Viscosity of cooking oils affects dishwasher cleaning performance Authors: Shaojin Ma, Yan Bai, Jiajia Shao, Xue Bai Published: 2026 ABSTRACT Edible oil is a primary energy source in human diets. This study investigates the effects of heating temperature on the physicochemical properties of edible oils and their impact on dishwasher cleaning performance. First, the density and dynamic viscosity of peanut oil, rapeseed oil, and soybean oil were measured at room temperature and after heating to 50, 100, 150, 200, and 250 °C. Accordingly, two oils with the greatest viscosity difference—soybean oil at room temperature (SO_25) and peanut oil heated to 250 °C (PO_250)—were selected for dishwasher cleaning performance tests. Residue penalty score (RPS) and re-deposition penalty score (RDPS) were defined to characterize the cleaning ability and anti-redeposition performance of dishwashers. The results indicated that PO_250 generally exhibited lower cleaning scores. In the quick wash mode, the RPS and RDPS of PO_250 and SO_25 were relatively similar; however, in the standard wash mode, both RPS and RDPS of PO_250 were higher than those of SO_25. Statistical analysis further revealed that changes in washing mode have a greater impact on cleaning scores for high-viscosity oil than for low-viscosity oil. Moreover, low-viscosity oil exhibited greater inter-machine variability in dishwasher cleaning scores compared with high-viscosity oil. Scanning electron microscopy (SEM) imaging showed that PO_250 formed an oil film on the ceramic surface, whereas SO_25 appeared as discrete particles. This study provides a scientific basis for developing regionally applicable dishwasher testing standards, contributing to the optimization of cleaning technologies.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-56996-3. Introduction Edible oils are one of the major energy resources for the human diet that play an indispensable role in the present agri-food system1. The sources of edible oils can be divided into two main categories: one comes from plants, commonly known as vegetable oils, such as sunflower oil, rapeseed oil, soybean oil, palm oil, coconut oil, etc., the other comes from animals, such as butter and lard. Vegetable oils are popular for their low cholesterol properties and health benefits. In contrast, animal fats are rich in saturated fatty acids, which make them less reactive with other chemicals (especially oxygen). On the other hand, the unsaturated fatty acids in vegetable oils have higher reactivity and are more easily digested by the human body2. Vegetable oils are indispensable worldwide, not only due to their everyday nutritional uses but also because of their strong connection to local environments—different regions produce various oilseed crops, each with its unique flavor and quality3,4. In daily diets, one common way of consuming vegetable oils is through heating, such as microwave heating or deep frying5. Heating can alter the physicochemical properties of vegetable oils. The chemical changes in cooking oils during heating can be broadly divided into three consecutive stages: First, primary oxidation products (namely hydroperoxides) are generated through various chemical reactions. These hydroperoxides then decompose to form secondary oxidation products such as aldehydes and ketones. Ultimately, tertiary oxidation products in the form of short-chain free fatty acids are produced. The reaction rate at each stage may be accelerated or decelerated by different influencing factors6. Additionally, physical properties such as color, foaming and viscosity of vegetable oil can also change due to deep frying. For instance, the color of oil darkens during the heating process, specifically manifested by a decrease in L (luminance or brightness of the image) and increases in a* (amount of red or green tones in the image. A large positive a* value corresponds to red/magenta) and b* (Amount of yellow or blue tones in the image. A large positive b* value corresponds to yellow.)7. When frying potato slices at boiling point temperature for varying durations, the viscosity of corn oil increases while that of mustard oil decreases8. These alterations in physical characteristics are correlated with chemical compositional changes in the vegetable oils9,10. Household dishwashers have become an indispensable partner in modern kitchens, making significant contributions to reducing manual labor. Different countries have established varying performance standards for dishwashers. For instance, EU regulations stipulate that dishwasher performance can be expressed through cleaning performance index (CPI), drying performance index, electricity consumption, water consumption, and cycle duration11. In contrast, China’s national standards exclude the time duration metric when grading dishwasher performance, utilizing only CPI, drying performance index, electricity consumption, and water consumption