Open access peer-reviewed chapter

Radioactive Isotopes in Soils and Their Impact on Plant Growth

Written By

Jelena Markovic and Svetlana Stevovic

Submitted: 21 June 2018 Reviewed: 05 October 2018 Published: 20 March 2019

DOI: 10.5772/intechopen.81881

From the Edited Volume

Metals in Soil - Contamination and Remediation

Edited by Zinnat Ara Begum, Ismail M. M. Rahman and Hiroshi Hasegawa

Chapter metrics overview

1,609 Chapter Downloads

View Full Metrics

Abstract

In 1999, Serbia was bombarded by NATO. One of the cities most affected by the consequences of bombardment with uranium is the city of Vranje, where the consequences are felt even today. Due to the influence of uranium, the mortality rate has increased. This paper presents the effects of some of the radionuclides that have contaminated the soil, as well as the connection between soil and plants that grow on that soil. The performed measurements of radionuclides (226Ra, 40K, 232Th, 238U, and 235U). The results show that the content of each of these radionuclides has different concentrations, but what is important is that some values are even below the detection limit, corn <0.06 235U on the location Korbevac and wheat <0.04 235U on the location Bujkovac. On the three and all of these gated locations, the calculated values of the transfer factors for 40K were in the range of 0.144–0.392, while in the case of 226Ra, the transfer factors ranged from 0.008 to 0.074. Only one value (0.051) was obtained for the transfer factor of 232Th. Specific activities of 137Cs, as well as uranium isotopes, in all the investigated cereal samples, were below minimal detectable activity concentrations. The ratio of radionuclides in soil and plants is of great importance for human nutrition.

Keywords

  • soil
  • plants
  • radioactive isotopes
  • monitoring
  • mortality

1. Introduction

Natural radioactivity in the environment, originating from the naturally occurring radionuclides of 232Th, 238U, and 235U radioactive series and 40K, largely contributes to the natural irradiation of man and biota, which can be external and/or internal (ingestion and inhalation). Natural radionuclides land characteristic of α and β radioactive decay [1]. The biggest number of radionuclides belongs to a radioactive series, which naturally has three. These three series start as radioisotope, so-called parent: 238U (series of 4n + 2), 235U (series of 4n + 3), and 232Th (string 4n). A series of successive radioactive decays occur from parents whose offspring core is also unstable and is subject to decay. The process of disintegration ends stable isotopes, and for those strings to the 206Pb, 207Pb, and 208Pb, respectively, from the radionuclide which does not belong to any of the radioactive series, the most important is the soil radionuclide 40K. Gamma radiation created during the radioactive decay of uranium and thorium series, as well as 40K, largely contributes to the natural irradiation of alive council (man and biota), which can be external and/or internal (ingestion and inhalation) [2]. The concentration of natural radionuclides depends on the composition of the soil. According to the report of the UNSCEAR, the medium activity concentration of 238U, 232Th, and 40K in the soil in the world amounts 33.45 and 412 mCi kg-1, respectively. The ranges of concentrations of 238U, 232Th, and 40K in the soil in Europe are 2–330, 2–190, and 40–1650 Bq kg−1 [3].

Besides the natural radionuclides, due to various human activities, different manmade radionuclides entered the environment. The most significant among them is 137Cs (T1/2 = 30y), found in the environment mostly as a result of the nuclear tests in the 1960s ties and the Chernobyl nuclear plant accident in 1986 [4]. 137Cs is bound in the surface layers of soil and is washed out and redistributed in the ecosystem for a longer period of time due to its long half-life. Thus, only a small amount of it is present in plants today. It is well-known that 137Cs isotopes take important part in the environment, due to their good assimilation by plants, which are used to feed the animals and finally human beings [5]. The reported values of 137Cs in the agricultural soil in the north part of Serbia, on several locations near the city of Novi Sad, are in the range of 1.5–12.6 Bq/kg [6].

