What SDS section 11 is required to contain 

Section 11 exists to translate a chemical hazard classification (established in Section 2) into the underlying toxicological evidence that supports it. Where Section 2 tells you the hazard category and the label elements that result from it, Section 11 tells you why—the study of data, exposure routes, and effects that justify the classification. 

Under OSHA’s Hazard Communication Standard, Appendix D specifies that Section 11 should address five broad categories of information: 

• Likely routes of exposure—inhalation, ingestion, skin contact, and eye contact, along with which route(s) are most relevant to how the chemical is actually used. 

• Symptoms related to exposure—the physical and physiological signs an exposed person might experience. 

• Immediate, delayed, and chronic effects from short-term and long-term exposure. 

• Numerical measures of toxicity—acute toxicity estimates, most commonly expressed as LD50 and LC50 values. 

• Carcinogenicity information, including whether the substance is listed by the National Toxicology Program (NTP), classified by the International Agency for Research on Cancer (IARC), or regulated as a carcinogen by OSHA. 

An important clarification: LD50 and LC50 map directly onto the “numerical measures of toxicity” requirement—they’re the values manufacturers are expected to report when data is available. NOAEL and LOAEL are not specifically named as required Section 11 data elements. They may appear because a manufacturer has chosen to include them, because repeated-dose study data exists and is relevant to chronic health hazard classification, or because the SDS author is being thorough. But their absence is not, by itself, a defect. 

Practical takeaway: If you pull up an SDS and NOAEL or LOAEL values are missing, that does not automatically mean the SDS is non-compliant or incomplete. It may simply mean that repeated-dose data wasn’t required for the classification, isn’t available, or wasn’t included by the author. Missing acute toxicity data (LD50/LC50) is a more meaningful gap, since those are explicitly tied to the required numerical toxicity measures.

LD50—Acute toxicity by ingestion or skin contact 

LD50 stands for “lethal dose, 50%. “It is the dose of a substance, administered in a single exposure, that is statistically expected to kill 50% of a test population of animals. The number is expressed in milligrams of substance per kilogram of body weight (mg/kg), which normalizes the dose across test subjects of different sizes. 

The “50” isn’t arbitrary—it reflects the midpoint of a dose-response curve. At lower doses, fewer than half of the test animals die; at higher doses, more than half die. LD50 is the statistically derived dose at which that mortality rate crosses 50%. Because it’s a midpoint estimate derived from a dose-response relationship (not a single observed data point), it typically comes with a confidence interval in the underlying study, even though SDSs usually report just the single value. 

Critically, an LD50 value is meaningless without knowing the route of administration and the test species it was derived from. A number reported as “LD50 = 300 mg/kg” tells you almost nothing on its own—you need to know whether that’s oral, dermal, rat, or rabbit before the number can be interpreted or compared to anything else.

Why a lower LD50 means greater toxicity 

LD50 operates on an inverse scale: the lower the number, the more toxic the substance, because it takes less of it to kill half the test population. 

Consider two substances: 

• Substance A has an LD50 of 5 mg/kg. 

• Substance B has an LD50 of 5,000 mg/kg. 

Substance A is far more acutely toxic—a tiny dose (5 mg per kilogram of body weight) is lethal to half the test subjects. Substance B requires a thousand times more material, by weight relative to body mass, to produce the same lethal effect. This inverse relationship trips people up constantly, especially those used to reading scales where “bigger number = worse.” With LD50, bigger numbers mean you’d need more of the substance to cause death — which generally means it’s less acutely dangerous, gram for gram. 

Why route and species matter 

LD50 values are route-specific and species-specific, and comparing across either dimension without adjustment is a common analytical error. 

Oral vs. dermal: A substance absorbed efficiently through the gut may behave very differently when applied to skin, where the stratum corneum acts as a barrier. Oral LD50 and dermal LD50 for the same chemical can differ by an order of magnitude or more, and neither predicts the other reliably. 

