Optimizacija sestave proizvoda na osnovi vodikovega peroksida

Godec, Veronika (author), Kolar, Mitja (mentor)

Cilj tega magistrskega dela je bila optimizacija sestave proizvoda na osnovi vodikovega peroksida z uporabo eksperimentalnih načrtov. Najprej sem pripravila presejalni eksperimentalni načrt. S četrtinskim faktorskim eksperimentalnim načrtom sem z 32 eksperimenti preverila, kateri izmed sedmih izbranih faktorjev signifikantno vplivajo na stabilnost vodikovega peroksida v formulaciji in na učinkovitost dezinfekcije formulacije. Z rankit metodo sem analizirala signifikantnost vplivov posameznih faktorjev in interakcij med dvema faktorjema. Rezultati analize so pokazali, da delna nevtralizacija signifikantno pozitivno vpliva na stabilnost vodikovega peroksida v formulaciji, medtem ko ima kislina 2 signifikantno negativen vpliv na stabilnost. Na učinkovitost dezinfekcije formulacije izrazito pozitivno vpliva kislina 1, signifikantno negativen vpliv pa imata delna nevtralizacija in interakcija med delno nevtralizacijo in kislino 1. Ostali faktorji in interakcije nimajo statistično pomembnega vpliva na preučevane odzive. Na podlagi dobljenih rezultatov presejalnega eksperimentalnega načrta in empiričnih opažanj sem pripravila optimizacijski eksperimentalni načrt. Za optimizacijo stabilnosti vodikovega peroksida v formulaciji, učinkovitosti dezinfekcije in penjenja formulacije sem z Box Behnkenovim eksperimentalnim načrtom za tri faktorje pripravila 16 eksperimentov. Z regresijo je program izračunal koeficiente modelov, ki sem jih nato ovrednotila s pomočjo analize variance. Glede na rezultate analize variance optimumov z dobljenimi modeli ni smiselno napovedovati, ker modeli niso statistično signifikantni. Model, ki vsebuje z regresijo izračunane koeficiente, se torej statistično pomembno ne razlikuje od modela, ki ga sestavlja le konstanta, enaka povprečnemu odzivu. Na stabilnost vodikovega peroksida v formulaciji nobeden izmed treh faktorjev nima signifikantnega vpliva, na učinkovitost dezinfekcije signifikantno vpliva le kislina 1, na penjenje pa le antipenilec. Kljub temu smo s pomočjo eksperimentalnih načrtov pridobili veliko novih informacij o formulacijah.

• Language: Slovenian
• Keywords: vodikov peroksid, optimizacija, eksperimentalni načrt, delni faktorski eksperimentalni načrt, Box-Behnkenov eksperimentalni načrt
• Work type: Master's thesis/paper
• Typology: 2.09 - Master's Thesis
• Organization: FKKT - Faculty of Chemistry and Chemical Technology
• Year: 2024
• PID: 20.500.12556/RUL-154569
• COBISS.SI-ID: 186459139
• Publication date in RUL: 21.02.2024

Secondary language

• Language: English
• Title: Optimization of the hydrogen peroxide-based product composition

Abstract

Objective of this master's thesis was optimization of the hydrogen peroxide-based product composition with design of experiments application. Firstly, screening design of experiments was used. One-quarter fraction factorial design with 32 experiments was used to investigate which of the chosen seven factors significantly affect stability of hydrogen peroxide in formulation and disinfectant efficacy of formulation. Rankit method was applied to analyze significancy of individual factor effects and two-factor interaction effects. The analysis results showed that partial neutralization has significant positive effect on stability of hydrogen peroxide in formulation whereas acid 2 has significant negative effect on stability. Acid 1 has markedly positive effect on disinfectant efficacy whereas partial neutralization and interaction between partial neutralization and acid 1 have significant negative effect on disinfectant efficacy. The other factors and interactions do not exhibit meaningful impact on the studied responses. Based on the given results of screening design and empirical observations, optimization design of experiments was prepared. To optimize stability of hydrogen peroxide in formulation, disinfectant efficacy and foaming of formulation Box-Behnken design with three factors was used to prepare 16 experiments. Model coefficients were calculated by software using regression and then evaluated by analysis of variance. According to analysis of variance results, optimum prediction with these models is not logical since the models are not statistically significant. A model containing coefficients calculated by regression is therefore not statistically significantly different from a model containing only a constant equal to the mean response. None of the three factors significantly affect stability of hydrogen peroxide in formulation, only acid 1 has significant effect on disinfectant efficacy and only antifoam agent has significant effect on foaming. Nevertheless, a lot of new information about formulations was gained.

• Keywords: hydrogen peroxide, optimization, design of experiments, fractional factorial design, Box-Behnken design