Investigating the Influence of Joint Reinforcements of Portland Cement Concrete Slabs Under Asphalt Concrete Overlays on Vertical Deflections Using Accelerated Pavement Testing

Abstract

Joint gaps between slabs in underlying Portland cement concrete (PCC) pavements are critical factors in the life span of asphalt concrete (AC) overlays. Therefore, being informed about the vertical deflections that may occur around the joints is important. In this study the joints were reinforced with three different materials: ethylene propylene diene monomer (i.e., rubber joint filler [RJF] reinforcement), bitumen-based sealant (i.e., rubber plate [RP] reinforcement), and steel plate (SP) (i.e., SP reinforcement), to limit the vertical movement of the underlying PCC. Tests were conducted at the accelerated pavement testing facility for two different AC overlay thicknesses, 50 mm and 100 mm, during 100,000 passages to investigate the influence of AC overlay thickness variations on slab vertical deflections and to also evaluate types of joint reinforcement on slab vertical deflections. For 50-mm thick AC overlay, the most effective joint reinforcements to prevent deformations caused by loading were RP, RJF, and SP; for a 100-mm thickness, this was determined to be RJF, SP, and RP, respectively.

Keywords

infrastructure pavements design and rehabilitation of asphalt pavements pavement design pavement layers pavement structural testing and evaluation deformation sustainable and resilient pavements sustainable pavements

In composite pavements with asphalt concrete (AC) over Portland cement concrete (PCC), the underlying PCC pavement provides strong support for the AC overlay thereby decreasing the potential for typical distresses seen in flexible pavements, such as fatigue cracking and rutting.

However, one of the problems of AC overlay applications on a PCC pavement is reflection cracking. This is a result of movements at the joints and cracks in the underlying pavement that create high stress concentrations in the overlay. This cracking propagates through the AC surface because of vertical, horizontal, or both vertical and horizontal movement of the PCC slabs, especially from the corners and free edges of the slabs ( 1 ). These movements at joints are caused by a combination of temperature and traffic loads ( 2 ). The repetitive slab-joint movements created by traffic cause concentrated stresses to occur under the AC overlay, which propagate toward the surface. The key to eliminating reflection cracking is to reduce the deflections produced in the overlay on existing PCC joints. For this purpose, existing joints can be strengthened before an AC overlay is applied.

Studies have been carried out on approaches to protect the overlay from reflection cracking, using methods such as interlayer systems, geotextiles/geogrids, steel grids, and so forth. The method of increasing the AC overlay thickness has also been evaluated ( 3 ). Based on the literature it appears that an AC thickness of 50 mm is necessary to substantially delay reflection of cracking ( 4 ). However, Merrill et al. emphasize that the asphalt overlay thickness on concrete pavement should not exceed 250 mm ( 5 ). It has been reported by several researchers that increasing AC pavement thickness will increase the formation of rutting (6 –10).

The use of short-term tests in the evaluation of pavements built for long-term use may not give realistic results ( 11 ). For this reason, use of tests that can adjust the environmental conditions and give fast and reliable results that closely reflect reality is valuable (12, 13). Accelerated pavement testing (APT) has been developed to meet this need, providing practical assessment methods that closely simulate service conditions in APT facilities. APT is a system that solves load and speed combinations in a very short time and in a very economical way compared with field studies, specifically, in relation to controlling the frequency of wheel loads, increasing axle loads, and controlling environmental effects. In APT facilities, variables such as temperature, wheel load, direction, maneuver, loading time, and so on, can be adjusted as desired. Thus, the performances of pavements at different loading and ambient conditions are very realistic ( 14 ).

Objectives

In this study, PCC pavement construction joints were reinforced with rubber ethylene propylene diene monomer (EPDM), bituminous seal filler, and a steel plate (SP) to limit PCC slab movement and reduce discontinuous areas in the pavement. At the APT facility, two different thicknesses (50 and 100 mm) of asphalt overlays were laid over the reinforced joints and then exposed to axle loading. The effects of the reinforcement materials and different overlay thicknesses on the vertical deflection of the PCC pavement were compared.

