58 RSD% 0.63% 0.051% 0.82% 1.0% 22 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Phan Thi Ngoc Trinh et al RSD of retention time and peak area were 0.63% and 0.82% (for AO); 0.051% and 1.0 % (for RB), respectively (n = 7) The method gained good precision According to AOAC, with RSD value ≤ 1.8%, the result by HPLC-PDA had good precision Method validation and quality control were ensured using blanks and spiked samples Fig HPLC chromatograms of spiked chicken meat sample (a) and chicken meat sample (b) Fig HPLC chromatograms of spiked bamboo shoots sample (a) and bamboo shoots sample (b) 23 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Tập 16, Số (2019): 16-28 There was a similarity in retention time of noise peaks in spiked samples and samples (Fig 6a, Fig 6b, Fig 7a, and Fig 7b) Meanwhile, using the peak purity test function of PDA detector, peaks of AO and RB in spiked samples were 100% purity (Fig 6a and Fig 7a) Consequently, the analytical results showed that AO and RB could be determined exactly in this method without the interferences of the food matrices Recoveries were determined as following: g of a chicken meat sample spiked with mixture standard, then the sample was kept in the refrigerator in 24 hours until analysis Table Recoveries of AO and RB in the recovery samples Concentration (µg/g) Peak area Recovery (%) AO RB AO RB 888.191 1954.000 91.52 93.27 900.504 1949.214 92.79 93.04 868.357 1866.296 89.48 89.08 891.151 1982.437 91.83 94.63 0.5 883.218 1928.280 91.01 92.04 885.340 1869.000 91.23 89.21 857.136 1861.447 88.32 88.85 Mean 90.9 ± 1.4 91.4 ± 2.2 1918.208 4345.709 102.54 100.80 1894.076 4334.987 101.25 100.55 1883.574 4292.576 100.68 99.57 1981.334 4432.542 105.91 102.82 1.0 1926.163 4384.205 102.96 101.70 1903.961 4163.081 101.77 96.57 1812.305 4251.641 96.87 98.62 Mean 101.7 ± 2.5 100.1 ± 1.9 2656.685 5997.764 91.51 94.70 2650.758 5788.187 91.31 91.39 2409.565 5354.113 83.00 84.54 2477.072 5661.422 85.33 89.39 1.5 2496.536 5682.338 86.00 89.72 2525.13 5983.245 86.98 94.47 2716.479 6169.421 93.57 97.41 Mean 88.2 ± 3.6 91.7 ± 4.0 24 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Phan Thi Ngoc Trinh et al The recoveries of AO and RB were determined by using this method and achieved in range from 83.00% to 105.91% Therefore, this analytical method could be applied to determine AO and RB in foodstuffs 3.5 Some analytical results This given method was applied to monitor AO and RB in nine food samples collected from some markets in Ho Chi Minh City Table Information of food samples Samples Name Location M1 Nhat Tao market Pickled collard greens M2 Tan Phuoc market M3 Do Dac market M4 Lotte Mart Chicken meat M5 Tan Phuoc market M6 Do Dac market M7 Nhat Tao market Bamboo shoots M8 Tan Phuoc market M9 Do Dac market The results showed that both AO and RB were absence in nine samples The sample M7 was sent to the analytical laboratory of Hoan Vu Scientific Technologies Company Limited for control analysis by LC-MS/MS method LOD of Hoan Vu laboratory’s method for AO and RB were 0.005 mg/L and 0.05 mg/L, respectively The result of LC-MS/MS method was totally appropriate to the result in this study 25 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Tập 16, Số (2019): 16-28 Fig HPLC chromatograms of spiked pickled collard greens sample (a) and pickled collard greens sample (b) Conclusion In this study, HPLC-PDA method was developed and validated for simultaneous analysis of AO and RB in food samples While the LOD and LOQ achieved by this method were 0.02 mg/L and 0.05 mg/L for two dyes, the recoveries ranged from 83.00% to 105.91% The method was applied to determine two dyes in nine samples (pickled collard greens, chicken meat, and bamboo shoots) and the analytical results shows that these dyes were absent from all monitored samples This study showed that HPLC-PDA is a reliable and fast tool for monitoring these dyes in foodstuffs in small and medium analytical laboratories Conflict of Interest: Authors have no conflict of interest to declare REFERENCES Adejumoke, A Inyinbor, Folahan, A., Adekola,b & Gabriel A Olatunji (2015) Adsorption of Rhodamine B Dye from Aqueous Solution on Irvingia gabonensis Biomass: Kinetics and Thermodynamics Studies 68, 115-125 Agents classified by the IARC Monographs (2019) International Agency for