![]() |
79 | Last Update: 12/23/99 |
1Rutgers University, Center for Advanced Food Technology, 63 Dudley Road, New Brunswick, NJ 08903 : 2Scientific Instrument Services, Inc., 1027 Old York Road, Ringoes, NJ 08551
Presented at PittaCon99 Meeting, Orlando, FL March, 1999
Abstract
Electron beam cured packaging samples were analyzed using an automated Short Path thermal desorption apparatus (AutoDesorb, Scientific Instrument Services, Inc., Ringoes, NJ), in conjunction with a gas chromatograph-mass spectrometer (Hewlett-Packard, Inc., Palo Alto CA). The samples were weighed and placed into custom-made glass tubes. The samples were then matrix spiked with internal standards. Next, the samples were purged of volatile products in a purge and trap sampling oven onto adsorbent resin-containing tubes (GLT tubes). The GLT tubes were then loaded into an automated Short Path thermal desorption apparatus for unattended introduction to the GC capillary column. The desorption temperature, purge gas flow rate, carrier gas flow rate, desorption temperature ramp time, and cryotrap cooling temperature are automatically controlled and can be independently defined for each sample. The volatile component profile is then identified by mass spectral analysis and quantified using internal standard methodology. Several organic components were identified including hydrocarbons from petroleum distillates, photoinitators from UV (ultraviolet) cure inks, plasticizers, and residual short chain monomers from the coating formulations. This series of analyses also serves to examine the effectiveness of the design, ruggedness, and performance of the newly developed AutoDesorb System.
INTRODUCTION
Lacquers, varnishes, and other clearcoats have traditionally been applied to various types of food packaging where they impart a glossy or semi glossy appearance. In addition to providing a topcoat, they also provide abrasion resistance and increased barrier properties. Due to their wide range of characteristics with regard to strength, barrier properties, and flexibility, their use has grown into a mainstay of the packaging industry. A major cost in manufacturing coatings, and one of the elements that has the greatest effect on quality is the cure process. The cure process is the portion of manufacture where the constituent monomers and oligomers are polymerized and crosslinked, and solvents are removed from the coating formulation.
Traditionally, the curing process has been carried out using thermal techniques. In this process, the preformed functional polymers are dissolved or dispersed in solvents. The solvent systems are either organic or water based. Organic solvent systems typically work more effectively, but have the undesirable effects of releasing high levels of VOC's into the environment and having a greater propensity for developing off odors. Oligomers, which facilitate the crosslinking process, are added along with pigments, wetting agents, flow agents, lubricants, plasticizers, polymerization aids and other compounds. The resultant formulation is then applied to a substrate and thermal energy is used to remove the solvent and to initiate the curing reaction processes. Finally, the excess heat is removed from the substrate. Thermal curing requires relatively long cure times and high energy consumption, creates residual thermal stresses, and requires expensive tooling.
An alternative to the thermal cure method is radiation curing. This technique has been in use since the 1930's. However, it was not until the late 1960's that commercial interest in the process became apparent. Radiation curing is a method of curing polymers with electron beam (EB) or ultraviolet (UV) radiation. These methods have the advantage of using solventless coating formulations, so there is no solvent removal step required. Also, curing time is on the order of seconds as opposed to hours for some thermal methods. To this date, little data exist on the VOC profile that is evolved from these coating formulations. Today, these formulations are in use on frozen dinner packaging, packaged meats, beverages, and other food and non-food applications.
The purpose of this paper is to quantify and identify volatile organic components from radiation cured packaging samples using an new automated Short Path thermal desorption GC accessory, AutoDesorb (Scientific Instrument Services, Inc.). This data can be used for quality control of the radiation cure process, determination of residual monomers, additives, and potential off-odor components. It can also be used to monitor migrants from the radiation cure process into foods, for toxicological assessment and studies regarding food contact law. Finally, the method and subsequent data set can be used to deformulate competitors' products.
Experimental Methods
Sample Preparation
Twelve samples of paperboard were clear coated with a proprietary acrylic formulation and donated by an anonymous manufacturer. From each sample a piece of 2 cm by 10 cm rectangular piece was cut. These were weighed, recorded, (approximately 400 mg each) and placed in a .25" outer diameter by 14" long glass tube and then plugged at either end with purified glass wool. The glass tube was then placed in a heated purge and trap sampling oven (Scientific Instrument Services) as shown in Figure 1. Stainless steel fittings were attached at either end of the tube to facilitate the attachment of the purge gas line at one end, and a glass lined stainless steel tube (GLT), packed with 100 mg of Tenax® TA was attached at the other end. The paperboard samples were matrix spiked with 1.0 ug of d-8 naphthalene in methanol using the solvent flush method, and were heated to 80° C for 30 minutes. During this time, a flow of 1.2 liters of helium was swept through the apparatus to sweep the released volatiles onto the GLT tubes. Additionally, eight samples of paperboard with printing and glossy clear coat were prepared as described above.
