Bitumen PG 52-10: Specification, Soft-Binder Applications and Export Supply

Read the designation as two promises about a road surface rather than as a product code. The 52 states that under AASHTO M320 this binder satisfies the high-temperature criteria where the average seven-day maximum pavement design temperature reaches 52 °C. The −10 states that it satisfies the low-temperature criteria where the minimum pavement design temperature falls to −10 °C. Neither figure is an air temperature; both describe the asphalt itself, and both are earned on a rheometer rather than assigned by inspection, inferred from a penetration figure or copied across from a refinery data sheet.

Three test temperatures follow from those two numbers and the grade name prints none of them. Shear testing runs at 52 °C on binder in two conditions, as supplied and after rolling thin film oven ageing. The fatigue criterion runs on pressure ageing vessel residue at 25 °C, and the low-temperature criteria run on that same residue at 0 °C. All three come out of the grade name with a calculator, so a PG 52-10 report can be audited by whoever is holding it. The arithmetic is worked in full further down this page.

Where PG 52-10 sits on the grid

M320 lays out high-temperature grades in six-degree steps — 46, 52, 58, 64, 70, 76 and 82 — and low-temperature grades in the same steps from −10 down to −46. PG 52-10 sits one step above the bottom of the high ladder and at the very top of the low one. Its useful temperature interval is 52 − (−10) = 62 °C, the narrowest span you will meet in ordinary trade. There is no advantage in dressing that up: this is the least demanding combination the standard is routinely asked to certify, and understanding that is the whole basis for buying it properly.

A two-sided envelope, and what the designation rule does to it

The common misreading of a performance grade is to treat it as a minimum that any stiffer binder automatically clears. M320 does not work that way. The high-temperature criteria are a floor on stiffness: the binder must remain stiff and elastic enough at 52 °C to resist permanent deformation. The intermediate and low-temperature criteria are ceilings: G*·sin δ must not exceed 5000 kPa at 25 °C, creep stiffness must not exceed 300 MPa at 0 °C, and the m-value must not fall below 0.300. Those three cap how stiff and how unrelaxing the aged binder is allowed to become.

Under M320 a binder is designated by the highest grade it actually satisfies. Correctly designated, PG 52-10 describes material whose continuous high grade is at or above 52.0 °C and below 58.0 °C — in other words, a binder that cannot reach 58. That is not a defect certificate. It is a description of a genuinely soft product, and on most PG 52-10 projects the softness is the reason for the purchase.

This is where the grade differs commercially from every other page in this cluster. When a specification asks for PG 64-22 and a supplier offers PG 70-22, the buyer has been given margin. When a specification asks for PG 52-10 because the design needs a low-stiffness binder — to hold chippings on a sprayed seal, to blend down against stiff reclaimed binder, to buy fatigue life in a thin low-volume pavement — and a supplier ships a stiffer grade against it, every line of the M320 table still reads as a pass and the design has quietly been defeated. Nothing on the certificate objects, because the certificate was never asked the question. A PG 52-10 requirement therefore has to be written as an exact grade, not as a minimum, and it has to be verified by the continuous grade rather than by the label.

What the designation leaves out

A performance grade is a temperature envelope for the binder and nothing more. Crude source, wax content and asphaltene fraction are outside its scope. So is adhesion: M320 says nothing about how the binder will bond to a particular aggregate, and it contains no test for moisture damage or stripping — which in the wet, cool, marine climates this grade belongs to is a leading cause of early failure. It is not a mix design. Two cargoes correctly labelled PG 52-10 can come from different feedstocks and behave differently in workability, ageing and adhesion while both being properly graded.

