00

by Brian Ruddell

Editor’s Note: Brian Ruddell, a New Zealander studying in Australia, was kind enough to write us about High Performance Liquid Chromatography tests that he performed on a number of chiles that he grew in his garden, as well as a famous hot sauce. The ‘Naga Jolokia’ he tested was being sold as “Indian PC 1” and is not a real ‘Naga Jolokia’, which now is believed to be a variety of the Caribbean Red (Capsicum chinense) pod type. This report is rather technical, but I have decided to leave it as is in the interest of science. –D.D.

I am doing a Diploma of Laboratory Technology at Challenger TAFE (Technical and Further Education, like your Institutes of Technology), Murdoch Campus, in Perth, Western Australia. As part of my studies, there is a requirement to complete a practical project involving something of interest to the student by using any of the analytical gear available (HPLC, GC, GC-MS, FAA, IC , etc.). It is meant to show one’s ability to perform independent work. As I am a keen chile grower, it was a perfect opportunity to combine the two interests.  I have the write-up pretty much done and it just needs a bit of polishing up, my supervisor is currently reading it and adding a few suggestions and comments (he seems impressed). The Diploma is a two year course and I am doing it in little over a year–which is a little bit hectic!

Also I ended up with a lot more information than I thought I would, and being a bit of a neophyte when it comes to writing up such projects and collating all the data, it has been a bit slow. During the analyses, I could not resist trying something else and doing “just one more run.” I ran analyses to compare results using different wavelengths that I have seen used in other methods. I used in my extraction methods the full AOAC method and also used acetonitrile sonication and ethanol sonication and compared the results for these. I found they had good agreement but acetonitrile was good when doing heavily coloured samples such as cayenne powder as it did not extract so many interfering substances. My runs were fairly long at 36 minutes, but I was able get excellent reproducibility and the need for using c-18 solid phase clean up cartridges was eliminated as the early eluting substances were gone before the capsaicin started to come off the

column. But I also did compare results with and without c-18 clean up cartridges. Some of the analytical work I have seen, although far shorter, had the capsaicin peaks as mere shoulder peaks and I did not like that. I suppose in a commercial laboratory, it is a case of time is money but as a student it was my own time and I was happy to use it.

Using the AOAC method, whole (less calyx) fresh chiles were used. Their dry weight was calculated by drying a number of that type of chile pods to get an average water loss. The Acetonitrile/sonication and the ethanol/sonication methods involved whole (less calyx) freshly dried chiles. All samples except the cayenne powder, Tabasco and placental tissues were of whole chiles, including seeds. When drying the fleshier rocoto and habanero pods, I put a couple of dozen cuts through the pod epidermis with a scalpel, taking care not to cut through the pod wall. A bit tedious, but with the long run times I had plenty of spare time and it gave excellent drying results.

With the placenta samples I made every effort to select the best looking bits from several pods as I really did want to know how hot the hottest part of the hottest chile is. I selected the sections where the white placenta was flecked with pale amber spots. I do not know but I assumed these may be centres of capsaicin production. Whatever the case, the results were quite spectacular. Although I am very confident in the 4,000,000+ Scoville Heat Units results, I hope that someone else will do a similar analysis to back up such a result, as I have not read of any results of this magnitude from plant material before. It is a good illustration of a point Dave DeWitt made in an article about how to get artificially high results from a chile sample–a little dried Savina placenta will go a very long way. One thing I wished to do but ran out of pods and time was to take a few pods and get a percentage of capsaicin in the placenta as compared to the total pod capsaicin content. Oh well, another time. Also, the removal of seeds is a way to bump up results, and the ‘Naga Jolokia’ (not) pod with the seeds removed would have had its SHUs doubled or nearly tripled, as the seeds were a very large percentage of the dry weight.

One thing I found which made life easy was the linearity of response over a wide range of concentrations. Using eight points from 7-150 ppm, my R values were from memory for both capsaicin and dihydrocapsaicin about 0.9994 . Using another sample of 300 ppm, the results still showed excellent linearity. Had time permitted, I would have like to have found how far this extended. It would save a bit of time not having to do so many dilutions! I had a Savina placenta sample which of course was miles out of the standard solutions calibration curve range, and using the equation for the line got a result. I then diluted it 10 times and ran it again using the equation for the line to get capsaicin figures. The results were within a couple of percent of each other, which would be pretty much be due to experimental error. I made my standard solutions from purified capsaicin and dihydrocapsaicin obtained from Sigma Chemicals. The nordihydrocapaicin and the homocapsaicin peaks were identified from comparing my chromatograms with those from a number of other experimenters and comparing the retention times. Following is a chart of my results.

The author would like to thank his supervisor, Mr. Ron Levitt. “Without his assistance, I would not have been able to do the study. He helped with advice and guidance when I submitted my proposals and took on supervision of my project when he was not obliged to.

Top of Page

00