Rat vs. rabbit (and other species): Metabolic rate, enzyme systems, gut physiology, and skin permeability vary across species. A chemical metabolized quickly by rat liver enzymes into an inactive form might be processed far more slowly—or into a more toxic metabolite—in another species. Regulatory toxicology defaults rat data for oral and inhalation studies and often rabbit data for dermal studies, largely for historical and practical reasons, but the species used should always be checked before drawing conclusions. 

Because of this, you cannot meaningfully compare “LD50 = 40 mg/kg” from one SDS to “LD50 = 40 mg/kg” from another unless both values share the same route and species. An apparent match in the raw number can mask two entirely different testing conditions. 

What LD50 does not tell you 

LD50 is narrowly scoped to a single endpoint—death from a single exposure—and it does not tell you: 

• Whether the substance causes chronic organ toxicity (liver, kidney, or neurological damage) from repeated low-level contact. 

• Whether it causes sensitization (allergic response) after repeated skin or respiratory contact. 

• What happens under repeated-exposure conditions typical of an industrial workplace. 

• Anything about long-term workplace safety at sub-lethal exposure levels. 

A chemical can have a relatively high (less acutely toxic) LD50 and still be a serious long-term hazard—a classic example is a substance that’s mildly toxic in a single high dose but causes cumulative liver damage with repeated low-dose exposure. This is exactly the gap that NOAEL and LOAEL are designed to address. 

The actual OSHA acute toxicity cut-off values 

Numerical toxicity thresholds circulate widely online, and not all of them reflect current regulatory classification criteria. Some are drawn from older, superseded toxicity scales (such as pre-GHS hazard rating systems) that don’t align with how OSHA’s Hazard Communication Standard actually classifies acute toxicity today. When evaluating whether a chemical falls into a given hazard category, the applicable cut-off values are the ones specified in the GHS-aligned classification criteria adopted by OSHA, not older informal toxicity charts. 

OSHA acute toxicity categories 

The following table reflects the acute toxicity cut-off values used for GHS hazard classification: 

Category 1 represents the most severe acute toxicity (lowest LD50/LC50 values); Category 4 represents the least severe within the classified range. Substances with LD50/LC50 values above the Category 4 upper bound generally fall outside acute toxicity classification altogether, though they may still be classified under other hazard endpoints. 

How toxicity categories affect labels 

Each acute toxicity category corresponds to specific required label elements: 

• Category → signal word: Categories 1–3 typically require the signal word “Danger,” while Category 4 uses “Warning.” 

• Category → pictogram: More severe categories (1–3, depending on route) generally require the skull-and-crossbones pictogram; Category 4 for some routes may use the exclamation mark pictogram instead. 

• Category → hazard statement: Each category has a corresponding standardized hazard statement (e.g., “Fatal if swallowed” for Category 1–2 oral toxicity, versus “Harmful if swallowed” for Category 4). 

This is the practical link between the raw LD50/LC50 number buried in Section 11 and the pictogram an employee actually sees on a drum or bottle.

Reading section 11 against section 2 

One of the most useful things an EHS professional can do with Section 11 is use it to sanity-check. Section 2. Consider a scenario where Section 11 reports an oral LD50 of 40 mg/kg, but Section 2 lists the substance as Acute Toxicity Category 3. 

Against the table above, an oral LD50 of 40 mg/kg falls within the Category 2 range (>5–50 mg/kg), not Category 3 (>50–300 mg/kg). This is a real inconsistency, and it’s exactly the kind of discrepancy a careful reader should flag. It could indicate a transcription error in the SDS, a classification based on a different (unreported) study with a different result, mixture-specific classification rules that don’t map directly to the pure-substance LD50, or simply an outdated SDS that hasn’t been corrected. Whatever the cause, cross-referencing Section 11’s numerical data against Section 2’s stated category is a legitimate and valuable audit step—and it’s one that’s easy to perform once you know how to read the cut-off table correctly. 

This kind of cross-checking becomes considerably harder to do consistently across a large chemical inventory without some kind of centralized system—a point worth returning to later in this article. 