The high-temperature tests: both at 52 °C, and 52 °C is not hot

The high-temperature grade is not derived from anything. It is the temperature the dynamic shear rheometer runs at. Two criteria apply there. On original binder, G*/sin δ must reach at least 1.00 kPa. On RTFOT residue — binder rolled in a glass bottle for 85 minutes at 163 °C in a moving air stream to AASHTO T240 — the same parameter must reach 2.20 kPa. Raising the bar on the aged sample is not the standard asking the same question twice with more severity. It concedes that the binder leaves the mixing plant stiffer than it arrived, and then insists the stiffened material go on resisting deformation through every summer that follows. In practice the aged line is the one that decides: material clearing 2.20 kPa after RTFOT has cleared 1.00 kPa unaged long before.

What deserves emphasis on this grade is how modest 52 °C is as a pavement temperature. A dark asphalt surface in direct summer sun sits well above the air temperature, and 52 °C at 20 mm depth is reached on days that most people would describe as pleasantly warm rather than hot. The whole commercial risk attached to PG 52-10 flows from that single fact, and it is dealt with in its own section below.

The intermediate temperature: why this grade is examined at 25 °C

Fatigue is checked at neither grade temperature. M320 sends it to a point four degrees above the mean of the two, on the reasoning that load-associated cracking is a mid-season distress belonging to spring and autumn rather than to a heatwave or a frost. Worked through for PG 52-10:

• (H + L) ÷ 2 + 4
• (52 + (−10)) ÷ 2 + 4
• 42 ÷ 2 + 4
• 21 + 4 = 25 °C

The ceiling applied there is 5000 kPa of G*·sin δ, measured on PAV residue. Two things about that expression trip buyers up. The modulus and the phase angle are multiplied here, where the rutting parameter divides them, and the limit is a maximum — a small number is the good result. The reasoning is that an aged binder which is stiff and elastic at once stores strain energy under each wheel pass instead of dissipating it as heat, and eventually returns that stored energy as a crack.

There is something specific to this grade worth noticing. Because the intermediate temperature is the mean of the two halves, it falls as the high grade falls. Across the whole −10 family the fatigue test runs at 25 °C for PG 52-10, 28 °C for PG 58-10, 31 °C for PG 64-10, 34 °C for PG 70-10 and 37 °C for PG 76-10. Bitumen is markedly softer at 37 °C than at 25 °C, so G*·sin δ shrinks as you climb the ladder and the ceiling becomes progressively easier to clear. PG 52-10 is therefore asked the most searching fatigue question of any grade on the −10 ladder — which matters less than it sounds, because a soft binder clears 5000 kPa at 25 °C comfortably anyway, but it does have one hard consequence for reading reports. A certificate offered as PG 52-10 that shows the fatigue parameter at 28 °C was graded as a PG 58-10, and one showing 22 °C was graded against a −16 low grade. Those are different tests on different assumptions, not rounding differences.

One coincidence causes recurring confusion in correspondence, particularly on this grade, because enquiries for PG 52-10 so often sit alongside enquiries for soft penetration grades. 25 °C is also the standard temperature for the penetration test (ASTM D5), for ductility and for specific gravity. The two have nothing to do with one another. Penetration is run on virgin binder and reports a depth in tenths of a millimetre; the M320 fatigue criterion is run on PAV residue — binder aged 20 hours at 2.1 MPa after already going through the RTFOT — and reports a modulus product in kilopascals. When a specification asks for “the 25 °C result” on a PG binder, ask for the parameter by name: G*·sin δ on PAV residue.

The low-temperature test: the beam is held at 0 °C, not at −10 °C

Every bending beam number on a PG 52-10 report is measured ten degrees above the low grade temperature — at 0 °C. That is the rule, not leniency, and not a laboratory that has misread your specification.

The instrument bends a small beam of PAV-aged binder at its midpoint and records two things at the sixty-second mark. Creep stiffness S may not exceed 300 MPa: the binder must not go so rigid that contraction on a cold night loads the mix beyond what it can carry. The m-value — how steeply that stiffness is still falling with time at sixty seconds, taken as the slope of log stiffness against log time — may not drop below 0.300: whatever stress does build, the binder has to be able to bleed it away.

