Bitumen Grades: The Complete Grading System Guide

Somebody put a sample of binder into a defined apparatus, ran a defined test at a defined temperature, and the number that came out became the name the cargo travels under. That is all a grade is. It follows that a grade name is only meaningful when the standard behind it is named alongside it — Bitumen 60/70 on its own leaves the acceptance criteria undefined, while Bitumen 60/70 to ASTM D946 is an enforceable contract term.

System 1 — penetration grading

The oldest system and still the commercial language of the export trade. A standard needle carrying 100 g is allowed to sink into a conditioned sample for 5 seconds at 25 °C, and the depth reached is reported in tenths of a millimetre (dmm). The governing specifications are ASTM D946 in the United States, AASHTO M20 which mirrors it, and EN 12591 in Europe. The test method itself is ASTM D5, EN 1426 or IS 1203 — three documents describing the same needle. Penetration and hardness run in opposite directions: a lower number is a harder binder.

System 2 — viscosity grading

Introduced because penetration measured at 25 °C says nothing useful about how a binder will flow at 150 °C in a mixing plant. IS 73:2013, published by the Bureau of Indian Standards, classifies paving bitumen by absolute viscosity at 60 °C in poise, measured to IS 1206 Part 2 or ASTM D2171. Read the requirement carefully, because it is routinely misquoted: IS 73 states absolute viscosity at 60 °C as a minimum — 800 poise for VG-10, 1600 for VG-20, 2400 for VG-30 and 3200 for VG-40 — and controls the other end of each grade through the penetration and softening point lines rather than through a stated viscosity ceiling. The neat symmetrical bands often printed against VG grades, 800–1200 and 2400–3600 and so on, do not come from IS 73 at all. They belong to the older US viscosity system, AASHTO M226, which used the same 60 °C measurement and named its grades AC-10 through AC-40 with tolerance bands of ±20 % — AC-10 at 1000 ± 200 poise, AC-20 at 2000 ± 400. Quoting an M226 band as though it were an IS 73 requirement is the most common error on Indian tender correspondence. India moved off penetration grading with the 2006 revision of IS 73, and the 2013 edition is the current one, which makes IS 73 the governing specification for one of the largest import markets in the world.

System 3 — performance grading

The output of the US Strategic Highway Research Program in the early 1990s. AASHTO M320, and its ASTM twin ASTM D6373, name the binder after the pavement it has to survive rather than after a laboratory reading. PG 64-22 is a binder that must still resist rutting at a 64 °C pavement design temperature and must still resist cracking at a -22 °C design temperature. The measurements behind it are rheological — dynamic shear rheometer and bending beam rheometer — rather than empirical, and the binder is tested in three states: as supplied, after short-term ageing, and after simulated long-term ageing.

System 4 — the Chinese numbered system

JTG F40, the Chinese technical specification for construction of highway asphalt pavements, keeps a penetration-based number — 30, 50, 70, 90, 110, 130, 160 — but adds two things the western penetration standards never carried. A quality letter (A, B or C) sets limits on wax content and solubility, and a climate zone code ties the acceptable grade to the summer and winter temperatures of the region where the road is being built. A Chinese enquiry that specifies only the number is incomplete.

Why the systems do not convert arithmetically

There is no formula that turns a penetration number into a viscosity in poise, and none that turns either into a PG designation. Penetration is a single-point empirical measurement at one temperature. Viscosity at 60 °C is a different single point. A PG grade is the result of five separate instrument tests across three ageing states. Two binders can share a penetration band and differ substantially in temperature susceptibility, ageing behaviour and low-temperature stiffness, because those properties depend on the crude source and the process route, not on the grade name.

What this means commercially is simple. Cross-reference tables tell you which grade to offer against a specification written in another system. Only a laboratory report on the actual batch tells you whether that grade satisfies it. Where a project specification names a grade, supply that grade and let the specifying engineer approve any substitution in writing before the cargo loads.

ASTM D946 controls less than buyers assume

ASTM D946 defines five penetration grades: 40–50, 60–70, 85–100, 120–150 and 200–300. For each grade it sets a penetration band, a minimum flash point by Cleveland open cup, a minimum ductility at 25 °C, a minimum solubility in trichloroethylene and a minimum retained penetration after the thin film oven test. What it does not set is a softening point. There is no ring and ball requirement anywhere in D946.

This matters more than it sounds. Softening point is the property that predicts hot-weather rutting, and it is the number a buyer in a 45 °C climate most wants to see. A cargo can be fully compliant with ASTM D946 and still arrive at the soft end of the softening point distribution. If softening point matters to your project — and in a hot climate it does — write it into the contract explicitly with a range, because citing ASTM D946 alone will not get it for you.