LC50: Acute toxicity by inhalation 

LC50 stands for "lethal concentration, 50%" and is the inhalation counterpart to LD50. Rather than a dose administered directly (as with oral or dermal exposure), LC50 describes the airborne concentration of a substance that is expected to kill 50% of a test population when they breathe it for a specified period—most commonly four hours under current GHS classification criteria. 

Why units matter 

Unlike LD50, which is consistently reported in mg/kg regardless of route, LC50 units vary depending on the physical form of the substance being tested: 

• Gases are reported in ppmV (parts per million by volume). 

• Vapors are reported in mg/L. 

• Dusts and mists are also reported in mg/L, though the underlying particle size and aerosol characteristics differ from vapors. 

Two SDSs reporting LC50 values in different units aren't necessarily inconsistent—they may simply reflect the different physical states in which a substance can exist (its vapor pressure, whether it's tested as a fine aerosol, etc.), and each unit convention is appropriate to that physical form. 

Why exposure duration matters 

Current GHS-based classification is built around a standardized four-hour exposure period. Older toxicological literature, however, sometimes reports LC50 data based on a one-hour exposure, a legacy of earlier testing protocols. A one-hour LC50 is not directly comparable to a four-hour LC50 without an appropriate conversion, since concentration and duration interact—a higher concentration for a shorter time can produce a similar toxic effect as a lower concentration over a longer time (a relationship often described qualitatively as similar to Haber's Rule, though real-world toxicant behavior doesn't always follow a strict linear concentration-times-time relationship). When comparing LC50 values across sources, checking the exposure duration is just as important as checking the units. 

LC50 and occupational exposure 

LC50 is particularly relevant for volatile liquids and gaseous substances, where inhalation is the dominant route of occupational exposure. A high-vapor-pressure solvent, for instance, may pose limited dermal risk under normal handling but a substantial inhalation risk in a poorly ventilated space—and LC50 data helps characterize that acute inhalation hazard. 

That said, LC50 should not be used, on its own, to select respiratory protection. That's a distinct decision process covered later in this article, but it's worth flagging here: a low LC50 signals that a substance is acutely dangerous to breathe, but it doesn't tell you what concentration is actually present in your workplace air, nor which respirator cartridge, filter class, or assigned protection factor is appropriate. 

NOAEL — The highest dose with no adverse effect 

NOAEL stands for "no observed adverse effect level." It's derived from repeated-dose toxicity studies, not single-exposure lethality studies, and it represents the highest tested dose at which no statistically or biologically significant adverse effect was observed in the study population, compared to controls. 

These studies are run over standardized durations depending on the toxicological endpoint being investigated: 

• 28-day repeated-dose studies (short-term, often used as a screening study). 

• 90-day (subchronic) studies, a common standard for characterizing repeated-dose effects. 

• Chronic studies, which may run a year or longer, sometimes spanning most of a test animal's lifespan, particularly when carcinogenicity or long-term organ effects are being evaluated. 

What NOAEL tells you 

It helps to contrast NOAEL directly against LD50, because they're answering fundamentally different questions: 

• LD50 asks: "What single dose is expected to cause death in half the test population?" 

• NOAEL asks, "At what repeated daily exposure level, over the duration of this study, did we see no observed adverse effect?" 

NOAEL isn't about death at all—it's about the threshold below which no measurable harm (organ damage, behavioral change, biochemical marker shift, reduced body weight gain, etc.) was detected across repeated administrations. 

The dose-spacing caveat 

This is a subtlety that's frequently glossed over, and it matters for anyone using NOAEL values in a risk assessment context. NOAEL is study-dependent—it is not a fixed biological constant for a given chemical. It is simply the highest of the specific doses that happened to be tested in that study, at which no adverse effect was observed. 

If a study tests doses of 10, 50, and 250 mg/kg/day, and no adverse effect is observed at 10 or 50 mg/kg/day but effects appear at 250 mg/kg/day, the NOAEL is reported as 50 mg/kg/day—not because 50 is some inherent biological threshold, but because it's the highest dose actually tested that didn't show an effect. A different study, testing doses of 10, 40, and 80 mg/kg/day, might have found effects starting at 40 mg/kg/day, producing a NOAEL of 10 mg/kg/day for the same chemical. The gap between studies isn't necessarily due to different chemical behavior—it can simply reflect different dose spacing chosen by the study designers. 