The ten-degree offset is an application of time–temperature superposition. In a viscoelastic material, holding a load longer and testing it colder do the same thing to the stiffness you measure, and the exchange rate between the two can itself be measured as a shift factor. That matters because thermal cracking is not an instant event. A pavement sheds heat through the small hours, contraction stress accumulates across that whole period, and the binder spends the entire time trying to relax it. Honouring that at −10 °C would mean loading each beam for something like two hours — impossible as a routine acceptance test, and it pushes beam deflection down towards the resolution limit of the instrument. SHRP used superposition to move the measurement somewhere workable: sixty seconds of loading at L + 10 °C returns the same stiffness as roughly two hours of loading at L. So the beam sits at 0 °C and is read after a minute. The measurement is at 0 °C; the behaviour it certifies is the binder’s at −10 °C over a realistic overnight cooling.

Almost no supplier explains this, so it is worth stating without hedging: on a PG 52-10 report, BBR at 0 °C is the expected and correct entry. Seeing it should reassure you. A BBR result reported at −10 °C means the laboratory tested at L instead of at L + 10, which would be the correct condition for a −20 grade that does not exist on the M320 ladder. No BBR result at all means the binder has not been performance graded, whatever the offer, the tank label or the drum says.

Which criterion actually binds, and why almost none of them do

On a demanding grade, reading a report is a matter of finding the one line that decides acceptance. On PG 52-10 there is barely a line to find. The RTFOT DSR at 52 °C is nominally the binding requirement, and it is a low bar — most straight-run paving binders clear it without effort. The fatigue ceiling at 25 °C is cleared easily by soft material. Both bending beam criteria at 0 °C are cleared by an enormous margin, because 0 °C asks a paving binder to relax at a temperature at which it is still, by its own standards, warm.

Two failures do occur and they are worth knowing. Waxy and paraffinic residues can miss the m-value even at 0 °C, because crystallising wax interferes directly with the stress relaxation the test measures — and a low m-value is invisible in penetration, softening point or any conventional consistency test. Material that has been heavily hot-stored or repeatedly reheated can miss the 5000 kPa ceiling at 25 °C, because oxidation raises stiffness and elasticity together. If a report fails on this grade, it will almost always be on one of those two lines.

The practical conclusion is uncomfortable but useful. A clean PG 52-10 certificate is evidence of compliance and almost no evidence of anything else. The discrimination between offers has to come from the continuous grade at both ends and from the conventional consistency figures, which is dealt with in the sourcing section.

Pavement temperature is a radiation balance, not an air temperature reading, and that is the fact that makes PG 52-10 rarer than its modest numbers suggest. A dark surface absorbs solar energy all day and re-radiates it at night. The SHRP model behind the Superpave grade maps computes the seven-day maximum design temperature 20 mm below the surface from two inputs: the seven-day maximum air temperature and the latitude of the site. Latitude appears because the solar energy arriving on a horizontal surface depends on the angle of the sun. The low-temperature grade is handled separately and is taken at the surface, where it tracks the air minimum closely.

Working that model backwards gives a useful sense of scale. To land on a 52 °C design temperature you need a seven-day maximum air temperature of roughly 29 °C near latitude 20, and of roughly 35 °C near latitude 50 — the same pavement temperature needs hotter air the further you are from the equator, because the sun is lower. Those figures are indicative outputs of the model at 50 % reliability, offered to show the shape of the relationship rather than to be used for design; a road authority’s own model, its reliability level and any local calibration all move them, and reliability moves them upward.

Three ways a place actually gets to 52 °C

Cold-current fog coasts. The most striking case, because it produces cool pavements in genuine deserts. Where a cold ocean current upwells against a low-latitude coast — the Humboldt along Peru and northern Chile, the Benguela along Namibia, the Canary Current along Morocco and the Western Sahara, the California Current along the central Californian coast — persistent marine stratus and cool onshore air cut the solar energy reaching the surface, while the same ocean removes any winter. Lima is close to the textbook case for this grade: summer air maxima in the high twenties, winter minima that never approach freezing, and a coastal fog season that suppresses the surface temperature for months.

