See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/321027780

Possibilities for efficient use of valuable materials from aluminium slag to

remove specific pollutants in Wastewater

Article  in  Journal of Environmental Protection and Ecology · January 2017

CITATIONS READS

8 289

14 authors, including:

Carmen Tociu Szép Robert

Institutul Naţional de Cercetare-Dezvoltare pentru Protecţia Mediului Sapientia Hungarian University of Transylvania
45 PUBLICATIONS   127 CITATIONS    64 PUBLICATIONS   654 CITATIONS   

SEE PROFILE SEE PROFILE

Ana-Maria Anghel Florica Marinescu

Institutul Naţional de Cercetare-Dezvoltare pentru Protecţia Mediului Institutul Naţional de Cercetare-Dezvoltare pentru Protecţia Mediului
55 PUBLICATIONS   202 CITATIONS    47 PUBLICATIONS   198 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Energy View project

TRAINING AND AWARENESS CAMPAIGNS REGARDING THE POTENTIAL HUMAN HEALTH AND ENVIRONMENTAL RISKS ASSOCIATED TO HAZARDOUS SUBSTANCES AND
WASTE (TARCHS) View project

All content following this page was uploaded by Carmen Tociu on 13 November 2017.

The user has requested enhancement of the downloaded file.

Journal of Environmental Protection and Ecology 18, No 3, 842–852 (2017)

Water pollution

POSSIBILITIES FOR EFFICIENT USE OF VALUABLE

MATERIALS FROM ALUMINIUM SLAG TO REMOVE
SPECIFIC POLLUTANTS IN WASTEWATER

C. TOCIUa*, R. SZEPb, A. M. ANGHELa, F. MARINESCUa, M. ILIEa,

E. HOLBANa, G. GHITAa, M. MATEIa, F. D. DUMITRUa, I. POPESCUa,
A. MONCEAb, L. LASLOa, A. I. DAESCUa, C. R. GH. POPESCUc
a
National Institute for Research and Development in Environmental Protection,
284 Spl. Independentei Street, 060 031 Bucharest, Romania
E-mail: tociucarmen@yahoo.com
b
Faculty of Economics, Socio-Human Science and Engineering, Sapientia
Hungarian University of Transylvania, 1 Piata Libertatii Street, 530 104
Miercurea Ciuc, Romania
c
University of Bucharest, 36–46 Mihail Kogalniceanu Blvd., Bucharest,
Romania

Abstract. The topic of this paper is part of the current concerns to reduce the impact of pollut-
ants on the aquatic environment by chemical treatment of wastewater in the context of sustainable
development for improving quality of life. Suspended solids can lead to deposition of sludge and
the occurrence of anaerobic fermentation with negative effects on the ecosystem organisms, when
untreated wastewater is discharged into the aquatic environment. Aluminium recycling presents a
series of challenges regarding care for preventing the environment pollution and use of metallurgical
waste for recovering valuable materials. Starting from the lab-scale experimental data on chemical
treatment of wastewater with recovered aluminium sulphate (AS), one tries to evaluate the coagulant
effectiveness in a more accurate and theoretical way through development of a statistical (logarith-
mic) correlations between the applied dose of recovered AS and the resulted removal index for total
suspended solids (TSS). Mathematical model enables generalisation of experimental information
obtained and it has direct applicability for the predictive calculations of the design and optimum
operation of wastewater treatment plants.
Keywords: aluminium sulphate, metallurgical slag, wastewater treatment, statistic correlation.

AIMS AND BACKGROUND

&KHPLFDOWUHDWPHQWRIZDVWHZDWHUE\FRDJXODWLRQÀRFFXODWLRQ. The wastewaters
have variable composition depending on the sources that generate the contamina-
tion of water after its use in human activities. The main monitored water quality
parameters concern the temperature, pH, organic substances, suspended solids,
nutrients, and other chemical/biological compounds1,2.
*
For correspondence.