This is precisely why more sophisticated risk assessment approaches increasingly favor benchmark-dose (BMD) modeling, which fits a mathematical model to the entire dose-response curve rather than relying on a single tested dose point. BMD modeling can extract more information from the same dataset and reduces the dependency on exactly which doses happened to be selected for testing. 

The practical implication: NOAEL should never be presented, or interpreted, as a universal "safe threshold" for a chemical. It's a study-specific data point, bounded by the doses that were actually tested, and it should be read with that limitation in mind. 

LOAEL — The lowest dose where effects appear 

LOAEL stands for “lowest observed adverse effect “level”—the lowest dose tested in a repeated-dose study at which a statistically or biologically significant adverse effect was observed. 

NOAEL and LOAEL are companion values from the same study, describing the two ends of the transition zone: the highest dose with no observed effect (NOAEL) and the lowest dose where an effect first appeared (LOAEL). Conceptually, you can picture a progression along the dose scale—no effect, no effect, NOAEL, [gap], LOAEL, effect, effect—where the gap between NOAEL and LOAEL represents an untested region where the true no-effect threshold actually lies, somewhere between those two doses. 

Why there is no standard NOAEL-to-LOAEL ratio 

This point deserves direct emphasis, because it’s a place where informal or outdated guidance sometimes overstates what can be assumed. There is no fixed, universal ratio between NOAEL and LOAEL—not 2×, not 3×, not 6×, or any other consistent multiplier that reliably holds across chemicals or studies. 

The size of the gap between NOAEL and LOAEL is determined entirely by dose spacing—the specific doses the study designers chose to test. If a study tests doses of 10, 100, and 1,000 mg/kg/day, and the NOAEL turns out to be 10 mg/kg/day with the LOAEL at 100 mg/kg/day, that’s a tenfold gap — not because the chemical’s biology dictates a tenfold relationship, but because those were the doses selected. A differently designed study on the same chemical, with tighter dose spacing (say, 10, 25, 50, 100 mg/kg/day), might identify an NOAEL of 25 mg/kg/day and a LOAEL of 50 mg/kg/day—a twofold gap for the exact same underlying toxicological reality, simply because the study had finer resolution. 

Any claim that NOAEL-to-LOAEL relationships follow a predictable multiplier should be treated with skepticism; it’s an artifact of how a specific study happened to be designed, not a biological rule. 

What happens when only LOAEL is available? 

Sometimes a study doesn’t establish a clean NOAEL—every tested dose, including the lowest one, produced some observed effect. In that situation, risk assessors typically apply an additional uncertainty factor (sometimes called a safety factor) when deriving a reference value for human exposure. This additional factor exists specifically because, without a NOAEL, the true no-effect threshold is unknown—it could be just below the LOAEL, or it could be substantially lower. The added uncertainty factor is a way of building in a margin of safety to account for that unresolved gap.

LD50 vs. LC50 vs. NOAEL vs. LOAEL 

It helps to boil this down to two fundamentally different questions being asked of a chemical: 

• Acute toxicity asks: “How dangerous is a serious single exposure?” 

• Repeated-dose toxicity asks, “What happens with repeated exposure over time?” 

A chemical can score very differently on these two axes. A substance might have a relatively unremarkable LD50—meaning it’s not especially dangerous from one large exposure—while still carrying a low NOAEL, meaning that even modest repeated exposure produces measurable harm over time. Conversely, a substance might be acutely lethal at very low single doses (a low LD50) but show minimal cumulative effect at sub-lethal repeated doses. Reading only one axis and assuming it describes the other is one of the most common and consequential misreadings of SDS toxicological data. 

What these numbers do not decide: PPE selection 

This deserves to be stated plainly and up front: Section 11 toxicity values should not be used by themselves to select personal protective equipment. LD50, LC50, NOAEL, and LOAEL describe hazard severity—they do not translate directly into a specific respirator, glove, or protective garment recommendation. 

LC50 does not determine respirator selection. 

