NOx O2 Correction Formula and Japan’s Reference O2 Levels

A boiler test report from Japan lists "NOx 60 ppm (O2 5% kansan)." Your air permit sets a limit in ppmvd at 3% O2, or in mg/Nm3 at 6% O2. You cannot compare the two numbers until they share the same oxygen reference, moisture basis, and units.
O2 kansan (O2換算) means "O2 correction." It is the calculation that restates a measured NOx or dust (particulate) concentration as if the flue gas contained a fixed reference amount of oxygen. Japan uses the formula C = Cs × (21 − On) / (21 − Os), and the reference level On is set by law for each type of facility.
This guide explains what the formula does and why it uses 21, then walks through a dilution example. It lists Japan's reference O2 levels, compares them with US, EU, and Indian practice where we could verify it, and shows how to restate a Japanese supplier's figure on your own basis. It ends with a checklist, questions for the supplier, and an email template.
Contents
- What O2 Correction Means on a Japanese Spec Sheet
- Why Emissions Are Corrected to a Reference O2
- Worked Example: Dilution Does Not Change the Corrected Value
- Japan's Reference O2 Levels by Facility Type
- How Japan's Reference O2 Compares with the US, EU, and India
- Dry Basis, ppm, and mg/Nm3: What Else Must Match
- Restating a Japanese Supplier's NOx Figure on Your Basis
- Checklist and Questions for the Japanese Supplier
- Limits of This Guide
- FAQ
- Sources
What O2 Correction Means on a Japanese Spec Sheet
Japanese specifications and test reports write the oxygen basis next to the value, for example "NOx 150 ppm (O2 4%換算)" or "O2 12%換算値." Read this as "corrected to 4% O2" or "at 12% O2." The legal reference level is called hyojun sanso nodo (標準酸素濃度), or standard oxygen concentration.
The formula comes from the Enforcement Regulations of Japan's Air Pollution Control Act (Taiki Osen Boshi Ho, 大気汚染防止法), in the notes to its emission-standard tables for dust and nitrogen oxides:
Japanese law states NOx as cm3 per m3 of flue gas, converted to 0°C and 1 atm. One cm3 in one m3 is 1 ppm by volume, so you can enter ppm directly. Dust uses the same formula, with the unit g per m3 of flue gas at 0°C and 1 atm.
The "21" is the oxygen content of air. Dry air is about 20.95% oxygen by volume, and Japan rounds this to 21. As shown later, US EPA rules use 20.9 instead.
Each of the four values in the formula has a different source:
Two values are measured, one is set by law, and one is the result.
The 20% cap on Os comes from the notes to Appended Tables 2 and 3-2 of the Enforcement Regulations (dust and NOx).
You do not choose On. It depends on the facility type and on the pollutant (NOx or dust). A boiler that burns only gas, for example, uses 5%; the main values are in the table later in this article.
Why Emissions Are Corrected to a Reference O2
The correction gives every facility the same yardstick. A boiler or incinerator burns fuel with oxygen from the air, and the oxygen that is not used leaves the stack with the flue gas. As a result, air is 21% oxygen, but flue gas contains less.
The figure below shows where the oxygen goes:
Combustion uses up oxygen from the air. Only what is left over reaches the flue gas.
The O2 level in the flue gas depends on how much excess air is supplied relative to the fuel (the air ratio).
The flue gas O2 level depends heavily on the air ratio the unit runs at. A higher air ratio means more oxygen in the flue gas, and a lower one means less.
The air ratio (kuki-hi, 空気比; also called excess air ratio or lambda) is the actual air supplied divided by the air needed for complete combustion.
Because the O2 level changes with operation, raw concentrations from two units cannot be compared on the same scale. Correcting every result to one fixed O2 level solves this.
A Numerical Example of Dilution
Take a flue gas flow of 100 Nm3/min at 3% O2. Nm3 (normal cubic meter) is a volume converted to 0°C and 1 atm.
If the gas contains 200 ppm NOx, the NOx flow is 100 × 200 × 10−6 = 0.02 Nm3/min. The oxygen flow is 3 Nm3/min.
A ppm value is only a concentration, so mixing in outside air can push it as low as you like. Add 100 Nm3/min of fresh air, and the total becomes 200 Nm3/min. That air carries 21 Nm3/min of oxygen, so total oxygen rises to 24 Nm3/min, or 12%.
