Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The use of a regression model in the variability analysis of the leachate quality from heaps of the production of the building materials

The use of a regression model in the variability analysis of the leachate quality from heaps of... An attempt to evaluate potential hazard for quality of manufactured construction elements and concrete in case of use of leachate as technological water was made in this paper. This may lead to negative impact on strength parameters of the prod‑ uct made in the analysed company. Raw industrial leachate must not be discharged into sewerage system due to increased concentrations of some mineral components; it must be initially purified. Selected elements of multiple regression module and principal components analysis were used to describe the above phenomena. The model we have developed uses selected significant parameters of pollutants contained in leachate from a process industrial waste heap. The model takes into account leachate conductivity and sulphate ions, in which increased lead compounds concentration was determined. The model was verified during its use. The variables used in the model explain the significant percentage of variability noted for the dependent variable. Keywords Leachate · Lead · Multiple regression · Principal components analysis Introduction established for identification of particular concrete quality parameters such as strength, flexibility, hardness and work ‑ Manufacturing of construction materials, in particular ability. It has been proved that concrete wash water featured concrete, is associated with generation of solid waste and increased alkalinity and solids content going beyond the lim‑ wastewater. Washing of concrete mixers and units generates its of popular standard ASTMC94, which caused that manu‑ wastewater polluting water and soil. Concrete washing water factured concrete was more porous. This limits the interest features high pH fluctuating between 11 and 12 due to high in application of technological water recycling mostly in the limestone content in the solution. This water contains dis‑ case of use of various admixtures in concrete production solved solids such as hydroxides and sulphates from cement, ASTM C94 (Asadollahfardi et al. 2015; Meena and Luhar chlorides from application of calcium chloride admixture, 2019). oils and greases from plant and machinery as well as small Leachates generated during construction materials pro‑ amounts of other chemicals associated with Portland cement duction fed into technological water make an additional hydration and derivates from chemical admixtures. Many source of pollution of technical concrete and prefabricated publications contain descriptions of use of concrete wash elements (Bonaccorsi et al. 2004; Chen et al. 2009; Dai et al. water for making mortar and concrete (Meena and Luhar 2009). Leachates fed into the manufactured product feature 2019; Vaičiukynienė et al. 2021; Taghizadeh et al. 2021; increased concentration of mineral components such as sul‑ Varshney et al. 2021). However, such materials must com‑ phates, carbonates, chlorides, nitrates and compounds of ply with the quality standards. The standards had been such heavy metals as chromium, lead, nickel, copper, cad‑ mium and zinc (Kurdowski 2010; Szymański et al. 2018). Examples of such waste and leachate management were * Beata Janowska found in one of the companies manufacturing construction beata.janowska@tu.koszalin.pl materials (Magistri and D’Arcangelo 2008). In the EU countries, the Landfill Directive, the Waste Faculty of Civil Engineering, Environmental and Geodetic Sciences Department of Water, Sewage and Waste Framework Directive 2008/98/EC, the Urban Wastewater Technology, Koszalin University of Technology, ul. Treatment Regulations Council Directive 99/31/EC and the Śniadeckich 2, 75‑453 Koszalin, Poland Vol.:(0123456789) 1 3 97 Page 2 of 14 Applied Water Science (2022) 12:97 Water Framework Directive 2000/60/EC belong to the main 2002; Nocuń‑ Wczelik and Łój 2006; Czarnecki et al. 2015; European regulations concerning waste landfilling and lea‑ Nalet and Nonat 2016). This has also impact on processes of chates management. They put an obligation on the mem‑ interaction between particular parameters/components con‑ ber states to regularly monitor leachate condition. Despite tained in the generated leachate that is reused (Król 2012; implementation of the said regulations, the scope of the Czarnecki et al. 2015) in replenishment water. Leachates parameters to be monitored has not been strictly defined. may also have negative impact on the natural environment It has been only stated that the analysed substances should unless subjected to a pre‑treatment before discharging to be selected based on the composition of the landfilled waste a sewerage system, in particular to the ground or aquatic (Brennan et al. 2016, 2017; Wdowczyk and Szymańska‑ environment (Magistri and D’Arcangelo 2008; Król 2012; Pulikowska 2020; Ida and Eva 2021). Czarnecki et al. 2015). In all member states, the regulations concerning the mon‑ This paper illustrates an effort to assess potential hazards itoring of wastewater discharge exist on the national level; for product (structural elements) quality and the phenom‑ subsequently, they are being supplemented or defined at the ena that occur in leachates from industrial waste dump. The regional or local level. For example, in Germany, Austria, objective of the research work was to determine statistically Romania, Ukraine and Czech Republic testing of heave met‑ the significant impact of particular components (independent als content is monitored only if there occurs a strong impact variables) on concentration of one of toxic lead compounds of industrial waste that is channelled via a sewerage system in tested leachate. Selected elements of mathematical statis‑ to any municipal wastewater treatment plant. Slovakia, Hun‑ tics in the form of the multiple regression module and main gary, Romania and Moldova keep monitoring heavy metals components analysis were used to describe those phenomena content in a more regular way within the self‑monitoring (Stanisz 2001). procedure. In Ireland, the parameters to be analysed in land‑ fill leachates are determined individually in the waste dis‑ posal licence. Their scope depends on the type and composi‑ Tested material and research methodology tion of the landfilled waste (Naveen et al. 2017; Wdowczyk and Szymańska‑Pulikowska 2020; Ida and Eva 2021). Materials and scope of research In the USA, the Environmental Protection Agency regu‑ lates wastewater discharge and treatment based on the Clean The leachate quality tests performed at the production unit Water Act (CWA). The National Pollution Discharge Elimi‑ site allowed to determine basic physicochemical indicators nation System (NPDES) issues licences for all discharges as well as heavy metals and alkaline leachates sampled for and wastewater treatment plants as well as sets the norms testing (Figs.1, 2) including: regarding quality of wastewater discharged to surface water and to municipal wastewater treatment plants. In China, • Nearby a sewer sump in A1 and A2 locations, detailed regulations regarding heavy metals content in From existing leachate pond near the construction waste wastewater and drinking water are also set at the national heap at A3 and A4 locations. level. In China, limits of the admissible heavy metals con‑ tent have been defined by the Chinese Ministry of Health (CMH) and The National Standard of China, in Kenya—by the National Environment Management Authority (NEMA) and Kenya Bureau of Standards (Kinuthia et al. 2020). In Canada, the standards and regulations pertaining to the col‑ lection, protection, treatment and elimination of leachates are very strict; they also put emphasis on regular monitoring of surface and ground waters nearby any landfill location (Naveen et al. 2017). According to Polish law, the concentration of the said metals must not exceed the admissible values in leachates discharged also to the aquatic environment or ground as well as to sewerage systems (Regulation 2006; Regulation A2 2019). Discharge of leachates featuring higher concen‑ A1 tration of those metals than pure replenishing water may catalyse generation of new crystallographic structures in the manufactured material having consequently negative Fig. 1 Leachate from the washing area of the production unit impact on strength parameters of the material (Baetzner 1 3 Applied Water Science (2022) 12:97 Page 3 of 14 97 described in the literature (Kowal and Świderska‑ Bróż 1996). Particular parameters of leachates (technological water) were being determined using mainly standard research meth‑ ods in accordance with Polish standards (Table 1). The content of total organic carbon (TOC) were deter‑ mined using the Vario Max CN macroanalyser. Samples of leachates for testing were mineralised using a mixture of acids (65% HN O 70% HClO and 30% H O ) using micro‑ 3, 4 2 2 wave energy (Milestone 1200 Mega apparatus). The contents of heavy metals and alkali metals were determined using FAAS technique (iCE 3500Z Thermo Scientific SOLAAR) A3 A4 (Czerniak and Poszyler‑ Adamska 2006; Szymański et al. 2018). Statistical analysis Fig. 2 Leachate from the industrial waste heap To evaluate tested pollutants, components occurring in par‑ ticular samples of leachate from concrete production STA- Locations A1 and A2 are specific for leachate originating TISTICA—multiple regression module—were used (Stanisz from the washing facility of the concrete production unit, 2001). To describe correlations between particular variables, whereas A3 and A4 are located at leachate pond nearby the an estimator of the correlation coefficient for tested features, waste heap. The tests were performed during 2014– as r correlation, was used. For this purpose, Fischer’s test 2018 period. Five (5) test series were effected in total. (F) verifying the hypothesis that the multiple correlation coefficient is significantly different from “0” was useful; this means that at least one model structural parameter (without Analytical methodology and test results free term) is different from zero. A number of notations that are specific for the analy ‑ Particular determinations were performed in accordance sis of results originating from statistical method research with Regulation of the Minister of Maritime Economy and were used in the presented paper (Stanisz 2001) and they Inland Navigation of 12 July 2019 Journal of Laws 2019 are: R —a multiple determination coefficient measuring item 1311, Regulation of the Minister of Construction Sec‑ which proportion of variation of the dependent variable is tor of 14 July 2006 Journal of Laws 2006 No 136, item 964 explained by the regression model calculated from the equa‑ (Regulation 2006; Regulation 2019) as well as methods tion: R = 1‑ [Rest SS/ Total SS], where Rest SS is a sum of Table 1 Standard research methods Parameter Analytical procedures—standard Analytical devices pH PN‑EN ISO 10523:2012 inoLab Multi Level 1 WTW Conductivity PN‑EN 27,888:1999 inoLab Multi Level 1 WTW Ammonium nitrogen PN‑ISO 5664:2002 – Nitrate (V) nitrogen PN‑C‑04576–08:1982 