1、Experimental Investigation of Bricks UnderUniaxial Tensile TestingBSTRACTSoftening is a gradual decrease of mechanical resistanceresulting from a continuous increase of deformation imposedon a material specimen or structure. It is a salient feature ofquasi-brittlematerials like clay brick, mortar, c
2、eramics, stoneor concrete which fail due to a process of progressiveInternal crack growth. Such mechanical behaviour iscommonly attributed to the heterogeneity of the material,due to the presence of different phases and materialdefects, such as flaws and voids. For tensile failure thisphenomenon has
3、 been well identified for concrete but veryfew results exists for clay brick. In the present paper, theresults of an extensive set of tests carried out at Universityof Minho and including three different types of back underniaxial tension will be presented. Both tensile strength andfracture energy a
4、re quantified, with recommendations for theadoption of practical values.INTRODUCTIONThe tensile behaviour of concrete and other quasi-brittlematerials that have a disordered Internal structure, such asbrick. can be well described by the cohesive crack modelproposed initially by HILLERBORG 1. This mo
5、del has beenwidely used as the fundamental model that describes thenon-linear fracture mechanics of quasi-brittle materials, e.g.2,3. According to this model and due to crackinglocalization, which is a characteristic of fracture process Inquasi-brittle materials, the tensile behaviour Is characteriz
6、edby two constitutive laws associated with different zones ofthe material during the loading process.see Figure 1. Theelastic-plastic stress-strain relationship of Figure la is validuntil the peak load is reached. It is noted that before thepeak Inelastic behaviour occurs due to micro-cracking andth
7、e energy dissipated in this process is usually neglected forthe calculation of the fracture energy. The stress-crackopening displacement relationship of Figure lb describesthe strain softening behaviour in the fracture process zoneafter the peak. The cohesive stress-opening displacementdiagram Is ch
8、aracterized by the gradual decrease of stressfrom ft maximum value, to zero, corresponding to theIncrease of the distance between the two edges of the crackfrom zero to the critical opening, u, The softening diagramassumes a fundamental role In the description of thefracture process and Is character
9、ized by the tensile strength,fr, and the fracture energy, Gr, which Is given by the areaunder the softening diagram, see Figure 1b. The criticalcrack opening, ue, can be replaced by the ductility index d,4 given as the ratio Grl fr, which represents the fractureenergy normalized by the tensile stren
10、gth. This parameterallows the characterization of the brittleness of the materialand is directly related to the shape of the descending portionof the stress-deformation diagram.There are several experimental methods that have beenused to measure the fracture properties (tensile strength,fracture ene
11、rgy and ductility Index) that allow the definition ofthe constitutive laws of the material, namely direct tensiletests, indirect tensile tests such as the three-point load test, and the Brazilian splitting test. Although tensile failureresults from a load combination and a multiplicity, of factors.m
12、eaning that direct tension is not the only cause of tensilecracking, a direct tensile test seems to be the moslappropriate test to characterize the basic failure mechanism(mode I) of quasi-brittle materials. This test Is defined as thereference method to follow (5j being adoptedin this work farthe c
13、haracterization of the tensile behaviour of bricks.Different issues related to the specimens and the testprocedures have been discussed in the past, namely thetesting equipment, the control method, thelocation of theLinear Variable Displacement Transducers (LVDTs), thealignment of the specimen and,
14、especially, the attachment ofthe specimens to the steel platens. The relevance of thelatter Is addressed In Figure 2 6. The behaviour inFigure 2a (rotating platens or hinges) Is justified by therotation of the specimen during the loading operation, wherethe crack proceeds from one side of the specim
15、en to theother side. In the case of Figure 2b using fixed (non-rotating)platens, a bending moment is introduced and multiple crackswill appear. This results in a slightly larger tensile strengthand a higher value of energy dissipated (fracture energy).Finally, It is noted that although the tensile s
16、trength andfracture energy are considered Intrinsic properties of thematerial, it Is well known that fracture properties are size andscale dependent 6,7.Tensile fracture parameters of masonry constituents,namely units and the mortar-unit interface, are keyparameters for advanced numerical modellingo
17、f masonryand for a deeper understanding of the behaviour of masonrystructures. in me present paper, an experimentalprogramme using three types of clay brick Is discussed withthe objective of increasing the data available in the literature.TEST SET-UP AND SPECIMENSTensile tests were performed with so
18、lid bricks produced byVale da Gandara, Portugal(S), hollow bricks produced by J.Monteiro e Filhos, Portugal (HP), and hollow bricksproduced by Suceram, Spain (HS). All bricks are extrudedand they were tested in vertical or thickness (V) and inhorizontal or length (H) direction resulting in six serie
