Yamada et al. 1993 — Small-scale survey of organotin compounds in Japanese household commodities

This March/April 1993 Journal of AOAC International technical communication by Sadaji Yamada and colleagues at the Aichi Prefectural Institute of Public Health and the Aichi Prefectural Government Department of Health (Nagoya, Japan) reports an organotin-compound (OTC) survey of 95 household commodity items purchased in Aichi Prefecture between 1990 and 1992. The survey covered the 13 household-product classes designated under Japan’s 1973 Law for the Control of Household Products Containing Harmful Substances (textile products including diapers, diaper covers, bibs, sanitary panties, underwear, socks, gloves, outergarments, and stockings; plus shoe polishes, adhesives, paints, and waxes) and assayed for six organotin species by flame photometric detection/gas chromatography (FPD/GC) with GC/MS confirmation. Tributyltin (TBT) and triphenyltin (TPT) — which Japan had banned for household-product use in 1979–1980 under the Law cited — were not detected in any of the 95 items. Mono- and/or di-substituted octyltin (DOT, TOT) and dibutyltin (DBT) compounds, which were not banned, were detected in 15 items: 6 of 10 diaper covers, 3 of 10 bibs, 2 of 10 sanitary panties, 1 of 5 outergarments, 1 of 5 stockings, 1 of 5 socks, and 1 of 10 shoe polishes. Diapers themselves, underwear, gloves, adhesives, paint, and wax were OTC-negative across all items tested. The single highest organotin loading observed was DOT at 1666.7 µg/g (≈ 0.17% by weight) in a polyester diaper cover; a second polyester diaper cover reached DOT at 902.8 µg/g with co-occurring TOT at 58.3 µg/g and DBT at 33.7 µg/g. The authors attribute octyltin and dibutyltin occurrence to intentional addition as silicone-coating catalysts (water-resistivity treatment for diaper covers, outergarments, and shoe polish) and to carry-over from poly(vinyl chloride) and polyurethane stabiliser additives.

Key numbers

Sampling frame (Methods, p. 437; Results Table 2, p. 440)

• Total items surveyed: 95 commercially available household commodities purchased 1990–1992 at retail stores in Aichi Prefecture, Japan; no more than two items per manufacturer per category.
• Category-by-category counts (number assayed / number with OTC detected):
  • Diaper 10/0
  • Diaper cover 10/6
  • Bib 10/3
  • Sanitary panty 10/2
  • Underwear 10/0
  • Socks 5/1
  • Gloves 5/0
  • Outergarment 5/1
  • Stockings 5/1
  • Shoe polish 10/1
  • Adhesive 5/0
  • Paint 5/0
  • Wax 5/0
• OTC detection frequencies by category: 6/10 (60%) diaper covers; 3/10 bibs; 2/10 sanitary panties; 1/5 outergarments; 1/5 stockings; 1/5 socks; 1/10 shoe polishes.
• Items with OTC detected: 15 of 95 (15.8%).

Per-species detection counts (Table 2, p. 440)

Reported as number of items in which each species was found, expressed as the propylated/derivatised form quantified by FPD/GC:

(Source: Table 2 columns DBT, TBT, DOT, TOT, TPT. The DPT column is absent from Table 2 because diphenyltin was dropped from the determinative procedure — see Methods.)

Per-item organotin concentrations (Table 3, p. 440; µg/g as chloride form)

Each value is the mean of duplicate determinations and is expressed as the dichloride for di-substituted species (DBTD, DOTD) and as the chloride for tri-substituted species (TOTC).

Range summary across detection-positive items (Results and Discussion, p. 440)