for classification. According to China’s national standards, the testing methodology for dishwasher cleaning performance is fundamentally aligned with the IEC standards (IEC 60436), which also serve as the reference testing protocol for the EU. The IEC standard stipulates the use of seven standardized food soiling (e.g., milk, tea, and margarine) and multiple standard tableware to evaluate the appliance’s cleaning and drying performance on these items. However, significant differences exist in dietary habits across countries. Using identical food soiling for dishwasher performance evaluation may lead consumers to question the appliance’s cleaning effectiveness in daily use. Particularly, margarine is rarely consumed as a primary fat source in China12. Heated vegetable oil represents the predominant form of dietary lipid intake in Chinese cuisine. Therefore, investigating dishwashers’ cleaning efficacy on heated vegetable oil is crucial for establishing performance testing standards aligned with China’s national context, and can promote the development of cleaning technologies that better accommodate local dietary habits by appliance manufacturers. In this work, the feasibility of using heated vegetable oil as a representative soiling for evaluating dishwasher cleaning performance was investigated. Peanut oil, rapeseed oil, and soybean oil were selected as test samples. The main contributions of this paper are outlined as follows: (a) The changes in density and viscosity of the three vegetable oils under different cooking temperatures (from room temperature to 250 °C) and after drying heating (80 °C for 2 h) were measured. (b) Two typical oils with the most distinct viscosity were used to soil tableware, and their cleaning performance was evaluated under both quick wash and standard wash modes in three household dishwashers. (c) Statistical analyses were applied to quantify the sensitivity of different oils to wash modes and dishwasher models. (d) SEM was employed to observe the microscopic morphology of oil residues on ceramic surfaces. Materials and methods Oil sample According to the survey results, peanut oil (PO), rapeseed oil (RO), and soybean oil (SO) are the three most popular edible oils in China (https://www.wjx.cn/report/321705173.aspx?sat=1). Therefore, these three oils were used as samples in the study. PO, RO, and SO were all purchased from online platforms. A total of two types of heated oil samples were prepared: cooking heated oil and drying heated oil. The purpose of preparing drying heated oil samples was to evaluate the impact of the drying process, as specified in the IEC standards, on the oil’s properties. When preparing the cooking heated oil, 240 g of edible oil was poured into a non-stick pan (EC30TYJ01, Zhejiang Supor Co., Ltd., Hangzhou, China) and heated using an induction cooker (C22S-N303, Hangzhou Joyoung Household Appliances Co., Ltd., Hangzhou, China) with a rated power of 2200 W. During the heating process, the temperature of the oil was monitored using a probe thermometer (THM C1, Delixi Electric Ltd, Zhejiang, China). When the temperature displayed on the thermometer reached the target temperature, heating was immediately stopped, and the pan was removed from the induction cooker. The oil was heated to 50, 100, 150, 200, and 250 °C, along with unheated oil, for subsequent preparation of drying heated oil. When preparing the drying heated oil samples, 30 mL of cooking heated oils and including unheated oil were transferred into a glass beaker and dried in a thermal cabinet (UN 450, Memmert (Shanghai) Trading Co., Ltd., Shanghai, China) at 80 °C for 2 h. The drying process for the heated oil was the same as the thermal drying procedure used during the dishwasher cleaning performance test. Both cooking heated oil and drying heated oil were cooled to room temperature after heating. They were stored in dark brown sealed glass bottles and refrigerated at 4 °C until analysis. During analysis, the samples were first removed from the refrigerator and allowed to reach room temperature before proceeding. Physical properties measurement Density The density of oils was calculated using a gravimetric method. The tests were carried out at room temperature (25 ± 2℃). The density of the oil was obtained by using Eq. (1):1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho =\frac{m}{V}{{ \times 1000}}$$\end{document} Where ρ is the density of oil sample (kg/m3). For each oil sample, 3 mL was transferred using a pipette into a bowl placed on a balance, and its mass, m (± 0.01 g) was recorded. Therefore, V (cm3) in Eq. (1) equal to 3. Duplicate measurements were conducted for all samples. Rheological properties The rheological characterization of the three types of vegetable oil before and after heating was investigated by using an Anton Paar MCR 301 rheometer (Anton Paar, Graz, Austria). The rheometer was equipped with a coaxial cylinder