A low LC50 tells you a substance is acutely dangerous to inhale. It does not tell you what concentration is actually present in a given workplace, and respirator selection depends on several additional factors: 

• Estimated or measured exposure—actual airborne concentration in the work environment, typically obtained through industrial hygiene sampling. 

• The applicable exposure limit—the OSHA PEL, ACGIH TLV, or other OEL—the exposure needs to be compared against. 

• IDLH status—whether the substance is classified as immediately dangerous to life or health at certain concentrations, which drives different respirator requirements (including whether supplied-air or self-contained breathing apparatus is required). 

• Assigned protection factor (APF)—the level of protection a given respirator class is rated to provide, which must be matched against the ratio of measured exposure to the applicable exposure limit. 

• Requirements under 29 CFR 1910.134—OSHA's Respiratory Protection Standard, which governs program elements including fit testing, medical evaluation, and cartridge change-out schedules. 

An LC50 value can inform how seriously to treat a potential overexposure, but the actual respirator selection process runs through this separate regulatory and industrial-hygiene framework. 

Dermal LD50 does not determine glove selection. 

Similarly, a dermal LD50 value tells you about acute lethality from skin absorption in a test species—it says nothing about which glove material will actually protect a worker's hands. Glove selection depends on: 

• Permeation—the rate at which a chemical migrates through glove material at a molecular level, even without a visible breach. 

• Breakthrough time—how long a glove material can be in contact with a specific chemical before permeation reaches a detectable level on the inside surface. 

• Glove material—nitrile, neoprene, butyl rubber, PVC, and other materials all have dramatically different chemical resistance profiles. 

• Manufacturer chemical-resistance data—glove manufacturers publish permeation and breakthrough data for specific chemical-material pairings, and this data source is the appropriate reference point. 

• ASTM F739—the standard test method used to generate permeation and breakthrough time data for protective clothing materials. 

• 29 CFR 1910.138 — OSHA's Hand Protection Standard, which requires employers to select gloves appropriate to the specific hazards present. 

What does section 11 contribute to PPE decisions? 

None of this means Section 11 is irrelevant to PPE decisions—it plays an important supporting role. Section 11 tells you about severity (how serious an exposure could be) and exposure routes (inhalation, dermal, and ingestion) that need to be controlled. What it does not tell you is which specific PPE product to purchase, what glove material to specify, or what respirator cartridge to issue. That determination requires combining Section 11's hazard information with exposure monitoring data, applicable exposure limits from Section 8, and product-specific performance data from PPE manufacturers. 

Where the enforceable exposure limit is SDS section 8 

It bears repeating, because it’s the single most common point of confusion in this whole topic: LD50 ≠ LC50 ≠ NOAEL ≠ LOAEL ≠ workplace exposure limit. None of the four toxicity metrics discussed above is itself an enforceable or recommended limit on how much of a substance a worker can be exposed to during a shift. That number, when one exists, lives in Section 8 of the SDS. 

OSHA PELs 

An OSHA Permissible Exposure Limit (PEL) is a regulatory limit on the concentration of a substance in workplace air, generally expressed as a time-weighted average over an 8-hour shift. Many current OSHA PELs have a somewhat dated origin: a substantial portion were adopted in 1971, when OSHA was first established, largely by incorporating existing consensus standards rather than developing new toxicological assessments from scratch—drawing heavily on the 1968 ACGIH Threshold Limit Values (TLVs) and relevant ANSI standards of the era. 

OSHA later attempted a comprehensive update to modernize hundreds of PELs in 1989, but that effort was challenged in court and largely vacated by a 1992 judicial decision (the Eleventh Circuit’s ruling in AFL-CIO v. OSHA), which found that OSHA hadn’t adequately supported the update for each individual substance under the rulemaking requirements that applied. As a result, most PELs reverted to their original 1971-era values. OSHA itself has publicly acknowledged that many of its PELs are outdated relative to current toxicological understanding, and the agency has periodically encouraged employers to consider more current voluntary consensus limits as a supplement—not a replacement—for regulatory compliance. 