The figure below shows the mass balance:
The amount of NOx stays the same. Only the concentration drops.
O2: 3 Nm3/min
NOx: 0.02 Nm3/min (200 ppm)
O2: 21 Nm3/min
NOx: 0 Nm3/min
O2: 24 Nm3/min
NOx: 0.02 Nm3/min (100 ppm)
The NOx flow is still 0.02 Nm3/min, so the concentration falls to 0.02 ÷ 200 = 100 ppm. The same amount of NOx leaves the stack, but the number is cut in half.
A concentration limit would mean little if dilution could meet it. O2 correction closes this gap by asking what the concentration would be at a fixed O2 level, such as 6%. Japan's Ministry of the Environment states that the standard-oxygen correction was adopted to prevent operators from meeting emission standards by diluting the flue gas with air.
Worked Example: Dilution Does Not Change the Corrected Value
Now correct both states from the previous section to 6% O2. The first is the basic calculation; the second confirms what happens after dilution.
Before Dilution
For flue gas at 3% O2 with 200 ppm NOx, enter Os = 3 and On = 6. The corrected value is about 167 ppm:
The measured O2 (3%) is below the reference (6%), so the corrected value is lower than the measured 200 ppm.
After Diluting with Excess Air
After adding 100 Nm3/min of fresh air, the gas is at 12% O2 with 100 ppm NOx. Corrected to 6% O2, the result is again about 167 ppm:
Because of this, adding air to increase the total gas volume no longer helps a facility slip under a concentration limit. The comparison below puts both states side by side:
The measured concentration drops by half, but both cases come to 167 ppm at 6% O2.
The corrected values match because all of the added oxygen came from the dilution air.
The ratio (21 − Os)/21 explains why. It is the share of the flue gas that is combustion gas, not dilution air. At 3% O2 that share is 18/21; at 12% it is 9/21.
Dividing a measured value by this share takes it back to the undiluted concentration, the "0% O2" equivalent. Both 200 ÷ (18/21) and 100 ÷ (9/21) come to about 233 ppm. O2 correction then multiplies by the reference share, 15/21 when On = 6: 233 × 15/21 ≈ 167 ppm.
To remember which goes where: the reference (On) is on top, and the measured value (Os) is on the bottom. The further the measured O2 is above the reference (the more diluted the gas), the more the corrected value rises above the measured one.
Japan's Reference O2 Levels by Facility Type
Japan sets On for each facility type in the Enforcement Regulations of the Air Pollution Control Act. NOx values are in the notes to Appended Table 3-2, and dust values are in the notes to Appended Table 2. Boilers split three ways by fuel: 5% for gas, 4% for liquid fuel, and 6% for solid fuel.
The table below lists the main facility types, checked against the regulation text on e-Gov, Japan's official law database:
Boilers split three ways by fuel. Diesel engines and gas turbines use high values, while gas and gasoline engines use 0%.
Taken from the notes to Appended Tables 2 and 3-2 of the Enforcement Regulations (e-Gov text as consulted in October 2026; check the version currently in force). For NOx, small boilers with a heat transfer area under 10 m2 form separate categories (solid fuel 6%, liquid fuel 4%). For dust, some facilities, such as boilers burning black liquor and facilities that use electricity as the heat source, use C = Cs with no correction.
For many of the main facility types, NOx and dust share the same On. The two tables group facilities differently, though, so check which entry your facility falls under for each pollutant.
A few special cases also apply to NOx. Some lead-oxide furnaces use C = Cs with no correction, and glass melting furnaces that burn with pure oxygen multiply the corrected value by an extra factor of 1/4. If equipment falls into one of these groups, ask the supplier which formula they applied.
Example: Checking a Gas-Fired Boiler
Suppose a gas-only boiler (On = 5) measures 8% O2 and 80 ppm NOx. The corrected value is about 98 ppm:
This 98 ppm is compared with the emission standard in Appended Table 3-2. For a gas-only boiler with a flue gas volume of 40,000 to under 500,000 m3N/h, the national standard is 100 cm3/m3 (100 ppm). The flue gas volume here is the maximum hourly volume converted to 0°C and 1 atm.
You can also work backward from the limit to find the raw reading that corresponds to it. Rearranging the formula gives Cs = C × (21 − Os) / (21 − On):
This back-calculation is a guide for reading the analyzer during operation. Compliance is decided against the standard and conditions that actually apply to the facility, which can include transitional rules based on installation date and stricter prefectural limits.
