A SEMINAR REPORT ON

ASPHALT RECYCLING METHODS & DESIGN OF PAVEMENT WITH CENTRAL PLANT

HOTMIX RECYCLING METHOD (CPHMR)

����

����

����

����

����

����

� �� �� �� � ��������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������

���� ��� � �� ����� ��� � �� ����� ��� � �� ����� ��� � �� � ����

����

��������������������������������������������

ABSTRACT: Signified the use of RAP towards the pavement rehabilitation

particularly for pavement recycling, discussed recycling methods and focused hot mix

recycling (CPHMR) – A case study of pavement design with central plant hot mix

recycled asphalt mixes is discussed where laboratory investigations carried over the

Recycled mix and virgin mix by comparing its Marshal values, Creep and Fatigue

performances. Finally the cost comparison is tabulated shows the immediate

implementation of recycling mixes in India for the economical benefits.

INTRODUCTION: Any in-service road needs maintenance. The maintenance of

roads can be done in the form of minor repair, overlay, recycling or reconstruction.

Given the fund constraint for any maintenance job, an engineer needs to equip himself

on various techniques of evaluation of pavement, alternative rehabilitation measures

and systematic approach to allocate maintenance funds.

Pavement recycling is expected to be a major thrust in pavement rehabilitation in

India, soon. Significant advancement has been achieved in mix design and

construction practices of bituminous pavement recycling in various countries.

Recycling of existing roadway materials is a highly attractive option for rehabilitation.

Recycling can not only save in costs of materials, but in most cases, can also lead to

significant savings in time of construction, and in all cases, lead to reduction in

disposal, and environmental benefits. Interestingly, not much of pavement recycling

works at present being carried out India, though basic initiative was taken quite

sometime back. The present workshop proposes to share the knowledge of state of the

art tools and techniques of bituminous pavement recycling which is vital for proper

utilization of the most appropriate recycling method for optimum utilization of funds

The bituminous pavement rehabilitation alternatives are mainly overlaying,

recycling and reconstruction. In the recycling process the material from deteriorated

pavement, known as reclaimed asphalt pavement (RAP), is partially or fully reused in

fresh construction.

Some of the advantages associated with pavement recycling are

� less user delay

� conservation of energy

� preservation of environment

� Reduced cost of construction

� Conservation of aggregate and binder

� Preservation of existing pavement geometrics etc.

� Recycled mix has higher resistance to shearing and scuffing, which in

turn increase the rutting resistance

� Chances of reflective cracking are found to be less with recycled mix

The purpose of the bituminous recycling is to regain the properties of the RAP, such

that it tends to perform as good as fresh mix. Thus, the process of bituminous

recycling involves mixing of the RAP, fresh bitumen, rejuvenators and new

aggregates in suitable proportions. Rejuvenators are low viscosity oily substance,

which helps to bring down the high viscosity of aged bitumen.

Recycling Methods:

� central plant recycling: If the RAP is modified at a plant, away from

construction site

� In-situ recycling. In this the RAP is modified in place, from where it is

available

If heat is applied then the process is known as hot mix recycling. In case of cold

mix recycling, old materials are conditioned using recycling agent (like, low viscosity

emulsion) without application of heat.

Fig.1 classification system

Based on the depth of removal: Surface recycling: Done when the top layers fail, and recycling is done to that failure

portion.

Full depth reclamation: Done when pavement layers up to base layer is removed and

constructed again.

Hot in-place recycling: � Initially the pavement intended to be recycled is heated to a higher

temperature using suitable heating arrangement. This facilitates easier

removal of materials. After heating, the pavement surface is scarified

to the required depth.

� Further, depending on the requirement fresh aggregate and binder are

added. The material is mixed well and compacted to the required

thickness. As this process consumes less time, least disruption to traffic

is caused. Also the transportation cost is less, as materials need not be

taken away.

� Machinery required for this purpose being bulky in nature, sufficient

right-of-way is required. This becomes an important consideration for

in-place recycling within the city areas.

Cold in place recycling � The pavement is scarified with a scarifier. The scarified material is

crushed to the required gradation. Then the required amount of fresh

aggregates and binder in cold form (emulsion or cutback) is added. It is

compacted and left for aeration.

� During this process additives like, cement, quick lime, fly ash may be

used.