Instrumentation
The samples were thermally desorbed using the AutoDesorb System (Scientific Instrument Services, Ringoes, NJ) a novel automated Short Path thermal desorption apparatus (Figure 2). In the system, the GLT tubes that contain the sample analytes from the preparative steps above are attached to a connecting tube head and needle prior to analysis. These are then placed on the sample carousel. The connector tubes contain a ball seal that protects the sample from contamination before analysis. When analysis is ready to begin, the carousel turns to advance the first sample into place. Then a pneumatic actuator extends to lift the first sample out of the carousel. Simultaneously, the connector head docks with the pickup assembly to lock it in place and depress the ball seal. The assembly is then retracted directly in-line with the GC injection port. Next, a flow of carrier gas (80 ml) is initiated to purge the GLT tube of residual oxygen and solvent. The pneumatic piston then injects the desorption tube assembly into the GC injection port where the needle serves as the transfer line to the GC. A pair of bilaterally operating heater blocks enclose around the sample cartridge to provide rapid heat transfer to the sample. The combination of flow through the desorption tube and heat from the blocks sweep the analytes through the needle and into the GC injection port. These analytes are then trapped at the front of the GC column using a Micro Cryotrap (Scientific Instrument Services, Inc., Ringoes, NJ) which is used to cool the column to subambient temperatures. A short 20cm x .53mm diameter fused silica precolumn is attached to the Micro Cryotrap to focus the analytes in a narrow band. After the analytes have been trapped onto the front of the column and the desorption cycle is complete, the blocks open, the system uninjects the sample, and the pneumatic arm places the cartridge back onto the carousel. After the column pressure equilibrates, the temperature program begins and the cryotrap heats to the specified temperature. The AutoDesorb then advances the carousel to the next sample to be analyzed. The AutoDesorb system automatically integrates and controls all features of the system suc has: cryotrap cooling/heating, purge gas flow, thermal desorption temperature (and ramp rate if applicable), and thermal desorption time. All parameters are fully integrated with Hewlett-Packard ChemStation software so that data only needs to be entered once. The AutoDesorb logs all analysis results for error detection and reporting.
The AutoDesorb was used in conjunction with an HP 6890 gas chromatograph
and 5973 mass selective detector. The mass spectrometer was operated in
electron impact (EI) ionization mode and scanned from 35-350 Da during
the GC run for the total ion chromatogram. The total ion signal was integrated
using Hewlett-Packard ChemStation software, and each of the chromatogram
peaks was library searched utilizing the United States National Institute
for Standards and Technology (NIST) library to identify the organic compounds.
For those peaks with no library match, manual interpretation was done using
comparison to analytical reference standards, proprietary mass spectral
database, and GC retention time index. The samples were analyzed unattended,
under the following AutoDesorb and GC-MS conditions:
|
|
|
| Initial Desorption Temperature | 250° C |
| Final Desorption Temperature | 250° C |
| Desorption Ramp Rate | 0 |
| Desorption Time | 5 minutes |
| Sample prepurge | 2 minutes |
| Initial cryotrap temperature | -65° C |
| Final cryotrap temperature | 280° C |
|
|
|
| Column | HP 35ms, 60m x 250m m x .25 m m |
| GC column initial temperature | 50° C |
| GC column final temperature | 250° C |
| GC column ramp rate | 10° C/minute |
| Injection split ratio | 100:1 |
|
|
|
| MS Mode | EI |
| Mass Range | 35-350 |
| Scans/second | 2.36 |
Results and Discussion
A representative GC-MS chromatogram from each of the sample types is shown in Figures 3 and 4. The peak assignments and quantitation data corresponding to these chromatograms are summarized in Tables 2 and 3. The concentration data are given in the tables in units of parts-per-billion on a weight to weight basis (PPB w/w) and in units of ug/cm2. Manufacturer's specifications are often given in mg/ream of packaging where 1 ream equals 500 square feet. Concentration data on a weight/surface area basis can then be calculated and extrapolated to mg/ream.
