The destabilization of dilute clay suspensions with labelled polymers

MISSING IMAGE

Material Information

Title:
The destabilization of dilute clay suspensions with labelled polymers
Physical Description:
xvi, 175 leaves : illus. ; 28 cm.
Language:
English
Creator:
Birkner, Francis Bruno, 1939-
Publication Date:

Subjects

Subjects / Keywords:
Sedimentation and deposition   ( lcsh )
Polymers and polymerization   ( lcsh )
Clay   ( lcsh )
Civil Engineering thesis Ph. D
Dissertations, Academic -- Civil Engineering -- UF
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis - University of Florida.
Bibliography:
Bibliography: leaves 169-174.
General Note:
Manuscript copy.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000559281
oclc - 13454042
notis - ACY4730
System ID:
AA00003565:00001

Full Text










THE DESTABILIZATION OF DILUTE

CLAY SUSPENSIONS WITH

LABELLED POLYMERS

















By
FRANCIS BRUNO BIRKNER









A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY










UNIVERSITY OF FLORIDA
August, 1965














ACKNOWLEDGMENTS


The author would like to express his sincere appreciation to his

committee chairman, Dr. A. P. Black, who provided the guidance, assis-

tance, and inspiration which were required to complete this dissertation.

The author is also deeply indebted to Dr. J. J. Morgan who gave his time

and advice whenever they were requested. Specific acknowledgment is

also due the following members of the Graduate Supervisory Committeez

Professors John E. Kiker, Jr., Thomas deS. Furman, George B. Morgan,

and Charles G. Edson. Sincere thanks are also extended to all the

author's colleagues in the water chemistry laboratory and to Mrs. J. G.

Larson for typing the dissertation.

Recognition and appreciation is also extended to Dr. B. G.

Dunavant, Department of Radiology, University of Florida, for use of

the department's liquid scintillation spectrometer; Peninsular Chem-

research Inc., Gainesville, Florida, for supplying the C14 labelled

cationic polymer; the Dow Chemical Company, Midland, Michigan, for

supplying the C14 labelled anionic and nonionic polymers, and the

United States Public Health Service whose financial support made this

study possible.















CONTENTS


Page


LIST OF TABLES. .

LIST OF FIGURES .


* 9 9 9 9 9 9 S 9 9 0

* 9 0 9 9 9 0 9 9 9 9


ABSTRACT . . .

CHAPTER

I. INTRODUCTION . .


II. THEORY .


S O O 9 0 9 0 9 O


Clays . . .
Structural Classification . .
Chemical Classification . .
Size of Clay Particles . ..
Origin of Surface Charge. .
Double Layer Structure .. .
Stability and Instability of Clay Suspensions.
Polyions .. .. ...
Polymer Configuration .. .. .
Ionic Charge and Charge Density .
Polymer Adsorption Models ... ..
Destabilization . ..
Coagulation . .
Flocculation .. .
Polymer Clay Interaction .. .

III. HISTORY . . .


Clay-Polymer-Water Systems .
Other Colloid-Polymer-Water Systems


* 9 9 9
* 9 9 9


IV. PURPOSE AND SCOPE . .

V. EXPERIMENTAL MATERIALS AND PROCEDURES .

Materials . .
Clays .. . .
Organic Polyelectrolytes . .
Calcium Chloride .


iii


V

ix

xiv


ACKNOWLEDGMENTS . .. . .










Page

Procedures . .. 42
Preparation of Clay Suspensions 42
Preparation of Organic Polyelectrolytes 44
Destabilization Experiments .. .. 45
Sample Preparation .. .. 45
Coagulation-flocculation . 46
Initial and residual turbidity measure-
ments . .. 46
Measurement of pH .. 47
Electrophoretic Mobility Determinations 47
Sample preparation . .. 47
Conductance measurement 48
Measurement of pH . 48
Measurement of electrophoretic mobility of
floc particles . .... 48
Residual Polymer Determinations 48
Liquid Scintillation Spectrometry 48
Scintillator system . 51
Standard curve calibration and sensitivity
and accuracy of the determinations 53
Sample preparation and counting pro-
cedures . .. 54
Polymer-Clay Kinetic Experiments 55

VI. DISCUSSION OF RESULTS AND CONCLUSIONS 57

Cationic Polymer-Clay Interactions 57
Anionic Polymer-Clay Interactions 107
Nonionic Polymer-Clay Interactions .. 117

VII. SUMMARY OF CONCLUSIONS ... 126

APPENDIX . . . 129

LIST OF REFERENCES . . 168

BIOGRAPHICAL SKETCH . .. 175












LIST OF TABLES


Table Page

1. Clay Suspension Concentrations . 43

2. Composition of Synthetic Water ........ 44

3. Liquid Scintillation Counting System 52

4. Values of Electrophoretic Mobility, Residual Turbidity,
and Polymer Adsorption at the Optimum Polymer Dosage
for Each Cationic, Anionic, and Nonionic Polymer-Clay
System ....... ................ 63

5. Summary of Adsorption Parameters of Cationic, Anionic,
and Nonionic Polymer-Clay Systems . 100

6. The Destabilization of a Kaolinite Clay Suspension with
Polymer No. 4. Initial Clay Concentration 14.9 mg/l. 130

7. The Destabilization of a Kaolinite Clay Suspension with
Polymer No. 4. Initial Clay Concentration 29.8 mg/l. *. 131

8. The Destabilization of a Kaolinite Clay Suspension with
Polymer No. 4. Initial Clay Concentration 73.2 mg/l. 132

9. The Destabilization of a Montmorillonite Clay Suspension
with Polymer No. 4. Initial Clay Concentration 34.4 mg/1. 133

10. The Destabilization of a Montmorillonite Clay Suspension
with Polymer No. 4. Initial Clay Concentration 66.8 mg/l. 134

11. The Destabilization of a Montmorillonite Clay Suspension
with Polymer No. 4. Initial Clay Concentration 144 mg/l 135

12. The Destabilization of a Kaolinite Olay Suspension with
Polymer No. 6. Initial Clay Concentration 14.9 mg/l. 137

13. The Destabilization of a Kaolinite Clay Suspension with
Polymer No. 6. Initial Clay Concentration 29.8 mg/l. 138

14. The Destabilization of a Kaolinite Clay Suspension with
Polymer No. 6. Initial Clay Concentration 73.2 mg/l. *. 139

15. The Destabilization of a Montmorillonite Clay Suspension
with Polymer No. 6. Initial Clay Concentration 34.4 mg/l. 140












16. The Destabilization of a Montmorillonite Clay Suspension
with Polymer No. 6. Initial Clay Concentration 66.8 mg/l. 141

17. The Destabilization of a Montmorillonite Clay Suspension
with Polymer No. 6. Initial Clay Concentration 144 mg/l. 142

18. The Destabilization of a Kaolinite Clay Suspension with
an Anionic Polymer. Initial Clay Concentration 33.3 mg/1.
CaC12 Concentration 25.0 mg/1. .. .. # 143

19. The Destabilization of a Kaolinite Clay Suspension with
an Anionic Polymer. Initial Clay Concentration 33.3 mg/1.
CaCl2 Concentration 250 mg/l. . 144

20. The Destabilization of a Kaolinite Clay Suspension with a
Nonionic Polymer. Initial Clay Concentration 33.4 mg/l.
CaC12 Concentration 25.0 mg/l. .. .. .. .. 145

21. The Destabilization of a Kaolinite Clay Suspension with a
Nonionic Polymer. Initial Clay Concentration 33.4 mg/1.
CaC12 Concentration 250 mg/l. . 146

22. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Kaolinite. Clay Concentration 14.9 mg/l. Polymer No. 4 147

23. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Kaolinite. Clay Concentration 29.8 mg/l. Polymer No. 4 148

24. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Kaolinite. Clay Concentration 73.2 mg/l. Polymer No. 4. 149

25. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Montmorillonite. Clay Concentration 34.4 mg/l. Polymer
No. 4 e . .. 150

26. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Montmorillonite. Clay Concentration 66.8 mg/1. Polymer
No. 4. . . . 151

27. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Montmorillonite. Clay Concentration 144 mg/1. Polymer
No. 4 . .. . 152

28. Selected Adsorption Data for Languuir Isotherm Plot. Clay
Kaolinite. Clay Concentration 14.9 mg/1. Polymer No. 6. 153

29. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Kaolinite. Clay Concentration 29.8 mg/1. Polymer No. 6. 154


Page


Table













30. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Kaolinite. Clay Concentration 73.2 mg/1. Polymer No. 6. 155

31. Selected Adsorption Data for Langnuir Isotherm Plot, Clay
Montmorillonite. Clay Concentration 34.4 mg/1. Polymer
No. 6. . . .* 156

32. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Montmorillonite. Clay Concentration 66.8 mg/l. Polymer
No. 6. t .. . r 157

33. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Montmorillonite. Clay Concentration 144 mg/1. Polymer
No. 6. e .. .. 158

34. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Kaolinite. Clay Concentration 33.3 mg/l. Anionic Polymer.
CaCI2 Concentration 25.0 mg/1. ...... .* 159
35. Selected Adsorption Data for Langmiir Isotherm Plot. Clay
Kaolinite. Clay Concentration 33.3 mg/l. Anionic Polymer.
CaCI2 Concentration 250 mg/1. .. 160

36. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Kaolinite. Clay Concentration 33.4 mg/1. Nonionic Polymer.
CaC2 Concentration 25.0 mg/1. . 161

37. Selected Adsorption Data for Langmuir Isotherm Plot. Clay
Kaolinite. Clay Concentration 33.4 mg/l. Nonionic Polymer.
CaCI2 Concentration 250 mg/ . 162

38. Stoichiometry of the Destabilization of Clay Suspensions
with Cationic Polymers ........ ... ...... 163

39. The Adsorption Kinetics of a Cationic Polymer Under Two
Different Intensities of Agitation. Cationic Polymer No. 6.
Clay Kaolinite. Polymer Dosage 40 pg/1. Clay Concentration
29.8 mg/1. . . . 164

40. The Effect of Time of Agitation Upon the Electrophoretic
Mobility of a Cationic Polymer-Clay Suspension. Cationic
Polymer No. 6. Clay Kaolinite. Polymer Dosage 500 1g/1.
Clay Concentration 29.8 mg/1. . 165

41. The Effect of Time and Intensity of Agitation Upon the
Destabilization of a Kaolinite Clay Suspension with a
Cationic Polymer. Initial Clay Concentration 29.8 mg/1.
Cationic Polymer No. 6. . .. 166


Page


Table













42. The Effect of Time and Intensity of Agitation Upon the
Destabilization of a Kaolinite Clay Suspension with a
Cationic Polymer. Initial Clay Concentration 73.2 mg/1.
Cationic Polymer No. 6... ...... 167


viii


Page


Table













LIST OF FIGURES


Figure Page

1. Three-layer Clay Unit Cell . 4

2. Atom Arrangement in a Kaolinite Unit Cell.(From van
Olphen10) . . 6

3. Atom ArSangement in a Montmorillonite Unit Cell.(From van
Olphen ). . . . 7

4. Stern Model of the Electrical Double Layer. (From Mysels13). 12

5. Selected Polymers .............. 17

6. Block Diagram of Tri Carb Model 314 EX Liquid Scintillation
Spectrometer. (From Packard Instrument Manual). ... 50

7. The Destabilization of a Kaolinite Clay Suspension with
Cationic Polymer No. 4. Initial Clay Concentration 14.9
mg/1 . . . 58

8. The Destabilization of a Kaolinite Clay Suspension with
Cationic Polymer No. 4. Initial Clay Concentration 29.8
mg/1 . . . 59

9. The Destabilization of a Kaolinite Clay Suspension with
Cationic Polymer No. 4. Initial Clay Concentration 73.2
mg/l . . . 60

10. The Effect of Initial Clay Concentration on the Destabili-
zation of Kaolinite Clay Suspensions with Cationic Polymer
No. 4. Initial Clay Concentrations:------14.9 mg/l;
29.8 mg/ 73.mg/; g/1 61

11. The Destabilization of a Montmorillonite Clay Suspension
with Cationic Polymer No. 4. Initial Clay Concentration
34.4 mg/1. . . . 64

12. The Destabilization of a Montmorillonite Clay Suspension
with Cationic Polymer No. 4. Initial Clay Concentration
66.8 mg/1 . . 65

13. The Destabilization of a Montmorillonite Clay Suspension
with Cationic Polymer No. 4. Initial Clay Concentration
144 mg/l . . . 66













14. The Effect of Initial Clay Concentration on the Destabili-
zation of Montmorillonite Clay Suspensions with Cationic
Polymer No. 4. Initial Clay Concentrations: ----- --
34.4 mg/1; -66.8 mg/1; ------ 144 mg/. 67

15. The Destabilization of a Kaolinite Clay Suspension with
Cationic Polymer No. 6. Initial Clay Concentration
14.9 mg/1. . . . 69

