Accepted Manuscript

Title: An improved method validation for rapid determination

of acrylamide in foods by ultra-performance liquid
chromatography combined with tandem mass spectrometry

Authors: Yu Zhang, Jingjing Jiao, Zengxuan Cai, Ying Zhang,

Yiping Ren

PII: S0021-9673(06)02380-6
DOI: doi:10.1016/j.chroma.2006.12.086
Reference: CHROMA 347155

To appear in: Journal of Chromatography A

Received date: 23-10-2006

Revised date: 8-12-2006
Accepted date: 15-12-2006

Please cite this article as: Y. Zhang, J. Jiao, Z. Cai, Y. Zhang, Y. Ren, An
improved method validation for rapid determination of acrylamide in foods by ultra-
performance liquid chromatography combined with tandem mass spectrometry, Journal
of Chromatography A (2006), doi:10.1016/j.chroma.2006.12.086

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.
 1 An improved method validation for rapid determination of
2 acrylamide in foods by ultra-performance liquid
3 chromatography combined with tandem mass
4 spectrometry
5 (Short communication)

t
rip
6

7 Yu Zhang a, Jingjing Jiao a, Zengxuan Cai b, Ying Zhang a, Yiping Ren a,b,*

sc
8

10
a

nu
Department of Food Science and Nutrition, College of Biosystems Engineering and

Food Science, Zhejiang University, Hangzhou 310029, Zhejiang Province, China

Ma
b
11 Zhejiang Provincial Center for Disease Prevention and Control, Hangzhou 310009,

12 Zhejiang Province, China

13
d

14 * Corresponding author. Tel.: +86 571 8723 5070.

pte

15 E-mail address: y_zhang@zju.edu.cn (Y. Zhang).

16
ce
Ac

Page 1 of 15
 1 Abstract

2 An improved method was validated for the determination of acrylamide in foods using

3 ultra-performance liquid chromatography (UPLC) coupled to eletrospray ionization tandem mass

4 spectrometry (MS/MS). This improved method supplies a rapid quantitative procedure of

5 acrylamide with a run time of only 3 min. Results of showed a good repeatability (RSD<4.5%) with

t
rip
6 50, 100, 250, 500 and 1000 µg/kg spiked concentrations in potato crisps in within-day (n = 5) and

7 day-to-day (n = 10) precision tests. Meanwhile, good recoveries (81.6%–99.0%) were obtained with

sc
8 the same spiked concentrations in acrylamide-free cereal samples (n = 3). The excellent method

10
nu
validation data and proficiency test results (Z-score: –0.1) of the official Food Analysis Performance

Assessment Scheme (FAPAS) suggested that the present quantitative method could be applied for
Ma
11 rapid determination of acrylamide in many investigations.

12

13 Keywords: Acrylamide; Rapid determination; Method validation; UPLC-MS/MS

d

14
pte
ce
Ac

Page 2 of 15
 1 1. Introduction

2 The discovery of acrylamide in heat-treated foods attracted wide attention and prompted

3 researchers to develop and validate rapid analytical methods for the quantification [1–3]. Some

4 review articles indicated that the mass spectrometry based chromatographic technology is the most

5 useful and recommended method for the acrylamide quantification [4–6]. Meanwhile, routine

t
rip
6 HPLC or GC method was successfully applied for the determination of acrylamide [7–9]. Some

7 other chromatographic methods including capillary zone electrophoresis [10] and LC with pulsed

sc
8 electrochemical detection [11] were also used to develop the analytical methods of acrylamide since

10
this year.
nu
The rapid analysis of acrylamide in food matrixes became more and more important when a huge
Ma
11 number of food samples need to be investigated. Therefore, the general pretreatment steps including

12 impurity removal, analyte extraction, derivatization, clean-up and chromatographic conditions

13 should be optimized to satisfy the requests of rapid determination. Gökmen and Şenyuva [12]
d

14 developed a generic method for the determination of acrylamide in various foods and harmonized
pte

15 the sample preparation protocols to increase the comparability of results. Mastovska and Lehotay

16 [13] optimized a fast and easy sample pretreatment procedure including combined extraction and
ce

17 clean-up. The procedures including defatting, extraction and separation were merged and achieved
Ac