If analyzer readings go into a report, the analyzer must also be properly calibrated.
Corrected values also help when operating conditions change. Preheating combustion air with an air preheater saves energy, for example, but you need to check that it does not raise the NOx concentration, and corrected values make the before-and-after comparison easier.
How Japan's Reference O2 Compares with the US, EU, and India
Other jurisdictions use the same idea, but often with different reference levels. The table below covers only the values we could confirm in the regulation text or a reliable secondary source. Your own permit or consent conditions take precedence over any general rule.
| Equipment | Japan (On) | US (examples) | EU (Industrial Emissions Directive) | India (example) |
|---|---|---|---|---|
| Gas-fired boiler | 5% | 3% O2, dry, in many state and local boiler rules (e.g., South Coast AQMD Rule 1146 series) | 3% (liquid and gaseous fuels) | Not confirmed |
| Oil-fired boiler | 4% | Check your permit | 3% | Not confirmed |
| Coal / solid-fuel boiler | 6% | Check your permit | 6% | 6%, dry (thermal power plants) |
| Gas turbine | 16% | 15% (40 CFR 60 Subpart KKKK) | 15% | Not confirmed |
| Diesel engine | 13% | Check your permit | Not confirmed | Not confirmed |
| Gas engine | 0% | 15% for spark-ignition engines when limits are in ppmvd (40 CFR 60 Subpart JJJJ) | 15% | Not confirmed |
| Waste incinerator | 12% | 7% (e.g., 40 CFR 60 Subpart Eb, Subpart EEEE) | 11% (3% for waste oil) | Not confirmed |
Three differences stand out:
- Engines. Japan corrects gas and gasoline engines to 0% O2, while US SI-engine rules and the EU use 15%. The same engine therefore shows a figure about 3.5 times higher on the Japanese basis (21/6 = 3.5).
- Gas turbines. Japan uses 16% and the US and EU use 15%. The same gas reads 20% higher at 15% O2 than at 16% (6/5 = 1.2).
- The constant. Japan and the EU write the formula with 21. US EPA rules use 20.9, for example Cadj = Cmeas × (20.9 − 7)/(20.9 − %O2) for 7% O2. The effect is small: correcting 200 ppm from 3% to 6% O2 gives 166.7 ppm with 21 and 166.5 ppm with 20.9.
Some limits are not concentrations at all. Subpart KKKK, for instance, also allows output-based limits in lb/MWh, which cannot be compared with a Japanese ppm figure through O2 correction alone.
Dry Basis, ppm, and mg/Nm3: What Else Must Match
The O2 reference is only one of four bases that must line up. The other three are moisture, unit, and reference temperature.
Dry vs. Wet Basis
US rules state ppmvd, "parts per million by volume, dry," and EPA defines both Cmeas and %O2 on a dry basis. EU limits are stated after correction for water vapor. Japanese compliance results are commonly reported on a dry gas basis (kawaki gasu, 乾きガス); one municipal incinerator report, for example, gives its values "on a dry gas basis, corrected to 12% O2."
Flue gas from natural gas firing contains a lot of water vapor, so a wet-basis value reads noticeably lower than a dry one. If a Japanese data sheet does not say "dry," ask. Do not apply the O2 formula to a wet NOx value and a dry O2 value at the same time.
ppm vs. mg/Nm3
Japan states NOx limits in volume (cm3/m3, or ppm). The EU and India state them in mass per volume (mg/Nm3), and EU limits express NOx as NO2. To convert ppm to mg/m3, multiply by the molar mass and divide by the molar volume at the reference temperature:
- At 0°C and 101.3 kPa (EU and Japanese "N" conditions): 1 ppm NO2 = 46.01 ÷ 22.41 ≈ 2.05 mg/Nm3
- At 20°C (68°F) and 101.3 kPa (US EPA standard conditions under 40 CFR 60.2): 1 ppm NO2 ≈ 1.91 mg/m3
Correct for O2 first or last; the order does not matter, as long as both values are on the same dry basis. What does matter is using the right reference temperature. A "Nm3" from Japan or the EU is at 0°C, while "standard" in US rules means 20°C.
