Whitening is a core function of cosmetics， and ingredients including acetyl glucosamine， tranexamic acid， nicotinamide， phenethyl resorcinol， glabridin， and ascorbyl tetraisopalmitate are widely used due to their clearly established whitening mechanisms. However， existing analytical methods face notable limitations： photodiode array detectors （PDA） cannot reliably detect weakly or non-ultraviolet （UV）-absorbing components； evaporative light scattering detectors （ELSD） suffer from inadequate sensitivity for low-concentration analytes； mass spectrometry （MS） is costly； and single-column cannot achieve the separation of ingredients with significantly different polarities. To address these challenges， this study established an analytical method for the determination of the six whitening ingredients in cosmetics using ultra performance liquid chromatography coupled with a photodiode array detector and a corona charged aerosol detector （UPLC-PDA-CAD）. For sample pretreatment： 0.2 g of sample was weighed， and 10 mL dichloromethane and 10 mL deionized water were added for vortex extraction of whitening ingredients. The mixture was then centrifuged for phase separation. The aqueous layer and dichloromethane layer were collected separately. The aqueous layer was washed twice with dichloromethane； all dichloromethane fractions were combined， concentrated to less than 1 mL under a nitrogen stream， and finally made up to volume with isopropanol. Both the aqueous layer and the reconstituted isopropanol solution were filtered through 0.22 μm hydrophilic polytetrafluoroethylene （PTFE） membranes before UPLC injection. For chromatographic separation， a Waters HSS T3 column （150 mm×2.1 mm， 1.7 μm） was selected. The column temperature was set at 40 ℃， the injection volume was 1 μL， and the flow rate was 0.3 mL/min. The mobile phase consisted of isopropanol， acetonitrile， and 20 mmol/L ammonium acetate solution （pH adjusted to 4.5 using formic acid）， using a gradient elution program. For detection， a tandem PDA-CAD system was used： PDA （scan range 200–400 nm） was employed for components with UV absorption， while CAD （nebulization temperature 35 °C， acquisition frequency 5 Hz） was used for components with weak UV absorption. Quantification was performed using the external standard method.Method validation results showed good linear relationships for all six whitening ingredients within their respective concentration ranges. The correlation coefficients （r） were all greater than 0.999. The limits of detection （LODs， S/N=3） were 5.0–50.0 µg/g， and the limits of quantification （LOQs， S/N=10） were 12.0–120.0 µg/g. Spiked recovery tests were conducted on negative cosmetic matrices （emulsion， cream， oil） at low， medium， and high levels. The recoveries ranged from 92.8% to 110.1%， and the relative standard deviations （RSDs， n=6） were 0.12%–5.45%， indicating excellent precision and accuracy. This method was applied to seven commercially cosmetics. The results revealed that all target whitening ingredients declared on the product labels were detected， with significant differences in their content. Nicotinamide was the most frequently detected compound， found in five products， and also exhibited the highest concentrations， ranging from 0.19% to 2.29%. Phenethyl resorcinol was detected in three products， with contents ranging from 0.02% to 0.52%. Ascorbyl tetraisopalmitate was detected in two products， at 0.09% and 3.08%， respectively. Acetyl glucosamine， tranexamic acid， and glabridin were each detected in only one product. In conclusion， this established UPLC-PDA-CAD method is simple， efficient， sensitive and accurate. It effectively overcomes the technical challenge of detecting whitening ingredients with widely varying polarities， offering reliable technical support for cosmetic quality control， regulatory supervision， and evaluation of product whitening efficacy and potential sensitization risks.