Stabilization of Peat by Infiltration of Reactants A feasibility study: infiltration of silica biopolymer suspension in peat
Hamer, D.A. den
MetadataShow full item record
Summary Construction on peat soils has proven to be a challenging task to civil engineers as this soil type is highly compressible. Especially in densely populated delta area’s as the Rijn or the Maas delta in the western of the Netherlands infrastructure needs to be constructed on soft non-bearing soil layers. Construction on soft soils like peat is frequently accompanied by high geotechnical risks and costs. In conclusion, a peat layer is often unsuitable to use as a founding material. Conventional stabilization techniques have several disadvantages, among which is a strong reduction in the water storage capacity of the peat layer. A novel stabilization method was proposed, which takes infiltration and reactive transport as the starting point. The goal was to strengthen the soil matrix without a significant loss of porosity. The aim was to create a silicate coating which encloses or at least connects the peat fibres, hereafter referred to as fibre encapsulation. It was proposed that encapsulation of the fibre alters the mechanical and chemical bulk properties of a peat layer. Stabilization should have been achieved by infiltration and transport of the reactive components (in-situ process). Transport distances well over 1 meter and low injection pressures increased the efficiency of the stabilization technique and thereby its applicability. The starting point for this research project was the patent “Soil Strengthening Composition” (Zon, 2007). This patent introduces an in-situ treatment method to strengthen the solid matrix of a peat soil. However, several steps in the patented technique are not feasible when applied in the field. Hence, optimization of the patented treatment method and evaluation of its feasibility under continuous flow conditions was necessary. The most critical aspects of the patented method, which were the subjects of optimization, are enumerated below. First of all, a highly concentrated source of silica was required that could function as injection fluid in a peat soil. A high silica load would reduce the number of flushes needed to stabilize a certain bulk volume. However, a distinctive characteristic of a peat soil is acidic to neutral pH conditions. The presence of silica in solution is restricted to concentrations less than 100 mg Si per liter in the pH range of 2 to 9.5 at soil temperature. To increase the efficiency of the treatment method the load of silica in the injection fluid had to be elevated well above the solubility product of amorphous silica. Therefore the phase transition of silica, from dissolved to solid form, needed to be inhibited or at least delayed. The biopolymer Celquat L200 was added to retard the polymerization process of silica and the subsequent growth of silica particles. Silica and the biopolymer were thereby the reactive components of the injection fluid. Secondly, a hardened layer of amorphous silica had to encapsulate the peat fibre. Mass transfer of silica from the pore fluid to fibre surface was therefore needed. Preservation of porosity was the aim. Hydraulic conductivity and water storage capacity of the layer had to be retained. Thirdly, infiltration was the method of choice to transport and place the reactive components. The horizontal hydraulic conductivity of a peat layer had to be 10-6 to 10-7 m/s or higher to obtain significant transport distances in one week. This was a crucial aspect to develop a promising in-situ stabilization technique. Objective and research questions The general objective was stabilization of peat through the formation of a silica based coating supplied by infiltration of the reactants. To reach the general objective, the critical aspects of the stabilization method as patented by Van der Zon (2007) were studied separately. The research parts concerned: (I) retardation of silica polymerization and silica particle growth, (II) attachment of silica to peat solids, and (III) infiltration of injection fluid trough peat. The related research questions are shortly described below. Retardation of silica polymerization and silica particle growth The injection fluid had to be a highly concentrated source of silicon at neutral to acid pH conditions. It was proposed that the concentration of dissolved and dispersed colloidal silica in the injection fluid could be elevated by the addition of the biopolymer Celquat L200. The following questions were formulated: - Does the biopolymer retard the polymerization process silica and the growth of silica particles? And what is the optimum composition of the injection fluid in terms of initial silica and biopolymer concentration? - What is the impact of dissolved and particulate organic matter on the efficiency of the biopolymer to retard silica polymerization and to retard silica particle growth? - What is the impact of the biopolymer on the zeta potential, as measure for the surface charge of a particle? Attachment of silica to peat fibre surface Attachment of silica to peat solids had to be obtained to improve the mechanical properties of the bulk material. However, repulsion between fibre surface and silica species in pore water was likely to oppose attachment of silica to fibre surface. In neutral to acid pH conditions, the surface charge of peat is negative. The surface charge of dissolved or colloidal silica is neutral to negative at pH values lower than 9.5. Equal charged surfaces result into repulsion. The proposed solution was again the use of cationic biopolymer Celquat L200 to initiate attachment between peat fibre and silica. The following research questions were formulated: - Does the biopolymer Celquat L200 initiate silica attachment to peat solids? And what is the optimum initial concentration of silica in the injection fluid to maximize attachment? Infiltration of injection fluid in peat To achieve transport distances well over one meter of the reactive components through peat three conditions need to be met. The following research questions were formulated: - Is the hydraulic conductivity of the peat material high enough to allow reasonable horizontal flow velocities with a limited pressure gradient (at least 10-6 to 10-7 m/sec)? - Is the volume fraction of pores that conducts flow in peat large enough to allow sufficient stabilization of the bulk volume? - What is the impact of infiltration of the reactive fluid on hydraulic conductivity of peat? Do the reactive components, silica and biopolymer, attach and precipitate preferably on the peat fibres, and not in the pore space where precipitation could lead complete clogging? Experimental Methods The feasibility of the proposed stabilization technique was determined based on laboratory research. To answer the specific research questions, three