See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/329209717

INFLUENCE OF TEMPERATURE ON ASPHALT STIFFNESS MODULUS

Conference Paper · June 2012

CITATIONS READS

0 74

3 authors, including:

Tatjana Rukavina Josipa Domitrovic

Faculty of Civil Engineering, University of Zagreb University of Zagreb
49 PUBLICATIONS   143 CITATIONS    13 PUBLICATIONS   18 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Pula airport runway rehabilitation project. View project

CETRA 2018 - 5th International Conference on Road and Rail Infrastructure View project

All content following this page was uploaded by Josipa Domitrovic on 27 November 2018.

The user has requested enhancement of the downloaded file.

 P5EE-143
INFLUENCE OF TEMPERATURE ON ASPHALT STIFFNESS MODULUS
Maja Halle1, Tatjana Rukavina2, Josipa Domitrovic2
1
Institut IGH, d.d., Zagreb, Croatia
2
Faculty of Civil Engineering, University of Zagreb, Zagreb, Croatia

ABSTRACT

It is well known that temperature has a great influence on performance of asphalt layers. Low temperatures raise the
asphalt stiffness modulus which finally results in low temperature cracking. On the other hand, the increase of the
temperature softens the asphalt binder and the stiffness modulus significantly drops. This, in combination with traffic
load, causes asphalt pavement distresses like rutting, corrugation and shoving.
In order to ensure reliability in asphalt pavement design, values of the design variables should, as much as possible,
match the actual pavement performance conditions. As the asphalt stiffness modulus is one of the input parameters, it is
important to know its values for temperatures that pavement will be subjected to during its life time. The temperature
range in Croatia usually varies between -10°C and 40°C.

According to the test method prescribed by European standard EN 12697-26 (Stiffness), Annex C (Test applying
indirect tension to cylindrical specimens, IT-CY), stiffness was tested at four different temperatures, on several types of
hot mixed asphalt mixtures. Based on the test results, temperature-stiffness correlations were obtained for the asphalt
mixtures typically used in Croatia, for bearing or wearing courses.

The application of the obtained stiffness modulus, as the parameter that describes the performance of asphalt layers
under different temperature conditions, will enable the optimization of the existing pavement design method in Croatia.

Keywords: asphalt mixture, stiffness modulus, temperature susceptibility, pavement design

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

1. INTRODUCTION

In Croatia, pavement design is performed according to HRN U.C4.012, after which critical stresses and strains are
verified by the mechanistic-empirical (analytically-based) analysis (by application of the BISAR computer program).
The verification is performed based on outdated and inaccurate data about material properties obtained by the elasticity
modulus tests, performed 40 years ago (Dormon and Edwards, 1967) 1. In Croatia no major changes have been
introduced in this field for a longer number of years, and with the traffic volume and the permissible axle load increased,
"well designed” pavements started to degrade sooner than expected.

Since the asphalt mixtures stiffness is the necessary parameter to assess stresses and strains in the pavement structure,
the goal of this research was to determine the stiffness modulus for several types of hot asphalt mixtures which are
presently used for construction of roads in Croatia, at the temperatures to which pavement will be subjected to during its
life time.

On the basis of the obtained results, the input parameters used in pavement design will be closer to the actual ones. .
This will contribute to reduction of eventual deviations between the stress and strain obtained by calculations and those
that occur in pavement layers as response to traffic load.

2. LABORATORY TESTING OF STIFFNESS MODULUS

In accordance with the standard HRN EN 12697-26 2, the asphalt mixtures stiffness modulus can be measured in
several ways. Because of the simplicity and quickness of measuring, as well as preparation of specimens (with respect to
other methods of measuring from the standard), and because of the availability of the measuring equipment, indirect
tensile test was selected (Annex C to the mentioned standard). 848 measurements of the stiffness modulus were carried
out, at four different temperatures and on 106 specimens prepared from seven asphalt mixtures. In the continuation, the
testing program, the measuring procedure and equipment are presented, and the tested materials and specimens are
described.