842
 Suspended solids are deposited in the bed of the basin receptor destroying
EHQWKLFRUJDQLVPVDQGDOWHULQJ¿VKVSDZQLQJVLWHV%LRGHJUDGDEOHRUJDQLFPDWWHU
decomposes, leading to the deoxygenation of river water. This has a direct impact
RQDOO¿VKDQGDTXDWLFRUJDQLVPVDQGLQGLUHFWO\LQÀXHQFHVWKHHPHUJHQFHRIKDUP-
ful decomposition products; the end result is a restructuring of the entire aquatic
biocoenosis. Due to the gas produced in the process of decomposition, quantities
of putrid sludge are entrained in the water, worsening further the situation and
producing an ugly landscape3,4.
7KHPHWKRGRIFKHPLFDOWUHDWPHQWE\FRDJXODWLRQÀRFFXODWLRQSURYLGHVDG-
vanced removal of undissolved impurities and part of those dissolved. Technologi-
FDOÀRZRIWKHSURFHVVLVEDVHGRQFKHPLFDOUHDFWLRQVVSHFL¿HGE\VWRLFKLRPHWULF
equations. This involves operations of preparation, dosing and mixing the reagents
in different proportions, within the treatment facilities and it is accomplished by
PHDQVRIVSHFL¿FHTXLSPHQW6HSDUDWLRQRIVROLGVE\WKHOLTXLGSKDVHLVGRQHE\
settling or centrifugation5,6.
One of the conventional chemicals used for coagulation are aluminium-based
salts, e.g. aluminium sulphate. When added to water, Al(III) ion hydrolyse to form
soluble species and solid precipitates, leading to the coalescence of colloids7,8.
Coagulation occurs when is carried out the optimum precipitation of Al(OH)3, i.e.
pH = 5–7.5. In order to increase the rate of particle aggregation and to facilitate the
IRUPDWLRQRIODUJHÀRFVZKLFKDUHDEOHWRIUHHO\VHWWOHFKHPLFDOVVXFKDVRUJDQLF
polymers (polyelectrolytes) are added.
3DUDPHWHUVWKDWLQÀXHQFHWKHSURFHVVRIFRDJXODWLRQÀRFFXODWLRQWUHDWPHQW
are chemical characteristics of the waste water (water pH, chemical composition)
and the hydrodynamic conditions (the duration and intensity of mixing, the dose
of coagulant, etc.)9.
Aluminium sulphate recovered from metallurgical slag. Secondary aluminium is
the metal obtained by processing aluminium scrap. This comprises waste result-
ing from production of castings, waste from mechanical processing, objects and
installations at the end of life cycle etc. Recovered waste is treated according to
their quality and features. Essential part of recycling is the furnace in which the
scrap melting is performed to obtain new raw materials10,11.
The recycling of aluminium is of high economic, environmental and social
importance. More than half of the aluminium used in the European Union is from
UHF\FOHGFRQWHQWVDQGWKLV¿JXUHLVLQFUHDVLQJFRQVWDQWO\12.
Secondary aluminium industry generates annually considerable amounts of
VODJEODFNVODJDQGVDOWFDNHZLWKDYDULRXVFKHPLFDODQGPLQHUDORJLFDOFRPSR-
sition13. The disposal of aluminium slag is a problem throughout the world14–16.
It contains many hazardous compounds and many compounds are water-soluble
PHWDOOLFR[LGHVDOOR\VFKORULGHVÀXRULGHVQLWULWHVFDUELGHVVXOSKLGHVHWFVR
WKDWODQG¿OOLQJLVFRQVLGHUDEO\H[SHQVLYH17–19.

843
 The melting of metal generates waste able to smother any activity if not
treated properly. Waste collection and processing have been developed as a paral-
lel economy becoming an important source of raw materials. Thus the removal
of metallurgical waste present in the industrial sites is very important in order to
achieve the two major goals of today’s society: protection of the environment and
the introduction into the economic circuit of recyclable materials. Usually, the
EODFNVODJFRQWDLQV±ZWDOXPLQLXPPHWDODQGWKHVDOWFDNHKDVUHVLGXDO
metallic aluminium content of 3–10 wt.% (Ref. 17).
7KLVSDSHUPDNHVDGLVFXVVLRQRQDQHI¿FLHQWPDQQHUWRLQFUHDVHWKHYDOXH
of the slag by obtaining an inorganic coagulant such as aluminium sulphate (AS)
which is possible to be used in the chemical treatment of wastewater, in order to
protect water resources. The results obtained for the removal of pollutants from
municipal wastewater are encouraging in this regard, only minor differences in
coagulation/precipitation effectiveness being observed between the recovered vs.
commercial AS (Ref. 20).
The technology referred to in this paper for recovery of aluminium from slag
in the form of aluminium sulphate is based on a chemical and hydrometallurgical
SURFHVVLQJPHWKRGRIVODJ,QWKH¿UVWVWDJHWKHVODJLVZDVKHGIRUWKHDGYDQFHG
removal of soluble salts and then the soluble aluminium is leached from slag with
VXOSKXULF DFLG7KLV WHFKQRORJ\ LV UHPDUNDEOH LQ WKDW LW LV D JUHHQ WHFKQRORJ\
resulting other by-products which have different uses21,22.

