Zhou et al. 2025 — Exogenous impurities in baijiu: sources, detection, and safety strategies (review)

Zhou, Hua, and Xu (Sichuan University of Science and Engineering; Flinders University; Sichuan Vocational College of Chemical Technology; Luzhou Key Laboratory) review exogenous impurities in baijiu, the Chinese solid-state-fermented distilled grain spirit. The review covers five impurity classes (pesticide residues, mycotoxins, heavy metals, plasticizers, and un-authorized sweeteners), their introduction pathways (raw materials, water, equipment, packaging, environment), Chinese national-standard regulatory frameworks (GB 2762-2022 for contaminants, GB 2763-2021 for pesticide MRLs, GB 2761-2017 for mycotoxins, GB 2760-2014/2024 prohibiting sweetener additions to baijiu, GB 5009-series analytical standards), analytical detection techniques (GC-MS, LC-MS/MS, ICP-MS, GFAAS, HPLC, ELISA), and control strategies. The review contributes no new measurements; heavy-metal concentration ranges in commercial baijiu are reproduced from a single underlying primary study (Newman et al. 2017, 47 samples from rural central China) and contextualised against Chinese regulatory caps.

Key numbers

This is a narrative review with no new measurements. All values reported below are reproduced from cited primary studies or named Chinese national standards; the review itself does not present pooled or independently verified data.

Heavy-metal concentrations in commercial baijiu (review section 2.2.1, Table 3, citing Newman et al. 2017 [ref 34])

The review reproduces heavy-metal concentration ranges measured in 47 samples of locally distilled baijiu obtained from rural central China:

The review’s own commentary (section 2.2.1, p. 6): “Cd concentration in Baijiu ranges from 0.75 to 19.55 µg/L, while Pb ranges from 0.35 to 580 µg/L, indicating greater variability in lead contamination across products.”

A separate cited study (Li Yongjiao et al., ref [29]) is summarised as finding that “less than 6% of the metal content was transferred into the distillate, while more than 94% remained in the fermented grains after distillation” during strong-aroma baijiu distillation; no numerical concentrations are reproduced for this study in the Zhou review.

Chinese regulatory caps for baijiu (review section 2.2.1, citing GB 2762-2022 [ref 35])

The review states that GB 2762-2022 sets a maximum residue limit for Pb in baijiu of 0.5 mg/kg and that “currently, there are no specific regulatory limits established for other heavy metals in Baijiu” (section 2.2.1, p. 6, and section 3.1, p. 11).

Dilution-water requirements (review section 2.1.3, citing Xiao 2018 [ref 25])

The review states that water used for diluting high-alcohol baijiu base liquor to lower-alcohol final products must meet, in addition to a total heavy-metals cap of <0.1 mg/L: As <0.01 mg/L, Cd <0.005 mg/L, Hg <0.001 mg/L, Se <0.01 mg/L, Pb <0.01 mg/L. The source citation is the Chinese textbook “Baijiu Production Technology, 3rd ed.” (Xiao, D.-G., 2018, Chemical Industry Press, Beijing), not a peer-reviewed primary measurement.

Pesticide residues in strong-flavor baijiu (review Table 3, citing Chen et al. 2024 [ref 10])

The review reproduces pesticide-residue concentration ranges from 10 sample groups of strong-flavor baijiu:

These pesticide values are outside HMI primary scope but are reproduced for completeness because the review presents them in the same Table 3 as the heavy-metals values.

Plasticizer (phthalate ester) concentrations (review section 2.2.2 and Table 3, citing Dong et al. 2019 [ref 40])

The review reports that 39 baijiu samples (from 20 Baijiu brands collected from local markets in China) tested positive in 100% of samples for five phthalates (DBP, DEHP, DMP, DEP, DIBP), with combined detected concentrations in the range 0.003–0.292 mg/kg. The per-analyte ranges from Table 3 are: DMP 0.05–0.12 mg/kg; DEP 0.04–0.05 mg/kg; DIBP 0.08–0.15 mg/kg; DBP 0.10–0.18 mg/kg; DEHP 0.13–0.24 mg/kg. China’s regulatory caps for baijiu (Document No. [2019] 214) are DBP ≤1 mg/kg and DEHP ≤5 mg/kg. Plasticizers are outside HMI primary scope.