2.1. Testing Program

Asphalt mixtures samples were delivered to the laboratory from several actual building sites in Croatia. Cylindrical
specimens with the diameter ~100 mm and height 60-65 mm were prepared in laboratory, according to the standard
Marshall procedure, with various number of compaction blows (Table 1).

Table 1. The asphalt mixtures testing program

The intention was to cover the largest possible number of asphalt mixtures that are currently used in road construction in
Croatia, as well as the largest possible number of specimens, in order to eliminate the possibility of excessive errors in
final results.

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

A total of 7 mixtures was tested, 16 specimens of each, except for the first one, for which 10 specimens were produced,
which comes to 106 specimens in all.

All the specimens were tested at four temperatures: +5OC, +15OC, +25OC and +35OC, with respect to the asphalt
pavement temperature range in Croatia (-8°C to +38°C,) and with respect to the equipment manufacturer's
recommendation (+0°C to +35°C) [3]. The four temperature measurements were selected as the optimum number for
obtaining the curve of the stiffness modulus dependence on temperature.

There were 848 measurements by indirect tensile test of the asphalt mixtures stiffness modulus made in all.

2.2. Materials

Asphalt mixtures specimens were prepared in the laboratory from the materials delivered from actual building sites,
depending on their availability.

Tests were conducted on:

 Four wearing course mixtures, two of which ware Stone mastic asphalt, with the maximum particle size 11 mm
(HS 11D (SMA)), both with polymer modified bitumen PmB 45/80-65 (HRN EN 14023). The remaining two were
Dense Graded Asphalt Mixtures, with the maximum particle size of 11 mm (HS-AB 11E), one of which was
prepared with pure bitumen 50/70 (EN 12591), and the other with polymer modified bitumen PmB 45/80-65
(HRN EN 14023).
 Bituminized base-wearing course mixture BNHS 16, bitumen 50/70, is the wearing and the base course, and can
be placed directly on the unbound granular base.
 The mixture VS16, bitumen 50/70, is placed in the pavement structure as the binding layer, between the wearing
and the base course.
 The bituminized base course from the mixture BNS 22s, bitumen 50/70, is placed in the pavement structure
under the wearing or binding layer.

Table 2 shows the composition and physical-mechanical properties of the mixtures tested in this paper.

2.3. Test Procedure

All the tests were conducted in 2010 and 2011 in Laboratory for Asphalt and Road Geomechanics, Institut IGH d.d.
Zagreb. Measuring of the stiffness modulus was carried out in accordance with the Croatian standard HRN EN 12697-
26 Bituminous mixtures - Test methods for hot mix asphalt - Part 26: Stiffness; Annex C: Indirect tensile test on
cylindrical specimen (IT-CY) 2.

2.4. Test Equipment

Testing of the stiffness modulus was performed on the "NU-10" device, Cooper Research Technology Limited , 1998
[3] (Figure 1). The device was adjusted to the proposal of the standard EN 12697-26, Annex C, current at the time. It
can be used for measurements of the stiffness modulus and the evaluation of fatigue cracking and plastic deformation
resistance of asphalt materials.

Testing can be performed on the specimens with the diameter of 10 or 15 cm, height 3 to 8 cm, with the target loading
time of 124 ms and the target horizontal deformation of 5 microns for the specimen with the 10 cm diameter. It is
assumed that the Poisson ratio is 0.35.

In the equipment manufacturer's instruction 3, performing of measurements at temperatures under 0°C is not
recommended due to accumulation of ice in the chamber and converters, which can have a considerable effect on the
test results and cause major damages to the equipment. It is recommended to perform the testing at 5°C, 15°C and 25°C,
to plot the stiffness modulus-temperature results on a logarithmic scale and to extrapolate them to lower values.
Measuring at temperatures above 35°C is also not recommended because the asphalt material is too soft.

