©
\
THE MECHANISM OF ORGANIC~
REMOVAL DURING COAGULATION ~
by
J. L. BERSILLON, Geol. Eng., D. Eng.
A ThesisSubmitted to the School of Graduate Studiesin Partial Fulfillment of the Requirements
for the DegreeDoctor of Philosophy
McMaster University
1983
J. L. Bersillon, 1983
~I
Il
1jl
iIi",,
l
!,
..,.j
II
\ .
THE MECHANISM OF ORGANIC
REMOVAL DURING COAGULATION
..
!
..
•
"
DOCTOR OF PHILOSOPHY (1983)(Civil Engineering)
..JMcMaster UniversityHamilton. Ontario
"The ~chanism ~f Organic Removal....During Coagulation
• TITLE:
AUTHOR:
•
SUPERVISOR:
'NUMBER OF PAGES:
\
•
Jean-Luc Bersi110nIngenieur Geo1ogueDocteur Ingenieur
Professor A. Benedek
xii, 188
i i
(Nancy. France)(Nancy, France)
"
. .
I
ABSTRACT
Coagulation is a common water treatme~t step pri-
mari ly designed to aggregate and thereby help remove t;:ue
particulate (turbidity causing) matter. Organic compounds
9f natural origin (i:e. humic and fulvic acids) have also
been observed to be',removed by Coagulation. This research
was designed primarily to identify the limittng mechanisms
responsible for the removal of natural organic~ by the coagu
latjon process. This identjficationis thought to be crucial
in the optimization of this important water treatment step
as it may help to maximize the benefits obtained from coagu-
...The examination of the literature related to this
topic suggests two possible removal routes:
(i) a reaction whereby the fulvic acids form an
original compound with the coagulating ion3+Al or one of its hydroxy complexes.
(ii) the adsorption of the fulvic acid molecules or
jons 0 tp_the s~rface of a solid precipitating
indepen ently of these compounds.
These two possibilities are examined in detail on theoretical
grounds. and two ollowing pieces of information are defined
iii
J
as being discriminatory with respec~ to ~he two removal
routes:
( i )l
The shape of the isotherm, as defined in tygi-
cal adsorp~ion studies.
(ii) The ligand number or OH/Al ratio of the
pr~cipitate, accessible by processing the
dissolved Aluminum data with respect to pH.
These considerations lead tb an experimental. design
allowing the convenient evaluation- of these characteristics.
An array of nine treatment dosages and four operating pH is
applied on two raw waters, using four Aluminum based coagu-
lants.
The results suggest that under these experimentpl
conditions (dosage between .1 and I mM III/L1 pH betw€en 5
and 8.5), the Fulvic acids are removed by Adsorption onto
Al ("OH)3' regardless of the type of coagulant. Increoising pH and
·the presen'ce of SuI fate in the coagul ant were found detrimental
to this adsorption. Increasing OH/Al ratio in the coagulant
is detrimental at low,pH, low dosage, and becomes beneficial
at ~eutral to mi ld 1y•
alkaline pH conditions., A two stage. '.-C .
treatment scheme was found efficient at neutral to mildly
alkaline pH, ~sing Alum.
i v l
/
ACKNOWLEDGEMENTS
I wish to thank first my advisor. Dr. Andrew Benedek
for his encouragements 'all along this research. and the
. technical ~nd frnancia~ support he offe~ed me. Without him.
this research cou·ld not have been done. Dr. J. R. Kramer's
I: thank also Dr. J. Y. Bottero for
collaboration 'on the characterization of
was greatly apprefi~ted.
were also very helpf~l.
The discussions
the fulvic material
wi th D~.' T'5'ezos
-
keeping me informed of the developments of his research.
Thi~ helped me in developing the present experimental design.
T~e technical as~~stance provided by the personnel
of Zenon Envi ronmerital Company. specially John Bancsi. Henri-
Behmann; Heather Donison and Mary Pejic was very helpful.
I would like to thank also Evan Diamadopoulos and
Roberto Narbaitz for the many discussions we.nad. technical
or. 0 the rw i s e .'
, Last. but not least. Lorraine Oneschuk's typing
sUlls ~elped to put this work.in its final format. I
thank her for he!' patience.
