By: Fanny Herrera, Luis Chahua and Elio Murrugarra, Knight Piésold Consultores, and David Reaño, Minera La Zanja.Abstract The expansion of the Pampa Verde waste dump is highly driven by the following constraints: (1) relatively limited options for laterally extending the existing buttress; (2) the need to satisfy both static and seismic stability; (3) the need to properly select high-strength reinforcement products; (4) the importance of testing specific materials to characterize the soilgeosynthetic interaction; and (5) the complex geometry of the overall system, in particular its global stability. The existing facility is stabilized by a geosynthetic-reinforced toe buttress that includes a compacted earth fill, with a slope of 2.1H:1V, overlying a mechanically stabilized earth wall (MSE or reinforced soil). The MSE structure includes upper and lower MSE wall sections with a horizontal step in between. The upper section is a wall reinforced with geogrid, while the lower section is a Terramesh system involving gabions reinforced with geogrids. The overall design approach involves the design of a geosynthetic-reinforced soil slope (RSS) that would use geogrids of high tensile capacity. The design guidelines for the proposed RSS are those outlined by the US Federal Highway Administration (FHWA-NHI-10-024). A key aspect of the proposed design is the proper evaluation of the soil vs. geosynthetic interface; a total of five geogrid products were considered for possible use in the construction of the geosynthetic-reinforced slope in the Pampa Verde project.IntroductionProject LocationThe La Zanja mining project, owned by Minera La Zanja, is located in the district of Pulan, province of Santa Cruz de Succhabamba, in the department of Cajamarca, Peru, at an altitude that varies between 2,800 and 3,800 meters above sea level (masl).The Pampa Verde waste dump is located southsouthwest of the Pampa Verde pit and comprises a total area of approximately 188,525 m2, after vertical expansion (effective area that does not include perimeter access, diversion channels and cut and fill slopes).BackgroundThe Pampa Verde waste dump was designed by Knight Piésold Consultores, considering that an unsuitable material stockpile would be encapsulated inside.In December 2013, the construction of the Pampa Verde wasted dump containment dike was finalized, consisting of a 29-m high compacted earth fill, with a slope of 2.1H:1V, overlying a mechanically stabilized earth wall (MSE or reinforced soil). The MSE structure involves upper and lower MSE wall sections with a horizontal step between them; the upper section is a wall reinforced with geogrid, while the lower section is a Terramesh system that includes gabions reinforced with geogrids. Figure 1 shows an aerial over-view of current conditions at the waste dump and Figure 2 shows an overview of the existing MSE structure and a detail of the existing rein- forced buttress.During the operation of the waste dump and the unsuitable material stockpile, several changes occurred, mainly due to the properties of the stored materials. Initially, the Pampa Verde waste dump was designed to store siliceous rock, but in practice, up to four different types of materials were stored: argillic, advanced argillic, massive silica and moder- ate silica.Therefore, Minera La Zanja requested Knight Piésold to redesign the Pampa Verde waste dump in order to implement the necessary measures to ensure its physical stability, which made it necessary to determine the properties of the materials through a geotechnical investigation. As a result of the redesign of the Pampa Verde waste dump, Knight Piésold developed a loading plan to conform the four types of the materials identified.To increase the storage capacity, Minera La Zanja requested the design of a vertical expansion of the dump to raise its elevation approximately 15 to 20 meters. The overall approach to stabilizing the vertical expansion involves the design and construction of a reinforced soil slope in front of the existing reinforced buttress. The raising of the Pampa Verde waste dump will allow an additional storage capacity of 0.94 million cubic meters, compared to its initial configuration.Characterization of the AreaLocal GeologyAt the local level, in the La Zanja project area there are mainly pyroclastic volcanic rocks and spills of the Lower Tertiary Lama formations and volcanic rocks from the Middle Tertiary Porculla formation. The rocks of the Upper Tertiary Huambo formation appear to the northwest, outside the limits of the project area.Geotechnical InvestigationBackgroundThe first geotechnical investigation was carried out between August 16 and October 11, 2010, in order to determine the geotechnical characteristics of the foundation surfaces where the Pampa Verde waste dump and associated structures would be built. The field works consisted of 6 drilling holes, 46 test pits, tests