**6. Properties of grouted soil, rock and building material**

312 Polyurethane

**Geopur® type**

company GME

 without reshaping of the grouted rock mass or with reshaping of the grouted rock mass.

> **foaming factor [-]**

considered as reshaping the rock mass as well.

surrounding of the borehole.

**5. Behavior of PUR resin in the grouted environment** 

is used in fissured rock and coarse grained soil like sand or gravel.

**Volume weight [kg/m3]** 

Grouting without reshaping of the rock mass may be of penetration or filling character. Penetration grouting works are performed in sandy soil or in constructions. Filling grouting

> **Water intake after 28days [vol. %]**

82/90 9 - 11 82 2,8 1,6 21 0,5 82/180 5 - 6 185 2,2 2,8 54 2,8 82/290 4 - 5 276 1,7 7,4 192 5,9 82/350 2 - 4 354 1,2 8,3 241 9,5 82/600 1,5 - 2 589 0,9 13,4 443 23,2 82/1000 1 - 1,2 1060 0,4 30,3 985 67,7

**Table 2.** Physical and mechanical parameters of the grouting system Geopur® produced by the

In case of grouting with reshaping of the rock environment a so called claquage occurs, which is in principle hydraulic fracturing of the rock well known from the oil and gas exploitation. Due to the high hydraulic pressure of the grouting media in the soil a spatial net of fissures is formed, which are subsequently filled with the grouting media. The length and width of fissures depends on the pressure of grouted resin, velocity of penetration and quantity of the grouting resin. Compacting grouting belongs among the grouting methods

Grouting PUR resin enters into the borehole as a mixture. The grouting material flows through the rock mass first as a liquid. After curing reaction start, gaseous CO2 is formed, which causes foaming of the mixture. In case of contact with moisture present in the soil or rock, the foaming is more intense, because the water reacts with the present isocyanate groups. Foaming causes increase of volume of the PUR mixture. The mixture is pushed into open structures of the rock mass and the viscosity of the mixture consecutively increases. The flowing stops, when the viscosity of the material is so high that further pumping is impossible, and the resin becomes hard foam. In case, that the pump is further operated, the pressure increases and the material density increases. In practice, this situation is indicated by significant pressure increase. Increase of the pressure may sometime cause opening of new structures for the grouting and continuing of the grouting. In case of formation of new openings the pressure drops. This may occur repeatedly until full grouting of the

**Flexural strength [MPa]** 

**Elasticity modulus [MPa]** 

**Compressive strength [MPa]** 

> During pressure grouting of PU grouting resins into soil, rock mass or fissured or defected constructions, new specific materials are formed. These materials have the properties of composite material and, taking into account their components character, are referred to as **geocomposites** (Snuparek & Soucek, 2000)**.**

> **In rocks or constructions** the grouted environment contains discontinuities. The geocomponent of the formed geocomposite is formed by blocks of the rock (or masonry), which are defined by combination of bedding surfaces, metamorphic foliation, fissures and etc.

> **In soil**, two basic types of geocomposites are formed by PU grouting: in case of non cohesive soil (sand-gravel), the geo-component of the geocomposite is built by solid grains or their aggregations of various size and shape. These contain grains of minerals and rocks, organic particles (shells of organisms, wood, carboniferous parts of plants and others) or parts of constructions (building material, metals, ash and others). In case of cohesive soil (clay, claystones, or siltstones), the geo-component of the geocomposite is formed by blocks of soil penetrated by a net of so called claquage fissures (fissures caused by hydraulic fracturing during the grouting), which are filled with the binding material.

> The binding material is represented in these geocomposites by hardened organic or organicmineral PU resin with various degree of foaming.

> Penetration of the grouting media through the inhomogeneous environment, and thus also the resulting properties of the formed geocomposite, is influenced by many factors. In case of geocomposites of PU resin – rock (soil) and PU resin – building material, the following factors have primary effect (Scucka & Soucek, 2007):


type and orientation of discontinuities, temperature of the environment, permeability (plastic+water+gas), adhesion of grouting media to the rock surface, composition of water, pore pressure.

