TY - JOUR
T1 - UV gradient combined with principal component analysis
T2 - Highly sensitive and specific high performance liquid chromatography analysis of cosmetic creams
AU - Gao, Wenhui
AU - Gray, Nicola
AU - Heaton, James
AU - Smith, Norman W.
AU - Jia, Yingmin
AU - Legido-Quigley, Cristina
PY - 2012/3/9
Y1 - 2012/3/9
N2 - HPLC has been employed to develop a method for the analysis of cosmetic creams, in particular the compounds hydroquinone, phenol and six preservatives have been studied. UV tuning was optimized as a gradient to achieve lower limits of detection compared to those of a previously validated method. In addition the chromatograms were then exported, aligned and visualized in a principal component analysis (PCA) model. The results were the highly efficient separation of the eight studied compounds. All the compounds showed good linear correlation coefficients (>= 0.9997), the detection limit was found to be in the range of 15-200 ng/mL, a 10-fold improvement for the preservatives on previous methodology and the average recovery was within limits between 83% and 117% with a relative standard deviation (RSD) less than 3.6% (n = 6). The PCA plot was constructed from the UV optimized cosmetic samples chromatograms from real samples, real samples that were spiked and quality controls. Quality controls contained the eight compounds and showed complete clustering in the PCA and three spiked samples containing six to seven toxic components clustered in the same quadrant. The method is highly sensitive and its potential use as a method that could be employed in the control of cosmetics, particularly those containing banned or suspected toxic additives, has been demonstrated.
AB - HPLC has been employed to develop a method for the analysis of cosmetic creams, in particular the compounds hydroquinone, phenol and six preservatives have been studied. UV tuning was optimized as a gradient to achieve lower limits of detection compared to those of a previously validated method. In addition the chromatograms were then exported, aligned and visualized in a principal component analysis (PCA) model. The results were the highly efficient separation of the eight studied compounds. All the compounds showed good linear correlation coefficients (>= 0.9997), the detection limit was found to be in the range of 15-200 ng/mL, a 10-fold improvement for the preservatives on previous methodology and the average recovery was within limits between 83% and 117% with a relative standard deviation (RSD) less than 3.6% (n = 6). The PCA plot was constructed from the UV optimized cosmetic samples chromatograms from real samples, real samples that were spiked and quality controls. Quality controls contained the eight compounds and showed complete clustering in the PCA and three spiked samples containing six to seven toxic components clustered in the same quadrant. The method is highly sensitive and its potential use as a method that could be employed in the control of cosmetics, particularly those containing banned or suspected toxic additives, has been demonstrated.
U2 - 10.1016/j.chroma.2011.08.077
DO - 10.1016/j.chroma.2011.08.077
M3 - Article
VL - 1228
SP - 324
EP - 328
JO - Journal of Chromatography A
JF - Journal of Chromatography A
ER -