16. The Destabilization of a Kaolinite Clay Suspension with
Cationic Polymer No. 6. Initial Clay Concentration
29.8 mg/1. . .. .... 70

17. The Destabilization of a Kaolinite Clay Suspension with
Cationic Polymer No. 6. Initial Clay Concentration
73.2 ag/1. . . .. 71

18. The Effect of Initial Clay Concentration on the Destabili-
zation of Kaolinite Clay Suspensions with Cationic Polymer
No. 6. Initial Clay Concentrations:----- -14.9 mg/l;
29.8 mg/l; ---73.2 mg/I ....... 72

19. The Destabilization of a Montmorillonite Clay Suspension
with Cationic Polymer No. 6. Initial Clay Concentration
34.4 mg/ . . 73

20. The Destabilization of a Montmorillonite Clay Suspension
with Cationic Polymer No. 6. Initial Clay Concentration
66.8 mg/1 . . 74

21. The Destabilization of a Montmorillonite Clay Suspension
with Cationic Polymer No. 6. Initial Clay Concentration
144 mg/1 . . . 75

22. The Effect of Initial Clay Concentration on the Destabili-
zation of Montmorillonite Clay Suspensions with Cationic
Polymer No. 6. Initial Clay Concentrations: -------
34.4 mg/1; 66.8 mg/1;- -- 144 mg/1. ... 76

23. Stoichiometry of the Destabilization of Kaolinite Clay
Suspensions with Cationic Polymers (0 Polymer No. 4;
+ Polymer No. 6) . . 78

24. Stoichiometry of the Destabilization of Montmorillonite
Clay Suspensions with Cationic Polymers ( Polymer No. 4;
+ Polymer No. 6). . .... 79


Figure


Page












25. Langmuir Adsorption Isotherm. Cationic Polymer No. 4;
Kaolinite Clay Concentration 14.9 mg/l. . 84

26. Langmuir Adsorption Isotherm. Cationic Polymer No. 4;
Kaolinite Clay Concentration 29.8 mg/l. . 85

27. Langmuir Adsorption Isotherm. Cationic Polymer No. 4;
Kaolinite Clay Concentration 73.2 mg/1. . 86

28. The Effect of Initial Clay Concentration on the Langmuir
Adsorption Isotherms of Kaolinite-Cationic Polymer Systems.
Polymer No. 4; Initial Clay Concentrations; ---- -14.9 mg/l;
29.8 mg/;--- 73.2 mg/1. 87

29. Langmuir Adsorption Isotherm. Cationic Polymer No. 4;
Montmorillonite Clay Concentration 34.4 mg/l. 88

30. Langmuir Adsorption Isotherm. Cationic Polymer No. 4;
Montmorillonite Clay Concentration 66.8 mg/. 89

31. Langmuir Adsorption Isotherm. Cationic Polymer No. 4;
Montmorillonite Clay Concentration 144 mg/l . 90

32. The Effect of Initial Clay Concentration on the Langmuir
Adsorption Isotherms of Montmorillonite-Cationic Polymer
Systems. Polymer No. 4: Initial Clay Concentrations:
---- 34.4 mg/1; 66.8 mg/; -----144 mg/l. 91

33. Langmuir Adsorption Isotherm. Cationic Polymer No. 6;
Kaolinite Clay Concentration 14.9 mg/1. .. 92


34. Langmuir Adsorption Isotherm. Cationic Polymer No. 6;
Kaolinite Clay Concentration 29.8 mg/. . 93

35. Langmuir Adsorption Isotherm. Cationic Polymer No. 6;
Kaolinite Clay Concentration 73.2 mg/1. . 94

36. The Effect of Initial Clay Concentration on the Langmuir
Adsorption Isotherms of Kaolinite-Cationic Polymer Systems.
Polymer No. 6; Initial Clay Concentrations: ---- -
14.9 mg/1; 29.8 mg/; -----73.2 mg/1. 95

37. Langmuir Adsorption Isotherm. Cationic Polymer No. 6;
Montmorillonite Clay Concentration 34.4 mg/1. 96

38. Langmuir Adsorption Isotherm. Cationic Polymer No. 6;
Montmorillonite Clay Concentration 66.8 mg/1. 97


Figure


Page












39. Langmuir Adsorption Isotherm. Cationic Polymer No. 6;
Montmorillonite Clay Concentration 144 mg/1. 98

40. The Effect of Initial Clay Concentration on the Langmuir
Adsorption Isotherms of Montmorillonite-Cationic Polymer
Systems. Polymer No. 6; Initial Clay Concentrations:
34.4 mg/1; -66.8 mg/1; ---
144 mg/1. . . . 99

41. The Adsorption Kinetics of a Cationic Polymer Under Two
Different Intensities of Agitation ( @ 20 rpm, 0 100 rpm).
Cationic Polymer No. 6; Polymer Dosage 40 pg/l; Kaolinite
Clay Concentration 29.8 mg/1. Polymer Dosage for Mobility
Data is 500 g/1. . 103

42. The Effect of Time of Mixing, Expressed as Total Number of
Paddle Revolutions, on the Destabilization of a Kaolinite
Clay Suspension (Initial Clay Concentration 29.8 mg/1) with
Cationic Polymer No. 6 at Two Intensities of Agitation
( --- ---- 20 rpm; 100 rpm). . 105

43. The Effect of Time of Mixing, Expressed as Total Number of
Paddle Revolutions, on the Destabilization of a Kaolinite
Clay Suspension (Initial Clay Concentration 73.2 mg/1) with
Cationic Polymer No. 6 at Two Intensities of Agitation
(---- --20 rpm; -100 rpm). ,........ 106

44. The Effect of Time of Mixing, Expressed as Total Number of
Paddle Revolutions, 6n the Destabilization of Two Kaolinite
Clay Suspensions (Initial Clay Concentrations: 0 73.2 mg/1;
29.8 mg/1) with Cationic Polymer No. 6 at 20 rpm Mixing
Intensity . . 108

45. The Effect of Time of Mixing, Expressed as Total Number of
Paddle Revolutions, on the Destabilization of Two Kaolinite
Clay Suspensions (Initial Clay Concentrations: 0 73.2 mg/1;
29.8 mg/1) with Cationic Polymer No. 6 at 100 rpm Mixing
Intensity. . . .. 109

46. The Destabilization of a Kaolinite Clay Suspension with an
Anionic Polymer. Initial Clay Concentration 33.3 mg/1;
CaCla Concentration 25.0 mg/1. . .. 110

47. The Destabilization of a Kaolinite Clay Suspension with an
Anionic Polymer. Initial Clay Concentration 33.3 mg/1;
CaC1, Concentration 250 mg/1. . 111


Figure


Page












48. The Effect of Calcium Chloride on the Destabilization of
Kaolinite Clay Suspensions (Concentration 33.3 mg/1)
with an Anionic Polymer. CaC1, Concentrations:---
25.0 mg/l; 250 mg/.... ... ......... 113

49. Langmuir Adsorptinn Isotherm. Anionic Polymer; Kaolinite
Clay Concentration 33.3 mg/l; CaC12 Concentration
25.0 mg/l. . . 114

50. Langmuir Adsorption Isotherm. Anionic Polymer; Kaolinite
Clay Concentration 33.3 mg/1; CaC1, Concentration
250 mg/l . . ....... 115

51. The Effect of CaC12 on the Langmuir Adsorption Isotherms
of Kaolinite-Anionic Polymer Systems. Kaolinite Clay
Concentration 33.3 mg/1. CaC1, Concentrations: -- --
25.0 mg/1' -250 mg/1. .............. 116

52. The Destabilization of a Kaolinite Clay Suspension with
a Nonionic Polymer. Initial Clay Concentration 33.4 mg/1;
CaC12 Concentration 25.0 mg/1. . 118

53. The Destabilization of a Kaolinite Clay Suspension with
a Nonionic Polymer. Initial Clay Concentration 33.4 mg/l;
CaC12 Concentration 250 mg/1 . .. 119

54. The Effect of Calcium Chloride on the Destabilization of
Kaolinite Clay Suspensions (Concentration 33.4 mg/1) with
a Nonionic Polymer. CaC12 Concentrations:
25.0 mg/l; --- 250 mg/. . 120

55. Langmuir Adsorption Isotherm. Nonionic Polymer; Kaolinite
Clay Concentration 33.4 mg/l; CaC12 Concentration
25.0 mg/l. . . .. 122

56. Langmuir Adsorption Isotherm. Nonionic Polymer; Kaolinite
Clay Concentration 33.4 mg/1; CaC12 Concentration
250 mg/1. . . . 123

57. The Effect of CaC12 on the Langmuir Adsorption Isotherms
of Kaolinite-Nonionic Polymer Systems. Kaolinite Clay
Concentration 33.4 mg/l. CaC1, Concentrations:-----
25.0 mg/l; 250 mg/l. .............. 124


xiii


Figure


Page











Abstract of Dissertation Presented to the Graduate Council in
Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy

THE DESTABILIZATION OF DILUTE CLAY SUSPENSIONS
WITH LABELLED POLYMERS

By

Francis Bruno Birkner

August 1965

Chairman: Dr. A. P. Black
Major Department: Civil Engineering

A detailed study has been made of the destabilization of dilute

clay suspensions with cationic, anionic, and nonionic polymers under

controlled conditions of pH, ionic strength, initial clay concentration,

and intensity and duration of solution agitation.
1k-
C labelled cationic, anionic, and nonionic polymers were

added to dilute kaolinite and montmorillonite clay suspensions, and the

clay-polymer systems were mixed at controlled intensities of agitation

for predetermined periods of time following which they were allowed to

sediment. Since anionic and nonionic polymers have been shown by others

to be ineffective in destabilizing dilute clay suspensions when used

alone, a simple coagulant, in this case CaC12, has been added to the

suspensions prior to the addition of polymer. The degree of destabiliza-

tion afforded by each polymer was evaluated by residual turbidity meas-

urements of the settled suspensions. Polymer residuals in the super-

natant liquid of the destabilized and settled clay suspensions were

determined by liquid scintillation counting techniques, and electro-

phoretic mobility measurements were made on the clay-polymer systems.


xiv










The destabilization of dilute clay suspensions with a cationic

polymer was found to occur over a limited polymer dosage range. An

optimum polymer dosage was found for each cationic polymer-clay system

investigated, and polymer dosages either larger or smaller than this

value resulted in incomplete destabilization of the suspension. The

width of the zone of good turbidity removal and the maximum degree of

suspension destabilization were found to depend on the type of clay

and the initial clay concentration of the suspension. In addition, a

stoichiometric relationship was found between the initial clay concen-

tration of the suspension and the optimum polymer dosage for each

cationic polymer-clay system investigated. The cationic polymer adsorp-

tion data fit the Langmuirian adsorption model very well and revealed

an inverse relationship between the initial clay concentration and the

saturation weight ratio of adsorbed polymer to clay. The adsorption of

cationic polymer molecules onto the clay particle surface reduced the

mobility of the clay particles and eventually reversed the sign of the

particle mobility at increased polymer dosages. Cationic polymer molec-

ular weight was found to have no significant effect on the degree of

suspension destabilization. However, the lower molecular weight polymer

did exhibit a narrower zone of optimum turbidity removal and consistently

showed a lower optimum polymer dosage than the higher molecular weight

polymer. The destabilization mechanism for cationic polymer-clay systems

is postulated to be a combined coagulation-flocculation reaction with

the cationic polymer acting both as a coagulant, in compressing the

double layers of the clay particles, and as a flocculant, in bridging











the particles via extended segments of adsorbed polymer molecules. The

kinetics of the polymer adsorption reaction have been found to be

extremely rapid. Consequently, the rate-controlling step in the overall

destabilization reaction is the flocculation reaction. The rate of

flocculation has been shown to depend upon the number of particles in

the suspension. However, the ultimate degree of turbidity removal

appears to be independent of the initial clay concentration and

intensity of solution agitation and dependent only upon the length of

the mixing process.

The destabilization of dilute clay suspensions with anionic and

nonionic polymers has been found to occur over a limited polymer dosage

range. However, a sufficient concentration of counterions must be

initially present in or added to the suspension in order to reduce

particle-particle, polymer-particle, and adsorbed polymer-polymer

repulsive forces so that interparticle bridging can occur.


xvi














I. INTRODUCTION


Turbidity in surface waters is composed of organic and inorganic

particulates of varying size and concentration. A large percentage of

the turbidity which occurs in surface waters is composed of various types

of clays which are deposited in streams and rivers through the process of

natural land erosion. These clay particles may range in size from a few

millimicrons to several microns, the very coarse particles remaining in

suspension only during flood flow.

Although the organic constituents of naturally occurring suspended

material may harbor pathogenic organisms, turbidity is removed from water

mainly for aesthetic reasons. Consequently, the United States Public

Health Service1 recommends that a potable water have a turbidity of less

than five units.

