CC BY
9
ISSN 2617-2909 (print) ISSN 2617-2119 (online)
J ourn. Geol. Geograph.
Geology, 28(3), 504-510. doi: 10.15421/111946
T. P. Mokritskaya, L.O. Nosova Journ. Geol. Geograph. Geoecology, 28(3), 504-510.
Forecasting suffusion deformation in dispersive soils
T.P. Mokrytskaya, L.O. Nosova
Oles Honchar Dnipro National University, Dnipro, Ukraine, mokritska@i.ua
Journal of Geology, Geography and Geoecology
Journal home page: geology-dnu-dp.ua
Received: 08.06.2019 Abstract. Suffusion is a dangerous geological process accompanied by the formation of
Received in revised form-. 24.06.2019 sinkholes, deformations of buildings and structures. Forecast of suffusion processes is a Accepted: WM7.2019 complex scientific problem since it is required to predict a complex process of the formation
of new soil structure experiencing certain changes while interacting with the flow. During the period of 12.02.2018 - 13.04.2018, a sample of Dnipro loessial soil was filtered with the salt solution in Darcy device. The experiment was carried out at the SRI of Geology of Oles Honchar DNU; it is the continuation of the research dealing with experimental study of the dispersive soils as complex natural systems. To study suffusion process in loessial soils, analysis of microaggregate and granulometric composition of a sample of undisturbed structure of Dnipro loessial horizon taken in Tunelna ravine outcrop has been performed. The sample was studied in different states: natural and changed (due to long-term filtration) ones. Results of the use of innovative method to determine values of the function of particle distribution in terms of their mass and calculations of fractal dimension of the function basing upon the microaggregate analysis have made is possible to forecast values of porosity coefficient and volumetric deformation corresponding to the new quality standards of a complex system - dispersive soil in the state of complete microaggregate disturbance. Algorithm of the microaggregate composition analysis according to the methodology (Riashchenko, 2010) is in the fact that the sample experiences different methods of preparation - thus, evaluations of soil dispersivity are different. Basing upon the obtained results, values of microaggregate coefficient have been calculated, and data on the number of aggregates and initial particles have been taken. It has been determined that the basic size of the aggregates is 0.01-0.005 mm; there is a fewer share of the aggregates of 0.05-0.01 mm; and the fewest share of the aggregates is represented by fine fraction. That indicates the changes in microaggregate composition of soil due to the carrying out of fine fractions and the disturbance of larger aggregates. Calculations of the values of volumetric soil deformation due to long-term salt solution filtration emphasize the fact that within the zones of technogenic contamination, possible aggregate decay due to chemical effect will result in the formation of structure with denser particle packing, i.e. compaction. If loessial layers with the state changed due to salt solution filtration are subject to mechanical effect, drastic soil loosening and loss of soil stability may be observed.
Key words: suffusion, Dnipro loessial soil, fractal
Прогнозування деформацп суфозн в дисперсних грунтах
Т.П. Мокрицька, Л.О. Носова
Днтровсъкий нащоналъний утверситет iMeni Олеся Гончара, Днтро, Украта, mokritska@i.ua
Анотащя. Суфозш - небезпечний геолопчний процес, що супроводжуеться формуванням провалш, деформащями будiвель i споруд. Прогноз суфозшних процеав представляе складну наукову проблему, так як потрбно спрогнозувати складний процес формування ново! структуры Грунту що зазнае змш при взаемодй з потоком. На протязi двох мгсящв з 12.02.2018 р. по 13.04.2018 року зразок дншровського лесового Грунту був тдвернутий фшьтрацй розчином соли у приладi Дарсг Експеримент виконувався в НД1 геологи ДНУ iменi О.Гончара та е подовженням наукового напряму - експериментального дослщження дисперсних Грунтш як складних природних систем. Для дослщження процесу суфозй в лесовому Грунт був виконаний анатз мжроагрегатного та гранулометричного складу зразка непорушено! структури дшпровського лесового горизонту, що був вщбраний з вщслонення у б. Тунельно!. Зразок був дослщжений у рiзних станах: у природньому та у змшеному в наслщок тривало! фшьтрацй. За результатами застосування нового методу встановлення значень функцп розподшу часток за !х масою, розрахунюв фрактально! розмiрностi функцй на пiдставi мжроагрегатного анатзу доведена можливють прогнозу значень
