Protocol

 

Author

Ian Wallis

Overview

Preparing samples for gas chromatography of leaf oils.

Background

This article concerns the preparation of samples for the analysis of terpenes and not the analysis of those samples by gas chromatography. There are many methods for sample preparation and this is just one of them. Indeed, we rarely use this method now and instead put a known mass of fresh leaf into a known volume of solvent (usually ethanol) containing an internal standard – dodecane or tridecane. We previously used steam distillation, with leaves immersed in water, but this almost certainly leads to loss and alteration of compounds. Clearly, there is a need for a standard method for extracting and analyzing terpenes so that the results reflect the concentrations in the starting material.

Materials/Equipment

• fine point scissors
• micro syringe (50 L)
• thick tweezers
• analytical balance
• auto pipette (1000 L) and tips
• butane fuel injector
• glass Pasteur pipettes prepared for the analysis
• flame torch
• glass GC vials with septum (ca 1.8 mL)
• leaf samples
• 50 mL volumetric flask
• safety glasses
• chemicals
  • hexane
  • tridecane
  • standards
• porcelain dry matter crucibles, dried in the oven and cooled in a desiccator

Units, terms, definitions

Density of tridecane = 0.756; density of hexane = 0.659

DM – Dry matter

Cineole – a monoterpene

Procedure

1. Prepare sample tubes from glass pipettes. Do this in bulk, preparing a separate tube for each duplicate of a sample. Use a flame torch on the highest setting to heat the pipette at the taper where the pipette becomes a capillary. Constantly twist the pipette in the flame and as the glass softens pull away the end. Store the prepared tubes in a beaker covered with foil.
2. Label the tubes before you begin weighing. Place them in order in an appropriate rack.
3. Check the analytical balance (4 decimal place = 0.1 mg) by ensuring that it is level and that the weighing platform is clean. If not, use a fine brush to clean it.
4. Weighing the samples. Note: oil analyses in duplicate; DM (dry matter) analyses on single samples. Weighing the samples is complicated because they are frozen. They will rapidly thaw and condensation followed by evaporation will occur. Thus, it is essential to weigh samples for dry matter at the same time that they are weighed for oil analysis. This necessitates working on small batches of samples (5) at a time.
   a) Oil analysis: weigh samples to the nearest 100 mg, keeping within a range of 97-103 mg. Lay the frozen sample on tissue to absorb water before weighing. Allow them to defrost. Cut away the stem and the main vein and slice the inner part of the leaf into 10 x 1 mm strips. Weigh these into the glass pipette, which should be stored on ice until sealing.
   b) DM analysis: weigh samples to the nearest 250 mg, keeping within a range of 240-260 mg. One sample is sufficient. Remove a porcelain crucible from the desiccator, record its number, place it on the balance and record its mass. Tare the balance. Weigh the sample.
5. Preparing the solvent. Note – This is a critical stage in the analysis.
   Tridecane (the internal standard) is extremely stable in hexane. Therefore, it is sensible to prepare a large volume of hexane reagent and to decant sufficient for each batch of analyses. Use a calibrated autopipette to transfer 500 l of tridecane to a clean 500 ml volumetric flask. Half fill the flask with hexane, stopper, invert 10 times and then fill to the mark with hexane. Invert a further 10 times. Better still, weigh the 500 l of tridecane into a smaller vessel on the 4-decimal place balance and then transfer it, with serial washings, to the 500 mL volumetric flask. Note: density of tridecane = 0.756; density of hexane = 0.659
6. Adding the solvent to the samples. Note: all work with solvents should be done in a fumehood
   a) Set the oven to 100^∘C
   b) Take the samples from the fridge and place them on dry ice. Add 400 L of solvent to each sample using the 1000 L autopipette.
   c) Seal the tubes with care by heating them just above the neck of the tube, using the procedure described above.
7. Incubating the tubes
   The tubes, which should be labeled, contain a sample and solvent and be liquid-tight are now ready for incubation. Place them in the oven at 100^∘C for 1h.
8. Transferring the solution.
   The GC vials must be clean. Clean them with a solvent (eg hexane) and dry them in the oven(see below).
   a) Label all vials and place them in a rack in numerical order.
   b) Score with a file and then snap off the end of the sample tube.
   c) Using a separate pipette for each sample, decant the solution from the sample tube into the GC vial.
   d) Add hexane up to the shoulder of the vial.
   e) Keep sample vials in the fridge until analysis.
   f) Retain the sample until the analysis is finished and the results calculated.

Cleaning GC vials

1. Separate the vials from the lids and septa (removed from vial lids).
2. Place vials in a bottle, fill with hexane (low grade), sonicate for 10 minutes. Repeat twice.
3. Sonicate the septa and lids in a similar way.
4. Dry vials at 110^∘C.
5. Dry lids and septa at 50^∘C.
6. Repeat these steps for both vials and septa and lids using MilliQ water.

Notes and troubleshooting tips

Calculations

A spreadsheet is attached at the base of this page. It requires that you first calculate the mass of the internal standard, tridecane, per mass of hexane (mg per mg) and put this in cell U1. Then you should change cell U2 to reflect the uL of hexane + tridecane you added before incubation.

The basic calculation is:

Concentration of terpene ((mg per g fresh leaf) = (peak area of terpene *correction factor/tridecane peak area) * mass of tridecane per g fresh leaf)

To express the results as mg per g dry leaf simply divide by the proportion of DM in the leaf.

Note the correction or response factor is the relative response of a substance compared to the internal standard. For example, if cineole has a response factor of 0.7647 this means that the peak area of the IS is 76.47 % the size of the peak area from an equivalent amount of a cineole. Thus, the result for cineole is effectively reduced to 76% of what it would be if no correction factor were applied.