Construction and Building Materials 96 (2015) 37–46

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmat

Use of basalt fibers for concrete structures

http://dx.doi.org/10.1016/j.conbuildmat.2015.07.1380950-0618/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail address: hmseliem@gmail.com (H.M. Seliem).

Cory High a, Hatem M. Seliem b,⇑, Adel El-Safty c, Sami H. Rizkalla a

a Department of Civil, Construction and Environmental Engineering, NCSU, Raleigh, NC, USAb Department of Civil Engineering, Faculty of Engineering, Helwan University, Cairo, Egyptc School of Engineering, University of North Florida, Jacksonville, FL, USA

h i g h l i g h t s

� BFRP bars had an ffu = 1,000 MPa, Ef = 45 GPa, and Ld = 32db.� Flexural members with BFRP bars are controlled by serviceability requirements.� ACI 440.1R-06 accurately predicts the nominal moment capacity.� ACI 440.1R-06 underestimates service deflection for low reinforcement ratios.� Basalt fibers increased f 0c & fr of concrete containing fly ash with low w/c.

a r t i c l e i n f o

Article history:Received 13 February 2014Received in revised form 1 July 2015Accepted 14 July 2015

Keywords:FibersBasaltFiber-reinforced concreteBondFlexureAverage residual strength

a b s t r a c t

This study investigated the use of basalt fiber bars as flexural reinforcement for concrete membersand the use of chopped basalt fibers as an additive to enhance the mechanical properties of con-crete. The material characteristics and development length of two commercially-available basalt fiberbars were evaluated. Test results indicate that flexural design of concrete members reinforced withbasalt fiber bars should ensure compression failure and satisfying the serviceability requirements.ACI 440.1R-06 accurately predicts the flexural capacity of members reinforced with basalt bars,but it significantly underestimates the deflection at service load level. Use of chopped basalt fibershad little effect on the concrete compressive strength; however, significantly enhanced its flexuralmodulus.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Basalt fibers are produced from basalt rocks, which are meltedat 1400 �C. Basalt fibers are environmentally safe, non-toxic, andpossess high stability and insulating characteristics [1]. BasaltFiber Reinforced Polymer (BFRP) reinforcing bars have beenrecently introduced as an alternative to steel reinforcement forconcrete structures and as external reinforcement for retrofittingof concrete structures. Unlike Carbon Fiber Reinforced Polymer(CFRP) and Glass Fiber Reinforced Polymer (GFRP) materials, basaltfibers have not been widely used. The limitation of their use may

be attributed to the lack of fundamental research and extensivetesting required to establish appropriate design recommendationsand guidelines. Chopped basalt fibers have been also introduced asan additive to concrete mixes to produce fiber reinforced concrete(FRC).

The research presented in this paper comprises two mainstudies. The first study evaluated the behavior of flexural con-crete members reinforced with BFRP bars. The study includedassessments of the mechanical properties and the bondstrength of two selected BFRP bars having two different sur-face deformations. The first BFRP bars were ribbed and thesecond were dented. The second study investigated the useof chopped basalt fibers as an additive to concrete mix toenhance the mechanical properties of hardened concrete. Twodifferent short basalt fiber products were investigated in thesecond study.

38 C. High et al. / Construction and Building Materials 96 (2015) 37–46

First study: flexural behavior of concrete members reinforcedwith BFRP bars

2. Background

Ramakrishnan et al. [1] investigated the use of basalt fiber barsfor reinforcing concrete members. Test results indicated that spec-imens reinforced with BFRP bars with short bond lengths exhibitedgradual slip prior to failure. Specimens with long bond lengthsexhibited sudden failures due to rupture of the BFRP bars.Patnaik [2] studied the flexural strength of 13 concrete beams rein-forced with BFRP bars and compared the measured failure loads tothose predicted by ACI 440.1R-06 guidelines. The study concludedthat prediction of moment capacities by ACI 440.1R-06 agrees wellwith the measured values. Ovitigala [3] investigated the behaviorof lightweight and normal weight concrete beams reinforced withBFRP bars. The study reported that ACI 440.1R-06 [4] predicted 77–93 percent of the measured moment capacities. In addition, thestudy reported higher deflections for BFRP-reinforced concretebeams in comparison to steel-reinforced concrete beams with thesame flexural capacity.

3. Mechanical properties of BFRP bars

A total of ten coupons, five coupons for each of the ribbed anddented bars, were tested in tension according to ASTM D7205[5]. The tension coupons had a 305 mm gripping length at eachend and a free length of 610 mm. The gripping length consistedof epoxy-filled steel pipes attached to each end of the test coupon.The elongation of the tension coupons was measured using a50 mm extensometer. The average engineering stress–strain rela-tionships of the ribbed and dented BFRP bars are shown in Fig. 1.The average measured cross-sectional area of the ribbed anddented BFRP bars is 109 mm2. The equivalent nominal diameterof both bars is approximately 12 mm. The measuredcross-sectional area and equivalent diameter of the bars weredetermined by volume water displacement according to ACI440.3R-04 [6]. It should be noted that the ribs were excluded fromthe measured area.