Inorganic coagulants have been used for many years to remove

turbidity from surface waters by coagulation and flocculation processes

followed by sedimentation and rapid sand filtration. Aluminum sulfate

is currently the most widely used inorganic coagulant for this purpose

although ferric sulfate is being used in some instances. However, ferric

sulfate finds its greatest use in the removal of color from water.2' 3

In the past decade or so, organic polyelectrolytes have had various

applications in many diversified fields. In the field of sewage treat-

ment they have been used mainly for the dewatering of sewage sludges

while the mining industry has used them for many years in ore dressing


-1-







-2-


processes. Soil scientists4' 5, 6 have also found that organic poly-

electrolytes are very effective as soil conditioners for improving the

porosity of soils. However, up to the present time, organic polyelectro-

lytes have found only limited use as coagulant aids in water treatment

processes.

In recent years the coagulation and flocculation of colloidal sus-

pensions by organic polyelectrolytes has become increasingly important,

since both laboratory and plant scale work have demonstrated their effec-

tiveness in extremely low concentrations. This has resulted in studies

by a number of investigators to determine, if possible, the basic mecha-

nisms by which these often spectacular results are obtained.

With the exception of a preliminary report by Ockershausen and

Peterman7 on the use of a cationic polymer for the destabilization of

kaolinite and montmorillonite clay suspensions, the work of Cohen et a.8

on the use of cationic, anionic, and nonionic polymers as coagulant aids

in removing turbidity from water, and the unpublished results of Kim9

dealing mainly with the destabilization of kaolinite clay suspensions

with various anionic polymers used in conjunction with alum, all previous

investigations were performed at rather high concentrations of various

types of suspended materials (i.e.,>400 mg/1). Consequently, very little

information is available on the physical-chemical interactions of poly-

electrolytes with clays in concentrations commonly encountered in

natural waters.

The present investigation is an attempt to elucidate the mechanisms

by which cationic, anionic, and nonionic polymers coagulate and/or floc-

culate dilute clay suspensions under controlled solution conditions.











II. THEORY


Clays


Clays are hydrous aluminum or magnesium silicates which are found

in nature combined with varying amounts of organic and inorganic impuri-

ties. The pure clay mineral is composed of two-dimensional arrangements

of silicon-oxygen tetrahedra and aluminum or magnesium-oxygen-hydroxyl

octahedra. The arrangement of these silicon and aluminum or magnesium

layers serves as one criterion for the classification of clay minerals.

Structural Classification

Clays can be classified on a structural basis as either two- or

three-layer minerals. For example, kaolinite is a two-layer clay con-

sisting of a tetrahedral silicon sheet and an octahedral aluminum sheet

which share common oxygen atoms. Montmorillonite, on the other hand, is

classified as a three-layer clay and consists of an octahedral aluminum

or magnesium sheet which shares oxygen atoms with two tetrahedral silicon

sheets. The combination of an aluminum sheet and one or two silica sheets

is termed a unit layer. Unit layers are stacked in a parallel arrange-

ment, and the distance between corresponding planes in adjacent unit

layers is called the "c-spacing" or "basal spacing." The arrangement of

atoms within each unit layer repeats itself in the lateral direction

forming a unit cell (see Figure 1). The differentiation between two-

and three-layer clays is relatively easy since their "c-spacings" are in

the order of 7.2 A and 9.2 A respectively.


-3-







-4-


C


Unit cell dimensions: A, B, C
Basal spacing = C
Unit layer: X
Cleavage plane: Y


------ Aluminum or Magnesium Sheet
Silicon Sheet


Fig. 1 Three-layer Clay Unit Cell.


Y Y







- 5 -


However, these distances will vary somewhat with the size of the

exchangeable cations located on each side of each unit layer.

Chemical Classification

Clays can also be classified according to the chemical composition

of the unit cell. For example, unsubstituted montmorillonite has a unit

cell formula of (OH)4Si8A14020 nH20 while kaolinite has a unit cell

formula of (OH)8Si4A14010. Diagrammatic sketches of the kaolinite and

montmorillonite unit cell structure and composition are shown in Figures

2 and 3.

Size of Clay Particles

The size of kaolinite and montmorillonite clay particles depends

upon the strength of the bonds joining unit layers since cleavage usually

occurs along unit layer planes. In the case of kaolinite clays, the unit

layers are hydrogen bonded between the oxygen and hydroxyl planes in

adjacent unit layers resulting in a relatively strong bond that is diffi-

cult to break. Consequently, kaolinite clays resist fairly well the

mechanical forces of nature and are found in rather large size fractions

(i.e.,>30) although smaller fractions do exist. In contrast, the unit

layers of montmorillonite are joined by rather weak bonds between adja-

cent oxygen planes which results in the formation of an excellent

cleavage plane. Consequently, montmorillonite clays occur in rather

small particle sizes in nature. Both types of clay particles have a

plate-like shape resulting from cleavage between unit layers. This

cleavage results in the breaking of chemical bonds which creates residual

valence forces on the clay particle surface. If the broken bonds emanate

from an oxygen atom, a cation from the surrounding medium is bound to the






-6-


N I.


/I


0 0 oN



1I00
N/ I / \


/ / t \ / 1 \/


N- N N


Tetrahedral
Sheet


,I
/


\k Octahedral
S Sheet


Legend


O Oxygen
O Hydroxyl
O Silicon
A Aluminum

Unit cell dimensions:

A = 5.15 A
B = 8.9A
C = 7.2A

Unit cell formula: (Al2(OH) (Si^05 )2

Unit cell weight: 516



Fig. 2 Atom Arrangement in a Kaolinite Unit Cell. (From van Olphen )






-7-


Tetrahedral
SSheet





'' Octahedral
S.00 1 i .. Sheet




Tetrahedral
Sheet




Legend

O Oxygen
O Hydroxyl
0 Silicon
/ Aluminum

.;it cell dimensions:

A = 5.15A
B = 8.9A
C = 9.2A

Unit cell formula: (Al2 (OH)2(Si205)2) 2 2' nH20

Unit cell weight: 720. Hydroxyl water: 5%.


Fig. 3 Atom Arrangement in a Montmorillonite Unit Cell (From van Olphen10)







-8-


lattice to satisfy the bond requirement. Similarly, the clay particle

may occasionally be fractured perpendicular to the unit layers thus

breaking bonds emanating from silicon and aluminum atoms. In this case

anions (e.g., OH") are bound to the fractured lattice.

Origin of Surface Charge

According to van Olphen10 a clay particle may obtain its surface

charge by two different methods. One is termed isomorphous substitution

which is the replacement of high valence cations in the clay lattice

structure, namely Si and A1 by lower valence cations. For example,

in the tetrahedral sheet tetravalent silicon is sometimes replaced by

trivalent aluminum while in the octahedral sheet, trivalent aluminum

may be replaced by g2 Zn2, Fe2, Na2, Li and others. This results

in a deficiency of positive charge in the clay lattice and the subsequent

adsorption on the clay surface of cations which are too large to be

accommodated within the lattice structure. In the case of montmorillonite

clays, adsorption of cations occurs on all unit layer surfaces while

kaolinite clays adsorb cations only on the exterior surfaces of the clay

particles. This phenomenon can be explained by the expansion of the mont-

morillonite lattice upon contact with water which allows cations to pene-

trate the particle and adsorb on interlayer surfaces. However, kaolinite

has a non-expandable lattice and consequently a smaller "c-spacing" which

physically prevents the accommodation of cations in the interior of the

clay particle. Therefore, all cations must be adsorbed on the exterior

surface of the clay particle.

The second method by which a surface charge may be obtained is the

preferential adsorption of certain ions from the solution phase at







- 9 -


specific adsorption sites on the clay surface resulting from residual

valence forces due to the exposure of inner lattice atoms at fractured

planes or from lattice atom bonds broken at the plane of cleavage, as has

been mentioned previously, This is sometimes referred to as the adsorp-

tion of "potential-determining ions."

The cation exchange capacity of a clay is its ability to exchange

cations with the solution phase. Originally it was thought that the

cation exchange capacity was due to the formation of anionic sites on the

ruptured platelet edges and cleavage surfaces of the clay particle.1

According to current opinion the cation exchange capacity results from

the adsorption of exchangeable cations on the clay surfaces because of

isomorphous substitution of lower valence cations for aluminum or silicon

in the crystal lattice. Consequently, montmorillonite clays have a much

higher cation exchange capacity than kaolinite clays because of the

larger degree of isomorphous substitution in the montmorillonite lattice.

Double Layer Structure

The net positive charge deficiency within the clay lattice, caused

by the isomorphous substitution of lower valence cations in the clay

lattice, is responsible for the attraction of charge-compensating cations

to the clay surface. These cations are distributed as a diffuse ion

cloud, or double layer, surrounding the clay particle. The mathematical

distribution of these ions in the double layer is given by the Boltzmann

distribution function and the Poisson equation which must be solved simul-

taneously and also satisfy the boundary conditions of the system. An

exact solution of the Boltzmann-Poisson equations is available13 for the

case of the infinite plane.

The thickness of the double layer depends upon the concentration







- 10 -


and valence of the counterions. That is, the thickness of the double

layer decreases as both the valence and concentration of the counterions

increase. According to Mysels,13 if the ratio of the particle radius

to the double layer thickness is ten or larger, the difference in curva-

ture between the inner and outer portions of the double layer will be

small, and an exact solution of the Boltzmann-Poisson equations for the

case of an infinite plane is justified.

Thus far, consideration has been given only to the double layer

associated with the flat negative surface of the clay platelet. However,

the edges of the clay platelet expose an entirely different atomic struc-

ture to the solvent phase and consequently have an entirely different

double layer. At the edges of the clay plate, silicon and aluminum bonds

are broken and an electric double layer is formed by the adsorption of
10 12
potential-determining ions. According to van Olphen and Michaels,

the edges of the clay plate possess a positive double layer in acid solu-

tion with aluminum ions acting as the potential-determining ions while

in neutral or alkaline solution the edges possess a negative double layer

with hydroxyl ions acting as potential-determining ions. This was demon-
12
strated by Thiessen, as quoted by Michaels,2 who showed by electron

micrographs that a negatively charged gold colloid is deposited on

kaolinite platelet edges while a positively charged gold colloid is

deposited on the platelet faces. Since the net electrophoretic potential

of the clay particles is negative, one may assume that the double layer

associated with the clay particle faces predominates. However, the exis-

tence of two oppositely charged double layers is very important in sta-

bility considerations of clay suspensions as will be shown later.







- 11 -


Although several models of double layer structure have been pro-

posed,13 the Stern model has been chosen to represent the double layer

of clay particles because it includes a layer of adsorbed counterions in

contact with the clay surface as well as a diffuse Gouy-Chapman layer

extending into the bulk of the solution, which more nearly represent the

specifically adsorbed and diffuse portions of the counterion configura-

tion about a clay surface (see Figure 4). The potential at the surface

of the particle is designated )J/ while the potential at the boundary

between the Stern and Gouy portions of the double layer is designated .

The only measurable potential in the double layer is the zeta,IJ, poten-

tial which is the potential at the plane of shear which separates the

water bound to the clay particle from the bulk of the solution. This is

the only potential wV-ich can be obtained from electrophoretic measure-

ments.

Stability and Instability of Clay Suspensions

The apparent stability of a colloidal clay suspension is attributed

to two phenomena.. The primary factor is the force of repulsion between

the double layers of two similarly charged clay surfaces. The other

factor which contributes to the stability of clay particles is the degree

of hydration of the clay surface which physically hinders the distance of

closest approach between particle surfaces. This factor is only important

when considering extremely short range particle interactions because the

adsorbed layer of water is only a few molecules thick.

The instability of a colloidal clay suspension is characterized by

the agglomeration of clay particles into masses which are large enough

to settle from the suspension. In natural clay suspensions this can occur






-12 -


Bound
Solvent






I G
@1 @'.i c


e


I~' G e




Stern Layer -Gouy-Chapman Layer


-Plane of Shear
I


Distance


Fig. 4 Stern Model of the Electrical Double Layer. (From Mysels13)







- 13 -


by "mutual coagulation" of the clay particles through edge face associa-

tion of the primary particles because the edges and faces of clay par-

ticles, under certain conditions, possess oppositely charged double

layers. Some energy must be available to promote interparticle collisions

so that edge face association may occur, and for small particles natural

Brownian motion is sufficient for this purpose. The other forces of

attraction responsible for the destabilization of clay suspensions are

known as van der Waals' forces which are relatively short range forces.

Although these forces are not very large for atom pairs, van Olphen10

points out that they become of considerable importance when the attrac-

tion between particles is concerned because the van der Waals, attraction

between atom pairs is additive. Therefore, the total attraction between

particles composed of large numbers of atoms is quite large and relatively

independent of electrolyte concentration.


Polyions

A polymer molecule is defined as a series of repeating chemical
14
units held together by covalent bonds.4 If the repeating units are of

the same molecular species,the compound is called a polymer. However,

if the molecule is formed from more than one molecular species, it is

termed a copolymer. The individual repeating units which make up the

molecule are called monomer units, and the molecular weight of the polymer

molecule is therefore the sum of the molecular weights of the individual

monomer units. The total number of monomer units is referred to as the

degree of polymerization of the polymer molecule.