Cool temperate oceanic margins. Cloud, wind and short warm spells cap the summer while the ocean keeps the winter mild: western Ireland and Scotland, the Faroes and the Norwegian coast, coastal British Columbia and the Pacific Northwest, southern Chile from Valdivia to Chiloé, Southland and Fiordland in New Zealand, and western Tasmania. The summer half of PG 52-10 fits these places well. The winter half often does not — on the Atlantic and Pacific margins the record frequently justifies a −16 or a −22 low grade, and the low side is the one to verify from data rather than from reputation.

Tropical highlands with a permanent spring. Elevation caps the maximum while latitude prevents a winter, giving places where the daily temperature range exceeds the annual one: Bogotá, Quito, Addis Ababa, Asmara, the Kenyan and Rwandan highlands, mid-elevation Andean corridors. Here the summer half is comfortably satisfied, and again the winter half is the one that needs checking — thin dry air under clear skies radiates heat away quickly after sunset, and many of these sites compute a −16 rather than a −10 low grade.

When the model returns less than 52, and the PG 46 problem

For the coolest of these sites the calculation returns a design temperature below 52 °C — the low forties is entirely possible for a high-altitude equatorial city or a fog-bound coast. M320 does carry a high-temperature step below 52: PG 46. But the standard’s own grade table pairs PG 46 only with the coldest low grades — −34, −40 and −46 — because a 46 °C design summer is computed almost exclusively in very cold continental interiors, where it is bought to hold an enormous span. PG 46-10 is not a listed combination in M320 and is not a product anyone makes. PG 52-10 therefore functions as the practical floor of the trade, and a site computing 46 or 49 °C buys 52 and accepts binder slightly stiffer than its climate requires. This is worth knowing before a specification is written around the assumption that something softer can be sourced.

The seven-day maximum is not the average, and this is where cool coasts get caught

The grade is driven by the hottest seven-day run in the record, not by the climate a place is known for. Several of the regions above have exactly the wrong kind of exception. Berg winds bring hot dry air off the interior onto the Namibian coast; föhn and offshore flows do the same on temperate maritime margins; a few days of desert air can arrive on the Moroccan or Californian coast in any year. A pavement that spends 355 days under marine cloud and ten days under offshore desert air is graded by the ten days. Add to that the local effects the regional maps cannot see — urban heat islands, thin dark surfacing on unshaded sun-facing gradients, sheltered urban canyons — and the case for taking the design temperature from the site’s own record rather than from a regional generalisation becomes concrete rather than procedural.

62 °C implies no polymer modification whatsoever

As a working rule across the industry, a useful temperature interval of up to roughly 92 °C is achievable with an unmodified straight-run binder from a suitable crude. Beyond about 92 °C, polymer modification is normally required to hold both ends at once. At 62 °C, PG 52-10 is not merely inside neat capability — it is further inside it than any other grade you are likely to buy. A modified binder offered against this grade under M320 is a question, not an upgrade. Either the specification called for modification for a reason that is not visible in the grade name, or a product the project never asked for has been substituted into the offer. The same arithmetic runs across almost the whole PG 52 family: even PG 52-40, at a 92 °C span, sits at the edge rather than beyond it, and only PG 52-46 at 98 °C ordinarily crosses into modified territory.

Traffic removes this grade rather than modifying it — except under M332

Climate is only half of the high-temperature decision under Superpave practice; loading is the other half. A slow transient load moves the high grade up one step of six degrees, a standing load moves it two, and a further step is sometimes considered where traffic volumes are very high. Applied here, a one-step bump gives PG 58-10 at a 68 °C span and a two-step bump gives PG 64-10 at 74 °C. Both are still comfortably neat binders. The practical consequence is that on this grade the traffic adjustment does not make the binder more complicated — it takes you off the grade altogether. PG 52-10 is defensible only where the climate is genuinely cool and the traffic is light and moving. Any part of a project with crawling or standing loads has already left this grade behind.

