Bitumen PG 70-10: Specification, Hot-Climate Fit and Export Supply

PG 70-10 means the binder has been shown by test to satisfy the high-temperature criteria of AASHTO M320 at an average seven-day maximum pavement design temperature of 70 °C, and the low-temperature criteria at a minimum pavement design temperature of −10 °C. Neither number is a description of the material and neither can be arrived at by judgement, by reading a penetration figure or by copying a refinery data sheet forward: they are the settings on the instruments, and a laboratory either got a result at those settings or it did not. The low-temperature half is read on binder that has already been through two artificial ageing steps, so it describes the binder several years into the road rather than the binder in the tank.

Everything else that appears on a PG 70-10 test report follows arithmetically from those two numbers, and none of it is written in the grade name. The dynamic shear rheometer runs at 70 °C, once on original binder and once on rolling thin film oven residue. The fatigue criterion runs on pressure ageing vessel residue at 34 °C. The bending beam rheometer runs on that same PAV residue at 0 °C. All three are recoverable from the grade name with a calculator, which means the test conditions on a certificate can be audited by anyone, without laboratory access. The arithmetic is set out in full further down this page.

−10 °C is the floor, not a choice

The low-temperature side of M320 stops at −10 and then steps downward in sixes: −10, −16, −22, −28, −34, −40, −46. −10 is the warmest step the standard contains. There is no PG 70-4 and no PG 70+2. That single fact changes how the second half of this grade name should be read. Nobody selects −10 because their site is expected to reach exactly −10 °C; they select it because the site does not come close to any of the colder steps and −10 is the mildest requirement that can be written down.

The practical consequence is that on this grade the low-temperature criteria stop being a climate requirement and become something else — a manufacturing check. They are still worth reading, and reading first in some cases, but for a reason that has nothing to do with winter. That reason is explained in the section on test temperatures below, and it is the single most under-used piece of information on a PG 70-10 certificate.

The 70 is where the grade is won or lost

Two criteria apply at 70 °C. On original binder, G*/sin δ must be at least 1.00 kPa. On RTFOT residue — binder rolled for 85 minutes at 163 °C under continuous air flow to AASHTO T240 — the same parameter must be at least 2.20 kPa. The higher limit on the aged residue is not a stricter version of the same question; it recognises that the binder stiffens on the way through the plant and requires the stiffened material to keep resisting permanent deformation through the summers that follow.

Of every line on the certificate, the RTFOT DSR result at 70 °C is the one that decides this grade. The fatigue criterion at 34 °C is comparatively easy to clear because the binder is soft at 34 °C. The bending beam criteria at 0 °C are cleared with a wide margin by almost any sound paving binder. Nothing else is close. When a batch fails PG 70-10, it fails at 70 °C on the residue, and when a supplier says a grade is difficult to make, this is the line they mean.

What the grade name deliberately hides

The two numbers describe a temperature window the binder has to survive, and they stop there. Crude source, asphaltene fraction and wax content are outside the scope; so is the mix, so is adhesion to a given aggregate, and so is moisture damage. On this grade the omission that costs buyers money is none of those: it does not say how the 70 °C was reached. A hard, high-asphaltene straight-run binder and a polymer modified binder built on a much softer base can both satisfy G*/sin δ at 70 °C, both are correctly labelled PG 70-10, and they are not the same product — in price, in storage behaviour, in mixing temperature, or in how they respond to a bus standing on them. Establishing which one is on offer is the central buying task for this grade, and M320 will not do it for you.

The high-temperature tests: both at 70 °C

The high-temperature grade is not derived from anything. The 70 is the dial setting on the dynamic shear rheometer, and both high-temperature criteria are read there. Original binder must reach G*/sin δ of at least 1.00 kPa at 70 °C; RTFOT residue must reach at least 2.20 kPa at the same 70 °C. A binder that clears 2.20 kPa aged has cleared 1.00 kPa unaged with room to spare, so in practice the residue result is the only one that binds.