Polyurethane Grouting Technologies 315

**According to spatial distribution of particles and pores** 


**According to angularity of** 


**According to size of pores in** 


**clastic particles** 


**rock or binder** 

**GEOCOMPOSITE TEXTURES** 

**According to distribution of rock particles** 



**According to rock grain size According to relative** 

**grain size** 

**binder** 

**Table 3.** Classification system for description of geocomposite structure and texture.


**According to distribution and morphology of bubble pores in the** 




**in the binder** 



**binder** 



binder material

binder material

**GEOCOMPOSITE STRUCTURES**

(fine, medium, coarse)

(fine, medium, coarse)

**According to quantity of** 

**According to ordering of building units According to character of penetration** 

**of the binder into the rock** 


**According to the level of filling of the space** 



Formed structure and texture of the geocomposite (usually very variable in case of PU geocomposites) is a result of the effect of the above mentioned factors. This variability depends on the bedding conditions of the grouted rock and on the parameters of the grouting process, mainly on the grouting pressure. Grouting pressure together with moisture cause for example significant zonal heterogeneity of the geocomposite in case of grouting of wet or saturated sand (mainly of lower permeability) (Aldorf & Vymazal, 1996).

Structural and textural variability of geocomposites significantly complicate the estimation of physical and mechanical properties of geotechnical constructions formed within the grouting process. Mainly the determination of strength and deformation properties of the geocomposite is problematic, because it is often hard to prepare standard laboratory testing specimens from the samples available and collected in situ by core drilling or excavation. In cases when it is impossible to prepare testing specimens from real in situ samples, model geocomposites are prepared by grouting into pressure tanks in the laboratory (Snuparek & Soucek, 2000). Physical and mechanical properties are subsequently determined on such prepared model samples. Qualitative and quantitative structural-textural parameters of the geocomposite are also analyzed by the methods of image analysis and are subsequently compared with parameters of real samples.

In the following text, basic types of structures and textures of geocomposites (with PU binding material) will be described and examples of determination of mechanical properties on real and laboratory prepared samples will be presented.

#### **6.1. Structure and texture of geocomposites**

Table 3 below presents a simple classification system for description of structure and texture of PU geocomposites according to various criteria. Some of the criteria are taken over from the modified system commonly used for analyses of structure and texture of sedimentary rocks in petrography (Pettijohn, 1975). We describe in more detail the categories created by the authors based on their long-term research. These include division of geocomposite textures according to the character of binding material penetration into the grouted soil (rock), division of structures according to quantity of binding material and description of the structure of the binding material in the geocomposite from the point of view of distribution, size and morphology of bubble pores.

#### *6.1.1. Character of penetration of the binder into the grouted rock*

According to penetration of the binder into the grouted soil or rock, the following textures or their combination may be distinguished:

 **honeycomb texture I.** - rock particle is surrounded by the binder and this has good adhesion to the rock surface (Fig. 2),


water, pore pressure.

compared with parameters of real samples.

on real and laboratory prepared samples will be presented.

*6.1.1. Character of penetration of the binder into the grouted rock* 

**6.1. Structure and texture of geocomposites** 

size and morphology of bubble pores.

or their combination may be distinguished:

adhesion to the rock surface (Fig. 2),

type and orientation of discontinuities, temperature of the environment, permeability (plastic+water+gas), adhesion of grouting media to the rock surface, composition of

Formed structure and texture of the geocomposite (usually very variable in case of PU geocomposites) is a result of the effect of the above mentioned factors. This variability depends on the bedding conditions of the grouted rock and on the parameters of the grouting process, mainly on the grouting pressure. Grouting pressure together with moisture cause for example significant zonal heterogeneity of the geocomposite in case of grouting of wet or saturated sand (mainly of lower permeability) (Aldorf & Vymazal, 1996). Structural and textural variability of geocomposites significantly complicate the estimation of physical and mechanical properties of geotechnical constructions formed within the grouting process. Mainly the determination of strength and deformation properties of the geocomposite is problematic, because it is often hard to prepare standard laboratory testing specimens from the samples available and collected in situ by core drilling or excavation. In cases when it is impossible to prepare testing specimens from real in situ samples, model geocomposites are prepared by grouting into pressure tanks in the laboratory (Snuparek & Soucek, 2000). Physical and mechanical properties are subsequently determined on such prepared model samples. Qualitative and quantitative structural-textural parameters of the geocomposite are also analyzed by the methods of image analysis and are subsequently

