Abstract
The poor interoperability of anthocyanin glycosides measurements by two pH differential methods is documented. Adams-Harbertson, which was proposed for commercial winemaking, was compared to AOAC Official MethodSM 2005.02 for wine. California bottled wines (Pinot Noir, Merlot, and Cabernet Sauvignon) were assayed in a collaborative study (n = 105), which found mean precision of Adams- Harbertson winery versus reference measurements to be 77 ± 20%. Maximum error is expected to be 48% for Pinot Noir, 42% for Merlot, and 34% for Cabernet Sauvignon from reproducibility RSD. Range of measurements was actually 30 to 91% for Pinot Noir. An interoperability study (n = 30) found Adams-Harbertson produces measurements that are nominally 150% of the AOAC pH differential method. Large analytical chemistry differences are: AOAC method uses Beer-Lambert equation and measures absorbance at pH 1.0 and 4.5, proposed a priori by Flueki and Francis; whereas Adams-Harbertson uses “universal” standard curve and measures absorbance ad hoc at pH 1.8 and 4.9 to reduce the effects of so-called co-pigmentation. Errors relative to AOAC are produced by Adams-Harbertson standard curve over Beer-Lambert and pH 1.8 over pH 1.0. The study recommends using AOAC Official Method 2005.02 for analysis of wine anthocyanin glycosides.
An international collaborative study on the First International Standard of bovine luteinizing hormone for immunoassay
