Effect of coar aggregate characteristics on concrete rheology
Jiong Hu a ,⇑,Kejin Wang b
a Department of Engineering Technology,Texas State University –San Marcos,San Marcos,TX 78666,United States
b
Department of Civil,Construction,and Environmental Engineering,Iowa State University,Ames,IA 50011,
United States
a r t i c l e i n f o Article history:
Received 11August 2009
Received in revid form 16August 2010Accepted 2September 2010
Available online 26September 2010Keywords:Aggregate Gradation Proportion Voids Rheology
a b s t r a c t
In the prent study,concrete was considered as a two-pha material,consisting of coar aggregate (CA)and mortar.Coar aggregate properties were characterized by fineness,uncompacted void and fric-tion angle.The combined effects of CA characteristics and mix design on the rheological properties of the corresponding concrete were investigated using a portable IBB concrete rheometer.Experimental results indicated that a higher CA and fine aggregate content normally result in higher concrete rheological parameters (yield stress and viscosity).For a given type and amount of mortar,concrete yield stress and viscosity generally increa with the uncompacted void content and friction angle but decread with the size (or fineness)of CA.Well gra
ded CA,generally having low uncompacted void content,pro-vides concrete with considerably reduced yield stress and viscosity when compared with single-sized CA.In addition,a multiple-parameter linear regression analysis was conducted to evaluate how different CA characteristics (fineness,uncompacted void and friction angle)and mix design parameters (mortar com-position,and CA volume fraction)affect concrete rheological behavior.
Ó2010Elvier Ltd.All rights rerved.
1.Introduction
Aggregate characteristics,such as size,gradation,shape,surface texture and volume fraction,all have significant effects on concrete rheology [1–3].The effects result from the aggregate interparti-cle forces (such as interlocking and friction among solid particles)and the particle movement in the liquid phas of fresh concrete [4–6].Geiker et al.have shown that the relative yield stress and viscosity of concrete significantly increa with incread coar aggregate (CA)volume fraction [7].The water requirement for con-crete decreas with incread aggregate particle size.Very fine aggregate requires more water for a given consistency.An optimal aggregate gradation provides a higher degree of packing and re-quires less paste to reach a given consistency since less cement paste is
needed to fill the space among the aggregate [8–10].Pre-vious rearch also indicates that friction among aggregate has a significant contribution to concrete rheology [11].Particles with a nearly spherical shape and a smooth surface texture provide more workable concrete.However,compared with the study of ce-ment paste and mortar,the study of concrete rheology is still lim-ited due to the difficulties in characterizing concrete aggregate and the limited equipment available for concrete rheology measure-ments.Very few aggregate parameters are applied in concrete rhe-ology study.
In the prent study,concrete was considered as a two-pha material,compod with mortar and CA.The two-pha approach can not only reduce the error of analysis caud from the wide range of aggregate size,but also provide practical advantages in concrete mix design since fine aggregate (FA)and CA are usually proportioned parated [12].In addition to proportions of concrete,CA properties were studied.The CA was characterized by grada-tion,uncompacted void and aggregate friction angle tests.The ef-fects of concrete material properties (CA characteristics)as well as mix design parameters (mortar composition and CA content)on concrete rheological behavior were studied.A multi-parameter linear regression analysis was conducted to study effects of differ-ent material and mix design parameters on concrete rheology.Bad on the regression analysis,the degree of importance of the aggregate properties and mix design parameters in concrete rheol-ogy were evaluated.2.Rearch significance
黄帝的儿子
Aggregate characteristics and content significantly influence rheology of concrete.Rational characterization of concrete aggre-gate has been challenging.In the prent paper,test methods for characterizing coar aggregate were explored,and the aggregate property parameters obtained from the tests were further ud to quantify the influence of aggregate (properties and content)on concrete rheology.Bad on the experimental results,a statisti-cal analysis was ud to evaluate the effect of original mix design and aggregate properties on rheological properties of concrete.朝鲜核
0950-0618/$-e front matter Ó2010Elvier Ltd.All rights rerved.doi:10.buildmat.2010.09.035
Corresponding author.Tel.:+15122456328;fax:+15122453052.
