Although food and drug packaging is often printed with colorful labels, there is limited knowledge about the migration of printed ink components into the products. With no global legislation available, and ill-defined terminology in the scientific community, formulating ink used in packaging becomes a challenge for suppliers and users who are concerned with potential migration of harmful components. Possible sources of migratable compounds in inks and coatings include UV photo-initiators, mineral oils, and resins. No single analytical method can detect all components in the inks due to the complexity of the compositions and solubility of individual components.

The aim of our study is to apply the most effective methodologies, using minimal extraction combined with instrumental analysis, to determine the amount of ink components potentially migrating into the contact adhesive layer of a transdermal patch with printing directly on it. Full details of the study are presented in a poster from SGS Life Science Services entitled ‘Migration of ink components into transdermal patches’.

The study was initiated by the extractables and leachables (E&L) studies group after the increase in requests to perform analysis on transdermal patches over recent years. Many clients are unaware how to satisfy the FDA with regards to E&L data. Therefore, the goals of the study were to develop and validate methodology that can provide an extractable profile by quantitative analysis. This work also involves the experimental strategy for the extraction and investigation of the potential migration of ink from backing print to adhesive layer under aggressive and exaggerated conditions.


Extractables testing involves monitoring compounds, some of which are volatile, and some of which are metals, so different instruments and methodologies are necessary to quantify each component. For volatile organic compounds (VOCs), the most appropriate technique is gas chromatography mass spectroscopy (GC-MS), for non-volatile compounds (NVOCs), liquid chromatography mass spectroscopy (LC-MS) is used, and for metals the method of choice is inductively-coupled plasma optical emission spectroscopy (ICP-OES).


One challenging aspect of the whole procedure was sample isolation as it was only necessary to analyze that portion of the patch that contacts the skin, and not the outside of the patch. To secure the right part of the patch, the fragile film needed to be removed from the remainder of the delivery patch and a small piece obtained from which a sample could be extracted for analysis.

The portion of the patch needing to be analyzed consists of a polymer matrix, presenting an added complication of background interference in the analysis. In the MS studies, considerable time was spent comparing the sample to a blank, locating the specific mass values of interest within the mass spectrum against the background noise, and determining their values in the sample.

Additionally, there are specific challenges involved in the testing of transdermal patches that need to be addressed. Firstly, we needed to take into account that sample matrices co-elute with target VOCs. Additionally, the methodology employed needed to take into account that the polyamide patch material and the photoinitiators employed and their decomposition products do not contain a chromophore since conventional liquid chromatogrphay with ultra-violet detection cannot be used to detect polyamide. Finally, no direct method is in practice for quantifying color dye.

Consequently, a key criterion in the choice of tests undertaken was the sensitivity of the methods. Typical analytical sensitivities required by the FDA are approximately 1-2 ppm, which can be challenging when analyzing a matrix. However, these values are above the limits of detection of the spectroscopic methods used, which are 0.2, 0.1 and 0.1 ppm for GC-MS, LC-MS and ICP-OES, respectively.

Sample Storage and Preparation

Unprinted patches were stored at controlled room temperature as negative control, whereas spiked recovery study and printed patches were stored at 40°C/75% RH for 6 months.

Three adhesive layers of each test article were separated from the backing films of the patches and transferred into the same scintillation vial, with 20 ml of methanol for solvent extraction. Then, the scintillation vial was capped and sonicated for 20 minutes. The extraction solvent was used ‘as is’ for LC-MS and GC-MS instrumental analysis, while the extraction solvent was digested with concentrated sulfuric acid and nitric acid prior to ICP-OES testing.


GC-MS was used to quantify benzyl alcohol due to sample matrices interference, and gas chromatography with flame ionization detection/mass spectrometry (GC-FID/MS) was used to quantify isopropyl alcohol. The detection limit for the method is 0.2 μg/ml and the spiked recovery is 102% for benzyl alcohol. (Table 1, Figure 1 and 2)

Ultra performance liquid chromatography with photodiode array detection/mass spectrometry (UPLC-DAD/MS) was used to quantify polyamide due to lack of chomophores. The limit of detection (LOD) and limit of quantification (LOQ) for the method is 0.1 and 0.3 μg/ml, respectively, and the spiked recovery is 89% for polyamide.
(Table 2 and Figure 3)

ICP-OES was used to quantify the calcium (Ca) element as represented for Red color dye. A known amount (~ 100 μg) of D&C Red color dye was prepared in concentrated nitric acid and sulfuric acid. The percentage recovery was 94% for Ca (D&C Red color dye). Since D&C Red color dye contains calcium (Ca) element, the result demonstrated that the stoichiometric ratio between calcium (Ca) element and the D&C Red color dye compound is 1:1. The limit of detection (LOD) and limit of quantification (LOQ) for the method is 0.1 and 0.3 μg/ml, respectively. The spiked recovery was 97% for the calcium (Ca) element. (Figure 3)


In the study, the calibration curves showed linearity in the range of 0.6-5.0 μg/ml for all GC-FID/MS, UPLC-DAD/MS and ICP-OES runs with t2 no less than 0.998 for all three instrumental analyses. The limits of detection and the recoveries of individual ink components from extraction samples were 0.2 μg/ml and 89-102%, respectively. These values indicate that the proposed methods would be useful for the quantification of ink migration from the transdermal patch.

The work undertaken demonstrates an efficient technique for analyzing extractable components from transdermal patches for instrumental analysis, while also developing sensitivity methods for GC-FID/MS, UPLC-DAD/MS and ICP-OES for analysis of benzyl alcohol and isopropyl alcohol, polyamide and metallic elements (Ca), respectively.

By evaluating each sample, and applying the best and most appropriate techniques and methodologies, the team is able to optimize instrumentation set up, sample preparation and analytical procedures; this ensures that samples are as fully characterized as possible, to meet the ever more stringent extractables & leachables analysis requirements of the FDA in all types of drug and medical products.


Kenneth Wong, Xinjie Song, Gayatri Trevedi and Theresa Burchfeld
SGS Life Science Services
Fairfield, NJ