• DOT (dioctyltin dichloride basis): detected in 14 of the 15 OTC-positive items (all except Sanitary panty B), with concentrations ranging from 24.4 µg/g (Diaper cover F) to 1666.7 µg/g (Diaper cover B).
• DBT (dibutyltin dichloride basis): detected in 4 items, ranging from 3.7 µg/g (Diaper cover F) to 33.7 µg/g (Diaper cover A).
• TOT (trioctyltin chloride basis): detected in 5 samples; the authors attribute its co-occurrence with DOT to TOT being a known impurity in commercially available DOT preparations rather than an independently added compound.
• TBT and TPT: not detected in any of the 95 items (consistent with the Japanese ban for household-product use in 1979–1980).
• Diaper cover A contained DOT at 902.8 µg/g (≈ 0.09% by weight) — the paper describes this as “near 1000 ppm DOT” — and diaper cover B reached DOT at 1666.7 µg/g (≈ 0.17% by weight), characterised by the paper as “a similar concentration” though it is in fact ≈ 1.85× the diaper-cover-A value. The authors describe DOT in diaper cover A, bib A, outergarments, and shoe polish as catalysts used with silicones to increase water resistivity of the textile or polish substrate, given the considerable amounts and the documented use of these products. On other samples (lower-loading bibs, stockings, socks, sanitary panties), the authors propose carry-over from poly(vinyl chloride) and polyurethane stabiliser/catalyst use during fabric or component manufacture.

Detection limits (Methods, p. 438)

• Detection limits were 0.2 ppm (≈ 0.2 µg/g) for DBTC, TBTC, and DOTD and 0.5 ppm for TOTC and TPTC.
• DPT was excluded from the determinative panel because recovery from textile products via the Florisil cleanup was poor and could not be improved within the scope of this study; the authors cite [Ishizaka et al. 1989] as having reached the same conclusion for fish matrices.

Recovery (Table 1, p. 438; mean ± SD across triplicate analyses)

Spike-recovery at 3.0 ppm in four item types:

DPT recovery from the two textile matrices (38.8% and 35.3%) is the basis for excluding DPT from the determinative panel.

Chromatographic retention times (Figure 1, p. 437)

Standard mixture of propylated derivatives gave the following FPD/GC retention times:

• Dipropyl-DBT 5.12 min
• Propyl-TBT 6.15 min
• Dipropyl-DOT 14.11 min
• Dipropyl-DPT 14.83 min
• Propyl-TOT 18.23 min
• Propyl-TPT 19.83 min

GC/MS confirmation (Figures 3–5, p. 439)

Mass-spectral confirmation for the three positively detected species used the following diagnostic peaks (base and fragment ions):

• Dipropyl-DBT: base peak m/z 221 [Sn(C₄H₉)(C₃H₇)]⁺; fragment peaks at m/z 121 [Sn]⁺, 165 [Sn(C₃H₇)₂]⁺, 179 [Sn(C₄H₉)₂]⁺, 207 [Sn(C₃H₇)₂]⁺, 235 [Sn(C₄H₉)₂]⁺, 263 [Sn(C₃H₇)₂(C₄H₉)]⁺, and 277 [Sn(C₃H₇)(C₄H₉)₂]⁺. Mass spectra of peak 1 in the diaper-cover A chromatogram matched the dipropyl-DBT standard.
• Dipropyl-DOT: mass spectra of peak 2 matched the dipropyl-DOT standard.
• Propyl-TOT: mass spectra of peak 3 matched the propyl-TOT standard.

Methods (brief)

Apparatus. FPD/GC: Shimadzu GC-8A gas chromatograph (Shimadzu Co., Ltd, Kyoto, Japan) with a flame photometric detector operated in the tin mode (610 nm filter) and a Shimadzu C-R1B integrator, using a megabore DB-17 column (15 m × 0.53 mm id, J&W Scientific, Inc., Folsom, CA 95630). Oven programmed 80→250 °C at 8 °C/min; injection port and detector 270 °C. Gas flow: nitrogen carrier 20 mL/min; hydrogen 0.5 kg/cm²; air 0.7 kg/cm². GC/MS confirmation: Hewlett Packard GC 5890 series II coupled to a JEOL JMS-AX505W mass spectrometer (JEOL, Tokyo, Japan), electron-impact mode, ionization voltage 70 eV, ionization current 300 µA, helium carrier 20 mL/min, separator and ion source 250 °C; same column programme as FPD/GC.

