|
ISSN 1991-8178
© 2010, INSInet Publication
Evaluation of Palm Oil Fuel Ash (POFA) on AsphaltMixtures
Muhamad Nazri Borhan,Amiruddin Ismail, Riza Atiq Rahmat
Sustainable UrbanTransport Research Centre,Department of Civil& Structural Engineering, Faculty of Engineering & Built Environment, Universiti Kebangsaan Malaysia, Malaysia.Abstract: Malaysia is the main crude palm oil supplier in the world. During the productionof crude palm oil, a large amount of waste material is generated, such as palm oil fibers (husk),shells and empty fruit bunches(EFB). After combustion, about5% of palm oil fuel ash (POFA) or boiler ash is produced.The mechanical properties of the modified asphalt mixtureswere examined and comparedwith a conventional mixture. The physical propertiesof POFA were analyzed first. Asphalt concretemixes having different percentages of POFA (0, 1, 3, 5 and 7%) as an additiveof the mineral filler were prepared, and the materials passed through a 0.075 mm sieve. These samples were characterizedusing the Marshall stability, resilientmodulus, static creep and dynamic creep tests and fatigue testing. The study showed that POFA-containing materials showed differentperformance levels but displayed more resistance to permanentdeformation compared to the controlmixtures.
Key words: Asphalt mixtures, palm oil fuel ash, mineral filler, permanent deformation.
INTRODUCTION
Malaysia is the main crude palm oil supplier in the worldand palm oil is the main agricultural industry in the country. Currently,Malaysia has about 4 million hectares of palm oil estate, of which Sabah state owns thelargest palm oil planted area with 1.2 million hectares (30% of the total planted area). In 2006 alone,Malaysia produced approximately 15 million tons of crude palm oil(MPOB 2006). Using an average oilextraction ratio (OER) of 20%, the estimated productionof fresh fruit branches (FFB) is about 75 milliontons per annum.
However, during the production of crude palm oil, a large amount of waste material is generated, such as palm oil fibers (husk), shells and empty fruit bunches (EFB) (Figure1a). According to palm oil millsestimation, for every 100 tons of EFB are discharged from the mill. This includes 8.25 milliontons of palmoil husks, 4.5 milliontons of palm oil shellsand 17.25 million tons of EFB produced as palm oil industry waste in 2006.
Some of the waste materials with high fuel value, such as palm oil husks and shells,can be reused as fuel to produce steam for generating electricity, which is required for extracting crude palm oil. After combustion,about 5% of palm oil fuel ash (POFA) or boiler ash is produced (Figure 1b). Normally a steam boiler requires a palm oil shell to husk ratio of 2:8 (20% palm oil shell plus 80% palm oil husk) to give the best performance of the boiler,considering environmentalpollution control. Based on this mixingcomposition, it is estimatedthat 10.31 million tons of palm oil waste (8.25 and 2.06 milliontons of palm oil husk and shell, respectively)generated about 516 thousandtons of boiler ash or POFA in 2006. Given the limited utilizationof POFA, this material has thus far been disposed as landfill material without any commercial returns.In the paper, the useof POFA is investigated as filler material in asphalt mixtures.
Additive or filler materials have been used in asphalt binders to design against or to repair pavement due to the following problems: surfacedefects (raveling and stripping), structural defects(rutting, shoving and distortion) and cracking (fatigueand thermal).
Many authorshave studied the effectsof mineral fillers, which are materials passing a sieve size of 0.075 mm, on the behaviorof asphalt mix (Anani and Al-AbdulWahhab 1982, Asi and Assa’ad2005, Kandhal et al. 1998, Karasahin and Terzi 2007). Differentfiller materials may have different mechanical properties in the asphalt mixture.Dukatz and Anderson(1970) have investigated eight different fillermaterials to investigate the mechanical properties of asphalt, and they found that different filler materialshave different effects onstiffness and have almost no effect on MarshallStability or void ratio. However,Mogawer and Stuart (1996)Corresponding Author: Muhamad NazriBorhan, Sustainable UrbanTransport Research Centre, Department of Civil & Structural Engineering, Faculty of Engineering & Built Environment, Universiti Kebangsaan Malaysia, Malaysia.