Soil is a complex material composed of mineral (inorganic) as well as an organic matter that originated from plant decomposition. It is a compact matter providing necessary micro- and macro-nutrition elements for plants to function and grow. Cereals as wheat, corn and barley are important component of everyday human diet [21]. Most of the radionuclide cereals are absorbed from soil, so the values of transfer factors are important in the studies of the transport and distribution of radionuclides in the “soil–plant–animal-human” chain, as well as in the evaluation of the radiation risk [7]. Transfer factors (TF) are crucial in the radionuclide transport models, in the environment as well as in the evaluation of the level of the specific activities of radionuclides in agricultural crops [8]. The main factors determining the level of TF are radionuclide itself, type of plant, type (physical and chemical characteristics) of soil, concentrations of stable chemical elements in soils [9], as well the local climate [10].The values of transfer factors should provide the basis for theoretical analysis on the different uptakes of elements not involved in physiological and biochemical processes in plants [11].

The main goal of this paper was to investigate transfer factor, because it can give crucial information about the possible quantity of radioactive and other toxic materials endangered for human’s health. Starting from the soil, through the plants, they enter in the food chain and consequently they reach in the human body and affect mortality. The transport processes in the “soil–plant” systems for radionuclides (226Ra,40K, 232Th, 238U, 235U) and 137Cs in the Pcinja District. Pcinja District is located in the southern part of the Republic of Serbia. It covers the city of Vranje and the municipalities Vladicin Han, Surdulica, Bosilegrad, Trgoviste, Bujanovac, and Presevo (Figure 1). The Pcinja District has not been investigated yet.

Figure 1.

Pcinja District.

Advertisement

2. Methodology

The samples of soils and cereals were collected in 2014, in the area of the city of Vranje, on three locations: the villages of Bujkovac, Korbevac, and Suvi Dol. The type of soils was the same on all the locations (the so-called gajnjaca). Gajnjaca belongs to well-drained soils, its chemical characteristics depending on the level of utilization, degree of erosion, chemical characteristics of the main substrate, and level of development. The content of humus in the gajnjaca soils is in the range of 2–5%. This type of soil in neutral or low acetous has a high capacity of adsorption, and the dominant ions in it are Ca and Mg. Its color is brown, reddish, or red depending of the content of aluminum and iron. It is very suitable for farming, wine growing, and afforestation.

2.1 Sampling sites and sample collection

About 11 samples of cultivated and uncultivated soils and 7 samples of cereals were collected. Sampling sites coordinates are present in Table 1. The samples of soils were taken from different depths that also differ from one to another sampling site. The depths that the soils were sampled from were 0–5 cm, 0–10 cm, and 0–20 cm on the sites Korbevac and Bujkovac and 0–5 cm, 5–10 cm, and 10–15 cm at the sampling site Suvi Dol. The samples of grain were taken from plots where the soil samples were taken. First, soil samples were taken for testing, on the land-sown cereals that are taken when they are ripe for examination. The position of the sampling sites is presented in Figure 2.

Site Coordinates Elevation
(m)
Sampling date
(North latitude) (East longitude)
Bujkovac 42°33′26” 22°00′35” 718 09.11.2014.
Korbevac 42°23′06” 21°44′24” 441 05.11.2014.
Suvi Dol 42°33′07” 21°56′05” 359 11.11.2014.

Table 1.

Coordinates of the location of soil samples.

Figure 2.

Sampling sites on the territory of the city of Vranje.

The radioactivity of the samples of soils was determined by gamma spectrometry in the Institute for Nuclear Sciences “Vinca” in the Laboratory for Radiation and Environmental Protection.

The mass of cereal and soil samples for analysis is necessarily 1 kg. The samples of soils were cleaned of mechanical impurities, stones and plant material, and dried at 105°C for 24 h. Samples of cereals were dried at room temperature and mineralized at 450°C. Soils were measured in Marinelli geometry (volume 500 ml) and cereals in cylinder bottles (volume 125 ml). The radioactive equilibrium was achieved in all the samples, as they have been sealed by bee wax and left for 30 days before measuring. The samples of grain were taken at the stage of full maturity of the technology and to hand. Samples of cereals (fruit cereal) were dried in air at room temperature for at least 3 weeks and then were crushed and mineralized at a temperature of 450°C for 24 hours, dry-ashing method [12].

The specific activity of natural 226Ra was determined by analyzing the spectra of its daughters [13], 214Pb and 214Bi, at the energies of 295, 352, 609, 1120, and 1764 keV [14]. Radionuclide 232Th was determined by its daughter 228Ac at the energies of 338 and 911 keV. The activities of 40K and 137Cs were determined at the energies of 1460 and 661.6 keV, respectively. The activity of 235U and 238U is determined by establishing a radiochemical equilibrium between 226Ra and 214Bi using photo-peak at energies around 186 keV-a [15, 16].