� The cold mix recycling takes care of local geometric correction of

pavement distresses like surface cracks being an in-situ process the

hauling cost is considerably low.

� The air quality related problems during construction is almost

negligible as compared to hot mix process

� Similar to hot in place recycling process, the machinery required being

bulky, sufficient maneuvering space should be available for operating

the equipment. Also, the lane needs to be closed for certain time so that

sufficient time is available for curing of freshly laid course.

� Moisture content (when bitumen emulsion is used) needs to be given

importance as it influences gradation control, mixing and workability

of recycled mix to a large extent

Hot central plant recycling

� In this process, RAP is combined with required quantity of bituminous binder,

and fresh aggregates in a hot mix plant. The resultant mix is heated to an

elevated temperature and mixed thoroughly. The hot mix is transported to

paving site, placed, and compacted to the required compaction level.

� The main advantage of this process is that the mix properties and performance

is comparable to that of virgin mix especially have noted that the quality

control due to the fact that constituents are mixed under controlled conditions

and it is possible to monitor mixing process continuously.

� Less workspace is required for laying the recycled mix; hence this is suitable

for the roads where the right of-way is somewhat restricted.

� The RAP should not be exposed to extremely high temperature as it causes

pollution due to smoke emission.

� As RAP is susceptible to moisture and care needs to be taken while storing it.

Cold central plant recycling

� This is the similar process as is the hot central plant mixing, except it does not

involve any heating, and therefore emulsion bitumen is used binder in most of

the cases.

� Precise control on the mixing time is important. Over-mixing may cause

premature breaking of emulsified bitumen and under-mixing results in

insufficient coating of aggregates.

CASE STUDY:

� The investigations are done for the hot mix recycling process since large

number of studies has been reported a conflicting conclusions like the

performance of hot mix recycled mixes in fatigue ,rutting or stiffness could be

better, worse or similar compared to the corresponding virgin mix.

� Two RAP samples are collected at Kanpur city for the recycling investigations

are carried out on the design of the recycled mix and laboratory tests Marshal,

fatigue and fatigue done and compared their performance with the virgin mix.

Laboratory investigation

The basic requirements of the recycled mix design can be summarized as follows: � The quantity of old aggregates and new aggregates are to be adjusted in such a

way that the resultant gradation of aggregates conforms to the specified

gradation.

� The quantity of the aged asphalt binder, virgin asphalt binder and the

rejuvenator, if any, are to be adjusted in such a way that the resultant viscosity

becomes equal to the desirable viscosity at operating temperature.

� The total quantity of asphalt binder should be adjusted in such a way that it

satisfies the desired asphalt binder quantity of the target mix.

� The other volumetric and strength parameters of the mix should also be

satisfied.

� The RAP samples are collected, proportioned, and mixed with virgin asphalt

binder and new aggregates, for various target bitumen contents. Standard

Marshall testing is conducted for estimation of the possible optimal binder

content. Further, creep and fatigue tests are performed on the recycled samples

in order to assess their performance.

� The same tests are conducted for virgin mixes, with same specification, in

order to have a comparative idea of mix performance. The schematic plan of

the Study is presented in fig.1

PROCEDURE& DESIGN CALCULATIONS:

� The representative RAP samples are cleaned for deleterious materials and

aged asphalt binder and old aggregates are separated using Centrifuge

Bitumen Extractor(CBE) as per ASTM D2172

� The average asphalt binder content of the RAP and the gradation of the old

aggregates present in RAP are found out. The virgin and extracted (aged)

binders are tested for their physical properties. The test results are given in

Table1, the gradation of RAP aggregates is presented in Fig. 2.

FIG.1 SCHEMATIC PLAN OF THE WHOLE STUDY

FIG.2 GRADATION OF RAP AGGREGATES FIG.3 COMPONENTS OF MIX AND SPECIFICATIONS

� One way for estimating optimum binder content is to calculate the binder

demand to cover the approximate surface area of the aggregates with an

average film thickness. For SDBC as per asphalt institute guidelines it is

approximately calculated as 5.5%of total weight of the mix.

� The proportion between the virgin and aged binder can be estimated by using

viscosity mixing rule as follows, so that the resultant mix achieves the target

viscosity at the reference temperature,

( ) ( ) ( )log loge t ob e o nb nn p n p n= +

Where,

,,t o nn n n Represent viscosity of target mix, aged and virgin binder at the

reference temperature.