In the following text, basic types of structures and textures of geocomposites (with PU binding material) will be described and examples of determination of mechanical properties

Table 3 below presents a simple classification system for description of structure and texture of PU geocomposites according to various criteria. Some of the criteria are taken over from the modified system commonly used for analyses of structure and texture of sedimentary rocks in petrography (Pettijohn, 1975). We describe in more detail the categories created by the authors based on their long-term research. These include division of geocomposite textures according to the character of binding material penetration into the grouted soil (rock), division of structures according to quantity of binding material and description of the structure of the binding material in the geocomposite from the point of view of distribution,

According to penetration of the binder into the grouted soil or rock, the following textures

**honeycomb texture I.** - rock particle is surrounded by the binder and this has good


Polyurethane Grouting Technologies 317

**Figure 5.** Doughy texture - OMR-binder has a character of dough pushed into gaps between the conglomerate grains, it does not fill fully the voids and does not stick fully to the grains.

through the rock (masonry) and is filled by the binder (Fig. 7),

spreads through the rock (masonry) and is filled with the binder.

filled with PUR-binder.

 **stringer texture** - a net of fissures (usually in all directions), not formed due to the grouting, spreads through the rock (masonry) and is filled with the binder (Fig. 6), **claquage texture** - a net of fissures, which was formed due to the grouting, spreads

 **diffusive texture** - the rock is penetrated by the binder "in diffusive way" in pores (Fig. 8), **barrier texture** - binder fills only the interconnected cavities and gaps between the grains, it does not penetrate through the barriers formed by the present fine-grained soil (Fig. 9).

**Figure 6.** Stringer texture - a net of fissures (usually in all directions), not formed due to the grouting,

**Figure 7.** Claquage texture – fine-grained soil fractured hydraulically with claquage fissure, which is

**Figure 2.** Honeycomb texture I. (PUR is surrounding the rock particle and sticks well to the rock surface).


**Figure 3.** Honeycomb texture II. (PUR is surrounding the rock particles, but sticks only partly to their surface).

**Figure 4.** Honeycomb texture III. (free rock particle can be taken out of the PUR-binder "tissue").

surface).

**Figure 2.** Honeycomb texture I. (PUR is surrounding the rock particle and sticks well to the rock surface).

**honeycomb texture II.** - rock particle is surrounded by the binder, but the binder sticks

 **honeycomb texture III.** - rock particle is surrounded by the binder, but the binder does not stick to the rock surface and is separated from the rock by a gap; free particle may

 **doughy texture** - the binder looks like pastry pushed into the gaps between the grains of the aggregate, it does not fill fully the gaps between the grains and does not stick

**Figure 3.** Honeycomb texture II. (PUR is surrounding the rock particles, but sticks only partly to their

**Figure 4.** Honeycomb texture III. (free rock particle can be taken out of the PUR-binder "tissue").

only partly to the rock surface (Fig. 3),

completely to the grains (Fig. 5),

be taken out from the "tissue" of the plastic binder (Fig. 4),

**Figure 5.** Doughy texture - OMR-binder has a character of dough pushed into gaps between the conglomerate grains, it does not fill fully the voids and does not stick fully to the grains.


**Figure 6.** Stringer texture - a net of fissures (usually in all directions), not formed due to the grouting, spreads through the rock (masonry) and is filled with the binder.

**Figure 7.** Claquage texture – fine-grained soil fractured hydraulically with claquage fissure, which is filled with PUR-binder.

Polyurethane Grouting Technologies 319

**type PUR 1** – binder is compact, vitreous, bubble pores occur only sporadically or are

 **type PUR 2** – isolated spherical or ellipsoidal bubble pores of similar size are suspended within the vitreous binder, bubbles have smooth walls, no collapsed walls occur (Fig.