PEL vs. TLV vs. REL 

Because of this history, an EHS manager should not automatically assume the OSHA PEL is the most protective available number for a given substance. It's worth comparing the PEL against: 

• ACGIH TLVs (Threshold Limit Values)—voluntary, non-regulatory guidelines updated more frequently based on current toxicological literature. 

• NIOSH RELs (Recommended Exposure Limits)—recommendations from the National Institute for Occupational Safety and Health, often more conservative than legacy PELs. 

In many cases, the TLV or REL for a substance is considerably lower (more protective) than the legacy PEL, reflecting decades of toxicological research that hasn't been incorporated into the regulatory PEL itself. Comparing all three, rather than treating the PEL as automatically the governing or most protective figure, is considered a best practice in industrial hygiene.

Where NOAEL/LOAEL fit into modern risk assessment 

While NOAEL and LOAEL aren’t themselves exposure limits, they are foundational inputs into how modern risk-based exposure guidance is derived: 

• EPA Reference Doses (RfDs)—derived by dividing an NOAEL (or a benchmark dose) by a series of uncertainty factors to account for interspecies variability, human variability, and other sources of uncertainty. 

• ECHA Derived No-Effect Levels (DNELs)—used under the EU’s REACH regulation, similarly derived from NOAEL/LOAEL data adjusted with uncertainty factors. 

• Uncertainty factors—standardized multipliers (commonly factors of 10, applied cumulatively) used to extrapolate from animal study data to a value considered protective for human populations. 

• Benchmark doses—as discussed earlier, an alternative to NOAEL that models the full dose-response curve rather than relying on a single tested dose. 

It’s important to keep this modern risk-assessment framework conceptually separate from the historical derivation of OSHA PELs. PELs, for the most part, were not built using this NOAEL/uncertainty-factor methodology—they largely trace back to the 1968 ACGIH TLV list adopted wholesale in 1971. RfDs and DNELs, by contrast, are generally built using the newer, more formal NOAEL/LOAEL-plus-uncertainty-factor (or benchmark-dose) methodology. Both approaches aim at protective exposure guidance, but they come from different regulatory traditions and different eras of toxicological methodology.

What changed under HazCom 2024—and the deadline That moved 

OSHA published its updated Hazard Communication Standard (HCS-2024) in the Federal Register on May 20, 2024, aligning the U.S. standard primarily with the 7th Revised Edition of the GHS and incorporating select elements of the 8th revision. The rule brought changes to hazard classification criteria, label elements, and SDS content requirements, replacing the prior 2012 update (which had aligned with the 3rd GHS revision). 

One point worth flagging explicitly: the acute toxicity cut-off values used to classify substances into Categories 1 through 4 — the same values in the table earlier in this article — were not changed by the 2024 update. Those cut-offs remain the same under GHS Revision 7 as they were previously. 

HazCom 2024 compliance dates: OSHA set a tiered compliance schedule for HCS-2024, with different deadlines depending on whether an organization is a chemical manufacturer/importer/distributor or an employer, and depending on whether it involves substances or mixtures. On January 15, 2026, OSHA published a final rule extending those deadlines by four months, citing the need for additional time to finalize compliance guidance for the regulated community. 

Under the extended schedule: 

• May 19, 2026—chemical manufacturers, importers, and distributors were required to bring SDSs and labels for substances into compliance with HCS-2024. This deadline has now passed. 

• November 20, 2026—the deadline by which employers must update workplace labeling, hazard communication programs, and employee training related to substances, reflecting whatever revised classifications and SDSs their suppliers have provided. 

• November 19, 2027—the manufacturer/importer/distributor deadline for mixtures. 

• May 19, 2028—the corresponding employer deadline for mixtures. 

Until each applicable deadline, OSHA's transition provision allows compliance with either the 2012 standard, the 2024 standard, or a combination of both. This staggered timeline matters directly to anyone maintaining a large SDS library: because the manufacturer deadline for substances has already passed while the employer deadline is still months away, and because mixture deadlines extend well into 2027–2028, organizations are likely to be holding SDSs classified under different versions of the standard simultaneously for an extended period. 

