Improved Recovery of Fungicide Carbendazim from Orange Juice

By: Olga Shimelis and Emily Barrey, Reporter US Volume 30.2

A QuEChERS sample preparation method followed by LC-MS/MS


The fungicide carbendazim can sometimes be used to combat mold on citrus fruits including oranges. Recently, low residue levels of the fungicide were reported to be present in bulk orange juice import shipments to the USA. Because this compound can persist through processing steps and may be found in the final consumer product, many government agencies have established maximum residue limits (MRLs) to ensure the safety of their population. For example, the carbendazim MRL in orange juice is 10 ppb in the USA, and 200 ppb in the European Union1-2.

Final consumer products include juice that is pulp-free, juice with pulp, or frozen concentrate. Each of these matrices contains a different amount of solids, which may interfere with sample preparation and/or analysis. For this work, we evaluated several sample preparation methods to determine if a single method could 1) achieve acceptable recoveries, and 2) reduce interference, for each matrix. These goals are desirable as they would allow food analysts to process various matrices in a single workflow.


Pulp-free orange juice, orange juice with pulp, and frozen concentrate samples were purchased from a local grocery store. Each matrix was subjected to three sample preparation methods prior to LC-MS/MS analysis. A comparison of the three methods is presented in Table 1. Method 1 is simple physical manipulation (centrifuge and filter), and is not suitable for the frozen concentrate sample. Method 2 is a solvent extraction that uses salts to drive analytes into the organic layer. Method 3 incorporates a QuEChERS cleanup following solvent extraction.

Table 1. Comparison of Methods

Method 1 (Filter and Shoot)
  1. Add 1 mL of orange juice to a 2 mL centrifuge tube
  2. Centrifuge at 10,000 rpm for 5 min
  3. Filter supernatant through a 0.45 µm PVDF filter
  4. Analyze via LC-MS/MS


Method 2 (Acetonitrile Extraction)
  1. Add 10 mL of orange juice, or 10 g of orange concentrate, to a 50 mL empty extraction tube (Product No. 55248-U)
  2. Add 10 mL acetonitrile, and mix well for 1 min
  3. Add contents of citrate extraction tube (Product No. 55227-U), shake well, and mix for 1 min
  4. Centrifuge at 3,200 rpm for 5 min
  5. Dilute the acetonitrile layer with water 1:1
  6. Analyze via LC-MS/MS


Method 3 (Acetonitrile Extraction + QuEChERS Cleanup)
  1. Follow Method 2 through step 4
  2. Mix 0.7 mL of the acetonitrile layer with contents of the PSA/C18 cleanup tube (Product No. 55288-U)
  3. Recover 0.25 mL of the supernatant and mix with 0.25 mL water
  4. Analyze via LC-MS/MS


The Quick-Easy-Cheap-Effective-Rugged-Safe (QuEChERS) cleanup uses bulk materials instead of traditional tube-based hardware, which simplifies the procedure3. The main sorbent used is a primary-secondary amine (PSA) which provides high capacity for the removal of sugars, organic and fatty acids, and polar pigments. Depending on the sample, C18 silica sorbent can be added for the cleanup of fat-containing samples, and/or carbon sorbent can be added for the cleanup of highly pigmented samples. Based on the interferences expected in orange juice samples, PSA/C18 sorbent was selected.

Unspiked and replicate spiked samples were prepared for each matrix using each method. Following sample preparation, all calibration standards and extracts were analyzed using LC-MS/MS on a short, narrow-bore, Ascentis® Express C18 column. Two mass transitions were monitored to identify carbendazim (m/z 192/160 and 192/132). The abundant response of carbendazim allowed an instrumental limit of detection of 0.1 ppb to be achieved.

Results and Discussion

Figure 1 shows chromatograms of an unspiked and a spiked replicate of the juice with pulp sample following sample preparation using Method 3. The choice of an Ascentis Express column allowed efficient chromatography in a short time. In fact, a benefit of this column line is that great efficiency can be obtained on any system, regardless if used with HPLC or UHPLC equipment.