UV/Vis 6715 spectrophotometer JENWAY Nitrate (III) nitrogen PN‑C‑04576–08:1982 UV/Vis 6715 spectrophotometer JENWAY Kjeldahl total nitrogen Titration method PN‑EN 25,663:2001 SCHOTT instruments titroline easy Chlorides PN‑ISO 9297:1994 – Total phosphorus PN‑EN ISO 6878:2006 UV/Vis 6715 spectrophotometer JENWAY Sulphates VI PN‑ISO 9280:2002 – Total dissolved solids PN‑EN 15,216:2010 – Total solids PN‑EN 872:2007 – Total suspended solids PN‑C‑04559–03:1972 – BOD (manometric method, PN‑EN 1899–2:2002 – COD Bichromate method PN‑ISO 15705:2005 – Cr 1 3 97 Page 4 of 14 Applied Water Science (2022) 12:97 squares error and Total SS is a total sum of squares. The represents strong negative correlation whereas + 1.00 2 2 multiple correlation coefficient is a root of R . Corrected R value represents strong positive correlation. Value 2 2 is interpreted just like R . The corrected R coefficient takes r = 0.00 indicates absence of the correlation. into account a variable number of the degrees of freedom for Total SS and Rest SS. The estimation standard error S provides information Results and discussion about the average quantity of empirical deviations of the dependent variable from the model‑derived value, defined Results of physicochemical analysis by formula: � The scope of the said research work is illustrated in Tables 2 e and 3. Table 2 comprises test results for basic physicochemi‑ i=1 S = cal indicators, whereas Table 3 comprises test results for n − 2 heavy and alkali metals. where the estimation standard error pertains to: y —inde‑ Tests performed on leachates from construction waste pendent variable values. y ̂ —model‑derived independent heap and technological devices of the concrete production variable value. e = y —y ̂ rests. S = estimation standard plant indicated increased contents of some heavy metals i i i e error. including chromium, lead and iron compounds. However, no mercury compounds were found therein. Zinc, cadmium, The statistics F (Fischer’s test) is used to verify the sig‑ nickel and manganese occurred in trace concentrations. Cal‑ nificance of the entire model. cium, and to a much lesser extent magnesium, dominated in The critical level of significance (p ) is a probability of the alkali metals group. The said leachates were also used for making an error consisting in refusal of the true null the preparation of technological water. Some volume of the hypothesis based on the observed sample values. Low leachates was being discharged to the combined sewerage values of the significance level (usually α < 0.05) indicate system. Environmental protection services had been indicat‑ low risk whilst the decision to reject the null hypothesis ing unlawful management of the leachates (Regulation 2006; is made. Regulation 2019). The correlation is a measure of association between two The research results indicate environmental protection or more variables. The correlation of two variables can problems as well as improper use of leachate for produc‑ be expressed numerically using the linear correlation tion of concrete and concrete products originating from that coefficient (r ). The linear correlation coefficient takes company. As it appears from Tables 2 and 3, such exces‑ values from − 1.00 to + 1.00 interval. The – 1.00 value sive values (average daily values) pertained, in accordance Table 2 Averaged results of Parameter Unit of measure Tested during the period of 2014—2018 physicochemical analysis of leachate samples Sample No 1 2 3 4 5 pH – 7.66 12.02 12.60 12.59 11.21 Conductivity μS/cm 318 1698 4560 6180 3189 Ammonium nitrogen mgN /dm 0.16 0.57 0.22 0.05 0.25 NH4 Nitrate (V) nitrogen mgN /dm 0.10 0.24 0.18 0.19 0.17 NO3 Nitrate (III) nitrogen mgN /dm 0.002 0.020 0.000 0.002 0.006 NO2 Kjeldahl total nitrogen mgN/dm 6.91 2.51 0.64 0.73 2.69 Chlorides mgCl/dm 18 28 42 50 34 Total phosphorus mgP/dm 0.98 1.71 0.68 1.52 1.22 Sulphates VI mgSO /dm 41.2 226.4 308.7 679.1 313.8 Tot. dissolved solids mg/dm 400 970 1500 1600 1117 Total solids mg/dm 4200 1200 10,250 4600 5062 Tot. suspended solids mg/dm 2700 363 8600 2400 3515 BOD mgO /dm 596 134 40 30 200 5 2 COD mgO /dm 3980 100 400 125 1151 Cr 2 TOC mgC/dm 1492 37 150 46 431 1 3 Applied Water Science (2022) 12:97 Page 5 of 14 97 Table 3 Averaged content of heavy and alkali metals in leachate sam‑ ions—Fig.  5 (correlation coefficient r = 0.986). Peculiar ples [mg/dm ] character of copper behaviour may also be interpreted as formation of this metal complexes with organic compounds. Parameter Tested during the period of 2014–2018 At the same time, the increase in the BOD load has shown Sample No negative correlation with Cu and positive correlation with 1 2 3 4 5 magnesium (r = 0.997; Tab.3) and iron (r = 0.990; Fig. 8.) ions. This does not mean that all heavy and alkali metals Cu 0.003 0.009 0.011 0.019 0.01 made similar combinations. Cd < 0.001 < 0.001 < 0.001 < 0.001 0.001 Thus, such leachate must not be drained into waters or Cr 0.006 0.020 0.363 0.019 0.102 ground nor used again as technological water for concrete Ni < 0.004 < 0.004 < 0.004 < 0.004 0.004 production (Regulation 2019). Due to occurrence of rela‑ Zn 0.003 < 0.001 < 0.001 < 0.001 0.001 tively high amounts of chromium and lead compounds in the Pb 0.005 0.015 0.025 0.048 0.023 leachate, it must not be used as a technological water com‑ Mn 0.014 < 0.001 < 0.001 < 0.001 0.004 ponent. Increased concentrations of heavy metals were also Fe 0.181 0.047 0.03 0.043 0.075 noted in leachate sampling locations A1 and A4, which can Ca 35.37 89.76 197.48 407.53 182.58 originate from higher volume of sewage sludge occurring Mg 3.33 0.06 0.077 0.045 0.878 there. Consequently, production of construction elements using technological water containing such leachate may with the above‑mentioned regulations, to: pH—sample No cause occurrence of unfavourable phenomena in produced 3 (12.60), suspended solids—sample No 3 (8600 mg/dm ), materials; therefore, there is a deterioration of their strength BOD —sample No 1 (596  mgO /dm), COD —sample No parameters (Czerniak and Poszyler‑ Adamska 2006; Gineys 5 2 Cr 3 3 1 (3980 mg/O dm ), TOC—sample No 1 (1492 mgC/dm ), 2011; Król 2012). sulphates—sample No 4 (676.1  mgSO /dm ) and calcium— Table 3, sample No 4 (407.53 mgCa/dm ). Statistical analysis of laboratory test results The test results presented in Table  2 indicate that lea‑ chate samples taken for testing at point A1 (Fig. 1) were Table 4 shows selected significant linear correlation coef‑ exposed to direct contact with the ground and vegetal mate‑ ficients. It was noted that there occurred a clear positive rial. Therefore, organic substances containing humic com‑ correlation of lead with: sulphates (r = 0.99), chlor ides pounds could migrate to the leachates. This has been indi‑ (r = 0.95), total dissolved solids (r = 0.88) and electrolytic cated by increased loads of BOD COD and TOC. It can conductivity (r = 0.96). High correlation was noted for chro‑ 5, Cr be reckoned that humic compounds formed metal–organic mium and total solids (r = 0.90) as well as total suspended combinations with, for example, iron and magnesium. solids (r = 0.93) (Table 4). Also significant negative values This may explain their high correlations. Occurrence of of the coefficient of correlation between iron and pH reaction calcium compounds favoured combinations with sulphate (r = − 0.99), total dissolved solids (r = − 0.89) and positive Table 4 Averaged linear Cu Cr Pb Fe Ca Mg correlation coefficients pH 0.798965 0.419140 0.718243 − 0.996361 0.660810 − 0.993451 Conductivity 0.943712 0.364665 0.959130 − 0.741467 0.954364 − 0.720229 Ammonium nitrogen − 0.341047 − 0.075248 − 483582 − 0.225064 − 0.558863 − 0.264870 Nitrate V nitrogen 0.518972 0.063616 0.378452 − 0.848332 0.294735 − 0.890233 Nitrate III nitrogen − 0.150449 − 0.407748 − 0.293376 − 0.209573 − 0.367645 − 0.285229 Kjeldahl total nitrogen − 0.841892 − 0.497105 − 0.786683 0.972432 − 0.742257 0.955850 Chlorides 0.950641 0.373938 0.953405 − 0.790957 0.941617 − 0.772290 Total phosphorus 0.354433 − 0.736170 0.259427 − 0.241784 0.209849 − 0.345671 Sulphates VI 0.992818 0.012041 0.995156 − 0.665358 0.985989 − 0.682669 Tot. dissolved solids 0.902790 0.487761 0.882325 − 0.890282 0.856769 − 0.866226 Total solids 0.145817 0.908333 0.216883 − 0.251460 0.246244 − 0.147146 Tot. suspended solids 0.055536 0.948552 0.113288 − 0.253781 0.135769 − 0.147503 BOD − 0.826407 − 0.430217 − 0.754290 0.990457 − 0.701250 0.984658 COD − 0.766111 − 0.299968 − 0.667138 0.984893 − 0.600994 0.997867 Cr TOC − 0.766313 − 0.299531 − 0.667350 0.984820 − 0.601212 0.997840 1 3 97 Page 6 of 14 Applied Water Science (2022) 12:97 with total nitrogen (r = 0.97) were found. High correlation of lead in leachate varies from approx. 0.005 to 0.048 mg Pb/ of iron with BOD (r = 0.99), COD (r = 0.98) and TOC dm , which partly correlates with sulphate VI ions indicator 5 Cr (r = 0.98) was noted. Strong correlations between total sus‑ within the limits from 41.2 to 679.1 mgSO/ dm . Strong pended solids and chromium (r = 0.92) were also noted— correlations for remaining components of the leachate are this suggests that chromium compounds are sorbed, to a illustrated in subsequent graphs presented in Figs. 4, 5, 6, major extent, by suspended solid particles. High correlation 7, 8. between lead compounds and sulphate VI ions (r = 0.96) Calcium compounds in examined technological water the indicates a probability of generation of sparingly soluble main element of alkaline components in leachate maintained PbSO . at the production plant. Occurrence of calcium compounds The determination coefficient calculated for lead and in cement, therefore, in leachate containing mineral additives sulphate compounds is R = 0.997. This association defines such as volatile silica ashes or granular blast furnace slag, which part of variability of the dependent feature can be containing hydrated calcium silicates, may lead to a synthe‑ interpreted using the developed model (Tab. 5). Examples sis of amorphous or semi‑ crystalline structures (Baetzner of statistical analysis results are illustrated in relevant graphs 2002; Batchelor 2006; Kudrowski 2010). It is believed that 4− (Figs. 3, 4, 5, 6, 7, 8). Particular correlations shown in those such structures composed of calcium ions and [SiO ] tetra‑ figures illustrate categorised graphs of dispersion along with hedrons are then significantly developed. In practical terms, regression line. Histograms of the variable distribution, with the increase in the CaO content in solution, the increase at frequency of occurrence of given parameter, were also in the Ca/Si ratio in the solid phase can be observed (Kur‑ depicted here. For example, Fig. 3 shows that concentration dowski 2010). It appears from the research work performed on leachates that this environment may also contain calcium compounds in CaSO form as well as lead compounds in Table 5 Multiple regression results sparingly soluble PbSO form, which was confirmed by 2− 2+ 2+ very high correlation of SO with Ca and Pb ions Statistics Value (r = ~ 0.99). Multiple correlation coefficient 0.9987 Particular graphs shown in Figs. 3, 4, 5, 6, 7, 8 are sup‑ Determination coefficient R 0.9975 plemented with histograms categorised with relations to