EXPERIMENTAL
The wastewater treated for removal of suspended solids was generated from various
VRXUFHVPHDWSURFHVVLQJRLOUH¿QLQJPHWDOOLFFRYHULQJDXWRVHUYLFHIXUQLWXUH
manufacturing, and urban wastewater.
The AS-based products (P1 … P30) lab-scale prepared from different sorts of
slag coming from foundries of Romania were tested by comparison with aluminium
sulphate Kemira ALG (EN 878:2004, Kemwater Cristal Comp.) considered ref-
erences coagulation agent, namely commercial AS. Agglomeration of coagulant
SDUWLFOHVWRIRUPODUJHDQGGHQVHÀRFVWKDWFDQEHHDVLO\GHFDQWHGJUDYLWDWLRQDOO\
was obtained by adding anionic polyelectrolyte FR1023 (Floerger Comp.). The
correction of pH to the optimum value (pH = 6.0–7.5) was made by using sodium
K\GUR[LGH7KHZRUNLQJVROXWLRQVZHUHSUHSDUHGDWWKHFRQFHQWUDWLRQVDVLQFRP-
mon practice in industrial activity of wastewater treatment.
([SHULPHQWDOVWXGLHVIRUFKHPLFDOWUHDWPHQWE\FRDJXODWLRQÀRFFXODWLRQZLWK
aluminium sulphate were carried out by using the Jar test23. This laboratory test is
intended to highlight the effectiveness of chemical treatment and it is considered
the best to determine operating conditions of treatment system tailored for each
VSHFL¿FW\SHRIZDVWHZDWHU

844
 The wastewater sample was introduced into a reaction vessel and mixed with
the aluminium sulphate under rapid mechanical stirring (160 rpm) for short time
(2 min) to ensure a homogeneous dispersion of the salt throughout the reaction
PDVV7KHÀRFFXODWLRQDJHQWZDVWKDQPL[HGVORZO\USPIRUORQJHUWLPH
PLQWRHDVHDJJORPHUDWLRQRIVXVSHQGHGVROLGVDQGVHWWOLQJRIWKHÀRFVGXULQJD
period of 120 min. The wastewater treated in this way has been analysed in order
to determine the degree of contaminants removal.

RESULTS AND DISCUSSION

This paper proposes the establishing of an analytical expression that enables to
determine on a theoretical way the coagulant effectiveness of aluminium sulphate
recovered from metallurgical slag. A statistical model is a theoretical description of
DUHDOSURFHVVDQGLWLVWKHEDVLFWRROIRUVROYLQJEDVLFWDVNVLQFKHPLFDOHQJLQHHULQJ
The measured initial and residual value of the TSS content in each category
of chemically treated wastewater are centralised in Table 1 together with the ap-
plied coagulant dose for every AS product obtained from different metallurgical
slag sources. As we can see, the applied AS product dose varies in the range of
100–6000 mg/l and it was found that the removed TSS relative index varies be-
tween 0.07–2.17 mg removed TSS/mg dose. Simple statistical indicators of data
variation were computed, namely the average value of 0.7016 mg removed TSS/
mg dose and the standard deviation of 0.5780 mg removed TSS/mg dose. These
results prove the good effectiveness of removed AS to wastewater treatment.
Table 1. TSS removal capacity of recovered AS products tested at treating wastewater from various
sources (mg removed TSS/mg AS product dose)
Recovered AS product No Coagulant dose (D) TSS Relative removed
(mg/l) TSS (RRS) mg re-
(mg/l) (mg Al3+/l) moved TSS/mg dose
1 2 3 4 5
Wastewater # 1 (initial load) 327
&RPPHUFLDO$6±EHQFKPDUN 1000 81.0 27 0.3000
P1–P7 1000 73.0–94.0 16–29 0.2980–0.3110
Wastewater # 2 (initial load) 425
&RPPHUFLDO$6±EHQFKPDUN 500 40.5 86 0.6780
P1–P7 500 36.5–47.0 71–94 0.6620–0.7080
Wastewater # 3 (initial load) 847
&RPPHUFLDO$6±EHQFKPDUN 6000 486.0 18 0.1382
P1-P7 6000 438.0–564.0 19–26 0.1368–0.1381
Wastewater # 4 (initial load) 287
&RPPHUFLDO$6±EHQFKPDUN 300 24.3 0.43 0.9200
P1–P7 300 21.9–28.2 0.44 0.8600–0.9300
Wastewater # 5 (initial load) 230
to be continued