 **type PUR 3** – partly collapsed bubble pores are suspended within isles of vitreous compact binder, bubbles are in contact, walls are of peel or shell character (Fig. 10c), **type PUR 4** – collapsed bubble pores with thin walls are in contact with each other and deform themselves, walls are of peel to honeycomb character. Vitreous compact binder

In case of organic-mineral resins, out of which mainly non foaming types are used in the geotechnics, the structure of the hardened resin has different character. The character strongly **depends on the intensity and time of mixing** of the input components. In case of good mixing, isolated or touching, regular spherical, white drops of polysilicious acid gel are densely distributed within the plastic mass. Irregularly distributed spherical or less regular pores of various sizes are also present in the structure (**type OMR 1**, Fig. 11a). In case of insufficient mixing time and intensity, an inhomogeneous mass is formed containing mineral part, which is irregularly distributed within the plastic mass (**type OMR 2***,* Fig. 11b).

**Figure 10.** Basic types of plastic binder structure with pores in PUR-geocomposites: (a) type PUR 1, (b)

(c) (d)

(a) (b)

not present at all (Fig. 10a),

is missing or is sporadic (Fig. 10d).

type PUR 2, c) type PUR 3, d) type PUR 4.

10b),

**Figure 8.** Diffusive porous zonal texture of geocomposite (crushed brick + PUR). A border formed by penetration of the binder into the pores of the brick fragments is visible on the bigger grain edges. Smaller brick fragments are fully penetrated by the binder.

**Figure 9.** Barrier texture – the binder fills the interconnected cavities between the grains, it does not penetrate through the barriers formed by the basic mass.

#### *6.1.2. Quantity of the binder in the geocomposite*

According to the quantity of the binder in comparison with the quantity of rock component in the geocomposite, the following structures can be distinguished:


#### *6.1.3. Distribution, size and morphology of bubble pores in the binder*

A specific feature of most grouting media on the basis of PU is increase of their volume by foaming. In order to describe the relative distribution and morphology of bubble pores in the foamed hardened PU binder, we use the following classification for both micro as well as macro evaluation (Scucka & Soucek, 2007).

 **type PUR 1** – binder is compact, vitreous, bubble pores occur only sporadically or are not present at all (Fig. 10a),

318 Polyurethane

**Figure 8.** Diffusive porous zonal texture of geocomposite (crushed brick + PUR). A border formed by penetration of the binder into the pores of the brick fragments is visible on the bigger grain edges.

**Figure 9.** Barrier texture – the binder fills the interconnected cavities between the grains, it does not

According to the quantity of the binder in comparison with the quantity of rock component

**basal structure** – rock particles are distributed in the abundant binder, particles are

**porous structure** – binding material fills the pores and voids in between the grains,

A specific feature of most grouting media on the basis of PU is increase of their volume by foaming. In order to describe the relative distribution and morphology of bubble pores in the foamed hardened PU binder, we use the following classification for both micro as well

 **contact structure** – binding material is present only in places of grain contact, **coating structure** – small amount of binder creates coating around the clastic grains.

Smaller brick fragments are fully penetrated by the binder.

penetrate through the barriers formed by the basic mass.

*6.1.2. Quantity of the binder in the geocomposite* 

grains are in contact with each other,

as macro evaluation (Scucka & Soucek, 2007).

separated,

in the geocomposite, the following structures can be distinguished:

*6.1.3. Distribution, size and morphology of bubble pores in the binder* 


In case of organic-mineral resins, out of which mainly non foaming types are used in the geotechnics, the structure of the hardened resin has different character. The character strongly **depends on the intensity and time of mixing** of the input components. In case of good mixing, isolated or touching, regular spherical, white drops of polysilicious acid gel are densely distributed within the plastic mass. Irregularly distributed spherical or less regular pores of various sizes are also present in the structure (**type OMR 1**, Fig. 11a). In case of insufficient mixing time and intensity, an inhomogeneous mass is formed containing mineral part, which is irregularly distributed within the plastic mass (**type OMR 2***,* Fig. 11b).

**Figure 10.** Basic types of plastic binder structure with pores in PUR-geocomposites: (a) type PUR 1, (b) type PUR 2, c) type PUR 3, d) type PUR 4.

Polyurethane Grouting Technologies 321

mechanics and building material mechanics, are applied for the testing (e.g. ISRM Commision , 1978), and these are adjusted to specific properties of the geocomposites.

