**2. Materials and Methods**

Commercially available, dry hydrated lime of the class CL 90-S (standard EN 459-1 [15]) was used as a binder, with its density being 2.22 g/cm3. Finely ground limestone (hereafter limestone filler or LS) composed of 95.3% calcite and 4.7% dolomite was selected as the main filler. The limestone filler particles had a density of 2.76 g/cm3; their maximum size was 100 μm, with 10%, 20%, 50%, and 90% of particles smaller or equal to 3 μm, 9 μm, 15 μm, and 40 μm, respectively. The water absorption of LS was 0.5%. The chemical compositions of the binder and limestone filler, determined by the X-ray fluorescence analysis (Bruker S8 TIGER, Anhovo, Slovenia) according to the EN 196-2:2013 standard [16], are given in Table 2.


**Table 2.** Chemical composition of the used hydrated lime (CL 90-S) and limestone filler (LS).

Thin-walled soda–lime–borosilicate glass microspheres (3M Glass Bubbles K1) were used as a density-reducing constituent of the grout, with their typical density being 0.125 g/cm3 and respective minimum and maximum densities being 0.10 g/cm3 and 0.14 g/cm3. The maximum size of the microspheres was 120 μm, with 10%, 50%, and 90% of particles smaller or equal to 30 μm, 65 μm, and 115 μm, respectively.

In order to obtain the adequate viscosity and injectability of the grout in a fresh state, a polycarboxylate ether-based superplasticiser (PCE-SP) with a relative density of 1.05 g/cm3 <sup>±</sup> 0.02 g/cm3 and a pH value of 5.5 ± 1.0 was used as a highly efficient water-reducing agent.

All materials were stored in a room at a controlled temperature of 20 ◦C ± 1 ◦C and relative humidity of 60 ± 5%. The grout mixtures were prepared using tap water at a temperature of 20 ± 1 ◦C.

In previous research [1], the grout mixture based on 1 volume part hydrated lime and 3 volume parts limestone filler, with the addition of the PCE-SP chemical admixture, showed the best behaviour in the fresh and hardened states. This grout composition was selected as the normal density reference mixture to which properties of compositions with reduced densities were compared. Grouts with reduced densities were designed in such a way that the gradually increasing part of the limestone filler in the reference mixture was replaced with the same volume of glass microspheres. Consequently, five different grout compositions were obtained (Table 3), with the following volumetric proportions between limestone filler and glass microspheres: 100%:0%; 67%:33%, 50%:50%, 33%:67%, and 0%:100%.

Two parameters of the reference grout mixture remained unchanged throughout different grout compositions: the hydrated lime content and the dosage of PCE-SP, calculated as a percentage of the total solid materials mass, i.e., the binder and the fillers, which was 0.5% [1]. The water content of the mixtures was adjusted to obtain adequate workability of each particular injection grout. The workability was evaluated by conservator via injection of the grout through a 10 mL syringe by applying minimum pressure, using the procedure described in [1].


**Table 3.** Composition of tested injection grout mixtures.

Note: Sample formulation is indicated by the following symbols: LS = limestone filler; GM = glass microspheres. The volume proportion (limestone filler and glass microspheres) is indicated by subscript numbers.

#### *Mixture Preparation and Testing Methods*

The grout mixtures were prepared with a simple handheld kitchen mixer, in order to simulate the preparation of injection grouts in the field. The mixer had a power of 300 W and five different mixing velocities. The metal whisk used was 8.5 cm long with a diameter of 4.6 cm. For compositions with glass microspheres, the microspheres were first mixed with 50% of the water content into a slurry. The binder and the limestone filler were then dry-mixed for 15 s at the low speed of 540 rpm. During the next 45 s (at 540 rpm), the microspheres slurry and 20% of the water content were added and mixing at low speed proceeded for a further 45 s. During the last 15 s of mixing at low speed, the PCE-SP and the remaining 30% of the water content were added to the mixture. After that, the mixing was stopped, and the sides of the mixer bowl were scraped; the mixer was then turned on again and mixing proceeded for 3 min at medium speed (1200 rpm).

For nonstructural grouts, several adaptations of commonly available standard test methods were needed. We followed the testing procedures proposed by Biçer- ¸Sim¸sir and Rainer [17] or Padovnik et al. [1]. A brief description of the testing methods is given in the continuation. The tests were carried out in the laboratory with controlled temperature (20 ± 1 ◦C) and relative humidity of the air (60 ± 5% RH). First, the methods to evaluate fresh grouts' properties are given, followed by the tests carried out on hardened grouts.

The wet density of each grout was determined according to an adapted EN 1015-6 [18] standard procedure. The volume of the mixture was reduced from 1000 mL to 100 mL, using a metal cylindrical vessel. The filling of the measuring vessel was carried out in the same manner as in the case of soft mortar [18]. The wet density was calculated as a quotient of measured mass of the grout and the 100 mL volume.

The mini slump flow test [19] was used to determine the consistency of the grouts. A truncated cone-shaped mould (according to EN 459-2 [20]) placed at the centre of a smooth plate was filled with fresh grout. The average spread of the grout after lifting the mould was measured [1].

A bleeding test was carried out according to the adapted ASTM C940 [21] standard procedure. The volume of grout used in the test was reduced from 800 ± 10 mL to 80 ± 1 mL. Apart from this, the standard procedure was followed. A graduated cylinder of 100 mL was filled with 80 mL of the grout. Change in the accumulation rate of bleed water on the surface of the grout was observed over a period of time. The bleeding was calculated as a quotient between the volume of final bleed water and the initial volume of the grout.