Mycotoxin concentrations (review Table 3, citing Wu et al. 2024 [ref 18])

The review reports concentrations of fumonisins (FB1, FB2, FB3), zearalenone (ZEN), and deoxynivalenol (DON) in 43 samples of commercial baijiu from the Chinese market: FB1 0.25–0.31 µg/L, FB2 0.27–0.30 µg/L, FB3 0.41–0.48 µg/L, ZEN 3.69–5.11 µg/L, DON 1.81–12.75 µg/L. Mycotoxins are outside HMI primary scope.

Unauthorized sweeteners (review section 2.3 and Table 3, citing Ma et al. 2020 [ref 54] and surveillance reports)

The review reports a national sampling and testing distribution among un-authorized sweetener detections in baijiu: sodium cyclamate 47%, neotame 26%, saccharin sodium 21%, sucralose 6%. A separate 30-sample survey of 12 baijiu brands collected from Chinese supermarkets reports per-analyte ranges of sucralose 0–8.45 mg/kg, neotame 0–1.07 mg/kg, and saccharin 0–3.22 mg/kg. Sweeteners are outside HMI primary scope.

Activated-carbon Pb removal (review section 2.2.1 control strategies, citing Wang et al. 2015 [ref 37])

The review states that nitric-acid-modified activated carbon achieves a 96.7% Pb removal rate from baijiu under ambient conditions with an 8-hour adsorption time. No starting or post-treatment Pb concentration is reproduced. Cited primary source: Wang, X.; Zeng, X.; Liu, B.; Zeng, L. Absorption of Pb in Baijiu (Liquor) by Active Carbon. Liquor-Mak. Sci. Technol. 2015, 6, 49–50 (in Chinese).

Evidence Fitness

This source is a narrative review of exogenous impurities in baijiu. Heavy-metal content enters only as cited concentration ranges from a single underlying primary study (Newman et al. 2017 [ref 34], 47 samples from rural central China). The review contributes no primary occurrence data and is not eligible as a pooling-engine input for any percentile work. It functions as: (1) a pointer to the Newman et al. 2017 underlying paper, which should be ingested directly for any defensible baijiu heavy-metal occurrence row in the future; (2) a structured summary of Chinese national-standard regulations applicable to baijiu impurities (GB 2762-2022 Pb cap 0.5 mg/kg, GB 2763-2021 pesticide MRLs in raw grains, GB 2761-2017 mycotoxin limits, GB 2760-2014/2024 sweetener prohibition, GB 5009.268-2025 analytical methodology); (3) a documentation of the contamination pathway from ceramic storage vessels (low pH and lower-proof alcohol increase Pb/Cd leaching from glazes), which is the same pathway already characterised in lead-glazed-ceramic kombucha leaching; and (4) a confirmation that the heavy-metal mass-balance during baijiu distillation strongly favors retention in the spent grains (>94%) over transfer to the distillate (<6%) per the cited Li et al. 2016 study. Reported public evidence label: Context only.

Methods (brief)

Narrative review (MDPI Beverages, Volume 12, Issue 1, article 2). No systematic search strategy, database list, inclusion/exclusion criteria, deduplication procedure, or PRISMA flow diagram is stated. The Data Availability Statement (p. 13) reads “No datasets were generated or analyzed during the current study.” The review compiles values from 61 cited references covering pesticide residues, mycotoxins, heavy metals, plasticizers, and food additives in baijiu and related fermented or distilled products. No analytical instrument or laboratory methodology is reported because the review presents no original measurements.

For heavy metals specifically, the review identifies the recognized analytical platforms as graphite furnace atomic absorption spectrometry (GFAAS) and inductively coupled plasma mass spectrometry (ICP-MS), specifying the latter as the standard method under GB 5009.268-2025 for Pb, Cd, Hg, and As in food (including alcoholic beverages such as baijiu). No instrument vendor, model, sample-preparation protocol, digestion method, calibration standard, or detection limit is reported in the review. The cited heavy-metal concentration ranges (Newman et al. 2017 [ref 34]) are reproduced as the underlying paper published them; the review does not state the underlying paper’s sample preparation, detection limits, sample size by analyte, or sampling locations beyond “locally distilled baijiu obtained from rural central China.”

