Donkey Milk Study: Flavor Compounds in Five Milk Powders (Zhou 2025)

Five laboratory-prepared milk powders were compared using two methods: 55 or 86 volatile compounds were detected. Donkey milk powder contained 58 of the 86 compounds, 25 of which were found only in donkey milk powder, including several ethyl esters.

The study in brief

• Original title: Differentiation and characterization of volatile compounds in five common milk powders using HS-GC-IMS, HS-SPME-GC–MS, and multivariate statistical approaches
• Authors: Zhou Y., Wang D., Zhao J. et al.
• Journal: Food Chemistry: X, 25, 102179 (2025)
• Country: China · Study type: Laboratory study

How it was investigated

One batch of raw milk from yak, donkey, camel, goat, and cow was obtained directly from producers in different regions of China in May 2024; the donkey milk came from a breeding cooperative near Jinan (Shandong Province). All samples were pasteurized identically (72–80 °C, 15 s), freeze-dried in the laboratory, and stored at −80 °C. The volatile compounds were determined using headspace gas chromatography-ion mobility spectrometry (HS-GC-IMS) and headspace solid-phase microextraction plus gas chromatography-mass spectrometry (HS-SPME-GC–MS, internal standard ethyl decanoate), each in triplicate.

Important figures

• 5 – Milk powders: yak, donkey, camel, goat, and cow (one batch each, freeze-dried)
• 55 – Signals per HS-GC-IMS, of which 53 were identified (incl. monomer/dimer signals)
• 86 – Volatile compounds per HS-SPME-GC–MS
• 58 – Of which detected in donkey milk powder (yak: 34, camel: 39, goat: 37, cow: 38; editorial count from Table 2)
• 25 – Compounds only in donkey milk powder according to Table 2 (text and Venn diagram: 24)
• 15 – Compounds in all five powders (HS-SPME-GC–MS)

Results

With HS-GC-IMS, 55 signals were detected, of which 53 were identified; several substances appeared as monomer and dimer signals. According to the fingerprint representation, donkey milk powder contained higher levels of, among others, 1-butanol, 2,6-dimethylpyrazine, 4-methyl-3-penten-2-one, butyl isobutyrate, 1-propanol, and hexanal than the other powders.

By peak volume, donkey and yak milk powders contained the highest proportions of ketones and esters. In the PCA of the GC-IMS data (PC1 47%, PC2 25%), donkey and yak separated clearly from camel, goat, and cow, which clustered closely together.

With HS-SPME-GC–MS, 86 compounds were found, 15 of which were in all powders. According to Table 2, 58 compounds were detectable in donkey milk powder, thus more than in the other powders (34–39).

Limitations of the study

• One batch per animal species: Only one batch of milk from one producer was examined for each animal species; the triplicate measurements are technical replicates. Differences may therefore also be due to origin, feeding, lactation stage, or sampling, and the model quality of the OPLS-DA is based on very few independent samples.
• Counting method in text vs. table: According to the text, 17 substances appeared as monomers and dimers (18 are listed), Table 1 contains 19 such pairs. The PCA is supposed to be based on "85" compounds identified by GC-IMS, although only 55 signals were detected.
• Only in one powder: differing numbers: The text and Venn diagram list 24, 7, 4, 1, and 1 compounds that occur only in donkey, yak, camel, goat, or cow milk powder, respectively. According to Table 2, these are 25, 7, 5, 1, and 2. The information on the distribution of individual compounds among the powders also differs between the text and the table.
• Units and assignments: No unit is given for the GC–MS values in Table 2, although an internal standard was used. Several database assignments for milk aroma are not plausible and are not categorized in the article.
• Laboratory powders, not commercial products: All samples were pasteurized and freeze-dried identically. The results therefore describe this laboratory process and cannot be directly transferred to commercially available powders with different processing or to fresh milk. No sensory evaluation was conducted.

Level of evidence: Basic research (laboratory) – describes the properties of milk, no statement about effects in humans.

Our assessment

The study provides an analytical aroma profile of laboratory freeze-dried donkey milk powder compared to four other types of milk. Since only one batch was examined for each animal species, the values are suitable as an initial orientation, not as general figures for donkey milk powder or individual commercial products.

Source: Zhou Y., Wang D., Zhao J., Guo Y., Yan W. (2025). Differentiation and characterization of volatile compounds in five common milk powders using HS-GC-IMS, HS-SPME-GC–MS, and multivariate statistical approaches. Food Chemistry: X. DOI: 10.1016/j.fochx.2025.102179 · Summarized by the DON DONKEY'S editorial team, as of October 2026.

End of the study summary. The study was conducted independently by DON DONKEY'S® and does not refer to our products. Our donkey milk is a food.