The water-retaining ability of fresh grout was evaluated by the standard procedure prEN 1015-8:1999 [22,23]. The fresh grout was subjected to suction provided by filter papers, resulting in a loss of water. The mass of water remaining in the grout was expressed as a percentage of the grout's initial water content and reported as the water retentivity.

The injectability test was carried out according to the adapted standard procedure EN 1771:2004 [24]. Crushed lime mortar was used as a granular material, with a water absorption coefficient of 1.4 kg/(m2·s 1 <sup>2</sup> ) after 10 min, and total and capillary porosities equal to 27% and 26%, respectively. The cumulative amount of the granular material passing through 1 mm, 2 mm, 3 mm, and 4 mm sieves was 5%, 12%, 42%, and 99%, respectively [1]. This test determined the ability of fresh grout to fill a capillary network of granular material in a dry or prewetted state.

The drying shrinkage test with mortar cups was used to determine the reduction in grout volume after drying. The procedure proposed in [17] was applied. The mortar for the mortar cups was prepared with 1 volume part lime putty and 2 volume parts limestone sand (0/1 mm). It had a water absorption coefficient of 1.8 kg/(m2·<sup>s</sup> 1 <sup>2</sup> ) after 10 min, and the total and capillary porosities equal to 32% and 25%, respectively. Each cup had an outer diameter of 75 mm and a height of 30 mm. Dry and prewetted mortar cups were filled with grout mixtures; dimensional changes, as well as the crack-pattern development, were observed as the grout dried at 20 ± 1 ◦C T and 60 ± 5% RH.

The adapted settlement column segregation test [25] was used in order to assess the stability of fresh grout mixtures. A cylindrical plastic tube with an internal diameter of 22 mm and a height of 375 mm with three holes to facilitate the collection of a sample from the top, middle, and bottom levels (Figure 1) was filled with fresh grout and covered. After an hour, a sample from the top of the column was transferred into the first glass container. The same procedure was applied for the middle and bottom levels. For each of the three levels, the wet density of the collected sample was determined according to the adapted standard procedure EN 1015-6 [18], previously described. The stability of fresh grout is of utmost importance for the compositions where a combination of glass microspheres and superplasticiser is used, since such mixtures are particularly sensitive to segregation.

**Figure 1.** Cylindrical plastic tube dimensions.

In order to determine the properties of the grouts in their hardened state, cylindrical moulds were used to cast the specimens (at least three samples for each property and age). They were left in the moulds for 48 h and subsequently cured under controlled laboratory conditions (relative humidity 60 ± 5% and temperature 20 ± 1 ◦C) until testing. The hardened properties were determined at the grouts' age of 90 days, with compressive strength being measured again at 365 days.

The dry density was determined according to EN 1015-10 [26], on 50/50 mm cylinders.

The total and capillary porosities of the 50/50 mm cylinders were evaluated according to the Swiss standard SIA 262/1:2003, Appendix A [27]. Each specimen was subjected to different intensities of water saturation. Total and capillary porosities were calculated from the test results.

Water absorption by capillarity was measured following RILEM test No. II.6 [28]. Dry 50/50 mm cylinders were placed 2 mm deep in water and weighed at the prescribed intervals. The weight change of the specimen was used to calculate the amount of water absorbed after a predetermined time; subsequently, water absorption coefficients after 10 min and 24 h were calculated.

Water-vapour resistance coefficient (μ) was determined according to standard EN ISO 12572:2001 [29], using the dry cup method. A cylindrical specimen, with a diameter of 100 mm and height of 20 mm, was put on top of a vessel (cup) containing a desiccant (calcium chloride, CaCl2), sealed to the vessel's rim, and placed in a humidity-controlled chamber at 23 ± 0.5 ◦C and 60 ± 3% RH. The rate of water-vapour transmission through the specimen from the controlled atmosphere to the inside of the cup was determined by periodic weighing of the cup with the desiccant and the specimen.

The compressive test was carried out according to the EN 1015–11/A1 procedure [30], on 50/50 mm cylinders.

The splitting tensile test was carried out according to the ASTM C496/C496M-11 procedure [31], on 50/50 mm cylinders.

The accelerated ageing was carried out on 50/50 mm cylinders, in order to assess the grouts' durability. The specimens were subjected to fourteen freezing–thawing and heating–cooling cycles, following the protocol in Figure 2. Before each cycle, the specimens were subjected to capillary absorption of 3% NaCl solution or distilled water for 30 min, following the RILEM test No. II.6 [28], described earlier. Results of the capillarity water absorption test showed that the amount of water absorbed after 30 min was close to the value measured at 24 h.

**Figure 2.** The protocol of the freezing–thawing and heating–cooling cycle used in the study.

Finally, to evaluate the re-attachment ability of the grout, a pull-off test—according to the standard EN 1015-12 [32]—was performed on panel sandwich models [1]. The models were prepared to simulate a smaller (2 mm) and a larger (5 mm) detachment of fine plaster (1:3 lime putty: fine sand 0/1 mm lime mortar) from the rough plaster (1:3 lime putty: coarse sand 0/4 mm lime mortar). At the age of one year, the simulated detachments were filled by the grout using a syringe.

#### **3. Results and Discussion**
