Abrasive erosion of concrete in hydraulic structures is mainly caused by rubble dragged by flowing water. Damages of concrete structures caused by the abrasion process are very severe and indicate the necessity of taking into account the influence of this process while designing concrete mixtures and building constructions. It should be emphasized that the unified and general criterion of acceptable range of destruction of such type structures is difficult to formulate. The second, very important issue is the methodology of testing concrete resistance to abrasive erosion. Examinations carried out so far has mainly dealt with the determination of concrete resistance properties. The process of abrasion was simulated artificially (i.e. shot blasting, sand blasting, abrasive disks) and did not reflect natural conditions of the environment. Obtained results, due to the application of various testing methods, cannot be assessed on the basis of comparative analysis.
Questions of concrete abrasive erosion simulation in laboratory conditions were, among others, paper subjects of A. Małasiewicz , A. Bania [2,3], E. Horszczaruk  and G. Haroske . The authors of the quoted papers agree that the vital impact on tests results are connected with the procedure of environment effects modelling. The importance of technological factors influence determining composition and the procedure of consolidation and curing of concrete exposed to abrasive erosion was emphasized in publications of T.C.Liu , T. Naik and others  as well as A. Nanni . So far, no comparative analysis has been carried out to indicate the importance of particular material and environmental factors on concrete abrasion caused by rubble dragged by flowing water.
The objective of the current research work was not only to build a device for simulating natural conditions of concrete abrasion, but also to work out a numerical model enabling to forecast disintegration progress of concrete structure according to hydraulic conditions of rubble movement.
2. Simulation of abrasive wear of concrete in laboratory conditions
The majority of devices described in professional literature designed to test abrasive resistance of concrete is used for simulating mechanisms of sand blasting [9,10] and grooving [8,11] with dry friction. Very few publications describe research done in conditions similar to the natural influence of environment with the applications of devices enabling to model the process of concrete abrasion with the mixture of aggregate and water [3,4,5]. Generally, the construction of these devices does not allow to model the movement of the whole grain composition of the rubble. Most often the mixture of small sand fractions and water is used as an abrasive; the mixture is ejected on the concrete surface under pressure and with high speed.
Abrasive erosion is a complex process and from the tribology point of view depends on the procedure of abrasive mixture reaction on the element undergoing abrasive actions. While analysing results of various tests one can compare only those that were based on the same tribological mechanisms of samples abrasion. These mechanisms are description by four basic external parameters: composition (hardness), size of abrasive grains, its speed and glancing angle on the sample. The alteration of one of these parameters causes the change of the abrasive mechanism and thus the comparative analysis of tests results is not possible. The lack of correlation between concrete tests results obtained by methods simulating different abrasive mechanisms was indicated, among others, in the paper . Concrete description by high resistance to abrasion, marked on Boehm's disk, that were exposed to abrasive action with the use of the mixture of aggregate and water showed low resistance or lack of abrasion resistance. Laboratory simulation of abrasion process in conditions similar to natural influence of the environment enables to assess properly the resistance of concrete to abrasive erosion.
Fig 1: Scheme of the device for testing abrasive erosion of concrete|
The scheme of the test stand for testing concrete resistance to abrasive erosion used by the authors of this paper is presented in figure 1. The device consists of steel drum of 155 cm in diameter and length of 228 cm; in its horizontal axle there is a fixed drive shaft with 36 beds for fixing tested concrete samples. The shaft with fixed samples is powered by electric engine. The belt transmission allows to regulate the number of drive shaft rotations simulation the speed of rubble flow dragged by water. The drum is filled with the mixture of aggregate and water, changed after each series of tests. Tested samples of concrete are fixed to the drive shaft in 3 or 6 rows, 6 pieces in each row. Geometrical relationships presented in figure 1 allow at the same time to abrade samples fixed only in one row.