with dynamic penetrometer of conical tip (DPL) and geological-geotechnical mapping. The second geotechnical investigation was carried out in June 2013 and consisted of 6 drilling holes, 9 test pits and 4 in-situ density tests that were carried out by the water replacement method. The third geotechnical investigation was carried out between June 5 and 25, 2014, to characterize the materials that were stored in the dump. 2 drilling holes, 12 test pits (sampling for large-scale grainsize tests) and 8 in-situ density tests by the water replacement method were performed on the advanced argillic and argillic materials.Additional geotechnical investigation was conducted in November 2014 to characterize the advanced argillic and argillic materials that had been conformed. 2 test pits, 2 large scale grainsize tests and 7 in-situ density tests were developed by the water replacement method. Between April and June 2017, a geotechnical investigation was carried out with the specific objective of designing the raising of the Pampa Verde waste dump, having characterized the stored materials and the foundation at the site of the projected reinforced soil slope.FieldworksThe geotechnical field investigation consisted of the execution of 7 drilling holes and 23 test pits. The vertical geotechnical drilling reached variable depths between 15.0 m and 90.0 m, in which Standard Pen- etration Tests (SPT) and Large Penetration Test (LPT) were carried out as well as in-situ permeability tests, Lefranc (in soils) and Lugeon (on rocks). In the test pits, which reached depths varying between 0.8 m and 5.6 m, 6 large-scale grainsize tests and 6 in-situ density tests were carried out using the water replacement method; in addition, detailed records were taken of the stratigraphy of the materials found; in-situ testing and sampling of disturbed and undisturbed soils were carried out for the laboratory testing.Georys Ingenieros conducted geophysical prospecting tests consisting of 14 surface wave measurements in multichannel arrays using the Multichannel Analysis of Surface Waves (MASW) method, 9 readings by the Microtremor Array Measurement (MAM) method and 3 MASW 2D lines. To monitor the water level in the unsuitable material stockpile and the Pampa Verde waste dump, 6 Casagrande piezometers were installed.Laboratory TestsLaboratory tests were developed to determine the properties of the materials, including the existing rock in the foundation. In order to evaluate the potential for acid drainage generation, geochemical tests were carried out using the Sobek Modified Method (ABAM), in the laboratory of ALS Envi ronmental Chemex (Peru). Laboratory tests of the geogrid and geogrid vs. soil interface tests (ASTM D 5321) were also conducted at TRI Environmental Inc. (TRI), Texas, USA, the results of which are pre- sented in Section 4.Design EarthquakeThere are three seismic hazard studies developed for the specific location of the La Zanja project, the last one of June 2017 prepared by ZER Geosystem Peru, which included the characterization of the seismogenic sources near the study site, the elaboration of the seismic model based on the Ground Motion Prediction Equation (GMPE), the evaluation of the seismic hazard through the probabilistic and deterministic method ologies, seismic disaggregation analysis and the generation of five synthetic accelerograms adjusted to site's Uniform Hazard Spectrum.The results of the probabilistic seismic hazard are presented in Table 1 for a soil type 'B', accordingly to the International Building Code (IBC). To evaluate the physical stability of structures for the storage of mining waste, it is recommended to use as a design earthquake that corresponds to a return period of 1 in 100 years, during the period of operation, a criterion that is accepted worldwide for the design of this type of structures. Accordingly, to the 'Environmental Guide for the Slope Stability of Solid Mine Waste Deposits' of the Ministry of Energy and Mines of Peru (MINEM), the seismic coefficient can vary from 1/2 to 2/3 of the peak of horizontal acceleration of the ground, that is, from 0.07 to 0.10. For the purposes of the seismic design (pseudo-static analysis) of the Pampa Verde waste dump rising, 0.12 was used with conservative criteria, for operat- ing conditions.Pampa Verde Waste Dump Vertical ExpansionDesign CriteriaThe design criteria used have been proposed by Knight Piésold in accordance with international standards and national requirements for this type of structures, which were accepted by Minera La Zanja, as presented in Table 2.Project constraintsThe constraints of the project are as follows:1. Limited options for laterally extend the existing buttress to avoid encroaching on existing mining facilities;2. The need to satisfy both static and seismic stability;3. The need to properly select high-strength rein- forcement products;4. The importance of testing