Same Report, Two Contradictory Conclusions: Why a Passing PV and a Short OSI Can Coexist
An oil leaves the refinery with a peroxide value of 0.8 meq O2/kg — comfortably inside the Codex limit of 10 and even inside the stricter voluntary benchmark of below 1 for fresh refined oil. The FFA reads 0.08%, well below the 0.3% refined-oil ceiling. On paper, the batch passes.
Six weeks later, the same oil is rejected by the customer for rancidity. The OSI, run on a retained sample, had already signalled the problem at the refinery gate: the induction time was roughly half of what the previous batch achieved under the same Rancimat conditions.
This is not a laboratory error. It is a predictable consequence of what each test actually measures. Peroxide value is a snapshot of hydroperoxides present at the moment of sampling. It is not a record of oxidative history, and it is not a forecast of remaining stability. Hydroperoxides accumulate, then decompose into aldehydes and ketones; a PV can fall while the oil is moving deeper into rancidity. OSI, by contrast, measures how long the oil resists reaching that secondary breakdown under accelerated conditions. The two numbers answer different questions, and a refinery that manages only the first will ship oil that fails on the second.
The diagnostic value of an oil quality test panel lies not in checking each specification independently, but in reading the four routine indicators — FFA, PV, OSI and smoke point — as a set that points to a specific stage of the refining equipment line. KMEC has over two decades of experience in oil refining equipment and process configuration. The sections below treat each indicator as a signal from a particular unit operation, then combine them into a troubleshooting matrix.
1. FFA: The Deodorizer’s Outlet Number, With Causes Upstream
1.1 What FFA Actually Measures
Free fatty acids are the product of triglyceride hydrolysis, triggered by chemical or enzymatic action and accelerated by heat and moisture. FFA is expressed as a percentage of oleic acid, or as acid value in mg KOH/g oil. Codex Stan 210 sets the refined-oil ceiling at 0.3% as oleic acid; cold-pressed and virgin oils are allowed far higher limits because they are not deodorized. The voluntary industry benchmark for refined oil is tighter still, in the region of 0.05%.
1.2 Where FFA Problems Originate
Hydrolysis can begin long before the refinery: poor seed storage, delayed pressing, or wet extraction conditions all raise the FFA load entering the line. Neutralization in chemical refining removes most of it. But for a plant running a full refining equipment line, the final FFA “exit” is the deodorizer.
Deodorization is vacuum-steam distillation. FFA is stripped along with volatile odor compounds under high temperature and deep vacuum. A 2017 study on rapeseed oil deodorization found that the acid value only reached the national standard once the deodorization temperature reached 230°C, and that extending time beyond the point where PV stabilised produced little further benefit for acid value. Below that window, stripping is incomplete; the oil may pass a smell test while the acid value remains stubbornly high.
Diagnostic logic: If refined FFA is persistently high and neutralization has been verified, the likely fault is not the neutralizer but the deodorizer’s temperature–vacuum–stripping steam match. A unit with uneven tray temperature distribution or weak vacuum ejector capacity can run at “standard conditions” on paper while delivering far lower FFA removal than design.
2. PV: The Most Misleading “Pass” on the Certificate
2.1 PV Is a Point-in-Time Measurement
Peroxide value measures hydroperoxide concentration, usually in meq O2/kg. Hydroperoxides are primary oxidation products; they rise, then fall as they decompose into aldehydes, ketones and acids. PV is therefore most informative for fresh oils at low oxidation levels — it is “most properly applied as a measure of the quality of fresh oils,” which should read 1 meq/kg or less. PV above 3 meq/kg is already treated by industry as oxidised, and the Codex refined-oil ceiling is 10 meq O2/kg.
2.2 Why PV Can Pass While Stability Fails
Deodorization removes hydroperoxides effectively: under vacuum and high temperature, they are volatilised or decomposed. A batch can therefore leave the deodorizer with a PV of 1 or below. But the same thermal treatment that removes hydroperoxides also destroys the oil’s endogenous antioxidants, principally tocopherols. The result is an oil that is chemically “clean” at the moment of testing but has lost much of its capacity to resist oxidation during storage.
This is the mechanistic basis for the opening scenario. PV answers “what oxidised material is present now?” OSI answers “how much defence remains?” A refinery that optimises only PV can systematically produce oil with a short shelf life while every individual PV result looks acceptable.
3. OSI: Quantifying the Defence That Remains
3.1 What the Rancimat / OSI Test Measures
The oil quality test known as OSI (Oxidative Stability Index), or Rancimat induction time, is an accelerated ageing method. Air is passed through the sample at constant elevated temperature; volatile secondary oxidation products are swept into a measuring vessel, and the time until the conductivity of the measuring solution begins to rise is recorded as the induction time. The standard methods are AOCS Cd 12b-92 and EN ISO 6886. Induction time is a relative stability indicator, not a direct shelf-life figure: as a rule of thumb, raising the test temperature by 10°C roughly halves the induction time.