5456
Aust. J. Basic & Appl. Sci., C(): CC-CC, 2010
investigated eight different fillermaterials, which were known in Europe,and they foundthat good quality fillers and poor qualityfillers did not affect the performance of mixtures. Asi and Assa’ad (2005) studied the affect of Jordanianoil shale fly ash on asphalt mixes. Replacing 10% of the mineral filler was the optimal replacement. Many differentwaste materials, such as fly ash (Ali et al.1996), coal ash (Eleni and Serji 1999), sewage sludge ash (Sayed et al. 1995) and marble waste dust (Karasahin and Terzi 2007),are used in asphalt mixtures as filler material. Severallaboratory tests, including Marshall mix design, resilient modulus, staticcreep and dynamic creep tests, were conducted to evaluate the performance of asphalt mixtures with POFA.The detailed testing program is shown schematically in Figure 2.
Fig. 1: Palm oil residues and palm oil fuel ash (POFA)
Fig. 2: Flow chart of the experimentalprogram
MATERIAL AND METHODS
The aggregate selected for the laboratory work was granite stone that was obtained from quarries around Selangor, which are mainly used for highway construction. The selected aggregate gradation was in accordancewith the Malaysia Ministry of PublicWorks(PWD)-recommended gradation for a heavytraffic wearing course, as shown in Table 1. Sieve analyses were carried out, and the availablegrading curve for the aggregateused in the study is shownin Figure 3.
Asphalt samples were collected from an asphalt cement producer in Malaysia. 80/100 penetrationasphalt cement was used in this study. Standard laboratorytest results for asphalt cement are shown in Table 2. Thefiller material (POFA) used in this research was from palm oil factories. Therefore, POFA was sieved from
0.075 mm. The analysis of POFA is presented in Table 3.
Marshall stability tests were carried out in order to find the optimumasphalt content (OAC) of the mixes. The Marshallmix design procedure (ASTM D1559) is currentlyfollowed in Malaysiafor asphalt concretemix design to determine the optimum asphalt content. The optimum asphalt content (OAC) was selected to produce
5457
Aust. J. Basic & Appl. Sci., C(): CC-CC, 2010
4% air voids. The OAC obtainedwas 5.35% OAC of the total mix weight. At this OAC, the Marshallstability, flow, voids filled with asphaltand voids in mineralaggregate values were checked to verify that they were within the specification limits of PWD for heavy traffic loads wearing course. Using the an OAC of 5.35%,asphalt concrete mixes with differentpercentages of POFA (0, 1, 3, 5 and 7%) as an additiveof the mineral filler, which passed a 0.075 mm sieve, were prepared.These samples were characterized using the Marshallstability, resilient modulus, static creep and dynamic creep tests and fatiguetesting.
Fig.3: Malaysia Ministryof Public Works (PWD) specifiedgradation limits and used gradation
|
0.075 1.5 0 - 3
|
Weight loss on heating (%) 0.6
Table 3: Physical and chemical properties of POFA Property Value
Specific gravity 2.22
Chemical composition (%)
Silicon dioxide 43.6
Aluminium oxide 11.4
Ferric oxide 4.7
Calcium oxide 8.4
Magnesium oxide 4.8
Sulphur trioxide 2.8
Sodium oxide 0.39
Potassium oxide 4.5
Loss on ignition 18
5458
Aust. J. Basic & Appl. Sci., C(): CC-CC, 2010
RESULTS AND DISCUSSION
Marshall Stability Test Result:
The Marshall method (ASTM D1559) was used for determining the optimal asphalt content forconventional and modified asphaltmixtures. The results of the Marshalltest are presentedin Table 4.
From this table, it can be seen that the samples with 0% POFA,after 30 minutes immersion in a water bath, have the highestMarshall stability values, followed by the 3, 5, 1 and 7% POFA samples, respectively. Although there is inconsistency in these results with respect to the POFA content, the difference between theMarshall stability valueswas not notably high. This might be attributed to the pozzolanic cementing nature ofPOFA.
