Bitumen 35/50: EN 12591 Specification, Applications and Export Supply

As with every penetration grade, the name is a measurement rather than a brand or a quality claim. The sample is conditioned to 25 °C, a standard needle carrying a 100 g load is released into it for five seconds, and the depth the needle reaches is read off in tenths of a millimetre. Any result from 35 to 50 makes the material a 35/50. The procedure is EN 1426, which is technically equivalent to ASTM D5 and IS 1203. European practice writes the unit as ×0.1 mm where American practice writes dmm; it is the same quantity under a different house style, and nothing turns on which one appears on the certificate.

Penetration and hardness run in opposite directions, so a lower number means a stiffer binder. On the EN 12591 ladder 35/50 sits one rung softer than 30/45 and one rung harder than 40/60, in the crowded middle-hard section of the range. It is the grade European practice reaches for when a pavement has to carry heavy, slow or channelised traffic through a warm summer, and it is the grade most often named for wearing and binder courses in Mediterranean and other warm-climate national specifications drawn from European standards.

What the grade is for

The engineering case for a harder binder is straightforward. Permanent deformation — rutting — is driven by the binder losing stiffness as the pavement warms and by traffic that loads the same wheel path repeatedly and slowly. Moving from 50/70 to 35/50 raises the softening point window by four degrees at both ends and raises the aged softening point floor with it, which buys resistance to that failure mode. The cost is paid at the other end of the temperature scale, in flexibility, fatigue tolerance and workability, and that trade is the whole reason a specification writer has to choose rather than simply reaching for the hardest available grade. The section on where 35/50 is the wrong choice, further down, sets out what the trade actually costs.

What EN 12591 controls that an ASTM data sheet does not

A buyer used to reading Gulf and South Asian export data sheets should understand that the difference between an EN order and an ASTM order is less a difference in the binder than a difference in what the producer is obliged to prove. EN 12591 asks for four things that an ASTM-based export data sheet usually does not carry:

• A two-sided softening point band for every grade. Determined by EN 1427, and for 35/50 the band is 50–58 °C. ASTM D946 sets no softening point requirement whatsoever; the softening point figures printed on export data sheets for ASTM grades are refinery and trading practice, not requirements of that standard.
• Ageing by RTFOT, not TFOT. Resistance to hardening is assessed by the rolling thin-film oven test to EN 12607-1, and three separate limits are placed on the residue: change of mass, retained penetration and softening point after hardening.
• A Fraass breaking point. EN 12593 measures the temperature at which a thin film of binder cracks when it is cooled and flexed. It has no counterpart anywhere in ASTM D946. Note how EN 12591 handles it: the Fraass limit belongs to the group of properties whose requirements are applied according to national circumstances and climate rather than to the group demanded of every grade in every country, which is exactly why it has to be named in the order.
• Solubility in toluene. EN 12592 uses toluene where ASTM D2042 uses trichloroethylene. The limit is the same 99.0 %, but the solvent is not, so a laboratory reporting solubility in TCE has not run the EN method.

All four are checkable on a Certificate of Analysis before a cargo moves. If a supplier offers 35/50 and the certificate carries only penetration, softening point, ductility and TCE solubility, that document is not evidence of EN 12591 conformity. It is an ASTM-style data sheet with a European grade name typed at the top.

Ageing, and why the limits tighten as the grade hardens

The ageing requirement is worth understanding on its own terms, because it is the part of the EN specification that most directly predicts how the pavement behaves after the mixer, and because its numbers move from grade to grade in a way the penetration band does not explain.

The rolling thin-film oven test to EN 12607-1 ages a continuously renewed film of binder inside rotating glass bottles with an air jet, at 163 °C. EN 12591 then places three limits on what comes out. Change of mass for 35/50 is capped at ± 0.5 % — note the plus-or-minus, because a binder can gain mass through oxidation as well as lose it through volatilisation, and a certificate reporting only a loss figure has answered half the question. Retained penetration must be at least 53 % of the original value, so a cargo entering the test at 42 ×0.1 mm must still measure at least 23 afterwards. And the softening point after hardening must be at least 52 °C, which puts a floor under the aged binder’s high-temperature stiffness.