Reagents. Tri-n-butyltin chloride (TBTC), triphenyltin chloride (TPTC), di-n-butyltin dichloride (DBTD), di-n-octyltin dichloride (DOTD; prepared from di-n-octyltin oxide per the official method [Standard Method of Analysis for Hygienic Chemists-With Commentary, Pharmaceutical Society of Japan, 1990]), and tri-n-octyltin chloride (TOTC) — sources: Tokyo Kasei Kogyo Co., Ltd, Tokyo, Japan (Bu/Oc tin standards); Fluka Chemie AG, Buchs, Switzerland (TOTC); Aldrich Chemical Co., Milwaukee, WI 53201 (DPTD). DOTD was purified on a 10% hydrated alumina column per the method of [Takami et al. 1988, Bunseki Kagaku 37:117–122]. Propylation reagent: n-propyl magnesium bromide (PMB), ≈2 M in tetrahydrofuran (Tokyo Kasei Kogyo). Phosphate–citrate buffer: 1.43 g Na₂SO₄·12H₂O, 17.3 g citric acid, 5.0 g NaCl dissolved in 800 mL H₂O; adjusted to pH 2.0 with 1 M HCl; final volume 1 L.

Extraction (sample-type-specific). Textile products: 1 g sample refluxed for 30 min at 70 °C in 75 mL methanol containing 0.05% HCl; 50 mL phosphate–citrate buffer and 50 mL H₂O added to the filtrate; CH₂Cl₂ extracted (2 × 30 mL); dried over anhydrous Na₂SO₄; evaporated. Water-soluble paint/adhesive: 1 g sample shaken vigorously with 20 mL MeOH and 1 mL HCl; filtrate treated with phosphate–citrate buffer and H₂O; CH₂Cl₂ extracted; evaporated. Lipo-soluble shoe polish/wax: 1 g sample homogenised with 2 mL MeOH, 50 mL 3 M HCl, and 10 g NaCl; extracted with 100 mL n-hexane–ethyl ether (4 + 6) twice; organic phase washed with 50 mL H₂O; evaporated.

Cleanup. Florisil column (Florisil PR, 3 g, 3 × 1.5 cm id) eluted with 40 mL ethyl ether (discarded fraction) followed by 40 mL ethyl ether–acetic acid (99 + 1) (collected fraction).

Derivatisation. Propylation: 1 mL n-hexane solution of the collected fraction combined with 1 mL PMB; 30 min at room temperature; 3 mL 0.5 M H₂SO₄ added dropwise to decompose excess PMB; reaction mixture transferred to a separatory funnel, flask rinsed with 10 mL EtOH, 70 mL H₂O added, n-hexane extracted (2 × 20 mL); organic phase dried over anhydrous Na₂SO₄; redissolved in n-hexane for FPD/GC and GC/MS analysis.

Quantitation. Calibration ranges (as propylated standards): dipropyl-DBT 0.2–5.0 µg/mL (as DBTD), dipropyl-DOT 0.2–10.0 µg/mL (as DOTD), propyl-TOT 0.5–15.0 µg/mL (as TOTC). Each detection-positive item was confirmed by GC/MS.

Spike recovery. OTC standards spiked at 3.0 ppm into four item categories (diaper cover, socks, adhesive, shoe polish); triplicate analyses; results in Table 1 (above). DPT recovery from textile products was 35–39%, the basis for dropping DPT from the determinative panel.

Limitations explicitly named by the authors. (i) Small sample frame — characterised by the authors as a “small-scale survey” rather than a definitive prevalence study; the conclusions note that the project does not reflect the prevalence in most items designated under the Law. (ii) DPT not determinable in textile products at the chosen recovery threshold. (iii) Dermal-absorption toxicology not assessed — the authors do not convert µg/g substrate concentrations into skin contact dose estimates because no examinations were performed on amounts contacting body surfaces and being absorbed. (iv) The authors flag that high concentrations were observed despite the small frame, and propose extending the survey with an improved DPT-recoverable sample preparation.