A special class of polymers which distinguish themselves from







- 14 -


ordinary polymer molecules by possessing ionizable functional groups

along the polymer chain are called polyelectrolytes or polyions. When

these groups dissociate the polymer molecules become charged either

positively or negatively, depending upon the specific functional groups

present. Polyelectrolytes which possess both positively and negatively

charged sites are called polyampholytes.

Polyelectrolytes may be formed by synthetic polymerization re-

actions in the laboratory, or they may be synthesized by living cells in

biological processes. Polymeric molecules such as starch, cellulose,

proteins, nucleic, pectic, and alginic acids are examples of polyelectro-

lytes synthesized by living cells. The synthetic polymerization of

polymers will not be discussed, and the reader is referred to specialized

texts on this subject.

Polymer Configuration

Polymer molecules may be either linear or branched in structure.

A linear chain consists of a series of in-line monomer units forming one

continuous chain, whereas a branched chain is characterized by side

chains emanating from the primary chain. The configuration of the poly-

mer molecule in solution depends partially on the architecture of the

molecule but mainly upon the balance between the electrostatic repulsion

between individually charged chain segments and the natural Brownian

motion of each segment. Consequently, at a given temperature the con-

figuration of the polymer molecule will depend upon the degree of ioniza-

tion of the polymer chain. Increasing the degree of chain ionization

increases the extensibility of the chain. This is also reflected in an

increase in the viscosity of the polymer solution as the chain becomes







- 15 -


more extended.

The addition of counterions to the polymer solution decreases the

extensibility of the chain and consequently the viscosity of the solu-

tion, and the polymer molecule assumes more of a random coil configura-

tion. Presumably the double layers of the charged chain segments are

compressed by the addition of counterions which reduces the mutual re-

pulsion between similarly charged segments, thus allowing the chain seg-

ments to assume a less extended configuration.

For ideal non-polar molecules, the only factor which determines

the shape of the polymer molecule is the Brownian motion of the indi-

vidual chain segments. Consequently the orientation of each segment

with respect to the preceding one is random while the length of the

individual segments is fixed. Therefore, assuming that the bonds be-

tween segments are completely flexible, the average distance between the

ends of the polymer molecule can be calculated on the basis of a random

walk model. This average distance can be calculated from the following

equation:

2= n12

where D = the average end to end distance of the polymer chain

n= the number of segments in the polymer chain

12 = the length of the individual chain segments


The above equation is applicable only for the ideal case of a

completely random coil, and nonidealities such as chain ionization,

inflexible joints, and counterion effects will all modify the end-to-

end polymer chain distance.






- 16 -


Ionic Charge and Charge Density

The sign of the charged sites on the polymer chain, if any, will

depend upon the nature of the functional groups on the polymer chain,

whereas the charge density or number of charged sites per polymer mole-

cule will depend not only upon the number of functional groups but also

upon the degree of ionization of these functional groups. This in turn

will depend upon solution conditions such as pH, ionic strength, tempera-

ture, etc. Figure 5 shows the chain structures for selected cationic,

anionic, nonionic, and polyfunctional polymers.


Polymer Adsorption Models

The extended segment theory of polymer adsorption was first pro-

posed by Jenckel and Rumbach15 to explain the adsorption of more polymer

onto a surface than could be accounted for by simple monolayer coverage.
16
Koral et al. studied the adsorption of polyvinyl acetate and

its monomer ethyl acetate on iron powder, tin powder, and activated

alumina from non-polar solvents and discovered that polymer adsorption

was twenty to forty times greater than the adsorption of monomer. From

these adsorption data, the shapes of the adsorption curves, and the

liquid-air area of a monomer unit, the authors concluded that the poly-

mer molecule must be attached to the adsorbent surface by from 1 to 10

percent of the possible bonding groups on the polymer molecule, with the

remainder of the polymer chain extending into the solution phase.

Gottlieb17 made a similar study of the adsorption of polyvinyl

acetate from non-polar solvents onto metal electrodes. He measured the

adsorption of polymer onto an electrode and the corresponding changes in

surface potential with increasing polymer adsorption. In all instances






- 17 -


I 0
I II I
O -z

U--U-Z

I
u


o u-z

0
I
U&z












I I
u

I+
U


I
u

I
U1


0"
i-
H 0
0
o

0 'ri
ag
0



















H
0





H H
o
0
H









0
a


*P *
LQ,-a
0s


(Y)
O
0

I I
u-OC-u
IcN

U O
I I
U-0-U

u
I










I(



I
I




Ic'j


ule
I
0

1


0)

-a H

. 0
0 o
at
<2


-i

H


0
0 0
cs

a)







- 18 -


it was shown that the surface potential of the electrode at first in-

creased linearly with polymer adsorption and then reached a plateau

where polymer adsorption was independent of surface potential. However,

maximum surface potential was attained long before the electrode became

completely saturated with polymer. From a similar study of the adsorp-

tion of octadecyl acetate, it was observed that the point at which the

adsorption of polyvinyl acetate became independent of the surface poten-

tial of the electrode corresponded to the maximum amount of octadecyl

acetate which could be adsorbed. From this it was concluded that at low

surface coverage all of the acetate groups of the polymer are in contact

with the adsorbent surface. However, when polymer adsorption becomes

independent of surface potential, the polymer molecules are adsorbed by

only a few segments with the remainder of the chain extending outward

from the adsorbent surface. The author also noted that the maximum

amount of polyvinyl acetate which could be adsorbed depended upon the

particular solvent employed. That is, maximum adsorption occurred from

the solvent in which the polymer was least soluble while the coverage

at which the surface potential became independent of electrode coverage

was relatively independent of the solvent system employed. This indi-

cates that at high surface coverage the adsorption of polymer molecules

depends upon their configuration in the solution phase.

From the shapes of the adsorption curves of polyvinyl acetate
18
onto chrome plates from benzene solution, Peterson and Kwei8 concluded

that for dilute polymer solutions adsorption was initially two dimen-

sional but changed to a three dimensional array with increasing time.

However, it was discovered that the polymer configuration was initially







- 19 -


three dimensional when adsorption occurred from more concentrated solu-

tions. The authors also indicated that polymer adsorption from dilute

solution was highly dependent on the rate of solution agitation and is

probably diffusion controlled.

Fontana and Thomas19 determined the fraction of polymer segments

of polyalkylmethacrylate attached to a silica surface by infrared spec-

trometry to be 0.36 at high surface coverage which was also substantia-

ted by film thickness determinations of the adsorbed polymer. The frac-

tion of attached segments was also found to be only slightly dependent

upon surface coverage and completely independent of a three-fold change

in molecular weight. The authors suggested that the adsorbed polymer

configuration is affected more by intramolecular interactions than by

intermolecular interactions since the fraction of segments adsorbed was

relatively independent of surface coverage. According to the equations

of Higuchi,20 which describe the segment density distribution as a func-

tion of a short range interaction parameter, the fraction of adsorbed

segments should have been close to unity for the strong polymer-surface

interactions present in this study. However, the authors explain that

the discrepancy between their results and theoretical predictions may

be due to steric effects caused by polymer backbone inflexibility or to

a lack of uniformity in the spacing of adsorption sites on the adsorbent

surface.

Simha, Frisch, and Eirich have developed a theoretical model for

polymer adsorption onto solid surfaces by considering the statistical

distribution of polymer segments on a solid surface from a polymer mole-

cule which has a random coil configuration. The adsorbed polymer chain






- 20 -


is pictured as being attached to the solid surface at a number of points

along the polymer chain with pendent loops extending into the solution

phase. The assumptions of this theory were summarized by Simha, Frisch,

and Eirich as follows:2

1. The adsorbing surface consists of a number of active sites

which are proportional to the surface area of the adsorbent.

2. The active sites on the polymer chain may each be part of a

single monomer unit, or occur once, on the average, in (r)

monomer segments making up the chain.

3. The area of an active adsorption site is equal to the area

of the statistical segment of the polymer.

4. Once a segment of polymer adheres to a given site, this site

will be considered as saturated and no further adsorption

will occur at that spot.

5. The solvent is not supposed to participate in the adsorption

reaction.

6. The theory considers only monomolecular adsorption from dilute

polymer solutions.

In a supplementary treatment of the same subject, Frisch and Simha22 con-

sidered the adsorption of polymer molecules from concentrated polymer

solutions by including the effect of chain interference near the surface

on the adsorption equilibrium. Both of the preceding treatments apply

only when the polymer-surface interaction energy is low. When large

interaction energies are involved, adsorption conditions may be quite

different, and the polymer chain may be completely deposited on the solid







- 21 -


surface. This has been treated by Frisch23 in a later paper. S.F.E.

theory predicts that for the case of weak polymer-surface interactions

the number of adsorbed segments will be proportional to the square root

of the molecular weight whereas for strong polymer-surface interactions

(i.e., for interaction energies>kT) the number of attached segments

is directly proportional to the molecular weight. The theory also pre-

dicts that adsorption of polymer from a poor solvent will be favored

over adsorption from a good solvent, and for a given molecular weight

the number of attached segments per polymer molecule will be greater

for polymer molecules adsorbed from a poor solvent because of the

smaller radius of gyration of the free coil.

In a recent theoretical treatment of polymer adsorption by

Silberberg24, 25 general adsorption equations were presented for any

strength polymer-surface interaction. It was also stated that the size

of the pendent chain loops does not depend upon the molecular weight of

the polymer but on the flexibility of the polymer chain. That is, a

relatively inflexible polymer molecule will be attached to the surface

by only a few segments and the resulting pendent loops will be longer

than for a polymer with a very flexible chain. Silberberg also states

that the terminal segments of an adsorbed polymer molecule tend to

remain attached to the adsorbing surface. However, this condition is

attained only when the entire polymer molecule comes to equilibrium with

the adsorbing surface, which may take a considerable length of time

depending upon the system involved.

The Simha-Frisch-Eirich (S.F.E.) isotherm for polymer adsorption,







- 22 -


neglecting polymer-polymer interactions on the adsorbent surface,is
--- -= KP
X(1-ee

where b is the number of segments per polymer molecule attached to the

adsorbent surface; e is the fraction of the adsorbent surface covered

by adsorbed segments; P is the activity of the polymer in solution, and

K is a constant. If the number of adsorbed segments is one (i.e.,, =1)

then the S.F.E. adsorption isotherm reduces to the well known Langmuir

isotherm
=bP
(1-e)

where b is the equilibrium constant for the adsorption-desorption re-

action.

Several investigators19' 26, 27 have discovered that polymer

adsorption data for various polymer-solvent-adsorbent systems fit the

Langmuirian model extremely well.

Theoretically, the S.F.E. isotherm initially rises more steeply

than the Langmuir isotherm but approaches its limiting value more slowly

than that of the Langmuir isotherm. Frisch et al.28 showed that data

for the adsorption of polystyrene on carbon fit the S.F.E. isotherm for

the conditionB = 50. However, a plot of the same data according to

the linearized Langmuir equation defines a very good straight line,

thereby implying possible Langmuirian behavior as well. Frisch and

Simha22 have indicated that although true Langmuirian behavior requires

thatA = 1, short chain polymers containing only a few hundred segments

per chain could possibly come very close to exhibiting true Langmuirian







- 23 -


behavior. They have also indicated that polymer adsorption may be

reduced to the Langmuirian type if the adsorbent surface is not covered
19
uniformly by adsorption sites. However, Fontana et al.19 discovered

that a polyalkylmethacrylate-silica system exhibited fairly good Lang-

muirian behavior even whenB was measured to be between 470 and 1660.

The authors attributed this to the insensitivity of the S.F.E. equation

to values of8 except at extremely low concentrations where the accuracy

of the adsorption data may be questionable from an analytical point of

view. Koral et al.16 also observed Langmuirian behavior at high polymer

concentrations but noted some deviation at lower concentrations.

In summary it can be said that the Langmuir adsorption model is

strictly correct only when = 1. However, the S.F.E. adsorption model

approaches the Langmuirian model as the adsorbent surface becomes

covered and the adsorbing molecules are attached to the surface by fewer

and fewer segments.


Destabilization

The destabilization of a clay suspension involves the removal of

the clay particles from the suspending medium. This will ultimately be

accomplished by the natural forces of instability inherent in clay-water

systems, but the process may be accelerated by utilizing coagulation

and/or flocculation reactions.

Recently La Her29 has differentiated between the terms coagulation

and flocculation, the former being responsible for particle destabiliza-

tion by a reduction of the repulsive potential of similar electrical







- 24 -


double layers surrounding two particles and the latter accomplishing

destabilization by a chemical bridging mechanism which enmeshes the

particles in a three-dimensional floc network. This differentiation

between the two processes is necessary since they are two distinct and

different mechanisms which may occur independently or concurrently,

depending upon the systems involved.