"
v
,/
,
TABLE OF CONTENTS
Chapter
I nt ro duc t ion
••Page
2 Literature Review2.1 Colour Causing Organics (
2.1.1 Nature and Occurrence2.1.2 Physical Chemical Properties2.).3 Reaction with Chlorine
2.2 Coagul ants -2.3 Aluminum Hydrolysis .
2.3.1 Hydrolysis of Soluble Species2.3.2 Complexation with Anion2.3.3 Precipitates2.3.4 Soluble .A1uminum Speciation
2.4 Organic-Aluminum Interplay -2.4.1 Reaction2.4.2 Adsorption '12.4.3 Coagulation
\
'. 3~-
4
Theoretical Expectations3.1 Possible Removal Pathways3.2 .The ,Reaction Hypothesis
3.2.1 General Considerations3.2,2 Straight Reaction3.2.3 Coprecipitation3.2.4 Formation' of an Irreversible
Precipitate3.3 The Adsorption,Hypoth~sis
3.3.1 TheLiga.ndNumber3.3.2 The Precipitate Loading
3.3.2.1 Most Usual Isotherm, EquationsI 3.3.2.2 Special Cases
3.3.2.3 Heterogeneity of the,Fu1vates
3.4 Significant Dependent Variables
Ex~erimental Methods4. Raw Waters Origin and Storage
.4.1.1 Fauquier Raw l/ater4.1.2 Distilled \later4.1.3 Fauquier Diluted Water4.1.4 Blanks
"
5567
1213151523242828293134
3535353539434~
44444546
4752
52
535353535454
•
-.
vi
•
Chapter
'.
54545557575859596161636364
66" 67
6772828~8695959698
9898
101105109109114...
.Page
pHSulfatethe
Reagents and Coagulants4.2.1 Stock Solutions4.~.2 Coagulant Operating SolutionsThe Jar Test Method .4~3.l Equipment4:3.2' Operating ConditionsSpecific Methods of Investigation4.4.1 Reversibility4.4.2 Kinetics4.4.3 Double TreatmentAnalytical Methods4.5.1 Equipment ..4.5.2 Measurements
•
4.5
4.4
4.3
4.2
Results and Discussion5:1 Ligand Number and Solubility
, 5.1.1 Caiculation of A13+ Activity5.1.2 Results5.1.3 Discussion
5.'2 Organic Removal5.2.1 Isotherms5.2.2 Influencing Factors
5.2.2.1 The Influence of5.2.2?2 The Influence of5.2.2.3 The Influence of
OH/Al ratio5.2.3 Discussion
5.,.3.1 Reversibility5.2'.3.2 Kinetics
5.2.4 Tentative Modelling5.3 Practical Impl ications
5.3.1 Double Treatment5.3:2 Precipitate Separability
5
I
(
l6 . Conclusions
7
·References
tor Further Research.lcatlon-o t e sorption ModelTreatment Process
116
11 9119120
121
Appendices
A. 1A.2
- Aluminum Determination MethodCharacterization of Colour Causing OrganicsA.2.l Iden!ification ~ the Material
129131131
vii
1./\).... -
Appendices' Page
A.2.2 1 em 10 cm DOC Correlation 133A254 nm/ A254 nm -A'. 2 . 3 Amounts.of DOC in the Raw Wa~ers , 135A.2.4 Discussion 135A.2 .5 References 140
A.3 General Observations 141A. 3. 1 Residual Organics 141A. 3.2 Residual Aluminum 148A.3.3 Physical Characteristics of the 151
PrecipitatesA.3.4 References 160
A.4 FRW Jar Test Results 161A.5 FDW Jar Test Results • 171A.6 Blank Runs Results 177A.7 Computer Programs 182
I
-.
j
vii i
.-
•
LIdT OF FIGURES
Figure Page
1.1 Place of the Coagulation Process inWater Treatment Plant
3,
2.1 Aluminum Speciation Diagram afterMay et al (1979)
19I
2.2 Aluminum Speciation Diagram afterParks (1972)
20
3. 1
3.2
Di fferent Removal Routes
Graphical Representation of the DispersionBrought by Fulvate Heterogeneity ."
36
42
3.3 Ionic Distribution Near the Surface(After Davis. 1980)
49
."
87
60
. 62
Adsorption Isotherms as a' Function of pH.FRW.Alum Runs
Modified Jar Test Schedule for theKinetic Experiment
-lo9[A1 3+] as a Function of pH = Blank Runs 74
-log[~13+] as a Function of pH = Blank Runs 77
-lo9[A1 3+1 as a Function of pH = FRW Runs 79
-log[A1 3+] as a Function of pH = FRW Runs 80
-log[A1 3+] as a Function of pH = FDW Runs 81
Jar Test Time Frame4. 1
4.2
5" 1
5.2
5.3
5.4
5" 5
5.6
"
\
".