The tension coupons exhibited a linear stress–strain relation-ship up to rupture of the bars. The average measured moduli ofelasticity of the ribbed and dented BFRP bars were approximately48.3 GPa and 41.4 GPa, respectively. Therefore an average valueof 45 GPa can be used for the modulus of elasticity. The averagemeasured ultimate tensile strength for both bars was approxi-mately 1000 MPa. The average measured rupture strains of theribbed and dented bars were 2.2% and 2.5%, respectively.

Stre

ss (M

Pa)

20

40

60

80

100

120

0

200

400

600

800

000

200

00 1StStrain

R

ain (%

RibbBa

(%)

bedrs

2

DDentBar

tedrs

3

Fig. 1. Typical stress–strain relationship of BFRP bars.

4. Bond strength of BFRP bars

4.1. Test specimens and test setup

Beam-end specimens were tested to assess the bond character-istics of the two types of BFRP bars with different surface deforma-tions (ribbed or dented). A total of eight specimens, four specimensfor each bar type, were tested. The development length accordingto the equation provided by ACI 440.1R-06 [4] for GFRP and CFRPbars was 762 mm, which is approximately equivalent to 65 timesthe bars diameter. Accordingly, four different bond lengths wereselected for this study, 380 mm, 610 mm, 1015 mm, and1270 mm, which are equivalent to 32, 51, 85, and 106 times thebar diameter, respectively. It should be noted that the four selectedtest bond lengths were longer than and shorter than that estimatedusing ACI 440.1R-06.

The concrete beam-end specimens had a total length of1524 mm to accommodate the longest bond length. The depth ofthe specimens was 610 mm to eliminate the influence of the com-pressed concrete zone on the bonded length of the bar. A width of305 mm was used to provide enough bearing strength. The speci-mens were cast with the BFRP at the bottom position and the spec-imens were rotated prior to testing. Details of specimens areshown in Fig. 2.

All tested bars were 2134 mm long to provide an embedmentlength of 1524 mm within the specimen and an overhang lengthof 610 mm to grip the bar. A PVC pipe was used at the unloadedend to break the bond and to provide the specified bond length.A 102 mm PVC pipe was used at the loaded end to avoid possiblelocalized failure of the concrete.

Two linear potentiometers were attached to the loaded end ofthe test bar to measure the elongation of the bar. Similarly, two lin-ear potentiometers were attached to the unloaded end (free end) ofthe BFRP bar tested to measure the slip of the bar. Elongation of thetest bars was measured using a 50 mm extensometer locatedwithin the free length of the bars. The test setup is shown in Fig. 3.

4.2. Test results and discussion

Test results of the ribbed bars ‘‘R’’ and the dented bars ‘‘D’’ aregiven in Table 1 including the observed failure mode (see Fig. 4),the maximum measured stress in the bar at failure, and the mea-sured concrete compressive strength at the day of testing. Testresults indicate that the ribbed and dented BFRP bars have similarbond strengths. A bonded length of 380 mm, which is equivalent to32 times the bar diameter, was found to be sufficient to developthe full strength of the BFRP bars used in this study.

5. Flexural behavior of BFRP-reinforced members

5.1. Test specimens

Six one-way slabs reinforced with ribbed BFRP bars only weretested in flexure up to failure. The specimens were 3658 mm long

STSTEEEEL GL GRI

38

BFRIP

81

BFRP

BO

mm

RP BA

BOND

m

BAR

NDED

R

D (

2

LEN

229

NG

9 mm

1

GTH

PVm

02

H VA

PVC B

mm

ARI

C BON

m

ES

OND

S)

D BRBREA

152

DE

EAKE

24 m

EBO

KERS

mm

OND

RS

m

DEDD

661010 mmmm

Fig. 2. Details of beam-end specimens.

Fig. 3. Schematic of bond strength test setup.

Table 1Summary of test results of bond strength study.

Bonded length(mm)

Failure mode Maximum stress(MPa)

Concrete strength(MPa)

R-380 Slip/BarRupture

1048 53.6

R-610 Bar Rupture 903 48.6R-1015 Bar Rupture 1000 48.6R-1270 Bar Rupture 814 NA

D-380 Bar Rupture 910 52.1D-610 Bar Rupture 807 52.1D-1215 Bar Rupture 889 51.4D-1270 Bar Rupture 869 51.4

152.4 mm

609.6 mm

BFRP REINFORCING BARS25.4 mm CLEAR COVER

Fig. 5. Typical cross-section of flexural specimens.

