Variability in capsaicinoid content and Scoville heat ratings of commercially grown Jalapeño, Habanero and Bhut Jolokia peppers. (1st November 2016)

• Record Type: Journal Article
• Authors: Sweat, Ken G.; Broatch, Jennifer; Borror, Connie; Hagan, Kyle; Cahill, Thomas M.
• Is Part Of: Food chemistry. Volume 210(2016)
• Page Start: 606
• Page End: 612
• Publication Date: 2016-11-01
• Subjects: Capsaicin -- Dihydrocapsaicin -- Jalapeño -- Habanero -- Bhut Jolokia -- Capsaicinoid -- Market basket study -- ANOVA -- Bootstrap -- Variation; Capsaicin (PubChem CID 1548943) -- Dihydrocapsaicin (PubChem CID 107982); Food -- Analysis -- Periodicals; Food -- Composition -- Periodicals; 664
• DOI: 10.1016/j.foodchem.2016.04.135
• Languages: English
• ISSNs: 0308-8146

Abstract

Highlights: Pepper varieties have a high degree of variability in Scoville heat ratings. Most variability was inherent to the pepper variety and not from the supplier. Different pepper varieties had different distributions of Scoville heat ratings. A larger sample size was necessary to assess maximum heat as opposed to average heat. Abstract: The variability of capsaicinoid content of three common, commercially-available hot pepper varieties, namely Jalapeño, Habanero and Bhut Jolokia, was investigated. For each variety tested, ten peppers were acquired from each of ten different suppliers resulting in 100 peppers per variety that were individually analyzed. The results showed that different pepper varieties had different distribution types. The Habanero peppers showed a normal distribution; the Bhut Jolokia showed a skewed distribution and the Jalapeño peppers showed a very skewed distribution. The source of variability was also different; the Habaneros were very consistent within a given pepper supplier so most of the overall variation resulted from differences between suppliers. The Jalapeño peppers were the exact opposite with a very high degree of variability within a given supplier and relatively low variation between suppliers. A bootstrap statistical simulation was conducted on the data to suggest a minimum number of peppers to analyze to characterize the variation in a population. The simulations indicated that small sample sizes are effective at estimating the mean concentrations, but a sample size of ten or more is necessary to describe the population and capture the high-end tail of the distributions, which are the very hottest peppers.