18 by IS-spiked sample homogenization together with hexane, water, acetonitrile, MgSO4 and NaCl so

19 that the time consumption of pretreatment was greatly shortened. Compared to the sample

20 preparation, few studies focused on the rapid quantification of acrylamide by the optimization of

21 chromatographic conditions. Ultra-performance liquid chromatography (UPLC) is made up of

22 available reverse phase chromatographic media with a 1.7 µm particle size along with a liquid

Page 3 of 15
1 system that can operate such columns at much higher pressures [14]. The major advantages of such

2 technology over conventional HPLC utilizing columns packed with 5 µm particles include

3 improved resolution within a shorter retention time and higher analytical sensitivity [15]. The aim of

4 this study is to develop a UPLC-MS/MS method for the determination of acrylamide in foods by the

5 methodology validation and proficiency test.

t
rip
6

7 2. Experimental

sc
8 2.1. Chemicals and reagents

10
nu
Acrylamide (>99%) and D3-labelled acrylamide (isotopic purity 99%) were purchased from

Sigma-Aldrich (St. Louis, MO, USA) and Cambridge Isotope Laboratories (Andover, MA, USA),
Ma
11 respectively. Ethyl acetate and formic acid (96%) were obtained from Tedia (Fairfield, OH, USA)

12 while methanol (HPLC-grade) was purchased from Merck (Whitehouse Station, NJ, USA). The

13 stock solution of acrylamide standard (1 mg/mL) was prepared by dissolving in ultra-purified water.
d

14 The working standard solutions for linear calibration were prepared by diluting the stock solution to
pte

15 the concentration sequences of 1, 5, 10, 50, 100, 150 and 200 ng/mL spiked with the internal

16 standard (100 ng/mL). Both the stock and working solutions were kept at 4 °C for a month.
ce

17
Ac

18 2.2. Samples

19 Certified reference test material (FAPAS T 3011 potato crisps) was obtained from CSL (Central

20 Science Laboratory, York, UK) to validate the present method in the proficiency test. In the present

21 study, a sample blank using suitable food matrix was prepared to estimate the recovery percentage.

22 In order to ensure reliable analytical results, the simulant food was analyzed separately to make

Page 4 of 15
 1 sure that they were free from any interfering contaminant.

3 2.3. Sample preparation

4 The procedure of sample preparation was according to our previous study with some

5 modifications [16]. Briefly, potato crisps (2 g) were finely grinded and weighted into 50 mL

t
rip
6 centrifuge tubes. The samples were spiked with acrylamide standard (250, 500 and 1000 µg/kg) to

7 investigate the recovery percentage of the method at this stage. For the method proficiency test and

sc
8 sample quantification, samples were spiked with 500 µL of D3-labelled acrylamide solution (1

10
nu
µg/mL) as the internal standard instead of acrylamide standard. Then, samples were defatted with

petroleum ether, extracted with 2 mol/L aqueous solution of sodium chloride, further liquid-liquid
Ma
11 extracted with ethyl acetate and purified by solid-phase extraction (SPE) with OASIS HLB

12 cartridges (6 mL, 200 mg) purchased from Waters Technology (Milford, MA, USA). Aliquots of the

13 eluant obtained from SPE were transferred into amber glass autosampler vials for UPLC-MS/MS
d

14 analysis.
pte

15

16 2.4. UPLC-ESI-MS/MS analysis

ce

17 Quantification of acrylamide in pretreated products was performed on a UPLC-MS/MS with the

Ac

18 electrospray positive ionization (ESI+). In detail, an ACQUITY UPLC quaternary pump system

19 equipped with the micro vacuum degasser, thermostated autosampler and thermostated column

20 compartment (Waters, Milford, MA, USA) was coupled with a Micromass Quattro Ultima

21 triple-quadrupole mass spectrometer from Micromass Company Inc. (Manchester, UK). The

22 analyte elution (injection volume: 10 µL) was carried out on a UPLC BEH C18 column (50 mm

Page 5 of 15
 1 length, 1.5 mm i.d., 1.7 µm particle size; Waters, Milford, MA, USA) maintained at 25 °C with a run

2 time of 3 min. The mobile phase was 10% methanol/ 0.1% formic acid in water with a flow rate of

3 0.2 mL/min. The conditions used for the electrospray source were as follows: capillary voltage, 3.5

4 kV; cone voltage, 50 V; source temperature, 100 °C; desolvation gas temperature, 350 °C;

5 desolvation gas flow, 400 L/h nitrogen; cone gas flow, 45 L/h nitrogen; and argon collision gas

t
rip
6 pressure to 3 × 10-3 mbar for MS/MS, which gave a highest acrylamide response in this study. The

7 collision energy (CE) was optimized for each multiple reaction monitored (MRM) transition. The

sc
8 collision energy for each monitored transition was optimized in MRM mode.