845
Continuation of Table 1
1 2 3 4 5
&RPPHUFLDO$6±EHQFKPDUN 100 8.1 22 2.0800
P1–P7 100 7.3–9.4 13–22 2.0800–2.1700
Wastewater # 6 (initial load) 273
&RPPHUFLDO$6±EHQFKPDUN 300 24.3 29 0.8133
P8–P12 300 17.8–22.1 26–41 0.7733–0.8233
Wastewater # 7 (initial load) 675
&RPPHUFLDO$6±EHQFKPDUN 1500 121.5 276 0.2660
P8–P12 1500 88.9–105.9 305–355 0.2133–0.2467
Wastewater # 8 (initial load) 510
&RPPHUFLDO$6±EHQFKPDUN 1000 81.0 135 0.3750
P8–P12 1000 59.3–73.8 96–140 0.3700–0.4140
Wastewater # 9 (initial load) 157
&RPPHUFLDO$6±EHQFKPDUN 100 8.1 12 0.0659–0.0697
P13–P17 100 5.7–9.3 16–25 0.0682
Wastewater # 10 (initial load) 370
&RPPHUFLDO$6±EHQFKPDUN 1000 81.0 25 0.3450
P18–P27 1000 52.2–103.0 29–52 0.3180–0.3410
Wastewater # 11 (initial load) 405
&RPPHUFLDO$6±EHQFKPDUN 1000 81.0 36 0.3690
P28–P29 1000 70.7–74.9 46–52 0.3530–0.3590
Wastewater # 12 (initial load) 340
&RPPHUFLDO$6±EHQFKPDUN 400 32.4 32 0.7700
P30 400 36.5 44 0.7400
Average 0.7016
Standard deviation 0.5780

Starting from the experimental data, one tries to determine a theoretical

description of the coagulation process through a statistical model to identify a
coherent relationship between the applied coagulant dose (D) and resulted TSS-
removal relative index (RRS). By studying several algebraic functions, the most
convenient dependence was found to be the log correlation of the form:
ln(1 + D)
y = RRS (relative removed TSS) = a + b (1)
D

where D is the mass of AS-based coagulant recovered from metallurgical slag

(mg/l); RRS – the mass of removed TSS relatively to the applied coagulant dose (mg
TSS/mg AS-product dose); k = [a, b@±WKHHPSLULFDOFRUUHODWLRQFRHI¿FLHQWYHFWRU
The expression of this nonlinear correlation corresponds actually to the
standard linear model y = a + bx, with p – 2 number of parameters. The vector
k of the model constants (see Table 2) was obtained by application of classical
linear regression procedure with the least squares estimator21. Experimental error
standard deviation (noise level) was estimated from replicated experiments and

846
LWVOHYHOLVıy >> 0.1, the value representing ca. 14% of the observed average (i.e.
average yexp = 0.7016, see Table 1).

2.50

experimental – recovered AS
TSS removed/dose (mg/mg)

2.00
empirical model
experimental – benchmark M
1.50

1.00

0.50

0.00
0 2000 4000 6000 8000

dose AS (mg/l)
Fig. 1. Correlation between the applied coagulant dose (D) and resulted TSS-removal relative index
(RRS)

