No access
Full-volume textual analysis: all 521 PDF pages, including front matter, eleven chapters, appendices, references and index. Twenty-eight selected technical/history pages plus covers received visual review. This is not a line-by-line independent proof of every equation, a rerun of numerical examples, or independent verification of cited publications. Printed anomalies are retained below as reuse cautions, not concealed.
Source argument
The volume presents distillation as a mature but evolving separation technology whose performance depends jointly on thermodynamics, transport, apparatus, operating conditions and design objectives. Its historical chapter emphasizes cumulative innovation; the engineering chapters move from equilibrium and rate descriptions to batch trajectories, heat integration and computational synthesis.
Evidence map
- Chapter 1: History of Distillation; printed 1–43, original PDF 22–64.
SRC-637 ch1 — Early apparatus evidence is not automatically beverage-spirit evidence ·
SRC-637 ch1, pp. 1–7 — Early apparatus evidence is not automatically beverage-spirit evidence
- Chapter 2: Vapor–Liquid Equilibrium and Physical Properties for Distillation; printed 45–95, original PDF 65–115.
SRC-637 ch2 — Mixture behavior requires more than pure boiling points ·
SRC-637 ch2, pp. 46–60 — Mixture behavior requires more than pure boiling points
- Chapter 3: Mass Transfer in Distillation; printed 97–143, original PDF 116–162.
SRC-637 ch3 — Real separation depends on finite transport ·
SRC-637 ch3, pp. 99–115 — Real separation depends on finite transport
- Chapter 4: Principles of Binary Distillation; printed 145–185, original PDF 163–203.
SRC-637 ch4 — Reflux is a separation variable, not a flavor score ·
SRC-637 ch4, pp. 158–172 — Reflux is a separation variable, not a flavor score
- Chapter 5: Design and Operation of Batch Distillation; printed 187–224, original PDF 204–241.
SRC-637 ch5 — Batch output follows a changing trajectory ·
SRC-637 ch5, pp. 191–198 — Batch output follows a changing trajectory
- Chapter 6: Energy Considerations in Distillation; printed 225–270, original PDF 242–287.
SRC-637 ch6 — Energy quantity and quality must be evaluated separately ·
SRC-637 ch6, pp. 226–247 — Energy quantity and quality must be evaluated separately
- Chapter 7: Conceptual Design of Zeotropic Distillation Processes; printed 271–303, original PDF 288–320.
SRC-637 ch7 — A design search cannot recover excluded alternatives ·
SRC-637 ch7, pp. 274–292 — A design search cannot recover excluded alternatives
- Chapter 8: Conceptual Design of Azeotropic Distillation Processes; printed 305–355, original PDF 321–371.
SRC-637 ch8 — Feasible balances do not guarantee a feasible separation profile ·
SRC-637 ch8, pp. 311–335 — Feasible balances do not guarantee a feasible separation profile
- Chapter 9: Hybrid Distillation Schemes: Design, Analysis, and Application; printed 357–381, original PDF 372–396.
SRC-637 ch9 — Hybrid selection must include the recovery problem ·
SRC-637 ch9, pp. 358–374 — Hybrid selection must include the recovery problem
- Chapter 10: Modeling of Distillation Processes; printed 383–436, original PDF 397–450.
SRC-637 ch10 — Rate-based models still need equilibrium and empirical support ·
SRC-637 ch10, pp. 394–412 — Rate-based models still need equilibrium and empirical support
- Chapter 11: Optimization of Distillation Processes; printed 437–496, original PDF 451–510.
SRC-637 ch11 — An optimum is conditional on model and search space ·
SRC-637 ch11, pp. 450–473 — An optimum is conditional on model and search space
Chapter-by-chapter analysis
Chapter1 — Norbert Kockmann, History of Distillation
DOI 10.1016/B978-0-12-386547-2.00001-6. The author treats innovation as an interaction of demand, materials, fabrication, heat/mass transfer and knowledge dissemination rather than a single inventor narrative.