Implications

• Certification: contributes no direct occurrence data eligible for a baijiu HMTc row (no baijiu product slug currently exists in the wiki taxonomy; see Verification notes). The cited Pb upper bound (580 µg/L = 0.58 mg/L) sits above the Chinese national cap of 0.5 mg/kg, indicating that real-world Pb contamination in rural-distilled baijiu can exceed the regulatory threshold; the underlying Newman et al. 2017 paper should be ingested directly for any defensible synthesis. The 96.7% Pb-removal capacity of nitric-acid-modified activated carbon (cited Wang et al. 2015) is a remediation pathway, not an occurrence finding.
• Courses: useful as a teaching reference for distilled-spirit contamination pathways, mass-balance during distillation (<6% of metal content transfers to distillate per cited Li 2016), ceramic-vessel leaching as a secondary contamination source (acidic ethanol-water matrix elutes Pb, Cd from glaze, with higher-proof spirits showing reduced leaching due to ethanol-glaze interactions forming low-solubility salts), and the role of solid-state fermentation and microbial enzymatic degradation in reducing pesticide residues (39.8–74.2% reduction through fermentation per cited Bai et al. Food Chem. 2021 [ref 9]; >90% during distillation per cited Bai PhD Thesis Hunan University 2021 [ref 17]).
• App: contributes no per-product concentration values; downstream consumers should source values from the original Newman et al. 2017 paper if a baijiu product is added to HMI scope.
• Discovery: the references most directly relevant to HMI primary scope are ref [34] Newman et al. 2017 (Alcohol. Clin. Exp. Res. 41, 207–215 — “Composition of Unrecorded Distilled Alcohol (bai jiu) Produced in Small Rural Factories in Central China”), ref [29] Li, Y.; Zhang, S.; Huo, D.; Zeng, X.; Tang, Y.; Wang, M. “Transfer of Metal Elements in the Distillation Process of Chinese Luzhou-Flavor Liquor” (Food Sci. 2016, 37, 156–161), ref [31] Turner, A. 2019 “Heavy Metals in the Glass and Enamels of Consumer Container Bottles” (Environ. Sci. Technol. 53, 8398–8404), and ref [37] Wang et al. 2015 on activated-carbon Pb removal from baijiu.

Provenance notes

Open-access article distributed under CC BY (license declaration on p. 1 of the PDF). Received 3 November 2025; revised 13 December 2025; accepted 18 December 2025; published 24 December 2025. Citation: Zhou, Y.; Hua, J.; Xu, L. Exogenous Impurities in Baijiu: Sources, Detection, and Safety Strategies. Beverages 2026, 12, 2. https://doi.org/10.3390/beverages12010002. Academic Editor: Dirk W. Lachenmeier. Funding statement: “Brewing Science and Technology Key Laboratory of Sichuan Province, grant number NJ2023-06, the Key Laboratory of Natural Products and Functional Food Development of Luzhou, grant number 2025-GNSP-1, and Wuliangye Industry University research cooperation project, grant number CXY2022ZR009.” Conflict-of-interest statement: “The authors declare no conflicts of interest.” Accessed via the Manual Fetch Discovery autopilot.

Evidence tier set to C. The source is a narrative review without systematic methodology, presents no new measurements, and reproduces heavy-metal concentration ranges from a single underlying primary study (Newman et al. 2017 [ref 34]). The review’s scope is the full impurity profile of baijiu (pesticides, mycotoxins, metals, plasticizers, sweeteners) with regulatory and analytical framing; heavy-metals occurrence is one section of five and is not independently verified.

The funding source includes a research cooperation project with Wuliangye Industry University (a research arm associated with one of China’s largest baijiu producers, named in the review’s body text in a non-contamination industry-initiative context). This is disclosed by the authors and noted here for transparency; the review’s regulatory and analytical content does not appear to be shaped by the cooperation, but readers should be aware of the affiliation when weighing the review’s recommendations on industry control strategies.

