Quality control of ice cream mix with near-infrared spectroscopy

Significance of the topic

Ice cream is a complex emulsion whose quality attributes—such as texture, flavor, stability, and caloric content—depend on precise control of fat, protein, sugars, and solids. Traditional laboratory analyses for these components are laborious, time-consuming, and require chemical reagents. Implementing a rapid, non-destructive, reagent-free technique enhances production efficiency and ensures batch-to-batch consistency vital for consumer satisfaction and regulatory compliance.

Objectives and study overview

This study evaluates the feasibility of near-infrared spectroscopy (NIRS) to simultaneously determine key compositional parameters in ice cream mix. The primary goals were:

• To develop calibration models for fat, total solids, protein, lactose, sucrose, and calories;
• To validate model performance against official reference methods;
• To assess the accuracy, precision, and speed of the NIR approach for routine quality control.

Methodology and instrumentation

The experimental setup and procedures included:

• Instrumentation: Metrohm NIR Analyzer operating in reflection mode over 1000–2250 nm using a small-cup accessory and single-point measurement;
• Sample set: Forty diverse ice cream mix formulations covering a broad compositional range;
• Reference methods: AOAC protocols for total solids (941.08), protein (930.33), and fat (932.06); high-performance liquid chromatography (HPLC) for lactose and sucrose; bomb calorimetry for caloric determination;
• Data handling: Metrohm software for spectral acquisition, preprocessing, and multivariate model development with leave-one-out cross-validation.

Key results and discussion

All six analytes exhibited strong correlation between NIRS predictions and reference values:

• Fat: R2 = 0.979; SEC = 0.24 %; SECV = 0.30 %
• Total solids: R2 = 0.979; SEC = 0.52 %; SECV = 0.58 %
• Lactose: R2 = 0.921; SEC = 0.06 %; SECV = 0.10 %
• Sucrose: R2 = 0.952; SEC = 0.33 %; SECV = 0.37 %
• Protein: R2 = 0.974; SEC = 0.11 %; SECV = 0.14 %
• Calories: R2 = 0.981; SEC = 2.83 kcal; SECV = 2.89 kcal

The high coefficients of determination and low standard errors confirm that NIRS can accurately predict compositional attributes within seconds, bypassing sample preparation and chemical reagents.

Benefits and practical applications

The NIR method offers multiple advantages over conventional analyses:

• Rapid, simultaneous quantification of multiple components;
• Non-destructive measurement without chemical reagents;
• Minimal sample preparation and reduced analysis time;
• Potential for integration into on-line or at-line production monitoring;
• Cost savings through reduced labor and consumables.

Future trends and opportunities

Advancements and potential developments include:

• Expansion of calibration libraries to include more product variants (e.g., low-fat, plant-based formulations);
• Integration with process control systems for real-time feedback and automated adjustments;
• Machine learning algorithms to improve robustness against matrix variability and temperature effects;
• Miniaturized handheld NIR devices for field or retail quality screening.

Conclusion

This application note demonstrates that near-infrared spectroscopy is a powerful tool for fast, accurate, and reagent-free quality control of ice cream mix. The developed models deliver reliable predictions for fat, solids, protein, lactose, sucrose, and caloric content, supporting efficient production workflows and consistent product quality.

References

• AOAC 941.08: Determination of Total Solids in Ice Cream
• AOAC 930.33: Official Method for Protein in Dairy Products
• AOAC 932.06: Official Fat Determination Method
• Metrohm Application Note AN-NIR-139: Quality Control of Ice Cream Mix