Effect of admixtures on fresh grout and two-stage (pre-placed aggregate) concrete A. Nowek, P. Kaszubski, H. S. Abdelgader and J.

Transcription

1 SC5162 Techset Composition Ltd, Salisbury, U.K. 10/31/2006 Effect of admixtures on fresh grout and two-stage (pre-placed aggregate) concrete A. Nowek, P. Kaszubski, H. S. Abdelgader and J. Górski Q1 According to two-stage (pre-placed aggregate) concrete technology, special grout is injected into forms or foundation trenches with aggregate, as backing, placed earlier. The fresh grout differs from the ordinary concrete mixture. Good quality two-stage grout can be prepared in high-speed mixers, for example an Ultramixer (3000 rpm). It is characterised by high fluidity, low sedimentation, good viscosity, intensive hydn and a notable increase in the cement particles surface. However, because of the unique and complex equipment involved, twostage concrete technology is still under-utilised. The aim of the investigation presented in this work is to design a mixture composition similar in features to Ultramixer grout. For this purpose, the new grout was prepared using a normal mixer (140 rpm) and some admixtures. The experimental tests were performed with different mix proportions. On the basis of the obtained results, an optimal composition has been proposed. A relationship between the grout produced by the Ultramixer and standard grout has also been elaborated. The new grout design method can improve two-stage concrete technology. Some preliminary two-stage concrete tests are also presented. Introduction Anna Nowek Gdańsk University of Technology, Gdańsk, Poland Paweł Kaszubski Gdańsk University of Technology, Gdańsk, Poland Hakim S. Abdelgader Al-Fateh University, Tripoli, Libya Jarosław Górski Gdańsk University of Technology, Gdańsk, Poland The main idea of two-stage (pre-placed aggregate) concrete is to fill up the cavities in coarse aggregate, which has been previously placed at the site of destination (ditch, foundation form etc.), by a special mixture. The process of filling starts from the bottom (Fig. 1). The main role of the grout is the cementing of the aggregate in a monolith. 1,2 The difference between two-stage concrete and traditional concrete depends on the concrete structure. In twostage concrete, the compressive stresses are transmitted first through the skeleton of the aggregate and, after deformation, through the hardened mortar (Fig. 2), whereas in normal concrete the stresses pass through the whole non-homogeneous material. The basic components of two-stage concrete (coarse aggregate, fine sand, cement, water, additives and admixtures) are the same as those used in traditional concrete. 3,4 In the two-stage concrete method, only about of concrete ingredients take part in the mixing process, which results in minimising the exploitation of machines and equipment. The cost reduction is connected with lower consumption of cement (up to 20 30) in comparison with traditional concrete of the same compressive strength. The cost is also reduced because the main components of two-stage concrete are ordinary stone aggregate and fine sand, from which gravel, the most expensive ingredient, is eliminated. The most important advantages of two-stage concrete are its properties, that is minimal drying shrinkage that results in very low volume change and higher modulus of elasticity. 5 8 Two-stage concrete technology can be applied to the construction of special foundations (e.g., an 18-storey building in Gdańsk, Poland 9 ), special buildings (e.g., atomic plants 10 ), massive underwater construction (e.g., refacing of Baker Dam, Colorado, USA, 1 piers of the Mackinac Bridge, USA, 1 dam in Czchów on the Dunajec river, Poland 9 )and other unusual projects and constructions. 2 To obtain high quality grout that can be applied in two-stage (pre-placed aggregate) concrete technology, special mixers are usually used, for example, the Ultramixer (UM). 