10 3. Results and discussion

nu
Ma
11 3.1. Comparison between UPLC-MS/MS and HPLC-MS/MS

12 Actually, a HPLC-MS/MS analytical method for the quantification of acrylamide in various

13 foods was validated in our previous study [16]. In the present work, this method was improved and
d

14 especially a higher sensitivity and better chromatographic efficiency observed from the
pte

15 UPLC-MS/MS methods were obtained compared to the HPLC-MS/MS analysis (Fig. 1). As a

16 whole, the UPLC based method could offer significant improvements in sensitivity, analytical speed
ce

17 and resolution, and was successfully applied in the acrylamide quantification. Considering the need
Ac

18 to push limits of quantification to be lower and lower levels in shorter and shorter time become more

19 critical to the conduct of optimization investigations, the performance of both chromatographic and

20 mass spectral components of LC-MS systems becomes more and more critical. The goal of

21 increasing overall performance of these techniques can be further enhanced by reducing constraints

22 imposed by the chromatographic separation [14]. It was approved from the chromatogram that the

Page 6 of 15
 1 hybrid particles used in UPLC columns often showed unique selectivity when compared to

2 conventional HPLC packings.

4 3.2. Optimization of UPLC columns and flow rate

5 Acrylamide is acknowledged as a strong polar molecule with poor retention and terrible peak

t
rip
6 shape in routine non-polar or weak polar LC columns. Most chromatographers have experienced

7 problems retaining and separating polar compounds, just as acrylamide, using conventional

sc
8 reversed-phase chromatography. Atlantis dC18 columns (Waters, Milford, MA, USA), which are a

10
nu
silica-based line of difunctionally bonded C18 columns that provide the optimal balance of retention

for polar and non-polar compounds in reversed-phase chromatography, are designed for these types
Ma
11 of challenging separations. In previous work, the Atlantis dC18 column (210 × 1.5 mm, 5 µm) was

12 successfully used for the acrylamide elution and the above problems were obviously mitigated

13 [16,17]. However, the analyte peak was still a bit asymmetric and tailing so that the quantification of
d

14 peak area was difficultly performed (Fig. 2C). Therefore, a UPLC BEH C18 column with the particle
pte

15 size of 1.7 µm was attempted to use in this study. The predicament of tailing peak was significantly

16 improved and the half-peak breath was consequently reduced (Fig. 2). Meanwhile, the comparison
ce

17 of flow rate between 0.2 and 0.3 mL/min was also observed. The former flow rate was selected
Ac

18 considering the improvement of peak sensitivity. Alternatively, the flow rate 0.3 mL/min was too

19 fast to perform an enough retention time and the sensitivity of analyte was not obviously

20 ameliorated.

21

22 3.3. Method validation

Page 7 of 15
 1 3.3.1. Linearity, limit of detection (LOD) and limit of quantification (LOQ)

2 Linearity was estimated by investigating the detection signals as a function of analyte

3 concentration, with the aid of a regression line by the method of least squares. The calibration

4 curve for the determination of acrylamide in foods by UPLC-MS/MS analysis (y = 0.514 x +

t
0.367) was linear over the range of 1---200 ng/mL (see the detailed concentration levels in Section

rip
6 0.9999 (n = 3). Some prepared samples
2
2.1.). The correlation coefficient (r ) obtained was •0.9999

7 containing relatively high levels of acrylamide were diluted before injection in order to match the

sc
8 linear range of calibration. The LOD of the quantitative analysis indicates the lowest level of the

10
nu
analyte that can be measured with definable statistical certainty in a sample, and equivalents to three

times the standard deviation of noise on analysis, while the LOQ is calculated from the
Ma
11 concentration of the analytes that provided signals equal to 10 times the signal to noise on analysis.