Wiki pages this source may touch

• lead
• cadmium
• arsenic
• mercury
• copper
• nickel
• aluminum
• chromium
• iron
• china-gb-2762-2022-contaminants

Verification notes

No baijiu product slug currently exists in the wiki product taxonomy. The products: frontmatter is left empty rather than routing the source to an inappropriate sibling (fermented-beverages-non-tea-based is the closest existing slug but is dominated by kombucha and other non-distilled fermented beverages, while baijiu is a distilled spirit produced via solid-state fermentation followed by distillation). Creating a baijiu product page requires Karen’s Step 0 Lock per CLAUDE.md Part 10 and is not in this ingest’s scope. The source still routes to multiple metals pages and to the China GB 2762-2022 contaminants regulation page via the metals: and jurisdictions: fields. A future Step 0 Lock for a baijiu product (or a broader “distilled spirits” product class) would benefit from also ingesting the underlying Newman et al. 2017 primary paper that supplies all the heavy-metal occurrence values reproduced here.

The metals: frontmatter lists Pb, Cd, As, Hg, Cu, Ni, Al, Cr, Fe. Pb and Cd appear with cited numerical concentration ranges in baijiu (Newman 2017). As, Hg, Cu are named in the review’s heavy-metals-of-particular-concern list (section 2.2.1, p. 6) without numerical concentrations in baijiu but with dilution-water regulatory caps. Ni, Al, Cr, Fe are named as distillation-equipment and storage-vessel contamination sources (section 2.2.1, p. 6) without numerical concentrations in baijiu. Per CLAUDE.md Part 14 speciation discipline, As is recorded without the iAs/tAs distinction because the review does not specify speciation (it uses “arsenic (As)” generically); Hg is recorded without the MeHg/tHg distinction for the same reason; Cr is recorded as elemental Cr without the Cr-VI specifier. The review’s analyte naming is at the parent-element level throughout.

The ingredients: frontmatter is empty. The review’s raw-material discussion names sorghum, rice, wheat, corn, and barley as baijiu grain inputs and reproduces pesticide and mycotoxin MRLs for these grains, but the review’s heavy-metals data concern the finished distilled baijiu, not raw-grain heavy-metal occurrence. The cited Li 2016 study on metal transfer during distillation reports a directional mass balance (<6% to distillate, >94% retained in spent grains) without per-grain numerical values reproduced in the Zhou review. No ingredient-level heavy-metal occurrence data are extractable; the ingredients: field is therefore not populated.

The matrices: list includes distilled-spirit (the baijiu finished beverage), fermented-beverage (the solid-state fermentation upstream of distillation), and food-contact-material-leachate (the ceramic vessel and equipment leaching pathway the review identifies as the dominant heavy-metals contamination source for finished baijiu, per section 2.2.1 citing refs [29], [31], [32], [33]). fermented-beverage and food-contact-material-leachate are used elsewhere in the corpus (e.g., the kombucha review and ceramic-vessel sources). distilled-spirit is a novel matrix string in this ingest and should be reconciled against the matrices controlled vocabulary on next lint pass; if not present, treat as a new-vocab flag for Karen’s matrix-vocabulary review (parallel to the products-slug flag above).

The jurisdictions: list contains CN (the review’s regulatory framework is entirely Chinese GB national standards; the review is authored by researchers at Chinese and Australian institutions writing about Chinese baijiu) and EU (the review briefly notes EU MRL alignment under Regulation EC No 396/2005 as an export-compliance consideration). No US regulatory content appears.