- Printed1–7: historical terminology covered operations now distinguished as extraction, sublimation, filtration and pressing. Mesopotamian pots are interpreted technically, but the text itself acknowledges alternate fermentation-seal uses. Do not equate apparatus interpretation with evidence of beverage alcohol. Ancient dates and the claim of no Arab apparatus innovations require specialist corroboration.
- Printed8–14: vernacular printed manuals dispersed practical knowledge; Brunschwig1500, Ulsted1526, Ryff1545, Gesner1555, Agricola1556. Heating, condenser design, sealing, materials, residence time and reflux shaped performance. Figures are reproduced with external rights; do not republish automatically. Medical recipes and historical metals are historical evidence, not practical recommendations.
- Printed14–20: laboratory work and industrial demand interacted; Cellier-Blumenthal1813, Pistorius1817, Perrier1822, Coffey1830, Savalle and Dubrunfaut form a cumulative development chain. These are source-reported dates, not independently verified priority judgments. Distinguish patent, improvement, adoption and throughput.
- Printed20–30: coal tar, gas, petroleum and air separation expanded requirements; quantitative design followed equipment practice. Hausbrand, McCabe–Thiele1925, HETP, packings and molecular distillation. Source has potential historical/technical slips: Rayleigh date needs corroboration; energy units p25 kW/kg should not be reused without checking; p26 labels crude-oil remainder low-boiling before describing cracking long-chain hydrocarbons. Do not silently reproduce these.
- Printed30–37: computation, physical-property data, structured packing, reactive/hybrid separation, research consortia, heat integration, dividing walls and process intensification. Technology forecasts are dated2014. Numerical performance claims are configuration-specific, not bourbon yield/flavor claims. p33 says1997 for Fidkowski/Krolikowski but reference109 is1987; flag internal date conflict.
- Printed37–43: 132 references inspected; secondary references are leads, not read sources. Some cross-volume chapter references label Fundamentals incorrectly (air distillation/packing etc.); obtain original before citing.
Chapter2 — Jürgen Gmehling and Michael Kleiber
DOI10.1016/B978-0-12-386547-2.00002-8. Printed45–95. Central argument: trustworthy property data and model parameters control separation predictions; models cannot be accepted because a simulator runs.
- pp46–51: phase equilibrium means component fugacities equal across phases. K=y/x; relative volatility is a ratio of K values, affected by composition and temperature. Activity-coefficient and equation-of-state approaches have different strengths. Pure boiling-point ordering does not fully specify mixture behavior. Positive deviations may lead to minimum-boiling azeotropes/miscibility gaps; negative deviations may produce maximum-boiling azeotropes.
- pp51–60: NRTL/Wilson/UNIQUAC fit binary observations; temperature dependence, infinite-dilution data, excess enthalpy and accurate pure vapor pressures matter. Poor fits can invent/remove azeotropes. Fig2.5 ethanol-water data are at343K, not a universal proof table. Unit conventions for interaction parameters must be checked. Minor components cannot be dismissed in whiskey merely because bulk separation specifications ignore them.
- pp60–68: cubic EOS, mixing rules, excess-Gibbs mixing rules and LLE; polar mixtures need appropriate models, not naive averaging. LLE may exist inside turbulent columns without macroscopic settling; equilibrium calculation must account for split phases.
- pp68–74: electrolyte salting-in/out, azeotropic equality of phase compositions, pressure/temperature dependence. Industrial salt/entrainer examples are not drinking-spirit blending instructions. No extrapolation to consumer salt additions or safety.
- pp74–83: UNIFAC family offers predictions from functional groups, modified form improves several domains; limitations include isomers, complex multifunctional molecules, aqueous hydrocarbon solubility and false miscibility gaps. Authors recommend experimental checks of key-component predictions. Their participation in model development and DDBST/consortium support should be disclosed as methodological perspective, not evidence of invalidity. Status/counts refer to2014.