specific materials to characterize the soil-geogrid interaction; and5. The complex geometry of the overall system, in particular its global stability.Reinforced Soil Slope DesignGeneralIn order to increase the storage capacity of the Pam pa Verde waste dump, it was proposed to extend the containment dike vertically by means of a soil reinforced slope with uniaxial geogrids. The crest of the reinforced soil slope will be 10.0 m wide, an up-stream slope of 2H:1V and will be supported by the existing containment dike, while the downstream slope will be 1H:1V. The total height of the reinforced slope will reach 41 m. At the foot of the reinforced soil slope, it was proposed to form a reinforcement embankment with a slope of 2H:1V. The waste material within the deposit will conform to a general slope of 2.5H:1V, in 10 m height lifts, with 1.4H:1V slope, having to maintain berms of 11 m wide between the waste material lifts. The additional volume to be stored will be 941 900 m3.Slopes GeometryThe section for the analysis has been considered one that runs longitudinally through the Pampa Verde waste dump, considered the most critical because it covers the largest amount of waste material and the steepest slope downstream of the existing dike (where the haul road passes). Likewise, the general slope of the projected reinforced slope has been considered. The locations of the section that was analyzed is shown in Figure 3.Materials PropertiesFor the geotechnical characterization of the materials involved in the slopes stability analysis, the results of the geotechnical investigation developed for the Pampa Verde mine waste design were used, as well as the results of previous geotechnical investigations (2010, 2013 and 2014). The properties of the different materials involved in the slopes stability analysis are presented in Table 3.Piezometric Level ConditionsThe piezometric level was defined based on the records of the piezometers installed in the 7 drilling holes of the geotechnical investigation carried out in June 2017. Two piezometric levels were considered:ν Variable depth between 26.4 and 36.8 m with re spect to the existing ground level. This piezometric level appears due to the moisture of the discharged material and the leaks that have occurred inside the existing waste dump.ν Variable depth between 38.0 and 75.2 m with respect to the existing ground level. This level is close to the foundation level of the deposit.Geogrids Pullout AnalysisThe RESSA version 3.0 computer program was used, which belongs to the Adama Engineering Inc. set of programs. The program allows to develop the stability analysis considering the type of translational failure through the interaction between the geogrid and the soil. The results of the laboratory tests of soil vs. uniaxial geogrid interface and the performance of the geogrid have been used. The allowable design stress of the uniaxial geogrids was 230 kN/m. Geogrids pullout analyzes were carried out under static and seismic conditions (pseudo-static analysis). Geogrids pullout analyzes results show a minimum static safety factor of 1.51 and a minimum pseudoestatic safety factor of 1.16.Stability Analysis of the Waste Dump Facility Slope stability analyses associated with the vertical expansion of the Pampa Verde waste dump were developed using the computer program SLOPE/W® version 7.23, for static and seismic conditions (pseudostatic analysis). The following cases were analyzed:ν Global failure of the downstream slope. Failures through the body of the reinforced soil slope and the current and projected waste dump.ν Local failure downstream of the toe of the waste dump. Failures through the soil reinforced slope with uniaxial geogrids.ν Global failure of the upstream slope. Slopes failures in existing and projected waste materials.The geotechnical model is shown in Figure 4 and the results of the slope stability analyzes of the Pampa Verde waste dump are presented in Table 4.Uniaxial Geogrids EvaluationTesting Methods and CriteriaThe results obtained would not be valid in the event that there was not an adequate interaction between the geogrids and the soil to be used in the construction of the slope reinforced with geogrid, so a proper selection of geogrids is particularly relevant in this project, mainly due to the following aspects:ν The structure is relatively high, which leads to the selection of high tensile strength geosynthetic products.ν Direct shear is a relevant failure mode for the configuration of this project. Consequently, the shear strength between the soil and the geogrids must be properly characterized, not only for pullout evaluation, but also for wedge analyses.