**Figure 12.** Testing specimens of geocomposites prepared by various methods: a) hand mixed mixture of sand + PUR poured into cylinder form with subsequent adjustment of frontal surfaces, b) sand grouted with PUR in pressure tank – cut out specimen of a prism shape of 50mm×50mm×100mm dimensions after uniaxial compressive strength test, c) cylinder shape specimen made from control core drilling, originating from grouted concrete foundation of high voltage pole, d) cube-shaped specimen –

(d) (e)

(a) (b) (c)

An example of PUR-geocomposite testing is an analysis of sample prepared by grouting in situ with Geopur® 082/90 PU grouting system into saturated sand and shale sandy breccia. Grouting works were performed during construction and excavation of an underground utility tunnel. The underground construction crossed non-coherent strongly saturated sand, where increased water inflows into the construction occurred with subsequent bursting of sand from the working face. Safety of the excavation works at this critical section was secured by creation of a protective "umbrella" above the excavation. This protective "umbrella" was made by the method of PU pressure grouting via perforated steel tubes. During the excavation one of the monolithic geocomposite bodies

coal parts grouted with PU in pressure tank, e) beam type specimen of 40mm×40mm×160mm dimensions during flexural strength test (specimen made of real sample of sand grouted with PUR)

**Figure 11.** Basic types of plastic binder structure in OMR-geocomposites: (a) type OMR 1, (b) type OMR 2.

### **6.2. Determination of mechanical properties of PUR-geocomposites**

#### *6.2.1. Preparation of samples and testing specimens*

Samples and testing specimens of PUR-geocomposites for laboratory testing of physicalmechanical properties are obtained by the following methods:


The choice of shape and size of the testing specimens is determined by the properties of particular geocomposite type. It depends mainly on the dimensions, shape and textural homogeneity of available geocomposite and also on the possibilities of cutting and machining with cutting or drilling tools. A high-speed abrasive water jet can be well used for cutting of large geocomposite samples (Hlavacek et al., 2009). For shaping of test samples, laboratory drilling machine with diamond bit and diamond saw are used.

#### *6.2.2. Laboratory tests of PUR-geocomposites*

There are no standard approaches in the field of laboratory testing of mechanical properties of PUR-geocomposites up to date. Corresponding methods and norms, used in rock mechanics and building material mechanics, are applied for the testing (e.g. ISRM Commision , 1978), and these are adjusted to specific properties of the geocomposites.

320 Polyurethane

OMR 2.

**Figure 11.** Basic types of plastic binder structure in OMR-geocomposites: (a) type OMR 1, (b) type

(a) (b)

Samples and testing specimens of PUR-geocomposites for laboratory testing of physical-

1. *by pouring and free foaming* – the simplest method, PUR-mixture is hand mixed with grouted material (sand, gravel, rock debris and others) and is poured into forms of required shape, in which it freely foams. Final shape of the testing specimen is adjusted by cutting off of overfoamed part of the sample (over the volume of the form) (Fig. 12a). 2. *by grouting into pressure tank* – testing samples of required dimensions and shape are

3. *by in situ test grouting*– PU mixture is grouted into the rock environment in situ, testing samples of required dimensions and shape are drilled or cut from the formed

4. *from real geotechnical projects* – during performance of grouting works in practice, test grouting is undertaken with subsequent sample collection of the grouted rock mass or construction, in some cases also control samples are collected in order to judge the

The choice of shape and size of the testing specimens is determined by the properties of particular geocomposite type. It depends mainly on the dimensions, shape and textural homogeneity of available geocomposite and also on the possibilities of cutting and machining with cutting or drilling tools. A high-speed abrasive water jet can be well used for cutting of large geocomposite samples (Hlavacek et al., 2009). For shaping of test

There are no standard approaches in the field of laboratory testing of mechanical properties of PUR-geocomposites up to date. Corresponding methods and norms, used in rock

samples, laboratory drilling machine with diamond bit and diamond saw are used.

**6.2. Determination of mechanical properties of PUR-geocomposites** 

*6.2.1. Preparation of samples and testing specimens* 

mechanical properties are obtained by the following methods:

drilled or cut from the formed geocomposite (Fig. 12b,d).

quality and effectiveness of the performed works (Fig. 12c).

*6.2.2. Laboratory tests of PUR-geocomposites* 

geocomposite, which is excavated after the test grouting (Fig. 12e).