Corrected determination coefficient R 0.9950 particular groups. Those phenomena were interpreted by F(2.2) 401.56 the multiple regression module made for parameters of the P 0.00248 Fig. 3 Categorised graph of Pb = 0.00174 + 0.68E-4 * sulphates dispersion with regression line correlation: r =0.995 for lead and sulphates 0,07 0,06 0,05 0,04 0,03 0,02 0,01 0,00 -0,01 -200 0 200 400 600 800 1000 02 4 2- Concentration SO [mg/dm ] 0.95 CI 1 3 Concentration Pb [mg/dm ] Applied Water Science (2022) 12:97 Page 7 of 14 97 Fig. 4 Categorised graph of Pb = -0.0191 +0.00123 *chlorides dispersion with regression line Correlation: r =0.953 for lead and chlorides 0,07 0,06 0,05 0,04 0,03 0,02 0,01 0,00 -0,01 51015202530354045505560 02 4 Concentration Cl [mg/dm ] 0.95 CI Fig. 5 Categorised graph of Ca = -7.226 + 0.60480 * sulphates dispersion with regression line correlation: r = 0.986 for calcium and sulphates -100 -200 0200 400600 8001000 02 4 2- Concentration SO [mg/dm ] 0.95 CI pollutants occurring in tested leachate originating from tech‑ contained in leachate from the industrial waste heap based nical concrete and construction elements production. on the example of electrolytic conductivity and sulphate This resulted in development of a model taking into ions, in which high concentration of lead compounds was account selected, significant parameters of pollutants found. Using this model, its verification was carried out. 1 3 Concentration Ca Concentration Pb 3 3 [mg/dm ] [mg/dm ] 97 Page 8 of 14 Applied Water Science (2022) 12:97 Fig. 6 Categorised graph of dis‑ Cr = -0.0618 +0.47E-4 * total suspended solids persion with regression line for correlation: r = 0.948 chromium and total suspended solids 0,45 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 0,00 -0,05 -0,10 -2000 2000 6000 10000 024 0 4000 8000 12000 Concentration total suspended solids [mg/dm ] 0.95 CI Fig. 7 Categorised graph of Cu = 0.00271 + 0.25E-4 * sulphates dispersion with regression line correlation: r = 0.993 for copper and sulphates 0,024 0,022 0,020 0,018 0,016 0,014 0,012 0,010 0,008 0,006 0,004 0,002 0,000 -0,002 -200 0 200 400 600 8001000 024 2- Concentration SO [mg/dm ] 0.95 CI originating from concrete and construction elements pro‑ Multiple regression module duction, allows for Fischer’s test (F) to be used. Lead is a toxic element; its content in cement and lea‑ Multiple regression, used as a tool in evaluation of mutual relationships between particular components of leachate chates is significant—it can migrate to any aquatic envi‑ ronment. High correlation between high Pb content and 1 3 Concetration Cu Concentration Cr 3 3 [mg/dm ] [mg/dm ] Applied Water Science (2022) 12:97 Page 9 of 14 97 Fig. 8 Categorised graph of Fe = 0.02284 + 0.26E-3 * BOD dispersion with regression line correlation: r = 0.990 for iron and BOD 0,22 0,20 0,18 0,16 0,14 0,12 0,10 0,08 0,06 0,04 0,02 0,00 -0,02 -100 0 100 200 300 400 500 600 700 800 02 4 Concetration BOD [mgO /dm ] 0.95 CI conductivity as well as sulphates (VI) content allowed for Those variables independently explain approximately 91% model development (Tab. 4). As a result, a model taking of the dependent variable variances after exclusion of impact into account some selected significant parameters of pol‑ of other independent variables. lutants contained in leachate originating from construction The adopted model, based on the correlation coefficient elements production, was developed. The said model takes values, can be written in the following form: into account leachate electrolytic conductivity and concen‑ Pb =  + conductivity +  sulphates (1) 0 1 2 tration of sulphate ions in this medium. Using this model, its verification was carried out. The use of statistical analysis for elaboration of test results allowed for finding of significant relationships Fischer’s test (F) between particular pollutants, which allowed for the defini‑ tion of: determination coefficient, corrected determination Table 6. contains basic parameters of statistical evaluation coefficient, F (Fischer’s) coefficient, critical significance including: the multiple correlation coefficient, determina ‑ level p and estimation standard error. Particular variables tion coefficient, corrected determination coefficient, F coef‑ in the model explain the significant percentage of the ficient, p critical significance level and estimation standard dependent variable (Tab. 5). The model significance was error. F statistic was used to check relevance of the entire additionally demonstrated based on the analysis of correla‑ model, where: tions between Pb measured values and theoretical values F (F-statistic)—statistic of the test verifying the hypoth‑ originating from Eq. (1). In the case of application of the esis that the multiple correlation coefficient is significantly regression analysis to evaluation of mutual relationships different from “0”, which means that at least one of the between particular leachate components several, and even a model structural parameters (without free term) is different dozen or so, factors can be considered—then, the multiple from zero. regression comes into play. The factors that were selected p (p-value)—critical significance level of the test verify ‑ in our case were the solution electrolytic conductivity and ing the hypothesis that the multiple correlation coefficient lead compounds concentration. This allowed for develop‑ is significantly different from “0”. ment of a model describing mutual relationships between Performed Fischer’s test F indicated the significance level those components. In the given case, the highest partial cor‑ p < 0.05; therefore, it required that at least one coefficient of relation coefficients of lead and sulphate ions compounds Eq. (1): β β or β should differ from zero. Estimated values 0, 1 2 was 0.995, whereas for the electrolytic conductivity: 0.959. of those coefficients in given case amount to b = 0.001291, 1 3 Concentration Fe [mg/dm ] 97 Page 10 of 14 Applied Water Science (2022) 12:97 Table 6 Summary of regression N = 5 Summary of the dependent variable regression b* Std. err. of b* b Std. err. of b t (2) p Free term 0.001291 0.000930 1.387607 0.299640 Conductivity 0.235808 0.098069 0.000002 0.000001 2.404511 0.138034 Sulphates 0.775100 0.098069 0.000053 0.000007 7.903602 0.015634 b*—are standardised beta coefficients for respective independent variables b = − 0.000002 and b = 0.000053 respectively. Therefore, the form of loadings of variable factors in the so‑called unit 1 2 there are two cases when p < 0.05. Based on these calcula‑ circle. To do this, an analysis taking into account a correla‑ tions, Eq. (2) was formulated as a model describing values tion matrix was performed. The internal area of the unit of the correlation coefficients; it can be written in the fol‑ circle represents admissible factor loadings. This means lowing form: that the closer to the circle is located the drawn variable, the better is its representation by graph factors. The major‑ Pb = 0.001291 − 0.000002 conductivity + 0.000053 sulphate ity of examined parameters of leachate mineral pollutants (2) (Fig. 9) (without heavy metals) is located in the circle. In Null hypothesis: H : b = 0, b = 0, b = 0 (none inde‑ 0 0 1 2 many cases, those correlations are very strong. An example pendent variable has significant impact on the dependent of correlation are associations of nitrogen (without ammo‑ variable). nium nitrogen) with B OD, COD , TOC, suspended matter, 5 Cr Alternative hypothesis: H : There is at least one b coef‑ 1 i total solids and significantly weaker between total phospho‑ ficient different from zero. rus and sulphates, chlorides, electrolytic conductivity and For the assumed significance level α = 0.05, the null pH. Sets of those variables show positive correlation. Such hypothesis p = 0.015634 < α = 0.05 can be rejected. correlation could not be observed in the case of heavy and Conclusion: there is an independent variable, which has alkali metals, which originated, partly, from low concentra‑ a significant impact on the dependent variable. tions of those leachate pollutants (Fig. 10). Nevertheless, It appears from the above considerations that using sta‑ quite close positive correlations between some heavy met‑ tistical methods for given set of data, significant impact of als and alkaline calcium could be seen. Negative correla‑ the independent variable on the dependent variable can be tions occurred between magnesium and chromium as well confirmed. as between nickel and manganese, cadmium, zinc and iron. This should be interpreted by absent, or occurring in trace concentrations, nickel. It is beyond any doubt that strong Principal components analysis correlation between lead compounds and sulphate as well as chloride ions indicated a possibility of generation of spar‑ Apart from use of descriptive statistics for data analysis, it was also possible to make graphs of 2 W type variables of ingly soluble compounds such as: PbSO (II) or PbCl (II) 4 2 and also CaSO as well as its hydrated form CaSO × 2H O. factor loadings. In this particular case, two variables (for X 4 4 2 and Y axes) were selected; then, a two‑dimensional disper ‑ It is quite possible that in alkaline leachate (pH > 8) copper compounds of Cu(OH) × 5H O type could be generated. sion graph was created. It is possible to obtain cumulative 2 2 eigenvalues and cumulative variance percentage. Part of This allowed for making of two graphs associated with the analysis of principal components, whereof one 2 W graph of the variance was isolated here by particular components. Those relationships are illustrated in Fig. 9. It can be noted factor loadings for objects (Fig. 9) and the second one 2 W graph of variable factors loadings (Fig. 10). Both graphs can that the component showing the highest eigenvalue explains 50.61% of total variance. The second one is 39.73% of total be aggregated and then a single graph, called biplot (Fig. 11), is obtained. This graph allows for penetration of the experi‑ variance. Therefore, two first components giving twodimen ‑ ‑ sional space explaining in total more than 90% variances mental data set structure, which facilitates evaluation of mutual interactions occurring in tested leachate and forecast were considered as significant. Such situation allowed for further conduct of the analysis solely based on two factor possible structures obtained through use of the leachate as a medium applied in the preparation of technological water. loadings. The newly created space, to which original points (variables and objects) were projected, made possible crea‑ Evaluation of such phenomena requires research by applica‑ tion of solid materials analysis techniques, for example XRF tion of proper data structure allowing to obtain the most significant information. X‑ ray fluorescence spectroscopy or XRD X ‑ ray diffraction, DTA differential thermal analysis and TG thermogravimetric Figure 9 shows an illustration of statistical analysis results for leachate from production of construction materials, in analysis (Lesko et al. 2001). The statistical analysis can be 1 3 Applied Water Science (2022) 12:97 Page 11 of 14 97 Fig. 9 Principal components Variables (axes F1 and F2: 90,34 %) analysis—projection