Coagulation

The term "coagulation" refers to the effect of counterions in

reducing the repulsion between similarly charged double layers. This

can occur either by the specific adsorption of counterions on the clay

surface with the subsequent reduction in the surface potential of the

particle or by the incorporation of these ions into the diffuse double

layer, which results in the compression of the double layer thickness.

When the double layers of the clay particles are sufficiently

compressed so that the closure between the particles is small enough for

van der Waals' forces of attraction to predominate over the repulsive

forces of the double layers, the particles coalesce and are said to

"coagulate," and the process is referred to as "rapid coagulation."

The rate of this coagulation reaction is dependent upon the number of

interparticle collisions. However, if the net interaction between the

particles still results in a net repulsive force(i.e., an energy barrier)

coagulation will not occur unless a sufficient amount of energy is im-

parted to the system to enable the particles to overcome this energy

barrier. The energy necessary to overcome this barrier may be supplied

by the particles natural Brownian motion or from mechanically induced







- 25 -


agitation of the suspension. Coagulation under these conditions is

termed "slow coagulation." The conditions of "rapid" and "slow coagula-

tion" have been treated by Smoluchowski in his classical paper.30

The degree to which the double layer is compressed depends upon

the concentration and valence of the added counterions as exemplified

by the Schulze-Hardy rule, which states that a bivalent ion is 50-60

times more effective than a monovalent ion, and a trivalent ion is 700-

1000 times more effective than a monovalent ion. Ionic size also

influences the interaction of both counterions and similiions with the

double layer as shown by the Hof.Ieister series, which lists various

cations and anions according to their coagulating power.

Flocculation

The term flocculationn" has not been used consistently by investi-

gators in different scientific fields. In the field of water works

engineering flocculation is considered the process in which pre-coagula-

ted primary particles are physically enmeshed in aluminum hydroxide flocs

formed upon the addition of alum to water under appropriate conditions

of pH and alkalinity. However, soil scientists use the term flocculationn"

to describe the process in which the repulsive potentials of the double

layers surrounding charged particles are reduced by the addition of

counterions. Still another connotation is given by colloid chemists to

the term flocculationn." They consider flocculation to be the aggrega-

tion of primary particles into loose haphazard arrangements having a very

porous structure in contrast to coagulated particles which have an orderly,

dense, compact arrangement.







- 26 -


Polymer Clay Interaction

According to La Mer31 flocculation is solely responsible for the

destabilization of negatively charged colloids by polyanions. However,

the destabilization mechanism for a given colloid-polymer system may be

flocculation, coagulation, or a combined coagulation-flocculation re-

action depending upon the nature and concentration of the colloid, physi-

cal and chemical characteristics of the polyion, chemical composition of

the suspending media, and the physical conditions of solution agitation.

The aggregation of particles into porous structural units with

organic polyelectrolytes has been described by several investigators,
32, 33, 34
and an interparticle bridging mechanism has been proposed2 33 34

Under the bridging theory it is postulated that the polymer molecules

attach themselves to the surface of the suspended particles at one or

more adsorption sites and that part of the chain extends out into the

bulk of the solution. When these extended chain segments make contact

with vacant adsorption sites on other suspended particles, bridges are

formed. The particles are thus bound into small packets which can grow

to a size limited only by the shear gradient imposed by the conditions

of agitation in the system and by the amount of polymer initially ad-

sorbed upon the surfaces of the suspended particles. If too many ad-

sorption sites on the suspended particles are occupied by polymer seg-

ments, bridging will be hindered or even totally inhibited if all sur-

face adsorption sites are occupied. Conversely, if too few adsorption

sites are occupied, the bridging between particles may be too weak to

withstand the shearing forces imposed by the conditions of agitation.














III. HISTORY


In the past two decades, investigations of the interactions of

organic polyelectrolytes with colloidal suspensions have covered a wide

range of polymer-colloid-solvent systems. Therefore, in order to achieve

some degree of continuity in the development of this chapter, those in-

vestigations which deal specifically with clay-polymer-water systems

will be reviewed first. This will be followed by a review of selected

investigations which are concerned with polymer interactions with

aqueous suspensions of materials other than clay.


Clay-Polymer-Water Systems

Ruehrwein and Ward32 were the first investigators to propose, in

1952, a polymer bridging mechanism for the flocculation of highly con-

centrated clay suspensions by anionic and cationic polyelectrolytes.

They postulated that the polymer molecules attach themselves to adjacent

clay particle surfaces thus binding the clay particles into small aggre-

gates. However, the specific interactions of anionic and cationic poly-

mer molecules with a montmorillonite clay were found to be quite differ-

ent. From x-ray diffraction measurements of the "c-spacing" of the

montmorillonite clay upon the addition of cationic and anionic polymer,

it was discovered that the cationic polymer molecules penetrated between

the unit layers of the clay particle causing the clay lattice to expand

while the anionic polymer produced no significant change in the


- 27 -






-28 -


"c-spacing." Consequently, it was concluded that the cationic polymer

molecules were adsorbed via a cation exchange mechanism onto the inter-

layer and exterior surfaces of the montmorillonite clay, whereas the

anionic polymer was probably adsorbed on the positively charged clay

edges.

Packter35 has indicated that polyanions are attached to posi-

tively charged clay edges by electrostatic linkages. However, infrared

studies by Holmes and Toth36 of the adsorption of polyanions onto mont-

morillonite, illite, and kaolinite clays showed the formation of strong

polymer-OH bonds which the authors suggested were due to the reaction of

amide, phenolic, or carboxyl groups on the polyanion with the OH groups

on the clay surface. Additional investigations of the adsorption of

polyanions onto clay surfaces were performed by Michaels and Morelos34

who concluded that adsorption probably occurred by hydrogen bonding

between un-ionized carboxyl or amide groups on the polymer chain and

oxygen atoms on the clay surface. Martin and Aldrich37 proposed a

similar mechanism for the attachment of polysaccharide molecules to soil

particles.

The adsorption of hydrolyzed polyacrylonitrile onto kaolinte clay

has been studied extensively by Mortenson38 39 40 and the adsorption

data were found to fit the Langmuirian model. The adsorption mechanism

was postulated to be either the formation of ionic bonds between the

ionized carboxyl groups on the polymer molecule and the exposed aluminum

atoms on the clay particle edges or the formation of base exchange salts

on the clay surface by the reaction of the carboxyl groups with adsorbed







- 29 -


surface cations. It was also noted that other anions in the solution

compete with polyanion molecules for adsorption sites on the clay sur-

face. Polymer adsorption was found to increase with a decrease in pH or

an increase in the sodium chloride concentration of the suspension. This

was attributed to a reduction of the negative potentials of both polymer

and clay and to a decrease in the size of the polymer coil which permits

closer packing of the polymer molecules on the clay surface.

McLaren41 studied the adsorption of various proteins onto kaolinite

clay and suggested that the polymer molecules were bound to the clay via

a combination of hydrogen bonding and coulombic attraction mechanisms.

In addition, a cation exchange mechanism was postulated for the adsorp-

tion of lysozyme since an equivalence was discovered between the number

of cationic groups on the enzyme and the cation exchange capacity of the
42
kaolinite clay. This was substantiated in a later study when it was

discovered that the adsorption of salt-free dialyzed lysozyme by a

sodium montmorillonite suspension in distilled water resulted in the

liberation of essentially all of the exchangeable sodium ions into the

solvent phase.

The effect of polymer chain length on the destabilization of

slurry concentrations of montmorillonite clay with polyacrylic acid was

studied by Warkentin and Miller.43 An optimum polymer dosage was

exhibited by both long- and short-chain polymers, but the short-chain

polymer showed a narrower optimum dosage range. This was attributed to

more complete adsorption of the short-chain polymer which resulted in

the complete saturation of the adsorption sites on the clay surface and







- 30 -


the redispersion of the clay particles. However, the long-chain polymer

was more effective at interparticle bridging as evidenced by the better

degree of destabilization afforded by this polymer. Increasing the clay

concentration of the suspension enhanced the degree of interparticle

bridging of both polymers presumably because of a decrease in inter-

particle distance. It was noted that the addition of sodium polymeta-

phosphate to the clay suspension prior to the addition of polymer

partially inhibited the destabilization reaction because it was adsorbed

onto the clay edge; thus reducing the number of adsorption sites avail-

,>le to the polymer molecules. Similar interactions with kaolinite clay

edges were proposed by Michaels12 and Dean and Rubins.4

Although the full extension of a polymer molecule aids the poly-

mer bridging mechanism, Michaels45 has shown from controlled hydrolysis

experiments, with 20 percent slurry concentrations of clay, that a poly-

anion should have a charge density which is sufficient to extend the

adsorbed polymer molecule so that interparticle bridging will be favored,

but the charge density should not be so great as to restrict polymer

adsorption because of the repulsive forces between the negatively charged

polymer and clay. Intramolecular association between hydrogen bonding

groups on the polymer chain also affects polymer chain configuration

and consequently interparticle bridging.

The destabilization of natural and synthetic waters of low sus-

pended solids concentration with cationic, anionic, and nonionic poly-

mers used in conjunction with alum was investigated by Cohen et al.

It was discovered that anionic and nonionic polymers had to be used in

conjunction with alum in order to remove the suspended turbidity, whereas











a cationic polymer was effective in reducing the suspended turbidity

when used either alone or in conjunction with alum. Each polymer

exhibited an optimum polymer dosage which produced maximum turbidity

removal, and polymer dosages either smaller or larger than this optimum

dosage resulted in incomplete removal of the suspended material. Tri-

polyphosphates and lignosulfonates interfered with the destabilization

reaction of anionic, nonionic, and cationic polymers. In the case of

the cationic polymer, this interference could be eliminated by increasing

the cationic polymer dosage which presumably completed or precipitated

the negatively charged interfering ions. However, increasing dosages

of anionic and nonionic polymers were ineffective in inactivating these

interfering ions.

Kim9 extended the work of Michaels45 to dilute clay suspensions

(.01 percent) and discovered that hydrolyzed polyacrylamide could not be

adsorbed onto the clay without the aid of alum, whereas nonhydrolyzed

polyacrylamide could be adsorbed either with or without the aid of alum.

From this it was postulated that for hydrolyzed polyacrylamide, an alumi-

num polymer complex was formed which adsorbed on the clay particle sur-

faces and bridged the particles together. Kim defined a "critical" zeta

potential ( 13 my) above which no flocculation was observed to occur.

However, flocculation of the clay particles would not occur if an ex-

cessive amount of polymer was adsorbed on the particle surface even

though the zeta potential of the particles was within this Icriticalu

zeta potential range. The differences between Kim's and Michaels'

results were attributed to the differences in the electrokinetic







- 32 -


properties of the suspensions used by each investigator. The effect of

a three-fold increase in the molecular weight of a cationic polymer on

the removal of the suspended clay turbidity was investigated but no

significant differences could be detected. However, it was noted that

increasing the cationic activity of the polymer reduced the optimum

polymer dosage and improved the degree of turbidity removal.

Ockershausen and Peterman' in a preliminary study of the destabili-

zation of kaolinite and montmorillonite suspensions with a cationic poly-

mer indicated that the optimum destabilization of the clay suspensions

occurred when the zeta potentials of the particles were reduced to be-

tween -31 and -7 mv depending upon the clay and cations present in the

suspension.

Other Colloid-Polymer-Water Systems

The effect of polymer charge density on the adsorption of a co-

polymer of vinyl acetate and crotonic acid was investigated by Schmidt
46
and Eirich. They found a common function relating the adsorptive

capacity of each copolymer of different charge density with an incre-

mental change in pH. This was interpreted as being due to the production

of equal ratios of adsorbing vinyl acetate and nonadsorbing (COO)" groups

with equal changes in pH.

From the adsorption of polyvinyl pyridine onto mercury electrodes

of varying sign and potential, Miller and Grahame4 estimated that even

at high negative potentials only one-fifth of the monomer units in the

polymer chain are actually attached to the mercury surface. The mode of

adsorption was considered to be partially specific and partially







- 33 -


electrostatic in nature because of the initial adsorption of the poly-

cation molecules when the potential of the mercury surface was still

slightly positive and the continued adsorption of polymer at increasing

48
negative mercury electrode potentials. Frisch and Stillinger .have

indicated that although there is a strong electrostatic attraction be-

tween positively charged polymers and negatively charged surfaces there

is also a great tendency for polymer molecules to adsorb on any surface

whose dielectric constant is greater than that of the solution.

As previously noted, the adsorption of excessive amounts of poly-

mer onto a particle surface results in the restabilization of the

colloidal material. Heller and Pugh26 49 in their work with gold sols

attribute this to both steric and electrostatic interactions between

polymer molecules adsorbed on the colloidal particles which inhibit the

formation of polymer bridges.