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

Table 2. Data on tested asphalt mixtures

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

Figure 1. The "NU-10" device for measuring the stiffness modulus by indirect tensile testing

3. TEMPERATURES OF AIR AND PAVEMENT

Since the mechanistic-empirical design model for determination of stresses and strains in the pavement structure is based
on material science, it is necessary to know the properties of the materials at the temperatures which they will be
subjected to. The response of the pavement structure is determined for three representative climate periods in the year:
summer, spring-autumn and winter.

The required temperature range was obtained by the analysis of monthly air temperatures in Croatia, and the data were
taken over from the official website of the Croatian Meteorological and Hydrological Service 4 and the Croatian
Bureau of Statistics 5. They refer to the meteorological stations in different climate zones in Croatia: Zagreb,
Varaždin, Osijek, Slavonski Brod, Rijeka, Zavižan, Ogulin, Gospić, Šibenik, Split and Dubrovnik. As the reference
values for summer and winter, the 10-year averages of monthly minimum and maximum air temperatures were taken
(data for the last 10 years). The highest average monthly maximum was recorded in Split (30°C), and the lowest in
Zavižan (-9°C). For spring and autumn, the average multiannual monthly air temperatures (DZS, DHMZ) were taken as
the reference values.

Croatia is situated in the northern temperate climate zone, and it is divided in several climate regions with respect to the
geographic position, the relationship between the land and the sea, and relief features. According to the Köppen's
division 7, the selected meteorological stations were classified by their geographic position into four climate zones
(Figure 2):

1. Central and Eastern Croatia, and Lika with the moderately warm and humid climate with warm summers (Cfb, marked
dark green on the map),
2. Istrian Coast, Kvarner and Dalmatian Hinterland with the moderately warm and humid climate with hot summers
(Cfa, marked light green on the map),
3. Dalmatia – coast and islands with the Mediterranean climate (Cs, marked red on the map), and
4. The highest mountainous areas with the snow-boreal forest climate (D, marked blue on the map).

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

Figure 2. The geographic distribution of climate types in Croatia according to W. Köppen in the standard period
1961-1990: Cfa, moderately warm and humid climate with hot summers; Cfb, moderately warm and humid
climate with warm summers; Csa, the Mediterranean climate with hot summers; Csb, the Mediterranean
climate with warm summers; Df, boreal climate (Šegota, Filipčić, 2003) 7,

The data about the monthly and seasonal air temperatures from meteorological stations in Croatia correspond to such a
division of Croatia by climate types.

Based on air temperatures, pavement temperatures were then determined, according to the Asphalt Institute equation
(1982) 8:
 1  34
Tasf  Tair  1   6 (1)
 z4 z4
Where:
Tasf - is the average monthly pavement temperature (°F) at the depth z,
Tair - is the average monthly air temperature (°F),
z - is the depth under the pavement surface on which temperature is to be determined (in)..

4. ANALYSIS OF LABORATORY TESTS RESULTS

Table 3 and Figure 3 show the mean values of the stiffness modulus of the tested asphalt mixtures, with differences in
compacting of specimens, according to temperatures of measurements.

All the specimens were tested in two perpendicular directions and if the average value obtained by testing of the
specimen in another direction deviated from the first one beyond the range from -20% to 10%, results were rejected [2],

The stiffness modulus obtained on the specimens compacted with 2*75 blows, i.e. on the specimens with less voids, are
greater than those of the modulus obtained for the specimens compacted with 2*50 blows. Only the results of BNHS 16
mixtures, at 5°C, deviate from this rule. The differences in compacting of the specimens had the smallest effect on the
stiffness modulus of the HS 11D (SMA) mixture specimens (1), i.e. the modulus obtained by more compacted specimens
are greater than those on the less compacted ones by 1% - 7%. The greatest differences of the stiffness modulus with
respect to compacting of specimens were obtained for the BNS 22s, bit 50/70 mixture, the modulus of which were on
more compacted specimens greater by 16% - 64% than those on less compacted specimens.