test system (C-CC27/T200). The oil sample was poured into the cylinder up to the marked line. At 20 °C, the viscosity of the oil was measured under a constant shear rate of 50 s⁻¹ in an isothermal and steady-state condition. Data was collected every 6 s for a total of 60 measurements. The average viscosity and shear stress values from the 55th to the 60th measurements were taken as the viscosity and shear stress of the oil sample. Duplicate measurements were conducted for all samples. Preparation of oil-soiled tableware Soup spoon and oval platters made of porcelain (National Reference Material for Electronics & Electrical Appliances Testing Industry Incubation) were used as the tableware to be separately soiled with pure oil and spinach oil mixture. The preparation method for the soup spoon contaminated with pure oil is as follows. A total of 20 g of pure oil was poured into a bowl. The soup spoon was dipped into the oil, ensuring that both the interior and exterior surfaces of the spoon were thoroughly coated. The oil-contaminated soup spoon was then placed in a plate, ready for subsequent dishwasher washing tests. For the preparation of the spinach-oil mixture, 1.50 g of freeze dried spinach (Sta-De Testmaterialien, Lauf, Germany) were weighed and mixed with 20 mL of water to ensure complete hydration of the spinach fragments. A portion of 7 g of the wet spinach fragments was then spread onto the surface of the oval platters. Following this, 2 mL of oil was added to the platters, and the spinach fragments were thoroughly mixed with the oil and evenly applied to the inner surface of the platters. The spinach-oil contaminated oval platters were subsequently placed into the thermal cabinet (UN 450, Memmert (Shanghai) Trading Co., Ltd., Shanghai, China) and dried at 80 °C for 2 h. The contamination methods for the soup spoon and oval platter mentioned above were adapted from the dishwasher testing guidelines and the preparation methods for margarine-contaminated soup spoons and margarine-spinach mixture-contaminated oval platters in the national standard GB/T 20,290 − 2016. Cleaning performance test After determining the physical characteristics of all edible oil samples, the two oil samples with the greatest differences were selected for cleaning performance testing. The cleaning performance test was conducted on three dishwashers (referred to as Dishwasher A (DW_A), Dishwasher B (DW_B), and Dishwasher C (DW_C)). DW_A (HRM1201, Haier, Qingdao, China) and DW_B (Reference, Míele, Gütersloh, Germany) each has 12 sets, so 6 soiled soup spoon and 1 soiled oval platter were prepared according to the method described in Sect. 2.3. DW_C (M30, Midea Group Laundry Appliance Manufacturing Co., Ltd., Fo Shan, China) has 6 sets, so 3 soiled soup spoon and 1 soiled oval platter were prepared. To ensure that the water’s interaction with the oil-soiled tableware surface during testing is as consistent as possible with the standard testing procedure, all tableware to be tested was placed in the designated position in the dishwasher along with the oil-soiled tableware. The arrangement of dishware and the amount of dishwashing powder used followed the guidelines specified in the dishwasher’s instruction manual. Details of tableware prepared for different types of dishwashers are presented in Table 1. Table 1Details of soiled tableware prepared for different types of dishwashers.Items beingcleanedType of soilNumber of items in Dishwasher1Number of items in Dishwasher2Number of items in Dishwasher3Rice bowl-61212Soup bowl-366Glass-366Mug-366Chopsticks-122424Soup spoonOil366Soup spoon-366Deep plate-366Shallow plate-366Condiment dish-366Large bowl-111Oval platterSpinach-oil mixture111Rice spoon-111Large soup spoon-111 Table 1Details of soiled tableware prepared for different types of dishwashers.Items beingcleanedType of soilNumber of items in Dishwasher1Number of items in Dishwasher2Number of items in Dishwasher3Rice bowl-61212Soup bowl-366Glass-366Mug-366Chopsticks-122424Soup spoonOil366Soup spoon-366Deep plate-366Shallow plate-366Condiment dish-366Large bowl-111Oval platterSpinach-oil mixture111Rice spoon-111Large soup spoon-111 Details of soiled tableware prepared for different types of dishwashers. The cleaning performance of the two oil samples was tested under the quick wash and standard wash modes on three dishwashers, with 10 repeated experiments for each mode, resulting in a total of 120 wash cycles (2 types of oil × 3 dishwashers × 2 modes × 10 repeated experiments). After each wash cycle, visual assessment was used to check whether there are any residual stains or re-deposited stains on each piece of dishware. Each item was assessed and scored individually. The scores range from 0 to 5, with 5 indicating no residue. The specific scoring method is outlined in Table S1. The average cleaning score for one test run was used to assess and compare the washing performance of dishwashers. The calculation formula is as follows:2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${C_i}=\frac{1}{N}\sum\limits_{{z=1}}^{{18}} {{C_z}}$$\end{document} where Ci is the average cleaning score for one test run. N is the number of all items. 