C. High et al. / Construction and Building Materials 96 (2015) 37–46 39

with a cross-section of 610 mm wide and 152 mm deep as shownin Fig. 5. The BFRP bars were spaced uniformly across the width ofthe specimens and had a clear concrete cover of 25 mm. All speci-mens were tested at concrete age ranging from 58 to 63 days andthe average measured concrete compressive strength was72.1 MPa at the day of testing. The balanced reinforcement ratioof the slabs was computed as 0.47 percent.

Two duplicate specimens reinforced with three bars weredesigned to fail in tension with a reinforcement ratio of 0.44 per-cent, which is 0.94 of the balanced ratio. Two duplicate specimensreinforced with seven bars were designed to fail in compressionwith a reinforcement ratio of 1.04 percent, which is 2.20 of the bal-anced ratio .The last two duplicate specimens were reinforced withfour bars, resulting in a reinforcement ratio of 0.59 percent, whichis 1.26 of the balanced ratio. The ribbed bars were selected in thisstudy due to their higher modulus of elasticity compared to thedented bars. The measured average cross-sectional area of theribbed bars was 109 mm2.

Fig. 4. Typical failure of bean-end specimens.

5.2. Test setup and instrumentation

The flexural specimens were tested in a four-point bending con-figuration with a test span of 3353 mm using one hydraulic actua-tor. The load was applied at two locations spaced 305 mm apart.The specimens were supported by a pin support at one end and aroller support at the other end.

Two string potentiometers were used to measure the deflectionat mid-span. Five PI-gages located at the mid-span section wereused to measure the strain of concrete. Two PI-gages, 50 mm apart,were placed on the top and two PI-gages on the bottom surfaces ofthe slabs. The fifth PI-gage was placed on the side face of the spec-imens at the depth of the BFRP bars to measure the concrete strainat the reinforcement level. Fig. 6 shows the test setup and theinstrumentation of typical specimen, BR-1.

5.2.1. Mode of failureThe under-reinforced specimens (qf = 0.94qfb) failed in abrupt

manner due to the complete rupture of all the BFRP bars as shownin Fig. 7(a). The test specimens with a reinforcement ratio slightlyhigher than the balanced ratio (qf = 1.26qfb) failed due to crushingof concrete in the compression zone of the constant momentregion as shown in Fig. 7(b). Partial rupture of the BFRP bars within

Fig. 6. Test setup and instrumentation of specimen BR-1.

Fig. 7. Flexural specimens at the conclusion of the test.

40 C. High et al. / Construction and Building Materials 96 (2015) 37–46

these two specimens was visibly evident at failure. Theover-reinforced specimens (qf = 2.20qfb) failed in compressiondue to crushing of concrete on the top surface of the specimen inthe constant moment region without any evidence of rupture ofthe BFRP bars as shown Fig. 7(c).

5.2.2. Load–deflection behaviorThe load–deflection behavior of the six test slabs is shown in

Fig. 8. All specimens behaved similarly up to first cracking. Aftercracking, the flexural stiffness was proportional to the BFRP rein-forcement ratio used in each category. The behavior of the load–deflection relationship was virtually linear up to failure is due tothe linear elastic nature of the BFRP bars.

The under-reinforced slabs had the least load carrying capacitywith an average failure load of 46.3 kN and a corresponding deflec-tion of 178 mm, which is slightly less than the slabs with a rein-forcement ratio slightly higher than the balanced ratio, due tothe sudden rupture of the bars. The slabs with a reinforcementratio slightly higher than the balanced ratio had an average failureload of 60.9 kN and corresponding deflection of 201 mm. Theover-reinforced slabs failed at the highest average load of 82.6 kNand a corresponding average deflection of 155 mm.

The load–deflection behavior of the test slabs was predictedusing the equations proposed by Bischoff and Gross with andwithout consideration of the tension stiffening [7], as well as theequation recommended by ACI 440.1R-06. The predicted load–deflection behaviors of the three different approaches arecompared to the measured behavior in Fig. 9 for the six tested slabswith different reinforcement ratios. The comparison clearly indi-cates that the equation proposed by Bischoff and Gross [7] without

Fig. 8. Load–deflection behavior of flexure specimens.

tension stiffening can accurately predict the behavior up to failure.The equations recommended by ACI 440.1R-06 and Bischoff andGross [7] with tension stiffening underestimate the deflection aftercracking. The behavior of both approaches highlights that the effectof tension stiffening of concrete is significantly reduced for flexuralmembers reinforced with BFRP bars due to their low modulus ofelasticity.