19、s withthe following notation: SV, SH; HPV, HPH; HSV, HSH.Table 1 gives the dimensions of the bricks and the freewater absorption.The net compressive strength of thebricks, along the extrusion direction was 78N/mm282N/mm2 and 58N/mm2,respectively for S. HP and HS.Here it is noted that these values ar
20、e merely indicative, asthe first two values were from independent tests by differentresearchers and insufficient Information about the testingprocedures is available, see (8,9. The third value ofcompressive strength was provided by the manufacturer.It is noted that: (a) bricks HP are extruded with t
21、he holesparallel to the larger dimension and bricks HS are extrudedwith the holes parallel to the smaller dimension; (b) bricksHP and HS have small grooves in the upper surface (sideopposite to the facing side) in order to increase adhesionbetween the unit and the backing mortar, see Flgure 3.Testin
22、g equipment and applied measuring devicesThe tests were performed in the laboratory of the Civil Engineering Department of University of Minho, using a CS7400 - S shearing testing machine. This machine has twoindependent hydraulic actuators, positioned in vertical andhorizontal directions. It has a
23、load cell connected to the vertical actuator with a maximum capacity of 25 kN, being particularly suited to small specimens (maximum size of 90 x 150 x 150mm). The adoption of a constant cross section for the specimens leads to uncertainty about the location of the micro-cracks. This represents the
24、usual supplementary difficulty for the control method of this type of test. Since thecontrol system allows only one Linear Variable Displacement Transducer (LVDT) as displacement control, it was decided to introduce, by means a diamond sawing machine, twolateral notches with a depth of 8mm and a thi
25、ckness of 3mm at mid height of the specimen in order to localize the fracture surface. With the notches, the stress and deformation distribution is no longer uniform, with stress and strain gradients occurring very localized near the notch tips. Since three-dimensional npn-uniform crack opening can
26、occur on tensile tests 10, the tensile test control using the average of the deformations registered on the four corners of the specimen is the most appropriate procedure, see Figure 4. However, the available equipment can only control one displacement transducer (LVDT), located at a notched side. T
27、he transducers have a measure base of 1mm with a linearity of 0.17% of the full stroke. A deformation rate of 0.5um/s was used in the tests. The force applied was measured on a load cell of 25kN maximum load bearing capacity, with an accuracy of 0.03%.After preparation of the specimens ends, glue ad
28、hesion conditions were enhanced by making a series of superficial slots with a saw. Then, the specimens were carefully fixed to the steel platens using an epoxy resin (DEVCOM) in such a way that the platens were kept perfectly parallel. Here, It Is noted that the steel platens are fixed (non-rotatin
29、g), meaning that load eccentricity Is not specimens. The only source of an issue for pnsmadc eccentricity is parallelism between the steel platens which we the lack of,uld induce a bending moment In the specimen in the clamping operation.Specimen dimensionsTaking into consideration the brick dimensi
30、ons and the test set-up, 40 x 40 x 70mm S brick specimens were extracted as shown In Figure 5. HP and HS bricks are hollow and, therefore, the specimens extracted from the bricks must be representative of the brick shell, a channel or U specimens,and the brick web 1 specimens, see Figure 6. Here, it
31、 is noted that the usage of channel specimens in questionable because a load eccentricity is introduced by the fact the top and bottom flanges are fully glued to the steel specimens. Nevertheless, because the end platens are fully fixed, the eccentricity is very low. a linear elastic FEM calculation
32、 Indicates that the normalized load eccentricity (measured byeccentricity / web width) is only 0.03.RESULTSFrom the force-elongation relationship obtained in the tensile tests, the following parameters were evaluated: tensile strength, fracture energy, and residual stress at ultimate scan reading. T
33、he notches reduce the Youngs modulus of the brick (Eb) by about 20% - 40% 11. As the measure ofEb is questionable, it is not shown here.Figure 7 illustrates the procedure adopted for evaluating the fracture energy, G,. In the cohesive crack model addressed above, the crack opening u is equal to the
34、total elongation, subtracted from the elastic deformation (u, = v x lmaes / E0) and the irreversible deformation u;, which accounts for inelastic effects during material unloading, in the vicinity of the macro-crack. Here, /means is the distance between the measuring points of the LVDT.The maximum f
35、orce recorded by the load cell was divided by the effective area of each specimen (notched cross-section), in order to determine the tensile strength.The fracture energy is identified with the work that is carried out to complete the separation of the two faces of the macro-crack, per unit of area.