The intermediate temperature: 34 °C for this grade

M320 places the fatigue criterion at the mean of the two grade temperatures plus four degrees. For PG 70-10:

• (H + L) ÷ 2 + 4
• (70 + (−10)) ÷ 2 + 4
• 60 ÷ 2 + 4
• 30 + 4 = 34 °C

At 34 °C the PAV residue must show G*·sin δ of no more than 5000 kPa. The parameter is a product rather than a quotient, and a low value is wanted: an aged binder that is simultaneously stiff and elastic stores strain energy under each wheel pass instead of dissipating it, and eventually gives that energy back as a crack.

34 °C is a warm place to ask a fatigue question. Anything with a colder low half is tested colder — PG 70-16 at 31 °C, PG 64-16 at 28 °C, PG 64-22 at 25 °C — and the only grades tested warmer are the hotter ones above this in the same climatic band, PG 76-10 at 37 °C and PG 82-10 at 40 °C. Bitumen is far softer at 34 °C than at 25 °C, so the 5000 kPa ceiling is generous here and the fatigue line is almost never the one that fails a PG 70-10 batch. That is not a reason to ignore it. It is the only mid-temperature check on the certificate, and on a grade that can legitimately be produced by making a binder very hard, it is the line that would register excessive hardness if the hardness were extreme. Read it as an over-stiffness screen rather than as a fatigue design number, and treat a result sitting close to 5000 kPa at 34 °C as a question worth asking rather than a pass to be filed.

One error pattern is specific to this grade. If a report offered as PG 70-10 shows the fatigue parameter measured at 31 °C, the laboratory graded that sample as PG 64-10. If it shows 37 °C, it graded it as PG 76-10. Neither is a rounding difference; each is a different test on a different grade.

The low-temperature test: 0 °C for a −10 grade

The bending beam rheometer for PG 70-10 runs at 0 °C, ten degrees warmer than the −10 °C in the grade name. This is correct, it is deliberate, and it is not the specification going easy on a hot-climate binder.

The BBR loads a small beam of PAV-aged binder in three-point bending and reads two values at the sixty-second mark. Creep stiffness S must not exceed 300 MPa: the binder must not be so rigid that thermal contraction generates more stress than the surrounding mix can carry. The m-value — the slope of the log stiffness against log time curve at that same sixty seconds — must be at least 0.300: whatever stress the binder does build, it must still be able to shed.

The ten-degree offset comes from time–temperature superposition. Bitumen is viscoelastic, so its apparent stiffness depends on how long a load is held as much as on how cold it is; cold and slow are interchangeable along a shift factor that can be measured. Thermal cracking is a slow event — a pavement cools through the small hours while stress builds and the binder tries to relax it — so reproducing it faithfully at −10 °C would require roughly a two-hour loading time per specimen. No acceptance laboratory can run a two-hour test on every beam, and at that duration the deflection has also run past the range the instrument reads reliably. So the Superpave developers traded ten degrees of temperature for the two hours of clock: sixty seconds of loading at L + 10 °C gives the same stiffness as approximately two hours of loading at L. The beam is therefore held at 0 °C and read at 60 seconds. The criterion is applied at 0 °C; the property it describes is the binder’s behaviour at −10 °C over a realistic cooling period.

Why the BBR still matters on a grade with no winter

The obvious objection to running any low-temperature test for a Gulf or Sahel project is that the pavement will never see 0 °C, let alone −10 °C. That objection is correct about the climate and wrong about the test.

At 0 °C, essentially every sound paving binder passes creep stiffness with an enormous margin — 300 MPa is a long way from where a normal binder sits at that temperature — and passes the m-value comfortably as well. So the test is not discriminating on climate grounds. What it discriminates on is how the binder was hardened. Stress relaxation is the first property that oxidation destroys. A binder that has been pushed to a 70 °C high grade by air blowing, by heavy processing or by prolonged hot storage loses m-value long before it loses stiffness, and it loses it at every temperature, including 0 °C. A material that cannot make 0.300 at 0 °C after PAV is not a binder with a marginal winter performance; it is a binder that has been oxidised into its grade.