5. 7
5.8
Adsorption Isotherms as a Function of pHFRW .. A1C1 3 Runs
Adsorption Isotherms as a Function of pHFRW.Alpo1 Runs
88
89
/'
Adsorption Isotherms as a Function of pHFRW.Alge1 Runs
ix-
90
Fi gu re
5.10
I
.•
Adsorption Isotherms as a Function of pH.FDW.Alum Runs-
E..age
92
5. 12
Adsorption Isotherms as a Function of pH. 93FDW.Al C1 3 Rims
2-Influence of S04 on the Residual Organics. 97Al doyage = 0.5 mM/L
5. 13
5.14
5.15
A2 . l'
A2.2
A2.3
A2.4
A3.1
,A3.2
A3.3
A3.4
A3.5
A3.6
pH as.a Function of Time,Mixing Time20 mn
pH as a Function of Time. Mixing Time200 mn
DOC as a Function of Time
Organics Light Absorption Spectra
UV Abso~bance as a Function of FRWDilution
UV Absorbance - DOC Correlation forFRW Experiments
OV Absorbance - DOC Correlation forFQW Expe ri men ts
Remaining DOC as a Function of AlumDosage and pl:!. Case of FRW
Remaining DOC as a Function of A1C1 3"Dosage and pH. Case of FRW
Remaining DOC as a Function of A1po1Dosage' and pH., C,ase of FRW
Remaining DOC as a Function of A1ge1Dosage and pH. Case of FRW ..
Remaining DOC as a F~nction of AlumDosage and pH. Case of.FDW
,RlW.Iilining DOC as a Function of A1C1 3Dosage and pH. Case of FDW
x•
,102
103
, -104,'
134
13p\
137
138
14"2
143
144
145
146
147
..,
xi
(
Table
2. 1
5. 1
5.2....
5. 3
5.4
5.5
5.6
5.7
5 .8
LIST OF TA8LES
Solubility Products [H+]3/[A1 3+] forthe Formation of Al (OH) 3
Characteristics of the Ions Used inthe Calculation of the Activity ofA13+ (after Truesdell et al, 19T4)
Reactions Accounted for in the Calculation of the Activity of A13+, andCorresponding Mass Action Law Constants
Slopes and Correlation CoefficientsDetermined by Linear RegressionAnalysis of log[A13+] vs pH
. . 3+Averages of the Ionic Product -1og[Al ]-3 pH as a Function of the Coagulant andthe Raw Water
Non Removable Fraction (X DOC o ) in FRW,and FDW, as a Function of the pH andthe Coagulant Nature
•Results of the ,Reversibility Experiment,Using Alum as Coagulant
Parameter Estimates for the FulvateAdsorption, Data ~
Double Treatment Experiments
,.
xii
"Page
26
70
73
83
84
94
100
,~
108
112
.\
CHAPTER 1
INTRODUCTION
The natural water bodies constitute the water
supply of all living communities. The human cOr:1munities
tap these resources for their needs, and most of the time,
this natural water must be treated to.fit to their use.
Among other undesirable substances, organic pollu-
tants are the target of the water treatment. These com-
pounds occur in most natural water sources. Natural
organics originate in soils, where part of them is dis-
solved by rain. These soil compounds referred to as humic
material, are responsible for the yellow to brown colour
of the marsh waters, as well as some river waters. On the
other hand, specific organic compounds found in natural
waters result directly from man's activity.
The need for the removal of these substances comes
from either ae'sthetic considerations (colour, taste, odour)
or their potential impact on public health. In recent
years, an increasing concern about'drinkingwater quality. .
promoted a research effort aimed at a better application
of the physical chemical treatment op.erations·through a
be~ter understanding of their underlying principles. These
1
2
procedures include Activated Carbon Adsorption, and several
chemical oxidation processes.
Interestingly, the most widely used treatment process,
i.e. coagulation - flocculation, received but little attep
tion as a potentially powerful process for removing soluble
organic contaminants. Although the interaction between
coagulants and the humic material has been investigated to
some extent, there is still a definite research need in this
area, especially regarding naturally coloured waters, as
outlined in a recent AWWA Committee Report (1979). On an
engineering point of view,' a better understanding of the
phenomena involved in the organic removal by the coagulation
process should lead to the formulation of a model. Th~s,
\modelization constitutes an essential tool, for the optimi-
zation of the process, and ultimately to the improvement of
the performances of-the process.
Such an improvement has consequences on all the
physical chemical processes involved in a water treatment
plant since the coagulation is the first step in this treat
ment, as shown in Figure 1-1. This statement has a special
value when Activated tarbon Adsorption is used. Assuming
that at equal cost, an improved coagulation process allows/--:
to increase the organic removal by this operation fr9m 75%
to 80%, a 5% improvement may look marginal. However, for
the subsequent finishing steps (GAC filtration, ozonation