of vari‑ Sulphates VI ables set in 2D factor loading Total phosphorus space Chlorides Ca Conductivity Pb Totaldissolved solids Cu 0.75 pH 0.5 Ni 0.25 Fe Hg Nitrate V nitrogen -0.25 Kjeldahl total nitrogen Total solids COD Mg BOD Cd -0.5 DOC Totalsuspended solids Nitrate III nitrogen Zn Mn Cr -0.75 Ammonium nitrogen -1 -1 -0.75-0.5-0.25 00.250.5 0.75 1 F1 (50,61 %) Fig. 10 Principal component Biplot (axes F1 and F2: 97,48 %) analysis (biplot)—projection of variables set and objects cluster‑ ing in 2D factor loading space Ca A3 Ni 3 Pb Cu Hg A1 Fe -1 Mg A2 -2 A4 Cr Zn -3 Cd Mn -4 -5 -6 -5 -4 -3 -2 -1 012345 6 F1 (64,38 %) 1 3 F2 (33,09 %) F2 (39,73 %) 97 Page 12 of 14 Applied Water Science (2022) 12:97 Fig. 11 Principal component Observaons (axes F1 and F2: 90,34 %) plot showing objects (A1‑ A4) clustering in 2D space (F1 and 6 A3 F2 factors scores) -1 A1 -2 A2 A4 -3 -6 -5 -4 -3 -2 -1 0123456 F1 (50,61 %) deepened due to outliers, applying the Bartlett’s test (Bartlett technological water for production of technical concrete 1954). The level of confidence p associated with the analysis and concrete elements. This may lead to negative impact of principal components equal to 0.00248 allows for such type on strength parameters of products manufactured in the operation. A preliminary analysis of projections of original analysed company. objects on the plane created by two first components (F 1 and F2), factor 1 and factor 2, was then performed (Lebart et al. • Raw post‑production leachate must not be discharged 1982). Already a superficial analysis indicates occurrence of into the sewerage system due to increased concentra‑ outliers, for example, Ni, Cr as well as ammonium nitrogen tions of some mineral components; it should be initially (Figs. 10, 11). Such outliers may have a negative impact on the purified. analysis result. Nickel occurred here in trace amounts whereas • A model was developed taking into account selected sig‑ Cr, with relation to other metals, in significant concentrations. nificant parameters of pollutants contained in leachate It can be noted that ammonium nitrogen NH ions were not from the industrial waste heap based on the example − − completely transformed into nitrate ions NO or NO , which of electrolytic conductivity and sulphate ions, in which 3 2 indicated also oxygen deficit in analysed leachate. Definitely, increased concentration of lead compounds was deter‑ the best way was to group the object factor loadings of the mined. Using this model, its verification was carried out. basic physicochemical analysis cases and selected groups of • The variables used in the model explain significant per ‑ heavy and alkali metals with locations of leachate sampling centage of variability noted for the dependent variable. (A1 and A4). The averaged concentration of pollutants in those • Inclusion of additional dependent variables into the locations had the highest value. At the same time, the high‑ model did not improve significantly particular model est amounts of accumulated sewage sludge were noted there, parameters. which originated from nature of the substratum, which played Model significance was additionally verified based on the the role of a sedimentation facility. analysis of correlations between Pb measured values and sulphates as well as electrolytic conductivity. • To describe the phenomena occurring in leachate from Conclusions construction materials production, elements of mathe‑ matical statistic, in the form of multiple regression mod‑ Examination of leachate from existing leachate pond ule and principal components analysis, can be used. located nearby construction waste heap indicated that the said leachate must not be used as a component of 1 3 F2 (39,73 %) Applied Water Science (2022) 12:97 Page 13 of 14 97 Funding The authors received no specific funding for this work. Czerniak A, Poszyler‑ Adamska A (2006) Assessment of heavy met‑ als immobilisation in cement‑ground composites used for con‑ struction of country roads. Acta Sci Pol Formatio Circumiectus Availability of data and material Data collected by the author and sec‑ 5(1):29–38 ondary data are cited. Dai Y, Qian G, Cao Y, Chi Y, Xu Y, Zhou J, Liu Q, Xu ZP, Qiao S (2009) Effective removal and fixation of Cr(VI) from aque‑ Declaration ous solution with Friedel’s salt. J Hazard Mater 170:1086–1092. https:// doi. org/ 10. 1016/j. jhazm at. 2009. 05. 070 Conflict of interest On behalf of all authors, the corresponding author Gineys T (2011) Influence de la teneur en elements métaliques du states that there is no conflict of interest. clinker sur les proprieties techniques et environnementales du ciment Portland–these. Dissertation, Université Lille Nord France Ethics approval Authors firmly abide by all ethical norms and follow Ida S, Eva T (2021) Removal of heavy metals during primary treatment the guidelines set by COPE. This article does not contain any studies of municipal wastewater and possibilities of enhanced removal: a with human participants or animals performed by any of the authors. review. Water 13:1121. https:// doi. org/ 10. 3390/ w1308 1121 Kinuthia GK, Ngure V, Beti D, Lugalia R, Wangila A, Kamau L (2020) Levels of heavy metals in wastewater and soil samples Open Access This article is licensed under a Creative Commons Attri‑ from open drainage channels in Nairobi, Kenya: community bution 4.0 International License, which permits use, sharing, adapta‑ health implication. Sci Rep 10:8434. https:// doi. or g/ 10. 1038/ tion, distribution and reproduction in any medium or format, as long s41598‑ 020‑ 65359‑5 as you give appropriate credit to the original author(s) and the source, Kowal LA, Świderska ‑ Bróż M, (1996) Water purification. PWN, provide a link to the Creative Commons licence, and indicate if changes Warsaw were made. The images or other third party material in this article are Król A (2012) Release of heavy metals from mineral composites con‑ included in the article's Creative Commons licence, unless indicated sidering environmental impact. Politechnika Opolska, Opole otherwise in a credit line to the material. If material is not included in Kurdowski W (2010) Concrete chemistry. PWN, Warsaw the article's Creative Commons licence and your intended use is not Lebart L, Morineau A, Fénelon JP (1982) Traitement des données permitted by statutory regulation or exceeds the permitted use, you will statistiques, méthodes et programmes. Dunod, Paris need to obtain permission directly from the copyright holder. To view a Lesko S, Leśniewska E, Nonat A, Mutin JC, Goudonnet JP (2001) copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . Investigation by atomic force microscopy of forces at the origin of cement cohesion. Ultramicroscopy 86:11–21 Magistri M, D’Arcangelo P (2008) New reducing agent for cement. Z K G 3:61–65 References Meena K, Luhar S (2019) Effect of wastewater on properties of con‑ crete. J Build Eng 21:106–112. https:// doi. org/ 10. 1016/j. jobe. Asadollahfardi G, Asadi M, Jafari H, Moradi A, Asadollahfardi R 2018. 10. 003 (2015) Experimental and statistical studies of using wash water Nalet C, Nonat A (2016) Retarding effectiveness of hexitols on the from ready‑mix concrete trucks and a batching plant in the pro‑ hydration of the silicate phases of cement: Interaction with the duction of fresh concrete. Constr Build Mater 98:305–314. https:// aluminate and sulfate phases. Cem Concr Res 90:137–143. https:// doi. org/ 10. 1016/j. conbu ildmat. 2015. 08. 053 doi. org/ 10. 1016/j. cemco nres. 2016. 09. 018 Baetzner S (2002) Ways of anallyzing iron (II) sulphate hydrate Naveen BP, Mahapatra DM, Sitharam TG, Sivapullaiah PV, Ramachan‑ in respect of it chromate reducing action in cement. Z K G dra TV (2017) Physico‑chemical and biological characterization 55(7):80–88 of urban municipal landfill leachate. Environ Pollut 220:1–12. Bartlett MS (1954) A note on the multiplying factors for various chi https:// doi. org/ 10. 1016/j. envpol. 2016. 09. 002 square approximations. J R Stat Soc A Stat 16:296–298 Nocuń‑ Wczelik W, Łój G (2006) Effect of lead compounds on alite Batchelor B (2006) Overview of waste stabilization with cement. Waste hydration Cement Lime. Concrete 11/73:343–350 Manag 26(7):689–698. https://doi. or g/10. 1016/j. w asman.2006. 01. Regulation of the Minister of construction sector of 2006 (Journal of Laws 2006 No 136, item 964) Bonaccorsi E, Merlino S, Taylor HFW (2004) The crystal structure of Regulation of the Minister of Maritime economy and Inland navigation jennite, Ca Si O (OH) 8H O. Cem Concr Res 34:1481–1488. 9 6 18 6 2 of 12 July 2019 (Journal of Laws 2019 item 1311). https:// doi. org/ 10. 1016/j. cemco nres. 2003. 12. 033 Stanisz A (2001) An Affordable Statistic Course Based on the STA ‑ Brennan RB, Healy MG, Morrison L, Hynes S, Norton D, Clifford E TISTICA PL Program on Examples of Medicine. StatSoft Polska, (2016) Management of landfill leachate: the legacy of European Krakow union directives. Waste Manage 55:355–363. https:// doi. org/ 10. Szymański K, Siebielska I, Janowska B, Sidełko R (2018) Variations in 1016/j. wasman. 2015. 10. 010 physical and chemical parameters of landfill leachates over time. Brennan RB, Clifford E, Devroedt C, Morrison L, Healy MG (2017) Desal Wat Treat 117:149–155. https://doi. or g/10. 5004/ dwt. 2018. Treatment of landfill leachate in municipal wastewater treatment plants and impacts on effluent ammonium concentrations. J Envi‑ Taghizadeh M, Asadollahfardi G, Salehi AM, Akbardoost J (2021) ron Manage 188:64–72. https:// doi. org/ 10. 1016/j. jenvm an. 2016. Feasibility study of reusing wash water and steel fibre simultane‑ 11. 055 ously on workability, mechanical properties and fracture tough‑ Chen QY, Tyrer M, Hills CD, Yang XM, Carey P (2009) Immobilisa‑ ness of concrete. Aust J Civ Eng. https:// doi. org/ 10. 1080/ 14488 tion of heavy metal in cement‑based solidification/stabilisation: 353. 2021. 18995 99 a review. Waste Manag 29:390–395. https:// doi. org/ 10. 1016/j. Vaičiukynienė D, Kantautas A, Tučkutė S, Manhanga F, Janavičius wasman. 2008. 01. 019 E, Ivanauskas E, Rudžionis Z, Gaudutis A (2021) The using of Czarnecki L, Deja J, Flaga K, Jasiczak J, Kurdowski W, Małolepszy concrete wash water from ready mixed concrete plants in cement J, Radomski W, Śliwiński J (2015) The frost resistance of bridge systems. Materials 14:2483. https://doi. or g/10. 3390/ ma141 02483 concretes. BTA 1(69):66–69 1 3 97 Page 14 of 14 Applied Water Science (2022) 12:97 Varshney H, Khan RA, Khan IK (2021) Sustainable use of differ ‑ Publisher's Note Springer Nature remains neutral with regard to ent wastewater in concrete construction: a review. J Build Eng jurisdictional claims in published maps and institutional affiliations. 41:102411. https:// doi. org/ 10. 1016/j. jobe. 2021. 102411 Wdowczyk A, Szymańska‑Pulikowska A (2020) Differences in the composition of leachate from active and non‑ operational munici‑ pal waste landfills in Poland. Water 12:3129. https:// doi. org/ 10. 3390/ w1211 3129 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Water Science Springer Journals

The use of a regression model in the variability analysis of the leachate quality from heaps of the production of the building materials

Loading next page...