- pp83–89: Antoine validity range, vapor pressure, enthalpy, density, transport properties and chemical equilibria. Reusable takeaway: explicit units, reference states and validity ranges. Equation2.36 reciprocal-density mixture relation needs basis verification before reuse: surrounding symbols use mole fraction, whereas mass-density mixture averaging needs consistent weights and nonideal-volume treatment. It must not be used directly as an ABV blend calculator.
- pp89–95: residue-curve topology and boundaries, pressure swing, decanter, membrane/adsorption hybrid and solvent approaches; selection depends on mixture properties. Fig2.26 caption/text conditions conflict (101.325kPa vs323.15K), so check visual/original before quantitative reuse. Bibliography55 entries read, original references not read. No direct quotes selected.
Chapter3 — Mass Transfer in Distillation
Ross Taylor and Hendrik A. Kooijman. PDF116–162, printed97–143. DOI10.1016/B978-0-12-386547-2.00003-X.
- pp98–110: phase-interface equilibrium is not bulk-phase equilibrium. Maxwell–Stefan uses chemical-potential gradients and multicomponent interactions; simple Fick concentration-gradient descriptions have restricted validity. Thermodynamic factors and diffusivity correlations require units, composition and temperature assumptions.
- pp110–126: low/finite-flux corrections, bootstrap conditions, film and surface-renewal models. Film is a conceptual resistance model, not a literal membrane. Multicomponent reverse/osmotic diffusion and barriers are possible; this is not proof of any particular resting-whiskey mechanism. Mixing-rule approximations are not a recipe for marrying-time estimates.
- pp126–139: tray geometry, jets, bubbles, splash zones, crossflow and mixing determine performance. Packed-column countercurrent behavior differs. HTU and HETP are conditional measures, not universal apparatus constants. No single correlation fits all internals and operating regimes.
- pp139–143: 90 references read as bibliography, not as separately acquired sources.
- Equation flags for later high-resolution check:3.33,3.88,3.98,3.137,3.172. Extraction loses signs; do not label all as printed errors. Contact-sheet review of3.98 appears to show2√Ds; independent derivation needed before numerical reuse.
Chapter4 — Principles of Binary Distillation
Eva Sorensen. PDF163–203, printed145–185. DOI10.1016/B978-0-12-386547-2.00004-1.
- pp147–153: bubble/dew curves, ideal/nonideal mixtures and azeotropes. K=yi/xi and activity coefficients must be interpreted together: statement onp148 that K generally lacks composition dependence is not generally valid for nonideal mixtures (comparep152 andchapter2). The claim that activity coefficients are highest at infinite dilution is not universal, particularly for negative deviations. Heterogeneous-mixture discussion approximates immiscible systems; partially miscible phases require full equilibrium treatment.
- pp153–158: differential/Rayleigh batch and continuous flash have distinct balances. Instantaneous vapor composition is not accumulated receiver-average composition, despite loose xD wording. Rayleigh whisky example explains redistillation; does not model congener cuts or all pot-still reflux.
- pp159–166: continuous rectification involves countercurrent vapor/liquid; reflux, stages and relative volatility jointly affect separation. Historical Cellier-Blumenthal1808 statement must reconcile withchapter1's1813 patent date, not collapse into a single priority assertion. Minimum reflux/infinite stages and total reflux/minimum stages are limits, not production settings. Economics depend on utilities and capacity.
- pp166–174: McCabe–Thiele is binary, steady-state, approximate constant molar overflow/equilibrium. Feed condition q and feed position affect construction. Ideal stage counts differ from actual trays. Feed-stage optimum illustration captions/prose appear reversed; verify before reproducing.
- pp174–183: multiple feeds, side withdrawals and direct steam modify balances. Partial condensers/reboilers may count as equilibrium stages, total condensers generally do not. Fenske–Underwood–Gilliland and Kirkbride are preliminary-design shortcuts with limitations; actual design needs rigorous checks. HETP applies to packing, tray efficiencies to staged contactors.
