Simultaneous Determination of Melamine

and Cyanuric Acid Using LC-MS with the

Acclaim Mixed-Mode WAX-1 Column
and Mass Spectrometric Detection
Leo Wang, Stacy M. Henday, Xiaodong Liu, Mark Tracy, and William C. Schnute, Dionex Corporation, Sunnyvale, CA, USA

ABSTRACT INSTRUMENT
Recent investigations of pet animal death and health problems have A Dionex Summit® HPLC consisting of:
revealed pet food contaminated by melamine and cyanuric acid.1 When P680 Pump
present together, melamine and cyanuric acid form an insoluble crystal
ASI-100 Autosampler
matrix that may cause kidney function failure.2 In addition, since con-
taminated wheat gluten, rice protein concentrate, and corn gluten used TCC-100 Column Oven
in animal feed can be also used in human food such as bread, pasta, UVD-340 UV-Vis Detector
baby food, etc., it is crucial to monitor for the presence of melamine and
cyanuric acid in raw materials as well as in suspicious animal tissue. A Dionex MSQ™ Plus single quadruple mass spectrometer with
electrospray ionization (ESI) source
Current methods for quantitative determination of melamine and cyanuric
acid include gas chromatography mass spectrometry (GC-MS) and Chromeleon® 6.8 Chromatography Management Software
liquid chromatography mass spectrometry (LC-MS). GC-MS requires
derivatization which is labor intensive and the reported LC-MS meth-
ods generally involve a long gradient chromatographic run as well as Chromatographic Conditions
column clean up. Analytical Column: Acclaim® Mixed-Mode WAX-1
(150 × 2.1 mm, 5 µm)
This presentation introduces a sensitive, simple, and high-throughput
method for simultaneous determination of melamine and cyanuric acid Flow Rate: Isocratic 90% acetonitrile (CH3CN) /
by LC-MS and uses stable isotope labeled internal standard (ISTD) for 10% 20 mM pH 4 ammonium acetate buffer (v/v)
quantification. at 0.25 mL/min
Column Temperature: 20 °C
Analytes were retained and separated on a recently developed Injection Volume: 5 µL
Mixed-Mode WAX-1 column3 which demonstrates the unique
selectivity and retention offered by mixed hydrophobic and ion exchange
retention mechanisms. Chromatographic run time was significantly MSQ Conditions
reduced to 8 min. and MS detection was applied to ensure selectivity
and sensitivity. Excellent linearities with R2 > 0.9995 through a range of Analysis Mode: Selected Ion Monitoring (SIM)
2–200 ng/mL were achieved for both analytes. Method detection limits with polarity switching
(MDL) were estimated to be 3.97 ng/mL for melamine and 3.32 ng/mL Cone Voltage: 50 V for all scans
for cyanuric acid. Dwell Time: 0.5 s for all SIM channels
Probe Temperature: 500 °C
The reliability of this method was evaluated against pet food extracts as
well as biological matrices: pork and fish tissue extractions. Pet food Scan Events: Two full scans (100–400 m/z in 250 ms)
samples were prepared by a fast, simple, and automated extraction pro- with different polarities were applied throughout
cess (results will be presented elsewhere). And biological samples were the entire analytical run time with 4 SIM
obtained from the U.S. FDA and prepared by its current protocol. scans (500 ms each) for analytes and internal
standards (ISTD), shown in Table 1.