**Figure 12.** Testing specimens of geocomposites prepared by various methods: a) hand mixed mixture of sand + PUR poured into cylinder form with subsequent adjustment of frontal surfaces, b) sand grouted with PUR in pressure tank – cut out specimen of a prism shape of 50mm×50mm×100mm dimensions after uniaxial compressive strength test, c) cylinder shape specimen made from control core drilling, originating from grouted concrete foundation of high voltage pole, d) cube-shaped specimen – coal parts grouted with PU in pressure tank, e) beam type specimen of 40mm×40mm×160mm dimensions during flexural strength test (specimen made of real sample of sand grouted with PUR)

(d) (e)

An example of PUR-geocomposite testing is an analysis of sample prepared by grouting in situ with Geopur® 082/90 PU grouting system into saturated sand and shale sandy breccia. Grouting works were performed during construction and excavation of an underground utility tunnel. The underground construction crossed non-coherent strongly saturated sand, where increased water inflows into the construction occurred with subsequent bursting of sand from the working face. Safety of the excavation works at this critical section was secured by creation of a protective "umbrella" above the excavation. This protective "umbrella" was made by the method of PU pressure grouting via perforated steel tubes. During the excavation one of the monolithic geocomposite bodies

was dug out for laboratory testing purposes (Fig. 13a). Cross cutting of the geocomposite body showed macroscopically visible zonal heterogeneity of the material (Fig. 13b). Using the methods of image analysis, it was found out, that the degree of foaming of PUR binder increases with the increasing distance from the grouting tube, and that the volume ratio of PUR binder in the geocomposite ranges from 40 to 45% in the various parts of the geocomposite body. Various consistencies of the binder and variable portion of coarse grained breccia grains were identified in the body of the geocomposite. Due to this heterogeneity, the compressive strength values tested on cube-shaped specimens cut from the geocomposite material ranged in relatively wide interval from 5 to 30 MPa (average 12 MPa) and the deformation modulus ranged in interval from 100 to 2000 MPa (average 700 MPa).

Polyurethane Grouting Technologies 323

**Figure 14.** Sample of model geocomposite (PUR+basalt aggregate) prepared in laboratory by grouting

**Figure 15.** Different types of textures of laboratory prepared model geocomposite formed due to different moisture level of the grouted material (PUR+basalt aggregate): a) dry aggregate, b) saturated

(a) (b)

*6.2.3. Current knowledge about the mechanical properties of PUR-geocomposites* 

Data about stress and strain properties of geocomposites with PU binder have not been yet evaluated in summary or statistically. Technical literature or company brochures offer information connected with particular applications under particular geotechnical conditions or from testing and comparison of individual grouting materials. A little bit more complex data and unified interpretation of observed parameters are presented by (Aldorf & Vymazal, 1996), where the properties of laboratory prepared and in situ prepared geocomposites are compared (sand grouted with PU and acrylate resin). Further, we present some conclusions deduced from the results of the above mentioned experiments and from

 PUR-geocomposites behave in comparison with common rock types extraordinarily, mainly in terms of considerable elasticity and plastic deformations. This feature is observed mainly behind the ultimate strength, when along with the relatively high values of longitudinal deformation (approx. 10 - 20% in case of grouted sand) residual

into pressure tank.

aggregate.

the knowledge of the authors in the field:

strength of the material remains significantly high.

**Figure 13.** Monolithic geocomposite body formed by GEOPUR grouting into saturated sand and shale sandy breccia (a) and a cross through the geocomposite – zonal heterogeneity of the material is visible (b).

An example of testing of model PUR-geocomposites, prepared in laboratory conditions by grouting into pressure tanks, is an analysis of the effect of grouted environment moisture to the resulting properties of the geocomposite (Scucka & Soucek, 2007). A geocomposite sample, laboratory prepared by grouting into pressure tank filled with loose rock material, is presented in Fig. 14. Grouting was performed into crushed basalt of defined grain size. The material was grouted by the Geopur® 082/1000 resin, which reacts during the curing process with water. Grouting was performed both into dry material and saturated material. Fig. 15 shows macroscopically visible differences in the texture of formed geocomposites. While during the grouting of dry material honeycomb type I texture is formed (good adhesion of binder to the rock particles) with slightly foamed binder *PUR 2* (see sec. 6.1.), in case of saturated grouted material, honeycomb type II texture is formed (only partial sticking of the binder to the rock particles) with strongly foamed binder *PUR 3.* The difference in moisture of the grouted material causes, that the compressive strength of saturated samples is in average lower by approx. 80% and the deformation modulus is lower by approx. 90% compared to the values of samples prepared by grouting into dry material.