Between the drum and the belt transmission there are the following parts: flexible coupling, torque meter with a collector and reducer. Forces of abrasive mixture reacting on the tested concrete samples cause torsion of the drive shaft whose deformations are measured with a set of extensometers. A signal from the collector of the torque meter is transmitted to the amplifier and than converted to numerical signal and recorded on a PC disk. Installed meter circuit allows to estimate the value of the torque moment of the drum's drive shaft:
In the equation (2.1), k means experimentally determined coefficient of shift, U means voltage obtained from the measurement in time t = t2 - t1. Figure 1 presents exemplary measurement of voltage alterations measured in time t = 4 seconds.
Defined values of moment Mp enable to determine forces P of the abrasive mixture reaction on the tested concrete samples:
Moment Mo introduced to the equation (2.2) is the moment of resistance caused by the dead movement of the system without the abrasive mixture, r is the arm of the acting force P, i is the number of samples fixed in one row of the drive shaft of the device.
Voltage measurement U during tests was made each time and in determined time periods, immediatelly before the inspection of abrasive concrete mass decrement change.
3. Program and methodology of scientific research
Two series of basic tests for abrasive wear of concrete were carried out. The first series were samples of portland cement without additions of CEM I 32.5 R, based on natural aggregate of stable graining up to 16 mm, characterized by variable value of water-cement ratio: w = 0.4, 0.5 and 0.6 and constant plastic consistence of the mixture. The samples were abraded at the constant speed of rubble movement v = 2.5 m/s.
Concrete samples of the second series of tests were made of cement and aggregate as for the first series with the constant ratio w = 0.5 and plastic consistence. The process of abrasive erosion was simulated with different speed of rubble movement - v=2.5,3.5 and 4.0 m/s.
Basic tests were carried out on cylindrical samples h = f= 80 mm. Sets of 18 samples were made from particular concrete mixtures; after 28 days of setting they were dripped to constant mass and than abraded with the mixture of aggregate and water. Abrasive mixture consisted of natural aggregate, graining 8 - 32 mm, to which water was added in 1:3 voluminal ratio. Selection of grain composition of the aggregate is based on the analysis of rubble transport conditions at assumed simulated speeds of flowing water. The process of the tested samples abrasion was carried out in 96 hours period. The check of the concrete samples mass was performed after 1,2,6,12,18,24,36,48 and 96 hours of the device functioning. Before the periodical check of the concrete samples mass the measurements of voltage alteration U of the meter circuit were made; the meter circuit enabled to define forces P of the abrasive mixture influence. Except for the basic tests the concrete samples of each series were exposed to the resistance tests after 28 days of setting as well as to absorbability tests.
4. Results of the research
Results of abrasive wear of concrete tests were presented in fig. 2 showing relative changes of the abrasive material mass in time function:Dm/mo = f(t). Particular mass decrements Dm/mo are the mean value obtained from 18 pieces of samples abraded at the same time. Mean values of standard deviation of relative samples mass decrement, determined in appointed check time, changed from s = 0.0026 after 1 hour of the abrasive mixture reaction to s = 0.0163 after 96 hours of abrasion.
Fig 2a: Changes of relative mass decrements, work of abrasive mixture- series I |
Fig 2b: Changes of relative mass decrements, work of abrasive mixture - series II|
Fig 2c: . Changes of relative mass decrements, power of abrasive mixture - series I |
Fig 2d: . Changes of relative mass decrements, power of abrasive mixture - series II|
Applied testing device enabled the check of abrasive conditions of the tested concretes. Determined force values P of the mixture reaction on the tested samples allow to define work W and abrasive power M of concretes from the particular series. These values, dependant first of all on the speed of rubble movement may be subject to slight alterations, even at its assumed constant speed. Presented results of tests show that changes in mass of the tested concretes caused by abrasive erosion are conditioned by both their composition and intensity of the abrasive mixture influence .
5. Numerical model
In the course of the mixture of aggregate and water reaction on concrete, the work is performed, some part of which is dissipated, but some part causes destructions of material. After marking the part of the work inducing the increase of the volume unit energy of the abraded concrete by a x W, the critical value called activation energy by E, the material volume of higher energy than activation energy E can be expressed by the equation:
Differential dependence defining the mass decrement of the abraded concrete according to the work of the abrasive mixture W is expressed by the equation (5.2):
Precise kinetics definition of the abrasive concrete wear, expressed in formula 5.2, should be search by assuming the parameters spectrum: a1, a2, ... an, determining local properties of abraded material. Assuming the discrete spectrum of parameters a, defining local "resistances" of material composition to abrasive wear, the equation (5.2) takes the following form:
where: pi determines the probability of parameter ai occurrence.