коефщенту пористостi та об'емно! деформацй якi вiдповiдають новому якiсному стану складно! системи - дисперсному Грунту у сташ повного порушення мжроагрегапв. Алгоритм дослiджень мжроагрегатного складу за методикою (Riashchenko, 2010) полягае в тому, що до зразка застосовують рiзнi засоби пiдготовки, завдяки чому оцiнки дисперсност Грунту е рiзними. На пiдставi отриманих результатiв, були розраховаш значення коефiцiенту мiкроагрегатностi, отримаш даннi про кiлькiсть агрегатiв та первинних часток. Встановлено, що агрегати мають розмiр в основному 0,01-0,005 мм, агрегат розмiром 0,050,01 мм присутш у меншо! кшькосп та кiлькiсть агрегатш тонких фракцш найменша. Це вказуе на змши мiкроагрегатного складу Грунту внаслiдок виносу дрiбних фракцiй та на порушення агрегатш бiпьших за розмiром. Розрахунки значень об'емно! деформаци Грунту внаслщок тривало! фшьтраци розчином солей вказують на те, що у зонах техногенного забруднення можливий розпад мжроагрегатш внаслiдок хiмiчного впливу призведе до формування структури з бiльш плитною упаковкою часток, до стискання. Якщо лесовi горизонти зi змiненим станом внаслщок фшьтраци сольового розчину будуть пiдвергнути механiчному впливу можливим е розпушення Грунтш, можливим е рiзке розпушення та втрата стiйкостi Грунту.
Ключовi слова: суфозiя, днтровський лесовий Грунт, фрактал
Introduction. Development of the experimental studies of the dispersive soil properties is the basis for successful solution of investment problems in the process of construction, especially within the zones characterized by complex geological processes. Suffusion is not often analyzed in engineering and geological studies. Nevertheless, it is known that suffusion phenomena within the areas of the available loessial soils is widely occurred and connected with subsidental, landslide, and erosive developments. Special attention is paid to the consideration of the processes in soil medium with the signs of fractal. Thus, studies of the processes of turbulence (Jafari et al., 2019), percolation (Ziani, 2019), peculiarities of soil behaviour within the shear zone (Lu et al., 2018) are carried out on the basis of the fractal theory elements. Wang et al. (2015) applied calculations of fractal size to define regularities of spatial changes in the properties of loessial soils to implement innovative recultiva-tion methods. Liu et al., 2013, determined the connection between fractal volumetric size of natural soil particles and amount of organic substance and argillaceous particles. Suffusion methods are developed on the basis of classic techniques (Khuzhaerov, 1994), (Nikiforov, 2000). Research by Gudehus, & Touplikiotis, (2017) substantiates importance of the analysis of soils as natural fractals as for solving the problem of forecasting "building - soil" geotechnical system. Thus, review of literature sources indicates the necessity to study suffusion as a natural fractal. The paper represents the results of experimental research of the soils basing upon innovative methodology that helps predict suffusion deformation by applying fractal theory for the results of determination of microaggregate composition. Materials and methods. Dispersive loessial soil masses of technogenically built-up territories are in constant dynamic transformation resulting in re-construction of the microstructure and changed physical, physical-mechanical, and hydrogeological properties of those soils. Increased level of ground water, trans-
mission of constant and temporal static and dynamic loads, changes in thermal regime result in changing states of soil mass and soil properties both within the zone of complete water saturation and within the aeration zone.
Experimental studies of suffusion effect upon the loessial soils are carried out in terms of loessial soil sample of Dnipro horizon taken within the outcrop of Tunelna ravine erosive system. Geological section of the ravine slope under consideration down to 42.0 m from the land surface involves Quaternary deposits of loessial complex represented by layering of loamy soils and loamy sands, reddish-brown loamy soils, and underlying Neogene deposits - clays and sands. Ancient and modern landslides are recorded within the right slope of the ravine.
Soils samples under study were taken from the wall of modern landslide disruption and erosive-washed scarp slope. Physical and physical-mechanical characteristics were determined for Black Sea-Dofinivskyi eolian-deluvial loamy soils (vd PIII pc+df), Bug eolian-deluvial loamy sands (vd P III bg), and Dnipro eolian-deluvial loamy sands (vd P II dn). The soil layers are irregular in their thickness and strike. Within the upper share of the slope, loessial thickness is from 38.8 down to 27.7 m; within the medium share, the thickness is down to 10.6 m; and within the lower share, the thickness is down to 0.0 m. In terms of hydrogeology, one Quaternary waterbearing level located in the loessial deposits of the right ravine slope (vd-e PIII-II pl + kd, vd PII dn, e PII zv) is recorded within the right Tunelna ravine slope down to the prospected depth. Results and discussion. The research involves modeling of the suffusion processes due to long-term filtration involving 10% solution of NaCl salt (Fig. 1).