Materials and Method

Materials

The tests at the facility were carried out on the road consisting of three sections: subgrade, PCC pavement, and AC overlay. All three test sections were selected to reflect a range of soil types in Turkey. The compacted subgrade soil classifications according to AASHTO classification were A-4, A-6, and A-7. The subgrade soils were fine-grained soils (CL, CH) according to the Unified Soil Classification System. According to the proctor test results the maximum dry density was 110.8 kN/m³ and optimum moisture content was 16%.

A 200-mm thick existing jointed plain concrete pavement was constructed in the APT facility in 2009. The existing PCC pavement was placed directly on the natural subgrade soil. Samples were taken from fresh concrete while the fresh concrete was poured. The 7- and 28-day compressive strength values of the samples are shown in Table 1.

Table 1.

Compressive Strength Values of PCC Samples

Samples (150 × 150 × 150 mm)	Compressive strength (MPa)
	7 days	28 days
Sample 1	18.9	26.02
Sample 2	21.3	26.12
Sample 3	16.3	24.05
Average	18.83	25.40

Note : PCC = Portland cement concrete.

The top layer consisted of an AC mixture with a nominal maximum aggregate size of 19.0 mm with an optimum binder content of 5.5% by the total weight of the mixture.

The PCC pavement consisted of four separate slabs divided by three transverse construction joints. The construction joint openings were 20 mm. Since the tests were to be performed by installing AC overlays of two different thicknesses on the PCC pavement, two half-joints were cut at a 1/3-depth of the PCC slabs in the longitudinal direction and the concrete pavement was divided into three separate lanes. Thus, the concrete slabs were formed to be 0.20-m thick, 1.90-m wide, and 4-m long. Similar sizes have been studied in the literature (15 –18).

The lanes were named the left side (LS) and the right side (RS). The AC overlay thicknesses of PCC pavement were 50 mm on the LS and 100 mm on the RS. To be able to adjust the AC thickness easily and to prevent the loads from affecting each other, a 1.5 m transition area was left between these two test lanes. The transverse construction joints in the 1st, 2nd, and 3rd lanes on the LS were labeled LJ₁, LJ₂, and LJ₃, respectively, and those on the RS as RJ₁, RJ₂, and RJ₃. Before the application of AC overlay, each of the three joint spacings between the PCC slabs was reinforced with different reinforcement materials.

The 1st joints (LJ₁ and RJ₁) were reinforced with rubber EPDM. This reinforcing process is referred to as rubber joint filler (RJF) reinforcement. The sealant material was melted by heating at 120°C to 130°C. The 2nd joints (LJ₂ and RJ₂) were reinforced by gluing a 3-mm thick EPDM RP over the upper surface after the rubber tiles (22 × 300 × 300 mm) were cut to size to fit into the joint. This reinforcing process was defined as rubber plate (RP) reinforcement. The 3rd joints (LJ₃ and RJ₃) were reinforced by gluing 3- × 200-mm steel sheets over the joints. This process is SP reinforcement (Table 2).

Table 2.

Characteristics of the Reinforced Joints

Test section	AC overlay thickness	PCC slab numbers		Joint numbers	Reinforcement
Left side	50 mm	1	2	LJ₁	RJF
		2	3	LJ₂	RP
		3	4	LJ₃	SP
Right side	100 mm	1	2	RJ₁	RJF
		2	3	RJ₂	RP
		3	4	RJ₃	SP

Note : PCC = Portland cement concrete; RJF = rubber joint filler; RP = rubber place; SP = steel plate.

In the RJF reinforcing process of the 1st construction joints on the LS and RS, debris were removed by directing compressed air within the joints. An insulated pool was created by adhering rubber EPDM to the joint-facing surfaces of PCC Slabs 1 and 2 on the LS and RS of the 1st joints (LJ₁ and RJ₁, respectively; Figure 1a). The joints were then filled with bitumen sealant material (Figure 1b).

