### Abstract

There exist differences in the values of the tensile strength of cement-treated soils among results from direct tension, splitting tension, and bending tests. This paper presents numerical simulations that serve to explain such differences in the tensile strength properties obtained from the three types of tests. Finite element (FE)-analyses were performed to simulate the three tests as boundary value problems. A simple elasto-plastic model that simulates the behavior of cement-treated sands under a general three-dimensional stress state was used for these analyses. The analytical results showed that the direct tension tests yield reliable values of the actual tensile strength that will be mobilized under an ideal condition of uniaxial tension. On the other hand, in the splitting tension and bending tests, the tensile strength values are calculated from an external load by assuming a linear-elastic behavior in estimating the stress distribution. However, these values do not correspond to the actual tensile strength. The splitting tensile tests underestimate the tensile strength because the shear failure occurs below the loading strip and restrains further increase of the applied load. On the other hand, the bending tests overestimate the tensile strength because of the redistribution of stresses induced by the strain-softening behavior on the tensile side.

Original language | English |
---|---|

Pages (from-to) | 657-674 |

Number of pages | 18 |

Journal | Soils and Foundations |

Volume | 47 |

Issue number | 4 |

Publication status | Published - 2007 |

Externally published | Yes |

### Fingerprint

### Keywords

- Bending test
- Cement-treated sand
- Direct tension test
- Finite element method
- Splitting tension test
- Tensile strength (IGC: El)

### ASJC Scopus subject areas

- Geotechnical Engineering and Engineering Geology
- Civil and Structural Engineering

### Cite this

*Soils and Foundations*,

*47*(4), 657-674.

**Evaluation of tensile strength of cement-treated sand based on several types of laboratory tests.** / Namikawa, Tsutomu; Koseki, Junichi.

Research output: Contribution to journal › Article

*Soils and Foundations*, vol. 47, no. 4, pp. 657-674.

}

TY - JOUR

T1 - Evaluation of tensile strength of cement-treated sand based on several types of laboratory tests

AU - Namikawa, Tsutomu

AU - Koseki, Junichi

PY - 2007

Y1 - 2007

N2 - There exist differences in the values of the tensile strength of cement-treated soils among results from direct tension, splitting tension, and bending tests. This paper presents numerical simulations that serve to explain such differences in the tensile strength properties obtained from the three types of tests. Finite element (FE)-analyses were performed to simulate the three tests as boundary value problems. A simple elasto-plastic model that simulates the behavior of cement-treated sands under a general three-dimensional stress state was used for these analyses. The analytical results showed that the direct tension tests yield reliable values of the actual tensile strength that will be mobilized under an ideal condition of uniaxial tension. On the other hand, in the splitting tension and bending tests, the tensile strength values are calculated from an external load by assuming a linear-elastic behavior in estimating the stress distribution. However, these values do not correspond to the actual tensile strength. The splitting tensile tests underestimate the tensile strength because the shear failure occurs below the loading strip and restrains further increase of the applied load. On the other hand, the bending tests overestimate the tensile strength because of the redistribution of stresses induced by the strain-softening behavior on the tensile side.

AB - There exist differences in the values of the tensile strength of cement-treated soils among results from direct tension, splitting tension, and bending tests. This paper presents numerical simulations that serve to explain such differences in the tensile strength properties obtained from the three types of tests. Finite element (FE)-analyses were performed to simulate the three tests as boundary value problems. A simple elasto-plastic model that simulates the behavior of cement-treated sands under a general three-dimensional stress state was used for these analyses. The analytical results showed that the direct tension tests yield reliable values of the actual tensile strength that will be mobilized under an ideal condition of uniaxial tension. On the other hand, in the splitting tension and bending tests, the tensile strength values are calculated from an external load by assuming a linear-elastic behavior in estimating the stress distribution. However, these values do not correspond to the actual tensile strength. The splitting tensile tests underestimate the tensile strength because the shear failure occurs below the loading strip and restrains further increase of the applied load. On the other hand, the bending tests overestimate the tensile strength because of the redistribution of stresses induced by the strain-softening behavior on the tensile side.

KW - Bending test

KW - Cement-treated sand

KW - Direct tension test

KW - Finite element method

KW - Splitting tension test

KW - Tensile strength (IGC: El)

UR - http://www.scopus.com/inward/record.url?scp=70849128502&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70849128502&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:70849128502

VL - 47

SP - 657

EP - 674

JO - Soils and Foundations

JF - Soils and Foundations

SN - 0038-0806

IS - 4

ER -