5.2.3. Measured strainThe measured concrete strain at the level of the BFRP bars of the

slabs with reinforcement ratio of 0.94qfb and 1.26qfb were alwayshigher than the strain of the bars at the same stress level inducedin the bars computed by sectional-analysis of the test slabs, asshown in Fig. 10. This behavior reflects slippage of the BFRP barsfor the slabs with reinforcement ratios equal to 0.94qfb and1.26qfb due to the high stress demand of the bars at these lowlevels of reinforcement. The slip of the BFRP bars was verified byremoving of the concrete cover and inspecting the slabs. The per-formance of the over-reinforced slabs did not exhibit similar slipbehavior. This behavior highlights the necessity of designingBFRP-reinforced members to fail in compression and with higherreinforcement ratio in comparison to the typical reinforcementratios used for steel.

6. Applicability of ACI 440.1R-06 guidelines

6.1. Nominal moment capacity

Measured flexural capacities were compared to those predictedaccording to the ACI 440.1R-06 guidelines, as shown in Table 2. Thenominal moment capacity was predicted using the measured con-crete compressive strength. Table 2 clearly indicates that ACI440.1R-06 can accurately predict the nominal moment capacityof concrete members reinforced with BFRP bars having differentreinforcement ratios.

6.2. Deflection at service level

The selected moment level used to compare the deflection atservice load (Ms) was estimated using the following equation:

Ms ¼£

aLoadMn

where Mn is the measured moment of the slab at failure. aLoad wasequal to 1.33 based on a dead-to-live load ratio of 2:1. According toACI 440.1R-06, the strength reduction factor ‘‘U’’ is 0.65 for qf -P 1.4qfb and 0.55 for qf 6 1.0qfb. Therefore, estimated servicemoment for the three categories of the test slabs were 0.41Mn,0.46Mn, and 0.49Mn for reinforcement ratio of 0.94qfb, 1.26qfb, and2.20qfb, respectively. The measured mid-span deflection at serviceload level, D, was compared to that predicted according to ACI440.1R-06 [4] for all test slabs as given in Table 3.

Serv ice

0

10

20

30

40

50

60

70

80

90

0 40 80 120 160 200 240

App

lied

Load

(kN

)

Midspan Deflection (mm)

S p e c i m e n U R 1S p e c i m e n U R 2B i s c h o f f ( W i t h o u t T e n s i o n S t i f f e n i n g )A C I 4 4 0 . 1 R - 0 6B i s c h o f f ( W i t h T e n s i o n S t i f f e n i n g )

Serv ice

0

10

20

30

40

50

60

70

80

90

0 40 80 120 160 200 240

App

lied

Load

(kN

)

Midspan Deflection (mm)

S p e c i m e n B R 1S p e c i m e n B R 2B i s c h o f f ( W i t h o u t T e n s i o n S t i f f e n i n g )A C I 4 4 0 . 1 R - 0 6B i s c h o f f ( W i t h T e n s i o n S t i f f e n i n g )

Serv ice

0

10

20

30

40

50

60

70

80

90

0 40 80 120 160 200 240

App

lied

Load

(kN

)

Midspan Deflection (mm)

S p e c i m e n O R 1S p e c i m e n O R 2B i s c h o f f ( W i t h o u t T e n s i o n S t i f f e n i n g )A C I 4 4 0 . 1 R - 0 6B i s c h o f f ( W i t h T e n s i o n S t i f f e n i n g )

Fig. 9. Prediction of load–deflection behavior of test slabs.

C. High et al. / Construction and Building Materials 96 (2015) 37–46 41

Table 3 shows that the ACI 440.1R-06 equation significantlyunderestimates the deflection at service load for members rein-forced with BFRP reinforcement ratios less than or approximatelyequal to the balanced ratio due to slippage of BFRP bars.However, the ACI 440.1R-06 predictions improve as the BFRP rein-forcement ratio is increased as the ratio of measured to predicted

deflection reduced from 2.19 to 1.22. This is due to the reductionof stresses in the BFRP bars and therefore, possible elimination ofslippage of the bars. Table 3 also indicates that the measureddeflection-span ratio (L/D) at service load level significantlyexceeds the permissible deflection limit under total service loadas recommended by ACI 318-11 [8]. This indicates that the designof flexural members reinforced with BFRP may also be controlledby serviceability requirements, due to the low modulus of elastic-ity of the bars.

Second study: basalt fiber-reinforced concrete (BFRC)

7. Background

Ramakrishnan et al. [1] investigated the use of short basaltfibers to enhance the material properties of concrete. The studyconcluded that basalt fibers can be easily mixed with concretewithout any balling or segregation. In addition, there was also anoticeable increase in the post-cracking energy absorption capac-ity and increase of the impact resistance.

Ma et al. [9] conducted an experimental program to investigatethe mechanical properties of concrete with the addition of basaltfibers that were pre-soaked in epoxy. Three different pre-soakedbasalt fiber lengths (10 mm, 20 mm and 30 mm) and three differ-ent fiber dosages (3000 g/m3, 5000 g/m3 and 7000 g/m3) wereexamined in this study. Test results showed that as the basalt fiberdosage and fiber length increased the measured slump decreased.According to this study, the presence of the pre-soaked basaltfibers did not significantly affect the compressive strength of theconcrete, however, use of fibers increased the concrete flexuralmodulus. This study concluded that adding 30 mm long basaltfibers to concrete at a dosage range of 3000 g/m3 to 5000 g/m3

resulted in improvement of the mechanical properties with anacceptable workability.