Montreux Symposium 2007 Presentation

Montreux Symposium 2007 Presentation 1
 Table 1. Scan Functions and Scan Events
Name Start Mass m/z End Mass m/z Time Range (min) Scan Time (milliseconds) Polarity Cone Voltage (V)

Positive Full Scan 100 400 0~8 250 pos. 50

Negative Full Scan 100 400 0~8 250 neg. 50

Name Mass Span Time Range (min) Dwell Time (milliseconds) Polarity Cone Voltage (V)
m/z m/z

SIM GROUP 1
Melamine 127 0.5 0~5 500 pos. 50
Melamine-15N3 130 0.5 0~5 500 pos. 50
SIM GROUP 2
Cyanuric Acid 128 0.5 5~8 500 neg. 50
Cyanuric Acid- C3 13
131 0.5 5~8 500 neg. 50

Structures of melamine and cyanuric acid are shown in Figure 1, along METHOD DEVELOPMENT
with their full scan spectra.
Chromatographic Optimization
Mobile phase composition: It is well understood that a mobile phase
108 N NH2
with higher organic content tends to retain polar analytes longer
127.1 H2N
under HILIC mode, and shows higher MS response under ESI condi-
N N tion. However, separation on a WAX-1 column includes weak anion
Melamine NH2 exchange, which is sensitive to the change of mobile phase in terms
%
of ionic strength and pH.3 To maintain method reproducibility, proper
168.1
buffer capacity is required so a mobile phase with 10% pH 4 NH4OAc
buffer in acetonitrile (v/v) was selected.
Relative Abundance

139.1
239.6 344.4
0
100 150 200 250 300 350 400 Buffer pH
108 m/z
128.2
OH O
When melamine and cyanuric acid are present in different moieties
H H
N N N N (organic, ionic or a mixture) under different pH conditions, longer
HO OH O N O
retentions were observed when each analyte was switched to a more
N
%
Cyanuric Acid H hydrophilic moiety (ionic form), which can be explained by an ionic
exchange mechanism. The effect of pH on retention was investigated
279.1
and the result is shown in Figure 2.
210.2
118.2 165.4 183.2 233.4 294.4 361.0
0
100 150 200 250 300 350 400
m/z
24476

Figure 1. Full scan spectra of melamine and cyanuric acid.

2 Simultaneous Determination of Melamine and Cyanuric Acid Using LC-MS with the
Acclaim Mixed-Mode WAX-1 Column and Mass Spectrometric Detection
 Column: Mixed-Mode WAX-1, 5 µm Melamine Column: Mixed-Mode WAX-1, 5 µm
Melamine - 1.62 600,000
Dimensions: 2.1 × 150 mm Dimensions: 2.1 × 150 mm
Mobile Phase: A = CH3CN, Mobile Phase: A = CH3CN, B = H2O,
B = 55 mM NH4OAc buffer C = 55 mM NH4OAc buffer pH 5.2
Condition 1 (Pink Traces): A:B (pH 6) = 90:10 Condition 1 (Pink Traces): A:B:C (11 mM) = 80:00:20
pH 4 Condition 2 (Blue Traces): A:B (pH 5) = 90:10 Condition 2 (Blue Traces): A:B:C (8.25 mM) = 80:05:15
Condition 3 (Purple Traces): A:B (pH 4) = 90:10 Condition 3 (Purple Traces): A:B:C (5.5 mM) = 80:10:10
Flow Rate: 0. 50 mL/min Condition 4 (Green Traces): A:B:C (2.75 mM) = 80:15:05

Intensity (Counts)
Column Temperature: 30 °C Flow Rate: 0. 25 mL/min
Melamine - 1.47 Inj. Volume: 10 µL Column Temperature: 30 °C
Detection: MSQ Plus Inj. Volume: 10 µL
pH 5 Melamine: pos. ESI SIM m/z = 127 Detection: MSQ Plus
Cyanuric Acid: neg. ESI SIM m/z = 128 Melamine: pos. ESI SIM m/z = 127
Cyanuric Acid: neg. ESI SIM m/z = 128
Melamine - 1.47

pH 6

0 1 2 3 4 5 6 7 0 3 6 9 12 15
20,000
12.42

Cyanuric Acid - 3.34

Cyanuric Acid - 5.40
Cyanuric Acid
pH 4

Intensity (Counts)
9.68

(NH4OAc) = 2.75 mM 8.27

pH 5
(NH4OAc) = 5.5 mM 7.40
Cyanuric Acid - 5.50
(NH4OAc) = 8.25 mM

(NH4OAc) = 11 mM

pH 6

0 3 6 9 12 15
0 1 2 3 4 5 6 7 Retention Time (Minutes)
Retention Time (Minutes)
24478
24477

Figure 2. Effect of mobile phase pH on retention time. Figure 3. Effect of buffer ionic strength on retention time.