 
/lp/springer-journals/the-use-of-a-regression-model-in-the-variability-analysis-of-the-2OZIdXEdjS

References (46)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
ISSN
2190-5487
eISSN
2190-5495
DOI
10.1007/s13201-022-01639-x
Publisher site
See Article on Publisher Site

Abstract

An attempt to evaluate potential hazard for quality of manufactured construction elements and concrete in case of use of leachate as technological water was made in this paper. This may lead to negative impact on strength parameters of the prod‑ uct made in the analysed company. Raw industrial leachate must not be discharged into sewerage system due to increased concentrations of some mineral components; it must be initially purified. Selected elements of multiple regression module and principal components analysis were used to describe the above phenomena. The model we have developed uses selected significant parameters of pollutants contained in leachate from a process industrial waste heap. The model takes into account leachate conductivity and sulphate ions, in which increased lead compounds concentration was determined. The model was verified during its use. The variables used in the model explain the significant percentage of variability noted for the dependent variable. Keywords Leachate · Lead · Multiple regression · Principal components analysis Introduction established for identification of particular concrete quality parameters such as strength, flexibility, hardness and work ‑ Manufacturing of construction materials, in particular ability. It has been proved that concrete wash water featured concrete, is associated with generation of solid waste and increased alkalinity and solids content going beyond the lim‑ wastewater. Washing of concrete mixers and units generates its of popular standard ASTMC94, which caused that manu‑ wastewater polluting water and soil. Concrete washing water factured concrete was more porous. This limits the interest features high pH fluctuating between 11 and 12 due to high in application of technological water recycling mostly in the limestone content in the solution. This water contains dis‑ case of use of various admixtures in concrete production solved solids such as hydroxides and sulphates from cement, ASTM C94 (Asadollahfardi et al. 2015; Meena and Luhar chlorides from application of calcium chloride admixture, 2019). oils and greases from plant and machinery as well as small Leachates generated during construction materials pro‑ amounts of other chemicals associated with Portland cement duction fed into technological water make an additional hydration and derivates from chemical admixtures. Many source of pollution of technical concrete and prefabricated publications contain descriptions of use of concrete wash elements (Bonaccorsi et al. 2004; Chen et al. 2009; Dai et al. water for making mortar and concrete (Meena and Luhar 2009). Leachates fed into the manufactured product feature 2019; Vaičiukynienė et al. 2021; Taghizadeh et al. 2021; increased concentration of mineral components such as sul‑ Varshney et al. 2021). However, such materials must com‑ phates, carbonates, chlorides, nitrates and compounds of ply with the quality standards. The standards had been such heavy metals as chromium, lead, nickel, copper, cad‑ mium and zinc (Kurdowski 2010; Szymański et al. 2018). Examples of such waste and leachate management were * Beata Janowska found in one of the companies manufacturing construction beata.janowska@tu.koszalin.pl materials (Magistri and D’Arcangelo 2008). In the EU countries, the Landfill Directive, the Waste Faculty of Civil Engineering, Environmental and Geodetic Sciences Department of Water, Sewage and Waste Framework Directive 2008/98/EC, the Urban Wastewater Technology, Koszalin University of Technology, ul. Treatment Regulations Council Directive 99/31/EC and the Śniadeckich 2, 75‑453 Koszalin, Poland Vol.:(0123456789) 1 3 97 Page 2 of 14 Applied Water Science (2022) 12:97 Water Framework Directive 2000/60/EC belong to the main 2002; Nocuń‑ Wczelik and Łój 2006; Czarnecki et al. 2015; European regulations concerning waste landfilling and lea‑ Nalet and Nonat 2016). This has also impact on processes of chates management. They put an obligation on the mem‑ interaction between particular parameters/components con‑ ber states to regularly monitor leachate condition. Despite tained in the generated leachate that is reused (Król 2012; implementation of the said regulations, the scope of the Czarnecki et al. 2015) in replenishment water. Leachates parameters to be monitored has not been strictly defined. may also have negative impact on the natural environment It has been only stated that the analysed substances should unless subjected to a pre‑treatment before discharging to be selected based on the composition of the landfilled waste a sewerage system, in particular to the ground or aquatic (Brennan et al. 2016, 2017; Wdowczyk and Szymańska‑ environment (Magistri and D’Arcangelo 2008; Król 2012; Pulikowska 2020; Ida and Eva 2021). Czarnecki et al. 2015). In all member states, the regulations concerning the mon‑ This paper illustrates an effort to assess potential hazards itoring of wastewater discharge exist on the national level; for product (structural elements) quality and the phenom‑ subsequently, they are being supplemented or defined at the ena that occur in leachates from industrial waste dump. The regional or local level. For example, in Germany, Austria, objective of the research work was to determine statistically Romania, Ukraine and Czech Republic testing of heave met‑ the significant impact of particular components (independent als content is monitored only if there occurs a strong impact variables) on concentration of one of toxic lead compounds of industrial waste that is channelled via a sewerage system in tested leachate. Selected elements of mathematical statis‑ to any municipal wastewater treatment plant. Slovakia, Hun‑ tics in the form of the multiple regression module and main gary, Romania and Moldova keep monitoring heavy metals components analysis were used to describe those phenomena content in a more regular way within the self‑monitoring (Stanisz 2001). procedure. In Ireland, the parameters to be analysed in land‑ fill leachates are determined individually in the waste dis‑ posal licence. Their scope depends on the type and composi‑ Tested material and research methodology tion of the landfilled waste (Naveen et al. 2017; Wdowczyk and Szymańska‑Pulikowska 2020; Ida and Eva 2021). Materials and scope of research In the USA, the Environmental Protection Agency regu‑ lates wastewater discharge and treatment based on the Clean The leachate quality tests performed at the production unit Water Act (CWA). The National Pollution Discharge Elimi‑ site allowed to determine basic physicochemical indicators nation System (NPDES) issues licences for all discharges as well as heavy metals and alkaline leachates sampled for and wastewater treatment plants as well as sets the norms testing (Figs.1, 2) including: regarding quality of wastewater discharged to surface water and to municipal wastewater treatment plants. In China, • Nearby a sewer sump in A1 and A2 locations, detailed regulations regarding heavy metals content in From existing leachate pond near the construction waste wastewater and drinking water are also set at the national heap at A3 and A4 locations. level. In China, limits of the admissible heavy metals con‑ tent have been defined by the Chinese Ministry of Health (CMH) and The National Standard of China, in Kenya—by the National Environment Management Authority (NEMA) and Kenya Bureau of Standards (Kinuthia et al. 2020). In Canada, the standards and regulations pertaining to the col‑ lection, protection, treatment and elimination of leachates are very strict; they also put emphasis on regular monitoring of surface and ground waters nearby any landfill location (Naveen et al. 2017). According to Polish law, the concentration of the said metals must not exceed the admissible values in leachates discharged also to the aquatic environment or ground as well as to sewerage systems (Regulation 2006; Regulation A2 2019). Discharge of leachates featuring higher concen‑ A1 tration of those metals than pure replenishing water may catalyse generation of new crystallographic structures in the manufactured material having consequently negative Fig. 1 Leachate from the washing area of the production unit impact on strength parameters of the material (Baetzner 1 3 Applied Water Science (2022) 12:97 Page 3 of 14 97 described in the literature (Kowal and Świderska‑ Bróż 1996). Particular parameters of leachates (technological water) were being determined using mainly standard research meth‑ ods in accordance with Polish standards (Table 1). The content of total organic carbon (TOC) were deter‑ mined using the Vario Max CN macroanalyser. Samples of leachates for testing were mineralised using a mixture of acids (65% HN O 70% HClO and 30% H O ) using micro‑ 3, 4 2 2 wave energy (Milestone 1200 Mega apparatus). The contents of heavy metals and alkali metals were determined using FAAS technique (iCE 3500Z Thermo Scientific SOLAAR) A3 A4 (Czerniak and Poszyler‑ Adamska 2006; Szymański et al. 2018). Statistical analysis Fig. 2 Leachate from the industrial waste heap To evaluate tested pollutants, components occurring in par‑ ticular samples of leachate from concrete production STA- Locations A1 and A2 are specific for leachate originating TISTICA—multiple regression module—were used (Stanisz from the washing facility of the concrete production unit, 2001). To describe correlations between particular variables, whereas A3 and A4 are located at leachate pond nearby the an estimator of the correlation coefficient for tested features, waste heap. The tests were performed during 2014– as r correlation, was used. For this purpose, Fischer’s test 2018 period. Five (5) test series were effected in total. (F) verifying the hypothesis that the multiple correlation coefficient is significantly different from “0” was useful; this means that at least one model structural parameter (without Analytical methodology and test results free term) is different from zero. A number of notations that are specific for the analy ‑ Particular determinations were performed in accordance sis of results originating from statistical method research with Regulation of the Minister of Maritime Economy and were used in the presented paper (Stanisz 2001) and they Inland Navigation of 12 July 2019 Journal of Laws 2019 are: R —a multiple determination coefficient measuring item 1311, Regulation of the Minister of Construction Sec‑ which proportion of variation of the dependent variable is tor of 14 July 2006 Journal of Laws 2006 No 136, item 964 explained by the regression model calculated from the equa‑ (Regulation 2006; Regulation 2019) as well as methods tion: R = 1‑ [Rest SS/ Total SS], where Rest SS is a sum of Table 1 Standard research methods Parameter Analytical procedures—standard Analytical devices pH PN‑EN ISO 10523:2012 inoLab Multi Level 1 WTW Conductivity PN‑EN 27,888:1999 inoLab Multi Level 1 WTW Ammonium nitrogen PN‑ISO 5664:2002 – Nitrate (V) nitrogen