18 represents the category of dishware. Cz represents the total cleaning score for the z-th category of dishware. The formula for calculating N is as follows::3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$N=\sum\limits_{{z=1}}^{{18}} {{n_z}}$$\end{document} where nz is the number of items in a particular category of dishware. The N value for DW_A and DW_B is 88, while for DW_C, it is 46. The formula for Cz is as follows::4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${C_z}=\sum\limits_{{c=0}}^{5} {{a_c}} {{ \times }}c$$\end{document} where c is the cleaning score, and ac is the number of items with a cleaning score of c. Equation (1) to (3) are referenced from GB/T 20,290–2016. Additionally, to evaluate the frequency and severity of residual stains and re-deposited stains, and to normalize results across dishwashers with different capacities, the Residual Penalty Score (RPS) and Re-Deposited Penalty Score (RDPS) were defined, with the calculation formulas as follows:5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$RPS=\frac{{\sum\limits_{{i=1}}^{{{M_{rs}}}} {(5 - S_{i}^{{rs}})} }}{{{N_{soiled}}}}$$\end{document}6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$RDPS=\frac{{\sum\limits_{{j=1}}^{{{M_{rds}}}} {(5 - S_{j}^{{rds}})} }}{{{N_{clean}}}}$$\end{document} where Mrs and Mrds are the total number of tableware items with residual stains and with re-deposited stains, respectively. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{S}_{i}^{\mathrm{rs}}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{S}_{j}^{\mathrm{r}\mathrm{d}\mathrm{s}}$$\end{document} are the observed scores (0–5) of the i-th item with residual stains (which only occur on dishware contaminated by an oil and spinach-oil mixture) and the j-th item with re-deposited stains (which only occur on dishware that was clean prior to washing), respectively. Nsoiled and Nclean represent the number of contaminated and clean dishware items before washing, respectively. A RPS and RDPS score of 0 indicates that the soiled tableware after washing receives a full score, and the clean tableware after washing receives a full score, respectively. Statistical analysis To quantify the differences in cleaning performance, the following statistical analyses were conducted: The sensitivity of each oil to wash modes was evaluated by comparing the cleaning scores of SO_25 and PO_250 between the quick wash and standard wash modes using paired t‑tests.The sensitivity of each oil to different dishwashers was evaluated by comparing the cleaning scores of SO_25 and PO_250 among the three dishwashers under the same wash mode using independent t‑tests for each pair (DW_A vs. DW_B, DW_A vs. DW_C, and DW_B vs. DW_C). The sensitivity of each oil to wash modes was evaluated by comparing the cleaning scores of SO_25 and PO_250 between the quick wash and standard wash modes using paired t‑tests. The sensitivity of each oil to wash modes was evaluated by comparing the cleaning scores of SO_25 and PO_250 between the quick wash and standard wash modes using paired t‑tests. The sensitivity of each oil to different dishwashers was evaluated by comparing the cleaning scores of SO_25 and PO_250 among the three dishwashers under the same wash mode using independent t‑tests for each pair (DW_A vs. DW_B, DW_A vs. DW_C, and DW_B vs. DW_C). The sensitivity of each oil to different dishwashers was evaluated by comparing the cleaning scores of SO_25 and PO_250 among the three dishwashers under the same wash mode using independent t‑tests for each pair (DW_A vs. DW_B, DW_A vs. DW_C, and DW_B vs. DW_C). For all t‑tests, 95% confidence interval (CI) were calculated to estimate the magnitude and precision of the differences. Statistical significance was accepted at p < 0.05. To describe the magnitude of the effect, the effect size (Cohen’s d) was used, which measures the strength of the phenomenon under study by calculating the number of standard deviations by which the means are separated13. Cohen’s d is calculated as follow:7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$d=\frac{{mean({C_1}) - mean({C_2})}}{{\sqrt {\frac{{S_{1}^{2}+S_{2}^{2}}}{2}} }}$$\end{document} When calculating the cleaning scores of SO_25 and PO_250 between the quick wash and standard wash modes, C1 and C2 represent the cleaning scores under the quick wash and standard wash modes, respectively, while S1 and S2 represent the standard deviations of the cleaning scores under the quick wash and standard wash modes, respectively. When calculating the cleaning scores of SO_25 and PO_250 among the three dishwashers under the same wash mode, C1 and C2 represent the cleaning scores of two dishwashers (corresponding to DW_A and DW_B, DW_A and DW_C, and DW_B and DW_C, respectively), while S1 and S2 represent the standard deviations of the cleaning scores of the two dishwashers. Scanning electron microscopy (SEM) measurement SEM studies were carried out using SEM (S-3400 N, HITACHI, Tokyo, Japan). Porcelain chip samples were placed on a specimen holder taped with double-sided scotch and sputter-coated with gold. Samples were transferred to a microscope after coating and were observed at 15.0 kV. To observe the microscopic morphology of oil stains on ceramic surfaces, customized cylindrical porcelain chips with a diameter of 10 mm and a thickness of 2.5 mm were prepared. The material of these chips was identical to that of standard tableware. One side of the chip was smooth, while the other side was rough. Three types of porcelain chip samples were prepared: chips contaminated with oil of the highest viscosity, chips contaminated with oil of the lowest viscosity, and clean chips (control group). For preparing the oil-contaminated chips, the chip was dipped in plant oil and then placed with the smooth side facing up in a glass beaker. The beaker containing the chip and oil was then placed in the aforementioned oven and dried at 80 °C for 2 h. The clean chips were also placed in the oven for simultaneous drying. After drying, the porcelain chips were removed from the beaker, and the rough surface was gently dabbed with lens cleaning paper (Hangzhou Special Paper Co., Ltd., Hangzhou, China) to absorb the oil, ensuring that the rough surface was clean to prevent slippage during testing. Subsequently, the porcelain chip was left standing vertically for 4 h, allowing the oil on the smooth surface to flow onto the cleaning paper. This process utilized gravity to remove excess and loosely attached liquid oil, ensuring that the surface observed was a relatively uniform, thin, and fixed film or residue, rather than a dynamic liquid layer. Results Physical properties of oil at different cooking temperatures Figure 1 shows the changes in density of PO, RO, and SO during cooking heating and drying heating. The absolute differences in density between PO, RO, and SO at room temperature and after heating to 250 °C were 90.00, 166.67, and 110.00 kg/m³, respectively. The density change patterns of PO and RO oils with temperature are quite similar. When heated from room temperature to 50 °C, their densities increase by 60 and 133 kg/m³, respectively, while the changes at 100 °C, 150 °C, 200 °C, and 250 °C were less pronounced. The density of all three oils increases under drying heating conditions (80 °C for 2 h) compared to before heating. Fig. 1Density of PO (A), RO (B), and SO (C) during cooking heating and drying heating. Density of PO (A), RO (B), and SO (C) during cooking heating and drying heating. Figure 2 shows the changes in rheological properties of cooking heated and drying heated samples of PO, RO, and SO. Figure 2 (A1), (B1), and (C1) show the viscosity curves as a function of temperature. It can be observed that for the crude oils, PO exhibits the highest viscosity, while SO has the lowest viscosity. During the cooking heating process, the viscosity of the three oils gradually increased when heated to 50, 100, 150, and 200 °C. However, when heated to 250 °C, the viscosity of all three oils showed a significant increase compared to the lower temperature treatments. Among these three oils, SO’s viscosity is the most sensitive to heating, with its viscosity increasing by 6.02 mPa·s when heated to 250 °C compared to room temperature. The viscosity sensitivity to heating for PO and RO is similar. PO’s viscosity increased by 3.71 mPa·s when heated to 250 °C, while RO’s viscosity increased by 3.81 mPa·s under the same heating condition. This is somewhat different from the findings of14, who indicated that peanut oil has better thermal stability, while soybean oil and rapeseed oil have relatively poorer thermal stability. Fig. 2Viscosity of PO (A1), RO (B1), and SO (C1) and shear stress of PO (A2), RO (B2), and SO (C2). Viscosity of PO (A1), RO (B1), and SO (C1) and shear stress of PO (A2), RO (B2), and SO (C2). In addition, previous studies have reported inconsistent findings regarding changes in oil viscosity during the heating process. For instance, observed that short-term heating (180 °C for 30 min) increased the viscosity of RO, sunflower oil, maize oil, palm oil, and coconut oil (CO). They attributed this increase to the intermolecular friction and chemical reactions triggered by the heat exposure16. Also reported that viscosity increased during repeated frying. However, it should be noted that their study did not involve simply heating the oil; rather, potatoes were fried in the oil. The researchers observed that the potato-to-oil ratio affected the viscosity of the oil. Contrary to the conclusions drawn in the two studies discussed above, Nduka et al.17 heated