Prediction of rutting formation in asphalt concrete pavement/Provezu susidarymo asfaltbetonineje dangoje prognozavimas/Risu veidosanas prognozesana asfaltbetona seguma/Roobaste tekke prognoos asfaltkatetes.

Haritonovs, Viktors ; Smirnovs, Juris ; Naudzuns, Juris 等

1. Introduction

Permanent deformations or rutting is the main type of damage to the AC pavement in Latvia. To eliminate it, a large section of the asphalt pavement must be renovated, which requires a lot of financial investment. In the current economic situation, this is inadmissible. In order to solve this problem, the in-depth laboratory study of deformation properties of the AC composition is required. Since 2007 the Construction Science Centre of the Riga Technical University has been participating in the State Joint Stock Company "Latvian State Roads" research programme "Research on Application of New Technologies", where this theme is included. Within the framework of the project, the Riga Technical University has acquired modern equipment to study the dynamics of appearance of permanent deformations. By applying the new technologies in Latvian circumstances, it is possible to timely evaluate the deformation properties of the AC mixture composition prior to its being laid on the road, by manufacturing the AC specimen close to the real circumstances and loading them on a road or a street, as well as to elaborate the AC mixture compositions, which are resistant to rut formation.

2. Aim and tasks of the research

Aim of the research is to investigate stability to strain of the AC mixture compositions applied for the surface of Latvian streets and roads under the heavy transport load by considering the local climatic circumstances.

To achieve this aim the following tasks have to be solved:

--designing of the AC mixture composition with the traditional aggregate and the modified and unmodified bitumen binder;

--investigation of the temperature and transport loads characteristic for Latvian circumstances, as the main external factors influencing formation of permanent deformation;

--determining the deformation properties of the AC specimens in the laboratory circumstances by the standard testing methods of the performance properties

--the wheel tracking tests (WTT), in conformity with LVS EN 12697-22:2007 Bituminous Mixtures Test Methods for Hot Mix Asphalt--Part 22: Wheel tracking and the cyclic pressure test (CPT), in conformity with LVS EN 12697-25:2005 Bituminous Mixtures--Test Methods for Hot Mix Asphalt--Part 25: Cyclic Compression Test;

--investigation of the dynamics of appearance of permanent deformations with the help of the VESYS model.

3. Methodology

AC is considered to be a very complicated material, as concrete (Maciulaitis et al. 2009). The properties of the hardened concrete depend on the selected raw materials. The results of the implemented research indicate that the most optimal solution is to use the coarse aggregate of multifractional or discontinuous fractional composition. The optimal composition of the concrete must be selected to ensure that the binding material is not overdosed. Analysis of the AC properties is made difficult too, due to many factors, which influence these properties, the main of these being: there is no constant load amount and its operation frequency, as well as the properties change considerably depending on the temperature and load nature. In accordance with researches of some scientists, the AC mechanical properties are the function of the load amount and temperature (Erkens 2002; Park et al. 2008; Sivilevicius, Sukevicius 2007). Therefore, elaboration of a mechanical model for the AC pavement, with observation of all factors influencing mechanical properties, is very complicated. There is a possibility of either to reduce a large amount of the influencing factors to several most important ones, or to perform the time consuming tests for determining other parameters and to summarise them in one model. The material models can be divided into three groups (Blab et al. 2004):

--rheological models;

--empiric correlation equations based on the experimental stage monitoring results;

--functional equations directly based on laboratory test results.

Plastic deformations from the repeated heavy transport load increase exponentially against the upper deformation boundary ([epsilon] = 20 mm). Growth of deformations from the cyclical load is non-linear. The internationally recognized VESYS method is chosen for permanent deformation prediction during experimental testing on the wheel tracking equipment, as well as on the cyclical press equipment, in accordance with the Standart EN 12697-25 method. The VESYS model states that the ratio of vertical plastic strain per cycle d[[epsilon].sup.p]/dN, to the resilient strain, [[epsilon].sub.r], is an exponential function of the number of load cycles, N (Eq (1), Fig. 1):

1/[[epsilon].sub.r] x [[epsilon].sup.P]/dN = [micro] x [N.sup.-e], (1)

where [e.sub.r]--elastic or/resilient deformation, mm; [[epsilon].sup.p]--permanent deformation, mm; N--the number of load applications, cycles; [mu]--parameter representing the constant of proportionality of strains; e--parameter indicating the rate of decrease.