A given facility might have some substances already reflecting HCS-2024 classifications, while related mixtures—or substances from suppliers who haven't yet updated their documentation—still reflect 2012-era Section 11 and Section 2 content. Tracking which version applies to which document, and confirming that classifications have actually been updated as suppliers reissue SDSs, is an ongoing administrative burden through at least 2028. 

Using section 11 well: A working checklist 

Pulling the discussion above into something usable day to day: 

• Read the route before the number. An LD50 or LC50 value is meaningless without knowing whether it’s oral, dermal, or inhalation and via which species. 

• Check inhalation units and duration. Confirm whether an LC50 is in ppmV or mg/L and whether it reflects a 4-hour or legacy 1-hour exposure period before comparing it to anything else. 

• Cross-check Section 11 with Section 2. Use the acute toxicity cut-off table to confirm the reported LD50/LC50 supports the stated hazard category. 

• Treat missing NOAEL/LOAEL as a data gap, not automatically a compliance defect. These values aren’t specifically required Section 11 elements. 

• Go to Section 8 for exposure limits. Section 11 tells you about the hazard; Section 8 tells you about the actual regulatory or recommended limit. 

• Never select PPE from Section 11 alone. Combine hazard severity with exposure monitoring data, applicable OELs, and manufacturer-specific PPE performance data. 

1. Re-review Section 11 when an SDS is revised. Classification changes—particularly as HCS-2024 compliance rolls out through 2026–2028—can shift acute toxicity categories, hazard statements, and required label elements even for chemicals you’ve handled for years. 

Keeping toxicological data current across a chemical inventory 

Reviewing a single SDS against this checklist is straightforward. Reviewing thousands of them, across an active chemical inventory with constantly rotating suppliers, is a different problem entirely. Supplier-issued revisions are easy to miss when they arrive as routine paperwork alongside a shipment, and a toxicological change buried in a revised Section 11—a new NOAEL from an updated repeated-dose study, or a reclassified acute toxicity category—can have downstream effects on required labeling, required PPE reassessment, and hazard communication training that go unnoticed if nobody is specifically watching for it. 

This is where a dedicated SDS management system earns its keep. Rather than relying on manual review of incoming paperwork, a well-built system can maintain authoritative, current versions of every SDS in an organization’s inventory, automatically flag when a supplier issues a revision, and surface classification changes that affect labeling or PPE requirements rather than leaving them buried in a document nobody reopens. Connecting that SDS data directly to the organization’s broader chemical inventory means a toxicological update to one SDS can be traced immediately to every location, process, and worker group where that chemical is actually in use—and the relevant people can be notified rather than discovering the change during the next audit. 