Data concerning changes in the filtration coefficient have been obtained, and calculations of salt content in the solution after the process stabilization have been performed. During the period of 12.02.2018 -13.04.2018, sample of Dnipro loessial soil was being
Fig. 1. Dynamics of changes in the filtration coefficient according to the experimental results in terms of stable
regime.
filtered in Darcy device. Layout and methodology of the research corresponded to the previously completed studies. Liquid volume was measured along with the determination of the filtrate composition in final phase of the experiment when stable regime was obtained (Fig.1). Calculations of the compositions of
Table 1. Results of chemical analysis of soil samples and filtrates.
different. Thus, sample one was subject to mechanical shaking during 2 hours, sample two was prepared with the addition of ammonia solution and boiled during one hour, and sample three was boiled during one hour with the addition of sodium pyrophosphate solution. After that, suspended matter was sampled from
Number ph hco-3 Mg2+ Ca2+ Cl- SO2- 4 Na+ + K+ i of ions, mg/ 100 g Dry residue, mg/100 g
mg-eq mg/ 100 g mg-eq mg/ 100 g mg-eq mg/ 100 g mg-eq mg/ 100 g mg-eq mg/ 100 g mg-eq mg/ 100 g
1 7.40 0.40 0.024 0.6 0.007 1.90 0.038 1.62 0.056 0.58 0.028 0.100 0.002 0.143 0.147
2 7.42 0.63 0.038 6.06 0.074 12.62 0.252 16.27 0.577 18.38 0.882 16.60 0.382 2.186 2.192
3 7.75 0.77 0.047 5.97 0.073 28.98 0.580 29.42 1.043 20.24 0.972 15.48 0.356 3.048 3.050
4 7.84 0.51 0.031 1.67 0.020 7.34 0.147 15.58 0.552 3.29 0.158 10.37 0.239 1.132 1.134
Table 2. Calculation of hypothetic salts.
Number Ca(HCO3)2 CaSO 4 MgSO4 Na2SO4 NaCl CaCl2 MgCl2
mg-eq % mg-eq % mg-eq % mg-eq % mg-eq % mg-eq % mg-eq %
1 0.80 15.38 1.16 22.31 - - - - 0.2 3.85 1.84 35.38 1.2 23.08
2 1.26 1.79 23.98 33.98 12.12 17.18 0.66 0.94 32.54 46.11 - - - -
3 1.54 1.53 40.48 40.13 - - - - 30.96 30.70 15.94 15.80 11.94 11.84
4 1.02 2.63 6.58 16.98 - - - - 20.74 53.51 7.08 18.27 3.34 8.61
hypothetic salts are represented below (Tables 1, 2).
Objective of the study was to determine changes in granulometric composition of the loessial soil sample due to long-term salt solution filtration. That process models the effect upon granulometric composition of chemical and mechanical suffusion. To study suffusion process within the loessial soil, microaggregate and granulometric composition of the regular-structure sample of Dnipro loessial horizon sampled within Tonelna ravine outcrop has been analyzed.
Algorithm of microaggregate study according to the methodology (Riashchenko, 2010) is as follows: the sample experiences different methods of preparation so that evaluations of soil dispersion would be
the depths of 25, 10, and 7 cm (in an hour interval) being determined by the standard and depending upon the temperature. Mass value of the average sample is used to define quantitative fraction content according to formula:
X=A*1000*(100-K)/(g0*V )
(1)
where, A - is mass of soil fraction within the sample of the known volume brought to the constant mass value, g;
g0 - is mass of absolute dry soil sample, initial value, g; Vn - is pipette capacity, cm3;
K - is total content of soil fraction of the size more
than 1 mm, %.
Results of the study of microaggregate composition of the sample being subject to long-term salt solution filtration have made it possible to determine certain changes in microaggregate soil composition. Percentage content of the particles according to the
Table 3. Percentage content of fraction particles.
have been defined. The obtained results have made it possible to develop cumulative curves of granulomet-ric composition of the samples having been prepared by three different methods. Those curves show the difference between the content of the particles of 0.10.05 and 0.05-0.01 mm depending upon the method
Method of sample preparation Percentage content of particles of the indicated fraction, mm
0.1-0.05 0.05-0.01 0.01-0.005 0.005-0.002 0.002-0.001 less than 0.001
Aggregate 43.353 21.763 31.433 0.000 2.713 3.274
Semi-disperse 33.580 64.053 0.000 2.037 3.030 0.095
Disperse 77.944 15.169 0.000 4.393 9.431 2.047
study results is represented below (Table 3).