3.2 What Shortens OSI — and What Does Not
The dominant controllable variable in a refinery is the deodorization temperature–time combination. A plant-scale study on physically refined rapeseed oil found that total tocopherol retention fell linearly with deodorization temperature: 91.5% at 220°C, approximately 80% in the 230–240°C range, and 54.7% at 270°C. Since tocopherols are the oil’s main natural chain-breaking antioxidants, this retention curve translates directly into OSI behaviour. A separate rapeseed study confirmed that tocopherol and β-carotene degradation at 270°C was far faster than at other deodorization temperatures, and that antioxidant capacity stabilised only after extended time at 250°C.
Two qualifications matter for honest diagnosis. First, deodorization is not the only step that affects OSI: neutralisation, bleaching and winterisation also remove or degrade minor components, and their cumulative effect must be considered when OSI falls. Second, deodorization can in some circumstances improve OSI by removing FFA and pro-oxidant metals. The correct inference from a short OSI is therefore not “the deodorizer is always the culprit,” but “the temperature–time–vacuum setpoint in deodorization is the first controllable variable to audit.”
4. Smoke Point: Where FFA and Phospholipids Overlap
4.1 Smoke Point Is Not a Single-Variable Function
Smoke point is the temperature at which visible blue smoke appears during heating. It is often treated as a proxy for FFA, and FFA does lower it: a 2016 study on first-grade soybean and rapeseed oils established log-linear relationships between FFA content and the percentage reduction in smoke point, with R² of 0.9753 for soybean oil and 0.8995 for rapeseed oil. But the same body of work produced a broader regression including acid value, peroxide value, phospholipids and moisture, with an R² of 0.848, and a later rapeseed-specific model that ranked moisture and volatile matter, peroxide value, phosphorus content, acid value and solvent residue in descending order of influence on smoke point.
4.2 What a Low Smoke Point With Normal FFA Tells You
If smoke point is low while FFA is within specification, the FFA–smoke point relationship cannot explain the gap. The more likely causes are incomplete degumming (residual phospholipids) or inadequate vacuum drying (moisture). Raising deodorization temperature in this situation is the wrong response: it may push FFA lower still while further degrading tocopherols, without fixing the underlying phospholipid or moisture problem. The correct audit is the degumming section and the vacuum dryer, not the deodorizer.
5. The Four-Indicator Troubleshooting Matrix
Reading indicators individually tells you whether a batch passed. Reading them together tells you which unit needs attention.
| Indicator pattern | Most likely process fault | First equipment/unit to audit |
|---|---|---|
| FFA high, PV normal, smoke point low | Deodorizer stripping efficiency insufficient | Deodorizer temperature profile, vacuum system, stripping steam rate |
| FFA normal, PV normal, OSI markedly short | Deodorization over-severe; tocopherol loss | Deodorization temperature setpoint and residence time (check against the 220–240°C retention curve) |
| FFA normal, PV high, OSI normal or short | Post-deodorization oxygen ingress | Nitrogen blanketing, pipeline seals, headspace oxygen in finished-oil tanks |
| FFA normal, smoke point low, PV/OSI acceptable | Incomplete degumming (residual phospholipids) or high moisture | Degumming section, vacuum dryer, bleaching adsorbent |
| FFA and PV both high, OSI short | Whole deodorizer underperforming, or upstream oxidative load too high | Full deodorizer audit plus feedstock/pretreatment oxidation control |
The logic behind the matrix is that FFA is the deodorizer’s outlet number, smoke point ties FFA to degumming and drying, PV reflects the balance between peroxide removal and re-generation, and OSI exposes damage to the antioxidant defence system. Only the combination localises the fault.
KMEC’s refining equipment line is configured around this joint reading. Deodorizer temperature, vacuum and stripping steam are not set against a single FFA target; they are checked simultaneously against PV removal, OSI retention and smoke point compliance. Degumming and bleaching windows are set from the combined smoke-point and phosphorus-residue feedback. For clients running both oil press machines and refining, the oxidative load carried from pressing and extraction — feedstock freshness, pressing temperature, residual solvent — propagates directly into the refinery’s PV and OSI results, so the diagnostic read should cover the chain from pressing through to the deodorizer.
6. FAQ
Q1: What are the Codex limits for FFA and PV in refined oil?
Codex Stan 210 sets refined-oil FFA at 0.3% as oleic acid and PV at 10 meq O2/kg. Cold-pressed and virgin oils are allowed 15 meq O2/kg. Industry voluntary targets for refined oil are tighter, commonly around FFA ≤0.05% and PV below 1 meq/kg for fresh oil.
