{"id":3119,"date":"2026-08-06T15:24:25","date_gmt":"2026-08-06T13:24:25","guid":{"rendered":"https:\/\/gekkophotonics.com\/?post_type=case-study&p=3119"},"modified":"2026-08-06T15:46:12","modified_gmt":"2026-08-06T13:46:12","slug":"liczba-kwasowa-zywic-alkidowych-kontrola-punktu-koncowego-estryfikacji","status":"publish","type":"case-study","link":"https:\/\/gekkophotonics.com\/en\/case-study\/liczba-kwasowa-zywic-alkidowych-kontrola-punktu-koncowego-estryfikacji\/","title":{"rendered":"Acid number of alkyd resins: real-time control of the esterification endpoint"},"content":{"rendered":"
\n
\n
Industry<\/dt>\n
Alkyd resin production<\/dd>\n<\/div>\n
\n
Scope<\/dt>\n
Esterification endpoint control: acid value and viscosity<\/dd>\n<\/div>\n<\/dl>\n

Client context<\/h2>\n

\u017bywice alkidowe to polimery estrowe produkowane przez estryfikacj\u0119 bezwodnika ftalowego, kwas\u00f3w t\u0142uszczowych i glicerolu w reaktorze pracuj\u0105cym powy\u017cej 220 stopni Celsjusza. S\u0105 baz\u0105 farb olejnych, alkidowych i alkidowo-akrylowych.<\/p>\n

Quality and repeatability are determined by hitting the esterification endpoint, which is directly governed by two parameters: acid value and viscosity. The acid value decreases as the reaction progresses, and its control is challenging because the kinetics vary depending on the batch of drying oils.<\/p>\n

Challenge.<\/h2>\n
    \n
  • The acid value was determined offline by titration (ISO 3682), requiring 30\u201360 minutes, which necessitates stopping the reactor and manually sampling the medium above 220 degrees.<\/li>\n
  • Variable reaction kinetics make it difficult to predict the end of esterification, and any overshoot of the target requires correction with excess solvent or results in an off-spec batch.<\/li>\n
  • The medium contains flammable organic solvents (white spirit D40, xylene), so the measurement must operate in an ATEX-rated zone.<\/li>\n<\/ul>\n

    Solution<\/h2>\n

    We performed the measurement using a Spectrally\u2122 X1 analyzer with an in-line immersion probe (785 nm laser), without sample extraction and under process-like conditions. The acid value has no single, dedicated spectral band, so we built a chemometric model (CNN network) correlating the entire Raman spectrum with the parameter, based inter alia on the ratio of acid\/ester carbonyl group bands.<\/p>\n

    The study was conducted in two stages: first at-line on samples, then with an immersion probe under process-like conditions. We also honestly noted the limitations, namely the labor-intensive model building and sensitivity to signal stability, which we are addressing before pilot implementation.<\/p>\n

    Results<\/h2>\n
    \n\n\n\n\n\n\n\n
    Metric<\/th>\nResult<\/th>\n<\/tr>\n<\/thead>\n
    Model vs. laboratory correlation (acid value)<\/td>\nR = 0.982\u20130.998; MAE error approximately 0.015 mg KOH\/g<\/td>\n<\/tr>\n
    Additional parameter from the same spectrum<\/td>\nviscosity, R = 0.988\u20130.998<\/td>\n<\/tr>\n
    Time to result<\/td>\nreal-time, instead of 30\u201360 min of titration<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

    Result of a feasibility study (technical proof), not implementation. Client anonymized.<\/p>\n

    What this means for your process<\/h2>\n

    If you produce alkyd resins or varnishes and control esterification based on titration, the same measurement allows you to move endpoint control directly into the reactor and obtain both acid value and viscosity simultaneously from a single spectrum, also in ATEX zones. This shortens the cycle and reduces the risk of off-spec batches.<\/p>\n

    The most reliable way to verify on a specific formulation is a short feasibility study on real samples from the given process.<\/p>","protected":false},"excerpt":{"rendered":"

    In a feasibility study at a resin manufacturer, we demonstrated that in-line Raman measurement predicts the acid value, and additionally the viscosity, with accuracy approaching laboratory levels, without titration and without stopping the reactor.<\/p>","protected":false},"author":1,"featured_media":3135,"menu_order":0,"comment_status":"closed","ping_status":"","template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":"","industry-grid":""},"class_list":["post-3119","case-study","type-case-study","status-publish","format-standard","has-post-thumbnail","hentry"],"blocksy_meta":[],"acf":[],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/gekkophotonics.com\/en\/wp-json\/wp\/v2\/case-study\/3119","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/gekkophotonics.com\/en\/wp-json\/wp\/v2\/case-study"}],"about":[{"href":"https:\/\/gekkophotonics.com\/en\/wp-json\/wp\/v2\/types\/case-study"}],"author":[{"embeddable":true,"href":"https:\/\/gekkophotonics.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/gekkophotonics.com\/en\/wp-json\/wp\/v2\/comments?post=3119"}],"version-history":[{"count":3,"href":"https:\/\/gekkophotonics.com\/en\/wp-json\/wp\/v2\/case-study\/3119\/revisions"}],"predecessor-version":[{"id":3163,"href":"https:\/\/gekkophotonics.com\/en\/wp-json\/wp\/v2\/case-study\/3119\/revisions\/3163"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/gekkophotonics.com\/en\/wp-json\/wp\/v2\/media\/3135"}],"wp:attachment":[{"href":"https:\/\/gekkophotonics.com\/en\/wp-json\/wp\/v2\/media?parent=3119"}],"curies":[{"name":"entry","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}