E-mail address:jiong.hu@txstate.edu (J.Hu).
The test and analysis results can provide rearchers and engi-neers with uful tools to evaluate and predict the effects of aggre-gate on concrete rheology.3.Experimental work 3.1.Material properties
ASTM Type I cement was ud as a binder in prent study,and its chemical composition and physic
al properties are listed in Table 1.Natural graded river sand with a fineness modulus (FM)of 2.92was ud as fine aggregate (FA).The absorption of FA was 1.60%,and specific gravities were 2.59and 2.63at the oven-dried (OD)and saturated surface dry (SSD)condition respectively.Crushed limestone with a 25mm (1in.)normal maximum size of aggregate (NMSA)was ud as CA.As en in Fig.1,three CA gradations (G1,G2,and G3)were employed,where G1and G3are the high and the low limits of ASTM C33‘‘Standard Specification for Concrete Aggre-gates”and G2is the middle point gradation between G1and G3.In addition,four single-sized CAs,retained on the 19.0mm (3/4in.),12.5mm (1/2in.),9.5mm (3/8in.),and 4.75mm (no.4)sieve but pasd the sieve one size higher than the specified sieve,were also ud.The specific gravity of the CA was 2.53at SSD condition and 2.45at OD condition.Absorption of the CA varied from 2.76%to 3.77%,depending on the aggregate particle sizes.The uncompacted void content and friction angle of the CA was also measured,the test procedures and results will be described later.3.2.Mix proportions
贽礼Different mortar proportions,CA gradations,and CA volume fractions (Vca)as shown in Table 2were considered in the concrete mix design.A total of 23concrete mixes with three different mortar proportions (M1:s /c =1.75,w /c =0.45;M2:s /c =2.21,w /c =0.45;and M3:s /c =2.60,w /c =0.50),three CA contents (Vca =35%,38%and 41%),and ven CA gradations (four single sizes at 19mm,12.5mm,
9.5mm and 4.75mm and three gradations at G1,G2and G3)were prepared.Note that the mortar proportions were orig-inally designed with the same water-to-cement rations (w /c )but different sand-to-cement ratios (s /c ).However,mixes with the highest s /c (M3)was chon to have slightly higher w /c than mixes with lower s /c so as to achieve acceptable flow ability.3.3.Mixing procedure
The concrete was mixed using a pan mixer bad on ASTM C192,‘‘Standard Practice for Making and Curing Concrete Test Specimens in the Laboratory”.CA and FA both at the SSD condi-tions and tap water at 23±2°C (73±3°F)were ud.3.4.Aggregate property measurement
CAs ud in prent study were sieved and recombined to ob-tain the designed gradation as described earlier.The uncompacted void content tests were performed according to ASTM C29‘‘Stan-dard Test Method for Bulk Density (‘‘Unit Weight”)and Voids in
Aggregate”.The void contents of CAs were calculated according to the mass of the aggregate required to fill a container of a spec-ified unit volume.Generally,angularity increas void contents while well-graded aggregate decreas void content.The void con-tent between aggregate particles affects paste and mortar require-ments in mix design.While a higher void content of aggregate usually requires more paste and mortar to provide concrete with the same workability,with the same mix design,aggregate with higher void content usually results in concrete with lower work-ability [3].
A simple method was also developed to estimate the friction angle of CA bad on a basic soil mechanics concept.Using the infi-nite slope stability analysis for dry conditions,the angle of repo at limit equilibrium conditions is equal to the angle of internal fric-tion of the material forming the slope [13,14].The maximum angle formed by the particle pile (i.e.,angle of repo)can be considered as a constant and correlated to the friction angle of the particles [15,16].ASTM C1444,‘‘Standard Test Method for Measuring the Angle of Repo of Free-Flowing Mold Powers,”also describes a friction angle test bad on the measurement of the angle of re-po.With the same concept,a ries of tests was conducted in prent study to measure the friction angle of various air-dry CAs using slope stability test,the friction angles of tested CAs were esti-mated from looly-falling aggregate piles that formed a maximum slope.