Borhan [10] studied the compressive and splitting tensilestrengths of BFRC with fiber volume fractions ranging from 0.1 to0.5 percent. Test results indicated that increasing the basalt fibercontent increased the splitting tensile strength of the concreteand did not affect the compressive strength, up to volume fractionsof 0.3 percent. Decreased compressive and splitting tensilestrengths were reported for fiber volume content equal to 0.5 per-cent. This study also reported a reduction in the concrete slump asthe basalt fiber volume content was increased.

8. Properties of basalt fibers

Two different short basalt fiber products were used in thisstudy. The first type consisted of chopped dry fibers, while the sec-ond type was produced by chopping precured fibers as shown inFig. 11. The first type of fiber is denoted as ‘‘D’’ for dry and the sec-ond type as ‘‘P’’ for precured. The properties of the two types offibers are given in Table 4.

9. Concrete mixes and specimens

Two concrete mixes were investigated in this study. Mix ‘‘A’’was a standard concrete mix with a target compressive strengthat 28 days of 20.7 MPa. Mix ‘‘B’’ had the same target compressivestrength, however contained fly ash and admixtures and therefore,less cement content. Details of the two mixes are given in Table 5.

For each mix, different contents of basalt fibers were used. Thetest matrix of the different batches is given in Table 6. Batches ‘‘1A’’and ‘‘1B’’ are control batches without any fibers. The two differenttypes of fiber were used together in different ratios. In order toproduce uniform batches, dry mixing of the constituents wasperformed prior to adding the water until blending of the

Fig. 10. Measured and computed strains in the BFRP bars.

Table 2Predictions of nominal moment according to ACI 440.1R-06 (kN-m).

Slab ID Rft. Ratio (qf) Measured ACI 440 Measured/ACI 440

Mn AVG

UR-1 0.94qfb 39.2 38.5 39.0 0.99UR-2 37.8BR-1 1.26qfb 48.2 49.4 51.4 0.96BR-2 50.5OR-1 2.20qfb 67.0 66.0 65.9 1.00OR-2 64.9

Table 4Properties of basalt fibers.

Property Dry fiber Precured fibers

Density 2.8 gm/cm3 1.9 gm/cm3

Diameter 13–20 lm 2.6 mmLength 24 mm 40 mmElastic modulus 89 GPa 43 GPaElongation at break 3.15% 2.20%

42 C. High et al. / Construction and Building Materials 96 (2015) 37–46

constituents was evident. Afterwards, the water containing all liq-uid admixtures was gradually added and the concrete was allowedto mix until a uniform consistency was achieved in a reasonabletime. Workability of all batches for mix ‘‘A’’ was higher than thatof mix ‘‘B’’, despite the use of admixtures in mix ‘‘B’’. This couldbe attributed to the high water/cement ratio used in mix ‘‘A’’,which was 0.55 compared to 0.38 used for mix ‘‘B’’.

Concrete cylinders of 102 � 204 mm and prisms of152 � 152 � 508 mm were cast from each of the concrete batches.Neither rodding nor internal vibration was performed during thecasting to avoid the disturbance of the distribution of the basaltfibers. However, the overfilled molds were vibrated using a

Table 3Predictions of deflection at service load according to ACI 440.1R-06 (kN-m and mm).

Slab ID Rft. Ratio (qf) Service moment Expe

Ms Ms/Mcr D

UR-1 0.94qfb 16.2 1.30 51UR-2 15.6 1.25 40BR-1 1.26qfb 22.3 1.78 75BR-2 23.3 1.86 69OR-1 2.20qfb 32.6 2.61 67OR-2 31.6 2.53 65

Fig. 11. Chopped Basalt fibers used for reinforc

vibration table until the concrete was adequately consolidated.The casted molds remained inside the laboratory at a constantambient temperature of 22 ± 2 �C and relative humidity of70 ± 5%. The cylinders and the prisms were de-molded 24 h aftercasting and cured by covering them with continuously wet burlapand plastic sheets to entrain the moisture until the designated testdate.

10. Test results and discussion

10.1. Compressive strength

The compressive strength of each batch was measured at 3, 7and 28 days in accordance with ASTM C39 [11]. Average measured

rimental ACI 440

DAVG L/D D DAVG Ratio

46 73 22 21 2.1920

72 47 41 44 1.6447

66 51 55 54 1.2252

ing concrete (tape measure shows inches).

Table 5Details of mix designs.