[FIGURE 1 OMITTED]

The material parameters [mu] and a are determined from the following expression:

[mu] = a x b/[[epsilon].sub.r], (2)

where a, b, [[epsilon].sub.r]--constants determined from testing, [alpha] = 1 - b.

Asphalt slabs are manufactured by the roller compaction machine in accordance with the Standard EN 1269732:2003 Bituminous Mixtures--Test Methods for Hot Mix Asphalt--Part 32: Laboratory Compaction of Bituminous Mixtures by Vibratory Compactor Method. Mechanical properties of the asphalt specimens manufactured in the laboratory are similar to those of the field compacted asphalt. For each type of asphalt, three slabs are made: two for the WTT and one for determining the resilient modulus. Thickness of the specimens corresponds to that of the field compacted asphalt layer, i.e. 40 mm. The WTT is performed in accordance with the Standard EN 12697-22 method. The equipment in the laboratory circumstances simulates the asphalt slab specimen load, which is close to the actual heavy transport load on the asphalt pavement. Testing is performed at +50[degrees]C--the warming up temperature of the asphalt pavement surface during the hottest summer days. The resilient modulus is determined by the indirect tensile test method in accordance with the Standard EN 12697-26:2004 Bituminous Mixtures--Test Methods for Hot Mix Asphalt--Part 26: Stiffness Method. The scheme of determining the resilient modulus is shown on Fig. 2.

4. Properties of AC and raw materials

4.1. Bitumen

Comparison of properties of the binders and evaluation of their conformity to requirements of the Roads Specifications 2005 (Haritonovs et al. 2005), which are the technical specifications for construction of Latvian streets and roads, are summarised in Tables 1 and 2. The modified bitumen binder has a lower penetration and the higher softening point temperature, the larger kinematic viscosity, but the lower fragility temperature, in comparison with the conventional bitumen binder B70/100.

[FIGURE 2 OMITTED]

4.2. Aggregate

The AC aggregate is selected in such a way as to include the main natural stone materials applied in manufacturing of AC in Latvia--dolomite, granite and diabase. The aggregate has its main physical and mechanical properties determined. Requirements of the Roads Specifications 2005 regulate conformity of the aggregate categories for construction of Latvian streets and roads depending on the motor road operation conditions, for instance, traffic volume (Table 3). Distribution of the aggregate properties into categories is provided in the Standard LVS EN 13043:2002 Aggregates for Bituminous Mixtures and Surface Treatments for Road, Airfields and Other Trafficked.

4.3. Compositions of the asphalt concrete mixtures

Five compositions of the dense graded AC mixture and two compositions of the SMA with the traditional aggregate and one reference mixture AC 11 with the Martin steel slag aggregate have been designed (Table 4). The optimal bitumen binder composition for the AC mixtures has been determined with the help of the Marshall method.

5. Results

For the designed AC and SMA asphalts, the deformation curves--permanent deformation growth is obtained on the WTT equipment, depending on analysis of the cycles. Table 5 provides summary of the material parameters [mu] and [alpha], deformation and resilient modulus for the AC specimen types used in the experiment.

The obtained results allow determining the max wheel tracking slope mm per 1000 load cycles for the AC specimens used in the research. Categories of the max wheel tracking slope mm per 1000 load ([WTS.sub.air]) cycles are given in the Standard EN 13108-1:2006 Bituminous Mixtures. Material Specifications. Asphalt Concrete. According this Standart, the max [WTS.sub.air] category is [WTS.sub.air]1, which means that the max wheel tracking slope per 1000 cycles is 1 mm. The estimated [WTS.sub.air] categories are provided in Table 6.

To determine the daily, weekly, monthly and annual growth of permanent deformations, the traffic volume data expressed in ESAL units are required, as the parameter N from Eq (1) is equal to the amount of ESAL units. ESAL is determined from the Eq (3):

ESAL = [f.sub.i] x G x AADT x 365 x N x F, (3)

where ESAL--equivalent single-axle load; [f.sub.i]--design line factor; G--growth factor; AADT--first year annual average daily traffic, vpd; N--number of axles on each vehicle; F--load equivalency factor for vehicle.

If accepting that the AC pavement design period on the A4 detour road (Baltezers-Saulkalne) in accordance with the project is 20 years and the annual traffic growth is 2%, the growth factor G will equal to:

G = [(1 + r).sup.n] - 1/r = 24.30, (4)

where r = i/100 - annual growth rate; i - growth rate, 2%; n - analysis period in years.