Ionic Strength Mass Spectrometric Optimization

A WAX-1 mixed mode column offers a unique selectivity for polar
analytes due to the weak anionic exchange mechanism. An analyte
with greater polarity is expected to demonstrate greater dependence on
mobile phase ionic strength. As shown in Figure 3, the retention time
of cyanuric acid changed from 12.42 min to 7.40 min by increasing the
mobile phase total ionic strength from 2.75 mM to 11.0 mM.
Mass spectrometric and chromatographic conditions were optimized in
combination. The chromatograms of a calibration standard are shown
in Figure 4 with the obtained optimization of resolution, sensitivity, and
throughput.

The combination of the three chromatographic factors offers greater

flexibility and control of the range of retention time for cyanuric acid,
shown in this study to be from approximately 6 to over 15 min.

Montreux Symposium 2007 Presentation 3

 60,000 SIM_01 126.75–127.25 m/z Melamine - 3.80 Cyanuric Acid Internal 'Cyanuric Acid-13C3'
800
Cyanuric acid: 11.1 ng/mL to 222 ng/mL
Cyanuric acid -13C3: 53.0 ng/mL

625 R2 = 0.9997
0
70,000 SIM_02 129.75–130.25 m/z
500

Area (% ISTD)
Intensity (Counts)

Melamine -15N3 - 3.80 375

0
4,000 SIM_03 127.75–128.25 m/z 250
Cyanuric acid - 5.97

125

0
1,600 0
SIM_04 130.75–131.25 m/z Cyanuric acid - 13C3 - 5.97 0 50 100 150 200 250
Amount (ng/mL)
24481

600 Figure 6. Calibration curve, Cyanuric acid.

0.02 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
Retention Time (Minutes)
24479

Figure 4. SIM chromatograms of a calibration standard under optimized Method Detection Limit (MDL)
conditions. Method detection limits were calculated by seven replicate injections of
calibration standards and calculated by the following equation:

METHOD VALIDATION MDL = t99% × S(n = 7)

Linearity and Calibration Curve

Where t is student’s t at 99% confidence intervals (t99%, n = 7 = 3.143) and
Calibration standards were obtained from the U. S. FDA with stable
S is the standard deviation. The MDLs of melamine and cyanuric acid
isotope labeled analytes (Melamine-15N3 and Cyanuric Acid-13C3) as
were calculated to be 3.97 ng/mL and 3.92 ng/mL, respectively.
internal standards (ITSD). These standards were used to generate
calibration curves and demonstrate linearity. Excellent linearity was
achieved throughout the range from 2 ng/mL to 200 ng/mL, as shown in Method Evaluation by Biological Samples
Figures 5 and 6. Chromatogram
Pork and fish tissue extracts (control and spiked samples) were
15
obtained from the U.S. FDA and prepared using current FDA protocol.
Melamine Internal 'Melamine- N3'
1,100 All biological samples were evaporated to dryness and reconstituted
Melamine: 2.34 ng/mL to 234 ng/mL
Melamine -15N3: 22.6 ng/mL
in mobile phase. Figure 7 shows the SIM chromatograms of a control
R2 = 0.9999 sample (extracted blank pork sample, ISTD spiked before extraction; and
800 spiked with melamine (melamine portion) or cyanuric acid (cyanuric
acid portion); and then evaporated to dryness and reconstituted in
Area (% ISTD)

600 mobile phase).

400

200

0
0 50 100 150 200 250
Amount (ng/mL)
24480

Figure 5. Calibration curve, Melamine.