- pp183–185: explicit acknowledgment that nonideality, multiple components and dynamic batch operation require stronger models;13 references read, not original reference texts.
- Numerical/editorial flags: PDF174 yi=Kiyi likely erroneous;179 total bottom balance sign;182 reversed q inequality;185 prose reverses V/L relationship and stripping slope;202 NC vsNC−1 column count inconsistency. Do not directly reuse extracted formulas. None is a tested whisky-design instruction.
Chapter5 — Design and Operation of Batch Distillation
Eva Sorensen. PDF204–241, printed187–224. DOI10.1016/B978-0-12-386547-2.00005-3.
- pp188–192: batch flexibility and traceability vs time-varying inventory/composition. Differential still differs from rectifying batch column with deliberate reflux. Continuous steady state and transient batch should not be conflated; azeotropes limit claims about separating arbitrary feeds into pure components.
- pp193–198: constant reflux gives changing instantaneous composition; constant composition normally needs increasing reflux. Optimal/cyclic policies, vapor loading and offcut recycle have recovery/time/capacity tradeoffs. Accumulated-average purity can hide within-run changes. General industrial offcuts are not exact equivalents of beverage heads/hearts/tails.
- pp199–205: regular, inverted, middle-vessel and multivessel arrangements; claimed performance improvements are particular studies, not general whisky savings. Optimization trajectories are open-loop until feedback measurements/controllers are included; measurement availability and nonlinear dynamics constrain practice.
- pp205–213: pressure swing, entrainers, membranes and reaction-integrated separation. Ethanol-water azeotrope figures are mole fraction and pressure-specific, not ABV/consumer-proof values. Industrial entrainer choices do not imply potable use. Reactive-distillation examples cannot establish beneficial reactions during bottle blending.
- pp213–220: shortcut/simple/rigorous/rate-based model distinctions, stiff DAEs, discrete and continuous decisions, objective-dependent optima, feasible versus infeasible parameterizations, genetic algorithms and Pareto tradeoffs. Define the goal before calling a process optimal. All future/commercial statements are2014 perspective.
- pp221–224:66 references read as bibliography only. Reflux optimization economics equation needs price-of-energy basis verification before implementation.
Chapter6 — Energy Considerations in Distillation
Megan Jobson. BeginsPDF242, printed225. DOI10.1016/B978-0-12-386547-2.00006-5. Complete through printed270/PDF287.
- pp226–237: heat, cooling and work have different quality/temperature and costs. Steam pressure levels, fired/thermal-fluid heating, refrigeration and power; site-specific examples are not current prices. Higher heating temperature and deeper refrigeration generally cost more. Life-cycle results depend on system boundary; retained for comprehensive indexing, not expansion of editorial scope.
- pp238–247: compare identical separation objectives. Pressure, pressure drop, stages/reflux, feed condition/location and condenser type interact. More trays may hurt vacuum energy through added pressure drop. Poor maintenance/over-reflux can defeat good design. None of these purity objectives guarantees better whiskey sensory quality.
- pp247–252: heat recovery requires hotter source, sufficient temperature approach and duty compatibility. Pinch analysis and total-site tradeoffs; increased column duty can coexist with lower purchased-energy cost. Hydrocarbon illustrative case explicitly has only partial heat-recovery feasibility at1000kPa.
- pp252–261: side-reboilers/condensers, double-effect, vapor recompression, HIDiC, thermal coupling, prefractionation and dividing walls. Fewer external heat duties can require higher-quality energy or capital/control complexity. Energy percentages and commercialization status are2014/case-specific. Do not equate double-effect integration with beverage double distillation.
- pp262–267: shortcut vsrigorous models, comparison metrics, reversible ideal and practical nonzero driving forces, exergy and multicomponent limits. Lowest vapor load is not necessarily lowest cost. Printed simple-payback definition appears inverted (benefit/investment rather than payback time); verify if reusing. Table6.3 currency-per-kJ likely unit issue; examples in text/GJ need visual check. Extracted subambient temperature minus signs absent; NEVER use extracted positive values.