PN‑C‑04576–08:1982 UV/Vis 6715 spectrophotometer JENWAY Nitrate (III) nitrogen PN‑C‑04576–08:1982 UV/Vis 6715 spectrophotometer JENWAY Kjeldahl total nitrogen Titration method PN‑EN 25,663:2001 SCHOTT instruments titroline easy Chlorides PN‑ISO 9297:1994 – Total phosphorus PN‑EN ISO 6878:2006 UV/Vis 6715 spectrophotometer JENWAY Sulphates VI PN‑ISO 9280:2002 – Total dissolved solids PN‑EN 15,216:2010 – Total solids PN‑EN 872:2007 – Total suspended solids PN‑C‑04559–03:1972 – BOD (manometric method, PN‑EN 1899–2:2002 – COD Bichromate method PN‑ISO 15705:2005 – Cr 1 3 97 Page 4 of 14 Applied Water Science (2022) 12:97 squares error and Total SS is a total sum of squares. The represents strong negative correlation whereas + 1.00 2 2 multiple correlation coefficient is a root of R . Corrected R value represents strong positive correlation. Value 2 2 is interpreted just like R . The corrected R coefficient takes r = 0.00 indicates absence of the correlation. into account a variable number of the degrees of freedom for Total SS and Rest SS. The estimation standard error S provides information Results and discussion about the average quantity of empirical deviations of the dependent variable from the model‑derived value, defined Results of physicochemical analysis by formula: � The scope of the said research work is illustrated in Tables 2 e and 3. Table 2 comprises test results for basic physicochemi‑ i=1 S = cal indicators, whereas Table 3 comprises test results for n − 2 heavy and alkali metals. where the estimation standard error pertains to: y —inde‑ Tests performed on leachates from construction waste pendent variable values. y ̂ —model‑derived independent heap and technological devices of the concrete production variable value. e = y —y ̂ rests. S = estimation standard plant indicated increased contents of some heavy metals i i i e error. including chromium, lead and iron compounds. However, no mercury compounds were found therein. Zinc, cadmium, The statistics F (Fischer’s test) is used to verify the sig‑ nickel and manganese occurred in trace concentrations. Cal‑ nificance of the entire model. cium, and to a much lesser extent magnesium, dominated in The critical level of significance (p ) is a probability of the alkali metals group. The said leachates were also used for making an error consisting in refusal of the true null the preparation of technological water. Some volume of the hypothesis based on the observed sample values. Low leachates was being discharged to the combined sewerage values of the significance level (usually α < 0.05) indicate system. Environmental protection services had been indicat‑ low risk whilst the decision to reject the null hypothesis ing unlawful management of the leachates (Regulation 2006; is made. Regulation 2019). The correlation is a measure of association between two The research results indicate environmental protection or more variables. The correlation of two variables can problems as well as improper use of leachate for produc‑ be expressed numerically using the linear correlation tion of concrete and concrete products originating from that coefficient (r ). The linear correlation coefficient takes company. As it appears from Tables 2 and 3, such exces‑ values from − 1.00 to + 1.00 interval. The – 1.00 value sive values (average daily values) pertained, in accordance Table 2 Averaged results of Parameter Unit of measure Tested during the period of 2014—2018 physicochemical analysis of leachate samples Sample No 1 2 3 4 5 pH – 7.66 12.02 12.60 12.59 11.21 Conductivity μS/cm 318 1698 4560 6180 3189 Ammonium nitrogen mgN /dm 0.16 0.57 0.22 0.05 0.25 NH4 Nitrate (V) nitrogen mgN /dm 0.10 0.24 0.18 0.19 0.17 NO3 Nitrate (III) nitrogen mgN /dm 0.002 0.020 0.000 0.002 0.006 NO2 Kjeldahl total nitrogen mgN/dm 6.91 2.51 0.64 0.73 2.69 Chlorides mgCl/dm 18 28 42 50 34 Total phosphorus mgP/dm 0.98 1.71 0.68 1.52 1.22 Sulphates VI mgSO /dm 41.2 226.4 308.7 679.1 313.8 Tot. dissolved solids mg/dm 400 970 1500 1600 1117 Total solids mg/dm 4200 1200 10,250 4600 5062 Tot. suspended solids mg/dm 2700 363 8600 2400 3515 BOD mgO /dm 596 134 40 30 200 5 2 COD mgO /dm 3980 100 400 125 1151 Cr 2 TOC mgC/dm 1492 37 150 46 431 1 3 Applied Water Science (2022) 12:97 Page 5 of 14 97 Table 3 Averaged content of heavy and alkali metals in leachate sam‑ ions—Fig.  5 (correlation coefficient r = 0.986). Peculiar ples [mg/dm ] character of copper behaviour may also be interpreted as formation of this metal complexes with organic compounds. Parameter Tested during the period of 2014–2018 At the same time, the increase in the BOD load has shown Sample No negative correlation with Cu and positive correlation with 1 2 3 4 5 magnesium (r = 0.997; Tab.3) and iron (r = 0.990; Fig. 8.) ions. This does not mean that all heavy and alkali metals Cu 0.003 0.009 0.011 0.019 0.01 made similar combinations. Cd < 0.001 < 0.001 < 0.001 < 0.001 0.001 Thus, such leachate must not be drained into waters or Cr 0.006 0.020 0.363 0.019 0.102 ground nor used again as technological water for concrete Ni < 0.004 < 0.004 < 0.004 < 0.004 0.004 production (Regulation 2019). Due to occurrence of rela‑ Zn 0.003 < 0.001 < 0.001 < 0.001 0.001 tively high amounts of chromium and lead compounds in the Pb 0.005 0.015 0.025 0.048 0.023 leachate, it must not be used as a technological water com‑ Mn 0.014 < 0.001 < 0.001 < 0.001 0.004 ponent. Increased concentrations of heavy metals were also Fe 0.181 0.047 0.03 0.043 0.075 noted in leachate sampling locations A1 and A4, which can Ca 35.37 89.76 197.48 407.53 182.58 originate from higher volume of sewage sludge occurring Mg 3.33 0.06 0.077 0.045 0.878 there. Consequently, production of construction elements using technological water containing such leachate may with the above‑mentioned regulations, to: pH—sample No cause occurrence of unfavourable phenomena in produced 3 (12.60), suspended solids—sample No 3 (8600 mg/dm ), materials; therefore, there is a deterioration of their strength BOD —sample No 1 (596  mgO /dm), COD —sample No parameters (Czerniak and Poszyler‑ Adamska 2006; Gineys 5 2 Cr 3 3 1 (3980 mg/O dm ), TOC—sample No 1 (1492 mgC/dm ), 2011; Król 2012). sulphates—sample No 4 (676.1  mgSO /dm ) and calcium— Table 3, sample No 4 (407.53 mgCa/dm ). Statistical analysis of laboratory test results The test results presented in Table  2 indicate that lea‑ chate samples taken for testing at point A1 (Fig. 1) were Table 4 shows selected significant linear correlation coef‑ exposed to direct contact with the ground and vegetal mate‑ ficients. It was noted that there occurred a clear positive rial. Therefore, organic substances containing humic com‑ correlation of lead with: sulphates (r = 0.99), chlor ides pounds could migrate to the leachates. This has been indi‑ (r = 0.95), total dissolved solids (r = 0.88) and electrolytic cated by increased loads of BOD COD and TOC. It can conductivity (r = 0.96). High correlation was noted for chro‑ 5, Cr be reckoned that humic compounds formed metal–organic mium and total solids (r = 0.90) as well as total suspended combinations with, for example, iron and magnesium. solids (r = 0.93) (Table 4). Also significant negative values This may explain their high correlations. Occurrence of of the coefficient of correlation between iron and pH reaction calcium compounds favoured combinations with sulphate (r = − 0.99), total dissolved solids (r = − 0.89) and positive Table 4 Averaged linear Cu Cr Pb Fe Ca Mg correlation coefficients pH 0.798965 0.419140 0.718243 − 0.996361 0.660810 − 0.993451 Conductivity 0.943712 0.364665 0.959130 − 0.741467 0.954364 − 0.720229 Ammonium nitrogen − 0.341047 − 0.075248 − 483582 − 0.225064 − 0.558863 − 0.264870 Nitrate V nitrogen 0.518972 0.063616 0.378452 − 0.848332 0.294735 − 0.890233 Nitrate III nitrogen − 0.150449 − 0.407748 − 0.293376 − 0.209573 − 0.367645 − 0.285229 Kjeldahl total nitrogen − 0.841892 − 0.497105 − 0.786683 0.972432 − 0.742257 0.955850 Chlorides 0.950641 0.373938 0.953405 − 0.790957 0.941617 − 0.772290 Total phosphorus 0.354433 − 0.736170 0.259427 − 0.241784 0.209849 − 0.345671 Sulphates VI 0.992818 0.012041 0.995156 − 0.665358 0.985989 − 0.682669 Tot. dissolved solids 0.902790 0.487761 0.882325 − 0.890282 0.856769 − 0.866226 Total solids 0.145817 0.908333 0.216883 − 0.251460 0.246244 − 0.147146 Tot. suspended solids 0.055536 0.948552 0.113288 − 0.253781 0.135769 − 0.147503 BOD − 0.826407 − 0.430217 − 0.754290 0.990457 − 0.701250 0.984658 COD − 0.766111 − 0.299968 − 0.667138 0.984893 − 0.600994 0.997867 Cr TOC − 0.766313 − 0.299531 − 0.667350 0.984820 − 0.601212 0.997840 1 3 97 Page 6 of 14 Applied Water Science (2022) 12:97 with total nitrogen (r = 0.97) were found. High correlation of lead in leachate varies from approx. 0.005 to 0.048 mg Pb/ of iron with BOD (r = 0.99), COD (r = 0.98) and TOC dm , which partly correlates with sulphate VI ions indicator 5 Cr (r = 0.98) was noted. Strong correlations between total sus‑ within the limits from 41.2 to 679.1 mgSO/ dm . Strong pended solids and chromium (r = 0.92) were also noted— correlations for remaining components of the leachate are this suggests that chromium compounds are sorbed, to a illustrated in subsequent graphs presented in Figs. 4, 5, 6, major extent, by suspended solid particles. High correlation 7, 8. between lead compounds and sulphate VI ions (r = 0.96) Calcium compounds in examined technological water the indicates a probability of generation of sparingly soluble main element of alkaline components in leachate maintained PbSO . at the production plant. Occurrence of calcium compounds The determination coefficient calculated for lead and in cement, therefore, in leachate containing mineral additives sulphate compounds is R = 0.997. This association defines such as volatile silica ashes or granular blast furnace slag, which part of variability of the dependent feature can be containing hydrated calcium silicates, may lead to a synthe‑ interpreted using the developed model (Tab. 5). Examples sis of amorphous or semi‑ crystalline structures (Baetzner of statistical analysis results are illustrated in relevant graphs 2002; Batchelor 2006; Kudrowski 2010). It is believed that 4− (Figs. 3, 4, 5, 6, 7, 8). Particular correlations shown in those such structures composed of calcium ions and [SiO ] tetra‑ figures illustrate categorised graphs of dispersion along with hedrons are then significantly developed. In practical terms, regression line. Histograms of the variable distribution, with the increase in the CaO content in solution, the increase at frequency of occurrence of given parameter, were also in the Ca/Si ratio in the solid phase can be observed (Kur‑ depicted here. For example, Fig. 3 shows that concentration dowski 2010). It appears from the research work performed on leachates that this environment may also contain calcium compounds in CaSO form as well as lead compounds in Table 5 Multiple regression results sparingly soluble PbSO form, which was confirmed by 2− 2+ 2+ very high correlation of SO with Ca and Pb ions Statistics Value (r = ~ 0.99). Multiple correlation coefficient 0.9987 Particular graphs shown in Figs. 3, 4, 5, 6, 7, 8 are sup‑ Determination coefficient R 0.9975 plemented with histograms categorised