Material Quantity

Mix design A Mix design B

Cement 310 kg/m3 220 kg/m3

Water 170 kg/m3 140 kg/m3

Fine aggregate 750 kg/m3 700 kg/m3

Coarse aggregate 990 kg/m3 1015 kg/m3

Fly Ash-Class C NA 145 kg/m3

AdmixturesAir-entraining 0.1035 L 0.1331 LWater-reducing NA 0.5028 LAccelerating NA 1.2Mineral NA 84 kg/m3

Table 6Test matrix of batches.

Mix Batch ID Fiber content (g/m3) P:D ratio

Type ‘‘D’’ Type ‘‘P’’

A 1A None None NA2A 1186 593 1:23A 1779 None 0:34A 1186 1186 2:2

B 1B None None NA2B 1186 593 1:23B 1779 None 0:34B 1186 1186 2:2

Table 7Test results of Basalt FRC.

BatchID

Compressive strength (MPa) Flexuralmodulus(MPa)

Averageresidualstrength(MPa)

3 days 7 days 28 days (f 0c) fr Ratio

fc Ratio fc Ratio f 0c Ratio

1A 14.6 1.00 17.0 1.00 23.1 1.00 3.4 1.00 NA2A 14.6 1.00 18.9 1.12 23.4 1.01 3.6 1.06 0.153A 15.4 1.05 19.4 1.14 24.4 1.06 4.0 1.16 0.204A 10.7 0.73 14.1 0.83 18.4 0.80 3.5 1.02 0.20

1B 11.3 1.00 14.1 1.00 21.2 1.00 3.0 1.00 NA2B 14.7 1.31 18.3 1.30 25.4 1.20 3.9 1.31 0.253B 14.3 1.27 18.1 1.28 24.4 1.15 3.5 1.18 0.154B 18.6 1.65 22.8 1.62 29.7 1.40 4.0 1.33 0.15

”A“xiM

67

63 64

61

6061 62

54

63 63 63

58

50

60

70

80

90

1A (0:0) 2A (1:2) 3A (0:3) 4A (2:2)

Perc

ent o

f fc'

Batch (ratio of fiber type P:D)

Fig. 12. Compressive strength at

C. High et al. / Construction and Building Materials 96 (2015) 37–46 43

compressive strengths, fc, at different ages for the two mixes withdifferent fiber content are given in Table 7. The reported strengthat ages of 3 and 7 days is an average of three individual tests, whilethe 28-day reported strength is an average of six individual tests.The ratios of the measured strength of the different batches tothe strength of the control batches 1A and 1B are also given inTable 7.

The early compressive strength at ages of 3 and 7 days of mixes‘‘A’’ and ‘‘B’’ as percent of the 28-day strength (f 0c) are graphicallyshown in Figs. 12 and 13, respectively. The early strength of mix‘‘A’’ did not exhibit an increase due to the use of the basalt fibers.However, there was a trend of increase of the early compressivestrength for mix ‘‘B’’ reflected by a ratio up to 65% and 77% forbatch ‘‘4B’’ at 3 and 7 days, respectively.

The measured compressive strength for all concrete batchesexceeded the target 28-day compressive strength of 20.7 MPa, withthe exception of batch ‘‘4A’’. Test results indicated that the con-crete compressive strength at 28 days (f 0c) of mix ‘‘A’’ was not sig-nificantly affected by the use of the basalt fibers with a maximumincrease of 6% for batch ‘‘3A’’ as compared to the control batch, 1A.However, the effect of the basalt fibers on fc

’ was evident for mix‘‘B’’ in comparison to mix ‘‘A’’. The 28-day compressive strengthof batch ‘‘4B’’ was increased by 40% in comparison to the controlbatch ‘‘1B’’. The increase of f 0c of mix ‘‘B’’ is also proportional tothe high fibers content used of both types of chopped fibers.

Test results indicate that adding basalt fibers to concrete mayslightly increase the 28-day compressive strength (f 0c) of concretecontaining fly ash and admixtures with low water-cement ratios.In addition, test results suggest that the early strength of concretecontaining fly ash and admixtures may be significantly increasedby using basalt fibers.

10.2. Modulus of rupture

The modulus of rupture at 28 days was measured using athird-point loading test in accordance with ASTM C78 [12].Flexural prisms of 152 � 152 � 508 mm were tested using a uni-versal testing machine and an apparatus for third-point loadingas shown in Fig. 14.

The average measured modulus of rupture for the two mixeswith different fiber content is given in Table 7. The reported mod-ulus of rupture is an average of three individual tests. It is evidentfrom test results that using basalt fibers could increase the flexural

”B“xiM

54

59 60

64

52

57 57

61

53

58 59

63

50

60

70

80

90

1B (0:0) 2B (1:2) 3B (0:3) 4B (2:2)

Perc

ent o

f fc'

Batch (ratio of fiber type P:D)

3 days as a percentage of f 0c .