soybean oil, peanut oil, and other oils at 185 °C for 1–7 h and found that the viscosity of the oils showed no consistent increasing or decreasing trend with heating time and temperature. The difference between the above studies and the present study is that the heating temperatures used in those studies did not vary over a wide range. In addition, the heating method used in this study involved heating the oil from room temperature to a target temperature followed by immediate cooling, which differs from the heating procedures employed in the studies mentioned above. It is speculated that the reason why the increase in the viscosity of the three oils measured after heating to 200 °C and subsequent cooling, compared with that at room temperature, was much smaller than the increase observed after heating to 250 °C and subsequent cooling is that some polymers are not generated in significant amounts until the oil temperature reaches 200–300 °C19. Dry heating (80 °C for 2 h) resulted in a slight increase in the viscosity of the oils20. demonstrated that the viscosity of fats increases with the number of reheating cycles, which may indirectly suggest that placing the oils into the oven for drying heating at 80 °C for 2 h after cooking heating causes the viscosity of all three vegetable oils to show an increasing trend. Figure 2 (A2), (B2), and (C2) present the shear stress curves as a function of temperature, showing that the trend in shear stress variation closely mirrors that of viscosity with respect to temperature. To select the two oils with the greatest differences in physical properties, one-way analysis of variance (ANOVA) was performed on the density, viscosity, and shear stress of the three vegetable oils at different temperatures. Table 2 presents the ANOVA results. The p-value for density was greater than 0.05, whereas the p-values for viscosity and shear stress were both less than 0.05, with viscosity showing the smallest p-value (1.3623 × 10⁻⁶). This indicates that there are significant differences in viscosity among the three vegetable oils at different temperatures. Therefore, in the subsequent cleaning performance tests, the two oils with the highest and lowest viscosities were selected. Table 2ANOVA results for the effect of temperature on the density and rheological properties of three vegetable oils.DensityViscosityShear stressp-value0.19171.3623 × 10− 61.4624 × 10− 6 Table 2ANOVA results for the effect of temperature on the density and rheological properties of three vegetable oils.DensityViscosityShear stressp-value0.19171.3623 × 10− 61.4624 × 10− 6 ANOVA results for the effect of temperature on the density and rheological properties of three vegetable oils. Effect of oil cooking temperature on cleaning performance From the results in Sect. 3.1, it can be seen that SO has the lowest viscosity at room temperature, while peanut oil heated to 250 °C has the highest viscosity. Therefore, the oil sample with the minimum viscosity at room temperature (denoted as: SO_25) and the maximum viscosity value (denoted as: PO_250) were used to contaminate the dishware, and washing was then performed. Figure 3 shows the average cleaning scores after a single test of all dishware using three dishwashers in both quick wash and standard wash modes. It can be seen that, in quick wash mode, the overall cleaning performance of the three dishwashers for different viscosity oil stains follows a similar pattern. That is, contaminating the dishware with SO_25, which has the lowest viscosity, results in a higher average cleaning score for one test run than contaminating with PO_250, which has the highest viscosity. In other words, the oil with lower viscosity (SO_25) is easier to clean than the oil with higher viscosity (PO_250). However, in standard wash mode, the results differ somewhat from those in quick wash mode. In the standard wash mode, the difference in overall cleaning performance between the two oils of different viscosities was significantly reduced. Additionally, the differences in cleaning performance among the different dishwasher models for oils of varying viscosities become smaller. Fig. 3Comparison of average cleaning scores for dishwashers under quick wash (A) and standard wash (B) programs with oils of varying viscosities. Comparison of average cleaning scores for dishwashers under quick wash (A) and standard wash (B) programs with oils of varying viscosities. When evaluating the cleaning score of each piece of tableware, it was found that re-deposited stains frequently appeared on the surface of clean tableware. Figure 4 shows the RPS and RDPS distribution for the three dishwashers under the quick wash and standard wash modes. RPS and RDPS reveal the interaction between the physical properties of oils and the dishwasher’s cleaning performance in terms of cleaning power and resistance to re-deposition. The higher the RPS and RDPS values, the worse the