2.2 Standard gamma spectrometry

The gamma spectrometry was performed on three HPGe detectors (CANBERRA) with relative efficiencies of 18, 20, and 50%; the resolution of all of the detectors was 1.8 keV at 1332 keV. For the samples of soils, the detectors were calibrated by a reference radioactive material—a silicone resin matrix, Czech Metrological Institute, Praha, 9031-OL-420/12, total activity 41.48 kBq on 31.08.2012 (241Am, 109Cd, 139Ce, 57Co, 60Co, 203Hg, 88Y, 113Sn, 85Sr 137Cs). The gamma-spectrometric measurements of radioactivity in soil samples was used and the ultra-low-background germanium detector-type GMX (extended energy range from 10 keV to 3 MeV-a manufacturer of ORTEC, the nominal efficiency of 32% in passive and active protection). Passive safety lead is made up of a thickness of 12 cm in the form of a cylinder and coated with a layer of tin and copper. The active protection (veto detectors) is the five plastic scintillation detectors which are anticoincidence mode working with HPGe detector and completely cover passive protection. Active protection lowers integral countdown in the background of a factor of three for the range from 50 to 2800 keV, which lowers the threshold of detection and is suitable for the measurement of environmental samples [17]. For cereal samples the detectors were calibrated with a secondary reference radioactive material in plastic boxes (volume 125 cm3) obtained from the primary reference radioactive material—Czech Metrological Institute, Praha, 9031-OL-427/12, type ERX, total activity 72.40 kBq on 31.08.2012 (241Am, 109Cd, 139Ce, 57Co, 60Co, 203Hg, 88Y, 113Sn, 85Sr 137Cs, 210Pb) [19].

The counting time was 60,000 s. The results are presented with the expanded measuring uncertainty for the factor k = 2, level of confidence for normal distribution 95%.

2.3 Transfer factor calculations

Transfer factor (TF) was calculated according to Eq. (1), defined as the ratio of specific activity of radionuclide in plant (Bq/kg dry matter) and specific activity in soil (Bq/kg) [18]:

TF = A p A s E1

where Ap is the specific activity of the radionuclide in plant [Bq/kg dry matter], and As is the specific activity of the radionuclide in soil [Bq/kg].

The change absorbed dose intensity into absorbed dose rate of gamma radiation from the natural radionuclides in soil was calculated according to Eq. (2). The annual effective dose was calculated according to Eq. (3).

D nGyh 1 = 0.462 × CRa + 0.604 × CTh + 0.0417 × CK E2

where CRa is the specific activity of 226Ra in soil, CTh is the specific activity of 232Th in soil, and CK is the specific activity of 40K in soil.

DE mSν = 0 . 7SνGy 1 x 0.2 x 365 x 24 x D E3

Advertisement

3. Results and discussion

The results of the gamma spectrometry analysis of soils at different locations (sampling sites) are presented in Table 2.

Bq/kg
Depth
(cm)
226Ra 232Th 40K 238U 235U 137Cs
Korbevac
0–5 43 ± 3 55 ± 4 730 ± 50 47 ± 8 2.7 ± 0.2 16 ± 1
0–10 45 ± 3 54 ± 4 730 ± 50 51 ± 9 2.4 ± 0.2 16 ± 1
0–20 38 ± 3 51 ± 4 690 ± 40 40 ± 8 2.4 ± 0.2 15 ± 1
Suvi Dol
0–5 38 ± 3 52 ± 4 490 ± 30 35 ± 8 1.7 ± 0.1 10.1 ± 0.7
5–10 33 ± 2 48 ± 3 470 ± 30 34 ± 9 1.7 ± 0.2 7.9 ± 0.6
10–15 37 ± 3 50 ± 3 460 ± 30 34 ± 8 1.9 ± 0.2 7.2 ± 0.5
Bujkovac
0–5 22 ± 2 30 ± 2 500 ± 30 25 ± 8 1.6 ± 0.2 17 ± 1
0–10 23 ± 2 30 ± 2 510 ± 30 25 ± 7 1.5 ± 0.1 18 ± 1
0–20 25 ± 2 29 ± 2 520 ± 30 22 ± 8 1.1 ± 0.1 17 ± 1

Table 2.