Posselt and Reidies50 found that the optimum cationic polymer

dosage for manganese dioxide hydrate suspensions increased with increas-

ing pH. This was attributed to an increase in the negative charge of the

manganese colloid with increasing pH. It was also observed that the

addition of divalent calcium ions to the suspension reduced the optimum

cationic polymer dosage and that polymer dosages in excess of the optimum

polymer dosage reversed the charge of the colloidal particles.

Next to clays, the destabilization of aqueous silica suspensions

has probably been studied more extensively than any other colloid system.

Linke and Booth51 have investigated the destabilization reactions between

slurry concentrations of silica and polyacrylamide with regard to







- 34-


adsorption and mixing parameters. Prolonged agitation of a flocculating

suspension was found to break up the flocs into smaller aggregates and

increase the amount of polymer adsorbed by the system. This was attri-

buted to the exposure of additional polymer adsorption sites as the

larger flocs were broken up. For polydisperse particle systems, the

authors noted that the flocculation reaction did not completely remove

the very small particle sizes. Measurements of the weight ratio of poly-

mer/solids showed that the small particles remaining in suspension had

a larger polymer/solids weight ratio than the settled flocs. Consequently,

it was postulated that the smaller particles were probably restabilized

by excessive polymer adsorption. It was discovered that the optimum

polymer/solids ratio was directly proportional to the surface area of

the solid and consequently dependent upon particle size. The effect of

particle molecular weight and ionic strength were to increase polymer

adsorption and decrease the optimum polymer/solid ratio respectively.

The effect of divalent calcium ions on the destabilization of

quartz suspensions with hydrolyzed polyacrylamide was studied by Jones

et al.52 It was discovered that the anionic polymer was ineffective in

destabilizing the colloidal suspension in the absence of a sufficient

concentration of calcium ions. The authors suggested that the function

of the calcium ion was to form a bridge between the polymer molecule

and the particle surface. It was also noted that branching of the poly-

mer chain for both anionic and cationic copolymers was deleterious to

the destabilization reaction presumably because of the lower extensibility

of a branched chain polymer as compared with a straight chain polymer.







- 35 -


Kragh and Langston53 suggested that polymer bridging is responsible

for the destabilization of quartz suspensions with gelatin and indicated

that maximum destabilization occurred when one-third of the saturation

amount of gelatin had been adsorbed. Although polymer adsorption did

not appear to be pH dependent, the destabilization reaction was definitely

found to be a function of the pH of the suspension. Therefore, since

the charge density and consequently the shape of the gelatin polymer

molecule is a function of pH, it was concluded that the configuration of

the polymer in the adsorbed state was the primary factor governing the

destabilization reaction.

La Mer and Smellie54' 55, 56- 57, 58, 59 have investigated the

flocculation, subsidence, and filtration of slurry suspensions of phos-

phate slices and have developed a mathematical theory for the polymer

bridging mechanism. Under this theory it is assumed that polymer adsorp-

tion is Langmuirian and that the rate of the flocculation reaction can

be expressed by a form of the Smoluchowski equation which is modified

to take into consideration the fraction of the particle surface covered

and uncovered by polymer molecules. This theory predicts maximum floccu-

lation when the fraction of the surface covered by polymer molecules is

one-half (i.e., e = 0.5). The adsorption-flocculation reactions have

been summarized by Healy and La Mer60 as follows:

1. Polymer + Solid----Polymer adsorbed until e = 0.5

2; n-unit Flocs- -Macro Floc

3. Macro Floc agitation n-unit Flocs

4. n-unit Flocs + Polymer-----Polymer further adsorbed to e = 1







- 36 -


The effect of time and intensity of agitation on the adsorption-floccula-

tion reaction have been studied by Healy.61, 62 It was observed that

increasing agitation caused a decrease in both the amount of polymer

adsorbed and the average floe size. This was attributed to a mechanical

desorption of polymer molecules from the surface of the particle and to

the mechanical disintegration of the floc particles, both of which were

caused by the shear forces in the fluid. However, this latter reaction

exposes additional surface for polymer adsorption which may or may not

occur, depending upon the strength of the bond between the polymer mole-

cule and the solid surface and the magnitude of the shear forces at this

point. It was also noted that the time of agitation at a fixed mixing

intensity does not alter the amount of polymer adsorbed but does decrease

the degree of flocculation. This was attributed to a redistribution of

the extended polymer molecules on the particle surface so that a large

fraction of the polymer chain segments are adsorbed on the solid surface

resulting in "contraction" and final disintegration of the floc.

Kane et al.63' extended the filtration studies of La Mer,

Smellie, and Healy to amorphous and crystalline silica suspensions which

were flocculated by various types of polymers. The authors found a

linear dependence between the optimum polymer dosage and the solids con-

centration of amorphous silica suspensions, which is not in agreement

with the results of La Mer, Smellie, and Lee5 for phosphate slime sus-

pensions where the optimum polymer dosage was found to vary with the

square of the solids concentration. In addition, the optimum polymer

concentration was found to be independent of the time of agitation which







37 -



is not in agreement with the results of Healy and La Mer62 for calcium

phosphate suspensions. In a later paper dealing with the flocculation

of silica suspensions with a cationic polymer and aluminum perchlorate,

Kane et al. showed that reducing the zeta potential of the suspended

particles to t 8 my resulted in optimum filtration resulting from

optimum flocculation.














IV. PURPOSE AND SCOPE


The purpose of the present study was to evaluate the effective-

ness and mode of action of very dilute aqueous solutions of cationic,

anionic, and nonionic polymers for the destabilization of dilute

colloidal clay suspensions under controlled conditions of pH, ionic

strength, initial colloid concentration, and intensity and duration of

solution agitation. Since anionic and nonionic polymers are ineffective

in destabilizing dilute clay suspensions, as evidenced by the work of

Cohen et a.8 and Kim,9 a simple coagulant such as CaC12 has been used

in conjunction with these polymers at concentrations of 25.0 and 250 mg/1.

The adsorption of polymer on the surface of the clay particles

has been measured and the corresponding changes in the electrophoretic

mobility of the clay particles and the degree of destabilization of the

suspension have been evaluated with regard to polymer dosage. An

attempt has also been made to measure the kinetics of the cationic poly-

mer adsorption reaction. The effect of initial clay concentration on

the destabilization reaction was investigated together with the effect

of time and intensity of solution agitation.

The applicability of the bridging mechanism proposed by Ruehrwein

and Ward32 to dilute clay suspensions has been evaluated, and a theoreti-

cal model for the destabilization reaction is proposed.

Lastly, the effect of molecular weight of a cationic polymer has

been studied with regard to adsorption, destabilization, and electrophoretic

mobility parameters.


-38 -














V. EXPERIMENTAL MATERIALS AND PROCEDURES


Materials


Clays

The clays used in this investigation were kaolinite 4 and mont-

morillonite 23, both obtained from Ward's Natural Science Establishment.

These clay samples were from the same lot as those used in previous

investigations by Black and Hannah and Black and Walters6 in their

studies of turbidity removal with aluminum sulfate and ferric sulfate

respectively. Hannah determined the cation exchange capacity of each

clay according to the procedure described in Official Methods of Analysis
68
of the Association of Official Agricultural Chemists. He reported the

cation exchange capacities of the montmorillonite and kaolinite clays to

be 115.0 and 8.7 milliequivalents/100 grams respectively. In addition,

the surface areas of the montmorillonite and kaolinite clays were cal-

culated from B.E.T. nitrogen gas adsorption data to be 71.0 and 15.8

S2/gm respectively. Additional information regarding the chemical and

crystallographic characteristics of these clays is available in the

publication Reference Clay Minerals.69

Organic Polyelectrolytes

The two cationic polymers used in this investigation, which were

designated Polymer No. 4 and Polymer No. 6, have intrinsic viscosities of


- 39 -









1.88 and 0.645 deciliters/gram respectively. According to the manufac-
turer, Polymer No. 4 has a number-average molecular weight greater than
50(10)3 while Polymer No. 6 has a number-average molecular weight of less
than 50(10)3. Both polymers are linear homopolymers of diallyldimethyl-
ammonium chloride which is labelled with C14 (specific activity 0.30

pc/mg) to permit the determination of microquantities of residual poly-
mer. This polymer consists of a linear chain of recurring N-substituted
piperidinium halide units alternating along the chain with methylene
groups. The ionized form of this cationic polymer is

C2


H2C HC CH--



H2C CH2



N+

H3C CH3 n
The anionic polymer used in this investigation was a 30 per-.
cent hydrolyzed polyacrylamide of specific activity 1.18Pc /mg. The


A product of Peninsular Chemresearch, Inc., Gainesville, Fla.
A product of the Dow Chemical Co., Midland, Michigan.


- 40 -






- 41 -


polymer is linear in structure and has a number-average molecular weight

of between four and six million. The ionized form of the polymer is



r"' ----S
-CH2 CH-- CH--CH-



c=o C=O




NH2 0-

-i-n

The number of carboxyl groups on the polymer chain depends upon

the degree of hydrolysis of the polymer. Under neutral or alkaline solu-

tion conditions the ionization of the carboxyl groups produces a nega-

tively charged polymer molecule. However, under extreme acid conditions

the ionization of the polymer is repressed, and protonation of the amide

groups results in the production of a polymer molecule with cationio

characteristics. The nonionic polymer used in this investigation was

also a C14 labelled polyacrylamide of specific activity 0.35 pc/mg. The

polymer is only 4 percent hydrolyzed thus giving the macromolecule

slight anionic characteristics in neutral or basic solution. This polymer


*A product of the Dow Chemical Co., Midland, Michigan. This polyZ
mer is classified by the manufacturer as a nonionic material and will be
referred to as such in all graphs, tables and text.







- 42-


has a number-average molecular weight of four to six million and is

identical with the previously mentioned polymer except for degree of

chain hydrolysis.

Calcium Chloride

The calcium chloride used in this investigation was reagent grade

material.


Procedures

Preparation of Clay Suspensions

The clays employed in this study were pulverized in a ball mill

for 24 hours and then sieved through a 200 mesh screen after which they

were suspended in deionized water. Forty liters of a 0.30 percent sus-

pension of each clay was prepared in deionized water and vigorously

stirred for 12 hours after which the suspension was allowed to settle

quiescently for 24 hours in order to separate the larger size fractions.

By assuming that the clay particles are spherical in shape and sediment

according to Stokes law, the particles remaining in suspension after a

24 hour settling period should be smaller than two microns in size.

After sedimentation, the supernatant liquid was siphoned off and stored

in polyethylene carboys. Quadruplicate samples of this stock were evap-

orated to dryness and the solids content determined gravimetrically.

Various size aliquots were then pipetted into individual glass containers,

stoppered and stored until needed. When these aliquots are diluted to

two liters with a slightly buffered synthetic water, which has an ionic

strength of 1.15(l0)-3, the clay concentrations shown in Table 1 are

obtained.






-43 -


Table 1

Clay Suspension Concentrations


Type of Clay


Clay Concentration (mg/l)


Kaolinite 4








Montmorillonite 23


73.2

29.8

33.4

33.3
14.9

144

66.8

34.4


The diluted clay suspensions had a pH of approximately seven. The syn-

thetic water was designed to simulate a typical surface water supply and

was prepared by adding the chemical constituents listed in Table 2 to

demineralized water.






- 44-


Table 2

Composition of Synthetic Water


Solution prepared by adding: Concentration of resulting ion

Chemical m mole/liter Ion Meq. ppm

NaHCO3 0.298 Ca++ 0.182 3.6

CaC12 0.091 Mg+ 0.166 2.0

MgSO4 0.083 Na+ 0.298 6.9

HC1 0.251 HCO3" 0.047 2.9
SO 0.166 8.

Cl 0.433 15.

H 1 x 10"7


Preparation of Organic Polyelectrolytes

The cationic polymers used in this investigation were received as

dry powdered materials which contained 96 percent pure polymer. The

samples contained trace amounts of acetone, which remained from the puri-

fication process, and were subjected to high vacuum prior to use. One

hundred milligram aliquots of pure polymer were weighed out and dissolved

in 100 ml portions of deionized water. Serial dilutions were then made

until samples of (10)1, (10 (1) (10)'3 mg/ml were obtained.

The anionic polymer was received as a liquid containing 74 percent

pure polymer or 7.280 mg of pure polymer. This entire quantity was dis-

solved in 250 ml of deionized water to obtain a solution concentration 6f







- 45 -


2.91 (10)-2 mg/ml. Serial dilutions were then made to obtain working

concentrations of 2.91 (10)-3 and 2.91 (10)-4 g/ml.

The nonionic polymer was received in solid form as 100 percent

pure polymer. The entire sample (144.90 mg) was dissolved in 250 ml of

deionized water which gave a solution concentration of 5.796 (10)- mg/mi.

Serial dilutions were then made to obtain solution concentrations of

5.796 (10)"2, 5.796 (10)-3. and 5.796 (lo)4 mg/ml.

All equipment which came in contact with pure polymer solutions

or polymer-clay suspensions was coated with Siliclad to prevent adsorp-

tion of polymer from the solution phase.