The next step to be completed after the estimation of statistical correlation

parameters (a, bLVTXDOLW\DQDO\VLVRILGHQWL¿HGVROXWLRQE\PHDQVRIWHVWVRIWKH
quality of statistical model. Model validation follows the standard methodology
described for explanation in this paper (see footnotes of Table 2)24–27. The results
of model adequacy tests and the estimated parameters quality tests are presented
in Table 2. These indicate an adequate regression model with experimental data,
WKDWPHDQVDJRRGPXOWLSOHFRUUHODWLRQFRHI¿FLHQWWRDVWDQGDUGGHYLDWLRQ
RIPRGHOSUHGLFWLRQVDURXQGRIWKHREVHUYHGYDOXHDQGDVDWLVIDFWRU\Ȥ2 ad-
HTXDF\WHVWIRUFRQ¿GHQFHOHYHO)ROORZLQJWKHLQWHUSUHWDWLRQRIWKHVWDWLVWLFDO
estimation theory, the estimated parameters quality tests show all model constants
DVEHLQJVLJQL¿FDQWLHVPDOOFRQ¿GHQFHLQWHUYDOVt-tests higher than the
FULWLFDOYDOXHRIFRQ¿GHQFHKLJKLQWHUFRUUHODWLRQFRHI¿FLHQWVDQGYDOXHV
RIWKHULGJHWHVWȜj / ı~ 2) much higher than the critical threshold 1.

Table 2. Model adequacy (a) and estimate quality (b) tests for the proposed algebraic correlation
(a) Model adequacy test
Adequacy test Test value Critical value Conclusion
R (note a) 0.9600 adequate
s (note b) 0.1675 шexp = 0.7016 adequate (s§GLQшexp)
Ȥ2 (note c) 2.8795 Ȥ2 (n – p; 0.95) = 100 adequate

847
(b) Estimate quality test
Param- Estimate FRQ¿- t-test Correlation matrix Ȝj/ı~ 2 Conclusion

r 0.0326
eter dence (note d) (note e) (note f) (note g)
a 0.1088 6.6389 1.0000 0.9570 1.78×102 VLJQL¿FDQW
b 37.1580 r 1.4960 49.4480 0.9570 1 8.00×105 VLJQL¿FDQW
Notes:
(a) R PXOWLSOHFRUUHODWLRQFRHI¿FLHQWHYDOXDWHGZLWKWKHUHODWLRQVKLS
n n n
™yi – ǔ)2 = (1 – R2™yi – ш)2; ш ™yi)/n;
i=1 i=1 i=1
(b) s2 ||yexp – ǔmodel||22/(n – p) – the model prediction variance; n = 68 – the number of experimental
points; p = 2 is the number of model parameters;
(c&DOFXODWHGȤc2 = s2ı2Ȥ2(df, qGHQRWHVWKHTXDQWLOHRIWKHȤ2-statistics with df degree of freedom
and qFRQ¿GHQFHOHYHO
(d&RQ¿GHQFHLQWHUYDORIWKHHVWLPDWHGSDUDPHWHUkj, computed with the relationship:

¦V
1 ª w yu º ª w yu º
2 « » « »
T

kj = kj ± {(V( k))jj)1/2}t(n – p; 97.5%); V(kˆ ) [p1, p ]

n

u ¬ w k ¼ [ p ,1] ¬ w k ¼ [1, p ]
^ ^

u 1

V – the estimate covariance matrix; k = [a, b]; t(df; q) – the quantile of the Student-statistics with
df degree of freedom and qFRQ¿GHQFHOHYHO

! t (n  p; 97.5%) = 2.042.
kˆ j
(e6WXGHQWWHVWIRUSDUDPHWHUVLJQL¿FDQFH t j =
[V (kˆ )] jj

R(kˆ ) [ p , p ] ^Rij `; Rij

(f) R is the parameter inter-correlation matrix calculated with the relationship:

 [1, 1] .
[V (kˆ )]ij
[V (kˆ )]ii [V (kˆ )] jj
(g7KHSDUDPHWHUULGJHVHOHFWLRQWHVWȜjaı2!±ZKHUHȜj are eigenvalues of matrix U.

¦¦ «¬ w k »¼
ª w yiu º ª w yiu º
« » V
~ 2 I; V min (V ui2 )
n r T

[ p , 1] ¬ w k ¼ [1, p ]
U (kˆ ) [ p , p ] ~2
u 1 i 1

7KHJUDSKLFDOUHSUHVHQWDWLRQVRIWKHPRGHOHUURUVLOOXVWUDWHGLQ)LJFRQ¿UP
WKHVDWLVIDFWRU\DGHTXDF\RIWKHVWDWLVWLFDOPRGHO3ORWVZLWKVXFKDOOXUHFRQ¿UP
the normal statistical distribution of experimental error, the linear character of the
log-correlation, and the absence of outliers in data set.