Research on Cancer, 1-123 Brazeau, J (2018) Identification and Quantitation of Water-Soluble Synthetic Colors in Foods by Liquid Chromatography/Ultraviolet-Visible Method Development and Validation ACS Omega, 3, 6577-6586 Dong-yang Chen, Hao Zhang, Jia-li Feng, Dong Zeng, Li Ding, Xian-jun Liu, & Bang-rui Li (2017) Research on the Determination of 10 Industrial Dyes in Foodstuffs Journal of Chromatographic Science, 55(10), 1021-1025 26 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Phan Thi Ngoc Trinh et al Le Thi Huong Hoa, Nguyen Thi Hoang Lien, Thai Nguyen Hung Thu, & Trinh Van Lau (2011) Method development of determination of illegal dye Rhodamine B in cosmetics Vietnam Pharmaceutical Journal, 51(10), 23-27 Nguyen Thi Kim Thuong, Pham Tuan Anh (2017, Sep-Oct) Study on determination Auramine O in food samples by differential pulse adsorptive stripping voltammetry Testing today - Journal of Vietnam Asscociation of Testing Laboratories, (1), 6-10 Peng Wang, Mingming Cheng, & Zhonghai Zhang (2014) On different photodecomposition behaviors of rhodamine B on laponite and montmorillonite clay under visible light irradiation Journal of Saudi Chemical Society, 18(4), 308-316 Ping Qi, Zhi-an Liang, Yu wang, Jian Xiao, Jia Liu, Qing-qiong Zhou, Chun-hao Zheng, Li-ni Luo, Zi-hao Lin, Fang Zhu, & Xue-wu Zhang (2016) Mixed hemimicelles solid-phase extraction based on sodium dodecyl sulfate-coated nano-magnets for selective adsorption and enrichment of illegal cationic dyes in food matrices prior to high-performance liquid chromatography-diode array detection detec Journal of Chromatography A, 1437, 25-36 Sabnis R.W (2010) Synthesis and Industrial Applications In Handbook of biological dyes and stains (pp 27-29) John Wiley & Sons, Inc., Publication Shruti Singh, Himani Shah, Ritika Shah, & Krishna Shah (2017) Identification and Estimation of Non-Permitted Food Colours (Sudan and Rhodamine-B Dye) In Chilli and Curry Powder by Rapid Colour Test, Thin Layer Chromatography and Spectrophotometry International Journal of Current Microbiology and Applied Sciences, 6(7), 1970-1981 Vietnamese Ministry of Health (n.d.) Decision Issuing the "Regulation of the list of additives permitted for use in food" In 3742/2001/QĐ-BYT Vietnamese Ministry of Heathy (n.d.) The Circular issued the list of food, food additives, food processing support substances and tools, packaging materials, containing food to be identified as a commodity code according to the list of exported and imported goods to Vietnam In 05/2018/TT-BYT Vietnamese Ministry of Science and Technology (n.d.) Foodstuffs - Determination of rhodamine B by high performance liquid chromatography (HPLC) In TCVN 8670: 2011 Xue Wang, Hui-Ling Duan, Shi-Yao Ma, Jun Wang, Han-Ying Zhan & Zhi-Qi Zhang (2018) Selective adsorption and separation of illegal cationic dyes from foodstuffs with anionic polyelectrolyte functionalization of metal-organic frameworks Food Additives & Contaminants: Part A, 35(10), 1870-1880 Yung-Yi Cheng, & Tung-Hu Tsai (2016) A validated LC–MS/MS determination method for the illegal food additive rhodamine B: Applications of a pharmacokinetic study in rats Journal of Pharmaceutical and Biomedical Analysis, 125, 394-399 27 TẠP CHÍ KHOA HỌC - Trường ĐHSP TPHCM Tập 16, Số (2019): 16-28 XÂY DỰNG VÀ ĐÁNH GIÁ PHƯƠNG PHÁP HPLC-PDA DÙNG ĐỂ XÁC ĐỊNH ĐỒNG THỜI AURAMINE O VÀ RHODAMINE B TRONG THỰC PHẨM Phan Thị Ngọc Trinh1, Nguyễn Thanh Thơi1, Huỳnh Thị Nhàn1, Nguyễn Ngọc Hưng1, Nguyễn Thị Tuyết Nhung1*, Lê Thị Hồng Vân2, Lê Văn Huấn2 Khoa Hóa học – Trường Đại học Sư phạm Thành phố Hồ Chí Minh Trung tâm Khoa học Cơng nghệ Dược Sài Gịn – Trường Đại học Y Dược Thành phố Hồ Chí Minh * Corresponding author: Nguyễn Thị Tuyết Nhung – Email: nhungntt@hcmue.edu.vn Ngày nhận bài: 18-3-2019; ngày nhận sửa: 11-4-2019; ngày duyệt đăng: 15-6-2019 TÓM TẮT Nghiên cứu phát triển phương pháp HPLC-PDA với mục đích xác định hai thuốc nhuộm Auramine O Rhodamine B mẫu thực phẩm Giới hạn phát giới hạn định lượng Auramine O (AO) Rhodamine B (RB) có giá trị 0,02 mg/L 0,05 mg/L Hệ số thu hồi AO RB mẫu thực phẩm có giá trị khoảng 83,00% đến 105,91% nồng độ chất phân tích 0,5 µg/g, 1,0 µg/g 1,5 µg/g Phương pháp phân tích thực đơn giản, có kết xác phù hợp để định lượng thuốc nhuộm thực phẩm Từ khóa: Auramine O, Rhodamine B, HPLC-PDA, thực phẩm 28