Wu et al. 2025 — Cadmium in the Soil–Tea–Infusion Continuum of Se-Enriched Gardens (Anhui, China)

This peer-reviewed field study, published in Foods (MDPI, CC BY 4.0), characterized cadmium migration through the soil-to-tea-tree-to-infusion continuum across 12 selenium-enriched tea gardens in southwestern Anhui Province, China, and evaluated whether co-occurring soil Se modulates Cd accumulation in tea leaves. The authors collected 36 topsoil samples and 216 tea-tree organ samples (fibrous roots, taproots, main stems, lateral stems, old leaves, young leaves), prepared tea infusions from young-leaf samples using traditional Chinese green-tea production methods, and quantified Cd and Se in all matrices by ICP-MS (Thermo Fisher iCAP RQ; LOD Cd 0.002 mg/kg, Se 0.005 mg/kg) per Chinese national standard GB 5413.21-2010. Mean soil Cd was 0.37 mg/kg (3.81× background) with 67% of gardens exceeding the Cd safety standard (pollution index Pi > 1.0); mean soil Se was 2.58 mg/kg (above the 0.40 mg/kg Chinese Se-enriched-soil standard). Cd in tea-tree organs followed a descending root-to-leaf gradient — fibrous root 1.17 > taproot 0.29 ≈ main stem 0.27 ≈ lateral stem 0.28 > old leaf 0.07 > young leaf 0.02 mg/kg — with all leaf values well below the 1 mg/kg Chinese tea Cd limit. Cd in tea infusion averaged 0.141 µg/L (mean leaching rate 13.33%), below the 5 µg/L Chinese permissible limit for food/beverage Cd. The authors reported a threshold antagonism: when fibrous-root Se < 2.0 mg/kg, Se and Cd accumulated together (positive correlation); when fibrous-root Se ≥ 2.0 mg/kg, Se inhibited Cd uptake (significant negative correlation). Estimated annual carcinogenic risk of Cd from single-brew tea consumption ranged 2.53 × 10⁻⁸ to 3.84 × 10⁻⁸, and a worst-case 100%-leaching scenario produced 1.60 × 10⁻⁷ to 5.03 × 10⁻⁷ — both within the ICRP-acceptable bound of 5 × 10⁻⁵.

Key numbers

Soil Cd (n = 36 samples across 12 gardens):

• Total Cd: 0.18–0.55 mg/kg; mean 0.37 ± 0.16 mg/kg (3.81× background)
• Available Cd (ACd, DTPA-extractable): 25.19–111.75 µg/kg; mean 70.90 ± 38.17 µg/kg
• Cd activation rate (ARCd = ACd/Cd): mean 20.93%
• Single-factor pollution index Pi (Cd standard = 300 µg/kg): 0.59–1.85; 33.3% of gardens below safety standard (Pi < 1.0), 67% above
• Significant correlations: ACd positively correlated with total Cd (p < 0.01), available K (p < 0.01), total N (p < 0.05); ARCd negatively correlated with soil pH (p < 0.01) and positively with AK and ASe (p < 0.01)

Soil Se (n = 36):

• Total Se: mean 2.58 ± 1.31 mg/kg (vs. 0.40 mg/kg Chinese Se-enriched-soil standard)
• Available Se (ASe, NaHCO₃-extractable, predominantly selenite): 18.53–99.20 µg/kg; mean 53.56 ± 29.90 µg/kg
• Se activation rate (ARSe): mean 2.27%

Cd in tea-tree organs (mg/kg dry weight; n = 36 per organ, 12 gardens × 3 replicates):

• Fibrous root: 0.51–1.97; mean 1.17 ± 0.47
• Taproot: 0.18–0.40; mean 0.29 ± 0.08
• Main stem: mean 0.27 ± 0.08
• Lateral stem: mean 0.28 ± 0.07
• Old leaf: 0.05–0.10; mean 0.07 ± 0.02
• Young leaf: 0.01–0.04; mean 0.02 ± 0.01
• Both old-leaf and young-leaf means are well below the Chinese tea Cd limit of 1 mg/kg
• Body-text note (p. 7) reports young-leaf range as “0.01 to 0.04 µg kg⁻¹”; Table 2 (mg/kg) and surrounding context confirm the correct unit is mg/kg — this is a within-paper unit typo, not a data error