There are 16% of the five-axle tracks on the A4 detour road. The amount of cars is 74%; still, their damage effect is very small, as the axle of one truck is equivalent to axles of 8 000-10 000 cars. Volume of the AADT on the A4 detour road is 10 000. Knowing the load equivalency factors for vehicles, AADT and the growth factor, the total ESAL is determined (Table 7).

ESAL for the first year--[ESAL.sub.0] equals to:

[ESAL].sub.o] = [summation] [ESAL.sub.i]/24.3 = 0.69 x [10.sup.6]. (5)

Daily_traffic = [ESAL.sub.o]/365 = 1890. (6)

The rut formation is assumed to take place during the period of April to September from 700 till 2100, when the asphalt pavement temperature can reach the high performance temperature-- > 45[degrees]C. ESAL for the period of April to September is 55% of the annual value, and from 700 till 2100 it is 85% of the daily value. According State Limited Liability Company "Latvian Environment, Geology and Meteorology Centre", percentage of the days with the high performance temperature during this period is 2%. The annual ESAL with the high pavement performance temperature is 6452.

By using Eq (2), the rut formation dynamics has been determined (Fig. 3).

6. Conclusions

In accordance with the obtained results, the maximally allowed rut depth on the asphalt pavement with the unmodified conventional bitumen is 25 mm (in Lithuania, for instance, it is 20 mm) is reached already during the first operation year of the asphalt pavement layer.

[FIGURE 3 OMITTED]

When performing the prediction research of permanent deformations, the climatic circumstances characteristic for Latvia and the transport load expressed in ESAL units have been taken into account.

The standard category [WTS.sub.air] of the conventional AC mixture exceeds one.

To achieve the more reliable results, validity of the method must be performed, for instance, comparison of the results obtained experimentally with the laboratory research.

doi: 10.3846/bjrbe.2010.05

Received 08 July 2009; accepted 7 January 2010

References

Blab, R.; Kappl, K.; Lackner, R. 2004. Models for Permanent Deformation for Bituminous Bound Materials in Flexible Pavement. Report No. SMA-05-DE11. 154 p.

Erkens, S. 2002. Asphalt Concrete Response (Acre: Determination, Modelling & Prediction). Delft: Delft University Press Science. 220 p. ISBN 9040723265

Haritonovs, V.; Naudzuns, J.; Smirnovs, J. 2005. Cela segu risu veidoSanas celonu noteikSana [Determining the Reasons for Formation of Pavement Rutting], in Proc of the Civil Engineering International Scientific Conference. May, 26-27, 2005, Jelgava, Latvia. LLU: 52-58.

Maciulaitis, R.; Vaiciene, M.; Zurauskiene, R. 2009. The Effect of Concrete Composition and Aggregates Properties on Performance of Concrete, Journal of Civil Engineering and Management 15(3): 317-324. doi:10.3846/1392-3730.2009.15.317-324

Park, D.-W.; Martin, A. E; Jeong, J.-H.; Lee, S.-T. 2008. Effects of Tire Inflation Pressure and Load on Predicted Pavement Strains, The Baltic Journal of Road and Bridge Engineering 3(4): 181-186. doi:10.3846/1822-427X.2008.3.181-186

Park, S. W. 2007. Effect of Stress-Dependent Modulus and Poisson's Ratio on Rutting Prediction in Unbound Pavement Foundations, Journal of Korean Geotechnical Society 23(3): 15-24.

Sivapatham, P.; Beckedahl, H. J. 2005. Asphalt Pavements with Innovative Polymer Modifications for Long Life Time and Low Maintenance Costs, in Proc of the 33rd CSCE Annual Conference (6th Transportation Specialty Conference). June 2-4, 2005, Toronto, Canada. Curran Associates, 198: 1-9.

Sivilevicius, H.; Sukevicius, S. 2007. Dynamics of Vehicle Loads on the Asphalt Pavement of European Roads which Cross Lithuania, The Baltic Journal of Road and Bridge Engineering 2(4): 147-154.