848
 a b

c d

Fig. 2. Residual plots (O) for the empirical model of equation (1)
a – observed y = RRS(exp., *) and predicted y = RRS(model, __ ) versus x = ln(1 + D)/D; b – residu-
als [RRS(exp.) – RRS(model)] versus ln(1 + D)/D; c – predicted y = RRS(model) versus observed
y = RRS(exp.); d – residuals [RRS(exp.) – RRS(model)] versus observed y = MSR(exp.)

In absolute terms, it is important to observe that the mass of removed TSS

(MRS, mg TSS/l) has a variation increasing logarithmically with coagulant dose
applied due to the transformed equation (1) in:
ln(1 + D)
yabs = MRS (mass of removed TSS) = a + b (2)
D

with the limits:

lim yabs = D[b0@ ’
Dĺ’

lim yabs = D[b0 + b1] = 0

Dĺ

$VZHNQRZWKHFKHPLFDOWUHDWPHQWRIZDVWHZDWHUE\FRDJXODWLRQÀRFFXODWLRQ
FDQEHGRQHLQGLIIHUHQWSRLQWVRIWKHSURFHVVÀRZDQGWKHUHPRYDOVWUDWHJ\RI
excess suspended solids must consider the obligation that quality parameters of
HIÀXHQWQHHGWRFRQIRUPWRWKHFXUUHQWUHJXODWLRQZKHQZDVWHZDWHULVGLVFKDUJHG
LQZDWHUERGLHV)RUH[DPSOHGXULQJWKHORZÀRZRIULYHURUZKHQWKHUHDUHVHD-
sonal variations in the pollutant loading of wastewater is necessary to improve
ZDVWHZDWHUWUHDWPHQWWKHFKHPLFDOSURFHVVLVVXI¿FLHQWO\ÀH[LEOHDQGFDQRSHUDWH
discontinuously) and one of the practical applications of derived correlation model
can be in this situation. If we desire to predict the necessary dose of aluminium
sulphate to remove suspended solids MRS = 500 mg/l from a wastewater, the
value of dose D is obtained by solving the nonlinear equation (2) that is D =

849
fct(yabs, a, b$SSO\LQJDXVXDOPHWKRGOLNHVXFFHVVLYHDSSUR[LPDWLRQUXOHUHVXOWV
a dose of aluminium sulphate D = 900 mg/l, i.e. a value never actually tested in
the laboratory (see Table 1).

CONCLUSIONS
7KLVSDSHUUHÀHFWVWKHFXUUHQWFRQFHUQVDERXWUHGXFLQJWKHLPSDFWRIFRQWDPLQDQWV
on water resources by removing them from wastewater, and to support an uncon-
ventional method of obtaining the process reagent (aluminium sulphate recovered
from metallurgical slag) through valorisation of waste with low uses.
Lab-scale tests performed in this study on various categories of wastewater
sources which have different load of pollutants reveal the very good properties
of recovered AS versus commercial AS obtained from bauxite, contributing to
agglomeration of the suspended solids from wastewater.
0DWKHPDWLFDOPRGHOVVXI¿FLHQWO\DGHTXDWHDQGZLWKVDWLVIDFWRU\SUHGLFWDELO-
ity are essential tools for analysing the process of wastewater treatment and offer
the possibility of theoretical studies by generalising the experimental results. The
most adequate type of statistical regression model has proven to be the model of
QRQOLQHDUFRUUHODWLRQWKDW¿WVWKHEHVWWRREVHUYHGGDWD±WKHPDVVRIUHPRYHG766
is logarithmic increasing with applied dose.
The mathematical model developed to characterize the chemical treatment
process of wastewater using aluminium sulphate recovered from metallurgical
slag has many practical advantages. This statistical model allows either predicting
of the coagulant dose necessary to remove the desired amount of pollutant from
wastewater or estimating the amount of pollutant that can be removed by applying
DNQRZQGRVHRIFRDJXODQW
7KHVWDWLVWLFDOFRUUHODWLRQSUHVHQWHGLQWKLVSDSHUZRUNLVYHU\XVHIXOWRVSHFLDO-
LVWVLQZDWHUPDQDJHPHQWIRUVROYLQJVSHFL¿FWDVNVVXFKDVGHVLJQFDOFXODWLRQRI
LQGXVWULDOSODQWVIRUFRDJXODWLRQÀRFFXODWLRQWUHDWPHQWDQGRSWLPXPRSHUDWLRQRI
WKHSURFHVVRQGLIIHUHQWDVSHFWVRIHI¿FLHQF\HFRQRPLFVZDVWHPLQLPLVLQJHWF