9,11 The UM consists, apart from the drive, of two basic elements: a cylindrical container and a turbine rotor on a vertical axis (Fig. 3). The basic rotor reaches rpm. The effect of UM action is high kinetic energy of mixing, being the result of three components of the mixed mass work: (a) translatory motion of the mixed mass along the circle of circumference of the main bin wall, in the horizontal plan (b) translatory motion of the mixed mass in spherical surface in the shape of funnel, in the vertical plan (c) acoustic vibn of the mixed mass, when the frequency depends on the number of revolutions, the number of rotor blades and the size of eccentric and rotor imbalance. The result of making the grout as indicated above was its special properties, namely high size reduction of cement batching, high fluidity and good viscosity. A practical measure of the mixture perfection was assumed to be the circumferential velocity of mass particles rotation and the mixing time. Though the UM was used for technical purposes, for example foundations, it revealed many disadvantages, for example, low efficiency (12 15 m 3 of concrete per hour) or high failure frequency of the engine that resulted in non-homogenous mortar and discontinuity of the whole process. These made this method competitive with traditional ones. The experimental work presented in this paper investigates the use of different types of grout admixtures in order to improve the # 2006 Thomas Telford and fib

2 main container electrical motor blader rotar fresh grout properties and strength of two-stage concrete. 12 A series of laboratory tests were carefully planned and carried out. Sedimentation and fluidity of the fresh grout and its compressive strength were examined during a period of 28 days. Taking account of the above, the grouts with the best properties and strength were selected. The aim of the research was to obtain mortar that satisfied the optimal UM grout property requirements without using special mixers. Some preliminary research into two-stage concrete is also presented. Details of the experiment Materials used For the prepan of all grout mixtures, Portland cement Type I 42.5 MPa, produced by the Górażdże cement plant (Poland) was used. The stone aggregate, having rounded shape and containing the fraction of 16/31.5mm, was obtained from Skowarcz quarry near (a) 4 Figure 3 (a) Ultramixer, and (b) scheme of mixing Gdansk (Poland). Just before making the concrete samples, the aggregate was washed in a stream of water in order to remove impurities. The aggregate was also subjected to cavity testing. For this purpose the volume of free space between the grains was measured. The void of the stone aggregate used was It should be pointed out that this is an important parameter because it defines the quantity of cement and admixtures in one cubic metre of the final two-stage concrete. The fraction of the stone aggregate was limited by the size of the mould ( ). The fine, clean sand was taken from the Borowiec quarry near Gdansk (Poland) and was divided into two fractions: 0/2 and 0/1 mm. Three types of superplasticisers were used in this investigation: Viscocrete 3 (Sika Poland), Muraplast FK-61 (MC Bauchemie) and Chrysofluid Optima 175 (Chryso Poland), all added at percentage of 0.7 by weight of cement. The additives used in the study were silica fume and fly ash, added at a percentage of 6 and 10 by weight of cement, respectively. Mixture proportions and specimen prepan The mixtures of the grout were prepared at w/c (water/cement) s of 0.4, 0.45, 0.5 and c/s (cement/sand) s of 1 : 0.5, 1 : 1, 1:1.5. The mixing was done in accordance with ASTM C192 using a normal mixer. The (b) rotor speed was 140 rpm. First, fine sand was added to a mixer, then the mixer was started and, after a few rotations, it was stopped. Next, the cement and the additives were added, and the mixer was again started and stopped after a few rotations. Finally, water and high-range water-reducing admixtures (HRWRA) were added. When all the ingredients were put into the mixer, the grout was mixed for 3 min. The properties of the freshly mixed grout, that is sedimentation and fluidity, were determined. In order to measure the sedimentation, glass cylinders of 1000 ml volume were used. Three cylinders were filled with grout in quantities of 500 ml and, after 3 hours, the free water and thin grout layers from the surface of the specimens were drawn off. The sedimentation (bleeding) is defined as a of the depth of the layers to the height of the grout measured after pouring it into cylinders. 3 The determination of the grout flow is linked directly to its capability of injection into the aggregate. Thus the required fluidity is one of the most important factors in twostage concrete grout. To measure the fluidity, the grout is placed in a steel cylinder of 200 ml volume (5 in diameter and 10 in height), and then poured onto a scale plate of 35 diameter from a distance of 1 (Fig. 4). The fluidity measure is the diameter of the grout spread. Six readings of the grout fluidity were taken in all measurement series, Q2