The rule of a decrease or an increase of the differences in modulus values (related to compaction rate) with
temperatures, was not noticed.

Mixtures for wearing courses have lower stiffness modulus than those of the mixtures for the binding, base and base-
wearing courses. The lowest stiffness modulus, at all temperatures, are those of the HS 11D (SMA) mixture specimens ,

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

compacted with 2*50 blows, while the highest are those of the BNS 22s mixture specimens, compacted with 2*75
blows.

Table 3. Stiffness modulus values Sm (MPa) of asphalt mixtures, with differences in specimens compaction, in
relation to temperatures

Stiffness modulus of asphalt mixtures depending on temperatures during measurrements

25000

20000
Stiffness modulus (MPa)

15000
NOT MEASURED

10000

5000

0
HS 11 D HS 11 D HS 11 D HS 11 D AB11E AB11E AB 11E, AB 11E, VS16, VS16, BNS 22s, BNS 22s, BNHS BNHS
(SMA) (SMA) (SMA) (SMA) PmB PmB bit 50/70, bit 50/70, BIT BIT bit 50/70, bit 50/70, 16, bit 16, bit
PmB PmB PmB PmB 45/80-65, 45/80-65, 2*75 2*50 50/70, 50/70, 2*75 2*50 50/70, 50/70,
45/80-65 45/80-65 45/80-65 45/80-65 2*75 2*50 2*75 2*50 2*75 2*50
(1), 2*75 (1), 2*50 (2), 2*75 (2), 2*50
Asphalt mixture 5°C 15°C 25°C 35°C

Figure 3. Asphalt mixtures stiffness modulus (MPa), with differences in compacting of specimens, at different
temperatures

As expected, the values of the stiffness modulus, of all the mixtures, were reduced with the increase of the temperatures
(Figure 4). The modulus values obtained at the maximum measured temperatures (35°C) were, on the average, by 91%
lower than the modulus values determined at the minimum measured temperatures (5°C), for all the tested mixtures. The
greatest modulus differences with respect to the temperature were manifested in the BNHS 16 specimens, compacted
with 2*50 blows, i.e. the stiffness modulus was decreased by 94% with the temperature increase from 5°C to 35°C. The
smallest modulus differences were those of the VS 16 specimens, compacted with 2*75 blows (the stiffness modulus

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

was decreased by 89% with the temperature increase from 5°C
Stiffness to 35°C).
modulus - Temperature

25000

20000
Stiffness modulus (MPa)

BNHS 1
15000 6, bit 50
BNHS 1 /70, No.
6 of comp
BNS 22s , bit 50/70, No. a cti on b
lows 2*5
, bit 50/7 of comp 0
BNS 22s 0, No. of a cti on b
, bit 50/7 compacti lows 2*7
VS16, B 0, No. of on blow 5
10000 IT 50/70 compacti s 2*50
VS16, B , No. of on blow
IT 50/70 compacti s 2*75
AB 11E , No. of on blow
, bit 50/7 compacti s 2*50
AB 11E 0 , No. of o n blows 2*
, bit 50/7 com pacti 75
5000 0, No. o on blow
AB11E f com pa s 2*50
PmB 45 cti on blo
HS 11 D /80- 65, N ws 2*75
(SMA) P o. of com
m paction blo
HS 11 D B 4 5 /80-65 (2 ws 2*50
0 (SMA) P ) , No. of
HS 11 D m B 45/80-65 co m paction b
(SMA) P (2) , No. lows 2*5
5°C m B 45/8 of comp
a 0
15°C H S 11 D (S 0-65 (1)
, No. of
c tion blow
s
25°C MA) Pm compacti 2 * 75
35°C B 45/80-6 on blow
5 (1) , No s 2*50
Temperature . of com
paction b
lows 2*7
5

Figure 4. Change of the stiffness modulus of the tested mixtures with temperature

The trend of a decrease of the stiffness modulus of the VS 16 mixture, and particularly of the BNS 22s and BNHS 16
mixtures, with an increase of temperatures, is different from that of HS 11D (SMA) and AB 11E mixtures. The mildest drop
of the curve is shown by the HS 11D (SMA) mixture (2), 2*50 compactor blows, and the sharpest by the BNHS 16 mixture
with 2*50 compactor blows.