Restating a Japanese Supplier's NOx Figure on Your Basis
If you have the raw readings (measured NOx Cs and measured O2 Os, both dry), apply your own formula and reference level directly. That is the cleanest method. If you only have a corrected figure, you can move it from one reference to another, because the measured O2 cancels out:
Follow these steps:
- Read the Japanese basis. Find the O2 reference (the "○% kansan" note), dry or wet, and the unit (ppm, or mg/m3N at 0°C).
- Find your basis. Take the reference O2, moisture basis, unit, reference temperature, and averaging time from your permit or consent, not from a general rule.
- Re-correct the O2 reference with the formula above, or recalculate from the raw readings.
- Convert units if your limit is in mg/Nm3 or mg/m3, using the reference temperature your rule specifies.
- Compare like with like. Check that the load condition, fuel, and averaging period of the Japanese test match what your limit covers.
Here are three fictional examples, worked with the constant 21:
- Gas-fired boiler. The supplier guarantees 60 ppm at 5% O2. At 3% O2: 60 × (21 − 3)/(21 − 5) = 60 × 18/16 = 67.5 ppm. In EU-style units (0°C, as NO2), that is about 67.5 × 2.05 ≈ 138 mg/Nm3 at 3% O2.
- Gas turbine. The test report shows 15 ppm at 16% O2. At 15% O2: 15 × (21 − 15)/(21 − 16) = 15 × 6/5 = 18 ppm.
- Gas engine. The data sheet shows 200 ppm at 0% O2. At 15% O2: 200 × (21 − 15)/21 ≈ 57 ppm.
The gas engine example shows why the basis matters. A Japanese figure that looks far above a 15% O2 limit may be well within it once restated, and a gas turbine figure that looks compliant may not be. Treat these results as screening only, and confirm the final comparison with your permit, the permitting authority, or the supplier.
Checklist and Questions for the Japanese Supplier
Use this checklist before you compare a Japanese emission figure with your limit:
- O2 reference of the Japanese figure (e.g., "O2 4% kansan") and of your limit
- Dry or wet basis for both the pollutant and the O2 reading
- Unit: ppm (cm3/m3N), mg/Nm3, or a mass-rate or output-based unit
- Reference temperature and pressure (0°C, 20°C, or 25°C; 101.3 kPa)
- NOx expressed as NO2 or as NO, if the limit is in mass units
- Formula constant: 21 or 20.9
- Load, fuel, and operating mode during the test
- Averaging time (instantaneous, 1-hour, daily) and measurement method (JIS K 0104 in Japan)
- Whether the figure is a guarantee, a design value, or a measured result
- Any special formula (no correction, or the 1/4 factor for oxy-fuel furnaces)
Questions to send to the Japanese side:
- What O2 reference is the NOx (or dust) value corrected to, and is it the Japanese standard O2 for this equipment type?
- Are both the NOx and O2 values on a dry basis?
- Can you provide the raw measured NOx and O2 values from the test, before correction?
- At what load, fuel, and ambient conditions was the test run, and over what averaging time?
- Can you restate the guarantee at our reference conditions (O2 %, dry basis, unit, reference temperature)?
You can adapt this email template for the request:
Subject: NOx emission data for [equipment / model] – basis confirmation
Dear [Name],
Thank you for the test report for [equipment / model]. The report lists NOx as [value] ppm "O2 [x]% kansan." Our permit sets the limit as [value] [ppmvd / mg/Nm3] at [y]% O2, dry basis, [reference temperature], [averaging time].
Could you please confirm whether the NOx and O2 values are on a dry basis, and send the raw measured NOx and O2 values with the load and fuel during the test? We would also appreciate it if you could restate the guaranteed NOx value at [y]% O2, dry basis, in [unit].
Best regards,
[Your name]
[Company]
Limits of This Guide
- The Japanese values are national standards from the Enforcement Regulations of the Air Pollution Control Act. Prefectures can set stricter limits, and older facilities may fall under transitional rules.
- Reference O2 levels outside Japan vary by rule, state, equipment size, and permit. The comparison table gives examples, not your applicable limit.
- Re-correcting a corrected figure assumes both are on the same dry basis and that the original correction used the standard formula. Working from the raw readings avoids that assumption.
- This article covers NOx and dust concentrations. Japan regulates sulfur oxides (SOx) differently (see the FAQ), and some limits are mass-rate or output-based.
- Compliance decisions belong to the facility's environmental team and the permitting authority. Use this guide to ask the right questions, not as a compliance determination.
FAQ
What if the measured O2 is above 20%?