3 Effect of admixtures on fresh grout and two-stage concrete 3 4 Figure 4 The pouring of the grout onto the fluidity scale plate that is three of the fresh grout and three after 30 min. 3 Next, the grout strength parameters were determined. The grout was placed in steel moulds to make small beams of dimension Three samples of each kind of grout were made. The tests were performed after 28 days of curing. In addition, two-stage concrete specimens were prepared and tested. The test specimens were cast for evaluation of the grout and concrete strength. First the aggregate was placed in cubic moulds of dimensions. The grout was injected, under gravity pressure, through the stone skeleton to fill the voids between aggregate particles. For this purpose hard plastic pipes were placed at the bottom of the moulds (Fig. 5). One day after casting, the concrete specimens were removed from the moulds and then immersed in a water tank for seven days. All specimens were then stored in the laboratory at C until the testing time of 28 days. 4 Figure 5 Stone aggregate placed in the mould and the grout injection pipe Test results Fresh grout properties. The compatibility of the fresh grout components (cement, additives and admixtures) was examined on the basis of preliminary selected admixtures that were expected to give satisfactory results. The silica fume and the fly ash were used as additives. The first part of the investigation was performed using a combination of all admixtures and additives, for both fractions of sand, at w/c 0.45 and c/s 1 : 1 (Table 1). The mean values of the w/c andc/s s were a basis for the future changes of these parameters and their influence on the grout properties. The following remarks can be formulated. (a) The physical properties of the grout with the silica fume additive are much better than those with the fly ash. The main problem regarding the fly ash was the instantaneous segregation of the components, that is the sand settled on the bottom and the clear cement mortar accumulated at the top. This made the measurement of the initial flow impossible and after 30 min. the segregation was even bigger. High fluidity resulted from the segregation of the components. The water that separated from the circumference of the grout mixture increased the diameter indicating the fluiditybyabout2 4(mortars3,4,7,8; Table 1). Another problem with the addition Table 1 Preliminary tests of mixture proportions No. Symbol Water/cement of fly ash was a high sedimentation, reaching over 5, and with the admix FK-61 it was even 55. The only superplasticiser that, except for large sedimentation, did not result in the mortar segregation was Chrysofluid (mortars 11, 12; Table 1). (b) At this stage of research, the superplasticiser FK-61 was abandoned due to discrepancies in the results: (i) the biggest sedimentation with fly ash admixed (mortars 7, 8; Table 1) (ii) the worst fluidity in comparison to the grouts with other admixtures, that is mortars 5, 6 (Table 1) gave unsatisfactory fluidity results (18 and 24, respectively), contrary to mortars 7, 8 which gave very high fluidity (31 and 29, respectively) as a consequence of big segregation of the specimens. (c) Another decision taken to reduce the number of the test samples was the abandonment of the fly ash. But, because of its low cost in comparison to the silica fume, it was determined to check its effect using different grout mix proportions. The influence of sand fraction on the grout properties. The determination of the influence of the sand fractions was based on an analysis of the grout properties for all combinations of w/c and c/s s. From the test results with the grout mixtures (using the c/s of 1 : 1, 1 : 0.5 and Additive Admixture Mean 1 D1 0/ : 1 sf V D1 0/ : 1 sf V D2 0/ : 1 fa V D2 0/ : 1 fa V D3 0/ : 1 sf F D3 0/ : 1 sf F D4 0/ : 1 fa F D4 0/ : 1 fa F D5 0/ : 1 sf C D5 0/ : 1 sf C D6 0/ : 1 fa C D6 0/ : 1 fa C /2 and 0/1, fractions of sand; sf, silica fume; fa, fly ash; V, Viscocrete 3; F, FK 61; C, Chrysofluid 175. Mean fluidity: Q3