ν Due to the potential contact between the geogrid reinforcements and acidic fill materials, chemical degradation considerations are more relevant than for conventional retaining structures.A proper evaluation of the soil vs geogrid interface begins with the establishment of the test conditions used for to determine the properties of the in terface between the uniaxial geogrids and the back fill material. The interface testing program was developed at TRI Environmental Inc. (TRI) geotechnical and geosynthetics testing laboratory, located in Austin, Texas.Geogrids Considered in the Testing ProgramA total of five geogrid products were considered for possible use in the construction of the geogrid- reinforced slope in the Pampa Verde project; for the purposes of this paper, we will refer to the geogrids as “Geogrid 1” to “Geogrid 5”. The polymers used in the manufacturing process and key results of the wide-width tensile tests (ASTM D 6637, Method B) are summarized in Table 5.The preliminary design considered an allowable tensile strength (design tensile strength) of 230 kN/m, which was the basis for the identification of the five geogrid products. It should be noted that the allowable tensile strength is defined as the ultimate tensile strength penalized by reduction factors (construction damage, degradation, creep); the reduction factors for each geogrid are different and established by certified documentation provided by the manufacturers. Table 6 summarizes the ultimate tensile strength as reported in tests performed on the TRI, the reduction factors, and the predicted allowable tensile strength.As shown in Table 6, Geogrid 1 and Geogrid 2 led to an allowable tensile strength that is slightly below the 230 kN/m originally considered in the preliminary design. Geogrid 3 resulted in an allowable tensile strength that is significantly below 230 kN/m. Finally, Geogrid 4 and Geogrid 5 met the allowable tensile strength considered in the preliminary design.Two other considerations are also important for the raising of the Pampa Verde dike:ν Compatibility of soil and geogrid strains: as will be discussed later, the soil peak shear strength is found to occur at a shear strain of approximately 5%. Consequently, unit stress at 5% strain will lead to improved performance. This is because although the tensile capacity of geogrids may continue to develop beyond a 5% strain, the soil shear strength would have already been achieved. Therefore, a relevant parameter to consider when comparing the different products is the secant stiffness at a tensile strain of 5%. With a secant stiffness of 6,240 kN/m, Geogrid 4 is the product that provides the best displacement compatibility with the fill soil. Geogrid 1 and Geogrid 1 provide a secant stiffness of approximately 3,500 kN/m. Finally, Geogrid 2 and Geogrid 3 provide a secant stiffness of less than 2,500 kN/m.ν Chemical resistance to acidic soils: an important aspect to consider in the selection process, which is directly related to the raw polymeric material used in the manufacture of geogrids, is related to the chemical resistance of the products. While polyester (PET) is susceptible to chemical degradation in basic environments (pH greater than 10) and acidic environments (pH less than 2), polyvinyl alcohol (PVA) offers comparatively high chemical resistance in both highly basic and acidic environments. While the actual borrowing source of the fill material may not be precisely defined, there is concern that the fill used in the geogrid-reinforced slope may possibly involve a comparatively acidic environment. Consequently, polymeric materials such as PP, HDPE and PVA will provide better chemical resistance than PET. Among the geogrids considered in this project, Geogrid 4 is the only product manufactured with a polymer that resists acidic environments (PVA); all other products are manufactured using PET, as this material allows manufacture of the high-strength geogrids required for this project. No PP or HDPE products have been identified that meet the tensile strength requirements for this project.Direct Shear Test of the SoilDirect shear tests of the soil (ASTM D3080) were conducted using samples sieved to a maximum particle size of ¾”. Tests were carried out at four different confining pressures (198, 400, 600, and 800 kPa). The tests were carried out in submerged conditions (with container flooded one hour before the start of the shear). Conditioning of the soil specimen involved application of the normal stress for a period of 15 minutes before shearing. The shear displacement rate was 0.1 mm/min, which was considered adequate to minimize the development of pore water pressures. The lower half of the direct shear box had dimensions of 457 x 305 mm and was sheared against a smaller fixed container (the upper half with dimensions of 305 x 305 mm). Consequently, no area correction was considered in the interpretation of the results. This setup is consistent with ASTM D5321 used for interface shear testing. The shear test typically took about 13 hours due to the comparatively small shear displacement rate. The test conducted at 800 kPa normal stress