,ob nbp p Represent fraction of aged and virgin binder, respectively

� The minimum rolling temperature100 C, as specified by MORT&H guidelines

is chosen as the reference temperature. The target viscosity has been chosen as

2320 MPa s. This is basically the viscosity value observed in 60/70

penetration grade of binder at the target temperature. It means the target of the

present mix design Problem is to achieve a viscosity of the recycled mix (at

reference temperature) similar to that of 60/70 grade of bitumen, by using

80/100 grade of bitumen as the recycling agent. Having known the proportion

between the aged and virgin binder, percentage of binder present in RAP the

remaining desired parameters can be computed with the equations given

below.

TABLE 2:

Percentage of virgin binder * R

n b bP P

Percentage of RAP ( *100)/R R RAPRAP b bP P P+

Peracentage of new aggregates 100-

1

0

1k

N kε� �= � �� �

Mix preparation Process1: old binder+ virgin binder= homogenous mixture of required viscosity.

This mixture+ old aggregates+ fresh aggregates= recycled mix

Prcoess2: (broken RAP + fresh aggregates ---heated @high temperature and for this

known amount of virgin binder are added) = recycled mix.

Actually it is believed that the realistic situation is somewhere in between the two

cases. The above two cases represents the extreme cases. Four possible samples are

taken based on the above processing methods namely S1-T1,S1-T2,S2-T1,S2T2.

MARSHAL TESTS: Approximate binder demand for SDBC estimated as 5.5%and it may not necessarily

be the optimum content of bitumen for the mix. So the tests are carried out at different

contents and results are tabulated.

Procedure:

� Marshal samples are prepared for bitumen contents

of4.5%,5%,5.5%,6%,6.5%

� It may be noted that as the target binder content changes, the

constituent proportions also gets changed and needs recalculation with

every time.

� The recycled mix is poured into marshal’s mold, and then compacted

with marshal hammer with 175 blows on each face. The compacted

marshal are tested after 24h curing .weigh measurements are done for

estimating the volumetric parameters and samples are kept in water

bath @60 C for 30 minutes and then test for Marshal stability value

and flow values using Marshal testing machine.

� Table 4 is the specifications given by MORT&H guidelines and table 3

shown are the results obtained for the tested samples and by comparing

them we can observe that the non-compliance is mainly with the

volumetric parameters .thus the optimum binder content for any of the

recycled mixes could not be conclusively established.

� Testes have been conducted for the virgin mix, using the mid-point

gradation of the SDBC mix as specified by MORT&H, and the

common zone of binder content for which all the Marshall parameters

are satisfied is found to be ranging between 5.0% and 6.4% (refer

fig.4). Any binder content within the feasible zone can be adopted as

design value. However, binder content, close to middle portion of the

upper and lower limit of the feasible zone, entails a more reliable mix

design so; a binder content of 5.5% can be safely assumed as optimum

binder content for the virgin mix in this study.

DISCUSSIONS

� It is seen that, the Marshall stability and flow values almost remained

within the permissible ranges and even comparable to the virgin mix. It

is only the volumetric parameter values that seem to go out of the

range.

� A possible reason can be that for the recycled mix design process

adopted in the present case, first the quantities of RAP, virgin binder

and new aggregates are determined, and later the gradation of the new

aggregates are adjusted so as to closely match with the mid-point

gradation of SDBC. Since, the quantities of the old and new aggregates

are fixed beforehand, and so also the gradation of the old aggregates

(i.e., RAP gradation), the only parameter that can be varied is the

gradation of new aggregates. By this process, the achieved gradation

may not necessarily match the mid point gradation, considering all

possible gradations of the new aggregates.

Thus, there remains a need of developing specifications for recycled mixes for Indian

gradation.

TABLE 3: Bitumen content(%)

Air voids(%)

VMA(%) VFB(%) Marshal stability(KN)

Marshal flow(mm)

4.5 5.04 17.8 71.6 7.0 1.66 5.0 3.59 16.7 78.5 10.7 1.80 5.5 3.43 16.7 79.5 8.8 2.20 6.0 2.43 15.8 84.6 7.6 2.30 6.5 2.43 14.9 83.7 7.0 2.35 TABLE 4: PARAMETER PERMISSIBLE VALUE Air voids (%) 3 to 5 VMA >14 VFB 65 to 78 Marshal stability(KN) >8.2 Marshal flow(mm) 2 to 4

FIG.4 schematic sketch showing feasible zones for SDBC virgin mix design CREEP TEST: � Static creep test is one of the tests that can characterize rutting potential of a mix.