The 2nd joints (LJ₂ and RJ₂) were reinforced with RPs. Firstly, RPs were placed upright in LJ₂ and RJ₂ to support the rubber EPDM sheets. These RPs were rubber flooring tiles that can be easily found on the market. The RPs, measuring 22 × 300 × 300 mm, were sized for the exact joint dimensions of 200-mm deep × 20-mm wide and placed tightly. The upper surfaces of the positioned RPs and an approximately 200-mm wide area on the PCC slab surface were covered with rubber EPDM sheet adhesive along LJ₂ and RJ₂, and a 3-mm thick rubber EPDM sheet was adhered to this area (Figure 2). EPDM, as a rubber-based material with high elasticity, is not greatly affected by environmental factors and does not lose its properties at high temperatures. This joint filling material was chosen for its capacity to adapt to the movements of the PCC slabs that would occur because of loading.

The SP reinforcement of the 3rd joints (LJ₃ and RJ₃) used a 3- × 200-mm SP that extended along the joints used. The SPs adhered over the joints with a high-temperature-resistant strong adhesive after debris were removed with compressed air within the joints (Figure 3).

In using SP that was rigid and RP, a flexible material, it was possible to investigate whether it was more effective to use rigid or flexible materials to prevent vertical slab movement and thus vertical slab deflection. The reinforcement materials were easily available, easily applied, low cost to produce and have high durability. After all the transverse joints of the existing PCC pavement were reinforced, it was ready for AC overlay construction (Figure 4).

The objective of this phase was to perform a comparison of the thickness of the AC overlays. Table 3 presents the AC overlays’ grain size distribution.

Table 3.

Aggregate Grain Size Distribution of Asphalt Concrete Overlay

Sieve, mm	Passing (%)
25	100
19	100
12.5	92
9.5	75.6
4.75 (No. 4)	45.4
2 (No. 10)	30.1
0.425 (No. 40)	13.5
0.18 (No. 80)	9.9
0.075 (No. 200)	5.2

The AC overlay with a thickness of 50 mm on the LS and 100 mm on the RS was constructed over the reinforced PCC pavement, shown in Figure 5. A roller was used for compaction of the AC overlay.

Experimental Program

The APT facility was surrounded by concrete walls and covered with a roof. The APT assembly was equipped to apply rolling wheel loads from a single axle with a dual wheel configuration. With one wheel passage, the maximum tire–pavement contact stress reached a typical value of 0.8 MPa (19 –21). The test parameters shown in Table 4 were used in the APT testing.

Table 4.

Accelerated Pavement Testing Parameters

Parameters	Values
Dual wheel load	64 kN
Tire pressure	0.8 MPa
Wheel size	295-80R-22.5
Wheel speed	2 km/h
Loading direction	Bidirectional loading
Ambient temperature	+20°C–21°C

A total of 100,000 passages was applied to the overlay, each of 1-s duration, to identify the long-term deflection response of the PCC pavement under AC overlay resulting from reinforcement materials at two different overlay thicknesses. During the tests, the ambient and AC pavement temperatures were monitored and changes of only a few degrees were allowed, to avoid temperature variation-induced behavior in the AC and PCC layers ( 22 ). The ambient temperature was generally +20°C–21°C during the tests.

The experiment included instrumentation being integrated into the loading system. This comprised thermocouples to measure the facility temperature, embedded pressure cells (Figure 6a) to provide a uniform contact pressure during the test, and linear variable differential transformers (LVDTs) (Figure 6b) to measure the PCC slabs’ surface deflections (Figure 7).

The vertical deflection data were obtained at the 25,000th and 100,000th passages. Two LVDT instruments were fixed for a joint at the corner of the adjacent slabs and a total of six LVDTs were used for each test section (Figure 7).

LVDTs were placed on each of the six joint zones of the partially exposed PCC pavement under the AC overlay to measure load-induced vertical deflections. The test sections were tested in the bidirectional mode. The wheel was driven across the joint consecutively at a speed of 2 km/h (to simulate heavy traffic), then the vertical deflections of the joints’ neighboring slabs were measured by the LVDTs. The LVDT measurements were analyzed for each of the three reinforcement materials under the two different overlay thicknesses. The positions and numbers of the LVDTs on the LS are shown in Figure 8 and a summary of the LVDT numbers are given in Table 5.

Table 5.