Basing upon the obtained results, coefficient values of microaggregation have been calculated; data on the amount of aggregates and initial particles have been obtained. It has been determined that the aggregates size is mostly 0.01-0.005 mm; aggregates of 0.05-0.01 mm are found less often; and fine-fraction aggregates demonstrate the least amount. That indicates certain changes in microaggregate soil composition due to the carrying out of fine fractions and the disturbances of larger aggregates. Content of the initial particles and content of the particles in microaggregates of the corresponding fractions have
Table 4. Percentage content of the initial particles
of sample preparation (Fig 2).
According to the methodology (Mokrytska, Tushev et al., 2018), values of fractal size of the functions of particle distribution in terms of their mass have been calculated (Fig.3). Values of the determination coefficient of linear trend equations indicate considerable importance of the obtained equations. Value of angle coefficient shows the effect of the preparation method upon the value of fractal size of the function of particles distribution in terms of their mass.
Next stage involves calculation of the porosity coefficient value of soil in terms of complete decay
Parameter M1 M1-a M2 M2-a M3 M3-a M4 M4-a M5 M5-a M6 M6-a
Percentage content of particles, % 43,353 34,591 15,169 - - - - 4,393 2,713 6,717 2,047 -
Fraction size, mm 0,10,05 0,050,01 0,010,005 0,0050,002 0,0020,001 Less 0,001
Note for Table 4:
M1 - content of the initial particles; M1-a - content of the particles in aggregates.
Table 5. Results of the calculation of forecast values of the porosity coefficient
Method of sample preparation for the analysis Value of fractal size of the function of particles distribution in terms of their mass D Claculation coefficient value Forecast value of the porosity coefficient k'
Aggregate 0.189 0.738 0.961
Disperse 0.071 0.892 0.668
Semi-disperse 0.212 0.711 1.019
been calculated (Table 4).
No availability of the initial particles of fine argillaceous composition and general decrease in the amount of fine particles comparing to the natural state
of samples microaggregates having experienced mechanical or chemical action of different degree (Table 5). Those values are the forecast ones indicating porosity value of the soil medium having experienced
N,
Fig. 2. Cumulative curves of granulometric composition of the sample of Dnipro loessial horizon having been subject to long-term salt solution filtration Notes for Fig. 2:
N, % - total number of particles; ln R - logarithm of particle size;
Row 1 - aggregate method of sample preparation (mechanical shaking during 2 hours);
Row 2 - semi-disperse method of sample preparation (with the addition of ammonia solution);
Row 3 - disperse method of sample preparation (with the addition of sodium pyrophosphate solution).
Table 6. Forecast values of the deformation of the sample of Dnipro loessial horizon.
Results of calculation of forecast deformation of the sample of Dnipro loessial horizon vdP II dn
Method of sample preparation Value of fractal size of the function of particles distribution in terms of their mass Porosity coefficient of a natural-state sample Porosity coefficient of a changed-state sample Volumetric sample deformation
Aggregate preparation 2.669 0.706 0.74 -0.034
Preparation with magnetic shaker 2.706 0.718 0.741 -0.023
Preparation as long-term water solution filtration 2.605 0.711 0.737 -0.027
Table 7. Forecast deformation values due to salt solution filtration
Method of soil preparation Porosity coefficient of a natural-state sample Porosity coefficient of a changed-state sample Volumetric sample deformation
Aggregate preparation 0.711 0.96 -0.249
Semi-disperse preparation 1.019 -0.308
Disperse preparation 0.668 0.04
long-term salt solution filtration and mechanical or chemical effect resulting in complete decay of all the microaggregates.
Comparison of the obtained values with the porosity coefficient values of natural-state soil (Table 5) and changed soil (Table 7) demonstrates the expected consequences of the effect of suffusion
process upon the porosity of Dnipro loessial horizon sample.
While analyzing the data, it should be noted that disturbance of microaggregate s results in the formation of looser structure; in this context, filtration is less important factor than the action of magnetic field or mechanical shaking.
ln v
■ +J5
^--
3.75
■ ■
-1-1-1-1-1-1-1-,
0 -] -1 -3 -4-5-6 -7 -8
LtiR
' Row l ■ Row 2 • Row J
Fig.3. Graphs of the dependence of logarithm of particle distribution function in terms of the mass upon the logarithm of particle sizes.