stain removal ability and the weaker the resistance to re-deposition, respectively. Fig. 4Distribution of Residual Stains Penalty Score (RPS) and Re-Deposited Stains Penalty Score (RDPS) for dishwashers under quick wash ((A) and (C)) and standard wash ((B) and (D)) programs. Distribution of Residual Stains Penalty Score (RPS) and Re-Deposited Stains Penalty Score (RDPS) for dishwashers under quick wash ((A) and (C)) and standard wash ((B) and (D)) programs. Under the quick wash mode, high-viscosity oils generally exhibited higher RPS and RDPS values than low-viscosity oils. However, the difference in RPS between the two viscosity levels was less pronounced than that in RDPS. This indicates that viscosity has a relatively minor effect on the removal of stains, but a more substantial impact on their redeposition onto other tableware. Specifically, oils with higher viscosity are more likely to redeposit onto other items. For the standard wash program, high-viscosity oils also showed overall higher RPS and RDPS values compared to low-viscosity oils, with both differences being more pronounced. Compared with the quick wash mode, the stain removal capability of the standard wash mode (as reflected by RPS) is more sensitive to variations in oil viscosity. Table S2 presents the statistical results of RPS and RDPS for each dishwasher under different washing programs and oil samples. Statistical analysis results Table 3 presents the statistical analysis of the cleaning score for low-viscosity oil (SO_25) and high-viscosity oil (PO_250) under the quick wash and standard wash modes. It can be observed that the effect size of switching washing modes is moderate to large for the low-viscosity oil, with absolute values ranging from 0.5 to 0.8, while a large effect size (absolute value > 0.8) is observed for the high-viscosity oil. This indicates that changes in the washing mode have a greater impact on the overall cleaning performance (including both stain removal and removal of redeposited residues) for high-viscosity oil. In addition, the t-values for both low- and high-viscosity oils exhibit a similar pattern. Table 3Statistical analysis of the sensitivity of SO_25 and PO_250 to wash modes.ItemSO_25 (Low-viscosity Oil)PO_250 (High-viscosity Oil)Cohen’s d-0.7110-1.0453t-value-2.8231-3.7984p-value0.00850.000795% CI[-0.0836, -0.0134][-0.1042, -0.0312] Table 3Statistical analysis of the sensitivity of SO_25 and PO_250 to wash modes.ItemSO_25 (Low-viscosity Oil)PO_250 (High-viscosity Oil)Cohen’s d-0.7110-1.0453t-value-2.8231-3.7984p-value0.00850.000795% CI[-0.0836, -0.0134][-0.1042, -0.0312] Statistical analysis of the sensitivity of SO_25 and PO_250 to wash modes. From the perspective of p-values, p < 0.05 for the low-viscosity oil, indicating that the differences in cleaning score between the two washing modes are not due to random variation. Furthermore, the p-value for the high-viscosity oil is even lower, suggesting a higher statistical reliability of the observed differences in cleaning score between the two washing modes. Table 4 presents the differences in cleaning score among three dishwashers for low-viscosity and high-viscosity oils under the same washing mode. For the low-viscosity oil, the Cohen’s d values between DW_A and DW_C, as well as between DW_B and DW_C, indicate large effect sizes, while a moderate effect size is observed between DW_A and DW_B. The t-values exhibit a similar pattern. For the cleaning score results of dishwashers using high-viscosity oil, the Cohen’s d values for DW_A vs. DW_B, DW_A vs. DW_C, and DW_B vs. DW_C correspond to moderate-to-large, large, and small effect sizes, respectively, with the t-values showing a similar trend. These results indicate that, under the same washing mode, the differences in cleaning performance among dishwashers are more pronounced for low-viscosity oil than for high-viscosity oil.。. Table 4Statistical analysis of the sensitivity of SO_25 and PO_250 to washing machines.ItemSO_25 (Low-viscosity Oil)PO_250 (High-viscosity Oil)DW_A vs. DW_BDW_A vs. DW_CDW_B vs. DW_CDW_A vs. DW_BDW_A vs. DW_CDW_B vs. DW_CCohen’s d-0.5717-1.3427-1.5867-0.6725-0.8501-0.3059t-value-1.8078-4.2459-5.0175-2.1268-2.6881-0.9673p-value0.07860.00010.00000.04000.01060.399595% CI[-0.0879, 0.0050][-0.1395, -0.0494][-0.0744, -0.0316][-0.0988, -0.0024][-0.1147, -0.0162][-0.0459, 0.0162] Table 4Statistical analysis of the sensitivity of SO_25 and PO_250 to washing machines.ItemSO_25 (Low-viscosity Oil)PO_250 (High-viscosity Oil)DW_A vs. DW_BDW_A vs. DW_CDW_B vs. DW_CDW_A vs. DW_BDW_A vs. DW_CDW_B vs. DW_CCohen’s d-0.5717-1.3427-1.5867-0.6725-0.8501-0.3059t-value-1.8078-4.2459-5.0175-2.1268-2.6881-0.9673p-value0.07860.00010.00000.04000.01060.399595% CI[-0.0879, 0.0050][-0.1395, -0.0494][-0.0744, -0.0316][-0.0988, -0.0024][-0.1147, -0.0162][-0.0459, 0.0162] Statistical analysis of the sensitivity of SO_25 and PO_250 to washing machines. Microscopic