required use of a smaller box (203 x 203 mm) to achieve the target normal stress.Soil vs Geogrid Interface Shear TestFigure 5 (a) to (e) shows the shear stress versus displacement results obtained for the four shear tests of the soil vs geogrid interface performed using each of the tested geogrid (Geogrid 1 to Geogrid 5, respectively).Determination of peak and residual shear strength values required careful interpretation. Consequently, only the results indicated with “red dots” were considered in the determination of the shear strength parameters. For Geogrid 1, the peak interface shear strength was characterized by an interface friction angle of 26.0 degrees and an adhesion intersection of 134 kPa. The residual shear strength of the interface was characterized by an interface friction angle of 24.7 degrees and a cohesion intersection of 89 kPa. The interaction coefficient for the peak interface shear strength was characterized by a friction coefficient of 0.84 and an adhesion coefficient of 11.75. Furthermore, the interaction coefficient for the residual interface shear strength was characterized by a friction coefficient of 0.99 and an adhesion coefficient of 44.68.For Geogrid 2, the peak interface shear strength was characterized by an interface friction angle of 20.4 degrees and an adhesion intersection of 115 kPa. The residual interface shear strength was characterized by an interface friction angle of 14.9 degrees and a cohesion intersection of 70 kPa. The interaction coefficient for the peak interface shear strength was characterized by a friction coefficient of 0.64 and an adhesion coefficient of 10.11. Furthermore, the interaction coefficient for the residual interface shear strength was characterized by a friction coefficient of 0.57 and an adhesion coefficient of 35.05.For Geogrid 3, the peak interface shear strength was characterized by an interface friction angle of 24.2 degrees and an adhesion intersection of 100 kPa. The residual interface shear strength was characterized by an interface friction angle of 24.5 degrees and a cohesion intersection of 77 kPa. The interaction coefficient for the peak interface shear strength was characterized by a friction coefficient of 0.78 and an adhesion coefficient of 8.80. Furthermore, the interaction coefficient for the residual interface shear strength was characterized by a friction coefficient of 0.98 and an adhesion coefficient of 38.79For Geogrid 4, the peak interface shear strength was characterized by an interface friction angle of 33.8 degrees and an adhesion intersection of 37 kPa. The residual interface shear strength was character ized by an interface friction angle of 28.6 degrees and a cohesion intersection of 33 kPa. The interaction coefficient for the peak interface shear strength was characterized by a friction coefficient of 1.16 and an adhesion coefficient of 3.23. Furthermore, the residual interface shear strength was characterized by a friction coefficient of 1.18 and an adhesion coefficient of 16.96.For Geogrid 5, it was observed that the test results performed at a normal stress of 800 kPa significantly affected the estimated friction angle. However, they were considered in the interpretation of the results. The peak interface shear strength was characterized by an interface friction angle of 41.0 degrees and an adhesion intersection of 0.00 kPa. However, obtaining this comparatively high interface friction angle was highly influenced by the 800 kPa test. The residual interface shear strength was characterized by an interface friction angle of 28.1 degrees and a cohesion intersection of 15.4 kPa. The interaction coefficient for the peak interface shear strength was characterized by a friction coefficient of 1.50 and an adhesion coefficient of 0.00. Furthermore, the interaction coefficient for the residual interface shear strength was characterized by a friction coefficient of 1.16 and an adhesion coefficient of 7.76.Figure 6 summarizes the results of the peak interface shear strength for the five soil vs. geogrid interfaces; the peak interface shear strength envelopes for the Geogrid 4 and Geogrid 5 (as well as the soil shear strength envelope) are represented by somewhat thicker lines. Figure 7 summarizes the residual interface shear strength results for the five soil vs. geogrid interfaces; the residual interface shear strength envelopes for the Geogrid 4 and Geogrid 5 (as well as the soil shear strength envelope) are represented by somewhat thicker lines in this figure.In the comparison of the different products, the most relevant parameters to evaluate, in relation to their interface shear strength characteristics, are the interface friction angle or the frictional coefficient of the interface shear strength. The highest interface coefficient for the peak interface shear strength was that of the Geogrid 5 (friction coefficient of 1.50), followed by the interface coefficient