As shown in Fig.2,a piece of paper was marked with a ries of circles,up to one meter (40in.)in diameter,and placed on a ground ba.CA samples (18–36kg,or 40–80lb,depending on the need for forming a maximum angle)were slowly poured onto the ground ba from a given height that was kept approximately 1cm (3/8in.)above the formed cone.The height was lected in order to form a pile with a maximum angle under least disturbing.A cone-shaped aggregate pile was gradually formed becau of the internal friction angle of particles.When the pile reached a height that no slop
e change could be visualized as more aggregate was added onto the pile (usually about 20–30cm or 8–12in.),the test was stopped.The slope of the aggregate pile was calculated from the diameter and height of the cone and defined as the friction an-
Table 1
Chemical composition and physical properties of cement.Composition
CaO
SiO 2Al 2O 3
Fe 2O 3
MgO K 2O Na 2O
羽绒服应该怎么洗
(Na 2O)eq.a
SO 3LOI b (%)64.220.8
5.55 2.25 1.91
0.50
0.190.52 2.96
0.82
Mean size =23.7l m
Fineness =399m2/kg Specific gravity =3.15
a (Na 2O)eq.=(Na 2O)+0.658(K 2O).b
LOS =loss of ignition.
J.Hu,K.Wang /Construction and Building Materials 25(2011)1196–12041197
Table 2
家庭教育理念Concrete mix proportions.
Mixture no.
Mixture ID
Cement,
kg/m 3(lb/yd 3)Water,kg/m 3(lb/yd 3)FA,kg/m 3(lb/yd 3)CA,kg/m 3(lb/yd 3)s /c w /c Mortar Vca (%)CA
酒香不怕巷子深
gradation/size Single-sized
1M1–Vca35%–19.0mm 446(752)202(340)781(1316)868(1463) 1.750.45M13519.0mm 2M1–Vca35%–4.75mm 446(752)202(340)781(1316)868(1463) 1.750.45M135 4.75mm 3M1–Vca41%–19.0mm 402(678)181(305)703(1185)1012(1706) 1.750.45M14119.0mm 4M1–Vca41%–4.75mm 402(678)181(305)703(1185)1012(1706) 1.750.45M141 4.75mm 5M2–Vca41%–19.0mm 352(593)1
58(266)777(1310)1013(1707) 2.210.45M24119.0mm 6M2–Vca41%–12.5mm 352(593)158(266)777(1310)1013(1707) 2.210.45M24112.5mm 7M2–Vca41%–9.5mm 352(593)158(266)777(1310)1013(1707) 2.210.45M2419.5mm 8M2–Vca41%–4.75mm 352(593)158(266)777(1310)1013(1707) 2.210.45M241 4.75mm Graded
9M1–Vca35%–G1446(752)202(340)781(1316)868(1463) 1.750.45M135G110M1–Vca35%–G2446(752)202(340)781(1316)868(1463) 1.750.45M135G211M1–Vca35%–G3446(752)202(340)781(1316)868(1463) 1.750.45M135G312M1–Vca38%–G2425(716)191(322)744(1254)937(1579) 1.750.45M138G213M1–Vca41%–G1402(678)181(305)703(1185)1012(1706) 1.750.45M141G114M1–Vca41%–G2402(678)181(305)703(1185)1012(1706) 1.750.45M141G215M1–Vca41%–G3402(678)181(305)703(1185)1012(1706) 1.750.45M141G316M2–Vca35%–G2390(657)177(298)862(1453)867(1461) 2.210.45M235G2Fig.2.Coar aggregate friction angle test.
Concrete
H-shape impeller
Speed 1
Speed 2
253mm
356mm
1 in. = 25.4 mm
(a)(b)
1198
J.Hu,K.Wang /Construction and Building Materials 25(2011)1196–1204
(a)revolution around the main shaft of the rheometer at speed I, and(b)rotation around the axis of the H-shaped impeller at speed II,where speed II was approximately2.20times of speed I(Fig.3b).
A load cell measured the reaction torque from the impeller,while a tachometer measured the rotation speed of the impeller.