EN 12591 controls softening point, and uses different bands

EN 12591 runs a wider ladder of grades from 20/30 through 250/330 and does specify a ring and ball softening point range for each one, measured to EN 1427. It also assesses ageing by the rolling thin film oven test (EN 12607-1) rather than the static thin film oven test used by ASTM.

The band boundaries do not line up. There is no EN 12591 grade called 60/70. A 60–70 dmm result straddles the 50/70 and 70/100 bands, and 50/70 is the grade normally offered against a 60/70 requirement in a European specification. Equally, 80/100 is not an ASTM D946 grade — the ASTM band is 85–100, which is why export cargo is quoted as 80/100 on refinery data sheets but tested against 85–100 when a D946 clause is invoked. Read those two facts together before you agree a substitution.

The retained penetration ladder

The most useful and least read line in ASTM D946 is retained penetration after the thin film oven test (ASTM D1754). The requirement tightens as the binder gets harder: 55 % for 40–50, 52 % for 60–70, 47 % for 85–100, 42 % for 120–150 and 37 % for 200–300. That single line is the standard’s durability clause. It answers the question ‘how much of this binder survives the mixing plant’, and a certificate that omits it has left out the half of the specification that governs whether the pavement is still intact in eight years.

Hard grades sit outside EN 12591

Very hard paving grades used in high-modulus asphalt are covered by EN 13924-1 rather than EN 12591, and are designated 10/20 and 15/25. If a European specification calls for a hard binder, check which of the two standards it cites before matching a grade to it.

What AASHTO M320 actually requires

A PG designation is earned across five instrument tests in three ageing states. On the binder as supplied: flash point minimum 230 °C (AASHTO T48), rotational viscosity maximum 3 Pa·s at 135 °C (AASHTO T316) so the binder can still be pumped and mixed, and dynamic shear rheometer G*/sin δ minimum 1.00 kPa at the high grade temperature (AASHTO T315). After short-term ageing in the rolling thin film oven (AASHTO T240): mass change maximum 1.00 %, and G*/sin δ minimum 2.20 kPa at the high grade temperature. After long-term ageing in the pressure ageing vessel (AASHTO R28, 20 hours at 2.1 MPa): G* × sin δ maximum 5000 kPa at the intermediate temperature, and on the bending beam rheometer (AASHTO T313) creep stiffness maximum 300 MPa with m-value minimum 0.300 at 60 seconds.

Two details trip buyers up. First, the bending beam test is run at the low grade temperature plus 10 °C, not at the low grade temperature itself — so a PG 64-22 is tested on the BBR at -12 °C. The 10 °C offset is a time-temperature superposition allowance, not a relaxation of the requirement. Second, the rotational viscosity limit of 3 Pa·s at 135 °C is a handling requirement, and heavily modified binders can fail it while passing everything else.

M332 and the traffic dimension

AASHTO M320 grades a binder against climate only. AASHTO M332 adds traffic through the multiple stress creep recovery test (AASHTO T350), which reports non-recoverable creep compliance Jnr at a 3.2 kPa stress level. Jnr is a compliance, so its unit is the reciprocal of a pressure: written 1/kPa on this page and kPa−1 on most laboratory reports, which are the same thing, and a larger number means a softer, more rut-prone response. The traffic designations are: S standard, Jnr max 4.5 1/kPa; H heavy, max 2.0; V very heavy, max 1.0; and E extremely heavy, max 0.5. A separate limit caps Jnr difference between stress levels at 75 %. A grade written PG 64E-22 therefore carries a climate half and a traffic half, which is why M332 designations cannot be read as if they were M320 designations.

JTG F40 and the number-plus-letter convention

The Chinese system starts from penetration, like ASTM and EN, but is structured differently. The grade number corresponds to a penetration band at 25 °C measured to JTG E20 method T 0604: 30 covers 20–40 dmm, 50 covers 40–60, 70 covers 60–80, 90 covers 80–100, 110 covers 100–120, 130 covers 120–140 and 160 covers 140–200. Note that the bands are 20 dmm wide and offset from the ASTM ladder, so Chinese 70 is not the same population as 60/70.

On top of the number sits a quality letter. Grade A is the specification for expressway and first-class highway surface and binder courses, Grade B for lower classes and lower structural layers, and Grade C for base courses and secondary roads. The letters differ mainly in purity and durability limits, and the two that matter commercially are wax content — a maximum of 2.2 % for Grade A, 3.0 % for Grade B and 4.5 % for Grade C by the distillation method — and solubility, at a minimum of 99.5 % for Grades A and B against 99.0 % for Grade C. Flash point minima also rise for the harder grades, at 260 °C for the 30, 50 and 70 numbers and 245 °C for 90.