with relations to Corrected determination coefficient R 0.9950 particular groups. Those phenomena were interpreted by F(2.2) 401.56 the multiple regression module made for parameters of the P 0.00248 Fig. 3 Categorised graph of Pb = 0.00174 + 0.68E-4 * sulphates dispersion with regression line correlation: r =0.995 for lead and sulphates 0,07 0,06 0,05 0,04 0,03 0,02 0,01 0,00 -0,01 -200 0 200 400 600 800 1000 02 4 2- Concentration SO [mg/dm ] 0.95 CI 1 3 Concentration Pb [mg/dm ] Applied Water Science (2022) 12:97 Page 7 of 14 97 Fig. 4 Categorised graph of Pb = -0.0191 +0.00123 *chlorides dispersion with regression line Correlation: r =0.953 for lead and chlorides 0,07 0,06 0,05 0,04 0,03 0,02 0,01 0,00 -0,01 51015202530354045505560 02 4 Concentration Cl [mg/dm ] 0.95 CI Fig. 5 Categorised graph of Ca = -7.226 + 0.60480 * sulphates dispersion with regression line correlation: r = 0.986 for calcium and sulphates -100 -200 0200 400600 8001000 02 4 2- Concentration SO [mg/dm ] 0.95 CI pollutants occurring in tested leachate originating from tech‑ contained in leachate from the industrial waste heap based nical concrete and construction elements production. on the example of electrolytic conductivity and sulphate This resulted in development of a model taking into ions, in which high concentration of lead compounds was account selected, significant parameters of pollutants found. Using this model, its verification was carried out. 1 3 Concentration Ca Concentration Pb 3 3 [mg/dm ] [mg/dm ] 97 Page 8 of 14 Applied Water Science (2022) 12:97 Fig. 6 Categorised graph of dis‑ Cr = -0.0618 +0.47E-4 * total suspended solids persion with regression line for correlation: r = 0.948 chromium and total suspended solids 0,45 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 0,00 -0,05 -0,10 -2000 2000 6000 10000 024 0 4000 8000 12000 Concentration total suspended solids [mg/dm ] 0.95 CI Fig. 7 Categorised graph of Cu = 0.00271 + 0.25E-4 * sulphates dispersion with regression line correlation: r = 0.993 for copper and sulphates 0,024 0,022 0,020 0,018 0,016 0,014 0,012 0,010 0,008 0,006 0,004 0,002 0,000 -0,002 -200 0 200 400 600 8001000 024 2- Concentration SO [mg/dm ] 0.95 CI originating from concrete and construction elements pro‑ Multiple regression module duction, allows for Fischer’s test (F) to be used. Lead is a toxic element; its content in cement and lea‑ Multiple regression, used as a tool in evaluation of mutual relationships between particular components of leachate chates is significant—it can migrate to any aquatic envi‑ ronment. High correlation between high Pb content and 1 3 Concetration Cu Concentration Cr 3 3 [mg/dm ] [mg/dm ] Applied Water Science (2022) 12:97 Page 9 of 14 97 Fig. 8 Categorised graph of Fe = 0.02284 + 0.26E-3 * BOD dispersion with regression line correlation: r = 0.990 for iron and BOD 0,22 0,20 0,18 0,16 0,14 0,12 0,10 0,08 0,06 0,04 0,02 0,00 -0,02 -100 0 100 200 300 400 500 600 700 800 02 4 Concetration BOD [mgO /dm ] 0.95 CI conductivity as well as sulphates (VI) content allowed for Those variables independently explain approximately 91% model development (Tab. 4). As a result, a model taking of the dependent variable variances after exclusion of impact into account some selected significant parameters of pol‑ of other independent variables. lutants contained in leachate originating from construction The adopted model, based on the correlation coefficient elements production, was developed. The said model takes values, can be written in the following form: into account leachate electrolytic conductivity and concen‑ Pb =  + conductivity +  sulphates (1) 0 1 2 tration of sulphate ions in this medium. Using this model, its verification was carried out. The use of statistical analysis for elaboration of test results allowed for finding of significant relationships Fischer’s test (F) between particular pollutants, which allowed for the defini‑ tion of: determination coefficient, corrected determination Table 6. contains basic parameters of statistical evaluation coefficient, F (Fischer’s) coefficient, critical significance including: the multiple correlation coefficient, determina ‑ level p and estimation standard error. Particular variables tion coefficient, corrected determination coefficient, F coef‑ in the model explain the significant percentage of the ficient, p critical significance level and estimation standard dependent variable (Tab. 5). The model significance was error. F statistic was used to check relevance of the entire additionally demonstrated based on the analysis of correla‑ model, where: tions between Pb measured values and theoretical values F (F-statistic)—statistic of the test verifying the hypoth‑ originating from Eq. (1). In the case of application of the esis that the multiple correlation coefficient is significantly regression analysis to evaluation of mutual relationships different from “0”, which means that at least one of the between particular leachate components several, and even a model structural parameters (without free term) is different dozen or so, factors can be considered—then, the multiple from zero. regression comes into play. The factors that were selected p (p-value)—critical significance level of the test verify ‑ in our case were the solution electrolytic conductivity and ing the hypothesis that the multiple correlation coefficient lead compounds concentration. This allowed for develop‑ is significantly different from “0”. ment of a model describing mutual relationships between Performed Fischer’s test F indicated the significance level those components. In the given case, the highest partial cor‑ p < 0.05; therefore, it required that at least one coefficient of relation coefficients of lead and sulphate ions compounds Eq. (1): β β or β should differ from zero. Estimated values 0, 1 2 was 0.995, whereas for the electrolytic conductivity: 0.959. of those coefficients in given case amount to b = 0.001291, 1 3 Concentration Fe [mg/dm ] 97 Page 10 of 14 Applied Water Science (2022) 12:97 Table 6 Summary of regression N = 5 Summary of the dependent variable regression b* Std. err. of b* b Std. err. of b t (2) p Free term 0.001291 0.000930 1.387607 0.299640 Conductivity 0.235808 0.098069 0.000002 0.000001 2.404511 0.138034 Sulphates 0.775100 0.098069 0.000053 0.000007 7.903602 0.015634 b*—are standardised beta coefficients for respective independent variables b = − 0.000002 and b = 0.000053 respectively. Therefore, the form of loadings of variable factors in the so‑called unit 1 2 there are two cases when p < 0.05. Based on these calcula‑ circle. To do this, an analysis taking into account a correla‑ tions, Eq. (2) was formulated as a model describing values tion matrix was performed. The internal area of the unit of the correlation coefficients; it can be written in the fol‑ circle represents admissible factor loadings. This means lowing form: that the closer to the circle is located the drawn variable, the better is its representation by graph factors. The major‑ Pb = 0.001291 − 0.000002 conductivity + 0.000053 sulphate ity of examined parameters of leachate mineral pollutants (2) (Fig. 9) (without heavy metals) is located in the circle. In Null hypothesis: H : b = 0, b = 0, b = 0 (none inde‑ 0 0 1 2 many cases, those correlations are very strong. An example pendent variable has significant impact on the dependent of correlation are associations of nitrogen (without ammo‑ variable). nium nitrogen) with B OD, COD , TOC, suspended matter, 5 Cr Alternative hypothesis: H : There is at least one b coef‑ 1 i total solids and significantly weaker between total phospho‑ ficient different from zero. rus and sulphates, chlorides, electrolytic conductivity and For the assumed significance level α = 0.05, the null pH. Sets of those variables show positive correlation. Such hypothesis p = 0.015634 < α = 0.05 can be rejected. correlation could not be observed in the case of heavy and Conclusion: there is an independent variable, which has alkali metals, which originated, partly, from low concentra‑ a significant impact on the dependent variable. tions of those leachate pollutants (Fig. 10). Nevertheless, It appears from the above considerations that using sta‑ quite close positive correlations between some heavy met‑ tistical methods for given set of data, significant impact of als and alkaline calcium could be seen. Negative correla‑ the independent variable on the dependent variable can be tions occurred between magnesium and chromium as well confirmed. as between nickel and manganese, cadmium, zinc and iron. This should be interpreted by absent, or occurring in trace concentrations, nickel. It is beyond any doubt that strong Principal components analysis correlation between lead compounds and sulphate as well as chloride ions indicated a possibility of generation of spar‑ Apart from use of descriptive statistics for data analysis, it was also possible to make graphs of 2 W type variables of ingly soluble compounds such as: PbSO (II) or PbCl (II) 4 2 and also CaSO as well as its hydrated form CaSO × 2H O. factor loadings. In this particular case, two variables (for X 4 4 2 and Y axes) were selected; then, a two‑dimensional disper ‑ It is quite possible that in alkaline leachate (pH > 8) copper compounds of Cu(OH) × 5H O type could be generated. sion graph was created. It is possible to obtain cumulative 2 2 eigenvalues and cumulative variance percentage. Part of This allowed for making of two graphs associated with the analysis of principal components, whereof one 2 W graph of the variance was isolated here by particular components. Those relationships are illustrated in Fig. 9. It can be noted factor loadings for objects (Fig. 9) and the second one 2 W graph of variable factors loadings (Fig. 10). Both graphs can that the component showing the highest eigenvalue explains 50.61% of total variance. The second one is 39.73% of total be aggregated and then a single graph, called biplot (Fig. 11), is obtained. This graph allows for penetration of the experi‑ variance. Therefore, two first components giving twodimen ‑ ‑ sional space explaining in total more than 90% variances mental data set structure, which facilitates evaluation of mutual interactions occurring in tested leachate and forecast were considered as significant. Such situation allowed for further conduct of the analysis solely based on two factor possible structures obtained through use of the leachate as a medium applied in the preparation of technological water. loadings. The newly created space, to which original points (variables and objects) were projected, made possible crea‑ Evaluation of such phenomena requires research by applica‑ tion of solid materials analysis techniques, for example XRF tion of proper data structure allowing to obtain the most significant information. X‑ ray fluorescence spectroscopy or XRD X ‑ ray diffraction, DTA differential thermal analysis and TG thermogravimetric Figure 9 shows an illustration of statistical analysis results for leachate from production of construction materials, in analysis (Lesko et al. 2001). The statistical analysis can be 1 3 Applied Water Science (2022) 12:97 Page 11 of 14 97 Fig. 9 Principal components Variables (axes F1 and F2: 90,34 %) analysis—projection of vari‑ Sulphates VI ables set in 2D factor loading Total phosphorus space Chlorides Ca Conductivity Pb Totaldissolved solids Cu 0.75 pH 0.5 Ni 0.25 Fe Hg Nitrate V nitrogen -0.25 Kjeldahl total nitrogen Total solids COD Mg BOD Cd -0.5 DOC Totalsuspended solids Nitrate III nitrogen Zn Mn Cr -0.75 Ammonium nitrogen -1 -1 -0.75-0.5-0.25 00.250.5 0.75 1 F1 (50,61 %) Fig. 10 Principal component Biplot (axes F1 and F2: 97,48 %) analysis (biplot)—projection of variables set and objects cluster‑ ing in 2D factor loading space Ca A3 Ni 3 Pb Cu Hg A1 Fe -1 Mg A2 -2 A4 Cr Zn -3 Cd Mn -4 -5 -6 -5 -4 -3 -2 -1 012345 6 F1 (64,38 %) 1 3 F2 (33,09 %) F2 (39,73 %) 97 Page 12 of 14 Applied Water Science (2022) 12:97 Fig. 11 Principal component Observaons (axes F1 and F2: 90,34 %) plot showing objects (A1‑ A4) clustering in 2D space (F1 and 6 A3 F2 factors scores) -1 A1 -2 A2 A4 -3 -6 -5 -4 -3 -2 -1 0123456 F1 (50,61 %) deepened due to outliers, applying the Bartlett’s test (Bartlett technological water for production of technical concrete 1954). The level of confidence p associated with the analysis and concrete elements. This may lead to negative impact of principal components equal to 0.00248 allows for such type on strength parameters of products manufactured in the operation. A preliminary analysis of projections of original analysed company. objects on the plane created by two first components (F 1 and F2), factor 1 and factor 2, was then performed (Lebart et al. • Raw post‑production leachate must not be discharged 1982). Already a superficial analysis indicates occurrence of into the sewerage system due to increased concentra‑ outliers, for example, Ni, Cr as well as ammonium nitrogen tions of some mineral components; it should be initially (Figs. 10, 11). Such outliers may have a negative impact on the purified. analysis result. Nickel occurred here in trace amounts whereas • A model was developed taking into account selected sig‑ Cr, with relation to other metals, in significant concentrations. nificant parameters of pollutants contained in leachate It can be noted that ammonium nitrogen NH ions were not from the industrial waste heap based on the example − − completely transformed into nitrate ions NO or NO , which of electrolytic conductivity and sulphate ions, in which 3 2 indicated also oxygen deficit in analysed leachate. Definitely, increased concentration of lead compounds was deter‑ the best way was to group the object factor loadings of the mined. Using this model, its verification was carried out. basic physicochemical analysis cases and selected groups of • The variables used in the model explain significant per ‑ heavy and alkali metals with locations of leachate sampling centage of variability noted for the dependent variable. (A1 and A4). The averaged concentration of pollutants in those • Inclusion of additional dependent variables into the locations had the highest value. At the same time, the high‑ model did not improve significantly particular model est amounts of accumulated sewage sludge were noted there, parameters. which originated from nature of the substratum, which played Model significance was additionally verified based on the the role of a sedimentation facility. analysis of correlations between Pb measured values and sulphates as well as electrolytic conductivity. • To describe the phenomena occurring in leachate from Conclusions construction materials production, elements of mathe‑ matical statistic, in the form of multiple regression mod‑ Examination of leachate from existing leachate pond ule and principal components analysis, can be used. located nearby construction waste heap indicated that the said leachate must not be used as a component of 1 3 F2 (39,73 %) Applied Water Science (2022) 12:97 Page 13 of 14 97 Funding The authors received no specific funding for this work. Czerniak A, Poszyler‑ Adamska A (2006) Assessment of heavy met‑ als immobilisation in cement‑ground composites used for con‑ struction of country roads. Acta Sci Pol Formatio Circumiectus Availability of data and material Data collected by the author and sec‑ 5(1):29–38 ondary data are cited. Dai Y, Qian G, Cao Y, Chi Y, Xu Y, Zhou J, Liu Q, Xu ZP, Qiao S (2009) Effective removal and fixation of Cr(VI) from aque‑ Declaration ous solution with Friedel’s salt. J Hazard Mater 170:1086–1092. https:// doi. org/ 10. 1016/j. jhazm at. 2009. 05. 070 Conflict of interest On behalf of all authors, the corresponding author Gineys T (2011) Influence de la teneur en elements métaliques du states that there is no conflict of interest. clinker sur les proprieties techniques et environnementales du ciment Portland–these. Dissertation, Université Lille Nord France Ethics approval Authors firmly abide by all ethical norms and follow Ida S, Eva T (2021) Removal of heavy metals during primary treatment the guidelines set by COPE. This article does not contain any studies of municipal wastewater and possibilities of enhanced removal: a with human participants or animals performed by any of the authors. review. Water 13:1121. https:// doi. org/ 10. 3390/ w1308 1121 Kinuthia GK, Ngure V, Beti D, Lugalia R, Wangila A, Kamau L (2020) Levels of heavy metals in wastewater and soil samples Open Access This article is licensed under a Creative Commons Attri‑ from open drainage channels in Nairobi, Kenya: community bution 4.0 International License, which permits use, sharing, adapta‑ health implication. Sci Rep 10:8434. https:// doi. or g/ 10. 1038/ tion, distribution and reproduction in any medium or format, as long s41598‑ 020‑ 65359‑5 as you give appropriate credit to the original author(s) and the source, Kowal LA, Świderska ‑ Bróż M, (1996) Water purification. PWN, provide a link to the Creative Commons licence, and indicate if changes Warsaw were made. The images or other third party material in this article are Król A (2012) Release of heavy metals from mineral composites con‑ included in the article's Creative Commons licence, unless indicated sidering environmental impact. Politechnika Opolska, Opole otherwise in a credit line to the material. If material is not included in Kurdowski W (2010) Concrete chemistry. PWN, Warsaw the article's Creative Commons licence and your intended use is not Lebart L, Morineau A, Fénelon JP (1982) Traitement des données permitted by statutory regulation or exceeds the permitted use, you will statistiques, méthodes et programmes. Dunod, Paris need to obtain permission directly from the copyright holder. To view a Lesko S, Leśniewska E, Nonat A, Mutin JC, Goudonnet JP (2001) copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . Investigation by atomic force microscopy of forces at the origin of cement cohesion. Ultramicroscopy 86:11–21 Magistri M, D’Arcangelo P (2008) New reducing agent for cement. Z K G 3:61–65 References Meena K, Luhar S (2019) Effect of wastewater on properties of con‑ crete. J Build Eng 21:106–112. https:// doi. org/ 10. 1016/j. jobe. Asadollahfardi G, Asadi M, Jafari H, Moradi A, Asadollahfardi R 2018. 10. 003 (2015) Experimental and statistical studies of using wash water Nalet C, Nonat A (2016) Retarding effectiveness of hexitols on the from ready‑mix concrete trucks and a batching plant in the pro‑ hydration of the silicate phases of cement: Interaction with the duction of fresh concrete. Constr Build Mater 98:305–314. https:// aluminate and sulfate phases. Cem Concr Res 90:137–143. https:// doi. org/ 10. 1016/j. conbu ildmat. 2015. 08. 053 doi. org/ 10. 1016/j. cemco nres. 2016. 09. 018 Baetzner S (2002) Ways of anallyzing iron (II) sulphate hydrate Naveen BP, Mahapatra DM, Sitharam TG, Sivapullaiah PV, Ramachan‑ in respect of it chromate reducing action in cement. Z K G dra TV (2017) Physico‑chemical and biological characterization 55(7):80–88 of urban municipal landfill leachate. Environ Pollut 220:1–12. Bartlett MS (1954) A note on the multiplying factors for various chi https:// doi. org/ 10. 1016/j. envpol. 2016. 09. 002 square approximations. J R Stat Soc A Stat 16:296–298 Nocuń‑ Wczelik W, Łój G (2006) Effect of lead compounds on alite Batchelor B (2006) Overview of waste stabilization with cement. Waste hydration Cement Lime. Concrete 11/73:343–350 Manag 26(7):689–698. https://doi. or g/10. 1016/j. w asman.2006. 01. Regulation of the Minister of construction sector of 2006 (Journal of Laws 2006 No 136, item 964) Bonaccorsi E, Merlino S, Taylor HFW (2004) The crystal structure of Regulation of the Minister of Maritime economy and Inland navigation jennite, Ca Si O (OH) 8H O. Cem Concr Res 34:1481–1488. 9 6 18 6 2 of 12 July 2019 (Journal of Laws 2019 item 1311). https:// doi. org/ 10. 1016/j. cemco nres. 2003. 12. 033 Stanisz A (2001) An Affordable Statistic Course Based on the STA ‑ Brennan RB, Healy MG, Morrison L, Hynes S, Norton D, Clifford E TISTICA PL Program on Examples of Medicine. StatSoft Polska, (2016) Management of landfill leachate: the legacy of European Krakow union directives. Waste Manage 55:355–363. https:// doi. org/ 10. Szymański K, Siebielska I, Janowska B, Sidełko R (2018) Variations in 1016/j. wasman. 2015. 10. 010 physical and chemical parameters of landfill leachates over time. Brennan RB, Clifford E, Devroedt C, Morrison L, Healy MG (2017) Desal Wat Treat 117:149–155. https://doi. or g/10. 5004/ dwt. 2018. Treatment of landfill leachate in municipal wastewater treatment plants and impacts on effluent ammonium concentrations. J Envi‑ Taghizadeh M, Asadollahfardi G, Salehi AM, Akbardoost J (2021) ron Manage 188:64–72. https:// doi. org/ 10. 1016/j. jenvm an. 2016. Feasibility study of reusing wash water and steel fibre simultane‑ 11. 055 ously on workability, mechanical properties and fracture tough‑ Chen QY, Tyrer M, Hills CD, Yang XM, Carey P (2009) Immobilisa‑ ness of concrete. Aust J Civ Eng. https:// doi. org/ 10. 1080/ 14488 tion of heavy metal in cement‑based solidification/stabilisation: 353. 2021. 18995 99 a review. Waste Manag 29:390–395. https:// doi. org/ 10. 1016/j. Vaičiukynienė D, Kantautas A, Tučkutė S, Manhanga F, Janavičius wasman. 2008. 01. 019 E, Ivanauskas E, Rudžionis Z, Gaudutis A (2021) The using of Czarnecki L, Deja J, Flaga K, Jasiczak J, Kurdowski W, Małolepszy concrete wash water from ready mixed concrete plants in cement J, Radomski W, Śliwiński J (2015) The frost resistance of bridge systems. Materials 14:2483. https://doi. or g/10. 3390/ ma141 02483 concretes. BTA 1(69):66–69 1 3 97 Page 14 of 14 Applied Water Science (2022) 12:97 Varshney H, Khan RA, Khan IK (2021) Sustainable use of differ ‑ Publisher's Note Springer Nature remains neutral with regard to ent wastewater in concrete construction: a review. J Build Eng jurisdictional claims in published maps and institutional affiliations. 41:102411. https:// doi. org/ 10. 1016/j. jobe. 2021. 102411 Wdowczyk A, Szymańska‑Pulikowska A (2020) Differences in the composition of leachate from active and non‑ operational munici‑ pal waste landfills in Poland. Water 12:3129. https:// doi. org/ 10. 3390/ w1211 3129 1 3

Journal

Applied Water ScienceSpringer Journals

Published: May 1, 2022

Keywords: Leachate; Lead; Multiple regression; Principal components analysis

There are no references for this article.