Procedure as shown in Fig.3a was ud in rheological measure-ment for all concrete mixes in prent study.In a test,in order to obtain a uniform sample,the concrete sample was pre-sheared at the impeller speed(speed I)of approximately0.2rev./s for30s. The impeller was then stopped for30s,and a mallet was ud to gently strike the side of the container for10times within this per-iod.After the rest period,the sample was subjected to a controlled-rate hysteresis loop,where the shear rate was incread from0to1 (rev./s)over a100-s period and subquently decelerated from1to 0(rev./s)over another100-s period.For simplification,the shear rates ud in this paper refer to the impeller speed of the main shaft of rheometer(speed I).
It is generally agreed that during a rheology test,the attractions and agglomerations of the particles in the tested sample are broken down during the period of the incread shear rate(up curve of the test result).After the attractions and agglomerations of the particles in the tested sample are fully broken down,the rheological behavior of the tested material generally agrees with the Bingham model,as shown by the down curve of the test result[17].Therefore,the linear portion of the down curve of a test result is commonly ud to calcu-late the Bingham parameters,yield stress and viscosity,of the tested material.The yield stress was determined by extending the linear portion of the down curve to the y-axis,which indicated the mini-mum stress required for a material to start aflow or
deformation [18].The viscosity was determined by the slope of the down curve, which is described as the resistance of the test material toflow under the increa of the shear rate[17].
Due to the complicatedflow pattern applied to the tested con-crete by the impeller(Fig.3b),it is difficult to obtain the exact va-lue of shear stress and shear rate of the tested concrete[3,19]. Therefore,the torque and the speed of the impeller were reported during the concrete rheology tests and their relationship was plot-ted.A typical IBB test result is shown in Fig.4b.The following equation was ud tofit the down curve from an IBB rheology test bad on Bingham model:
T¼GþHÂNð1Þwhere T was the torque acting on the impeller,and N was the rota-tion speed of the impeller.G was the interception of the linear por-tion of an IBB down curve(with the rotation speed in the range of 1–0.04sÀ1)and the y-axis.H was the slope of the linear portion of the IBB down curve.According to the report from ACI committee 238[20],although the device does not allow direct calculation of yield stress and plastic viscosity in fundamental units due to the complicatedflow pattern,the down curves of the torque-rotation speed were found tofit well with Bingham pattern.The slope(H) and interception(G)obtained from IBB measurements were consid-ered to be directly proportional to yield stress and viscosity of con-crete respectively.In the prent paper,G is therefore ud to describe the yield stress and H is ud to describe the plastic viscos-ity of the tested concret
e.The reliability of the rheology test was evaluated by repeated tests.Three batches of concrete with one same mix design were prepared at the early state of rearch,the rheology test were performed on each of the mixes.The coeffi-cients of variation(COV)of the yield stress and viscosity parameters were calculated respectively by dividing the standard deviation by mean.Results showed that the COV of the rheology parameters from the concrete rheology test are below15%.
4.Test results and discussions
4.1.Aggregate properties
Uncompacted void content and friction angles,together with FM,of different CAs are listed in Table3.The results showed that, as expected,graded aggregates generally had lower void content (44.48–45.53%)than the single-sized aggregates(47.06–49.89%). Larger single-sized aggregates had a larger friction angle,and no significance difference in friction angles was obrved among three graded CAs ud in prent study.
4.2.Concrete properties
Slump,and rheological parameters bad on linear regression from down curves from rheology mea
surements of all23different mixtures are summarized in Table4.Results showed that the R2 values from linear regressions of down curves from rheological measurements were all higher than0.92.The high coefficient of determination indicated great consistency of theflow pattern with Bingham model.The rheological parameters(interception and vis-cosity)from rheology tests of the concrete mixtures are discusd in the following ctions.
4.3.Effects of mortar and CA content
Rheological parameters of concrete made with same CA grada-tion(G2)but different mortars and CA contents were summarized in Fig.5.Results show that for a given mortar,both yield stress (interception,G)and viscosity(slope,H)of the corresponding con-crete incread with the CA content.This is becau more CA par-ticles were in the concrete and less mortar was available to coat them for a betterflow.