4 Simultaneous Determination of Melamine and Cyanuric Acid Using LC-MS with the
Acclaim Mixed-Mode WAX-1 Column and Mass Spectrometric Detection
 80,000 SIM_01 126.75–127.25 m/z Melamine - 3.80 Table 2. Method Reproducibility
Retention Time Observed Amount
Intensity (Counts)

Analyte
Mean (minutes) % RSD Mean (ng/mL) %RSD
Melamine 3.80 0.00 61.2 1.38
Cyanuric Acid 5.64 0.74 141 1.02
0
18,000
SIM_02 129.75–130.25 m/z
Melamine -15N3 - 3.80
Intensity (Counts)

CONCLUSION
• A simple and fast LC-MS method has been developed for simultane-
ous determination of melamine and cyanuric acid.
0
8,000 SIM_03 127.75–128.25 m/z
• Melamine and cyanuric acid can be retained and resolved with satis-
Cyanuric acid - 5.68
factory retention time and total analytical cycle time.
Intensity (Counts)

• Reproducibility has been demonstrated.

• MDLs of <5 ng/mL were achieved by mass spectrometric detection.
• Excellent linearity was achieved with R2>0.9995 within 2 to
200 ng/mL.
2,000 SIM_04 130.75–131.25 m/z Cyanuric acid -13C3 - 5.68
• This method has been applied to the analyses of biological samples
Intensity (Counts)

prepared by current U.S. FDA protocols.

600 ACKNOWLEDGEMENTS
0 1 2 3 4 5 6 7 8
Retention Time (Minutes) The authors would like to thank Alex Krynitsky, of the U. S. FDA, for
24482
preparing and providing standards and samples for method evaluation.
Figure 7. SIM chromatograms of FDA control samples after evaporation and
reconstitution.
REFERENCES
Reproducibility 1. Pet Food Recall/Tainted Animal Feed, U.S. FDA, May 31, 2007.
Chromatography may become distorted after repeated injections and 2. Melamine and Cyanuric Acid Interaction May Play Part in Illness and
interferences may arise from late eluting unknown compounds from pre- Death from Recalled Pet Food, AVMA, May 1, 2007.
vious injections. Since these interferences may require extra measures, 3. Dionex Corporation, Acclaim Mixed-Mode WAX-1 Column,
such as gradient long runs or column clean up, it is important to evalu- LPN 1899, February 2007.
ate method reproducibility. In this study, method reproducibility was
evaluated by repeated injections of melamine and cyanuric acid control
samples spiked respectively with the relevant analyte. The result is
shown in Table 2. Excellent reproducibility was observed for melamine,
both for retention time and observed amount. A slight variation was ob-
served for cyanuric acid retention time, which may be explained by the
presence of formic acid residue from the sample preparation process.

MSQ is a trademark of Thermo Fisher Scientific, Inc.

Acclaim, Chromeleon, and Summit are registered trademarks of Dionex Corporation.

Passion. Power. Productivity.

Dionex Corporation North America Europe Asia Pacific
1228 Titan Way U.S. (847) 295-7500 Austria (43) 1 616 51 25 Benelux (31) 20 683 9768 (32) 3 353 4294 Australia (61) 2 9420 5233 China (852) 2428 3282 India (91) 22 28475235
P.O. Box 3603 Canada (905) 844-9650 Denmark (45) 36 36 90 90 France (33) 1 39 30 01 10 Germany (49) 6126 991 0 Japan (81) 6 6885 1213 Korea (82) 2 2653 2580 Singapore (65) 6289 1190

Sunnyvale, CA
94088-3603
South America
Ireland (353) 1 885 1700 Italy (39) 02 51 62 1267 Switzerland (41) 62 205 9966
United Kingdom (44) 1276 691722
Montreux Symposium 2007 Presentation
Taiwan (886) 2 8751 6655
5
LPN 1991-01 10/07

(408) 737-0700
Brazil (55) 11 3731 5140 www.dionex.com ©2007 Dionex Corporation