After introducing the designation Q = M×
a, where M stands for abrasive power (W= M ×
t), the speed of concrete wear is defined by the following dependence:
Gamma distribution was assumed for the description of variable Q (a) and the following expression defining the kinetics of abrasive erosion of concrete was formulated:
Integrating the expression (5.5) within limits from t = 0 to t we obtain the dependence determining the mass decrement during the process of abrasion:
where: defines the expected value of parameter Q.
The expected value of the variable a is expressed by the equation:
Dependence (5.6) transformed to the form:
at the designated coefficient values a and l approximates properly the results of experimental examinations. According to the accepted assumptions, the mass decrements of experimental substances made of concrete characterized by settled, constant composition are the work function of the abrasive mixture W and the random variable a dependent on concrete composition. Concretes made from mixtures of different ratio w=w/c are characterized by various expected values of the random variable a (fig.3).
Fig 3: Changes of relative mass decrements of I series concrete samples according to the work of the abrasive mixture|
Higher values of parameter E(a) of concretes with higher ratio w show their lower resistance to abrasive erosion. When one considers different speeds of the abrasive mixture - series II, the concrete samples of settled, constant composition are characterized by one expected value of the random variable a: E(a)=1.6x10-5 kg/ kJ independent of rubble movement speed (fig.4).
Fig 4: Changes of relative mass decrements of II series concrete samples according to the work of the abrasive mixture|
- The innovative and unique achievement of this research work is revealed in the presented methodology of testing of concrete abrasion with the mixture of aggregate and water, simulating natural mechanisms typical for the environment. Providing the testing device with a system recording forces of rubble influence on the abraded concrete allows to describe quantitatively the current interactions and carry out a scientific analysis of the observed cause-effect relationships.
- The results of the laboratory tests show that the mass decrements of concretes made of the mixture consisted of settled, constant composition can be expressed in the function of the abrasive mixture work and the material parameter dependent on its composition and independent of the intensity of the environmental influence. The above results create perspectives of forecasting durability of hydraulic concrete structures exposed to abrasive erosion according to material conditions and hydraulic conditions of rubble movement.
- A. Małasiewicz, Ścieralność betonu hydrotechnicznego przy różnych kątach padania na próbkę rumowiska niesionego przez wodę, Rozprawy Hydrotechniczne, PAN, 32, 1973.
- A. Bania, Bestimmung des Abriebs und der Erosion von Betonen mittels eines Gesteinsstoff-Wassergemisches, Dissertation B, TH Wismar, 1989.
- A. Bania, Wpływ składu mieszaniny ściernej na wielkość erozji betonu, Zeszyty Naukowe PS, 478, 1992.
- E. Horszczaruk, Zużycie betonu wywołane działaniem mieszaniny kruszywa i wody, Praca doktorska, Szczecin, 1999.
- G. Haroske, Erosionsverschleiss an Betonoberflächen durch Geschiebetransport, Dissertation, TH Wismar 1998.
- T. C. Liu, Maintenance and Preservation of Concrete Structures, Report 3, Abrasion-Erosion Resistance of Concrete. Tech. Report No. C-78-4, U.S., Army Engineer Waterways Experiment Station, Vicksburg 1980.
- T. R. Naik, S. S. Singh, M.M. Hossain, Abrasion Resistance of Concrete as Influenced by Inclusion of Fly Ash, Cement and Concrete Research, vol.24, 1994.
- A. Nanni, Abrasion Resistance of Roller Compacted Concrete. ACI Materials Journal, Nov.-Dec. 1989.
- ASTM C 418-90: Standard Test Method for Abrasion Resistance of Concrete by Sandblasting, 1990.
- A. Misra, I. Finnie, On the size effect in abrasive and erosive wear, Wear 65,1981.
- ASTM C 779-89a: Standard Test Method for Abrasion Resistance of Horizontal Concrete Surfaces, 1989.