4 4 Nowek et al. 1:1.5 and variations in proportions of other parameters, i.e. w/c, fraction of sand, additive and admixture) it can be stated that: (a) A change of the sand fraction does not affect the physical properties of the grout. In most cases the differences in fluidity of the grout are approximately 1 2 or there is no difference at all. (b) As the w/c grows, the segregation and sedimentation of the grout mixtures rise, especially when fly ash is added. Instantaneous higher fluidity is caused by segregation of the components (the water that separates is collected around the circumference). The influence of additives on the fresh grout properties. (a) Much better results, in view of the segregation of components and sedimentation are obtained with silica fume added, compared to fly ash, by using Viscocrete and Chrysofluid. (b) Silica fume gives small fluidity for all c/s s and creates a very thick grout mixture when the w/c is low (w/c ¼ 0.4). (c) The grouts with silica fume result in much lower sedimentation than grouts with fly ash. No segregation was observed and the sedimentation was up to 5. (d) Grouts with fly ash result in high sedimentation and segregation of components. Good results have been achieved only for c/s 1 : 1.5 and all w/c s. Better compatibility of the fly ash with Portland Table 2 Properties of mixture with and without Viscocrete No. Symbol Water/cement cement, of high sand content, has been proved. Effect of admixtures on the grout properties. (a) The effect of the admixture action is mainly determined by the type of additive used. On the other hand, it is not possible to adequately estimate the influence of the c/s or the w/c s because, in each case, the combination of those parameters must be considered. (b) Viscocrete yields better results than Chrysofluid when the silica fume is added. This holds for both fluidity and sedimentation. (c) A much better sedimentation (1 3) with the remarkable number of grouts under investigation was achieved after 30 min. due to delayed action of the superplasticiser. This effect was observed using both the superplasticisers, especially in the proportion of c/s ¼ 1:1. The influence of proportions and composition on the compressive strength of the grout. (a) The fraction of sand (0/2 and 0/1 mm) influences neither the grout properties nor the compressive strength. (b) For recipes of grouts depending only on the type of admixture, higher compressive strength values were obtained using Viscocrete instead of Chrysofluid (see Table 3). (c) In the case of the grout recipes differing only in the type of additives, higher compressive strength values were obtained when silica fume was used in comparison to fly ash (Table 3). Additive Admixture Mean 1 D1 0/ : 1 sf V D1 0/2 m : 1 sf 10 3 EY1 0/ :0.5 sf V EY1 0/2m 0.4 1:0.5 sf Mean fluidity: D1 0/2 and EY1 0/2 are mixtures that were selected for the two-stage concrete production (see Analysis of two-stage concrete test ); m, samples of mortars of the same composition of components that the mortars used for the concrete grout prepans without admixtures; sf, silica fume; V, Viscocrete 3. Modified recipes of grout mixtures The basic part of the research was followed by more modified tests to check the effect of increasing content of admixtures and additives upon the grout mixture performance. In the first step, the quantity of admixture was increased. Two recipes (mortars 1 and 5; Table 3) were selected for the investigation and the fly ash content was changed from 10 up to 15, 20 and 25 (Table 3). 10 fly ash content is a normal quantity (i.e. used in all cases except the modified ones analysed at this point). These refer to mortars 1 and 5 of Table 3, where the samples are compared to a higher ash quantity. The aim of the variation of the admixture content was to check if the sedimentation and segregation of the grout components could be reduced or eliminated. According to satisfactory results obtained in the previous experiments, and the high cost of this additive, there was no need to increase the dosage of silica fume. Next, the content of the superplasticisers was increased using both Viscocrete and Chrysofluid. The quantity was changed from 0.7 up to 1 (Table 3) in order to get better fluidity. The grout fluidity did not exceed 20 when low proportions of w/c (even reaching 0.4; mortars 9 and 11; Table 3) were applied. (a) Larger quantities of fly ash with Viscocrete or Chrysofluid did not improve the results. A small increase in fluidity was accompanied by large sedimentation and a decrease in strength (Table 3). (b) Increasing the content of admixtures resulted in an improvement of the grout performance. Slightly better effects were observed with Viscocrete compared to Chrysofluid. It can be stated that applying admixtures in the proportion of 1 to the weight of cement causes an increase in fluidity up to 40 for Viscocrete and 30 for Chrysofluid. This appears promising for future research, as satisfactory fluidity results can be obtained at low w/c (a reduction of the water content in grout leads to better results in strength, tightness and frost resistance).