Mix “A” Mix “B”

76

8381

79

71

79 78

73

73

8179

77

50

60

70

80

90

1A (0:0) 2A (1:2) 3A (0:3) 4A (2:2)

Perc

ent o

f fc'

Batch (ratio of fiber type P:D)

68

7376

77

65

7172

76

67

7274

77

50

60

70

80

90

1B (0:0) 2B (1:2) 3B (0:3) 4B (2:2)

Perc

ent o

f fc'

Batch (ratio of fiber type P:D)

Fig. 13. Compressive strength at 7 days as a percentage of f 0c .

Fig. 14. Flexural test specimen prior to testing.

Fig. 16. ARS specimen prior to the initial loading.

44 C. High et al. / Construction and Building Materials 96 (2015) 37–46

modulus of the concrete for both mixes. However, the increase wasmore pronounced for mix ‘‘B’’ in comparison to mix ‘‘A’’. Thisbehavior may be due to the effectiveness of mixing basalt fiberswith concrete mixes containing fly ash and admixtures to reducethe water-cement ratio.

Fig. 15. Normalized flexu

ACI 318-11 code [8] relates the modulus of rupture (fr) and thecompressive strength (f 0c) of normal weight concrete as:

ral strength of BFRC.

Fig. 17. Normalized average residual strength of BFRC.

C. High et al. / Construction and Building Materials 96 (2015) 37–46 45

f r ¼ 0:62�ffiffiffiffif 0c

q

The measured modulus of rupture was normalized by the squareroot of the 28-day compressive strength (f 0c) as shown in Fig. 15.The normalized flexural strengths reveal that the modulus of rup-ture of concrete containing basalt fiber is higher than that of normalweight concrete without fibers. The increase in the rupture strengthfor mix ‘‘A’’ was proportional to the fiber content and on averagewas slightly higher than that for mix ‘‘B’’.

10.3. Average residual strength (ARS)

The toughness (or capability to resist crack opening) of the BFRCwas evaluated using the average residual strength (ARS) testaccording to ASTM C1399 [13]. ARS provides a measure of thepost-cracking strength of the concrete; as such strength may beaffected by the use of fiber-reinforcement. The test method wasdeveloped particularly for FRC with low fiber content to avoidthe instability of tested beams at the onset of cracking, as typicallyseen during the withdrawn ASTM C 1018 test method [14]. ARStesting was performed using prisms of 100 � 100 � 350 mm,which were cut from the 152 � 152 � 508 mm prisms using anabrasive concrete saw approximately 24 h before testing. Fig. 16shows an ARS specimen positioned in the testing apparatus priorto testing.

The average measured ARS for the two mixes with differentfiber content is given in Table 7. The reported values are an averageof five individual tests. ARS test values were normalized by thesquare root of the 28-day compressive strength (f 0c) as shown inFig. 17. The ARS values of the tested BFRC specimens appear tobe low compared to other studies of FRC using different type offibers [15]. Based on these limited tests, the results indicate thatthe use of basalt fibers did not enhance the average residualstrength of concrete.

11. Summary and conclusions

The research program presented in this paper comprises twostudies. The first study evaluated the flexural behavior of concretemembers reinforced with BFRP bars. As part of the first study, themechanical properties and bond strengths of two BFRP bars wereinvestigated. The applicability of ACI 440.1R-06 design guidelinesfor predicting the deflection and strength of BFRP-reinforced

concrete members was discussed. The second study investigatedthe effect of using two different types of chopped basalt fibers toenhance the characteristics of concrete.

Based on the test results, the following conclusion can be made:

1. The ribbed and dented BFRP included in this study had an aver-age ultimate tensile strength of approximately 1000 MPa withan average modulus of elasticity of 45 GPa. The bond strengthof both BFRP bars is essentially the same and the developmentlength is approximately equal to 32 times the bar diameter,which is significantly less than the predicted developmentlength using the ACI440.1R-06 equation.

2. Slippage of the BFRP bars could occur for low reinforcementratio in the range of the balanced ratio. Slippage of the barscan be avoided by using high reinforcement ratio of at leastdouble the balanced reinforcement ratio. This behavior can beattributed to the high stress demand on the bars of specimenswith low reinforcement ratios. This behavior also highlightsthe necessity for designing BFRP-reinforced flexural membersto fail in compression.

3. Design of flexural members reinforced with BFRP bars may becontrolled by serviceability requirements due to the low modu-lus of elasticity of the bars.

4. ACI 440.1R-06 accurately predicts the nominal moment capac-ity of flexural members reinforced with BFRP bars. ACI440.1R-06 significantly underestimates the deflection at serviceload for under-reinforced and balanced reinforcement ratios.However, the ACI 440.1R-06 deflection prediction improves asthe BFRP reinforcement ratio is increased.