Compare that last figure with the neighbours. For 30/45 the aged softening point floor is 54 °C, for 35/50 it is 52 °C and for 40/60 it is 50 °C. Across the EN 12591 range the floor tracks two degrees above the bottom of each grade’s fresh softening point band, which is a compact way of saying that the standard expects a harder grade to still be a harder grade after the mixing plant has had its way with it. This matters commercially far more than it looks, because it is one of the two lines that genuinely separate three grades whose penetration bands overlap. That is the subject of the next two sections.

The one thing to settle before reading any 35/50 certificate

EN penetration bands are deliberately wide, and adjacent bands deliberately overlap. A binder measuring 40 ×0.1 mm is inside 30/45, inside 35/50 and inside 40/60 at once, and if its softening point lands between 52 and 56 °C it satisfies all three softening point bands as well. Nothing on that certificate tells you which grade the material is, because on those two lines it is all three. Which grade it can be sold as depends on the ageing and low-temperature limits, on what the producer actually tested against, and on what the contract named. Read every 35/50 certificate against the grade written in your own purchase order rather than against whichever grade heading the producer printed on the page.

Start with the arithmetic

Take the three bands as EN 12591 publishes them and intersect them.

• Penetration. 30/45 covers 30–45, 35/50 covers 35–50 and 40/60 covers 40–60. All three contain 40 to 45 ×0.1 mm. A binder measuring 40, 42 or 45 is inside every one of them.
• Softening point. 30/45 requires 52–60 °C, 35/50 requires 50–58 °C and 40/60 requires 48–56 °C. All three contain 52 to 56 °C.

So a certificate reporting a penetration of 42 ×0.1 mm and a softening point of 54 °C satisfies the two headline requirements of 30/45, 35/50 and 40/60 simultaneously. There is no ambiguity to resolve and no mistake to find. On those two lines the material genuinely is all three grades, and a producer could truthfully head the same page with any of the three names.

Then find where the three grades stop being the same

They separate on the lines a buyer skims: the ageing residue and, where it is called up, the low-temperature limit. Those are set per grade, and they move as the grade name changes.

• Softening point after hardening (EN 1427 on the EN 12607-1 residue). Minimum 54 °C for 30/45, 52 °C for 35/50, 50 °C for 40/60.
• Retained penetration after RTFOT. Minimum 53 % for 30/45 and for 35/50, 50 % for 40/60.
• Fraass breaking point (EN 12593), where the national annex calls it up. Maximum -5 °C for 30/45 and for 35/50, and a more demanding maximum of -7 °C for 40/60. Confirm the figure for your grade against the current edition, because this property sits in the group applied according to national circumstances.

Note that the Fraass line runs the other way from the ageing lines. Naming the harder grade tightens the ageing floor and relaxes the brittleness limit; naming the softer grade does the reverse. No grade in this cluster is simply stricter than the others on every line, which is precisely why the choice cannot be made by picking the hardest name available.

A worked example, because the arithmetic is the argument

Consider a batch with these measured values: penetration 42 ×0.1 mm, softening point 54 °C, retained penetration after RTFOT 55 %, softening point after hardening 53 °C, Fraass breaking point -6 °C.

• Against 35/50: penetration inside 35–50, softening point inside 50–58, retained penetration above 53 %, aged softening point above the 52 °C floor, Fraass below the -5 °C ceiling. It passes every line.
• Against 30/45: penetration and softening point still pass, but the aged softening point of 53 °C is below the 54 °C floor that 30/45 requires. It fails.
• Against 40/60: penetration, softening point and both ageing lines pass comfortably, but the Fraass result of -6 °C is above the -7 °C ceiling that 40/60 requires where that property is called up. It fails.

That batch is a 35/50 and nothing else, and no part of that conclusion could have been reached from the penetration figure. Move the numbers slightly — aged softening point 55 °C and Fraass -9 °C, on the same 42 and 54 — and the identical material clears all three grades outright. That is the honest position: the overlap is real and it is complete, and whether a given batch is one grade, two or three depends on lines that most certificates report last and most buyers read least.

What follows for the buyer

Four things follow, and each is worth acting on before a contract is signed rather than after a cargo has sailed.

1. The grade in the contract is a commercial choice as much as a technical one. Where a project specification names 35/50 and a producer holds material that would satisfy 35/50, 30/45 and 40/60 alike, nothing physical is changed by writing one name rather than another. What changes is the limit table the cargo will be judged against, the tests the producer has to run and be prepared to defend, and the argument available to each side if a result at destination lands near an edge. Choose the name your specification requires, and be aware that the name is doing work the material is not.