Carbendazim spiked (1 ppb) and unspiked

Figure 1. LC-MS/MS Analysis of Unspiked and Spiked Orange Juice with Pulp

HPLC Conditions
add 10 mL of orange juice with pulp to a 50 mL empty extraction tube (Product No. 55248-U); add carbendazim at 1 ppb to ‘spiked’ sample; add 10 mL acetonitrile; mix well for 1 min; add contents of citrate extraction tube (Product No. 55227-U); shake well; mix for 1 min; centrifuge at 3,200 rpm for 5 min; mix 0.7 mL of the acetonitrile layer with the contents of the PSA/C18 cleanup tube (Product No. 55288-U); recover 0.25 mL of the supernatant; mix with 0.25 mL water
column: Ascentis Express C18, 5 cm x 2.1 mm I.D., 2.7 µm particles (Product No. 53822-U); mobile phase: (A) 10 mM ammonium acetate in water; (B) 10 mM ammonium acetate in methanol; gradient: 0-1 min: 30% B; 1.5-3.5 min: 100% B; 3.5-7 min: 30% B; flow rate: 0-1 min: 0.3 mL/min; 1.5-7 min: 0.5 mL/min; temp.: 30 °C; detector: MS, ESI(+), MRM, m/z 192/160, 192/132; injection: 5 µL


Very low levels of carbendazim were detected in two of the three unspiked samples. Note that the levels observed are below the MRLs for both the USA and European Union. Average recovery and %RSD values were calculated and are summarized in Table 2, by matrix and method. As stated previously, Method 1 is unsuitable for a solid matrix, such as the frozen concentrate. Based on the data presented, Method 3 achieves the first goal of this work– to identify a single method which can achieve acceptable recoveries for each matrix.

Table 2. Average Recovery and %RSD (in parentheses) from 1 ppb Spiked Samples (n=3)

Matrix Method 1 Method 2 Method 3
Juice, Pulp-Free 25% (79) 49% (7) 73% (5)
Juice, With Pulp 45% (18) 64% (14) 73% (15)
Concentrate, Frozen n/a 36% (13) 72% (20)

The methods were also evaluated as to which best removed matrix interference. This was quantified by comparing actual and expected concentrations for spiked samples. The closer the actual measured concentrations are to expected values, the lower the ion-suppression. The observed ion suppression values are displayed in Figure 2. As shown, sample cleanup using Method 3 results in the least amount of ion-suppression and outperforms the others in this aspect.

Comparison of matrices by method

Figure 2. Ion-Suppression Values for Each Method


We presented a single method, involving fast and effective QuEChERS cleanup, suitable for various orange juice matrices. The choice of PSA/C18 sorbents reduced interferences, resulting in greater recoveries and less ion-suppression. Because the QuEChERS method is fast, it only adds a few minutes to the sample preparation procedure. Sample processing using QuEChERS with PSA/C18 allows analysts to accurately process various matrices in a single workflow, with very little additional sample preparation time.

Legal Information

Ascentis is a registered trademark of Sigma-Aldrich Co. LLC




  1. US FDA web site, (accessed 6-Mar-2012).
  2. EU Pesticide Database cfm, (accessed 6-Mar-2012).
  3. Lehotay, S. J. Quick, Easy, Cheap, Effective, Rugged and Safe (QuEChERS) Approach for the Determination of Pesticide Residues. Proceedings AOAC Annual Meeting, St. Louis, MO USA, 2004.
  4. EN15662:2008, Foods of plant origin - Determination of pesticide residues using GC-MS and/or LC-MS/MS following acetonitrile extraction/partitioning and cleanup by dispersive SPE - QuEChERS-method.
  5. AOAC Official Method 2007.01, Pesticide Residues in Foods by Acetonitrile Extraction and Partitioning with Magnesium Sulfate.


Related Links