Since the temperatures to which asphalt mixtures will be exposed during pavements design life go beyond the
temperature range within which the stiffness modulus were measured, those values of the stiffness modulus were
extrapolated.

Table 4 shows, along with the values of the stiffness modulus determined by tests (from Table 3.), interpolated and
extrapolated values of the stiffness modulus at temperatures from -5°C to +45°C, at 5 degree steps. The values for
temperatures higher than 5°C were obtained using exponential function based on four known points. This fixed points of
the curve are the modulus values obtained by measurements at the temperatures of 5°C, 15°C, 25°C and 35°C. For
temperatures lower than 5°C, linear extrapolation was made, because the curve obtained for higher temperatures
becomes very steep at lower temperatures and the obtained modulus values are unrealistically high (over 50000 MPa).

The results obtained for two HS 11D (SMA) mixtures, with the same working composition, are very similar. The
differences are somewhat greater on the specimens compacted with 2*50 blows. Further in the paper, the mean values of
the two tested mixtures were taken for the stiffness modulus of the HS 11D (SMA) mixtures at different temperatures.

Table 4. The stiffness modulus of asphalt mixtures (MPa) depending on temperatures, determined by
interpolation and extrapolation based on the measured values

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

Since the number of blows for production of Marshall specimens depends on the expected intensity of traffic on the road
section in which the asphalt mixture will be placed (Table 5), the results in Tables 6 and 7 are presented depending on
the anticipated traffic load.

Table 5. Marshall criteria for mixture design of finishing and base course layers (number of compaction blows)

Tables 6 and 7 show the values of the stiffness modulus for tested mixtures, presented by seasons and climate regions of
Croatia in which mixtures will be laid. With respect to the position of the layer in the pavement structure (the
approximate depth at which the mixture will be placed), for each mixture, the depth at which its temperature was
determined, was selected. For wearing courses, the temperature was determined at the depth of 2 cm, for the binding
layer at 6 cm, and for the base course at 10 cm under the pavement surface.

For roads with light and middle-intensity traffic load, the stiffness modulus are presented in Table 6, which contains the
results obtained on the specimens compacted with 2*50 blows. For roads with heavy and very heavy traffic load, the
stiffness modulus are presented in Table 7., which shows the results obtained on the specimens compacted with 2*75
blows.

The stiffness modulus read from Tables 6 and 7 can be used in determining the stresses and strains in critical points of
the pavement structure (for example, for use in the BISAR, computer program) with respect to the geographic location
at which the mixtures will be laid, the position in the pavement structure, and the anticipated traffic load. It is
recommended to round the values of stiffness modulus to the nearest hundred for the purpose of the calculation.

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

Table 6. The stiffness modulus of mixtures (MPa) depending on seasons and the geographic location, for roads
with very light, light, and middle-intensity traffic load

* For the HS 11D(SMA) and AB 11E wearing course mixtures, the temperature was determined at the depth of 2 cm, for that of the binding layer, VS
16, at the depth of 6 cm, and that of the base coarse, at 10 cm under the pavement surface
**The binding layer (VS 16) is embedded on motorways, roads with very heavy and heavy traffic load, so that the values of its stiffness modulus are
not presented in this Table.

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

Table 7. The stiffness modulus of mixtures (MPa) depending on seasons and the geographic location, for roads
with heavy and very heavy traffic load

* For the HS 11D(SMA) and AB 11E wearing course mixtures, the temperature was determined at the depth of 2 cm, for that of the binding layer, VS
16, at the depth of 6 cm, and that of the base coarse, at 10 cm under the pavement surface
**The BNHS mixture is placed exclusively on roads intended for light or very light traffic load, so that the values of its stiffness modulus are not
presented in this Table.