2. A certificate has to be read against the grade you named, not the grade it is headed with. This is the practical failure mode. A producer with a batch at 43 ×0.1 mm may head the certificate 40/60 because that is what the previous customer bought, and offer it against your 35/50 order on the basis that the penetration and softening point results fall inside your band. They may well do. But if the batch was produced and released against the 40/60 limit table, the aged softening point may have been accepted at 51 °C — compliant for 40/60, below the 52 °C floor for 35/50 — and nobody has done anything wrong except offer you a certificate written for a different grade. Check the ageing lines against your grade every time.

3. Band edges are softer than they look. EN 1426 and EN 1427 both publish precision statements giving the spread expected between two results in one laboratory and between two laboratories. A batch whose true penetration is 40 ×0.1 mm sits exactly on the lower edge of the 40/60 band and comfortably inside 35/50; a re-test at destination reporting 39 puts it outside 40/60 with neither laboratory being wrong. Any grade claim that depends on a value at the very edge of a band is fragile for that reason alone. Where a cargo is likely to test near an edge, name ASTM D3244 or an equivalent agreed practice in the contract for combining a supplier result and a receiver result into an assigned test value, which turns a future stand-off into a calculation.

4. Do not accept the word equivalent in place of a grade name. The clause 35/50 or equivalent has no test attached to it and no agreed meaning, and both parties will read it in their own favour on the day it matters. If a substitution to 30/45 or 40/60 is genuinely acceptable to the engineer, have it approved in writing, by name, with the standard and the edition cited, before production. The wider cross-system picture is set out on the grade equivalence reference.

Which of the three a specification writer should actually name

Set the commercial question aside and the engineering distinction between the three is real, even though the bands overlap. 30/45 is the base and binder course grade: the deeper, hotter-running layers where stiffness is the whole point and the surface behaviour of the layer matters least. 35/50 is the grade for heavy traffic wearing and binder courses, where the layer has to resist rutting and also has to survive as a running surface. 40/60 is general highway paving in warm climates, the grade for a road that is warm rather than hot and busy rather than punished. Where the design traffic, the layer and the climate point clearly at one of those three descriptions, name that grade. Where they point between two of them, the overlap is not a problem to be solved — it is the standard saying that either name will be served by the same material, and that the choice can be made on availability and on which limit table you would rather be holding.

Cold climates and cold winters

This is the first and largest exclusion. The Fraass breaking point limit for 35/50 is a maximum of -5 °C, against -8 °C for 50/70, -10 °C for 70/100 and -12 °C for 100/150. Those figures are the standard’s own statement of what each grade is expected to tolerate before a thin film of it cracks. A binder qualified to crack at -5 °C is a warm-climate binder, and putting it into a pavement that sees hard frosts invites low-temperature transverse cracking — a failure mode that appears as regularly spaced cracks across the carriageway, admits water into the structure, and cannot be repaired by anything short of resurfacing. Where winters are genuinely cold, move down the ladder to 50/70, 70/100 or softer, or specify against measured low-temperature data rather than against a penetration grade at all.

Thin surfacings, surface dressing and chip seals

Spray-applied and thin-layer work needs a binder that will wet the aggregate, hold the chippings and tolerate the strains a very thin layer sees. 35/50 is too stiff for that duty. It does not wet chippings readily at practical spray temperatures, it gives up chippings under early trafficking, and in a thin layer it has no capacity to absorb strain. Surface dressing and thin surfacings are the territory of the softer grades and of bituminous emulsions, and no amount of care with spray rates compensates for starting with the wrong binder.

Emulsion and cutback feedstock

Cationic emulsions classified under EN 13808, and the cutbacks used for prime coats, are normally made from softer base binders. A 35/50 base produces an emulsion whose residual binder is stiff, slow to develop adhesion at ambient temperature and unforgiving in tack and prime applications where the film has to wet an existing surface rather than resist a wheel load. Where the job is a prime or tack coat, the grade question is answered by the emulsion specification, not by choosing the hardest available binder.

Thin pavements and weak or flexible foundations

Deformation resistance is only the governing risk when the structure is thick enough and the foundation stiff enough that rutting is what will actually happen. On a thin asphalt layer over a flexible or marginal foundation the governing distress is fatigue cracking from repeated bending, and a stiffer binder makes that worse rather than better: it carries more of the bending stress and tolerates less strain before it cracks. Lightly trafficked rural roads, thin overlays on old flexible pavements and surfacing over unbound or weak bases are all cases where a softer, more strain-tolerant binder outlasts a harder one, however counterintuitive that seems.