5 Effect of admixtures on fresh grout and two-stage concrete 5 Table 3 Influence of increasing the additive and admixture dosage on the grout properties No. Symbol Water/cement Additive Admixture Mean Mean fluidity: Compressive strength: MPa 1 D6 0/ : 1 fa 10 C D6 0/ : 1 fa 15 C D6 0/ : 1 fa 20 C D6 0/ : 1 fa 25 C D2 0/ : 1 fa 10 V D2 0/ : 1 fa 15 V D2 0/ : 1 fa 20 V D2 0/ : 1 fa 25 V 9 E1 0/ :1 sf V E1 0/ :1 sf V E6 0/ :1 fa C E6 0/ :1 fa C , 15, 20, 25 are 10, 15, 20, 25 content of additive; 0.7, 1 0.7, 1 content of admixture; sf, silica fume: fa, fly ash; V, Viscocrete 3; Q4 C, Chrysofluid 175; modified grout mixtures are marked in grey. Because of a great number of basic tests performed previously, only a few further modifications in grout composition were made. Consequently, only qualitative conclusions have been reached. Optimal grout recipes. The analysis of the test results allowed to find some optimal parameters: sedimentation up to 3 without segregation of the components; fluidity ; compressive strength of grout over 50 MPa. Two grout mixtures were chosen for further investigation of two-stage concrete. The selected grout mixtures fulfilled the requirements mentioned above. The recipes of the grout mixtures for one cubic metre of concrete are presented below. The void of aggregate was (a) D1 0/2: w/c ¼ 0.45; (c þ sf)/s ¼ 1:1; silica fume added at a rate of 6 by weight of cement; Viscocrete 3 added at a rate of 0.7 by weight of cement (Portland cement CEM I 42.5 MPa) sand kg cement kg water kg silica 22.2 kg Viscocrete 3 2.6kg Total grout needed for 1 m 3 of stone aggregate to produce concrete: 907.9kg (b) EY1 0/2: w/c ¼ 0.4; c þ cf/s ¼ 1:0.5; silica fume added at a rate of 6 by weight of cement; Viscocrete 3 added at a rate of 0.7 by weight of cement (Portland cement CEM I 42.5 MPa) sand cement water silica Viscocrete kg kg kg 28.7 kg 3.3kg Total grout needed for 1 m 3 of stone aggregate to produce concrete: 897.4kg (i) The silica fume appeared to be a better additive in comparison with fly ash and resulted in better physical properties of the grout, higher strength, lower sedimentation and a better fluidity (ii) A better effect of HRWRA was observed when Viscocrete was added compared to Chrysofluid. Comparison between grout produced by an Ultramixer and an ordinary (standard) mixer The results of the research were compared with the results obtained by the high-speed Ultramixer (UM) (3000 rpm). 3,4 The mixing time using of UM was 2, 4 and 6 min, whereas using the standard mixer it was 3 min. Thus, it was reasonable to compare the results for grouts mixed by the UM in 4 min. The grouts chosen for the test had the same w/c and c/s (w/c ¼ 0.4, 0.45, 0.5 and c/s ¼ 1 : 1, 1 : 1.5). During the test attention was paid to sedimentation, fluidity and compressive strength of grout. The fraction of sand used (0/2 mm) in the investigation (Table 4) was not taken into considen for obvious reasons high rotations of the UM crumbled sand into ash fractions. (a) Fluidity of grout was the most important parameter under investigation. The grouts produced by the standard mixer with addition of HRWRA have indicated better fluidities in comparison with the grouts made by the UM. A difference was noted only in mortar 7 (Table 4). In a large number of grouts compared, the difference was not significant, yet between mortar no. 10 and 11 it was over 15 and between mortar 1 and 3 or 1 and 4, it reached 7 and 6, respectively (Table 4). (b) As can be seen in Table 4, the compressive strength is lower in the grouts prepared with the UM than made using the standard mixer in each case. This results from the different types of cement used: Portland cement 35 (Górażdże) in the case of the UM; Portland cement 42.5

6 6 Nowek et al. Table 4 Comparison of the properties of fresh grout produced by the Ultramixer and the normal mixer No. Sand fraction: mm Admixture Additive Water/cement Mean Mean fluidity: Compressive Strength: MPa 1 0/2 B : /2 C fa 0.4 1: /1 V fa 0.4 1: /1 V 1 sf 0.4 1: /2 B : /2 V sf : /2 B : /1 C sf 0.5 1: /2 V sf 0.5 1: /2 B : /2 V fa 0.4 1: /2 B : /2 V fa : /2 V sf : /2 B : /2 C sf 0.5 1: /2 C fa 0.5 1: B 2, 2 content of Betoplast 1 admixture; V 1, 1 content of Viscocrete 3 admixture; V, C, 0.7 content of Viscocrete 3 and Chrysofluid 175 respectively; sf, silica fume; fa, fly ash; V, Viscocrete 3; C, Chrysofluid 175; grout mixtures prepared with the UM were marked in grey. Q4 (Górażdże) with the standard mixer. Nevertheless, the difference in compressive strength of grout was between 30 and 40. (c) The results of the sedimentation tests are better for the grouts produced in the UM due to the greater pulverising capability of their components. Sedimentation of the grout does not exceed 3 but similar satisfactory results can be achieved with a standard mixer, if the components are appropriately selected. It should be