New Fully Automated System Simplifies Dioxin Sample Preparation, Saves Time and Reduces Costs

Project in co-operation with LCTech: Transfer to an automated system

The automated systems that had been on the market for a few years were not deemed to be viable alternatives by the CVUA-MEL scientists: Over and again, the Bernsmann team encountered during their evaluations the problem of very expensive consumables and a high solvent consumption, which by far exceeded that needed for their own manual processing. “In addition, enquiries amongst other users highlighted that the systems were maintenance-prone, took up a lot of space and needed to be placed underneath an extraction hood. All these points prevented us from the use of such a system to date.” Instead, their own methodology of manual processing has been developed over the years with the development potential now fully exhausted to be able to process the samples as fast as possible.

In order to find a way to shorten the processing times even further, Bernsmann established contact with LCTech GmbH in the Bavarian Dorfen. After some discussions, it was decided that together with the LCTech experts in the field of sample preparation systems, the manual CVUA-MEL method would be transferred to an automatic system, which would forgo the problems of other existing systems. “During the development we needed to merge the best analytical approach with what was technically possible,” explains Bernsmann. “One of the biggest challenges was to successfully implement the solvent exchange option between the individual column combinations in an automated system.”

Low detection limit through high fat content 

The device, which emerged from the collaboration, is characterised by a particularly simple mode of operation. Preparatory work includes only the connection of collection vessels to the output lines and clipping the ready-to-use columns into the holder – the other system’s standard screw connections are omitted, the locking is done by the system itself. Also the placement under an extraction hood is not necessary owing to the closed system design. “Before the samples are processed, the fat is extracted first. There are different ways to do this, e.g. by a Soxhlet-Extraction“, says Bernsmann. “Here fore, the fat together with the internal standards is dissolved in 5 mL hexane and 2 mL toluene and under rinsing with hexane, the solution is quantitatively transferred into the sample loop. ”Injection into the sample loop is the only manual step of the entire preparation process, everything else runs fully automatic after you start the device. 

Afterwards, the inserted columns are conditioned and the fat/solvent mixture is put onto the first column, a silica gel column, which is sulphuric acid impregnated. Here, fats and other substances will be oxidised. Depending on the fat matrix, the DEXTech™ device can process up to 5 g fat fully automatically. Generally, 2 g fat is sufficient to achieve the required low detection limits. “Dioxins and PCBs stay intact on the sulphuric acid column and are rinsed with hexane onto the second column, the Florisil column,” explains Bernsmann. On this column PCBs and dioxins are separated. The latter remain with hexane on the Florisil column, whilst PCB is transferred onto the first PCB-carbon column. There, the non-ortho-PCBs are retained with hexane, while the mono-and di-ortho-PCBs are collected in the first fraction. The non-ortho-PCBs are eluted from the carbon column with toluene using the reverse flow or backflush procedure.

Optimal Separation and Fractionation

“What’s new is the transfer of the dioxins from the Florisil-phase with dichloromethane/hexane (1:1) to another carbon column using the reverse flow method. Through this method much solvent is saved and the dioxin fraction can be cleaned once more”, explains the expert. The dioxins stick to the second carbon column, interfering substances can be removed with the solvent. From this carbon column, the dioxins will then be eluted again via backflush with toluene. The resulting three fractions only need to be evaporated and then measured. Owing to the optimum separation and fractionation of the dioxins and the different groups of PCB, a superimposition of the toxicologically particularly relevant non-ortho-PCB with mono-ortho- and di-ortho-PCB is avoided during the subsequent chromatographic analysis, and an incorrect evaluation is prevented from the start. From raw extract to the finished fractions, this process takes only 96 mins, including drying with nitrogen. The reaching of the program end can be made audible with the sound of a signal tone. 

This universal method is suitable for food and feed samples as well as for environmental samples and enables any matrix to be processed. The method is pre-set in DEXTech™ and therefore can be easily applied to all. “Method development is no longer necessary. However, the system allows to adapt individual processing steps or to program a new method via touch screen, such that the device can be used very flexibly,” says Bernsmann. A conversion from a 4- into a 3-column system is possible; dummy columns are included specifically for this purpose.
 

Low solvent consumption and elimination of labour-intensive steps

Applying the standard method, the DEXTech™ uses less than 640 mL solvent including column conditioning, which is a huge saving compared to manual processing, and also to other systems. The low consumption is supported by specially constructed, rotatable switching valves with dead space free passages, which also prevent clogging of the system. A carry-over from one sample to another is not possible, as for example the sample loop can be rinsed very fast and effectively in the process. Compared to the manual methods, evaporation steps between sample loading and elution are negated with the DEXTech™. Preliminary tests of the materials used as well as filling, preparing and rinsing of the columns are no longer necessary. Consequently, all labour and time intensive steps are now removed. “Furthermore, all connections are clearly laid out in the viewing area in front of the user. They can be easily visually checked and also repaired, if necessary,” says Bernsmann. The base of the DEXTech™ measures only 30 x 57 cm, such that several devices can be placed side by side without any problems whereby the throughput can be further increased.

The system could already impress with its performance at the interlaboratory testing (LVU) “Feed Fat 2013” of the European Reference Laboratory EURL in Freiburg. All analytical results came very close to the consensus values, which were determined by the EURL for LVU from data of all participating laboratories. “The future development of the DEXTech™ system will be aimed at a further reduction of the time needed for sample clean-up and thereby even more solvents will be saved,” explains Bernsmann. “In addition, we would like to work on the reduction of materials whilst maintaining the same performance.”