Where a specification is written to AASHTO M332 instead, the grade carries a traffic letter — PG 52S-10, PG 52H-10, PG 52V-10, PG 52E-10 — and the G*/sin δ requirement on RTFOT residue is replaced by a multiple stress creep recovery test to AASHTO T350, reporting non-recoverable creep compliance Jnr and percent recovery at the high grade temperature. The Jnr limits at 3.2 kPa are max 4.5 kPa⁻¹ for S, 2.0 kPa⁻¹ for H, 1.0 kPa⁻¹ for V and 0.5 kPa⁻¹ for E, with a lower value meaning greater rutting resistance, and M332 also caps the difference in Jnr between the 0.1 kPa and 3.2 kPa stress levels at 75 %. Because M332 handles traffic by tightening Jnr at 52 °C rather than by raising the test temperature, the grade name stops moving — but the modification question returns through the back door. A soft unmodified binder tested at 52 °C produces a high Jnr and very little recovery, so the V and E letters normally require elastomeric modification even though the span does not. PG 52S-10 is the direct counterpart of M320 PG 52-10. Establish which standard governs before ordering; an M320 certificate contains no Jnr values and they cannot be derived from it.

The continuous grade — and here you are checking that it is not too high

On a demanding grade you ask for the continuous grade to see how much margin you have. On this one you ask for the opposite reason. A binder correctly designated PG 52-10 has a continuous high grade of at least 52.0 °C and below 58.0 °C, and a continuous low grade at or below −10.0 °C. That first band is six degrees wide, and in a grade bought for softness the difference between a binder that continuous-grades at 52.6 and one that grades at 57.4 is the difference between the product the design assumed and something noticeably stiffer.

There is a further practical point. Most commercially available paving bitumen continuous-grades well above 52 on the high side, so material that genuinely grades at 52 is a deliberately soft production stream rather than a default. A cargo labelled PG 52-10 that continuous-grades at 63 is compliant with every line of the M320 table and wrong for a sprayed seal or a high-RAP blend. Ask for the continuous grade at both ends. It belongs on the test report, it costs the laboratory nothing extra to state, and on this grade it is the only number that distinguishes two offers which look identical on a specification sheet.

“Meets or exceeds PG 52-10” is not a compliant statement here

It fails on two separate grounds. The first is the design argument already made: where the grade was chosen for low stiffness, a stiffer binder defeats the purpose while passing the paperwork. The second is formal, and it is the one to put in writing. Under M320 the intermediate criterion is grade-specific. A binder graded PG 64-10 demonstrated G*·sin δ below 5000 kPa at 31 °C, not at 25 °C; a PG 58-10 demonstrated it at 28 °C. Neither has shown that it clears the ceiling at this grade’s intermediate temperature. It may well do so — it has not been tested to say so. Grade envelopes overlap in some directions and not in others, and this is one of the directions in which they do not.

The clean request is not for a conformity statement against a grade the binder may never have been tested to. It is for a grading report on the tank or batch being loaded, stating the actual DSR and BBR results at 52 °C, 25 °C and 0 °C, together with the continuous grade at both ends.

A penetration grade cannot be converted into a PG grade

No table, factor or rule of thumb turns 120/150 or 200/300 into PG 52-10. A penetration figure is one needle, one temperature, one unaged sample. It reports nothing about how the binder behaves in shear at 52 °C, nothing about how it relaxes on a beam at 0 °C, and nothing whatever about what twenty hours in a pressure ageing vessel will do to it. What does exist is a tendency — soft penetration grades land on a 52 °C high grade far more often than 60/70 does, which is why the two enquiries so often arrive in the same email — and a tendency is not a conversion. Put two cargoes of 200/300 from different crude sources on a rheometer and their continuous high grades can be several degrees apart, while the low end cannot be guessed at from the penetration figure at all.