Implications

• Certification (HMTc). Adds 1990–1992 Japanese occurrence data for organotin substrate-content in household textile products and shoe polish to the HMI tin/organotin evidence base. Diaper covers in particular reached 0.1% w/w DOT (1666.7 µg/g) in one polyester sample, with DOT-positive in 6 of 10 diaper covers. For HMTc certification of categories under the wiki product taxonomy that overlap with the surveyed items (bibs, diapers-and-components, period-underwear-reusable-garments, infant-clothing), this paper is the legacy Japanese-baseline complement to the Stringer 2001 multi-country dataset (stringer2001-greenpeace-pvc-children-products). This page does not propose HMTc thresholds.
• App. Not directly relevant to ingredient contamination_profile data because no food matrices were sampled. Potentially relevant to a future textile/household-article screening surface that uses material composition (polyester, polyurethane, poly(vinyl chloride), cotton) and product category as a DOT/DBT/TOT contamination-likelihood predictor — particularly for diaper-cover sub-formats.
• Courses. Useful case study for the regulatory-history teaching module covering Japan’s 1973 Law for the Control of Household Products Containing Harmful Substances, the 1979–1980 Japanese phase-out of TBT and TPT in household products, and the contrast between regulated (TBT/TPT) and unregulated (DBT/DOT/TOT) organotin species in the same product categories. Also a methodological case study for tin-mode FPD/GC with PMB propylation and Florisil cleanup of textile/lipo-soluble matrices.
• Cross-reference. This paper is the temporally earliest organotin-in-consumer-product dataset in the HMI corpus and the only Japanese-jurisdiction entry. It is the analytical-methodology and country-coverage predecessor of stringer2001-greenpeace-pvc-children-products, which extended organotin-speciation surveys to PVC consumer products across 20 countries in 2001 and identified DBT and DOT as the predominant species — replicating the Yamada 1993 pattern of DBT/DOT predominance over TBT/TPT in non-marine consumer-product matrices.

Wiki pages this source may touch

• tin
• organotins
• bibs
• diapers-and-components
• period-underwear-reusable-garments
• infant-clothing