Note for Fig. 3:
Ln V, % - total volume of particles;
Ln R - logarithm of particle size;
Row 1 - aggregate method of sample preparation;
Row 2 - disperse method of sample preparation;
Row 3 - semi-disperse method of sample preparation.
According to the results of previous studies due to filtration with water solution, increased coefficient of soil porosity should be expected. Calculations of the values of soil volumetric deformation due to long-term filtration with salt solution show the fact that possible microaggregates decay within the zones of technogenic contamination due to chemical effect will result in the formation of structure with denser particles packing, i.e. compaction.
If loessial horizons with the changed state due to salt solution filtration are subject to mechanical effect, there is possible soil loosening (even, drastic loosening) and loss of soil stability.
Thus, comparison of the results shows considerable changes in the soil state due to suffusion within the zones of technogenic contamination. Certain regularities of changes in soil porosity in the state of complete degradation in terms of complete decay of microaggregates due to mechanical suffusion or processes of solution and carrying out of particles during salt solution filtration have been defined. It has been demonstrated that density deformation is possible only in the context of disperse preparation in terms of maximum decay of microaggregates. Conclusions and their discussion. The paper represents the results of long-term experiment on determining regularities of changes in loessial soil state within the zones of technogenic action.
Effect of chemical composition of the solution upon the regularities of changes in granulometric and microaggregate soil composition has been confirmed.
It has been proved that suffusion may result in loss of soil strength and compaction.
For the first time, forecast values of the porosity coefficient of loessial Dnipro horizon have been obtained; the values will correspond to new soil state.
For the first time, the experimental results have made it possible to prove the possibility to obtain forecast values of the properties of the medium experiencing certain changes due to suffusion.
References
Gudehus G., Touplikiotis A., 2018. On the stability of geotechnical systems and its fractal progressive loss Acta Geotech. 13: 317. https://doi. org/10.1007/s11440-017-0549-x Ziani H., Abbeche, K., Messaoudene, I. et al., 2019. KSCE, J Civ Eng 23: 1028. https://doi.org/10.1007/ s12205-019-0051-0 Jafari F. K., Asgari M. S., Pishkoo A. The Fractal Calculus for Fractal Materials, 2019. Fractal Fract, 3(1), 8; https://doi.org/10.3390/fractalfract3010008 Khuzhaerov B. J., 1994. A model of multicomponent grouting and suffosion filtration Eng Phys Thermophys, Vol. 66, Issue 4, pp. 373-379. https://doi.org/10.1007/BF00853459 Liu Y., Gong Y., Wang X. et al., 2013. Volume fractal dimension of soil particles and relationships with soil physical-chemical properties and plant species diversity in an alpine grassland under different disturbance degrees Arid Land 5: 480. https://doi.org/10.1007/s40333-013-0184-9 Lu S., Tang, H., Zhang, Y. et al., 2018. Bull Effects of the particle-size distribution on the micro and macro behavior of soils: fractal dimension as an indicator of the spatial variability of a slip zone in
a landslide Eng Geol Environ 77: 665. https://doi. org/10.1007/s10064-017-1028-1 Mokritskaya T.P., , Tushev A.V., Samoylich K.A. et al., 2019. Bull Eng Geol Environ 78: 3729. https:// doi.org/10.1007/s10064-018-1361-z Nikiforov A.I., , 2000. J Modeling of suffosion of waterbearing strata Eng Phys Thermophys 73: 959. Vol. 73, Issue 5, pp. 959-965 https://doi.org/10.1007/ BF02681587
Wang G. L. J., Xuan L., Xinhui L., Xiaosai L., Yaqiong R., Jing W., 2018. Liming Dong Partitioning and geochemical fractions of heavy metals from
geogenic and anthropogenic sources in various soil particle size fractions. Geoderma,Vol. 312, 15 February, Pages 104-113. https://doi. org/10.1016/j.geoderma.2017.10.013 Wang J., Zhang M., Bai Z. et al., 2015. Multi-fractal characteristics of the particle distribution of reconstructed soils and the relationship between soil properties and multi-fractal parameters in an opencast coal-mine dump in a loess area. Environ Earth Sci 73: 4749. https://doi.org/10.1007/ s12665-014-3761-0