J.Hu,K.Wang/Construction and Building Materials25(2011)1196–12041199
It is generally agreed that yield stress and viscosity of the mor-tars incread with s /c ,which was primarily due to the fact that the high friction between fine aggregate particles caud by the less paste available to coat the FA particles [20,21].It can be obrved in the Fig.5that the degrees of the increas in concrete rheolog-ical parameters,especially yield stress,were affected by the mortar properties.For the mortar having low sand content (s /c =1.75),the corresponding concrete had low yield stress and viscosity.As the CA volume fraction incread from 35%to 41%,the concrete yield stress and viscosity incread slightly (interception from 2.18to 3.18Nm and slope from 5.37to 7.24Nm S,respectively).For the mortar having high sand content (s /c =2.60),the yield stress and viscosity of the corresponding concrete were also high,and the ef-fects of CA content on the concrete rheology parameter,especially on the yield stress,became more significant.When the CA volume fraction incread from 35%to 41%,the corresponding concrete yield stress term and viscosity term incread from 4.03to 8.98Nm and from 5.19to 8.15NmS,respectively.
重生之国王的微笑Results suggest that both CA content and mortar composition had significant effects on concrete yield stress and viscosity.The two factors are important in concrete mix design for a workable concrete.
4.4.Effect of CA size and gradation
In prent study,concrete was considered to be a two-pha material compod with CA and mortar.The role of mortar in con-crete can be considered as bonding CA particles together,and fill-ing the voids among the CA particles.The amount of mortar required for a workable concrete depends on the amount of voids
Table 3
Coar aggregate characterization.
Single-sized aggregate Graded aggregate 19.0mm (3/4in.)
12.5mm (1/2in.)9.5mm (3/8in.) 4.75mm (no.4)G1G2G3Fineness modulus (FM)8.007.007.00 6.007.277.06 6.85Uncompacted voids (%)47.0648.1147.5849.8944.8744.4845.53Friction angle (°)
47.14
45.9
43.99
42.37
44.62
43.58
43.20
Table 4
Concrete test results summary.
Mixture no.
Mixture ID
Slump,mm (in.)G (Nm)H (Nm S)R 2Single-sized
1M1–Vca35%–19.0mm 165(6.50) 2.57 6.060.9552M1–Vca35%–4.75mm 165(6.50) 3.34 5.720.9953M1–Vca41%–19.0mm 121(4.75) 4.998.020.9384M1–Vca41%–4.75mm 102(4.00) 4.417.980.9895M2–Vca41%–19.0mm 64(2.50)8.718.900.9266M2–Vca41%–12.5mm 44(1.75)12.699.120.9757M2–Vca41%–9.5mm 32(1.25)13.5510.840.9868M2–Vca41%–4.75mm 38(1.50)14.2710.120.984Graded
9M1–Vca35%–G1197(7.75) 2.13 4.810.98510M1–Vca35%–G2216(8.50) 2.18 5.370.97911M1–Vca35%–G3210(8.25) 1.99 5.060.99812M1–Vca38%–G2191(7.50) 2.60 5.730.98513M1–Vca41%–G1171(6.75) 3.45 6.420.98714M1–Vca41%–G2178(7.00) 3.187.240.98515M1–Vca41%–G3191(7.50) 3.227.490.98616M2–Vca35%–G2140(5.50) 4.59 5.780.98817M2–Vca38%–G2102(4.00) 5.70 6.300.95018M2–Vca41%–G1102(4.00) 6.257.500.96219M2–Vca41%–G276(3.00)7.828.290.97220M2–Vca41%–G3114(4.50) 5.297.070.97821M3–Vca35%–G2127(5.00) 4.03 5.190.98622M3–Vca38%–G251(2.00) 6.837.140.98123项目发起人
M3–Vca41%–G2
38
(1.50)
8.98
8.15
0.982
Notes :G is the interception (yield stress term),H is the slope (viscosity term),and R 2is the coefficient of correlation of linear regression.
Vca=35%Vca=38%Vca=41%
M1 (s/c=1.75 w/c=0.45), G2M2 (s/c=2.21 w/c=0.45), G2M3 (s/c=2.60 w/c=0.50), G2
Vca=35%Vca=38%Vca=41%
M2 (s/c=2.21 w/c=0.45), G2M3 (s/c=2.60 w/c=0.50), G2
1200
J.Hu,K.Wang /Construction and Building Materials 25(2011)1196–1204