Specific activity of radionuclides in soil samples at different depths and sampling sites [Bq/kg].

The results of the calculated absorbed dose intensity and the annual effective doses from natural radionuclides in soils are presented in the table and in Table 3.

Depth
(cm)
D(nGyh−1) DE(mSν)
Korbevac
0–5 83.53 0.102
0–10 85.85 0.105
0–20 73.89 0.091
Suvi Dol
0–5 69.39 0.085
5–10 63.84 0.078
10–15 66.48 0.081
Bujkovac
0–5 49.13 0.061
0–10 50.75 0.062
0–20 50.01 0.061

Table 3.

Absorbed dose intensity D(nGyh−1) and the annual effective doses DE(mSν) from natural radionuclides in soils.

There are no significant differences among the specific activities of natural radionuclides in soils regarding the sampling depth of the soil at the specific location, i.e., the differences are within the measuring uncertainty. The same applies for the specific activities of 137Cs—their values do not differ significantly regarding the sampling depth of the soil at the specific location. As it has been detected only in traces, it does not present a risk of being accumulated in plants and human diet [20].

For all of the locations, the specific activities of 226Ra are in the range of 22–45 Bq/kg, while for 232Th the values are in the range of 29–55 Bq/kg. For 40K, the specific activities cover the interval from 460 to 730 Bq/kg, for 238U the activities are in the range of 22–51 Bq/kg, and for 235U in the range of 1.1–2.7 Bq/kg. The specific activities of 137Cs cover the interval of 7.2–17 Bq/kg. The uneven distribution of cesium within the same area is mainly due to the relocation and washing out effects in the soil.

There are no significant differences among the specific activities of natural radionuclides between the locations (Table 2). The lowest values of the specific activities for 226Ra, 232Th, 238U, and 235U are obtained at Bujkovac, the highest ones at Korbevac. The values are within the range of the literature data of the specific activities of natural radionuclides in soils reported for the region of former Yugoslavia [6]. Compared to the other locations, the specific activity of 226Ra is lower only in soils sampled at Bujkovac.

The values of the calculated absorbed dose intensity are in the range of 49.13–85.85 nGy/h, while the annual effective doses range from 0.061 to 0.105 mSv/h and are within the values reported for other regions in the country [6].

The results of the levels of natural radionuclides and 137Cs in cereals are presented in Table 4.

(Bq/kg)
Sample 226Ra 232Th 40K 238U 235U 137Cs
Korbevac
Wheat 2.2 ± 0.4 2.6 ± 0.8 150 ± 10 < 2 < 0.2 < 0.06
Corn 0.4 ± 0.1 < 0.2 108 ± 7 < 1 < 0.06 < 0.03
Suvi Dol
Wheat 0.30 ± 0.07 < 0.1 106 ± 7 < 0.6 < 0.04 < 0.02
Corn < 0.2 < 0.2 68 ± 5 <1 < 0.09 < 0.03
Bujkovac
Wheat 0.37 ± 0.07 < 0.2 102 ± 7 < 1 < 0.04 < 0.02
Corn 1.4 ± 0.3 < 0.4 89 ± 7 < 2 < 0.1 < 0.06
Barley 1.7 ± 0.2 < 0.2 200 ± 10 < 2 < 0.1 < 0.07

Table 4.

Specific activity of radionuclides in grain samples [Bq/kg dry matter].

Table 5 presents the means of the specific activities of the radionuclides in cereals sampled on the investigated locations.

Radionuclide Cereals (Bq/kg)
Mean value Interval
Min Max
226Ra 1.06 MDA* 2.2
232Th 2.6 MDA 2.6
40K 118 68 200
238U The values are under MDA
235U
137Cs

Table 5.

Mean values of the radionuclides’ specific activities in cereals [Bq/kg dry matter].

MDA—minimal detection limit.


The specific activity of 232Th (2.6 Bq/kg dry matter) presented in Table 5 refers only to the sample of wheat from the village of Korbevac. In all the other samples of cereals, the specific activity of this radionuclide is under MDA. The specific activities of 238U, 235U, and 137Cs in all investigated samples of cereals are under the MDA, too.

The values of calculated transfer factors are presented in Table 6. The values of the specific activity in soil used to calculate the transfer factors were the mean specific activity of the radionuclides for the different sampling depth at the location.