Table4: Effect of POFA on the MarshallStabilities Sample
% POFA number Stability Average
| 0 | 1 | 12.364 | |
| | 2 | 11.014 | 11.704 |
3 11.734
| 1 | 4 | 9.452 | |
| | 5 | 8.879 | 9.055 |
6 8.834
| 3 | 7 | 10.948 | |
| | 8 | 10.954 | 10.815 |
9 10.543
| 5 | 10 | 9.941 | |
| | 11 | 9.866 | 9.796 |
12 9.582
| 7 | 13 | 8.415 | |
| | 14 | 8.752 | 8.617 |
15 8.684
Please (1961) indicatedthat the fact that the Marshall stability decreasesas the binder content decreases below the optimum content value is an indication that the Marshall stability alone is not adequate to describethe ability of mixes to resist deformation.Please further indicatedthat the Marshall stability is significantly dependent on the binder viscosity,but also claimed that, because of the involvement of particle friction, therelationship is not a simple proportionality. Therefore, the Marshall quotientmay not be a good indicatorfor measuring permanent deformation for asphalt concretemixtures.
Resilient Modulus,Mr, Test Result:
The resilient modulus is the most important variable for mechanisticdesign of pavement structures. It is the measure of pavement response in terms of dynamic stressesand the corresponding strains. In recent years, there has been a change in philosophy in asphalt pavement design from the more empirical approach to the mechanistic approach based on elastic theory. AASHTO in 1986 this mechanisticapproach in the form of elastic theory is being used by increasingnumbers of highway agencies.Elastic theory-based design methods require the elastic properties of pavement materials as input. The resilient modulus of asphalt mixtures, measured in the indirect tensilemode (ASTM D4123), is the most popular form of stress–strainmeasurement used to evaluateelastic properties. The resiliency modulus, along with other information, is then used as input tothe elastic theoriesmodel to generate an optimum thickness design.Therefore, the effectiveness of the thickness design procedure is directly related to the accuracyand precision in measuringthe resiliency modulusof the asphalt mixture.The accuracy and precision are also important in areas where resilient modulus is usedtoas an index for evaluating stripping, fatigue and low temperature crackingof asphalt mixtures (Brown andFoo 1991, Kulash 1994, Tian et al.1998).
Threesamples for each POFA concentration were tested with the diametral resilient modulus (Mr) test at a test temperature of 40°C. Figure 4 shows the Mr values obtained for all the tested mixes at 40°C. It indicates that POFA modification improved the diametral resilientmodulus of the modified mixes compared to the conventional mixture for the test temperature. The average resilient modulus of the conventional mixture at
40°C was found to be 1,142MPa; this value increased to 1,395MPa for the 5% POFA mixes.For the 1, 3 and 7% POFA mixes, therewas an increase in the resilientmodulus values,but not as high as for the 5%POFA mixes.
According to the resilientmodulus test, specimens incorporating 5% POFA had higher elasticitymodulus,
22% at 40°C, and these mixtures had the lowest cracking resistance. The resultsindicate that the stiffness modulus valuesof the modified mixtures are higher than that of the conventional mixture.
5459
Aust. J. Basic & Appl. Sci., C(): CC-CC, 2010
Fig.4: Comparisonof Mr Values at 40°C
Static Creep Test Result:
A static creep test is conducted by applyinga static load to a specimen and then measuringthe permanent deformation of the specimen afterunloading. The test was used to determine the permanent deformationof the asphalt mixtures.The observed permanent deformation of the asphaltmixtures was then correlated withthe rutting potential. The creep deformation of a cylindrical specimen under a uniaxial static load was measuredas a functionof time, and the sample dimensionsand test conditions were standardized. Permanent deformationrisk was greater under heavy loads and high temperature, so the following test parameters were selected: the uniaxial load was 425 KPa (0.4 MPa), the temperature was 40°C, and the load duration was 3600 s.
The values of static creep compliance obtainedfrom the test are given in Figure 5. The strainsof the asphalt concrete mixtures initially increased rapidly, but became stable later during the hour-long test. According to the static creep results,the 5% POFA mixtureshowed better performance than the other asphalt mixes. The addition of POFA has improvedthe performance of the modifiedasphaltic concrete.
Fig.5: Comparisonof static creep behaviour of the differentmixes at 40°C
Dynamic Creep Test Result:
The dynamic creep test is a test that applies a repeatedpulsed uniaxial stress on an asphaltspecimen and measures the resulting deformations in the samedirection using linear variabledifferential transducers (LVDTs). The applied stress on the specimen was a feedbackhaversine pulse. The pulse width durationwas 100 ms,and the rest period before the application of the next pulse was 900 ms. Experiments were performed at a
5460
Aust. J. Basic & Appl. Sci., C(): CC-CC, 2010
40°C test temperature. Samples were exposed to a 780 N (100 kPa) starting load. An average 1100 N (138 kPa)load was used for the duration of the test. The skin and core temperatures of the specimen duringthe test weremonitored by two thermocouples, whichwere inserted in a dummy specimen located near the specimen under test. The testing was continued until the maximum axial strain limit reached 10,000 microstrains or until 10,000cycles, whichever occurredfirst (Gabrielson 1992).