That makes the m-value at 0 °C the cheapest available screen on production route, on a grade where production route is the main commercial question. It also makes one particular omission serious: a certificate that leaves the BBR out on the grounds that the destination has no winter has removed the only line that would have caught an over-blown product. Ask for the BBR even though the climate does not need it, and read the m-value before the stiffness.

Two further error patterns are worth naming. If BBR is reported at −10 °C, the laboratory tested at L rather than at L + 10 — the condition that would suit a −20 low grade, which M320 does not contain — and the numbers cannot be measured against the M320 limits at all. If there is no PAV step and the fatigue or BBR figures are quoted on RTFOT residue, they are not M320 results at all.

A 70 °C pavement design temperature does not require 70 °C air. It requires a dark, low-albedo surface under intense solar radiation for long summer days, with a thick asphalt layer that stores heat and releases it slowly. Air temperatures in the mid-forties, sustained across a week, will do it; where air passes 50 °C — as it does in Kuwait, southern Iraq, parts of the Saudi Eastern Province and the lower Indus plain — the pavement figure is reached without argument. The M320 high temperature is the average of the seven hottest consecutive days measured near the top of the layer, so it is not a freak reading. It is a condition the surface returns to every summer.

Where the low side genuinely does not matter

The low-temperature grade is governed almost entirely by two things: distance from a large body of water and elevation. PG 70-10 belongs to sites where both work in the binder’s favour — low-lying, coastal or near-coastal, and often humid at night, so that the surface has neither the altitude nor the clear dry sky it would need to radiate down towards freezing.

• The Arabian Gulf littoral. Kuwait, Bahrain, Qatar, the eastern province of Saudi Arabia, the UAE coastal strip from Abu Dhabi through Dubai and Sharjah, and the Omani Batinah coast. Extreme summer surface temperatures with humid nights that hold the minimum far above any low-temperature grade step.
• Southern Iraq and the head of the Gulf. The Basra region and the lower Mesopotamian plain, which combine some of the highest air temperatures recorded anywhere with a low, flat, near-sea-level setting.
• The Red Sea coast. The Saudi Tihama and Jeddah, the Sudanese and Eritrean coast, Djibouti and the coastal plain of Yemen. Hot, humid and effectively frost-free.
• The Sahel and the southern Sahara fringe. Northern Nigeria, Niger, Chad, Mali and central Sudan. Very high insolation, and winter nights that cool sharply in the dry season but from a much higher daytime base than an elevated plateau.
• The lower Indus plain and western India. Sindh and southern Punjab in Pakistan, Rajasthan and Gujarat. Some of the hottest sustained summer conditions in the region, with winters that are mild at low elevation.

The boundary of this grade is worth stating as plainly as its centre. Move inland and upward and the high grade often stays at 70 while the low grade drops to −16. The Saudi interior, the Jordanian plateau and the Anatolian southeast can produce the same summer pavement temperature and still record sub-zero winter nights, and there the honest grade is PG 70-16 with its bending beam test at −6 °C. The high grade is set by insolation, latitude and surface; the low grade is set by elevation and the sea. The two halves of a PG grade come from different physics, and a site can be extreme in one and unremarkable in the other.

What failure looks like in this climate

Where the low-temperature side is a formality, essentially all binder-attributable distress is rutting: shear flow in the upper hundred millimetres of the pavement, showing as a depression in each wheel path with a shoulder of heaved material beside it. Densification under traffic contributes early in the life of the layer, but the deep ruts that force an intervention are shear, and shear in asphalt is a viscous, time-dependent process. The hotter the binder, the less resistance it offers, and the longer the load sits, the further the material moves.

The secondary consequences are what actually close a road. A rut holds water, water in the wheel path causes hydroplaning and loss of skid resistance, and standing water eventually finds its way into the structure through the cracks that open along the rut shoulders. In a climate where the surface can be above 60 °C for weeks at a time, an under-graded binder does not fail gradually over a decade; it can produce a measurable rut in a single summer on a heavily loaded approach.