36、It is not possible to determine the exact crack opening for which the stress value transferred becomes zero, due to long tail exhibited by the softening branch of the stress-opening crack. For the calculation of the fracture energy, the value of the fracture energy Is usually calculated as the resul
37、t of the sum of two quantities, one quantity being measured and the other quantity estimated. The measured value of fracture energy Gf,means is directly computed as the area under the stress- crack opening diagram up to a nominal value of the peak strength (or the ultimate value). The estimated valu
38、e Gi, (b) in hollow bricks, the tensile strength of the shell is higher than that of the web. Moreover, the average results in the brick specimens are fairly constant taking into consideration that three different brick manufacturers were involved. Therefore, for practical purposes the following rec
39、ommendations seem possible: (a) the tensile strength of brick is around 5% of the compressive strength (with values found around 4N/mm2 in the direction parallel to extrusion and 3N/mm2 in the direction perpendicular to extrusion); (b)the ductility index is around 0.018mm (meaning that the fracture
40、energy found is around 0.08 and 0.06N/mm, respectively parallel and perpendicular to the extrusion direction). The values found apply solely for solid bricks and must be reduced for hollow bricks, according to the volume of holes.ACKNOWLEDGMENTSThe present work was partially supported by project GRO
41、W- 1999-70420 “Industrialised solutions for construction of reinforced brick masonry shell roofs“ funded by European Commission.单轴拉伸试验下砖的实验研究摘要转化是来自在一个材料样本和结构逐步减少机械阻力的过程，这是粘土砖、砂浆、石材等准脆性材料具体到一个渐进过程的显著特点。其破坏的原因是内部裂纹的增长。由于缺陷和空洞的存在，这些特性通常材料的异质性。在混凝土中，拉伸破坏现象已得到确定，但是这种破坏很少存在粘土砖中。在目前的论文中，米尼奥大学进行了一系列拉伸试验，改试
42、验还包括三个不同类型砖的单轴拉伸。这三种试验保过抗拉强度、断裂能量的量化和实用价值采纳的建议。引言混凝土和其它准脆性材料懒神行为有一个无序的内部结构材料，如砖。改象可以很好地描述最初有希勒勒提出的去裂纹模型，改模型已经作为最基本的模型用于解释准脆性材料的非线性断裂。依据这个模型，准脆性材料的一个特点就是开裂的位置不同，这是拉伸材料在不同部位的拉伸特点，见图 1。直到达到高峰负荷，弹塑性应力应变关系图是有效的。据悉，非弹性行为的高峰值发生是由于微裂过程中消耗的能量通常被忽略。应力开裂张拉位移关系图 1b 介绍了在断裂过程区的应变后峰转化行为。凝聚力应力张开位移座高峰压力逐渐减少直到为零，与其相对
43、应的裂纹的两个边之间距离增加从零到关键的开裂点。软化图在描述假设的基础性作用断裂过程抗拉强度特点的断裂能量，即由该地区给予的软化图，简图 16.关键性裂纹张拉可以代替延性指数 D；其代表了能源正常化的抗性强度。此参数允许脆性材料的表征和和降部分的形状直接关系到应力变形图。已经有几个用于测量断裂性能的实验方法对材料直接拉伸实验和间接拉伸实验本构关系，这意味着直接拉伸不是破坏的唯一原因。直接拉伸实验似乎是最适合的测试表征准脆性材料的实效机理。这个测试定义为可参考的方法。样本组织和测试程序已经在过去发表过，即测试设备，控制方法，线性可变位移传感器的安放位置。后者在图 2 中心理问题的相关性，在图 2
44、 的案例中，用固定压板，弯矩和多个裂缝会出现。这样的结果产生于一个稍大的抗拉强度和更高的能量值消散。最后，其指出虽然抗拉强度和断裂在属性材料内考虑，但是，众所周知，砌体成分断裂依赖于大小和规模，即单位砂浆设备接口一个实验程序使用三种类型砖在文献中体现目标数据的增加。拉伸断裂参数的砖石成分，即单位和砂浆设备接口，是关键参数先进的砖石结构的数值模拟并为砖石结构的特性有更深入的了解。我在本论文中，实验程序使用三种类型的粘土砖讨论，文献提供的目标数据的增加的。测试设置的标本由河谷达拉进行的实心砖的拉伸实验，左右的砖都是挤压的，他们测试是直的，厚的，水平的和长度方向六大系列。表 1 给出了砖的尺寸和自由水吸收。砖的净抗压强度在沿挤出方向分别是 78N/mm2，82N/mm2 和 5882N/mm2。在这里需指出：这些指
