Numerical modeling of seismic table testing of tailings with and without inclusion

Behnam Ferdosi, Michael James, Michel Aubertin, Frederick Pretot

In the proceedings of: GeoRegina 2014: 67th Canadian Geotechnical Conference

Session: Mining Geotechnics

ABSTRACT: modeling of seismic table testing of tailings with and without inclusion Behnam Ferdosi, Michael James, Michel Aubertin & Frederick Pretot* Department of Civil, Geologic and Mining Engineering, Ecole Polytechnique, Montreal, Quebec, Canada *Now with SNC-Lavalin Company, Montreal, Qc, Canada ABSTRACT A new co-disposal method consisting of the construction of inclusions made of waste rock in tailings impoundments was proposed by Aubertin et al. (2002) to provide drainage and reinforcement. Research on this innovative disposal method has included intermediate scale physical models and numerical analyses. This paper presents 2D dynamic numerical simulations of instrumented seismic table testing of tailings, with and without inclusion. The goal of the ongoing numerical modelling is to validate and calibrate the use of the UBCSAND model in FLAC using the tests results. The results of the numerical simulations of the seismic table testing indicate that there is an encouraging agreement with the actual experimental data. There are however some differences regarding the rate of excess pore water pressure generation. Key results are presented and briefly discussed in this article. RÉSUMÉ Une nouvelle méthode de co-disposition qui consiste à construire des inclusions de roches stériles dans les parcs à résidus miniers a été proposée par Aubertin et al. (2002) afin de contribuer au drainage et au renforcement. Les travaux de recherche menés sur cette méthode de disposition novatrice comprennent des essais sur modèles physiques à échelle intermédiaire et des analyses numériques. Cet article présente des simulations numériques en conditions e menés sur des résidus miniers avec et sans inclusion.. Le but des analyses numériques en cours est de valider et calibrer le modèle UBCSAND implanté dans FLAC sur la base de ces essais. Les résultats des simulations numériques des essais sur table sismique montrent un accord encourageant avec les données expérimentales. Il y a toutefois des différences au niveau de la vitesse de génération des pressions 1 INTRODUCTION Impoundments used for storage of hydraulically-deposited tailings may be highly susceptible to statically or seismically-induced liquefaction that can cause failure. There have been several notable failures of tailings impoundments due to the seismic activity that induced excess pore water pressure generation and liquefaction in the retained tailings (Ishihara, 1984; ICOLD, 2001; WISE, 2014). Due to their permanence and the potential for, and effects of, liquefaction of tailings, their dynamic behavior, even in regions of relatively low seismicity, must be considered in stability and performance evaluations. Several studies , such as Mittal and Morgenstern (1975), Ishihara et al. (1980, 1981), Graga and Mckay (1984), Troncoso and Verdugo (1985), Troncoso (1986), Wijewickreme et al. (2005), Al-Tarhouni (2011), and James et al. (2011), have been conducted using cyclic simple/direct shear or cyclic triaxial testing of tailings, so their dynamic behavior is fairly well understood. Nonetheless, as the vast majority of research has been conducted on the dynamic behavior of naturally occurring soils, the important differences between the behaviors of soils and tailings have raised questions regarding the application of existing methods of analysis and evaluation (e.g. the simplified method of liquefaction analysis of the UBCSAND model) to the latter, which may require further calibration. Adalier et al. (2003) conducted seismic table testing of non-plastic silt with and without stone columns (inclusions). It was shown that these columns retarded the pore water pressure generation within the silt during cyclic loading. This reduction was attributed to the increased rigidity of the mass (silt and inclusions) which reduced shear strains. The effect of drainage via the sand columns during cyclic loading was minimal due to the relatively low hydraulic conductivity of the silt and short duration of the tests. Similarly, waste rock inclusions in tailings impoundments are expected to provide drainage during tailings deposition (Bolduc and Aubertin, 2013, 2014) and reinforcement (James and Aubertin 2012) during seismic loading. The UBCSAND model was developed by Byrne and his collaborators (Byrne et al. 1995; Puebla1995; Puebla et al.1997; Beaty and Byrne1998, 2011; Naesgaard 2011) for simulation of the pore water pressure generation and liquefaction of sandy soils. James (2009) used the UBCSAND model to simulate the cyclic simple shear testing of tailings over a range to confining stresses (100 to 400 kPa) and cyclic stress ratios (0.075 to 0.15) and found reasonably good agreement with respect to excess pore water pressure generation,shear strain development and the number of cycles of loading required to induce liquefaction. Pépin et al. (2012a, 2012b) conducted nine seismic shaking table tests on tailings including 5 tests with an inclusion consisting of more rigid and/or more pervious

RÉSUMÉ: ical modeling of seismic table testing of tailings with and without inclusion Behnam Ferdosi, Michael James, Michel Aubertin & Frederick Pretot* Department of Civil, Geologic and Mining Engineering, Ecole Polytechnique, Montreal, Quebec, Canada *Now with SNC-Lavalin Company, Montreal, Qc, Canada ABSTRACT A new co-disposal method consisting of the construction of inclusions made of waste rock in tailings impoundments was proposed by Aubertin et al. (2002) to provide drainage and reinforcement. Research on this innovative disposal method has included intermediate scale physical models and numerical analyses. This paper presents 2D dynamic numerical simulations of instrumented seismic table testing of tailings, with and without inclusion. The goal of the ongoing numerical modelling is to validate and calibrate the use of the UBCSAND model in FLAC using the tests results. The results of the numerical simulations of the seismic table testing indicate that there is an encouraging agreement with the actual experimental data. There are however some differences regarding the rate of excess pore water pressure generation. Key results are presented and briefly discussed in this article. RÉSUMÉ Une nouvelle méthode de co-disposition qui consiste à construire des inclusions de roches stériles dans les parcs à résidus miniers a été proposée par Aubertin et al. (2002) afin de contribuer au drainage et au renforcement. Les travaux de recherche menés sur cette méthode de disposition novatrice comprennent des essais sur modèles physiques à échelle intermédiaire et des analyses numériques. Cet article présente des simulations numériques en conditions

Access this article:
Canadian Geotechnical Society members can access to this article, along with all other Canadian Geotechnical Conference proceedings, in the Member Area. Conference proceedings are also available in many libraries.

Cite this article:
Behnam Ferdosi; Michael James; Michel Aubertin; Frederick Pretot (2014) Numerical modeling of seismic table testing of tailings with and without inclusion in GEO2014. Ottawa, Ontario: Canadian Geotechnical Society.

@article{GeoRegina14Paper125,author = Behnam Ferdosi; Michael James; Michel Aubertin; Frederick Pretot,title = Numerical modeling of seismic table testing of tailings with and without inclusion ,year = 2014}