NOMENCLATURE

a, b = statistical correlation parameters

D = mass of AS based coagulant recovered from metallurgical slag
f = estimation objective function
I = identity matrix
k = parameter vector of the correlation model
MRS = mass of removed TSS
n = number of measured experimental points
p = number of parameters in the statistical model
r = number of observed variables
R = Inter-correlation matrix of the parameter

850
U = PRGL¿HGLQWHUFRUUHODWLRQPDWUL[RIWKHSDUDPHWHU
RRS = mass of removed TSS relatively to the applied coagulant dose
s2 = model prediction variance
t = Student statistical distribution
V = parameter variance-covariance matrix
x = independent variable
y = dependent variable
*UHHNOHWWHUV
Ȥ
O
= statistic distribution

V
= eigenvalues of the matrix U
= noise level
ı
~
= minimum noise level
Superscripts:
^ = estimated/predicted value
Abbreviations:
AS = aluminium sulphate
TSS = total suspended solids

Acknowledgements. Financial support from the ANCS – CNMP (Romania), project number PCCDI
1486/2007–2010 ‘Hazardous waste processing technology from the secondary aluminum industry
to obtain coagulants used to purify water in order to prevent environmental pollution and natural
UHVRXUFHFRQVHUYDWLRQ¶LVJUDWHIXOO\DFNQRZOHGJHG

REFERENCES
 0,/,()0$5,1(6&8**+,ğĂ*'(È.*67Ă1$6(05$,6&+,$VVHVVPHQW
of Heavy Metal in Water and Sediments of the Danube River. J Environ Prot Ecol, 15 (3), 874
(2016).
2. V. DAESCU, E. HOLBAN, R. ISBASOIU: Issues Concerning the Consequences of the Ap-
plication of Emission Mitigation Measures and Mercury Utilisation Stipulated in EU Strategy
on Mercury. J Environ Prot Ecol, 12 (3A), 1275 (2011).
3. D. M. ROSIORU, A. OROS, L. LAZAR: Assessment of Heavy Metals Contamination in Bivalve
Mytilus galloprovincialis Using Accumulation Factors. J Environ Prot Ecol, 17 (3), 825 (2014)
4. V. E. ARTEMYEV: Geochemistry of Organic Matter in River-Sea Systems. Springer, London,
1996.
5. R. D. LETTERMAN, A. AMIRTHARAJAH, C. R. O’MELIA: Coagulation and Flocculation
LQ:DWHU4XDOLW\DQG7UHDWPHQWWKHG0F*UDZ+LOO1HZ<RUN
0,0$5&8609/$'0$0,7,8,QÀXHQFHRI'LIIHUHQW*DOYDQLF6OXGJH7\SHVRQWKH
([WUDFWLRQ(I¿FLHQF\RI&KURPLXP,RQV$GY0DW5HV, 1143, 108 (2017).
7. P. M. BERTSCH, D. R. PARKER: Aqueous Polynuclear Aluminium Species. In: The Environ-
mental Chemistry of Aluminium (Ed. G. Sposito). 2nd ed, CRC Press, Florida, 1996.
8. J. GREGORY, J. DUAN: Hydrolyzing Metal Salts as Coagulants. Pure Appl Chem., 73, 2017
(2001).
9. G. TCHOBANOGIOUS, F. L. BURTON: Wastewater Engineering. Treatment, Disposal and
5HXVH0F*UDZ+LOO1HZ<RUN
0(6&+/(6,1*(5$OXPLQLXP5HF\FOLQJ&5&3UHVV1HZ<RUN