| Table 2 - Volatiles from Printed and Coated Paperboard Sample | |||
| Retention Time | Assignment | Concentration in PPB (w/w) | Concentration in wt/area (ng/cm2) |
| 6.69 | branched c-10 hydrocarbon | 17.2 | 0.92 |
| 7.42 | tetramethyl pentadecane | 80.9 | 4.33 |
| 10.64 | nonanal | 46.8 | 2.51 |
| 10.71 | dodecane | 56.3 | 3.01 |
| 12.14 | decanal | 46.3 | 2.48 |
| 12.18 | tridecane | 80.8 | 4.33 |
| 12.87 | cyclic polydimethyl siloxane oligomer (column artifact) | 87.3 | 4.67 |
| 13.04 | d-8 naphthalene (internal standard) | 933.7 | 50.00 |
| 13.39 | triisopropyl benzene | 80.9 | 4.33 |
| 13.57 | tetradecane | 330.2 | 17.68 |
| 13.65 | 1-tetradecene | 113.1 | 6.06 |
| 14.15 | p-tert-butyl phenol | 89.3 | 4.78 |
| 14.34 | pentadecane | 94.0 | 5.03 |
| 14.70 | Kodaflex® type plasticizer | 89.3 | 4.78 |
| 14.87 | pentadecane | 300.4 | 16.09 |
| 15.38 | methyl pentadecane | 86.2 | 4.62 |
| 15.50 | benzophenone | 154.2 | 8.26 |
| 15.76 | hexanediolmonoacrylate | 124.7 | 6.68 |
| 16.10 | hexadecane | 220.8 | 11.82 |
| 16.16 | 3-hexadecene | 166.8 | 8.93 |
| 16.19 | 1-hexadecene | 97.8 | 5.24 |
| 16.79 | 2-methyl hexadecane | 39.7 | 2.13 |
| 17.08 | diisopropyl naphthalene | 135.3 | 7.25 |
| 17.28 | heptadecane | 85.0 | 4.55 |
| 17.36 | tetradecanal | 199.3 | 10.67 |
| 17.19 | hexendioldiacrylate | 51.0 | 2.73 |
| 17.97 | hexanedioldiacrylate (HDODA) | 407.0 | 21.79 |
| 18.39 | octadecane | 35.4 | 1.90 |
| 18.50 | hexadecanal | 84.0 | 4.50 |
| 19.49 - 19.73 | tripropyleneglycoldiacrylate (TPGDA) | 747.0 | 40.00 |
| 20.53 | 2,4 diphenyl-4-methyl-1-e-pentene | 29.3 | 1.57 |
| 20.71 | trimethylolpropanetriacrylate (TMPTA) | 87.8 | 4.70 |
In the samples we see the presence of several acrylates. These are residual short chain monomers and oligomers that evolve due to incomplete curing or radiolytic decomposition products. Due to the coating formulation, we see hexanedioldiacrylate (HDODA), hexendioldiacrylate, ethoxylated hexanedioldiacrylates, and trimethylolpropanetriacrylate. In the printed sample there are several hydrocarbons such as decane, nonane and branched chain hydrocarbons. These products may arise from residual lubricants used in paperboard processing, ink solvents for printing, binders, clays, and sizing agents. In addition we see benzophenone, a photoinitiator used in UV-cure inks. They are molecules that, when light energy in a given wavelength hits it, will cleave homolytically to give a free radical. The free radical generation is to initiate and accerlate the polymerization/crosslinking chain reaction. The final class of compounds seen is phthalates, which are used as plasticizers and/or viscosity control agents.
Conclusion
The purge and trap or dynamic headspace technique followed by thermal desorption has been utilized for the identification and quantification of the volatile organic compounds in electron beam cured coating formulations. This analytical method can be used to analyze these samples unattended through the use of a new automated system. The data presented here is novel, because currently little data is publicly available on the volatiles from radiation cured packaging. The method can easily be integrated for quality assurance purposes in manufacturing facilities or buyer acceptance. Since this system has the capability to run unattended, it can be used to automatically develop methods by varying sample sizes, purge time, desorption time, cryotrap cooling temperature, and GC column temperature program. Since several samples can be analyzed, the reproducibility and precision of the data obtained can be assessed using statistical methods. The purge and trap method followed by automated Short Path thermal desorption can be applied to a variety of sample matrices, including spices, herbs, air, integrated circuits, and foods. Preliminary data with this system has been generated on hydrocarbon and antioxidant standards, deuterated surrogate internal standards, spices (saffron) and food (cookies).
![]() Yttria coated filament at start |
![]() Yttria coated filament after 16,000 cycles |
New Pumps Added to the List of LOW COST S.I.S. Rotary Vane
Vacuum Pumps - Save 35 - 50% off the price of comparable vacuum pumps.
Adixen (Alcatel) ACP Series Compact Dry Roughing Pumps - a frictionless pumping module, optimized for operation without internal lubricant.
- Direct Replacement for the Shimadzu QP5000 Filament.
PCI Analytics Products - product solutions for all Analytical Instrumentation
SilTite FingerTite, the next generation GC capillary ferrule system without the use of any tools.
(25% discount until Oct 31)
Direct Replacement for the Heated Capillary in the Thermo-Fisher Scientific TSQ & LCQ Mass Spectrometers now available
- Ideal for use in analytical instrumentation
- Will improve your laboratory workflow and reporting confidence in many ways
DigiVac Bench Top Vacuum Gauges
- Vacuum instrumentation with everything you need to go to work
Imtakt HPLC Columns
-offer a wide variety of unique stationary phases in their HPLC
columns
Programmable syringe pumps for controlled flow. Model NE-300 priced at only $275.00
NEW: Wiley 9th Edition - 660,000-850,000
spectra (a 65% increase).
NEW: Agilent 5973/75 Replacement Source
Heater and Temperature Sensor
NEW: SIS Direct Replacement
Filament for the Amatek/Dycor RGA Leak Detectors and Quadrupoles
(p/n 95381VE), which tests have shown increase service life.
NEW: SIS
Replacement Filter Elements For Edwards Oil Mist Eliminators -
Save up to 25% off the price of comparable Edwards replacement
elements.
Identify mass spectra via library
search with the fully-evaluated, most widely used mass spectral
reference library, compiled by the National Institute of Standards
and Technology (NIST).
View and capture images or video at
20x/400x power on a PC computer via attached USB cable.