12 The LOD and LOQ evaluated by the statistical software of MassLynx v4.0 (Micromass, Manchester,

13 Lancashire, UK) was calculated as 1 and 3 µg/kg, which adequately satisfied the trace quantification
d

14 of acrylamide in foods by the UPLC method.

pte

15

16 3.3.2. Precision
ce

17 The precision of this method was evaluated by the within-day and day-to-day (continuous 10 days)
Ac

18 determination of acrylamide. Excellent within-day precision data (n = 5) and day-to-day precision

19 data (n = 10) were obtained based on the analysis of five potato crisps spiked with 50, 100, 250, 500

20 and 1000 µg/kg of acrylamide standard respectively. Results showed that the RSD ranged from

21 0.6% to 4.5% for the day-to-day precision tests, and 0.4% to 3.2% for the within-day precision tests.

22 Results of the within-day and day-to-day precision data indicated good reproducibility for the

Page 8 of 15
 1 UPLC-MS/MS analysis of acrylamide.

3 3.3.3. Spiked recovery

4 Recovery of the method was investigated in three tests employing the method of standard

5 addition. The matrix was an acrylamide-free cereal sample supplied by Chinese Center for Disease

t
rip
6 Control and Prevention (Beijing, China). Five different levels of acrylamide standard (50, 100, 250,

7 500 and 1000 µg/kg) were added to this food matrix, which were used for the estimation of sample

sc
8 recovery in the low, intermediate and high spiked levels. The average recoveries in three spiked

10
nu
levels were shown in Table 1. Excellent percentage recoveries (79.6%–99.0%) were obtained with

acceptable variation (RSD%: 1.3%–4.2%) according to the present method.

Ma
11

12 3.3.4. Proficiency test

13 The analysis was integrated within the scope of an authorized proficiency test controlled by the
d

14 official Food Analysis Performance Assessment Scheme (FAPAS) for accreditation. In our previous
pte

15 study, the HPLC-MS/MS method was successfully validated in-house with repeated (n = 6) analysis

16 of a crispbread sample used in a FAPAS ring trial (Central Science Laboratory, UK; Series 30
ce

17 Round 11, T3011, robust mean value 1404 µg/kg). In this proficiency test, the result from our
Ac

18 laboratory (No. 021) for acrylamide (1381 µg/kg) in dispatched test material with a Z-score of –0.1

19 seemed satisfactory, which fulfill requirements from the organization [18]. In the proficiency test of

20 this study, the acrylamide content in the same reference test material was repeatedly quantified by

21 the present UPLC-MS/MS method. Results showed that the acrylamide content in the test material

22 (crispbread sample) was determined as 1399 µg/kg (n = 6) with a Z-score of –0.1. Such result

Page 9 of 15
 1 demonstrated the accuracy of the UPLC-MS/MS method used for the quantification of acrylamide.

3 4. Conclusion

4 An improved UPLC-MS/MS method for the rapid quantification of acrylamide has been

5 validated to be a fast and suitable method compared to previous HPLC-MS/MS method. This

t
rip
6 improved method supplies a rapid quantitative procedure of acrylamide with a run time of only 3

7 min and shows high sensitivity, good linearity over a wide range and good reproducibility. Since this

sc
8 method was not tested in routine operation, the optimization work including pretreatment

10
nu
procedures, recipes on the interference of unforeseen impurities, clean-up steps and instrumental

parameters still should be performed in various food products. Further studies will focus on the trial
Ma
11 of trace analysis and validation of the present method in some difficult food matrixes such as coffee.

12

13 Acknowledgements
d

14 This study was financially supported by National Natural Science Foundation Council of China
pte

15 (grant No. 30540016).

16
ce

17 References
Ac

18 [1] D.S. Mottram, B.L. Wedzicha, A.T. Dodson, Nature 419 (2002) 448.

19 [2] R.H. Stadler, I. Blank, N. Varga, F. Robert, J. Hau, P.A. Guy, M.-C. Robert, S. Riediker,

20 Nature 419 (2002) 449.

21 [3] E. Tareke, P. Rydberg, S. Karlsson, M. Eriksson, M. Törnqvist, J. Agric. Food Chem. 50

22 (2002) 4998.