Se in tea-tree organs (mg/kg dry weight):

• Fibrous root: mean 1.71 ± 0.96
• Taproot: mean 0.88 ± 0.58
• Main stem: mean 0.63 ± 0.17
• Lateral stem: mean 0.40 ± 0.22
• Old leaf: mean 0.80 ± 0.58
• Young leaf: mean 0.38 ± 0.41

Cd enrichment coefficients (EC-Cd = organ/soil):

• Fibrous root EC-Cd: 3.49 (range 1.03–6.90), significantly highest (p < 0.05)
• Taproot, main stem, lateral stem EC-Cd: ~0.93, 0.86, 0.88 (statistically indistinguishable)
• Young-leaf EC-Cd: 0.02–0.11; old-leaf EC-Cd: 0.06–0.22
• All EC-Se < 1 (Se absorption capacity generally weaker than Cd in this system)

Cd transport coefficients (TC-Cd, organ-to-next-organ ratio):

• Fibrous root → taproot: 0.29
• Taproot → main stem: 0.97
• Main stem → lateral stem: 1.16
• Lateral stem → old leaf: 0.26
• Lateral stem → young leaf: 0.07
• Two major bottlenecks: fibrous root → taproot (0.29) and lateral stem → young leaf (0.07)

Cd in tea infusion (first brewing; 5 g green tea + 100 mL 100 °C deionized water, 10 min infusion, n = 12 gardens):

• Cd in infusion: 0.116–0.175 µg/L; mean 0.141 ± 0.019 µg/L
• All gardens below Chinese permissible limit for Cd in food/beverage (5 µg/L)
• Cd leaching rate (Cd-infusion / Cd-tea-leaf): 6.98–19.35%; mean 13.33 ± 3.86%

Se in tea infusion:

• Se in infusion: mean 5.18 ± 3.50 µg/L
• Se leaching rate: mean 22.62 ± 8.19% (approximately double Cd leaching rate)

Cd carcinogenic health risk (US EPA model adapted for tea consumption):

• Single-brew assumption (using measured infusion Cd, n = 12): annual carcinogenic risk Rig = 2.53 × 10⁻⁸ to 3.84 × 10⁻⁸ (mean 3.30 × 10⁻⁸)
• 100%-leaching worst-case (all leaf Cd transferred to infusion): annual risk 1.60 × 10⁻⁷ to 5.03 × 10⁻⁷ (mean 2.68 × 10⁻⁷)
• Both values are below the ICRP maximum acceptable risk of 5 × 10⁻⁵
• Risk-model parameters: average daily tea intake 0.0114 kg, Cd carcinogenic slope factor 6.1 mg/kg·d⁻¹, 70-year exposure duration

Se threshold-effect finding (fibrous-root Se vs. fibrous-root Cd):

• When fibrous-root Se < 2.0 mg/kg: positive Cd–Se correlation (R² = 0.20, p < 0.05)
• When fibrous-root Se ≥ 2.0 mg/kg: significant negative correlation (R² = 0.88, p < 0.01)
• Authors interpret as Se → SeO₃²⁻ → Se²⁻ → insoluble Cd–Se complex formation above the threshold, lowering Cd uptake

Linear-regression Se–Cd antagonism (other organs):

• Significant antagonism (p < 0.01): lateral stems (R² = 0.66)
• Significant antagonism (p < 0.05): main stems (R² = 0.19), young leaves (R² = 0.44)
• No significant linear relationship: taproot, old leaf

Random-forest predictors of young-leaf Cd (top tier, p < 0.01): soil organic matter (SOM), Cd in fibrous root, TC-Cd (fibrous-root → taproot), available Cd (ACd), TC-Cd (lateral-stem → young-leaf), total N.