The third element is the climate zone code, a two-part designation in which the first digit is the summer high-temperature zone and the second the winter low-temperature zone. JTG F40 tells the designer which grade numbers are acceptable in which zone, so the same road class takes a different binder in Guangdong than in Heilongjiang. A Chinese enquiry quoting only a number is missing two of the three pieces of information needed to price and certify it correctly.

The wax point is the one to remember

Wax content is the single clearest illustration of how differently these systems think. ASTM D946 does not limit wax at all, and EN 12591 does not carry a wax limit in its harmonised requirement table either — where a European project controls wax, the limit comes from a national annex or a project clause, never from the grade name. JTG F40 limits it explicitly and tiers the limit by road class. A binder can be perfectly compliant 60/70 under ASTM and be rejected in China on wax alone. If you are shipping to a Chinese buyer, get wax content tested and reported on the Certificate of Analysis before the cargo loads, not after.

Nothing about wax content

No penetration grade name in any system carries wax information. ASTM D946 sets no limit on paraffin wax, and the harmonised requirement table of EN 12591 does not set one either. Wax raises temperature susceptibility, so a waxy binder softens faster on a hot day and stiffens faster on a cold night than its grade name implies. It also interferes with adhesion to aggregate and promotes low-temperature brittleness. The test exists — EN 12606-1 by distillation, or JTG E20 method T 0615 in the Chinese system — but of the four grading systems on this page only JTG F40 makes it a grade acceptance criterion, at 2.2 %, 3.0 % and 4.5 % maxima for Grades A, B and C. If wax matters in your market, it has to be a contract line, because the grade name will not carry it for you.

Nothing about adhesion to your aggregate

Stripping is a property of the binder and the aggregate together, not of the binder alone. A binder that bonds well to a basalt will not necessarily bond to a siliceous river gravel or a limestone from a different quarry. No line in any binder specification predicts this reliably. The tests that do are run on the actual mixture: AASHTO T283 for moisture-induced damage, ASTM D3625 boiling-water stripping, or EN 12697-11. Where stripping has been a problem on previous work, run one of those on your materials and budget for an anti-stripping additive or hydrated lime rather than hoping a certificate covers it.

Nothing about ageing beyond what the residue tests show

Penetration and viscosity grading assess ageing with a short-term test only — TFOT under ASTM D1754 or RTFOT under ASTM D2872 — which simulates the few hours a binder spends in a hot mixing plant. Neither system requires the pressure ageing vessel, and only the performance grading system does. So a compliant 60/70 or VG-30 certificate tells you the binder survives the plant. It does not tell you how it behaves after eight summers in the pavement, because nobody tested for that.

Nothing about crude source or process route

Two cargoes can both be genuine 60/70 and come from entirely different places: straight-run vacuum residue, a semi-blown product, or a blend brought into band with a softer stream. They will share a penetration reading and differ in temperature susceptibility, ageing rate and compatibility with polymer. Penetration index, calculated from penetration and softening point together, is the cheapest partial view of this, and it is worth asking for when a new source is being introduced.

Nothing about handling safety, which the grade name actively obscures

Safety is set by the family, not by the grade number, and getting this wrong has consequences that a specification dispute does not.

• Flash point method is not interchangeable. Paving and oxidized grades are tested by Cleveland open cup (ASTM D92). Cutbacks must be tested by Tag open cup (ASTM D3143 or D1310). A Cleveland result quoted on a cutback certificate means the wrong test was run on the product whose safety case depends on it.
• Cutbacks release flammable vapour at ambient temperature. MC-30 and MC-70 can flash at 38 °C, below the shade temperature of a summer afternoon in most export markets. No open flame, no hot work nearby, no heating with a direct-fired burner.
• Emulsions and cutbacks fail in opposite directions. An emulsion is destroyed by freezing and a cutback is hazardous in heat, so a single storage rule written for “bitumen” will ruin one of them. Two grades that sit adjacent in the master table can need opposite storage instructions.
• Oxidized and enamel grades are worked hotter than paving grades. A softening point above 100 °C means application temperatures a paving crew would never encounter, and the grade number is the only hint of it.

The hazards that belong to hot bitumen as a material rather than to any particular grade — steam eruption where water is trapped beneath hot product, carbonisation and fire from a heating coil that is not submerged, and contact burns that keep transferring heat after the splash has stopped — apply identically to every row of the master table, which is precisely why no grade name warns you about them. Working temperatures and the controls that go with them belong on the individual grade and family pages, and the supplier Safety Data Sheet for the specific product is the document to work from. Read it before the cargo arrives rather than filing it when it does.