- pp267–270: all47 references inspected as bibliography only.
Chapter 7 — Shah and Agrawal, printed 271–303 / PDF 288–320
Systematic synthesis of near-ideal multicomponent separation trains. Sharp and nonsharp splits, regular/plus/subcolumn configurations and thermal coupling enlarge the candidate space beyond simple heuristics. Search cannot discover configurations excluded at generation. Matrix enumeration counts apply within defined rules, not to every possible plant. Minimum vapor duty is a screening surrogate, not a complete measure of cost, energy quality, controllability or safety. Partial coupling and retained exchangers can be advantageous. Dividing-wall arrangements combine functional columns in one shell; they do not establish a whiskey-specific sensory advantage. Petroleum savings examples are modeled potentials, not measured whiskey outcomes. Fifty-five references read as bibliography only.
Chapter 8 — Skiborowski, Harwardt and Marquardt, printed 305–355 / PDF 321–371
Nonideal synthesis proceeds through candidate generation, shortcut evaluation and rigorous optimization. Boiling-point order is insufficient. Residue curves, distillation lines, boundaries, pressure, entrainers, phase stability and recycle determine feasibility. Overall balances and collinearity alone are insufficient; continuous composition profiles matter. Total-reflux and finite-reflux feasibility differ. More reflux is not a universal cure: extractive systems can have upper and lower feasible reflux limits. Shortcut pinch methods have assumptions and can misclassify or mis-rank alternatives; rigorous MESH solutions still depend on thermodynamics and initialization. Solvent selectivity alone ignores regeneration and heating quality. Acetone/chloroform/DMSO industrial examples are not potable-whiskey recipes. Reported energy and total-annual-cost savings differ; preserve the objective and boundary. All 137 bibliography entries read, not their underlying publications.
Chapter 9 — Babi and Gani, printed 357–381 / PDF 372–396
Hybrid means integrated separations including pressure swing, extractive, reactive and membrane operations; it is not interchangeable with marketing terminology for whiskey stills. Agent screening must include recovery, operating range and environmental/health constraints, not selectivity alone. Heuristic and driving-force methods are preliminary; property and membrane models require validation. A claim of a unique concave driving-force maximum should not be generalized to all nonideal systems. Software descriptions are 2014 context, not verified current availability. Figure 9.7 pressure/boiling-temperature labels and Table 9.7 feed/flow values merit visual checking before reuse. Thirty-two references read as bibliography only.
Chapter 10 — Kenig and Blagov, Modeling of Distillation Processes
Models simplify reality. Equilibrium models are efficient but only indirectly represent actual internals; efficiencies are empirical. Rate-based models resolve finite mass/heat transfer; CFD and hydrodynamic analogy require additional data and computational effort. Complexity must fit the question and validation evidence. Infinite-reflux/height limits are ideal feasibility tools, not real plant settings. Check sign/time orientation before comparing residue-curve equations between chapters. No exact equations are certified for implementation.
Chapter 10 completed PDF397–450, printed383–436. HTU/HETP values are conditional on flows, composition, packing and wetting, not intrinsic constants transferable unchanged. Rate-based does not eliminate equilibrium: it applies at the interface, with finite transport in the bulk/films. Maxwell–Stefan captures cross-diffusion; pseudo-binary approximations require care. Dynamic batch/start-up models require holdup and consistent initialization. Hydrodynamic analogy uses observed geometry and flow patterns, not arbitrary simplification; CFD remains dependent on interfaces and closure assumptions. Microdistillation refers to submillimeter structures, not craft production. Reactive and membrane separation chapters are industrial context, not consumer blending prescriptions. The appendix develops Franklin's transformation and Underwood-root geometry under constant relative volatility and molar overflow; this is not a universal whiskey-congener model. All 152 references inspected as references only.
