Investigation of the effect of sample preparation methodology on accuracy and repeatability in honey color measurement by spectrophotometric technique

The color of honey is a fundamental physical attribute that influences numerous factors ranging from consumer preferences and marketing strategies to quality classification and legal standards. Therefore, accurate, repeatable, and standardized determination of honey color is of great importance for both scientific research and industrial applications. In this thesis study, the effects of sample preparation steps (dilution, temperature, sonication, centrifugation, and waiting time) on CIELAB color parameters and the CIEDE2000 (ΔE₀₀) color difference were systematically investigated to enable the objective, accurate, and repeatable measurement of honey color using spectrophotometric techniques. Honey samples were prepared at solid content levels of 30%, 40%, 50%, and 60%, and analyzed at 25, 35, and 45 °C. The measurements were conducted based on tristimulus (X, Y, Z) values obtained from transmission spectra recorded within the range of 380–780 nm. The flower honey samples used in the study had a moisture content of 20% and were obtained from different colonies; no statistically significant difference was found between their moisture levels (p > 0.05), and all samples were determined to be homogeneous and suitable for analysis. In the first stage of the research, the effects of dilution ratio and measurement temperature on color parameters were examined. As the solid content increased, L* values decreased while a* and b* values increased; the ΔE₀₀ color difference between the 30% and 60% samples was calculated as 8.39. The 30% solid content provided better optical transmittance and yielded more stable measurement results. An increase in temperature led to a significant rise in L* values, particularly in samples with lower concentration. Centrifugation was found effective for particle removal, resulting in an increase in L* and a decrease in a* values. A ΔE₀₀ value of 19.21 was reached after 30 minutes of centrifugation at 10000 ×g, indicating a pronounced impact on color measurements. Sonication led to increases in L*, b*, and C* values; a ΔE₀₀ value of 5.05 was recorded at 360 seconds. After 120 seconds, changes in color parameters began to slow, and optical equilibrium was reached at 180 seconds. Waiting time between measurements (0–60 minutes) revealed optical instability, particularly within the first 10 minutes. Increasing trends in L*, a*, b*, and C* values were observed; the ΔE₀₀ value exceeded 2 in the first 2 minutes and reached 3.96 by the 60th minute. These findings underscore the necessity of controlling measurement timing with precision. With respect to measurement repeatability, higher solid content and lower temperature yielded more stable results. A sonication duration of 120–180 seconds was found sufficient, and centrifugation at 10000 ×g for 20–30 minutes provided more consistent particle removal compared to other conditions. Although the Pfund scale remains a fast and economical tool for routine honey color classification, the results of this study revealed that perceptible color differences (ΔE₀₀ > 2) may still occur among samples classified under the same Pfund category. This highlights the limited sensitivity of the Pfund method in detecting subtle color variations. Therefore, integrating ΔE₀₀ color difference calculations within the CIELAB color space and applying standardized sample preparation protocols will improve both the accuracy and comparability of honey color measurements. Overall, the findings demonstrated that the standardization of sample preparation steps is critically important for ensuring both accuracy and repeatability in the spectrophotometric determination of honey color.