Hand-laid work, small patching and difficult laying conditions

35/50 needs higher mixing and compaction temperatures than the softer grades and gives a shorter workable window as the mat cools. In machine-laid work on a warm day with rollers immediately behind the paver, that is manageable. In hand-laid work, in small patches, in windy or cold conditions, at night, or where haul distances are long, it is a recipe for a mat that stops responding to the roller before target density is reached. Density lost at laying is not recoverable, and an under-compacted layer with a hard binder ages faster and cracks earlier than a well-compacted layer with a softer one.

Mixes with a high recycled asphalt content

Reclaimed asphalt brings aged, stiffened binder into the mix, and the blended binder is harder than the virgin binder that went in. Where the recycled content is significant, the usual answer is to select a softer virgin grade so that the blend lands where the design intends. Specifying 35/50 into a high-RAP mix compounds two sources of stiffness and produces a binder nobody designed, with low-temperature and fatigue behaviour nobody has measured.

Where the real problem is heavier than a penetration grade can solve

There is a limit to what hardening a straight-run binder achieves. Where the combination is extreme — very high pavement temperatures, very slow or stationary heavy loads, or a rutting problem that has already appeared on a road built with a hard grade — the answer is usually a polymer modified binder certified to EN 14023, or a high-modulus mix design, rather than another step down the penetration ladder. A harder unmodified grade improves high-temperature stiffness and degrades everything else. A modified binder is specified precisely because it separates those two outcomes.

Where the specification names a different grade

This one is procedural rather than technical, and it is the most common of all. Substituting 35/50 for a specified 40/60 or 50/70 because it looks stronger is a defect, not an upgrade. The specification writer chose a softening point band and a set of ageing and low-temperature limits, and a harder grade fails to deliver some of them — the Fraass limit above all, which becomes less demanding as the grade hardens. Any substitution needs the written approval of the engineer who will sign the pavement off, naming the substituted grade and its standard, obtained before shipment rather than after discharge.

Where the destination specification is not European

Two cases. Where the project is written to ASTM D946, 35/50 is not a grade of that standard and there is no limit table for a certificate to prove compliance against; the nearest named ASTM grade is 40-50, and the comparison table above sets out where the two differ. Where the project is written in performance grades, no penetration grade converts to a PG designation on paper: PG requires dynamic shear rheometer and bending beam rheometer results on that specific binder, and two 35/50 cargoes from different crude sources can carry different low-temperature grades. Ask for the test report, as the performance grading guide sets out, rather than for a conversion.

The general rule

Choose the softest binder that will resist the deformation the site will actually impose. That sentence is the whole of binder selection for unmodified paving grades. Deformation resistance is bought with stiffness, and stiffness is paid for with cracking resistance, fatigue life and buildability. 35/50 is the right answer when heavy slow traffic on a warm pavement is genuinely the governing risk. Where it is not, a harder grade is a cost with no benefit attached, and it usually presents its bill in the third or fourth winter.

Name the grade, the standard and the methods

The single most effective thing a buyer can do is name the methods rather than only the grade. A purchase order that says Bitumen 35/50 invites a data sheet built around ASTM methods with a European grade name at the top. A purchase order that says Bitumen 35/50 to EN 12591, batch Certificate of Analysis to report penetration by EN 1426, softening point by EN 1427, flash point by EN ISO 2592, solubility by EN 12592, and the EN 12607-1 RTFOT residue with change of mass, retained penetration by EN 1426 and softening point after hardening by EN 1427, together with the Fraass breaking point by EN 12593 leaves nothing to interpretation. Every item in that sentence is something an engineer at destination will look for, and every one of them has to be arranged before production, because none can be added retrospectively to a cargo that has already loaded.

Require the certificate to be reported against 35/50 specifically

This is the requirement peculiar to this grade, and it follows directly from the overlap. Because a batch can satisfy 30/45, 35/50 and 40/60 at once on penetration and softening point, the grade printed at the top of a certificate is not proof that the material was released against your grade’s limit table. Two sentences in the contract close the gap. First, that the Certificate of Analysis shall state the grade as 35/50 to EN 12591 and report every property against the requirements for that grade. Second, that the ageing residue results — retained penetration and softening point after hardening — shall be reported as measured values rather than as pass or fail, so that you can check them against the 53 % and 52 °C floors yourself. A certificate that reports only compliance statements cannot be checked against a different grade’s limits, and that is precisely the check you need to be able to make.