851
 11. J. A. TENORIO, D. C. ESPINOSA: Effect of Salt/Oxide Interaction on the Process of Aluminium
Recycling. J Light Met, 2, 89 (2002).
(XURSHDQ$OXPLQLXP$VVRFLDWLRQ 2UJDQL]DWLRQRI(XURSHDQ$OXPLQLXP5H¿QHUVDQG5HPHOW-
ers: Aluminium Recycling – the Road to High Quality Products. 1st ed, Brussels, www.eaa.net,
2005.
13. L. ZHANG: State of the Art in Aluminium Recycling from Aluminium Dross. J Light Met,
Proceedings of Sessions, TMS Annual Meeting, Texas, 2006, 931–936.
14. Y. XIAO, M. A. REUTER, U. BOIN: Aluminium Recycling and Environmental Issues of Salt
Slag Treatment. J Environ Sci Health, 40, 1861 (2005).
15. G. DEAK, V. DAESCU, E. HOLBAN, P. MARINESCU, G. S. TANASE, R. CSERGO, A. I.
DAESCU, S. GAMAN: Health-environment Relation – a Key Issue of Romanian Environmental
Protection. J Environ Prot Ecol, 16 (1), 304 (2015).
'0&2&$57$61($078$05(6(7$5'($&&DUFLQRJHQLF5LVN(YDOXDWLRQIRU
+XPDQ+HDOWK5LVN$VVHVVPHQWIURP6RLOV&RQWDPLQDWHGZLWK+HDY\0HWDOV,QW-(QYLURQ6FL
Technol (IJEST), 13 (8), 2025 (2016).
$*,/0DQDJHPHQWRI6DOW&DNH*HQHUDWHGDW6HFRQGDU\$OXPLQLXP0HOWLQJ3ODQWVE\'LVSRVDO
LQD&RQWUROOHG/DQG¿OO&KDUDFWHULVWLFVRIWKH&RQWUROOHG/DQG¿OODQG6WXG\RI(QYLURQPHQWDO
Impacts. Environ Eng Sci, 24, 1234 (2007).
&0$5,$&72&,8*0$5,$,PSURYHPHQWRI$TXDWLF3ROOXWDQW3DUWLWLRQ&RHI¿FLHQW
Correlations Using 1D Molecular Descriptors – Chlorobenzene Case Study. Chem Pap, 67, 173
(2013).
19. K. E. LORBER, H. ANTREKOWITSCH: Treatment and Disposal of Residue from Aluminium
Dross Recovery. In: Proc. of the 2nd International Conference on Hazardous and Industrial Waste
Management, Chania-Crete, 5–8 October 2010.
&72&,8(',$&8&0$5,$0LQLPL]DWLRQRI&KHPLFDO5LVNE\8VLQJ5HFRYHUHG$OX-
minium from Metallurgical Slag as Coagulant in Wastewater Treatment. Environ Eng Manag J,
13, 429 (2014).
21. R. TEODORESCU, M. ROMAN, C. TOCIU, M. GHEORGHE, A. CRISAN: The Technology
for Aluminium Recovery from Various Aluminium Slag for Environmental Depolution. Metal
Int, 14, 131 (2009).
22. R. TEODORESCU, V. BADILITA, M. ROMAN, V. PURCARU, P. CAPOTA, C. TOCIU,
M. GHEORGHE, A. CRISAN: Optimization of Process for Total Recovery of Aluminium from
Smelting Slag. 2-Removal of Aluminium Sulfate. Environ Eng Manag J, 13, 7 (2013).
23. Z. SATTERFIELD: Jar Testing. Tech Brief, The National Environmental Services Center at West
Virginia University, Morgantown, West Virginia, 2005.
24. G. MARIA: Statistical Data Analysis and Correlations. Distributions and Estimators. Printech
Publishing House, Bucharest, 2008.
25. D. M. BATES, D. G. WATTS: Nonlinear Regression Analysis and Its Applications. Wiley, New
<RUN
*0$5,$&&5,67,1$56$/&('26)(<2GH$=(9('2'DWDEDQN7UDQVIHURI
Information, Shortcut and Exact Estimators Used in the Wastewater Biological Treatment Process
,GHQWL¿FDWLRQ&RPSXW&KHP(QJ24, 1713 (2000).
27. G. MARIA, D. W. T. RIPPIN: A Note Concerning Two Techniques for Complex Kinetic Pathway
Analysis. Chem Eng Sci, 48, 3855 (1993).
Received 7 May 2017
Revised 15 June 2017

852

View publication stats