This involves application of known amount of static load for a specified duration at

constant temperature. Since the optimum binder content of the recycled mixes

could not be established conclusively from Marshall Test, it was decided to carry

out creep test over a range of binder content.

� Recommendations made by Shell pavement design manual have been used for

performing creep test. This involves loading and unloading for a period of 1 h each

at a temperature of 40 C.

� Schematic diagram of setup is shown in fig.5 it consists of a loading frame and a

pair of dial gauges. Loading frame is designed in such a way that it transfers the

load to the specimen axially. The samples are placed in between the smooth

surfaced ceramic plates with the flat surfaces horizontal. The dial gauges are fixed

at two places on the ceramic plates. The dial gauges are placed independent of the

loading frame which helps in accurate measurement of axial deformation.

� The samples for creep test are prepared in similar way the samples are prepared for

Marshall testing. Each sample is tested after curing for 24 h. calculated amount of

load is placed at the end of the loading frame such that stress of 0.1 MPa is

developed in the sample. After a period of1 h, the load is removed. The

displacements are noted at different time intervals using dial gauges over the entire

period. A typical variation (for virgin mix sample with6.5% binder content) of

axial deformation with time (i.e., creep curve) is shown in Fig.6 The recoverable

strain values at different binder contents is plotted for various mixes in Fig.7

DISCUSSION: � Recoverable strain gradually increases with the binder content and then again starts

decreasing for all types of mixes. From Fig. 7 it is seen that for most of the mixes,

maximum recoverable strain is observed between 5.5% and 6.0% of binder

content.

� Further, the permanent and recoverable strain components are compared in Fig. 8

for all types of mixes at particular binder content, chosen as 5.5% in the present

case.

� It is seen that even though permanent strain component in virgin mix is less,

recoverable strain in recycled mix is comparable to that of virgin mix.

Schematic diagram of creep test set up

Fig.6 Variation of deformation with time in creep test for virgin mix sample with

6.5% bitumen content.

Fig.7 Variation of recoverable strain with binder for different mixes.

FATIGUE TEST:

� Beam fatigue test is one of the tests that can characterize the fatigue behaviour of

the mix. In this case a binder content of 5.5% is chosen for fatigue testing for all

types of mixes. Constant strain amplitude fatigue testing is carried out to

characterize fatigue behaviour of recycled mixes. Fatigue testing machine consists

of a motor, cam, load cell, LVDT, beam holding arrangement and a data

acquisition system. Motor is used to apply sinusoidal loading on the specimen. The

cam arrangement is used to impart displacement to the specimen, which in turn

varies the strain in the beam being tested.

� A load cell is used in measuring the load applied and a LVDT is used to measure

displacement. Since constant strain amplitude fatigue test is carried out, the

maximum displacement is kept constant for the entire period of testing on a

particular beam. The beam holding arrangement is used to fix the beam rigidly in

place. A photographic view of the setup is given in Fig. 9.

� Asphalt mix is prepared and poured into the mould (length 380 mm, breadth 78

mm) in three layers and it is compacted with 175 blows with standard Marshall

Hammer. This number of blows is arrived at by comparing beam density with the

average density of standard Marshall Specimen at same binder content. These

beams are tested for fatigue performance after 24 h of curing. Each beam is fixed

to the fatigue testing machine and sinusoidal loading is applied. Load

corresponding to half the initial load is noted as failure load and corresponding

number of repetitions is also noted. Similar procedure is repeated for other test

samples at different strain levels to develop the fatigue curve. Temperature during

test is maintained 30 ± 2 _C.

A GENERAL REGRESSION BASED FATIGUE EQUATION CAN BE EXPRESSED AS

1

0

1k

N kε

� �= � �� �

Where,

N = number of load repetitions the beam can sustain till failure

∈ = tensile strain

0, 1k k = regression coefficients

Fig. 10 presents the fatigue curves of the various mixes. The regression coefficients of

the fatigue equation, as well as the average initial stiffness values for all these types of

mixes are presented in Table 5

DISCUSSION: � From Fig. 10 it can be seen that at lower strain levels, the fatigue lives of the

recycled mixes are better or similar to that of virgin SDBC mix. However, at

higher strain level, the opposite trend is observed. Fatigue performance of S2-T1 is

observed to be poorest of all other mixes.