The inspection scope has to include the slow tests

A standard quality-and-quantity scope from an independent surveyor covers penetration, softening point and the quick lines. A full RTFOT with penetration and softening point run on the residue, and a Fraass determination, are extra laboratory work measured in days rather than hours. If they are not written into the inspection order and into the production plan, they will not appear on the report — and that, rather than any failure of the binder, is the usual reason an otherwise sound EN cargo arrives with an incomplete certificate. Sample drawing should follow a named practice, with sealed retained samples held by both parties. Size the retained quantity for the tests the argument might actually turn on: rechecking penetration and softening point consumes very little material, while re-running an RTFOT and then testing the residue consumes a great deal more. Agree the quantity in the contract rather than discovering at dispute stage that the sealed tin will not stretch to the two tests that matter. On drummed cargo, draw across the whole load instead of from whichever drum sits nearest the container door; the sampling procedure page sets out the practice.

Read the certificate for internal consistency

Penetration and softening point track each other for a straight-run binder of a given crude and process route. A sample reporting 45 ×0.1 mm with a softening point well above 58 °C is not a superior 35/50 — it is out of grade, and it suggests material that has been air-blown or blended rather than produced as a straight-run paving bitumen. Reading the two numbers together costs nothing and is the most useful authenticity check available to any penetration-grade buyer. A certificate whose values land exactly on the specification limit batch after batch is a template rather than a test report. The certificate of analysis guide covers what a genuine document looks like.

CE marking and the Declaration of Performance

Inside the European Union and the EEA, EN 12591 does double duty: it is the technical specification, and it is also the harmonised standard sitting behind CE marking and the Declaration of Performance under the Construction Products framework. That paperwork is generated at the producing plant through initial type testing and factory production control, and it travels with the product — a trading house cannot attach it to a cargo in transit, and it answers a different question from whether the binder meets the numbers in the table above. If your destination sits inside that area, establish before committing tonnage who issues the Declaration of Performance for the material you are buying and what obligations attach to the party whose name appears on the import. The harmonised standards list is revised periodically, so take the position from your importer, your customs broker or the relevant national authority rather than from any supplier’s assurance. Outside that regulatory area the picture is simpler: a great many markets buy to EN 12591 with no CE obligation at all, because the standard is being used purely as a technical specification, and what the engineer wants is a Certificate of Analysis reported against the EN methods.

Packing and container loading

Packing sets container count, the handling plant needed at discharge, waste disposal and how quickly the cargo can be put to work, so it moves landed cost more than most buyers expect. The figures are governed by the packaging rather than by the binder and are the same for any paving grade:

• New steel drums, 150 kg net: 80 drums in a 20-foot container, about 12 MT per FCL.
• New steel drums, 180 kg net: about 14.4 MT per 20-foot FCL.
• New steel drums, 185 kg net: about 14.8 MT per 20-foot FCL.
• Jumbo or meltable poly bags, 1 MT each: 20 bags, about 20 MT per 20-foot FCL.

Exact counts vary with drum dimensions, container type and destination axle weight limits. Where drums are used, write new into the contract — reconditioned drums are cheaper and are the single most common origin of contamination claims at discharge — and confirm that drum tare is excluded from the invoiced net weight. Jumbo bags suit a buyer with the plant to melt them and leave less packaging waste per tonne to dispose of, while drums suit a site with limited handling equipment. The packaging and logistics and shipping pages set out the options in full.

Origin, documents and terms

Material is offered from the Middle East. The commercial file for a shipment is the same as for any bitumen cargo — commercial invoice, packing list, bill of lading, Certificate of Origin from the chamber of commerce, Technical Data Sheet, Safety Data Sheet and the batch Certificate of Analysis, with an independent inspection report where the contract calls for one. Settle the Incoterm before discussing price, because FOB, CFR and CIF put the cost of freight, insurance and risk transfer in different places, and a headline figure quoted on one basis tells you nothing about a figure quoted on another. The Incoterms reference sets out which term carries which obligation. Send quantity, packing, destination port and Incoterm together with the specification you are working to, and state plainly whether the EN test suite is required, because that decision has to be taken before production rather than at the discharge port.