Grade bumping: PG 70-10 is very often already the bumped grade

Superpave practice adjusts the high-temperature grade for loading as well as for climate. The established guidance is one step — six degrees — for slow transient loads, two steps for standing loads, and a further step considered for very high traffic volumes. The low-temperature grade is never adjusted for traffic, so a bump moves only the top number and widens the span.

Applied to this climatic band, the arithmetic runs like this:

• A coastal site whose temperature data produces PG 64-10, carrying slow or heavy traffic, is bumped one step to PG 70-10. A large share of all PG 70-10 sold is exactly this: not a climate grade but a traffic grade sitting on top of a PG 64-10 climate.
• A site whose data genuinely produces PG 70-10, with slow transient loading, is bumped one step to PG 76-10 — a span of 86 °C.
• The same site with standing loads is bumped two steps to PG 82-10, a span of 92 °C, which is at the outer edge of what any binder does without polymer.

The reason standing loads justify two steps rather than one is loading time, not load magnitude. Rutting accumulates with the duration a stress is applied, and a stationary or crawling wheel applies its load for orders of magnitude longer per pass than a wheel at highway speed. A bus bay served by a few dozen buses a day can accumulate more loaded seconds per square metre in a month than a motorway lane does with thousands of moving trucks. At signalised intersections the problem compounds, because braking and acceleration add a horizontal shear component on top of the vertical load, applied repeatedly at exactly the same spot in the approach lane. That is why intersections, bus stops, roundabout entries, toll plazas and weighbridge approaches rut while the open highway between them stays flat, and why they are frequently designed and tendered with a different binder grade from the corridor they belong to.

Two practical points follow. First, settle the traffic adjustment before the grade goes into the contract, because a bump changes the product being bought and the price basis with it. Second, where a project is written to AASHTO M332 rather than M320, traffic is handled differently: the grade keeps its 70 °C designation and gains a letter — PG 70S-10, 70H-10, 70V-10, 70E-10 — with the RTFOT DSR line replaced by a multiple stress creep recovery test to AASHTO T350 reporting non-recoverable creep compliance and percent recovery at 70 °C. Those values are not present on an M320 certificate and cannot be derived from one, so establish which standard governs before ordering.

The 80 °C span implies no modification — but it is not the whole test

The span is the standard first test of whether a grade needs a polymer. Two grade temperatures far apart demand a binder that is stiff in summer and still relaxable in winter, and those requirements pull against each other; the industry working figure is that an unmodified straight-run binder from a suitable crude will hold a useful temperature interval of up to roughly 92 °C, and that past that point an elastomer is normally needed to keep both ends at once. PG 70-10 spans 80 °C, two full grade steps inside that boundary. Nothing about this grade’s span implies polymer modification, and it should not be priced as though it does.

That reading is right and it is only half the picture, because the span measures the gap and says nothing about the height. A binder can have all the range in the world and still not be stiff enough at the top. Below about 64 °C the height is rarely the problem and the span really is the whole answer. By 70 °C it is not: a great many streams that would span 80 degrees without trouble still cannot put 2.20 kPa on the DSR at 70 °C after RTFOT, and when the vacuum tower will not deliver it, the producer has to intervene. There are two ways to intervene.

Route one: base stiffness

The first route is to start from a harder, more asphaltic residue — a stiffer vacuum bottom from a suitable crude. In penetration terms a neat binder that grades PG 70 tends to be a hard one, closer to a 40/50 or harder than to a 60/70, because without a polymer the only way to hold 70 °C is to be stiff at every temperature.

A neat PG 70-10 has genuine advantages. It is homogeneous, so there is no separation mechanism and no agitation requirement in storage. Mixing and compaction temperatures can be read from its own viscosity–temperature chart at the conventional equiviscous targets of roughly 0.17 Pa·s for mixing and 0.28 Pa·s for compaction. It is priced as a straight-run product. Its limitations follow from the same stiffness: it resists rutting by being rigid rather than by being elastic, so deformation that does occur is not recovered; it needs higher plant temperatures than a softer binder, which costs it a little more ageing on the way through; and its margin on the m-value at 0 °C is narrower than a softer binder’s, because it starts closer to the limit.