Verification notes

• Source identification. Yamada, S., Fujii, Y., Mikami, E., Kawamura, N., Hayakawa, J. (Aichi Prefectural Institute of Public Health, Nagare 7-6, Tsuji-machi, Kita-ku, Nagoya 462, Japan), and Aoki, K., Fukaya, M., Terao, C. (Aichi Prefectural Government, Department of Health, San-no-maru 3-1-2, Naka-ku, Nagoya 460-01, Japan). “Small-Scale Survey of Organotin Compounds in Household Commodities.” Journal of AOAC International Vol. 76, No. 2, pp. 436–441 (March/April 1993). Technical Communications section. Received December 23, 1991; accepted June 26, 1992. ISSN 1060-3271; issue identifier JAINEE 76(2) 257-460 (1993).
• DOI. No DOI located in the PDF or on the article’s title page. AOAC International journal articles from the early 1990s often lack registered Crossref DOIs; flagged with no_doi_assigned: true pending future verification against the Crossref or AOAC backfile.
• Author attribution. All eight authors are named on the article title page. The first five (Yamada, Fujii, Mikami, Kawamura, Hayakawa) are at the Aichi Prefectural Institute of Public Health; the last three (Aoki, Fukaya, Terao) are at the Aichi Prefectural Government Department of Health. The acknowledgments thank “Ikai Y.” for GC/MS operation.
• Source-tier rationale. evidence_tier: A: peer-reviewed publication in the Journal of AOAC International (Technical Communications section; the section retains peer review). Quantitative measurements use validated analytical chemistry (FPD/GC with GC/MS confirmation), explicit LODs per analyte, triplicate spike-recovery in four matrix types, and a controlled standard panel from named suppliers.
• Source-type rationale. source_type: peer-reviewed: Journal of AOAC International is a peer-reviewed scientific journal of AOAC International (the Association of Official Analytical Chemists).
• License rationale. copyrighted-third-party: AOAC International holds copyright on the 1993 volume per the masthead (“Copyright, 1993, by AOAC International, Inc. All rights reserved”). This page reproduces numerical findings under fair-use scholarly excerpting.
• Frontmatter metals: [Sn] is correct — the elemental analyte family is tin in all six measured organotin species. Per CLAUDE.md Part 14, the frontmatter abbreviation vocabulary uses Sn for tin. The speciation distinction (DBT, DOT, TOT all detected; TBT and TPT not detected; DPT not determinable) is captured in the Key numbers and Methods sections and in the [[metals/organotins]] wikilink in the “Wiki pages this source may touch” section. “organotins” is not in the abbreviation vocabulary, so the routing path is via the body wikilink section, matching the pattern used on stringer2001-greenpeace-pvc-children-products.
• Frontmatter products: field. Four slugs selected from the 2026-05-18 taxonomy snapshot, all reflecting direct sample classes in Table 2/Table 3:
  • bibs (10 items surveyed; 3 OTC-positive — Bib A 312.5 µg/g DOT, Bib B 59.7 µg/g DOT, Bib C 48.9 µg/g DOT);
  • diapers-and-components (10 diapers surveyed all OTC-negative + 10 diaper covers surveyed of which 6 were OTC-positive — Diaper covers A–F; this slug covers diapers and their components including diaper covers);
  • period-underwear-reusable-garments (10 “sanitary panties” surveyed; 2 OTC-positive — Sanitary panty A 144.4 µg/g DOT, Sanitary panty B 5.5 µg/g DBT. In Japanese consumer-product terminology, sanitary panty (生理用ショーツ) is a reusable menstrual-management garment, which maps to the period-underwear-reusable-garments slug);
  • infant-clothing (the diaper covers and bibs are within the infant/young-child clothing category; included as the broader umbrella because diaper covers are functionally a clothing layer used during infancy).
  • Test categories with no matching slug in the current 2026-05-18 taxonomy: adult underwear, socks, gloves, outergarments, stockings, shoe polishes, adhesives, paints, and waxes. Per Phase 1 frontmatter discipline I do not invent slugs for these; the test results are documented in the page body but are not routed by products: until matching slugs exist. This matches the discipline applied on stringer2001-greenpeace-pvc-children-products for the same family of dataset.
• Frontmatter ingredients: [] is correct — no food ingredients are involved.
• Frontmatter matrices: [] is correct — no food matrix is sampled. All values are reported as µg/g (≈ mg/kg, ≈ ppm) by weight of the consumer-product substrate, expressed as the dichloride for di-substituted organotin species and as the chloride for tri-substituted species.
• Frontmatter jurisdictions: [JP] reflects the Japanese sampling jurisdiction (Aichi Prefecture retail purchases, 1990–1992) and the Japanese regulatory context (Law for the Control of Household Products Containing Harmful Substances, Law No. 112, October 12, 1973; Ministry of Health and Welfare Ordinance No. 34, September 26, 1974, which banned TBT and TPT for household products in 1979–1980).
• Frontmatter near_duplicates points to stringer2001-greenpeace-pvc-children-products, the 2001 Greenpeace multi-country PVC-consumer-product survey that also performed organotin speciation (eight species) and reported DBT/DOT as the predominant non-marine organotin species in consumer items including bibs, diaper covers, and clothing. The two datasets are methodologically and thematically near-sibling rather than literally duplicate (different countries, different decades, different speciation panels) but cite-key reading benefits from cross-linking.
