CDSI-001 Testing Protocol

Document number: CDSI-001 Version: 1.0 (Founding Draft) Status: Draft — open for scientific review Effective date: TBD (post-MOU with HIIMR) Supersedes: None Reference framework: ALS Environmental service request P1605022 (Nov 2016, Divine Tribe V3 ceramic donut atomizer)

1. Scope

This protocol specifies the procedure CDSI uses to characterize the volatile and semi-volatile compounds released by a cannabis concentrate vaporizer when operated under realistic consumer-loading conditions. It applies to:

• Ceramic donut atomizers
• Quartz banger / dab inserts heated by external sources (glass ovens, ball-vape heaters)
• Cartridge-style atomizers (510-thread, pod-style, proprietary)
• Dry-herb chambers operated as concentrate-on-screen platforms

Out of scope (covered by separate protocols, future versions): - Battery and electrical safety (deferred to UL 8139 reference) - Combustion-mode dry-herb vaporizers - Edible or topical preparation hardware

2. Reference standards

This protocol draws methodological reference from:

• NIOSH Manual of Analytical Methods, 1500-series (volatile organic compounds, sorbent tube collection)
• EPA Method 8260D (VOC by GC-MS)
• EPA Method 6020B (trace metals by ICP-MS)
• ALS Environmental service request P1605022 (2016 baseline test, this institute’s founding artifact)
• ASTM E2825-12 standards for indoor-air emissions methodology

CDSI does not claim equivalence to any single reference. The protocol is composite by design — VOC capture and metals analysis are not currently combined in any published cannabis-hardware standard, which is the gap this protocol exists to close.

3. Apparatus

3.1 Test chamber. Stainless steel or borosilicate glass enclosure, total volume 10–25 L, with controlled airflow (50–500 mL/min adjustable) and thermal stability ±2 °C. Single-pass airflow through the chamber to sorbent collection.

3.2 Device fixturing. Non-reactive PTFE or stainless steel mount holding the device under test in a consumer-realistic orientation. Power delivered via the manufacturer’s specified power source (mod, battery, AC supply). Power profile logged.

3.3 Sorbent tubes. Tenax TA for VOC capture (C5–C26 range). Dual-bed (XAD-2 / Tenax) tubes for SVOC. Backup tubes monitored for breakthrough.

3.4 Particulate filter. PTFE membrane filter (0.22 µm) downstream of the chamber outlet for metals collection.

3.5 Thermocouple instrumentation. K-type thermocouples on the heating element exterior, the device chamber wall, and the gas stream 10 cm downstream.

3.6 Analytical instruments. GC-MS for VOC/SVOC analysis; ICP-MS for trace metals on filter digests. Both instruments calibrated daily against NIST-traceable standards.

4. Sample preparation

4.1 Device samples. Three units of each device, drawn from production stock without notice to the manufacturer. Serial numbers logged. No pre-conditioning.

4.2 Concentrate matrix. Three certified concentrate matrices used in parallel runs: - Matrix A: Distillate (>90% THC, terpene content <2%) - Matrix B: Live resin (THC 65–80%, terpene content 5–10%) - Matrix C: Cured rosin (THC 60–75%, terpene content 4–8%)

All matrices sourced from California-licensed manufacturers, third-party-tested, lot numbers documented. CDSI maintains a frozen reference inventory.

4.3 Loading. Manufacturer-recommended quantity per session, single load per test run. No re-dosing during the run.

5. Procedure

5.1 Pre-run baseline. Empty chamber sampled for 10 minutes; sorbent and filter analyzed. Background VOC and metals must be below the protocol detection limits or the run is invalid.