laboratory experiments were performed. A batch experiment was performed to research the design of the injection fluid (Retardation of silica…). A batch experiment in the presence of peat was performed to research attachment of silica to peat (Attachment of silica…). At last, an infiltration experiment was performed to investigate the transport of the reactive components and to evaluate the effect of injection on the porosity of a peat column (Infiltration of injection fluid…). Peat material as used in the Attachment and Infiltration experiments originated from Bellingwedde. The infiltration test an extra type of peat was tested that originated from location close to Zegveld. Peat from Bellingwedde was a reed-sedge bog peat. Peat from Zegveld was a fen peat classified as sedge type of peat. The materials and methods used to answer the research questions, as formulated for the three separate research parts, are shortly described in this section. Retardation of silica polymerization and silica particle growth The polymerization and aggregation of silica was monitored in the presence of the biopolymer Celquat L200. Flasks were prepared with an initial silica concentration of 100, 300, 600 or 1250 ppm SiO2 using sodium metasilicate as the source of silicon. These flasks contained a biopolymer to silica weight ratio of 1 to 1, 0.5 to 1 or 0.1 to 1. The polymerization reactions were induced by neutralizing the super saturated alkaline silica solution (from pH >12 to 7.5) and monitored by the time-dependent depletion of dissolved silica and time-dependent particle formation and growth. The experiment was completed 113 hours after pH adjustment. Attachment of silica to peat fibre surface The attachment efficiency of silica to peat solids was determined at initial concentrations of 60, 100, 300, 600 and 1250 ppm as SiO2; and a biopolymer to silicate wt. ratio of 1. The distribution of silica and biopolymer between the liquid and the solid phase in presence of peat was the subject of research. The amount of silicate and biopolymer in the solid phase was calculated from the difference between the initial dissolved concentration before adjustment of pH and the final dissolved concentration after exposure to peat. The starting point of the attachment test was the moment the pH was adjusted to 7.5. The dissolved concentration of silica and biopolymer were measured after 65 hours and 113 hours of incubation. Infiltration of injection fluid in peat To derive the horizontal hydraulic conductivity of Bellingwedde and Zegveld peat a constant head test was performed. By in line electrical conductivity measurements and analyses of effluent composition at specific time intervals breakthrough curves of tracer infiltration and elution were constructed. Five peat columns were treated. Injection and elution of sodium chloride solution (0.09 and 0.07 M) was performed. Injection and elution of saturated silica solution (217.5 and 199.5 ppm SiO2) was performed; followed by the injection and partial elution of reactive colloidal suspensions (1247 ppm and 705 ppm SiO2/ ppm L200). And the fifth column was infiltrated with a biopolymer solution of 1962 ppm L200. Conclusions The conclusions drawn per research part are described below. These conclusions are directly related to the results of the experimental tests. In the next section the implication of these conclusions for the feasibility of in-situ stabilization of peat soil by infiltration of reactants, is given. Retardation of silica polymerization and silica particle growth The biopolymer Celquat L200 does effectively retard the polymerization process of silica. Dissolved silica concentrations of 300 to 400 ppm SiO2 were achieved. The efficiency in which the biopolymer retards the polymerization of silica does not depend on initial silica concentration and biopolymer dosage. The biopolymer Celquat L200 does effectively retard the growth of silica particles and this does depend on initial silica concentration and biopolymer dosage. Colloidal suspensions were formed during the 113 hours of incubation. Optimum composition of the injection fluid is obtained at initial silica concentration of 600 ppm SiO2 and the 600 ppm biopolymer Celquat L200. Attachment of silica to peat fibre surface The biopolymer Celquat L200 effectively adsorbs to peat in the presence and absence of silica; 90% to 99% of the biopolymer initially added was removed from solution. Attachment of silica to peat solids in the presence of the biopolymer Celquat L200 is effective; 79% to 90% of silica was removed from solution in the presence of the biopolymer Celquat L200. Infiltration of injection fluid in peat It appeared not to be possible to obtain sufficient infiltration of the injection fluid in the peat columns (with permeabilities of 10-6 and 10-7 m/s). A hard transparent gel was observed at the inlet on the interface between the porous disc and the peat. Penetration of the gel was in the order of millimeters. It appeared that the attachment was too fast to get sufficient infiltration. This has to be researched further before the method can be used in the field. Implications for the Application Solely based on present results in-situ stabilization of peat is not feasible, though the performed research is far from complete. If significant transport distances of the reactants can be obtained in-situ stabilization of peat might be feasible. And significant transport distances might be achieved if the attachment of silica is delayed. It should however be noted, that the intrinsic hydrologic properties of a peat soil complicates infiltration of reactants – irrespective of the properties of the reactants. Given the relatively low hydraulic conductivity of peat, the small pore volume that actually conducts flow and the heterogeneity of the aspects on small and bulk scale, the question arises if the method could be efficient and under which conditions. That is, efficient in the period of infiltration needed and the bulk strength obtained within this period. In-situ stabilization would provide a solution for a niche of the construction-market on soft soils. The focus is at applications where time is not a constrain. Treatment could then be applied as long-term method; with the advantage of preservation of water storage capacity of the peat layer, and low burden to the surroundings, as opposed to the common applied long term method of preloading. A better assessment could be made if in-situ stabilization or even mixed in place stabilization technique is the method of choice for a specific type of peat; or if they are at all efficient methods to apply; in the case chemical and botanic characteristics of peat are known. This aspect should be acknowledged when aiming for optimization of the mechanical properties of peat – and therefore included in geotechnical research on behavior and stabilization of peat soils.