Destabilization Experiments

Sample preparation. The aforementioned suspended clay aliquots

were transferred to two liter volumetric flasks and the containers rinsed

several times to assure that all the clay had been transferred. Pre-

determined amounts of NaHCO3, CaCl2, MgSO4, and HC1 were then added

(Table 2) to each sample and the samples diluted to two liters with

deionized water. The suspensions were then mixed on a magnetic stirrer

for 30 minutes after which a one liter sample from each suspension was

transferred to a 48 oz. square jar and subsequently used in the jar test

experiment. The remaining one liter sample was retained for mobility

determinations which were performed later. A series of six samples were
**
run simultaneously on the multiple jar test stirrer for each jar test

experiment.



A product of Clay-Adams, New York, N. Y.

Manufactured by Phipps and Bird, Richmond, Va.







- 46 -


Coagulation-flocculation. While the clay suspensions were being

stirred at 100 rpm, a 50 ml burette was used to add a predetermined

quantity of polymer to all jars. The concentration of the polymer solu-

tion added to the jars depended on the particular polymer dosage employed.

It was found convenient to make polymer additions of 50 ml or less, but

in all cases the polymer was added in as large a volume as possible in

order to attain better accuracy. The suspensions were stirred at 100 rpm

for 20 minutes after which the rate was reduced to 15 rpm for an addi-

tional 20 minutes. For experiments involving the anionic and nonionic

polymers, a simple coagulant (CaC12) was added to the clay suspensions

prior to the addition of the polymers. Predetermined amounts of a stock

CaC12 solution (1 ml = 10 mg) were pipetted into beakers from which they

were transferred to the clay suspensions, while the suspensions were

being stirred at 100 rpm. Five minutes later the anionic or nonionic

polymers were added to the suspensions. Thereafter, the mixing procedure

was the same as mentioned above.

After the prescribed periods of rapid and slow mixing, the paddles

were removed from the jars, and the suspensions were allowed to settle

for 15 minutes. At the end of the sedimentation period 250 ml samples

were withdrawn from each jar by suction at a level 1 1/2 inches below the

surface of the liquid in the jar. These samples were then used for

residual turbidity and residual polymer determinations.

Initial and residual turbidity measurements. Turbidity of the
*
samples was measured with a Lumetron Model 450 Filter Photometer. For



Manufactured by Photovolt Corp., New York, N. Y.







- 47 -


this work, relative turbidity values of sufficient accuracy have been

obtained by using a 75 mm light path and a 650 mp filter. Values so ob-

tained were converted to turbidity units by means of a curve prepared

for each clay on standards referred back to the Jackson Candle Turbidi-

meter in a manner set forth in Standard Methods.70

The degree of destabilization of the clay suspensions is reflected

by a decrease in the residual turbidities of the suspensions after the

coagulation, flocculation, and sedimentation processes have been com-

pleted. Since mutual coagulation can be experienced by montmorillonite

clays and to some extent by kaolinite clays, the initial optical turbidi-

ties of the clay aliquots increased upon aging as the clay particles grew

in size. Consequently, all resultswere expressed as residual optical

turbidities rather than as percent of the initial suspension turbidity.

Measurement of pH. The final pH of the destabilized clay sus-
*
pensions was measured with a Beckman Model G pH Meter.

Electrophoretic Mobility Determinations

Sample preparation. The remaining one liter aliquot of the two

liter sample initially prepared for each clay sample was stirred thor-

oughly on a magnetic stirrer for five minutes prior to the addition of a

predetermined amount of polymer. The clay-polymer suspension was then

mixed for two minutes and the particle mobilities immediately determined.

When CaC12 was used in conjunction with polymer, the CaCl2 was mixed with

the clay suspension for five minutes prior to the addition of polymer.



Manufactured by Beckman Instruments, Inc., Fulleiton, Calif.







- 48 -


Conductance measurement. Specific conductance was determined

according to the procedure described in Standard Methods.70 A Model

1B-2A Impedance Bridge with a pipette-type conductivity cell having a
-1
cell constant of 1 cm- was used for the measurements.

Measurement of pH. The pH of each suspension was measured

immediately before the mobility determinations were made.

Measurement of electrophoretic mobility of floc particles.

Mobility measurements were performed with a horizontal glass Briggs cell
71
according to the procedures developed by Black and Smith.7 Several

improvements in this technique which have recently been described by
72
Black and Smith2 were also incorporated in this study.

Twenty individual particle mobilities, ten in either direction,

were used to obtain an average particle mobility for each sample. The

calculated mobilities are time-averaged rather than velocity-averaged

mobilities.

Residual Polymer Determinations

Liquid Scintillation Spectrometry. This technique is extremely

useful for low level counting of beta as well as gamma emitting radio-

nuclides. Self absorption is virtually eliminated when the sample is

dissolved with scintillators in a suitable solvent, and "4TT" geometry

is maintained as long as the sample and scintillators remain as one

homogenous solution. Liquid scintillation spectrometry allows the use

of relatively large samples with high counting efficiency and sensitivity.



Manufactured by Heath Company, Benton Harbor, Michigan.







- 49 -


The liquid scintillation spectrometer used in this investigation

was a Tri Carb Model 314 EX. A simplified block diagram of the instru-

ment circuitry is shown in Figure 6. The machine functions in the

following manner when counting in the coincidence mode of operation:

1. The two photomultiplier tubes on either side of the sample

vial detect a light pulse simultaneously when it is emitted

by the scintillator solution.

2. The photomultiplier tubes convert the light pulse into an

electrical pulse and amplify the pulse height.

3. The pulse heights are further amplified in the preamplifier

assemblies.

4. One pulse passes out of one of the preamplifiers and into an

electronic coincidence circuit and system logic.

5. The other pulse passes out of the second preamplifier and

into channel I and II amplifiers, through discriminator cir-

cuits, and into the logic.

6. Both pulses must enter the logic system within a very short

time interval in order for one ionization event to be re-

corded in the channel I and II scalers.

The coincidence circuitry of the spectrometer reduces most of the

background counts arising from photomultiplier tube noise since thermal

noise pulses, which occur at random in the two photomultipliers, are not

likely to coincidein time. The photomultiplier tubes are also operated



Manufactured by Packard Instrument Co., La Grange, Ill.







- 50 -


Legend

1. Sample
2. Photomultiplier
3. Preamplifier
4. Channel I Amplifier
5. Channel II Amplifier
6. High Voltage
7. Monitor Amplifier
8. Discriminator B


9.
10.
11.
12.
13.
14.
15.
16.


Discriminator A
Discriminator D
Discriminator C
Discriminator A'
Logic System
Coincidence System
Channel I Scaler
Channel II Scaler


Fig. 6 Block Diagram of Tri Carb Model 314 EX Liquid Scintillation
Spectrometer. (From Packard Instrument Manual)







- 51 -


at low temperatures to help reduce thermal electron emission by the

tubes.

Discriminator circuits are also employed to pass pulses of pre-

determined amplitudes which helps to reduce the background count. For

example, if discriminator A is set for a 10 volt pulse height and dis-

criminator B is set for a 100 volt pulse height, only input pulses be-

tween 10 and 100 volts will be passed through the discriminator circuits

to the system logic. Therefore, the lower discriminators on each chan-

nel of the instrument, A and C, should be set to reject the small pulses

which originate from tube noise while the upper discriminators, B and D,

should be set to reject pulse heights larger than those which occur from

normal beta energies (i.e., high energy background radiation). Discrim-

inator A' in the coincidence circuit should also be set to reject small

pulses which arise from tube noise. A more rigorous treatment of this
73 74
subject is given by Davidson and Feigelson and Bell and Hayes.

Scintillator system. A scintillator system consists of a solvent,

primary scintillator, and sometimes a secondary scintillator. The sol-

vent must have the ability to accept energy from ionizing radiation and

transfer it to a dissolved scintillator molecule. Organic scintillators

and solvents have been found to be very suitable for this purpose. How-

ever, they cannot be used for aqueous samples unless some modifications

are made to the scintillator system. Consequently, many scintillator-

solvent systems have been tailored for individual situations. One impor-

tant consideration which must be taken into account when counting aqueous

samples at the low temperatures required for proper photomultiplier tube







- 52 -


operation, is that the water must not "freeze out" of the scintillator

solution. Partial freezing of the sample would alter the "4TW" geometry

of the system which would result in a decrease in the over-all counting

efficiency of the sample. Since aqueous samples are very severe quench-

ing agents, this also adds to the difficulties in counting these types

of samples.

The scintillator system chosen for the present work was developed

by Bruno and Christian.75 The composition of the scintillator system

is given in Table 3.




Table 3

Liquid Scintillation Counting System



Chemical Quantity


p-dioxane 1000 ml

Ethylene glycol monoethyl ether 200 ml

PPO 12.2152 g

POPOP 0.6108 g

Napthalene 61.08 g


2,5-Diphenyloxazole (Scintillation Grade). A product of Packard
Instrument Co., La Grange, Ill.

**l,4-bis-2-(5-Phenyloxazolyl)-Benzene (Scintillation Grade). A
product of Packard Instrument Co., La Grange, Ill.







- 53 -


This system is capable of holding 25 percent aqueous samples at 0C with

a C14 counting efficiency of 55 percent. Only high purity reagents were

used since any impurities could result in undeterminable amounts of

quenching.

If this scintillator solution is exposed to air, the dioxane and

ethylene glycol monoethyl ether form peroxides which are known to be

quenching agents. Therefore, care was taken to carefully stopper the

scintillator solution after use, and a fresh solution was prepared

monthly. No reduction in the counting efficiency of the scintillator

system was observed over a one month period.

Standard curve calibration and sensitivity and accuracy of the

determinations. The preparation of serial dilutions of the organic poly-

electrolytes has been described previously. For each polymer a series of

standard samples were prepared from the aforementioned dilutions and

counted for 100 minutes in the Tri Carb liquid scintillation spectrometer.

The standards ranged in concentration from 10 lg/1 to 6000 ig/1. A

linear relationship between polymer concentration and net cpm was ob-

tained for all three polymers over the concentration range indicated

above. In all cases the accuracy of the determinations was found to be

3 percent. However, the sensitivity of the method depends upon the

specific activities of the individual polymers, since natural background

radiation places a lower limit on the smallest concentration of polymer

which can be detected. In order to reduce the total background counts

as much as possible, polyethylene counting vials were used instead of

glass vials which contain radioactive potassium. Polyethylene counting







- 54 -


vials gave a constant background count of 26 cpm whereas the glass vials

gave counts of between 44 and 90 cpm, depending upon the percent of

radioactive potassium in the glass. Consequently, the sensitivity of

the method for the determination of residual cationic, nonionic, and

anionic polymer was found to be 3.0 ig/l, 3.0 Ag/l, and 1 g/l respec-

tively.

The Tri Carb Liquid Scintillation Spectrometer was found to be

extremely stable over periods as long as six months. The slope of the

standard calibration curve always remained within the t 3 percent

accuracy of the method. However, a new calibration curve was always

prepared when repairs were made to the spectrometer or when the experi-

mental work on a single polymer exceeded two months.

Sample preparation and counting procedures. After the coagula-

tion, flocculation, and sedimentation processes were completed, a 250 ml

sample of the supernatant liquid was removed from each jar by suction

at a level 1 1/2 inches below the surface of the liquid. This sample

was analyzed for residual polymer in the following manner:

1. A 50 ml sample of the destabilized and settled suspension

was centrifuged for 20 minutes at 3700 x g.

2. A 5.0 ml sample of the centrifugate was pipetted into 15 ml

of scintillator solution in a polyethylene counting vial.

3. The sample was then counted for 100 minutes at a temperature

of 00C, a voltage of 1250 volts, and predetermined discrimina-

tor and coincidence circuit settings.

4. The net cpm for the sample were determined by subtracting the







- 55 -


average background activity, as determined from blank samples

counted at the beginning and end of the counting period, from

the gross cpm of the sample containing the radioactive polymer.

5. The concentration of the polymer in the sample was then read

from the appropriate calibration curve of net cpm vs polymer

concentration.

Each jar test consisted of six individual experiments so a series

of six samples and two blanks were counted immediately following the end

of the jar test experiment. Therefore, the total counting time per jar

test was approximately 14 hours, Daring this period of time no signifi-

cant adsorption of polymer on the walls of the counting vial could be

detected. However, it was found that significant adsorption did occur

after 36 hours. Consequently, the time between sample preparation and

the end of the counting period was never permitted to exceed 14 hours.

Polymer-Clay Kinetic Experiments

The kinetic experiments consisted of measuring the rate of polymer

adsorption from the solution, the rate of change in the electrophoretic

mobility of the clay particles upon the addition of polymer to the clay

suspension, and the effect of initial clay concentration and rate and

time of mixing on the destabilization of the clay suspensions. All of

the above mentioned experiments were performed using a suspension of the

kaolinite clay and cationic polymer No. 6.