of Geogrid 4 (friction coefficient of 1.16). In addition, the highest interface coefficient for the residual interface shear strength was that of the Geogrid 4 (friction coeffi- cient of 1.18), followed by the interface coefficient of Geogrid 5 (friction coefficient of 1.16). Overall, Geogrid 4 and Geogrid 5 were the products that provided the highest interface shear strength performance; Geogrid 1 was also found to provide a good interface shear response. In contrast, Geogrid 2 and Geogrid 3 produced comparatively low interface shear strength results.Selection of Reinforcing GeogridProper selection of the geogrid to be used to mechanically stabilize the geogrid-reinforced slope requires evaluation of numerous factors. To objective ly evaluate these various considerations, a value engineering approach was adopted, often used by the FHWA to assess the merits of different alternatives retaining structures. Specifically, the following factors were identified as relevant for the different geogrid products considered as alternatives:ν Shear strength properties of the soil vs geogrid interface.ν Compatibility with deformation.ν Performance in acidic environments.ν Documented manufacturing quality control and reduction factors.ν Anticipated quality of technical support during the final stages of design and installation.ν Tradition in the use of the geogrid products in geoenvironmental applications.The tensile strength requirement was not taken as a factor, as the allowable tensile strength was a min imum requirement for consideration of the five geogrid products. Furthermore, cost was not considered among the factors for selection and consequently only technical considerations were weighed in this evaluation.Because not all factors are equally relevant, weighted ratings (WR) of 1 to 3 were assigned for each selected factor. Consequently, a WR of 3 was assigned to interface properties, a WR of 2 was assigned to performance in acidic environments and to manufacturing quality control, and a WR of 1 was assigned to the remaining selection factors. Selected factors are shown in Table 7.For each geogrid considered, qualitative weights (QR) ranging from 1 to 4 were subsequently assigned accordingly to the merit of each geogrid for each factor selected; the QR values are also shown in Table 7. Finally, the weighted scores were obtained by multiplying the WR by the QR, as summarized in Table 7, which also shows a final score for each geogrid alternative.As presented in this evaluation, Geogrid 4 was found to be the most appropriate geogrid alternative, with an aggregated score of 40. Geogrid 5 and Geogrid 1 were identified as somewhat distant secondrate alternatives, with aggregated scores of 31 and 21, respectively. Finally, Geogrid 2 and Geogrid 3 were identified as the least appropriate alternatives, with aggregate scores of 19 and 18, respectively.It is recommended that Geogrid 4 be selected as the geogrid reinforcement for the reinforced soil slope designed to stabilize the Pampa Verde waste dump. This selection is supported by the various considerations summarized in the value engineering approach documented in Table 7.Conclusions and RecommendationsBelow are the conclusions and recommendations derived from the engineering of the vertical expansion design of the Pampa Verde waste dump:1. Slope stability analyses indicate that the new configuration of the Pampa Verde waste dump will remain stable for static and seismic conditions.2. Five different types of geogrids were evaluated. The results of the laboratory tests indicated that, accordingly to its mechanical properties (resistance and deformation), it was recommended to use Geogrid 4, whose raw material is polyvinyl alcohol (PVA).3. The Senior geotechnical design reviewer of the vertical expansion of the Pampa Verde waste dump considered the efforts involved in the geotechnical characterization and engineering evaluations carried out by Knight Piésold to be particularly robust. Complementing these efforts with the selection of an appropriate geogrid product is expected to lead to a safe and well performing vertical expansion of the Pampa Verde waste dump.4. Carry out the laboratory tests of the geogrid to be used in the construction of the reinforced soil slope, if a different geogrid is used than the recommended one, in order to review the slope stability analyzes and verify the design of the reinforcement.ReferencesKnight Piésold Consultores S.A., 2018, Crecimiento del Depósito de Material Estéril Pampa Verde- Informe de Diseño, Rev. 0, Peru.Koerner, R.M., 1998, Designing with Geosynthetics, 4th Edi- tion, Prentice Hall Inc., New Jersey.National Highway Institute, November 2019, “Design and Construction of Mechanically Stabilized Earth Walls and Reinforced Soil Slopes"TRI/Environmental, Inc., 1998, Interface Friction/Direct Shear Testing & Slope Stability Issues Short Course, June 25-26. Austin, Texas.