5.2 Operating profile. Three duty cycles per device (3 replicates × 3 matrices = 9 runs minimum): - Cycle 1 (low): manufacturer’s lowest temperature setting, 5-second draws every 30 seconds, 20 draws total - Cycle 2 (mid): 480 °F target, 5-second draws every 30 seconds, 20 draws total - Cycle 3 (overshoot): manufacturer’s highest setting, 5-second draws every 30 seconds, 20 draws total

5.3 Capture. Continuous airflow through sorbent tubes for the duration of the run plus a 5-minute purge after the final draw. Particulate filter in line throughout.

5.4 Post-run. Sorbent tubes sealed, refrigerated, analyzed within 14 days. Filters digested with HNO3/HCl per EPA 3050B and analyzed within 30 days.

6. Target analyte list (initial)

6.1 VOC / SVOC (GC-MS)

• Carbonyls: formaldehyde, acetaldehyde, acrolein, glyoxal
• Aromatics: benzene, toluene, ethylbenzene, xylenes, styrene
• Terpene degradation products: alpha-pinene oxide, limonene oxide, p-cymene, methyl-furan
• Cutting agent indicators: vitamin E acetate, propylene glycol, polyethylene glycol breakdown products
• Heater material indicators: silicones, phthalate plasticizers

6.2 Trace metals (ICP-MS)

Lead, cadmium, chromium (total and hexavalent in separate aliquots), tin, arsenic, mercury, nickel, copper, zinc, antimony.

Detection limits and quantitation limits published with each report; minimums informed by USP <232> heavy-metals limits as a starting reference.

7. Quality assurance

7.1 Blanks. Every run includes a chamber blank (no device), a sorbent blank, and a filter blank. Detection above blank by 3× signal-to-noise minimum to call a finding.

7.2 Spikes and recoveries. Each batch of 10 runs includes one spiked sample at a known concentration; recovery must fall within 70–130% for the data to be valid.

7.3 Replicate variance. Across the three replicate units of the same device model, RSD must be <30% for any reported analyte. Higher variance is reported but flagged as low confidence.

7.4 Chain of custody. Each device is photographed on receipt, sealed in tamper-evident packaging, and tracked from intake to disposal. Custody log included in every report.

8. Reporting

Every CDSI-001 test produces a public report containing:

• Device identification, manufacturer, lot, serial numbers, photographs
• Concentrate matrices used, sources, lot numbers
• Full operating profile and instrument logs
• Analyte tables with detection limits, quantitation limits, observed values, RSD
• QA/QC results (blanks, spikes, recoveries)
• Chain of custody
• Conflict of interest disclosure (cover page)
• Pass/fail determination against the published acceptance criteria

Reports are published to the CDSI public registry within 90 days of intake, regardless of outcome. The manufacturer may submit a written response within 30 days of report finalization; the response is appended to the public report verbatim.

9. Acceptance criteria

CDSI-001 acceptance criteria (Tier 2 “Certified” mark — see CDSI-MARK-001):

Failure on any single category prevents Tier 2 certification but does not prevent publication; failed reports are public.

10. Open issues for scientific review

This document is a founding draft. Items flagged for HIIMR Scientific Advisory Board review before V1.0 finalization:

• Matrix selection: are three concentrate matrices sufficient, or should solventless rosin be split from cured rosin?
• Duty cycle realism: consumer-use telemetry data is sparse; should we partner with a manufacturer to anonymize real-world session profiles?
• Hexavalent chromium speciation: ICP-MS does not distinguish total from hexavalent; do we add IC-ICP-MS for speciation or accept total-chromium as the working surrogate?
• Terpene degradation product list: the initial list is conservative; degradation pathway literature should drive expansion.
• Acceptance criteria: USP and OEHHA references are starting points borrowed from adjacent fields; the SAB should review whether cannabis-specific limits should be derived independently.

Document history

• 2026-04-24 — V1.0 founding draft authored by Matt Macosko (Founder, CDSI)
• TBD — V1.0 issued after HIIMR Scientific Advisory Board review

This protocol is open for scientific review and replication. Comments to: protocols@cdsi.io (pending domain registration). Until then: matt@ineedhemp.com.