Chapter 11 — José A. Caballero and Ignacio E. Grossmann
PDF451–510, printed437–496, DOI10.1016/B978-0-12-386547-2.00011-9. FUG, group and aggregate models screen candidates; MESH stage models, MINLP and GDP represent coupled discrete equipment and continuous operation decisions. The optimal solution is conditional on objectives, included alternatives, property models and mathematical assumptions. Nonconvex local solutions are not global guarantees. GDP removes equilibrium constraints from bypassed nonexistent trays, reducing numerical problems, but initialization and bounds still matter. STN distinguishes transformations from equipment; thermodynamic equivalence does not imply identical pressure losses or operability. Full thermal coupling can lower vapor requirements while demanding higher-grade heating, larger sections and more complex control. Cost, energy quantity and energy quality must remain separate. Historical industry counts and consumption figures are not current estimates. Appendix explains relaxations, OA/GBD, branch-and-bound, big-M and convex-hull formulations; claimed bounds depend on appropriate convexity/global solution assumptions. Main111 and appendix28 references read. Index PDF511–520 inspected for retrieval, not treated as new evidence.
Visual verification and publication cautions
Visual pages: 24,25,28,38,67,69,78,108,118,131,135,139,174,182,189,215,250,264,309,337,384,394,412,441,468,483,484,489 (plus covers previously).
Confirmed in printed pixels, not merely extraction: Table4.1 partially vaporized q inequality reversed; Figure4.17 caption feed-above/below mismatches diagrams/prose; Figure9.7 says1kPa with atmospheric-looking temperatures and duplicate acetone/chloroform56.1 labels; Tables9.7/9.8 disagree on280 vs230.28 flow and feed species label; Eq10.16 integrates dy for liquid expression; reflux relation printed m/(m−1) with m<1; Eq11.66 condenser inequality and Eq11.68 duty sides require correction before computation. Do not use these as calculation-ready equations. Engineering concepts remain usable with source-specific limitations; no numerical design is certified. Chapter1's broad ancient terminology and dates need independent historical corroboration before timeline use. No direct quotations selected, no figures republished.
Researcher synthesis
For our whiskey website, use this volume to explain separation, reflux, batch-versus-continuous behavior, theoretical stages, equipment interactions and model limitations. It complements rather than replaces the existing Vogelpohl source: No access and its literature note Distillation: The Theory — full-source synthesis. Do not merge their bibliographic records.
The volume does not directly test consumer blends, marrying time, sensory preference, or the flavor superiority of still types. Chemical purity is not our sensory objective. Engineering solvents and industrial purity case studies must not become drinking-spirit instructions.
Historical material provides leads for the History section, not permission to overwrite established chronology. Distinguish an apparatus interpretation, a published description, a patent, commercial adoption and beverage use. No website history pages were changed by this processing operation.
Durable ideas
Historical apparatus is not proof of a distilled drink
A mixture cannot be separated by reading a boiling-point list
Batch distillation must be understood as a trajectory
Theoretical stages are not literal distillation passes
The lowest heat duty need not be the best design
Model complexity is not validation
Assessment and reuse boundaries
Strong specialist secondary source for engineering mechanisms and design frameworks. Chapters include the authors' own methodological perspectives and illustrative case studies. Original bibliography items were not separately read merely because their references appear here. Software status, prices, industry counts, savings and forecasts are dated to the volume; verify current values before publishing current claims.
No direct quotations selected. No figures reproduced in public content. All extracted equations are non-authoritative for numerical implementation until checked against the page, definitions and an independent derivation or trusted reference. Full source processing does not certify every statement in the source as true.
Open questions
- Which specific historical apparatus/date claims survive comparison with the library's archaeological and historical sources?
- Which engineering mechanism statements have direct whiskey-congener or sensory evidence in the existing corpus?
- If a calculation is needed later, which original formula, unit basis and operating range will be independently validated?
These are future-use qualifications, not missing PDF pages or unread chapters.