It follows that a grading report attaches to a particular quantity of binder, not to a product name. Last quarter’s report describes last quarter’s tank and is evidence about material you are not buying. What makes a PG specification enforceable is a DSR and BBR report carrying the same tank or batch reference, sampling date and test date as the Certificate of Analysis for your own cargo. Where that report does not exist, the grade on the offer is a claim rather than a measurement and should be treated as one during price comparison. It is worth settling at the enquiry stage whether a grading report on the actual batch can be produced, because this export trade is quoted overwhelmingly in penetration grades and a PG designation is normally the result of grading a specific production batch.

Handling a soft binder: heat it less, and watch it in transit

The 3 Pa·s limit at 135 °C says nothing about how the road will perform; it exists so the binder can be pumped through a terminal and will wrap itself around aggregate at a plant. The temperatures you actually mix and compact at come from the binder’s own viscosity–temperature chart, read at the conventional equiviscous targets of about 0.17 Pa·s for mixing and 0.28 Pa·s for compaction. A PG 52 binder reaches those viscosities at markedly lower temperatures than a stiffer binder from the same source, so carrying a 60/70 plant recipe across unchanged means heating it harder than it needs, ageing it in the mixer for no benefit and giving away part of the softness that was bought. Take the figures from viscosity data for the binder actually supplied. The equiviscous approach is not valid for polymer modified binders, so if an M332 traffic letter moves the order to a modified product, the supplier’s recommended range replaces the calculation.

Two points are specific to soft grades and rarely appear on a data sheet. First, every reheating cycle stiffens the binder, and on a grade bought for softness that is a direct loss of the property being purchased — repeated draining and reheating of drums is worth avoiding here more than on any other grade. Second, a soft binder is more sensitive to ambient heat in transit and storage than a 60/70. Cargo that transits or transships through a hot climate, or sits in a container standing in the sun at a port, can soften enough to deform drums and consolidate under stack load, even though the material itself is unharmed once remelted. Keep drums out of direct sun, limit stack heights, and consider bulk, tank container or jumbo bag packing where the destination has the facilities, since it removes the problem entirely.

Checking the certificate before shipment

• Every DSR and BBR result has its test temperature printed beside it. A value without a temperature cannot be checked against any grade.
• The high-temperature DSR results are at 52 °C. Results at 58 °C belong to a PG 58 grade.
• Ageing before the second DSR is RTFOT to AASHTO T240 / ASTM D2872, not TFOT to ASTM D1754. TFOT is the penetration-grade procedure; M320 criteria are written on RTFOT residue.
• The RTFOT residue limit is 2.20 kPa, not the 1.00 kPa unaged figure applied twice.
• PAV conditioning to AASHTO R28 / ASTM D6521 is present, at 20 h and 2.1 MPa. Fatigue and BBR results quoted on RTFOT residue alone are not M320 results.
• The fatigue result is at 25 °C. At 28 °C the sample was graded as a PG 58-10; at 22 °C it was graded against a −16 low grade.
• The BBR is at 0 °C, not at −10 °C, and the m-value is reported and not only creep stiffness. On waxy sources the m-value is the line that fails even at 0 °C.
• Viscosity is in Pa·s by AASHTO T316 / ASTM D4402, not centistokes by a kinematic method — a different test, not convertible to the M320 limit.
• Flash point is Cleveland open cup, ASTM D92. Tag open cup is the cutback method and its appearance on a paving binder certificate is a safety-case error.
• The continuous grade at both ends is stated, and the report carries a tank or batch number, a sampling date and a test date that tie it to your cargo.
• Penetration, softening point and specific gravity are supplied alongside the PG report. They are on every refinery certificate in this trade already, they cost nothing, and on a grade where four of the five criteria are close to formalities they tell you what kind of binder is behind the pass.