Enter 20%. Under Japan's rules for dust and NOx, an Os above 20% is treated as 20%, so a reading of 20.5% goes into the formula as 20. As Os approaches 21%, the denominator 21 − Os approaches zero and the corrected value becomes extremely large.
Facilities whose On is defined as equal to Os, or that use C = Cs, are not corrected at all.
What does "corrected to 0% O2" mean?
It is the concentration before any dilution by air. With On = 0, the formula becomes C = Cs × 21/(21 − Os). Japan uses 0% for gas engines and gasoline engines.
For example, a gas engine at 10% Os with 300 ppm NOx gives 300 × 21/11 ≈ 573 ppm, and that is the value compared with the standard. For the same gas, a 0% figure is always larger than a 6% figure, so match the O2 basis before comparing reports.
Does Japan apply O2 correction to SOx?
No. Japan regulates sulfur oxides with the "K-value" method, which sets an allowable hourly emission rate from the effective stack height and a regional constant K. The formula is q = K × 10−3 × He2, where He is the corrected stack height.
Designated areas also have total-emission controls and fuel-use standards. Sulfur in the fuel can also cause low-temperature corrosion in boilers.
How is O2 correction related to the air ratio?
The air ratio m can be estimated from flue gas oxygen as m ≈ 21/(21 − O2). On that basis, the factor (21 − On)/(21 − Os) reads as "air ratio at measurement ÷ reference air ratio."
The air ratio formula is an approximation, however, so the two do not match exactly. O2 correction is a regulatory calculation, while the air ratio is an operating value for combustion control.
Running at a lower air ratio, without going too low, reduces the heat carried away by the flue gas.
How do I convert ppm at 15% O2 to ppm at 3% O2?
Multiply by (21 − 3)/(21 − 15) = 3, or by (20.9 − 3)/(20.9 − 15) ≈ 3.03 if your rule uses 20.9. A value of 9 ppm at 15% O2 is about 27 ppm at 3% O2. Both values must be on a dry basis.
Sources
- Correction formula, standard O2 levels, 20% cap, and NOx emission standards: e-Gov, Enforcement Regulations of the Air Pollution Control Act (Japanese) (Articles 3–5, Appended Tables 2 and 3-2)
- Prefectural limits stricter than national standards: e-Gov, Air Pollution Control Act (Japanese) (Article 4)
- Purpose of the correction (preventing dilution with air): Ministry of the Environment, Japan, "Dust Regulations" (Japanese)
- NOx concentration and total-emission controls: Ministry of the Environment, Japan, "Hazardous Substance Regulations" (Japanese)
- SOx K-value regulation: Ministry of the Environment, Japan, "Sulfur Oxide (SOx) Regulations" (Japanese)
- Japanese results reported on a dry gas basis (example): Sano City, incinerator emission measurement results (Japanese)
- Air preheating and checking NOx: Ministry of the Environment, Japan, SHIFT program, measure 121222 "Preheating Combustion Air with Boiler Flue Gas" (Japanese)
- Oxygen content of dry air (20.95%): NASA, "Earth Fact Sheet"
- US 7% O2 correction with the 20.9 constant (dry basis): 40 CFR 60.2975; 40 CFR 60.58b (Subpart Eb)
- US gas turbine limits in ppm at 15% O2 or lb/MWh: 40 CFR 60 Subpart KKKK, Table 1
- US spark-ignition engine limits in ppmvd at 15% O2: 40 CFR 60 Subpart JJJJ, Table 1
- US standard conditions (20°C, 101.3 kPa): 40 CFR 60.2
- Boiler limits at 3% O2, dry (state/local example): South Coast AQMD, Rules 1146/1146.1 working group presentation
- EU reference O2 for combustion plants (6%, 3%, 15%) and reference conditions: Directive 2010/75/EU, Annex V, Part 1 (legislation.gov.uk copy)
- EU waste incineration at 11% O2 (3% for waste oil), NOx as NO2: Directive 2010/75/EU, Annex VI (legislation.gov.uk copy)
- India thermal power plants corrected to 6% O2, dry basis (Gazette amendment as reproduced in a tribunal filing): National Green Tribunal, affidavit in OA 203 of 2022
Original article in JapaneseThis guide is adapted from a MONOSCOPE article written for readers in Japan: 排ガスのO2換算とは?NOx濃度の計算式・計算方法と標準酸素濃度