References

1. 29 CFR 1910.1200 — Hazard Communication. Office of the Federal Register, eCFR. https://www.ecfr.gov/current/title-29/subtitle-B/chapter-XVII/part-1910/section-1910.1200
2. 29 CFR 1910.1200 Appendix A — Health Hazard Criteria (Mandatory). OSHA. https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.1200AppA
3. 29 CFR 1910.1200 Appendix C — Allocation of Label Elements (Mandatory). OSHA. https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.1200AppC
4. 29 CFR 1910.1200 Appendix D — Safety Data Sheets (Mandatory). OSHA. https://www.osha.gov/sites/default/files/appendix_d.pdf
5. OSHA. Hazard Communication Standard, final rule, 89 FR 44144 (20 May 2024). https://www.osha.gov/hazcom/rulemaking
6. OSHA. Hazard Communication Standard, final rule extending compliance dates, 91 FR 1695 (15 January 2026). https://www.federalregister.gov/documents/2026/01/15/2026-00653/hazard-communication-standard
7. OSHA. HCS 2024 compliance date extension notice. https://www.osha.gov/hazcom/rulemaking/extension
8. 29 CFR 1910.1000—Air Contaminants, Tables Z-1, Z-2, and Z-3. OSHA. https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.1000
9. 29 CFR 1910.134 — Respiratory Protection. OSHA. https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.134
10. 29 CFR 1910.138 — Hand Protection. OSHA. https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.138
11. NIOSH. Pocket Guide to Chemical Hazards. Centers for Disease Control and Prevention. https://www.cdc.gov/niosh/npg/pgintrod.html
12. AFL-CIO v. OSHA, 965 F.2d 962 (11th Cir. 1992).
13. ACGIH. Threshold Limit Values for Chemical Substances and Physical Agents and Biological Exposure Indices, current edition.
14. ECHA. Guidance on Information Requirements and Chemical Safety Assessment, Chapter R.8: Characterization of dose-concentration response for human health. https://echa.europa.eu/documents/10162/17224/information_requirements_r8_en.pdf
15. US EPA. Benchmark Dose Technical Guidance, EPA/100/R-12/001 (June 2012). https://www.epa.gov/risk/benchmark-dose-technical-guidance
16. CCOHS. What are LD50 and LC50? https://www.ccohs.ca/oshanswers/chemicals/ld50.html
17. ASTM F739. Standard Test Method for Permeation of Liquids and Gases through Protective Clothing Materials under Conditions of Continuous Contact.
18. NTP. Report on Carcinogens. National Institute of Environmental Health Sciences. https://ntp.niehs.nih.gov/whatwestudy/assessments/cancer/roc
19. IARC. Monographs on the Identification of Carcinogenic Hazards to Humans. https://monographs.iarc.who.int/
20. ATSDR. Toxicological Profiles. https://www.atsdr.cdc.gov/toxprofiledocs/index.html

Conclusion 

Section 11 of an SDS packs a lot of technical information into a small space, and the four metrics discussed here are the ones most likely to be misread. LD50 and LC50 describe acute toxicity — the danger from a single, short-term exposure, expressed as a dose (mg/kg) or airborne concentration (ppmV or mg/L) expected to be lethal to half a test population. NOAEL and LOAEL describe repeated-dose toxicity—what happens with ongoing exposure over days, weeks, or months—and they're inherently study-dependent rather than fixed biological constants. Neither pair should be confused with Section 8's workplace exposure limits — the OSHA PELs, ACGIH TLVs, and NIOSH RELs that actually govern how much of a substance a worker can be exposed to. 

The core message worth carrying away from all of this: none of the four-toxicity metrics discussed here—LD50, LC50, NOAEL, or LOAEL—is a permission threshold for worker exposure. Each one is a piece of evidence that feeds into a larger picture of chemical hazard and risk, but reading any single number in isolation, without its route, species, duration, and units, will lead to the wrong conclusion more often than not. 

FAQ 

1. Is a lower LD50 more dangerous or less dangerous? 

More dangerous. LD50 operates on an inverse scale—a lower value means a smaller dose is expected to be lethal, indicating higher acute toxicity. 

2. Does an SDS have to include NOAEL and LOAEL? 

No. NOAEL and LOAEL are not specifically required Section 11 elements under OSHA's Hazard Communication Standard. LD50 and LC50, by contrast, are the expected numerical measures of acute toxicity when such data is available. 

3. Why is my SDS showing an LC50 in mg/L when another shows ppm? 

Units depend on physical form. Gases are typically reported in ppmV, while vapors, dusts, and mists are typically reported in mg/L. This reflects the substance's physical state during testing, not necessarily an inconsistency between documents. 

4. Can I use the LC50 to decide which respirator to issue? 

Not by itself. Respirator selection depends on measured or estimated workplace exposure, the applicable exposure limit, IDLH status, assigned protection factors, and the requirements of 29 CFR 1910.134—not the LC50 alone. 

5. Is the OSHA PEL a safe exposure level? 

Not necessarily the most protective one available. Many OSHA PELs date to 1971 and haven't been substantively updated since a 1992 court decision vacated a broader modernization effort. Comparing the PEL to more current ACGIH TLVs and NIOSH RELs is considered good practice. 

On a tiered schedule. The manufacturer/importer/distributor deadline for substances (May 19, 2026) has already passed. The employer deadline for substances is November 20, 2026. Mixture deadlines follow in November 2027 (manufacturers) and May 2028 (employers).