Viktors Haritonovs (1), Juris Smirnovs (2), Juris Naudzuns (3)

Dept of Road and Bridge, Riga Technical University, Azenes 16/20, 1048 Riga, Latvia E-mails: (1) vh@e-apollo.lv; (2) smirnovs@mail.bf.rtu.lv; (3) juris.naudzuns@inzenierbuve.lv

Table 1. Comparison of properties of the bitumen

binders B70/100 and ModBit 80B

Bitumen
properties Results

 B 70/ Modbit
 100 80B Standard

Penetration 25 71 59 LVS EN 1426
[degrees]C, x
0.1 mm

Softening 47.70 68 LVS EN 1427
point,
[degrees]C

Paraffin wax 1.10 -- LVS EN 12606
content, %

Kinematic 322 350 LVS EN 12595
viscosity 135
[degrees]C,
[mm.sup.2]/s

Flash and fire 320 349 LVS EN 22592
points,
[degrees]C

Frass breaking -21.20 -16 LVS EN 12593
point,
[degrees]C

Solubility in 99.27 -- LVS EN 12592
toluol, %

Dynamic 146 -- LVS EN
viscosity 60 12596
[degrees]C,
Pa/s

Density, 1.0066 -- LVS EN ISO 3838
g/[cm.sup.3]

Resistance to
hardening under
the influence
of heat and 135
[degrees]C

Mass change -0.050 0 LVS EN 12607

Penetration 25 72.20 40 LVS EN 1426
[degrees]C, x
0.1 mm

Softening 51.80 71 LVS EN 1427
point,
[degrees]C

Softening point 4.10 3.0 LVS EN 1427
change,
[degrees]C

Table 2. Evaluation of the bitumen B 70/100
conformity to requirements of the Standard
LVS EN 12591:2000 Bitumen and Bituminous
Binders--Specifications for Paving Grade
Bitumens

 Requirement

Index Standard Conventional Modified

Penetration 25 LVS EN 1426 70-100 50-70
[degrees]C, x
0.1 mm

Softening LVS EN 1427 43-51 > 53
point,
[degrees]C

Paraffin wax LVS EN 12606 [less --
content, % than or
 equal
 to] 2.2

Frass breaking LVS EN 12593 [less > -15
point, than or
[degrees]C equal
 to] -10]

Solubility in LVS EN 12592 > 99.0 --
toluol, %

Kinematic LVS EN 12595 > 230 --
viscosity 135
[degrees]C,
[mm.sup.2]/s

Flash and fire LVS EN 22592 > 230 > 235
points,
[degrees]C

Elastic LVS EN 22592 -- > 50
reverse, %

Hardening
LVS EN 12607-1

Mass change % [less than < 0.5
 or equal to]
 0.8

Permanent LVS EN 1426 > 50 > 35.4
 penetration,
 25[degrees]
 C, x 0.1 mm

Elastic LVS EN 22592 -- > 50
 reverse, %

Storage
stability

Softening -- < 5
 point
 change,
 [degrees]C

Penetration LVS EN 13399 -- < 9
 change
 25[degrees],
 x 0.1 mm

 Results

Index Standard Conventional Modified Evaluation

Penetration 25 LVS EN 1426 71 59 Conforms
[degrees]C, x
0.1 mm

Softening LVS EN 1427 47.7 67.7 Conforms
point,
[degrees]C

Paraffin wax LVS EN 12606 1.1 -- Conforms
content, %

Frass breaking LVS EN 12593 -21.1 --16 Conforms
point,
[degrees]C

Solubility in LVS EN 12592 99.27 -- Conforms
toluol, %

Kinematic LVS EN 12595 322 -- Conforms
viscosity 135
[degrees]C,
[mm.sup.2]/s

Flash and fire LVS EN 22592 320 349 Conforms
points,
[degrees]C

Elastic LVS EN 22592 -- 88
reverse, %

Hardening
LVSEN 12607-1

Mass change % -0.05 0 Conforms

Permanent LVS EN 1426 72.2 40 Conforms
 penetration,
 25[degrees]
 C, x 0.1 mm

Elastic LVS EN 22592 -- 84 Conforms
 reverse, %

Storage
stability

Softening -- 1.9 Conforms
 point
 change,
 [degrees]C

Penetration LVS EN 13399 -- 6 Conforms
 change
 25[degrees],
 x 0.1 mm

Table 3. Conformity of the coarse aggregate to the
requirements of the Roads Specifications 2005
at AADT > 3500 vpd

Index Standard Requirement

 Dense
 graded
 AC SMA Lim

Flakiness LVS EN [less [less 11
index 933-3 than or than or
 equal equal
 to] 15 to] 15

Resistance to LVS EN [less [less 25
fragmentation 1097-2 than or than or
(Los Angeles) equal equal
 to] 20 to] 20

Magnesium LVS EN [less [less 6
sulphate 1367-2 than or than or
test equal equal
 to] 18 to] 18