Numbers of the LVDTs

Test section	PCC slab numbers		LVDT numbers		Joint numbers	Reinforcement
LS (50-mm thick)	1	2	LVDT-L1	LVDT-L2	LJ₁	RJF
	2	3	LVDT-L3	LVDT-L4	LJ₂	RP
	3	4	LVDT-L5	LVDT-L6	LJ₃	SP
RS (100-mm thick)	1	2	LVDT-R1	LVDT-R2	RJ₁	RJF
	2	3	LVDT-R3	LVDT-R4	RJ₂	RP
	3	4	LVDT-R5	LVDT-R6	RJ₃	SP

Note : PCC = Portland cement concrete; LVDT = linear variable differential transformer; LS = left side; RS = right side; RJF = rubber joint filler; RP = rubber plate; SP = steel plate.

Results and Analysis

The mean vertical deflection data taken from the LVDTs on both sides of a joint were examined and the effects of the joint reinforcement materials on the joint deflections were investigated. One passage of the test truck represented two loadings, as outbound and the return. It took approximately 60 s for the truck to complete one passage. Vertical deflection graphics prepared according to the data obtained from LS and RS are shown in Figures 9 and 10, respectively. The graphs show the displacement amplitudes in the form of seating and upward movements obtained at passages of around 25,000 and 100,000 for different joint reinforcements. The LVDTs operated at a precision of 1% of a millimeter and were calibrated to produce negative values in the case of pavement seats, otherwise they produce positive values.

In this study, amplitude refers to the total upward and downward distance made by the PCC slab when the load was applied and lifted. Increases in vertical deflection amplitudes cause faulting movement in PCC pavements under repeated axle loads. In this case, reflection cracks resulting from loading occurred earlier in the AC overlay on the joint.

As seen in Figure 9, the greatest movement on the LS was in the adjacent slabs of the 2nd joint reinforced with SP for both passages at 25,000 and 100,000. When Figure 10 was examined, it was observed that the maximum slab movements in the RS occurred in the adjacent slabs of the RP-reinforced joint.

Table 6 shows the vertical deflection values obtained from LVDTs at the 25,000th and 100,000th passages.

Table 6.

Vertical Deflection Values Obtained According to Reinforcement Materials

Side	No. of passages	RJF reinforcement				RP reinforcement				SP reinforcement
		LVDT-1		LVDT-2		LVDT-3		LVDT-4		LVDT-5		LVDT-6
		(+)	(−)	(+)	(−)	(+)	(−)	(+)	(−)	(+)	(−)	(+)	(−)
LS	25,000	0.00	0.92	0.76	0.81	0.14	0.91	0.03	0.64	0.55	0.83	0.16	1.02
LS	100,000	0.00	0.53	0.37	0.95	0.45	0.57	0.42	0.45	0.61	0.89	0.23	0.93
RS	25,000	0.17	0.42	0.02	0.27	0.38	0.43	0.30	0.59	0.15	0.38	0.02	0.29
RS	100,000	0.21	0.51	0.21	0.45	0.50	0.60	0.37	0.79	0.29	0.61	0.08	0.70

Note : (+) represents upward movement and (−) represents seating movement. RJF = rubber joint filler; RP = rubber plate; SP = steel plate; LVDT = linear variable differential transformer; LS = left side; RS = right side.

Regardless of the thickness of the overlay and the number of passages, the deformation magnitudes in the seating form were higher than in the upward form. The seating and upward movement magnitudes obtained from 50-mm thick overlay were higher than were obtained for 100-mm thick overlay for each joint. The mean vertical deflections decreased as the AC overlay thickness increased from 50 to 100 mm. It was expected that decreasing the overlay thickness would increase joint deflections on PCC pavement, because of the dynamic interaction between the pavement and the wheel. The thickness of the overlay should be increased to account for the decreased strength of the layer that has deteriorated and to protect it from excessive stresses or deflections. With an increase in overlay thickness, applied loads will be distributed over a larger area of the pavement layers, decreasing the stresses and deflections imposed on them.