• Sample-size field. sample_n: 95 reflects the 95 items surveyed across the 13 product categories.
• Brand-firewall (Part 12, strict reading locked 2026-05-17). No brand names appear in the Key numbers section of this page. The Yamada 1993 paper itself does not name retailer or manufacturer brands; items are referred to generically by category plus an alphabetic suffix (Diaper cover A through F, Bib A through C, Sanitary panty A and B). Method-vendor names (Shimadzu, Hewlett Packard, JEOL, Tokyo Kasei Kogyo, Fluka, Aldrich, J&W Scientific) are retained in the Methods section under Exception 2 (scientific-method vendor/material names) per the audit-prompt rule lock.
• Wiki/HMTc firewall (Part 2). The Implications section flags Japan’s regulatory context (Law for the Control of Household Products Containing Harmful Substances; the 1979–1980 TBT/TPT phase-out for household products) without proposing HMTc thresholds for tin or organotins in textile categories. No HMTc certification level is endorsed or critiqued.
• Speciation flag. Reported organotin species are: DBT (di-n-butyltin, quantified as dichloride DBTD), TBT (tri-n-butyltin, quantified as chloride TBTC; not detected), DOT (di-n-octyltin, quantified as dichloride DOTD), TOT (tri-n-octyltin, quantified as chloride TOTC), DPT (di-n-phenyltin, excluded from the determinative panel because of poor textile-matrix recovery), TPT (tri-n-phenyltin, quantified as chloride TPTC; not detected). All values in Tables 2 and 3 are reported as the chloride/dichloride form (the propylated derivatives quantified by FPD/GC are back-calculated to the chloride basis of the standards). The page preserves the source’s chloride-basis reporting throughout; no µg/g→ppb conversion or speciation substitution is applied.
• Folder context vs paper scope. The PDF lives under raw/manual-fetch/Kimi_Agent_Download Corruption Issue/household_papers/03_Dish_Soap_Dishwasher/ in the Kimi corruption-issue raw tree, but the paper itself does not study dish soap or dishwasher detergents — the sampling frame is the 13 household-product classes designated under Japan’s 1973 Law, dominated by textile items (diapers, diaper covers, bibs, sanitary panties, underwear, socks, gloves, outergarments, stockings) plus shoe polishes, adhesives, paints, and waxes. The folder name reflects the Kimi-agent’s batch-organisation scheme during the corruption-issue rescue, not the paper’s content. No dish-soap or dishwasher-detergent category appears in Table 2 or Table 3 of the paper.
• Date and unit conventions. All organotin concentrations are reported in µg/g by weight of the consumer-product substrate (numerically equal to mg/kg and to ppm). Each Table 3 value is the mean of duplicate determinations and is expressed as the chloride/dichloride form of the species named. Detection limits (Methods) are reported in ppm and converted to µg/g for the body. Retention times are reported in minutes. Mass-spectral peaks are reported as m/z values.
• Internal-source consistency check. Table 2 (frequency) and Table 3 (concentrations) are mutually consistent on all 15 OTC-positive items. Table 2 column totals are: 4 DBT-positive (verified against Table 3: Diaper cover A 33.7, Diaper cover B 8.9, Diaper cover F 3.7, Sanitary panty B 5.5 — 4 entries); 14 DOT-positive (verified against Table 3: 14 numerical entries in the DOT column, with Sanitary panty B being the lone OTC-positive item without a DOT entry); 5 TOT-positive (verified against Table 3: Diaper cover A 58.3, Diaper cover C 128.5, Diaper cover D 91.3, Diaper cover E 33.6, Stockings 14.2 — 5 entries); 0 TBT-positive and 0 TPT-positive.
• Raw integrity. raw_sha256 = d307067002c8b40a3b7214009609d0f54402e56804c825045a4afe8d5670c7b0 computed from the PDF at the raw path.
• Audit subagent findings applied (2026-06-03). Fresh-context Agent subagent audit returned verdict REVISE with 2 ⚠️ findings (Check 1 — Numerical fidelity); verified each against the PDF and applied:
  • ⚠️ Applied: The GC/MS confirmation list for dipropyl-DBT was missing the m/z 235 fragment, listed the m/z 207 fragment with the wrong composition string ([Sn(C₃H₇)₂(C₄H₉)]⁺ when the source has [Sn(C₃H₇)₂]⁺), and listed the m/z 263 fragment with the same composition string as m/z 277 ([Sn(C₃H₇)(C₄H₉)₂]⁺ when the source has [Sn(C₃H₇)₂(C₄H₉)]⁺ at m/z 263). PDF p. 440 (“A base peak at m/z 221 [Sn(C₄H₉)(C₃H₇)]⁺ and fragment peaks at m/z 121 [Sn]⁺, 165 [Sn(C₃H₇)₂]⁺, 179 [Sn(C₄H₉)₂]⁺, 207 [Sn(C₃H₇)₂]⁺, 235 [Sn(C₄H₉)₂]⁺, 263 [Sn(C₃H₇)₂(C₄H₉)]⁺, and 277 [Sn(C₃H₇)(C₄H₉)₂]⁺ isotope peaks were observed for peak 1”). Corrected the fragment list to the full seven-peak source-faithful sequence. The subagent identified one of the two wrong composition strings (263); the deeper re-read against the PDF surfaced the missing 235 peak and the 207 transposition as well.
  • ⚠️ Applied: The Results-and-Discussion bullet “Diaper covers A and B both contained DOT near 1000 µg/g (≈ 0.1% by weight)” understated diaper cover B’s 1666.7 µg/g value. The paper’s narrative itself paraphrases diaper cover B as having “a similar concentration” to diaper cover A’s 902.8 µg/g, but the table values differ by ≈ 1.85×. Rewrote the bullet to give each diaper cover its own concentration and to flag the paper’s own loose paraphrase of B as “similar.”
  • Checks 2 (slug vocabulary), 3 (speciation/methods), 4 (Part 12 brand firewall), and 5 (Part 2 wiki/HMTc firewall) all returned ✅ clean from the subagent and were spot-checked clean here.

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.