Figure 6 shows the relationship betweenthe number of cycles and the axial accumulatedpermanent deformation for the four tested groups (0, 1, 3, 5 and 7% POFA).The 5% POFA mixes showedthe best performance, followed by the 3%, 1%, 7% and 0% POFA mixes, respectively.The values of creep stiffnesscompliance obtained from the test are given in Figure 7. Specimens incorporating POFA showed higher creep stiffness compared to the conventional mixture. Therefore, it can be inferred that the POFA samples have less rutting potential compared to the conventionalmixture. Therefore, POFA, due to its pozzoloniccementing properties, improved the creep resistance of the asphalt concretemixes. The addition of POFA improved the performance of the modifiedasphaltic concrete and increased resistance to permanent deformation.
Fig. 6: Comparisonof dynamic creep behaviourof the different mixes at 40°C
|
Numerous investigations have been carried out on incorporating mineral filler modifiedbitumen to improve the performance of bituminous composites.Most of the results obtained from laboratory and full-scale trials demonstrateto varying extentsan improvement in the performance of these modified bituminous mixes in terms of increasedresistance to permanent deformation, improvement in fatigue life, improved durability and resistance to moisturedamage.
Fatigue Test Result:
Fatigue properties are importantbecause one of the principal modes of pavement failure is fatigue-related
5461
Aust. J. Basic & Appl. Sci., C(): CC-CC, 2010
cracking, called fatigue cracking.Therefore, an accurate prediction of fatigue propertieswould be useful in predicting overall pavement life. Figure 8 shows the results of these tests. In this figure, regression lines are drawn through the mean of the samples at each strain level. The results show a normal linear relationshipbetween the logarithm of the applied initialtensile strain and the logarithmof the fatigue life.
Fig. 8: Comparisonof fatigue behaviorof the different mixes at 40°C
The fatiguedata were analyzedby running a regression analysisto determine the fatiguerelationship parameters in the followingform:
εt = I(Nf)S (1)
where εt = initial tensile strain, Nf = number of load repetitions to failure, I = antilog of the intercept of the logarithmic relationship, and S = slope of the logarithmic relationship.Regression parameters for Eq. (1) areshown in Table 5. Analysisof the fatigue results shows significant improvement in fatiguelife of the mixeswith 5% addition of the mineral filler with POFA followed by the 3, 7, 0 and 1% POFAmixes, respectively. Therefore, POFA, due to its cementingproperties, improvedthe fatigue properties of the asphaltconcrete mixes.
Table5: Parametersfor Fatigue Equationof Asphalt Mixtures POFA content (%) Regression factor Coefficient of determination R²
-------------------------------------------------------------
I S
| 0 | 1964.8 | -0.2682 | 0.9901 |
| 1 | 2388.4 | -0.3239 | 0.9936 |
| 3 | 2430.7 | -0.2547 | 0.9981 |
| 5 | 2552.4 | -0.2288 | 0.9534 |
7 2623.1 -0.3019 0.9688
The fatigue behavior of POFA-modified mixes was found to be significantly improved compared to conventional mixtures. The increase in fatigue life, as observedfrom laboratory fatigue test results, is nearlytwofold. Considering the improvement in other mix parameters, such as better elastic recovery of the modified binder, field fatigue livesof the POFA-modified mixes can be expectedto be at least two times longer thanthat of normal asphaltmixes.
Conclusion:
For the mixturesevaluated in this study, the following conclusions are derived.
• Marshall stability values were found to be generallyhigher than that of the conventional mixture. As faras the Marshall test results are concerned for both conventional and modifiedmixtures, the Marshallquotient may not be a good indicator of measuring permanent deformation for asphalt concretemixtures.
5462
Aust. J. Basic & Appl. Sci., C(): CC-CC, 2010
• According to the resilient modulus test, POFA modificationimproved the diametral resilientmodulus of themodified mixes compared to that of the conventional mixture. The presence of the POFA in the concrete mixtures increasedthe elasticity modulus and the stiffnessof the mixture.
