Content
Citation(s) & Footnote(s)
Book(s)No access › No access
Source argument
Charring partially disrupts cellulose crystallinity, followed by loss of cell-wall carbohydrates from spirit-exposed oak; a model experiment supports a cellulose contribution to solution glucose.
Evidence map
Gollihue 2018 — Stave layers and sampling limits — pp.2,10; Fig.1; supplementary Fig. S1
The authors distinguish C (charred interior), P (thermally modified and penetrated wood), R (red-line boundary), and O (outer wood). Red-line location varies within and across staves, so anatomical sampling is not a fixed universal depth. Five staves per treatment and two ten-year barrels do not make every measured subsample an independent barrel. The five-year comparison held rye, whereas six- and ten-year samples involved bourbon.
X-ray and microscopy results support reduced crystalline organization in charred oak and strong loss of detectable C-layer cellulose after aging. Figure 2 gives relative crystallinity about 26.15 in new C versus 59.54 outer, and 3.09 in ten-year C versus 59.24 outer. Table S1 defines the index as percent; the narrative incorrectly attaches diffraction-angle notation to the index. Fluorescence and crystallinity are not direct measurements of dissolved bourbon sugar.
Updegraff results report C-layer cellulose 308.0 micrograms/mg in new wood and 6.2,9.1,6.8 after 5,6,10 years. Deeper ten-year layers retain much more cellulose. These sparse, nonrandom age groups do not establish a smooth aging-rate curve. Sulfuric-acid results support the broad C-layer decline, but several published deeper-layer means and one new-C standard error do not reconcile with supplied workbook observations.
TFA and sulfuric-acid hydrolysis give different neutral-sugar quantities and statistical contrasts. TFA did not show a significant new-C versus new-O total-sugar difference; sulfuric analysis did. Subsequent aging effects depend on assay and contrast. These are hydrolyzed alcohol-insoluble wood measurements, not ready-to-drink whiskey sweetness. The authors discuss incomplete TFA hydrolysis and acid-driven sugar degradation as methodological explanations.
A cellulose-depleted wood treatment yielded less glucose in a heated ethanol model than buffer-treated wood. This indirectly supports a cellulose contribution, without tracing labeled cellulose into commercial bourbon or measuring sensory effects. The model used toasting, not charring, and heated flammable ethanol in laboratory equipment; it is not a home-aging instruction. Results describe 180 minutes of initial sampling while Methods says 360; the published plot ends 180.
The paper extrapolates approximately 300 g of C-layer glucose building blocks and 1.3 g/L in a standard barrel. These are not measured bottle concentrations; 300 g divided by the stated 200 L barrel capacity would be 1.5 g/L before other assumptions. The authors explicitly say the proposed downstream ethyl-glucoside/HMF pathway has not been shown in whiskey solution. Do not turn it into a claim that a bourbon contains a measured amount of sweet sugar.
Independent grouping of all supplied cells yields TFA sample counts 8-15 per group, not uniformly 12. TFA group means and standard errors largely match Table S2 to displayed precision. Sulfuric new groups contain 8 observations, not 10; ten-O neutral sugars have two missing dashes. Some sulfuric values differ materially: ten-O glucose G51:G58 mean 47.531 versus published 26.9; ten-P xylose I29:I38 mean 276.979 versus 256.4; ten-R xylose I39:I48 mean 253.073 versus 291.6; ten-O cellulose J49:J58 mean 267.258 versus 298.0. Preserve both reported and recomputed values, do not silently repair the source. This discrepancy is not evidence of misconduct.
Seven authors are credited; Wheatley declares an employment-based financial interest and supplied production knowledge/material. This is primary evidence for the described study, not primary historical verification of the 1826 Corlis quotation or Roman-era spirits claim. The cited 50-80% flavor-from-wood estimate is background, not measured here. The 57 references are research leads, not independently processed sources.
Researcher synthesis
Separate solid-wood measurements, laboratory extraction, and extrapolated bottle chemistry. Preserve differences between assays and disclose non-reconciled published numbers. The work informs maturation mechanisms, not sensory sweetness predictions or home-blending recipes.
Assessment
All 12 PDF pages, five supplementary tables, five supplementary figures, and 1,184 populated cells across two workbook sheets read. All 12 article pages and 11 supplement render pages visually inspected. Group summaries independently calculated; not a reproduction of full statistical analysis.
Nonrandom production staves, five-year rye mixed with bourbon ages, within-barrel replication, assay differences, incomplete raw-data coverage and material workbook/table discrepancies. Wheatley declared employment financial interest. No sensory panel.
Why do sample counts and several sulfuric-acid values differ between workbook and tables? No author contact authorized or performed. Exclude unresolved precise claims from publication rather than invent a reconciliation.
Completion checklist
Flavor-profile reconstruction is not maturation provenance.