Transfer factor
Cereal 226Rap/226Ras 232Thp/232Ths 40Kp/40Ks
Korbevac
Wheat 0.052 0.051 0.209
Corn 0.009 0.151
Suvi Dol
Wheat 0.008 0.224
Corn 0.144
Bujkovac
Wheat 0.016 0.200
Corn 0.061 0.174
Barley 0.074 0.392

Table 6.

Value of transfer factor for cereals.

As some of the obtained values of the radionuclides, specific activities in cereals were under MDA; transfer factors were calculated only for 40K, 226Ra, and 232Th. The calculated values of the transfer factors for cereals indicate that 40K and 226Ra are the main radionuclides transferred into the cereal grain. The TF for 40K (0.144–0.392) are higher than the TF for 226Ra and 232Th by an order of magnitude (0.00–80.074 for TF (226Ra)). The TF for 40K can be rather high, as is known and reported in the literature [4]. Other radionuclides do not accumulate in the plant in more significant amounts [12]. This is mostly due to the discrimination in uptake of essential and nonessential elements, exhibited by the plant [12]. Also, it is reported that small percentage of the total activity found in the plant is accumulated in the root system, while 1–16% is accumulated in the grain [15]. The addition of phosphate to soil reduces the availability of thorium for root uptake through the formation of phosphate salts that have low solubility [15]. Regression analysis, reported in [15], showed that thorium availability to wheat was negatively related to soil pH and positively related to soil organic matter, cationic exchange capacity, and clay content. In comparison to the literature, it can be seen that the obtained TF for cereals in Pcinja region are in agreement with the results obtained in other parts of the world [7, 12, 16], while they are lower by the order of magnitude in comparison to the TF reported for the plants that are principally grass pasture, where the stem and leaves were analyzed (TF(Ra) = 0.17, TF(Th) = 0.058, TF(K) = 1.3 [17]).

Advertisement

4. Conclusion and recommendation

The specific activities of the radionuclides in soil, at all the investigated locations, were in the range from 22 to 45 Bq/kg for 226Ra, from 29 to 55 Bq/kg for 232Th, 460 to 730 Bq/kg for 40K, from 22 to 51 Bq/kg for 238U, from 1.1 to 2.7 Bq/kg for 235U, and from 7.2 to 17 Bq/kg for 137Cs. The obtained specific activities for 236Ra, 232Th, and 40K in “gajnjaca” soil are in good agreement with the values obtained for the other types of soils [4]. The differences between the specific activities of a radionuclide in soil samples from different depths are within the measuring uncertainties, and the ratio of specific activities for 235U/238U suggests the natural origin of uranium. The activities of radionuclides in cereals also do not differ from the values obtained by other authors. Distribution of radionuclides from the soil into the plant depends on the bioavailability of minerals in the soil, the root structure of the investigated plant and the processes in the plant tissue. The calculated values of TF for cereals indicate that 40K and 226Ra are the main radionuclides that are transferred in cereals. This evaluation is most important for the production of foodstuffs diet with low contents of radionuclides.

On the location of Korbevac, Suvi Dol, and Bujkovac the calculated values of TF for 40K were in the range of 0.144–0.392; for 226Ra the values of transfer factors were in the range of 0.008–0.074. It should be noted that the evaluated activities refer to the content of radionuclides in dry plant matter and that the activities in the fresh plants are on the average four to five times lower due to the water content. For other natural radionuclides and for 137Cs, the TF have not been calculated as the specific activities of these radionuclides in cereals were under the MDA. The results presented in this paper are the preliminary investigations of the contents of radionuclides in soils and cereals in the region of Pcinja. As the transfer factors in the “soil-cereal” system were determined only for the specific type of soil, the investigations should continue for other types of soils and cereals mostly used in animal and human diet. The measurements presented in this manuscript are the first to be conducted in the region of Pcinja, thus providing the results that can be used as a baseline for the future measurements and monitoring.