� The stiffness values of Sample 1 recycled mixes are observed (refer Table 5) to be

higher than virgin mix. For Sample 2 it is observed to be reverse.

TABLE 5

Regression coefficients and initial stiffness of different mixes

Mix type Number of

samples

tested

Regression

coefficients

R*R Avg. initial

stiffness(Mpa)

K0 K1

Virgin mix 10 2.8150 1.3355 0.91 1477

S1-T1 9 0.5647 1.5458 0.92 1711

S1-T2 9 0.0631 1.8755

0.92 1723

S2-T1 10 0.7872 1.3573 0.94 773

S2-T2 10 0.1951 1.6928 0.78 738

PAVEMENT DESIGN AND COMPARISION:

� The individual mixes showed different values of stiffness moduli and fatigue

performance parameters. The overall effect of these two parameters can only be

assessed when pavement design is performed.

� For performing pavement design a suitable shift factor, derived from long term

pavement performance data, needs to be adopted. Different shift-factor values are

suggested by various researchers. The shift factor for asphalt pavement can even

vary between 5 and 700.

� In absence of field performance data of the recycled mixes developed, a value of

100 is chosen as shift factor for all the fatigue equations for comparison of various

pavement designs. The values of k0 of fatigue Eq. for various mixes are

accordingly modified

� For the present pavement design, a three layer pavement section is assumed, made

up of asphalt surfacing, unbound granular layer and subgrade. Literature suggests

that Poisson’s ratio generally ranges from 0.35 to 0.5 for asphalt layer, 0.2 to 0.45

for unbound granular layer and 0.3 to 0.5 for clayey subgrade.

� Thus, in the present case, typical Poisson’s ratio values as 0.35, 0.3 and 0.3 are

chosen for asphalt surfacing, unbound granular layer and subgrade, respectively.

The stiffness moduli of granular layer and subgrade have been assumed as 150

MPa and 50 MPa, respectively.

� The design traffic is assumed to be 20 million standard axles. Using the fatigue

equations, the allowable strains are obtained for the design traffic. For a given

thickness of granular layer, some tentative thickness value of bituminous surfacing

layer is assumed and the tensile strain value is calculated using linear multi-layer

elastic analysis of pavement structure. The thickness of the bituminous surfacing

layer is then adjusted such that the computed strain value is comparable to the

allowable strain value and thus the design thickness value is obtained. Therefore,

the design thicknesses from fatigue considerations obtained are compared in Table

� Now, it is possible to compare the material cost of the construction of asphalt layer

made up of recycled mix and virgin mix. Since S2-T1 design thickness is

considerably high, compared to other mixes, it has been neglected for further cost

computation.

� It is assumed that same machinery and manpower can be used for production of

virgin mix as well as recycled mix. Operations that have been considered in the

cost estimate are milling and transportation of RAP, production of asphalt mix

using hot mix plant, transportation of mix thus produced to construction site and

compacting using different rollers. It is assumed that the thickness of existing

pavement is such that there is no excess or deficit of RAP collected.

� A typical calculation is given in Appendix A. Standard schedule of rates issued by

Central Public Works Department (CPWD), Government of India has been used

for computation of the costs.

� Thus the construction cost of asphalt layer of 7 m wide and 1 km long stretch is

tabulated in Table 7.

DISCUSSION:

• It can be seen that cost of construction with recycled mix could be economical

compared to virgin mix. Percentage saving in the present example problem varied

between12.1% and 54.6% for different mixes.

• It may be noted that these differences in material costs across various mixes is due

to (i) differences of design thickness, as well as (ii) differences in the mutual

constituent proportions.

CONCLUSIONS: • Mix design for recycled asphalt mix has been performed through Marshall, creep

and fatigue tests. The parameters obtained are used in pavement design and the

economy in alternative designs is evaluated considering the material cost towards

constituent proportion as well as design thickness.

• With the recycled mixes we can have a savings of about 12% to 54%. So there is

an emergent need to utilize the technology and optimize the benefit of resources.

REFERENCE: • Pavement recycling mix design of central hot mix recycling asphalt. • Pavement recycling by Wilson • A work shop on pavement recycling conducted at IIT Kanpur during Dec’05.