5. The equation proposed by Bischoff and Gross [7] without ten-sion stiffening can accurately predict the deflection of flexuralmembers reinforced with BFRP bars up to failure. The effect oftension stiffening of the concrete is negligible for flexural mem-bers reinforced with BFRP bars due to their low modulus ofelasticity.

6. Using basalt fibers slightly increased the 28-day compressivestrength (f 0c) of concrete containing fly ash and admixtures witha low water-cement ratio. In addition, the early compressivestrength of concrete containing fly ash and admixtures may sig-nificantly increase due to the use of basalt fibers.

7. Use of basalt fibers resulted in an increase of the modulus ofrupture of the concrete. However, the increase in the flexuralstrength was more pronounced for concrete mix containingfly ash, admixtures, and with low water-cement ratio.

46 C. High et al. / Construction and Building Materials 96 (2015) 37–46

Acknowledgments

The authors would like to acknowledge the Florida Departmentof Transportation for funding this project and the National ScienceFoundation Center of Integration of Composites into Infrastructure(CICI) at NCSU, Grant No. 2009-1644. Thanks are also due to thestaff of the Constructed Facilities Laboratory of NCSU for their helpthroughout the experimental program.

References

[1] V. Ramakrishnan, Neeraj S. Tolmare, Vladimir B. Brik, Performance Evaluationof 3-D Basalt Fiber Reinforced Concrete & Basalt Rod Reinforced Concrete. FinalReport for Highway IDEA Project 45, Transportation Research Board, 1998, pp.79.

[2] A. Patnaik, S. Adhikari, P. Bani-Bayat, P. Robinson, Flexural performance ofconcrete beams reinforced with basalt FRP bars, in: 3rd fib. InternationalCongress, Washington, D.C., May 2010, pp. 3821–3833.

[3] T. Ovitigala, M.A. Issa, Flexural behavior of concrete beams reinforced withbasalt fiber reinforcement polymer (BFRP) Bars, in: 11th InternationalSymposium on Fiber Reinforced Polymer for Reinforced Concrete Structures,Guimarães, Portugal, June 2013, pp. 249–260.

[4] ACI 440.1R-06, Guide for the Design and Construction of Structural ConcreteReinforced with FRP Bars. ACI Committee 440, American Concrete Institute,Farmington Hills, MI, USA, 2006, pp. 44.

[5] ASTM D7205/D7205M-06 (Reapproved 2011), Standard Test Method forTensile Properties of Fiber Reinforced Polymer Matrix Composite Bars, ASTMInternational, West Conshohocken, PA, USA, 2011, pp. 13.

[6] ACI 440.3R-12, Guide Test Methods for Fiber-Reinforced Polymer (FRP)Composite for Reinforcing or Strengthening Concrete and MasonryStructures. ACI Committee 440, American Concrete Institute, FarmingtonHills, MI, USA, 2012, pp. 40.

[7] Peter H. Bischoff, Shwan P. Cross, Equivalent moment of inertia based onintegration of curvature, ASCE – J Compos. Constr. 15 (3) (2011) 263–273.

[8] ACI 318–11. Building Code Requirements for Structural Concrete andCommentary. ACI Committee 318, American Concrete Institute, FarmingtonHills, MI, USA, 2011, pp. 503.

[9] Jianxun Ma, Xuemei Qiu, Litao Cheng, Yunlong Wang, Experimental researchon the fundamental mechanical properties of presoaked basalt fiber concrete,in: Lieping Ye, Peng Feng, Qingrui Yue (Eds.), Advances in FRP Composites inCivil Engineering, Springer, Berlin, Heidelberg, 2011, pp. 85–88, <http://link.springer.com/chapter/10.1007/978-3-642-17487-2_16>.

[10] Tumadhir M. Borhan, Thermal and mechanical properties of basalt fibrereinforced concrete, Proc. World Acad. Sci., Eng. Technol. 76 (2013) 313.

[11] ASTM C39/C39M-12a, Standard Test Method for Compressive Strength ofCylindrical Concrete Specimens, ASTM International, West Conshohocken, PA,USA, 2012, pp. 7.

[12] ASTM C78/C78M-10, Standard Test Method for Flexural Strength of Concrete(Using Simple Beams with Third-Point Loading), ASTM International, WestConshohocken, PA, USA, 2010, pp. 4.

[13] ASTM C1399/C1399M-10. Standard Test Method for Obtaining AverageResidual-Strength of Fiber-Reinforced Concrete. ASTM International, WestConshohocken, PA, USA, 2010, pp. 6.

[14] N. Banthia, J.F. Trottier, Test methods for flexural toughness characterization offiber-reinforced concrete: some concerns and a proposition, ACI Mater. J. 92(2) (1995) 48–57.

[15] Jean-Francois Trottier, Michael Mahoney, Dean Forgeron, Can synthetic fibersreplace welded-wire fabric in slabs-on-grade, Concr. Int. 24 (11) (2002) 59–68.