Oxford University Cave Club

1995 Expedition: Picos de Europa, Spain

Boca del Joon


Science Report - Dye Tracing

On 5th August 500g of fluorescein was placed in the streamway near the bottom of F64. The detectors consisted of 25g bags of activated charcoal, which was bought from an aquarium shop. These were placed at various locations in C3 and at Culiembro as summarised below.

Detector  Location                                       Placed   Removed     

    A     C3 Main stream downstream from t-junction.     5/8/95   11/8/95     
    B     C3 First inlet upstream from t-junction.       5/8/95   11/8/95     
    C     C3 Second inlet upstream from t-junction.      5/8/95   11/8/95     
    D     C3 Third inlet upstream from t-junction.       5/8/95   11/8/95     
    E     C3 Fourth inlet upstream from t-junction.      5/8/95   11/8/95     
    F     C3 Main stream upstream from inlet G.          5/8/95   11/8/95     
    G     C3 Fifth inlet upstream from t-junction.       5/8/95   11/8/95     
    1     Culiembro. In Cares stream, upstream from      20:00    08:00       
          resurgence.                                    4/8/95   5/8/95      
    2     Culiembro. In resurgence, above main Cares     5/8/95   18/8/95     

In the absence of a fluorescein spectrometer I developed an alternative method of identifying fluorescein in samples extracted from dye detectors. Thin layer chromatography (TLC) was the obvious choice. In TLC the sample to be analysed is spotted onto the bottom of a plate coated with a thin layer of silica gel. Solvent is then allowed to run up the plate by capillary action. The ease with which different components are eluted up the plate is dependent primarily on the polarity of the individual components. For a given component under specified conditions of solvent and plate coating the retention factor (r.f.: the ratio of distance eluted by the component to the distance travelled by the solvent front) is a good analytical guide. Fluorescein, under UV light (220 nm), has a strong yellow fluorescence. This, together with the r.f. offers a cheap, reliable technique for the analysis of fluorescein. This system may offer several advantages over the use of fluorescence spectrometry. The materials and hardware are cheap, portable and accessible. There is no need to obtain a measure of background fluorescence. While obtaining relative quantitative values via TLC is unreliable, it would be trivial to extend this technique to high performance liquid chromatography (HPLC) which is quantitative and a much more sensitive and powerful technique. While it is difficult to quantify the relative sensitivity of TLC versus fluorescence spectrometry without access to a spectrometer I did carry out quantitative work and was easily able to obtain clear results on samples as small as 0.01g (1x10-8g).

The r.f. for fluorescein is for the free acid. If the dye detector is extracted into potassium hydroxide it will be potassium salt, which is not eluted under these conditions. Treatment with a mineral acid is required to give the acid. The absorption of the dye is affected by acidity; under acidic conditions the extinction coefficient drops to a minimum at pH ~3.5.

The bags had been stored damp from September to December and had in some cases become mouldy and fallen apart. However there was no possibility of cross contamination.

Dye Detection

The bags of charcoal were extracted into 70 ml portions of potassium hydroxide in ethanol for 2 hours. At this stage some of the bags gave the clear and obvious colouration and fluorescence associated with fluorescein. The ethanolic solution was decanted from the bags and treated with 3.5 M hydrochloric acid to pH 1-2. This was to give a sample suitable for (TLC). A sample of the acidic ethanol solution was spotted onto a TLC plate (Merck 1.05735) and eluted with ethyl acetate. Fluorescein gives a distinctively yellow fluorescence when viewed under a UV lamp (220 nm) with an r.f. of 0.54. Samples showing no response at this stage were concentrated under reduced pressure and a second TLC run under the same conditions. The only sample to give a positive test under these conditions was a very weak response from the detector placed at Culiembro. The results for all detectors are summarised below.

Detector   EtOH extract       TLC     TLC on concentrated  

    A         Strong       Positive           ---          
    B            X             X               X           
    C            X             X               X           
    D            X             X               X           
    E            X             X               X           
    F         Strong       Positive           ---          
    G            X             X               X           
    1            X             X               X           
    2            X             X           Very weak       

It is clear from these results that F64 forms part of the upstream catchment of C3, which itself drains into Culiembro.

Jim Ramsden


From these results it is obvious that the water at the end of F64 drains into C3, which then resurges at the Culiembro resurgence. Although further dye tracing experiments would be desirable to reinforce these conclusions there are several good reasons why these existing results should be accepted.

The surveys shows that the end of F64 is over 100m above the C3 streamway and is heading towards the area that upstream C3 is heading for. The end of the C3 streamway is heading directly towards the Cares gorge.

The detectors placed in the main stream of C3 gave strong positive results, whereas the detectors in the inlets above the main stream were all negative.

The detector in the resurgence several kilometres away was weakly positive, as would be expected from the large amount of dilution. The detector in the main stream upstream of, and below the resurgence was negative.

Although some of the detector bags were mouldy and damaged each bag was stored inside three polythene bags, so there was no danger of cross contamination.

There is much work that remains to be done in C3 next year. No attempt was made this year to try to get a throughput time for the C3 water. Now we know that the time is definitely less than one week (and this was from F64, so the time from C3 would almost certainly be less), it should be possible, given a willing volunteer, to get a more accurate estimate. It would also be useful, though probably impracticable to carry out a trace from C3 to 2/7. This would tell us whether the water down C3 is a new piece of the Culiembro jigsaw or whether it is simply a part of the 2/7 water that we have already found. However, this would require a large amount of equipment and person-power.

Overall the dye tracing should be seen as a great success. It is the first positive trace obtained from a Top Camp cave and many thanks must be given to those who spent time and effort on this experiment.

Dave Lacey

Contents: Next Article (Medical Report)