Conclusions

Comparisons of the vertical deflections from reinforcement materials through APT have been presented, and fairly good results were obtained, which could help achieve a better understanding of the effects of AC overlay for PCC. Focusing on the effects of overlay thicknesses and the types of joint reinforcement materials, the following observations were made:

For 25,000 and 100,000 loadings, the slabs that moved the most were on the LS (50-mm thick overlay), whose joints were reinforced with SP, and on the RS (100-mm thick overlay), with slabs reinforced with bitumen-based sealant (i.e., RP reinforcement). The deflection amplitudes in the slabs generally decreased with a doubling of the AC overlay thickness (i.e., from 50 to 100 mm). Considering that slab movements accelerate reflection crack propagation into the AC overlay, to delay this, increasing the AC overlay thickness was appropriate; however, the cost implications of this approach should be taken into account.

When the average seating deflections in adjacent plates were examined; for the LS with 50 mm AC overlay, it was observed that the biggest seating movement was in the 3rd joint, reinforced with SP on both passages, and the smallest seating deflection was in the joint reinforced with RP. In 100-mm thick overlay on the RS, it was observed that the 1st and 3rd joints reinforced with RJF (i.e., joints reinforced with rubber EPDM) and SP, respectively, had less seating and upward movements than the 2nd joint reinforced with RP in both passages.

Regardless of the overlay thickness and the number of passages, seating movements were greater than upward movements, and mean deflections decreased as overlay thickness increased. The 100-mm thick AC overlay has been shown to slow the occurrence of vertical deflection but could result in higher costs and thicker pavement structures.

For 50-mm thick AC overlay, in order of effectiveness, RP, RJF, and SP were the most suitable joint reinforcements to prevent deformations caused by loading; for 100-mm thickness, the order was determined to be RJF, SP, and RP.

This study was carried out in a laboratory environment; it would therefore be beneficial to substantiate the findings in the field under different climatic conditions. Furthermore, use of a nonreinforced control joint would facilitate comparisons. The use of dowel bars between PCC slabs could also be investigated. A more comprehensive study would be possible by changing the reinforcement materials used, the loading conditions, vehicle speeds, the soil conditions, concrete strengths, and AC overlay properties. Moreover, the formation and development of reflection cracks could be tracked along the loading transitions using different measuring instruments.

Footnotes

Author Contributions

The authors confirm contribution to the paper as follows: study conception and design: M. T. Seferoğlu, A. G. Seferoğlu, M. Çelik; data collection: M. T. Seferoğlu, M. Çelik; analysis and interpretation of results: M. T. Seferoğlu, A. G. Seferoğlu, M. Çelik, M. V. Akpinar; draft manuscript preparation: A. G. Seferoğlu. All authors reviewed the results and approved the final version of the manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Scientific and Technological Research Council of Turkey (Grant no. 217M481).

ORCID iDs

Muhammet Çelik

Muhammet Vefa Akpinar

References

Cook

Ellis

Reflection Cracking on Airfield Pavement-A Design Guide for Assessment, Treatment Selection and Future Minimization. Defense Estates, West Midland, 2005.

Al-Qadi

Elseifi

Greene

Mechanisms and Mitigation Strategies for Reflective Cracking in Rehabilitated Pavements. TRB Webinar, 2015. https://onlinepubs.trb.org/Onlinepubs/webinars/150824.pdf.

Mukhtar

M. T.

Interlayer Stress Absorbing Composite (ISAC) for Mitigating Reflection Cracking in Asphalt Concrete Overlays. PhD thesis. University of Illinois, Urbana-Champaign, 1994.

Cleveland

G. S.

Button

J. W.

Lytton

R. L.

Geosynthetics in Flexible and Rigid Pavement Overlay Systems to Reduce Reflection Cracking. FHWA/TX-02/1777. Texas Department of Transportation Research and Technology Implementation Office, Austin, 2002.

Merrill

Dommelen

A. V.

Gaspar

A Review of Practical Experience Throughout Europe on Deterioration in Fully-Flexible and Semi-Rigid Long-Life Pavements. International Journal of Pavement Engineering, Vol. 7, No. 2, 2006, pp. 101-109.