• The addition of POFA improved the performance of the modified asphalticconcrete in terms of therheological behavior (i.e., increase in stiffness). Therefore,POFA, due to its pozzolonic cementingproperties, improved the creep resistance of the asphaltconcrete mixes.
• Incorporating POFA into bitumen greatlyincreases the fatigue properties of the asphalt concretemixes.
The fatigue behavior of POFA-modified mixes was found to be significantly improved compared to conventional mixture. The increase in fatiguelife, as observed fromlaboratory fatigue test results, is nearly twofold. Considering the improvement in other mix parameters, such as better elastic recoveryof the modified binder, the field fatiguelives of the POFA modified mixes can be expectedto be at least two times longerthan that of normal asphalt mixes.
• The replacement of the mineral filler by POFA can reach up to 5% without impairing the performance properties of the asphaltconcrete mixes.
REFERENCES
Ali, N., J.S. Chan, S. Simms, R. Bushman and A.T. Bergan, 1996. Mechanistic evaluation of fly ash asphalt concrete nixtures. Journalof Metarials in Civil Engineering, 8: 19-25.
Anani, B. and H. Al-Abdul Wahhab, 1982. Effects of baghouse fines and mineral fillers on properties of asphalt pavements. Transportation ResearchBoard, 843: 57-64.
Asi, I. and A. Assa’ad, 2005. Effect of Jordanian oil shale fly ash on asphalt mixes. Journal of Materialsin Civil Engineering, 17: 553-559.
Brown, E.R. and K.Y. Foo, 1991. Evaluationof variability in resilient modulus test results.Journal Test
Evaluat., 19: 1-13.
Dukat,z E.L. and D.A. Anderson, 1970. The effect of various fillers on the mechanical behavior of asphalt and asphaltic concrete. TransRes Rec., 38: 46-58.
Eleni, V.C. and N.A. Serji, 1999. Coal ash utilizationin asphaltconcrete mixtures. Journal of Materialsin Civil Engineering., 11: 295-301.
Gabrielson , R.J., 1991. Evaluationof hot mix asphalt (HMA) static creep and repeated load tests, PhD. Thesis, Auburn University. Alabama.
Kandhal, P., C. Lynn and F. Parker, 1998. Characterization tests for mineral fillers related to performance ofasphalt paving mixtures. NCAT Rep. No. 98-2, NationalCenter for AsphaltTechnology, Auburn.
Karasahin, M. and S. Terzi, 2007. Evaluation of marble waste dust in the mixtureof asphaltic concrete. Constr and Build Mater.,21: 616-620.
Kulash, D.J., 1994. Toward performance–based specificationsfor bitumen and asphaltmixtures. Proc
Institut Civil Eng – Transport., 105: 187-94.
Malaysia Palm Oil Board,2006. Malaysian Oil Palm Statistics.
Mogawer W.S., K.D. Stuart, 1996. Effect of mineral fillers on propertiesof stone matrix asphalt mixtures. Trans Res Re., 1530:86-94.
Please, A., 1961. Use of the Marshall test for evaluating dense bituminous surfacings.J Appl. Chem., 11:
73-80.
Sayed, M.H., I.M. Madany and R.M. Buali, 1995. Use of sewage sludge ash in asphaltic paving mixtures in hot regions. ConstrBuild Mater., 9(1): 19-23.
Tian, P., M.M. Zaman, G. Joakim and G. Laguros,1998. Variation of resiliency modulus of aggregate base andits influence on pavement performance.J Test Evaluat., 26: 329-35.
5463
Do you like this article? Share It!
0 comments to "Palm Oill Fuel Ash POFA study"
Popular Posts
-
The Fuji Instax Mini 25 and Instax Mini 7S are a couple of moment picture zoom lens utilizing Fuji's Visa measured moment picture. In th... -
The above picture is one of the most famous picture in the world. If you are the requester band Rage Against The Machine, the picture above... -
The hotel room sports a modern yet cozy design and everything you need is provided for. The bed is super comfortable and remains the thing I...
-
Radisson - Blu Hotel , Berlin Germany is the world's largest cylindrical aquarium , named Aquadom . It is 25 meters ...
Post a Comment