References

  1. 1. Gilmor GR. Practical Gamma-Ray Spectrometry. New York, USA: John Wiley and Sons; 2008
  2. 2. UNCSEAR. (United Nations Scientific Committee on the Effects of Atomic Radiation), Report to the general assembly with Scientific Annexes. Annex B: Exposure of the Public and Workers from Various Sources of Radiation. New York, USA; 2010
  3. 3. Bossew P, Kirchner G. Modeling the vertical distribution of radionuclides in soil. Part 1. The converse dispersion equation revisited. Journal of Environmental Radioactivity. 2004;73:127-150
  4. 4. Solecki J, Chibowski S. Determination of transfer factors for 137Cs and 90Sr isotopes in soil–plant system. Journal of Radioanalytical and Nuclear Chemistry. 2002;252(1):89-93
  5. 5. Karacan F. The simple radiochemical determination of 90Sr in environmental solid samples by solvent extraction. Journal of Radioanalytical and Nuclear Chemistry. 2011;288:685-691
  6. 6. Bikit I et al. Monitoring of Radioactivity in Soil on the Territory of Novi Sad during the Year 2012. Faculty of Science, University of Novi Sad; 2012. (in Serbian)
  7. 7. Selvasekarapandian S et al. Natural radionuclide distribution in soil Gudlaore, India. Applied Radiation and Isotopes. 2000;52(2):299-306
  8. 8. Pulhani VA et al. Uptake and distribution of natural radioactivity in wheat plants from soil. Journal of Environmental Radioactivity. 2005;79(3):331-346
  9. 9. Schimmack W et al. Spatial variability of fallout– 90Sr in soil and vegetation of an alpine pasture. Journal of Environmental Radioactivity. 2003;65(3):281-296
  10. 10. Abu-Khadra et al. Transfer Factor of Radioactive Cs and Sr from Egyptian Soils to Roots and Leafs of Wheat Plant. IX Radiation & Protection Conference, Egypt; 2008
  11. 11. Bikit M et al. Transportation of Natural Radionuclides from the Soil into Plants. Proceedings of XIII International Symposium on Protection against Radiation, Becici. 1995. pp. 221-224 (in Serbian)
  12. 12. Sarap N, Rajacic M, Jankovic M, Nikolic J, Dolijanovic Z, Todorovic D, Pantelic G. The Distribution of the Total Beta and Specific Activities of 40K in Vegetative and Generative Organs of Maize. Proceedings of the XXVIII DZZSCG, Institute of Nuclear Sciences Vinca; 2015. pp. 70-73
  13. 13. Popovic D et al. Contents of radionuclides in soils in Serbia: Dose calculations and environmental risk assessment. Advances in Environmental Research. 2012;6:91-134 (in Serbian)
  14. 14. Dragovic S et al. Assessment of gamma dose rates from terrestrial exposure in Serbia and Montenegro. Radiation Protection Dosimetry. 2006;121:297-302 (in Serbian)
  15. 15. Roesseler CE, Smith ZA, Bolch WE, Prince RJ. Uranium and radium 226 in Florida phosphate materials. Health Physics. 1979;37:269-277
  16. 16. Kim KH, Burnet WC. 226Ra in Phosphate nodules from the Peru/Chile seafloor. Geochimica et Cosmochimica Acta. 1985;49:1073-1081
  17. 17. Bikit I, Slivka J, Mrdja D, Zikic-Todorovic N, Curcic S, Varga E, et al. Simple Method for Depleted Uranium Determination. Japanese Journal of Applied Physics (JJAP). 2003:5269-5273. (in Serbian)
  18. 18. Vandenhove H et al. Proposal for new best estimates of soil-to-plant transfer factor of U, Th, Ra, Pb and Po. Journal of Environmental Radioactivity. 2009;100(9):721-732
  19. 19. Leo WR. Techniques for Nuclear and Particle Physics Experiments. Berlin Heidelberg, New York: Springer-Verlag; 1994
  20. 20. Vukasinovic I et al. Depth distribution of 137Cs in anthrosol from the experimental field “Radmilovac” near Belgrade, Serbia. Archives of Industrial Hygiene and Toxicology. 2013;64:425-430. (in Serbian)
  21. 21. Džoljić J, Stevović S, Todorović D, Polavder S, Rajačić M, Nikolić JK. Natural and artificial radioactivity in some protected areas of South East Europe. Nuclear Technology & Radiation Protection Journal;XXXII(4 (December 2017)). ISSN 1451-3994

Written By

Jelena Markovic and Svetlana Stevovic

Submitted: 21 June 2018 Reviewed: 05 October 2018 Published: 20 March 2019