Finn

Saraf

Kulkarni

Nair

Smith

Abdullah

The Use of Distress Prediction Subsystems for the Design of Pavement Structures. Proc., 4th International Conference on the Structural Design of Asphalt Pavements, Michigan, 1977, pp. 3-38.

Wijeratne

Sargious

Prediction of Rutting in Virgin and Recycled Asphalt Mixtures for Pavement Using Triaxial Tests. Association of Asphalt Paving Technologists Proceedings, Vol. 56, 1987, pp. 111-129.

Zhang

D. L.

Prediction of Rutting in Asphalt Pavement. China Journal of Highway and Transport, Vol. 8, No. 1, 1995, pp. 23-29.

Zheng

N. X.

Niu

S. S.

X. Q.

Temperature, Axle Load and Load Frequency Model of Rutting Prediction of Heavy-Duty Asphalt Pavement. China Journal of Highway and Transport, Vol. 22, No. 3, 2009, pp. 7-13.

10.

Xiaoping

Zhang

Niu

Meng

Development of a Rutting Prediction Model for Asphalt Pavements With the Use of an Accelerated Loading Facility. Road Materials and Pavement Design, Vol. 17, No. 1, 2016, pp. 15-31.

11.

Steyn

W. J.

Hugo

Perspectives on Trends in International APT Research. In The Roles of Accelerated Pavement Testing in Pavement Sustainability ( Aguiar-Moya

Vargas-Nordcbeck

Leiva-Villacorta

Loría-Salazar

, eds.), Springer International Publishing, Cham, 2016, pp. 211-225.

12.

Jameson

G. W.

Sharp

K. G.

Vertessy

N. J.

Full-Depth Asphalt Pavement Fatigue Under Accelerated Loading: The Mulgrave (Victoria) ALF Trial, 1989/1991. Australian Road Research Board, Melbourne, Victoria, 1992, pp. 1-50.

13.

Metcalf

J. B.

Application of Full- Scale Accelerated Pavement Testing. Transportation Research Record: Journal of the Transportation Research Board, 1996. 235: 1-116.

14.

Jones

An Overview of Accelerated Pavement Testing. Proc., 4th International Conference on Accelerated Pavement Testing, Arizona, 2011.

15.

Carmichael

Marienfeld

Synthesis and Literature Review of Nonwoven Paving Fabrics Performance in Overlays. Transportation Research Record: Journal of the Transportation Research Board, 1999. 1687: 112-124.

16.

Perez

S. A.

Balay

J. M.

Tamagny

Petit

Accelerated Pavement Testing and Modeling of Reflective Cracking in Pavements. Engineering Failure Analysis, Vol. 14, No. 8, 2007, pp. 1526-1537.

17.

Ozer

Al-Qadi

L. I.

Wang

Leng

Characterization of Interface Bonding Between Hot-Mix Asphalt Overlay and Concrete Pavements: Modelling and In-Situ Response to Accelerated Loading. International Journal of Pavement Engineering, Vol. 13, No. 2, 2011, pp. 181-196.

18.

Yin

Full-Scale Test of Thermally Induced Reflective Cracking in Airport Pavements. Road Materials and Pavement Design, Vol. 16, No. 1, 2015, pp. 119-132.

19.

Al-Qadi

I. L.

Wang

Full-Depth Pavement Responses Under Various Tire Configurations: Accelerated Pavement Testing and Finite Element Modeling. Journal of the Association of Asphalt Paving Technologists, Vol. 78, 2009, pp. 721-760.

20.

Greene

Toros

Kim

Byron

Choubane

Impact of Wide-Base Single Tires on Pavement Damage. Transportation Research Record: Journal of the Transportation Research Board, 2010. 2155: 82-90.

21.

Huang

Evaluating Pavement Response and Performance With Different Simulative Tests. PhD thesis. Virginia Polytechnic Institute and State University, 2017.

22.

Killingsworth

. Concrete Jointing and Details: Thickness is Only the Start. National Ready Mixed Concrete Association. https://www.ltrc.lsu.edu/ltc_16/pdf/presentations/16-Concrete%20Pavements%20(Part%202)-Jointing%20in%20PCC%20Construction.pdf. Accessed September 25, 2021.