Filler LVS EN [less [less 2.4
< 0.063 mm 933-1 than or than or
 equal equal
 to] 2 to] 2

Nordic test LVS EN [less [less 18
 1097-9 than or than or
 equal equal
 to] 10 to] 10

Index Result Evaluation

 D GR MTS

Flakiness 9 12 3 Conforms
index

Resistance to 13 12 18 Lim
fragmentation does not
(Los Angeles) conform

Magnesium 0.9 - 2 Conforms
sulphate
test

Filler 1.4 1.1 0 Lim does
< 0.063 mm not conform

Nordic 13 8 4 Lim and D
test does not
 conform

Table 4. Compositions of the dense
graded AC mixture and SMA

 Aggregate fraction
AC d-D, mass, %
mixture
type 11-16 5-11 8-11

AC 11/Lim (3) -- 37.7 --
AC 11/Gr (4) -- -- 51.5
AC 11/D (5) -- -- 21.9
AC 11/Ref (6) -- 29.8 --
AC 16/Lim 20.9 29.5 --
SMA 16/Gr 39.9 -- 28.3
SMA 11/D -- -- 51.7

 Aggregate fraction
AC d-D, mass, %
mixture
type 5-8 2-5 0-5

AC 11/Lim (3) -- 11.3 37.7 (1)
AC 11/Gr (4) 20.7 51.5 (70% 0-5)
 (30% 2-5)
AC 11/D (5) 7.6 1.9 60.2
AC 11/Ref (6) -- -- 42.9
AC 16/Lim -- 1.0 37.1 (2)
SMA 16/Gr 9.5 -- 14.1
SMA 11/D 17.9 0.9 15.1

 Bitumen
AC
mixture Dolomite B70/ Mod
type powder 100 Bit

AC 11/Lim (3) 7.6 5.7 --
AC 11/Gr (4) 3.9 4.7 --
AC 11/D (5) 3.8 4.6 --
AC 11/Ref (6) 6.5 6.8 --
AC 16/Lim 6.6 4.9 --
SMA 16/Gr 7.3 -- 5.9
SMA 11/D 8.5 5.5 --

Note: (1)--natural washed sand; (2)--crushed
sand; (3)--dolomite; (4)--granite; (5)--diabase;
(6)--martin steel slag.

Table 5. AC deformation parameters [micro] and a

 AC mixture compositions

Material AC AC AC AC
parameter 11/Lim 11/Gr 11/D 11/Ref

E, MPa 11.5 33.4 32.2 115.3

[[epsilon]
.sup.p], mm 20 ** 13.1 * 14.0 * 11.2

[[epsilon]
.sup.r],
(10-2) mm 58.4 43.2 31.9 13.2

[micro] 0.008 0.003 0.01 0.45

[alpha] 0.933 0.9830 0.977 0.470

Specimen,
mm 50

 AC mixture compositions

 SMA
Material AC SMA 11/D_
parameter 16/Lim 11/D Mod

E, MPa 55.8 68.6 164

[[epsilon]
.sup.p], mm 16.5 *** 7.61 3.0

[[epsilon]
.sup.r],
(10-2) mm 32.1 23.5 3.13

[micro] 0.02 0.52 2.54

[alpha] 0.825 0.446 0.201

Specimen,
mm

Note: * [[epsilon].sup.p] = 20 mm per 1700 cycles;
** [[epsilon].sup.p] per 5000 cycles;
*** [[epsilon].sup.p] per 10 000 cycles.

Table 6. Wheel tracking slope
[WTS.sub.air].

 [WTS.sub.air]
 factual
AC mixture (mm/1000 load
type cycles)

AC 11/Lim 3.11
AC 11/Gr 2.57
AC 11/D 6.87
AC 11/Ref 0.49
AC 16/Lim 1.50
SMA 11/D 0.56
SMA 16/Gr_ModBit 0.06

Table 7. Total ESAL estimation parameters

 Design
Number line
of tracks factor, Growth
axles [f.sub.1] factor, G AADT

2 axles
3 axles
4 axles 0.5 24.3 10 000
5 axles
6 axles

 Load
Number Equivalency ESAL in Total
of tracks factor for each ESAL x
axles vehicle, f group, x [10.sub.6]

2 axles 0.007 0.20
3 axles 1.050 0.46
4 axles 1.500 2.70 16.7
5 axles 1.760 12.50
6 axles 1.820 0.80