Brand handling per CLAUDE.md Part 12. Three categories of brand mention appear in the review and have been handled as follows in this wiki page:

• Regulatory-event exception (Part 12, Exception 1): The review names Gansu Pukang Liquor Group (2025 enforcement action for unauthorized sodium cyclamate) and Zuiyuanchun Liquor Group (2024 enforcement action for unauthorized sodium cyclamate) in section 2.3.1 as subjects of public-record regulatory enforcement actions. These are regulatory events, not contamination rankings, and are within Part 12’s exception. They have NOT been reproduced in the wiki-page body because sweeteners are outside HMI primary scope and the regulatory-event framing pertains to an analyte class HMI does not track; preserving the brand mentions on the wiki page would add brand-identifying content without serving HMI’s primary scope. The exception permits but does not require inclusion; non-inclusion is the conservative choice here.
• Industry-initiative mention (NOT a regulatory event): The review names Maotai Group, Wuliangye Group, and Kongfuyan Group in section 2.1.1 (grain cultivation initiatives) and section 4 (pure baijiu pursuit) in a non-contamination industry context. These are NOT subjects of regulatory enforcement actions and the mentions do not fall under Part 12 Exception 1. The wiki-page body refers to these as “leading baijiu producers” and “major brewers” rather than naming them. The Wuliangye Industry University research-cooperation funding link is disclosed in Provenance notes for transparency about the review’s funding source per standard academic disclosure norms, not as a brand ranking.
• Scientific-method vendor mention (Part 12, Exception 2): The review does not name specific instrument vendors, models, reference materials, or statistical software for analytical methods. Generic method names (GFAAS, ICP-MS, LC-MS/MS, HPLC, GC-MS, ELISA, IC, UHPLC-MS/MS, QuEChERS) are reproduced as needed because they are method classes, not vendor names. No Part 12 Exception 2 deletion was required.

No HMTc threshold proposals, no consumer-audience translations, no risk advisories, and no synthesis claims of the form “this confirms the literature consensus that…” appear in the body, per CLAUDE.md Part 2 wiki/HMTc firewall. The Pb 580 µg/L upper bound exceeds the Chinese national cap of 0.5 mg/kg (500 µg/L) by a small margin in the highest rural-distilled samples; this observation is reported as a unit comparison, not as an HMTc threshold recommendation. The Newman 2017 underlying paper measures rural-factory (non-recorded, non-commercially-traded) baijiu specifically, which is a distinct population from commercially-traded brand-name baijiu, and the upper-tail value should be read in that population context rather than generalised to the full commercial baijiu market.

The unit harmonisation across the Pb and Cd values: the review reports both in µg/L (parts-per-billion equivalent for an aqueous matrix), while the Chinese regulatory cap is in mg/kg. For a distilled-spirit matrix near 1 g/mL density, µg/L and µg/kg are equivalent to within ~1%; both are equivalent to ppb. The 0.5 mg/kg regulatory cap corresponds to ~500 µg/L. No unit conversion was applied; both bases are reported as the underlying sources gave them, with the harmonised µg/L-vs-mg/kg comparison flagged inline.

Audit subagent (2026-06-06) ran the five-check audit against this page and the underlying PDF and returned a PROMOTE verdict with three minor advisory flags. Two were verified and applied: (1) the editorial “(DDT)” gloss next to Chofenotane in the pesticide-table row was removed because the PDF does not state DDT-equivalence — Chofenotane appears unannotated in PDF Table 3 (p. 10); (2) the Bai-2021 attribution was split so the 39.8–74.2% fermentation reduction is attributed to Bai et al. Food Chem. 2021 [ref 9] (PDF p. 3) and the >90% distillation degradation is attributed to Bai PhD Thesis Hunan University 2021 [ref 17] (PDF p. 4) — the prior conflation was a citation-mapping error and is now corrected. The third advisory (matrices-vocabulary verification for distilled-spirit) was applied as an in-page flag for Karen’s matrix-vocabulary review rather than a content correction. The audit also confirmed numerical fidelity to PDF Table 3 across heavy metals, phthalates, mycotoxins, pesticides, and sweetener distributions; confirmed Part 12 brand-firewall handling on Maotai/Wuliangye/Kongfuyan (generalised), Gansu Pukang/Zuiyuanchun (Exception-1 subjects, non-inclusion), and Wuliangye funding disclosure (Provenance placement); and confirmed Part 2 wiki/HMTc firewall — no threshold recommendations, no consumer advisories, no synthesis-consensus framing.

Page history

The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.