10

Page 10 of 15
1 [4] T. Wenzl, M.B. de la Calle, E. Anklam, Food Addit. Contam. 20 (2003) 885.

2 [5] R.H. Stadler, G. Scholz, Nutr. Rev. 62 (2004) 449.

3 [6] Y. Zhang, G.Y. Zhang, Y. Zhang, J. Chromatogr. A 1075 (2005) 1.

4 [7] V. Gökmen, H.Z. Şenyuva, J. Acar, K. Sarıoğlu, J. Chromatogr. A 1088 (2005) 193.

5 [8] E.K. Paleologos, M.G. Kontominas, J. Chromatogr. A 1077 (2005) 128.

t
rip
6 [9] Y. Zhang, Y. Dong, Y.P. Ren, Y. Zhang, J. Chromatogr. A 1116 (2006) 209.

7 [10] E. Bermudo, O. Núñez, L. Puignou, M.T. Galceran, J. Chromatogr. A 1120 (2006) 199.

sc
8 [11] I.G. Casella, M. Pierri, M. Contursi, J. Chromatogr. A 1107 (2006) 198.

10
nu
[12] V. Gökmen, H.Z. Şenyuva, J. Chromatogr. A 1120 (2006) 194.

[13] K. Mastovska, S.J. Lehotay, J. Agric. Food Chem. 54 (2006) 7001.

Ma
11 [14] M.I. Churchwell, N.C. Twaddle, L.R. Meeker, D.R. Doerge, J. Chromatogr. B 825 (2005)

12 134.

13 [15] O.Y. Al-Dirbashi, H.Y. Aboul-Enein, M. Jacob, K. Al-Qahtani, M.S. Rashed, Anal. Bioanal.
d

14 Chem. 385 (2006) 1439.

pte

15 [16] Y. Zhang, J.J. Jiao, Y.P. Ren, X.Q. Wu, Y. Zhang, Anal. Chim. Acta 551 (2005) 150.

16 [17] H. Ono, Y. Chuda, M. Ohnishi-Kameyama, H. Yada, M. Ishizaka, H. Kobayashi, M. Yoshida,

ce

17 Food Addit. Contam. 20 (2003) 215.

Ac

18 [18] FAPAS Acrylamide Report 3011, 2005, FAPAS proficiency testing, April,

19 http://www.fapas.com/ fapas.cfm.

20

11

Page 11 of 15
 1 Figure Captions

3 Fig. 1. Comparison study on the determination of acrylamide by (A) UPLC-MS/MS and (B)

4 HPLC-MS/MS. The chromatographic conditions of UPLC-MS/MS analysis were described in

5 Section 2.4. The chromatographic conditions of HPLC-MS/MS analysis were according to our

t
rip
6 previous study [15].

sc
8 Fig. 2. Chromatograms performed on different LC columns and flow rates: (A) ACQUITY UPLC

10
nu
BEH C18 column (50 × 1.5 mm, 1.7 µm), 0.3 mL/min; (B) (A) ACQUITY UPLC BEH C18 column

(50 × 1.5 mm, 1.7 µm), 0.2 mL/min; (C) Atlantis dC18 column (210 × 1.5 mm, 5 µm), 0.2 mL/min.
Ma
11
d
pte
ce
Ac

12

Page 12 of 15
1 Table 1

2 Spiked recovery of the UPLC-MS/MS method (n = 3)a

Spiked level (µg/kg) Detected level (µg/kg) Recovery (%) RSD (%)

50 40.8 81.6 4.0

100 83.4 83.4 2.9

t
250 218.8 87.5 3.4

rip
500 494.8 99.0 4.2

1000 915.6 91.6 1.3

sc
a
3 The food matrix used in the recovery test was an acrylamide-free cereal sample supplied by

4 Chinese Center for Disease Control and Prevention.

5 nu
d Ma
pte
ce
Ac

13

Page 13 of 15
Figure 1

pt
cri
us
an
dM
pte
ce
Ac

Page 14 of 15
Figure 2

pt
cri
us
an
dM
pte
ce
Ac

Page 15 of 15