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

 5. SUMMARY

Based on the analysis of the results obtained by measurements of the stiffness modulus by indirect tensile test, on seven
different asphalt mixtures and at four temperatures, on the total of 106 specimens, the following was observed:
- Values of the stiffness modulus, of all the tested mixtures, decrease with the temperature increase.
- The stiffness modulus-temperature curves of the tested mixtures are the steepest at low temperatures, and the mildest
at high temperatures.
- The content of cavities is inversely proportional to the stiffness modulus, i.e. the specimens compacted with 2*75
blows have a greater stiffness modulus than those compacted with 2*50.
- The wearing course mixtures have lower values of the stiffness modulus than the mixtures for the binding, base
course and base-wearing course layers.
- The increase of the stiffness modulus with the decrease of temperatures is slower for the mixtures of wearing
courses.
- The values of modules obtained by measuring in the first direction were higher than those in the other direction in
65% of the cases.

6. CONCLUSION

Until today, no systematic research of the asphalt mixture stiffness modulus and its temperature dependence have been
conducted in Croatia. Therefore, this values are supposed on the basis of tests carried out 40 years ago (Dormon and
Edwards, 1967) [1], on the mixtures that were used at the time, and under the conditions in which they were used. In this
paper, stiffness modulus of asphalt mixtures currently in use in Croatia were determined, with the purpose of obtaining
representative values for use in pavement design.

The results indicate that at all (particularly low) temperatures, the tested mixtures were considerably stiffer than it has
been supposed until now. That means that the values of the stiffness modulus used in determination of the critical
stresses and strains, were considerably lower than the actual ones, so that there is a possibility that the calculated critical
stresses were underestimated (particularly for the winter period). In the cases where critical stresses exceeded the
allowed ones, there is a great chance of low temperature cracks appearance. Better knowledge of asphalt mixtures
performance will enable preventing of early appearance of damages due to the traffic load and the environment
conditions. Also, the realistic values of parameters entered in the calculations will, in the end, contribute to selection of
the most suitable combination of quality and the price of materials.

Using results shown in Tables 6 and 7, where asphalt stiffness modulus are presented with respect to the location at
which the mixture will be (or has been) placed, the season temperature and the expected traffic load, the procedure of
determining the response of the pavement structure to loads will be simpler and more reliable than the existing one.

With respect to the fact that the deviations of tests results of particular specimens from the mean values were within 10%
in most cases (in 96% of the cases), it can be concluded that indirect tensile testing is a reliable method for measuring
the asphalt mixtures stiffness modulus

REFERENCES

1 Developments in the application in practice of a fundamental procedure for the design of flexible pavements,
Dormon G.M., Edwards J.M., Proceedings of the second international conference on the design of asphalt
pavements, pp. 99–107, 1967
2 HRN EN 12697-26, Bitumenske mješavine - Ispitne metode za asfalt proizveden vrućim postupkom (Bituminous
mixtures - Test methods for hot mix asphalt - Part 26: Stiffness; Annex C: Indirect tensile test on cylindrical
specimen (IT-CY) (EN 12697-26:2004), First Edition, December 2001
3 NU-10 & NU-5, The Operation and Maintenance Manual, Cooper Research Technology Limited, 1998
5 http://klima.hr/klima.php?id=mjes, 2011.
6 http://www.dzs.hr/Hrv/publication/stat_year.htm, 2011.
7 Köppenova podjela klima i hrvatsko nazivlje (Köppen's Climate Division and Croatian Terminology), Šegota T.,
Filipčić A., Geoadria, vol. /1, pp. 17-37, 2003
8 Design of Full Depth Asphalt Airfield Pavements, Witczak, M.W., Proceedings - Vol. I, Third International
Conference on the Structural Design of Asphalt Pavements, pp 550-567, London, 1972

5th Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul

View publication stats