(average 700 MPa).

(b).

material.

was dug out for laboratory testing purposes (Fig. 13a). Cross cutting of the geocomposite body showed macroscopically visible zonal heterogeneity of the material (Fig. 13b). Using the methods of image analysis, it was found out, that the degree of foaming of PUR binder increases with the increasing distance from the grouting tube, and that the volume ratio of PUR binder in the geocomposite ranges from 40 to 45% in the various parts of the geocomposite body. Various consistencies of the binder and variable portion of coarse grained breccia grains were identified in the body of the geocomposite. Due to this heterogeneity, the compressive strength values tested on cube-shaped specimens cut from the geocomposite material ranged in relatively wide interval from 5 to 30 MPa (average 12 MPa) and the deformation modulus ranged in interval from 100 to 2000 MPa

**Figure 13.** Monolithic geocomposite body formed by GEOPUR grouting into saturated sand and shale sandy breccia (a) and a cross through the geocomposite – zonal heterogeneity of the material is visible

(a) (b)

An example of testing of model PUR-geocomposites, prepared in laboratory conditions by grouting into pressure tanks, is an analysis of the effect of grouted environment moisture to the resulting properties of the geocomposite (Scucka & Soucek, 2007). A geocomposite sample, laboratory prepared by grouting into pressure tank filled with loose rock material, is presented in Fig. 14. Grouting was performed into crushed basalt of defined grain size. The material was grouted by the Geopur® 082/1000 resin, which reacts during the curing process with water. Grouting was performed both into dry material and saturated material. Fig. 15 shows macroscopically visible differences in the texture of formed geocomposites. While during the grouting of dry material honeycomb type I texture is formed (good adhesion of binder to the rock particles) with slightly foamed binder *PUR 2* (see sec. 6.1.), in case of saturated grouted material, honeycomb type II texture is formed (only partial sticking of the binder to the rock particles) with strongly foamed binder *PUR 3.* The difference in moisture of the grouted material causes, that the compressive strength of saturated samples is in average lower by approx. 80% and the deformation modulus is lower by approx. 90% compared to the values of samples prepared by grouting into dry

**Figure 14.** Sample of model geocomposite (PUR+basalt aggregate) prepared in laboratory by grouting into pressure tank.

**Figure 15.** Different types of textures of laboratory prepared model geocomposite formed due to different moisture level of the grouted material (PUR+basalt aggregate): a) dry aggregate, b) saturated aggregate.

#### *6.2.3. Current knowledge about the mechanical properties of PUR-geocomposites*

Data about stress and strain properties of geocomposites with PU binder have not been yet evaluated in summary or statistically. Technical literature or company brochures offer information connected with particular applications under particular geotechnical conditions or from testing and comparison of individual grouting materials. A little bit more complex data and unified interpretation of observed parameters are presented by (Aldorf & Vymazal, 1996), where the properties of laboratory prepared and in situ prepared geocomposites are compared (sand grouted with PU and acrylate resin). Further, we present some conclusions deduced from the results of the above mentioned experiments and from the knowledge of the authors in the field:

 PUR-geocomposites behave in comparison with common rock types extraordinarily, mainly in terms of considerable elasticity and plastic deformations. This feature is observed mainly behind the ultimate strength, when along with the relatively high values of longitudinal deformation (approx. 10 - 20% in case of grouted sand) residual strength of the material remains significantly high.

 The ratio of rock grains to the PU binder, distribution of grains, grain size and the possibility of formation of porous foamed material have significant influence to the values of parameters of physical-mechanical properties of the geocomposite. These factors are always very variable at in situ conditions and depend on the bedding conditions of the rock (local porosity, structure, permeability, moisture and etc.). It is therefore necessary to take into consideration during the laboratory testing mainly the parameters of samples of lower volume weight.

Polyurethane Grouting Technologies 325

**Figure 16.** Original state of the retaining wall

**Figure 17.** Anchoring works

**Figure 18.** Situation before final completion of the works