A variant of this route deserves a warning. A high grade can also be reached by partial air blowing or semi-blowing — oxidising a softer residue to raise its softening point until it clears 70 °C. It works, it is cheap, and it is the reason the BBR m-value at 0 °C is worth insisting on for this grade. Blowing raises stiffness and destroys relaxation capacity at the same time. A product that reaches its high grade this way tends to show a poor m-value after PAV, a fatigue result at 34 °C that is closer to the 5000 kPa ceiling than it should be, and a tendency to embrittle in service in a way that no amount of hot weather will disguise.

Route two: polymer modification

The second route is to disperse an elastomer, usually a styrene-butadiene-styrene block copolymer, into a conventional and much softer base binder, sometimes with a crosslinker. At a few percent by weight the polymer forms a continuous network through the binder phase. That network carries load at high temperature, which lifts the high grade sharply, while the base binder underneath stays soft and keeps its relaxation behaviour intact.

The practical differences from a neat binder are substantial. A polymer modified PG 70-10 recovers a meaningful part of each deformation instead of only resisting it, which is precisely what matters where wheels stand still. It normally holds more low-temperature margin, because the base was never hardened. Against that, it costs more; it can separate if it is held hot and static for long periods, so it needs agitation or circulation and a defined storage regime; its viscosity behaviour is non-Newtonian, so the equiviscous method for setting mixing and compaction temperatures is not valid and the supplier’s recommended range replaces the calculation; and it is less tolerant of repeated reheating.

Why M320 cannot tell the two apart, and what can

Here is the crux. The high-temperature criterion under M320 is G*/sin δ. That parameter rewards stiffness and elasticity together, and a binder that is simply very stiff can satisfy it with almost no elastic character at all. The parameter was never designed to distinguish an elastic binder from a merely rigid one, and under a standing load — where recovery between load applications is the whole point — the distinction is exactly what the pavement cares about. This limitation is one of the reasons AASHTO M332 and the multiple stress creep recovery test exist: non-recoverable creep compliance and percent recovery at 70 °C separate the two routes immediately, where G*/sin δ does not.

So if the project has standing or slow traffic, do not stop at an M320 certificate. Ask for MSCR data to AASHTO T350 at 70 °C, and ask for elastic recovery to ASTM D6084 or EN 13398. If the answer to either is unavailable, that is itself informative.

Ask for the continuous grade

The M320 envelope is six degrees wide at each end, so passing tells you less than it appears to. Two binders can both be genuine PG 70-10 and sit in very different places inside it: one continuous-grading at 75.4 on the high side, the other at 70.3. Both are correctly labelled. The first is very nearly a PG 76 and will tolerate a hot spell at the plant, a stiff RAP fraction or an unusually severe summer without falling out of grade; the second has three-tenths of a degree in hand. On this grade the high-side continuous grade is the number worth asking for, because the high side is where the grade is actually earned. It belongs on the test report and it is the first thing to request when two offers look identical on the specification sheet.

A penetration grade cannot be converted into a PG grade

There is no arithmetic that turns 40/50, or 60/70, or any other penetration grade into PG 70-10. Penetration at 25 °C carries no information about behaviour at 70 °C, none about behaviour at 0 °C, and none at all about the binder after PAV ageing. A hard penetration grade is a plausible candidate for a 70 °C high grade and nothing more: two 40/50 cargoes from different sources can continuous-grade several degrees apart at the top, and one may reach 70 °C while the other stops short. The relationship runs one way and it runs weakly.

The consequence is that a PG grading report is batch-specific and does not transfer. A grading report from an earlier campaign describes that campaign, not the tank now being loaded for you, and on a grade where the high side may sit less than a degree inside the step, that distinction is the whole margin. Where a specification is written in PG, ask for the DSR and BBR results tied to the same tank or batch number as the Certificate of Analysis for your cargo. A grade claimed without a report tied to your tank is a claim nobody has tested.

• PG 76-10 — the next high-temperature step, and the point at which modification usually enters
• PG 76-16 — the same hot end with a much colder low grade