Q2: Why can an oil pass the peroxide value test but still turn rancid quickly?
PV measures hydroperoxides present at sampling. Deodorization can remove these while simultaneously destroying tocopherols. The oil then has a low PV but a depleted antioxidant defence, which OSI detects as a short induction time. PV describes the present; OSI describes the remaining resistance.
Q3: What does the Rancimat / OSI test actually measure?
It measures induction time under accelerated oxidation: air is passed through the oil at constant elevated temperature, and the time until volatile secondary products raise the conductivity of a measuring solution is recorded. Standard methods are AOCS Cd 12b-92 and EN ISO 6886. It is a relative stability indicator, not a direct shelf-life figure.
Q4: How much tocopherol is lost at different deodorization temperatures?
In plant-scale physically refined rapeseed oil, total tocopherol retention falls from 91.5% at 220°C to about 80% at 230–240°C, and to 54.7% at 270°C. The relationship is approximately linear with temperature, which is why the deodorization setpoint is the first variable to audit when OSI is short.
Q5: Is there a fixed minimum deodorization temperature for FFA removal?
No single universal figure applies. In one rapeseed study, acid value only reached the national standard at 230°C. For high-oleic sunflower oil, however, deodorization at 220–235°C with total sparge steam above 2.0% is reported to deliver high tocopherol retention together with low free and trans fatty acid contents. The window depends on oil type and whether refining is chemical or physical.
Q6: What should I check first if smoke point is low but FFA is within specification?
Check degumming and drying, not the deodorizer. Residual phospholipids and moisture both depress smoke point independently of FFA. Regression models that include phosphorus and moisture alongside acid value explain smoke point variation better than FFA alone. Raising deodorization temperature in this case risks further tocopherol loss without addressing the cause.
References
- Čmolík, J., Pokorný, J., Réblová, Z. & Svoboda, Z., “Tocopherol retention in physically refined rapeseed oil as a function of deodorization temperature,” European Journal of Lipid Science and Technology, 2008 — total tocopherol retention 91.5% at 220°C, ~80% at 230–240°C, 54.7% at 270°C.
- Wu, Y., Fang, B., Wu, H., Zhang, M. & Shen, Y., “Establishment of Mathematical Relationships Between Smoke Point and Minor Compounds in Vegetable Oils Using Principal Component Regression Analysis,” Chemical Engineering Transactions, Vol. 51, 2016 — ln smoke point = 4.790 − 0.021×ln(acid value) + 0.014×ln(PV) − 0.051×ln(phospholipids) − 0.105×ln(moisture), R² = 0.848.
- Fang, B., Zhu, N., Wang, Y. & Luan, X., “Establishment of the mathematical model between free fatty acid and smoke point in soybean oil and rapeseed oil,” Journal of the Chinese Cereals and Oils Association, 31(10), 2016 — FFA vs smoke-point reduction: R² 0.9753 (soybean), 0.8995 (rapeseed).
- Fang, B., Luan, X., Li, X. & Wang, Y., “Establishment of Mathematical Equations of Microconstituents and Smoke Point in Rapeseed Oil Based on Principal Component Regression Analysis,” Journal of the Chinese Cereals and Oils Association, 32(4), 2017 — ln smoke point model with R² = 0.926; influence ranking: moisture > PV > phosphorus > acid value > solvent residue.
- Zhou, R., He, R., Ju, X., Wang, B., Wu, Y. & Xu, F., “Effect of Deodorization Process on Microconstituent and Antioxidant Capacity of Rapeseed Oil,” Chinese Institute of Food Science and Technology Annual Conference, 2017 — acid value reaches national standard at 230°C; tocopherol and β-carotene degrade much faster at 270°C than at other temperatures.
- Martincic et al., cited in Tocopherol — an overview, ScienceDirect — high-oleic sunflower oil deodorization at 220–235°C with total sparge steam above 2.0% for high tocopherol retention and low free/trans fatty acid content.
- Codex Alimentarius Commission, Codex Stan 210 — refined oil FFA 0.3% as oleic acid; PV 10 meq O2/kg for refined oils.
- US AMS / AOCS reference data on OSI and PV — PV most properly applied to fresh oils, target ≤1 meq/kg; PV above 5 significantly reduces shelf life of finished products; OSI per AOCS Cd 12b-92.
- Metrohm Application Bulletin 204/3 e, “Oxidation stability of oils and fats — Rancimat method” — AOCS Cd 12b-92 and EN ISO 6886; induction time definition; 10°C temperature increase approximately halves induction time.
For deodorizer setpoint optimisation, FFA/PV/OSI/smoke-point joint interpretation, or refining equipment line configuration, contact the KMEC engineering team.