Methods (brief)

Field-survey design across 12 Se-enriched tea gardens in southwestern Anhui Province, China (Golden Tea Belt, ~30 °N). Three 5 m × 8 m quadrats per garden; five tea trees per quadrat; sampling depth 80 cm. Plant material was divided with ceramic scissors into fibrous roots, taproots, main stems, lateral stems, old leaves (fully developed, > 2.5 cm), and young leaves (one bud with two leaves, < 1 cm); rinsed in deionized water; dried to constant weight; ground in a Cd/Se-free stainless-steel grinder; sieved. Topsoil (0–20 cm) samples pooled per garden (12 × 3 = 36 samples). Tea infusions prepared by traditional Chinese green-tea production from young leaves, then 5.000 g of finished green tea + 100 mL boiling deionized water (100 °C) in a 200 mL Erlenmeyer flask, 10 min infusion, filtered. Cd and Se quantified by ICP-MS (Thermo Fisher iCAP RQ, Waltham, MA, USA); LOD Cd 0.002 mg/kg, Se 0.005 mg/kg; quantification per Chinese national standard GB 5413.21-2010 (Determination of Multiple Elements in Foods). Soil available Cd (ACd) and available Se (ASe) measured by DTPA and 0.5 mol/L NaHCO₃ extraction respectively (Shaheen et al. 2018; Yang et al. 2021 methods), 5.00 g soil + 10/25 mL extractant, 180 rpm × 2 h at 20 °C, then ICP-MS. Conventional soil indicators (pH, SOM, TP, AK, TN, AP) per Lu (2000). QA/QC: parallel-sample relative error < 3% (one in three samples randomly duplicated); three blanks per batch; methodology validated against China’s Standard for Geochemical Evaluation of Land Quality (DZ/T 0295-2016). Statistical analysis: SPSS 26.0; mean ± SD with n = 3 per group; Shapiro–Wilk normality; one-way ANOVA + LSD post-hoc; significance at p < 0.05; Pearson correlation in Origin 2021; random-forest predictor analysis via rfPermute in RStudio 4.3.1. Carcinogenic risk model from US EPA (USEPA 2009 Exposure Factors Handbook): Dig = 0.0114 × Ci × µ / 70; Rig = [1 − exp(−Dig × qig)] / 70; qig = 6.1 mg/kg·d⁻¹.

Limitations

• Single-county design. All 12 gardens sit in one Se-enriched tea-growing area in southwestern Anhui Province. Findings — particularly the 2.0 mg/kg fibrous-root Se threshold — may not transfer to Se-enriched tea regions with different parent geology (the authors attribute Se–Cd co-occurrence here to Lower Cambrian carbonaceous siliceous black shale).
• Generic risk-model parameters. The US EPA carcinogenic-risk model used generic body weight, exposure duration, and tea-consumption parameters; the authors explicitly recommend that future region-specific surveys collect southern-Anhui-specific tea-drinking patterns to refine the estimate, especially across age and gender strata.
• Single first-brewing infusion. Infusion data are from one 10-min brew at 100 °C; multi-brew or longer-steep behavior of Cd was not characterized.
• Lab-scale tea processing. Young leaves were processed into green tea within a laboratory setting. The authors note that commercial tea processing using metal machinery may add exogenous Cd, which would raise both leaf Cd and infusion risk relative to the lab-processed values reported here.
• Cd speciation not characterized in plant tissue. Total Cd is reported; subcellular distribution (vacuole vs. cell wall vs. plasma membrane) and Cd–Se complex stoichiometry are inferred from prior literature rather than measured.
• Geographic anonymization. The study area is described as “a certain County” in southwestern Anhui; the reference list ([40] Long et al. 2018) suggests Shitai County but the body text does not name it.

Verification notes

Source page newly created from raw/Manual Fetch Discovery/wu2025-matcha-heavy-metal.pdf (Manual Fetch Discovery folder). The PDF filename includes “matcha” but the paper is not about matcha — it characterizes green tea from cultivars ‘Chuye’ and ‘Liuye’ processed by “traditional green tea production methods.” The cite-key was set to wu2025-se-enriched-tea-cadmium-anhui to reflect the actual subject; the misleading raw-handle is preserved for filesystem traceability. Three identity checks ran clean: DOI grep (grep -rl "10.3390/foods14183156" wiki/sources/) returned no hits; raw_handle grep returned no hits; ls wiki/sources/ | grep wu2025 returned only wu2025-climate-meghan-freshwater-fish-china.md (different paper). No commercial brand attributions in the source; ICP-MS vendor (Thermo Fisher iCAP RQ) and statistical-software names (SPSS, Origin, RStudio rfPermute) are method-vendor mentions allowed under Part 12 Exception 2. Se data are reported here as context for the Cd–Se interaction finding; HMI tracks Cd, not Se, so Se appears only where it modulates Cd behavior.

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.