remarked that HRWRA were added in an amount of 0.7 by weight of cement. As further modification of the grout recipes has shown, it is possible to significantly improve the grout parameters by increasing the HRWRA dosage. Analysis of two-stage concrete test It should be pointed out that the main subject of the research was the property of the grout mixture itself, yet the investigation of twostage concrete was of preliminary importance. The results of the compression tests are shown in Table 5. The sample strengths scatter substantially. The reason for this scatter can be connected mainly to the dimensions of mould. The moulds usually used for the grout or ordinary concrete testing are too small for the two-stage concrete testing where the stones (aggregate) dominate in the mixtures. The next important reason for the scatter of the results is the small number of the samples (see Table 5). The investigation was performed using rounded stone aggregate. Cracking of the samples appeared at of strength during compression tests by hydraulic press. Despite this, the samples retained their form until the end of test. That proves that the stresses are transferred through the aggregate Table 5 Compressive strength of two-stage concrete Symbol D1 0/ EY1 0/ Compressive strength: MPa skeleton. The process of the sample destruction was not violent or explosive. One of the crushed specimens is presented in Fig. 6. Summary and conclusions This paper presents experimental results of investigations into the effect of types of additive and admixture on the grout that can be used in two-stage concrete technology. As the analysis of tests results show, the aim of the work was achieved. It is technically possible to get the same, or even better, fluidity of grouts with no or low sedimentation 4 Figure 6 Compressive strength test fracture across the rounded aggregate particles

7 Effect of admixtures on fresh grout and two-stage concrete 7 Q5 without the necessity to use a high-speed mixer. This is possible due to the fast progress of the chemistry that allows the replacement of mechanical performance of a high-speed mixer with the chemical action of HRWRA to achieve the required properties of the grout. Another important feature that can be eliminated is the problem of sand fractions. It was expected that the required fluidity of the grout could be obtained by limiting the sand fraction to a mere 1 mm. Yet it occurred that the action of HRWRA was so significant that there was practically no difference in applying sand fractions between 0/2 and 0/1 mm. In view of the results presented above, the application of two-stage concrete sounds very promising. The biggest disadvantage the limited efficiency and access to the equipment was eliminated. It is necessary to continue the research in order to specify the optimal proportions of grout components and the kind of admixtures and additives to be applied to specific types of cement. Comprehensive tests of the two-stage concrete (mechanical properties, durability and others) should be also performed. References 1. ACI Committee 304. Guide for the use of preplaced aggregate concrete for structural and mass concrete applications. American Concrete Institute, Detroit, reapproved Warner J. Preplaced aggregate concrete. Concrete International. September 2005, Abdelgader H. S. Effect of quantity of sand on the compressive strength of two-stage concrete. Magazine of Concrete Research, 1996, 48, No. 177, Abdelgader H. S. How to design concrete produced by a two-stage concreting method. Cement and Concrete Research, 1999, 29, No. 3, Abdelgader H. S. and Górski J. Influence of grout proportions on modulus of elasticity of two-stage concrete. Magazine of Concrete Research, 2002, concrete. Journal of Materials in Civil Engineering ASCE, 2003, 15, No Abdelgader H. S. and Ben-Zeitun A. E. Effect of grout proportions on tensile strength of twostage concrete measured by split and doublepunch tests. Structural Concrete, 2004, 5, No. 4, Abdelgader H. S., Ganaw A. and Fahema A. H. Influence of curing types on the compressive strength of two-stage (pre-placed aggregate) concrete. International Conference on Future and Challenges for Urban Development. Cairo, Egypt, December Braun K. Polcrete-two stage concrete method. Scientific Papers of the Technical University of Gdańsk, No. 115, Civil Engineering, No. 16, Gdańsk, Davis H. E. High-density concrete for shielding atomic energy plant. American Concrete Institute, 1958, 54, No. 11, Abdelgader H. S. Applications of two-stage concreting methods. Scientific Papers of the Technical University of Gdańsk, No. 570, Civil Engineering, No. 55, Gdańsk, , No Nowek A. and Kaszubski P. Two-stage concreting Q6 6. Abdelgader H. S. and Górski J. Stress strain method. MSc work, Technical University of relations and modulus of elasticity of two-stage Gdańsk, Gdańsk, 2003 (in Polish).