structure of oil on porcelain surfaces As shown in Fig. 5 (B) and (C), after the same pretreatment, discrete particles were observed on the surface of porcelain plates contaminated with SO_25 under SEM, whereas no similar particles were detected on the surface of plates contaminated with PO_250. This difference may be attributed to changes in the physical properties of the two oils induced by the heating process. The low-viscosity oil was not subjected to heating prior to coating onto the porcelain plates; instead, it was dried and heated together with the plates after application. During the drying and heating process, volatile components in the low-viscosity oil may have evaporated, leading to the aggregation of the remaining high-boiling-point components and ultimately the formation of discrete particles. In contrast, the high-viscosity oil had already undergone high-temperature treatment prior to coating, during which significant oxidation and polymerization reactions had occurred21,22. When this high-viscosity oil with low fluidity was applied to the porcelain surface, it was unable to undergo further liquid-phase aggregation. Fig. 5SEM of porcelain chips. (A) Clean porcelain chips; (B) SO_250-contaminated porcelain chip; (C) PO_250-contaminated porcelain chip. SEM of porcelain chips. (A) Clean porcelain chips; (B) SO_250-contaminated porcelain chip; (C) PO_250-contaminated porcelain chip. As shown in Fig. 5 (A) and (C), the surface morphology of the porcelain plates contaminated with PO_250 observed under SEM exhibits a macroscopic protrusion distribution similar to that of the clean plates. However, the overall surface appears smoother than that of the clean porcelain. This is likely because the high-viscosity oil forms a continuous film on the surface, thereby covering the underlying microstructural features of the porcelain. Discussion The above results in Table 3 indicate that changes in the washing mode have a greater impact on the overall cleaning performance for high-viscosity oil than for low-viscosity oil, and that the cleaning score under the standard wash mode is significantly higher than that under the quick wash mode. This may be attributed to the longer duration and stronger hydrodynamic forces acting on the tableware in the standard wash mode, whereas the quick wash mode significantly reduces washing time, water flow intensity, and temperature (Tewes et al., 2024). In addition, low-viscosity oil exhibited greater inter-machine variability in dishwasher cleaning scores compared with high-viscosity oil (Table 4). This may be because the overall cleaning scores for high-viscosity oil were consistently lower across all dishwashers. Under such conditions, the true differences between samples are difficult to distinguish effectively, thereby compressing the variation in cleaning scores among different dishwashers. In contrast, the cleaning scores for low-viscosity oil spanned a wider range (as shown in Fig. 3), making it more sensitive to differences in internal design among dishwashers. Therefore, for standardized dishwasher performance testing, low-viscosity oil is more suitable as a soiling material. Conclusion In this work, the effects of instantaneous heating temperature on the physicochemical properties of edible oils, as well as their impact on dishwasher cleaning performance, were investigated. By measuring the density and rheological properties of peanut oil, rapeseed oil, and soybean oil at room temperature and after heating to 50, 100, 150, 200, and 250 °C, it was found that the rheological properties of all three oils changed only slightly when heated from room temperature to 200 °C, whereas an increase in viscosity was observed at 250 °C. The overall trends in rheological behavior were similar among the three oils. However, there was no significant difference in the density of the three vegetable oils at different temperatures, with a p-value of 0.1917. Subsequently, two oils with the most pronounced viscosity difference (SO_25 and PO_250) were selected to evaluate cleaning performance under quick and standard wash modes across three dishwashers. The results showed that the cleaning performance for high-viscosity oil (PO_250) was more sensitive to washing mode, whereas that for low-viscosity oil (SO_25) was more influenced by dishwasher model. SEM analysis showed that low-viscosity oil soils formed granular residues on the ceramic surface, whereas this phenomenon was not observed on the ceramic surface contaminated with high-viscosity oil soils. This study may provide a preliminary reference for future revisions of dishwasher performance test methods tailored to Chinese dietary habits. Before being formally applied to the test method, future studies should further increase the number of dishwashers tested and continue investigating the storage stability of oils with different viscosities. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2