The rate of polymer adsorption from the solution was determined by

measuring the amount of polymer remaining in the solution after pre-

determined times of mixing. The amount of polymer adsorbed by the clay







- 56 -


is considered to be the difference between the polymer dosage and the

residual polymer concentration in the centrifugate of the polymer-clay

sample. Similarly, the rate of change in the electrophoretic mobility

of the clay particles was measured at predetermined time intervals after

the addition of polymer and the initiation of mixing.

The effect of initial clay concentration and rate and time of

mixing on the destabilization of the clay suspensions were evaluated by

mixing the suspensions at a given intensity of agitation for a pre-

determined period of time, following which the suspensions were allowed

to settle for 15 minutes after which samples for residual turbidity

determinations were withdrawn.













VI. DISCUSSION OF RESULTS AND CONCLUSIONS


Cationic Polymer-Clay Interactions

Figures 7 to 14 show the results obtained for the destabilization

of kaolinite and montmorillonite clay suspensions with cationic polymer

No. 4. In these figures electrophoretic mobility, residual turbidity,

and polymer adsorption are plotted against polymer dosage for each sus-

pension. Optimum polymer dosage is defined as that dosage which produces

maximum destabilization of the clay suspension (i.e., minimum residual

turbidity).

The kaolinite clay-cationic polymer No. 4 interactions depicted

in Figures 7, 8, 9, and 10 show that for each clay concentration the

amount of polymer adsorbed increased linearly at first and then approached

a limiting value which was dependent upon the initial clay concentration.

As the polymer dosage was increased the residual turbidity of each sus-

pension decreased and reached a minimum value which was dependent upon

the initial clay concentration of the suspension. However, as the

initial clay concentration was increased, the zone of successful clay

destabilization became broader, and the maximum degree of destabilization

was increased (Figure 10). At polymer dosages in excess of the optimum

polymer dosage the suspensions were gradually restabilized,and at

extremely high polymer dosages restabilization became complete. For each

clay concentration, the electrophoretic mobility of the clay particles

was reversed and approached a limiting positive value which appeared to


- 57 -






- 58 -


+2
0 0
S+1

0
g0 /-




0 2 4 6 8 10
Polymer Dosage (pg/1)(10)2
120

100 .




20
80


A-1 40

0 20
0 I I I I
0 2 4 6 8 10
Polymer Dosage (Ig/l)(10)2
4

00-



0 0 I

0 2 4 6 8 10
Polymer Dosage (Clg/1)(10)2

Fig. 7 The Destabilization of a Kaolinite Clay Suspension with
Cationic Polymer No. 4. Initial Clay Concentration 14.9 mg/l.








- 59 -


o 1 2 3 4 5 6 7 8
Polymer Dosage (pg/1)(10)2


o 1 2 3 4
Polymer


9 10 11


5 6 7 8 9 10 11

Dosage (pg/) (10i)2


oo
2',
00


$4--
0 ho


0 1 2 3 4 5 6 7 8 9 10 11
Polymer Dosage (pg/1)(10)2


The Destabilization of a
Cationic Polymer No. 4.


Kaolinite Clay Suspension with
Initial Clay Concentration 29.8 mg/l.


Y-,-'*






t -I----- I I


04
o
*$



P 0 V
o v
11H
Mt


120

100


$4
E-4


4


Fig. 8 -







- 60 -


0 1 2 3 4 5 6 7 8
Polymer Dosage (pg/1)(10)2


0 1


2 3 4 5 6 7 8 9
Polymer Dosage (pg/1)(10)2


9 10 11


10


Polymer Dosage (pg/l)(10)2


Fig. 9 The Destabilization of a
Cationic Polymer No. 4.


Kaolinite Clay Suspension with
Initial Clay Concentration 73.2 mg/l.


o



~o
0
H h


120

100

80

60

40

20


cl
U)
0


1o

00
H

0 bD
H

sE


o











- 61 -


+2

+1

0


-1


0 1 2 3 4 5 6 7 8
Polymer Dosage (Ig/1)(10)2


9 10 11


3 4 5 6 7 8
Polymer Dosage (rg/1)(10)2


Fig. 10 -


The Effect of Initial Clay Concentration on the Destabilization
of Kaolinite Clay Suspensions with Cationic Polymer No. 4.
Initial Clay Concentrationst ---- 14.9 g/1;
29.8 mg/1; 73.2 ag/1.


.01 do-






0 1 2 3 4 5 6 7 8 9 10 11
Polymer Dosage (pg/l) (10)2


120

100

80

60

40

20


t4
NI

*r -
pa


0

ri






- 62 -


be relatively independent of the initial clay concentration (Figure 10).

It was also noted that the amount of polymer required to reverse the sign

of the mobility (Figure 10) was dependent upon the initial clay concen-

tration of the suspension. Since the amount of polymer adsorbed by the

clay is dependent upon the number of available adsorption sites in the

system, one would expect smaller concentrations of clay to require less

polymer to reverse the mobility of the particles. Similarly, the total

number of adsorption sites in a system will govern the amount of polymer

adsorbed before the system is restabilized. However, the effect of

initial clay concentration on the maximum degree of turbidity removal

is purely a kinetic phenomenon as increasing the clay concentration

increases the probability of interparticle collisions which leads to

more favorable conditions for interparticle bridging. This will be dis-

cussed later in this chapter in more detail. For the kaolinite clay

concentrations investigated, optimum destabilization was found to occur

when the mobility of the particles was reduced to between 0.0 and -0.7

P/sec/v/cm depending on the initial clay concentration of the suspension
(Table 4).

Figures U1, 12, 13, and 14 show the results obtained for the

destabilization of suspensions of a montmorillonite clay with cationic

polymer No. 4. Although these polymer-clay interactions are somewhat

similar to those presented for kaolinite clay, there are several signifi-

cant points of difference. First, the mobility data show that maximum

destabilization of the montaorillonite suspensions occurred when the

mobility values of the clay particles ranged from -0.3 to 40.6 p/sec/v/cm






- 63 -


Table 4

Values of Electrophoretic Mobility, Residual Turbidity, and
Polymer Adsorption at the Optimum Polymer Dosage for Each
Cationic, Anionic, and Nonionic Polymer-Clay System


Clay Optimum Electro-
Ooncen- Polymer Polymer Residual phoretic
Polymer tration Dosage Adsorbed Turbidity Mobility
(mg/1) (pg/1) (pg/1) (T) (p/sec/v/cm)

Cationic No. 4 K 73.2 143 130 2 -0.7
K 29.8 62 50 3 0.0
K*, 14.9 30 25 13 -0.4
M 144 890 800 0 +0.5
M 66.8 670 625 1 +0.6
M 34.4 147 120 0 -0.3

Cationic No. 6 K 73.2 98 90 5 +0.1
K 29.8 45 45 8 +0.3
K 14.9 30 30 13 -0.1
M 144 555 450 0 +0.6
M 66.8 570 500 0 +0.5
M 34.4 171 155 1 +0.1

Anionic K 33.3 12 11 5 -0.6
CaCa1 (250 mg/1)

Anionic K 33.3- -
CaCl, (25.0 mg/1)

Nonionic K 33.4 240 240 5 -0.4
CaCla (250 mg/1)

Nonionic K 33.4 -
CaCaI (25.0 mg/1)

Kaolinite
ontorillonite
Montmorillonite






- 64 -


o

0 a
%#


Polymer Dosage (pg/1)(10)3


Fig. 11 The Destabilization of a Montmorillonite Clay Suspension
with Cationic Polymer No. 4. Initial Clay Concentration
34.4 mg/1.


+1



-1
-2
S2 3 4
Polymer Dosage (prg/1)(10)3
60

50

40 -

30

20

10
00
0 / I I II
0 1 2 3 4 5
Polymer Dosage (pg/1)(10)3
2






0
I-


0 .t1 it


t
-I
-'-
U)


,..


0
k\






- 65 -


Polymer Dosage (pg/l)(lO)3


0 1 2 3 4 5
Polymer Dosage (pg/l)(10)3


1 2 3 4 5
Polymer Dosage (Ag/1)(10)3


Fig. 12 -


The Destabilization of a Montmorillonite Clay Suspension
with Cationic Polymer No. 4. Initial Clay Concentration
66.8 mg/1.


0
::j -~

0o t

0 A^
M .
1


0


K <_ 000001 1


I


Oh
00


IM.
r!V






-66 -


0







0 1 2 3 4 5 6
Polymer Dosage (pg/1)(10)3


00

r*
0t
rIT


Fig. 13 -


0 1 2 3 4 5 6 7
Polymer Dosage (pg/1)(10)3

The Destabilisation of a Montmorillonite Clay Suspension
with Cationic Polymer No. 4. Initial Clay Concentration
144 g/1.


OlO

0
2y'5'
Q


120

100


2^

-rt
47
1:9'*







- 67 -


0 1 2 3
Polymer Dosage


0 1


4 5 6 7
(ig/l)(10)3


2 3 4 5 6 7
Polymer Dosage (pg/1)(10)3


Fig. 14 -


The Effect of Initial Clay Concentration on the Destabilization
of Montmorillonite Clay Suspensions with Cationic Polymer No. 4.
Initial Clay Concentrations: ----- 34.4 mg/1;
66.8 mg/l; ----- 14 mg/1.


O

EA
0

0 -
g^


4.


120

100

80

60

40

20


0o






- 68 -


depending on the initial clay concentration of the suspension (Table 4).

This behavior differs somewhat from that observed by Black and Hannah,

who found that throughout the pH range of optimum flocculation of a mont-

morillonite clay with alum, mobility values were always slightly negative.

It was also observed that as the initial clay concentration was increased

the zone of successful clay destabilization became broader, but the maxi-

mum degree of destabilization remained unchanged (Figure 14). However,

the montmorillonite clay suspensions exhibited a much broader zone of

successful turbidity removal than the kaolinite clay suspensions. If it

is assumed that polymer adsorption occurs by cation exchange, one would

expect a montmorillonite clay suspension to exhibit a broader zone of

maximum turbidity removal than a suspension containing an equal weight

of kaolinite clay because the montmorillonite clay possesses a much higher

cation exchange capacity. Consequently, more polymer can be adsorbed

before the clay particles become saturated and the system restabilized.

Figures 15 to 22 show the results obtained for the destabilization

of kaolinite and montmorillonite clay suspensions with cationic polymer

No. 6. The kaolinite clay-cationic polymer No. 6 interactions are very

similar to those mentioned previously for cationic polymer No. 4 with the

exception that optimum destabilization was found to occur when the mobil-

ity of the particles was reduced to between -0.1 and +0.3 ir/sec/v/cm

(Table 4). In addition, the montmorillonite clay-cationic polymer No. 6

interactions are also very similar to those presented for cationic polymer

No. 4 with the exception that optimum destabilization occurred when the par-

ticle mobility was reduced to between 40.1 and +0.6 p/sec/v/cm as shown in







- 69 -


+2 ,
00 0
+1

o 1








60
+> on
-= 40 -1
,M 1
-2 I I I I I I I
0 1 2 3 4 5 6 7 8 9 10 11
Polymer Dosage (pg/1)(10)2
12 0 ..... ..... .....------------




















4)
100
+ 80

6-


*I 0-.-
60


r 20 o o o o -

0 i I I 1 I I I
0 1 2 3 4 5 6 7 8 9 10 11
Polymer Dosage (pg/l)(10)2





S
00 3







0 1 2 3 4 5 6 7 8 9 10 11
Polymer Dosage (Ig/l)(10)2

Fig. 15 The Destabilizatlon of a Kaolinite Clay Suspension with
Cationic Polymer No. 6. Initial Clay Concentration 14.9 mg/1.








- 70 -


0 1 2 3 4 5 6 7 8 9 10 11
Polymer Dosage (pg/1)(10)2


0 1 2 3 4
Polymer


5
Dosage


6 7 8
(pg/1)(10)2


9 10 11


00
o o


0 0

Ed.


0 1 2 3 4 5 6


7 8 9 10 11


Polymer Dosage (pg/1)(10)2


Fig. 16 The Destabilization of a
Cationic Polymer No. 6.


Kaolinite Clay Suspension with
Initial Clay Concentration 29.8 mg/l.


o
-H
0
o +J>
*r1 0
+POM
a .
H


120

100

80


60

40


20


*1~I
-o
$4

E*4
f-i.--

4-,
(I)
G)







- 71 -


0 1 2 3 4 5
Polymer Dosage


6 7 8
(pg/1)(10)2


9 10 11


0 1 2 3 4 5 6 7 8 9 10 11
Polymer Dosage (pg/l)(10)2


3 4 5 6 7 8
Polymer Dosage (rg/1)(10)2


The Destabilization of a
Cationic Polymer No. 6.


Kaolinite Clay Suspension with
Initial Clay Concentration 73.2 mg/l.


0

a) >>e
0 ,) 0
Uo

4) 1-
H- *-


0I
7 ^ i-----,I- II-


120

100

80

60


40


20


*ri


E0


10
8,(i
00
VU)r-4
54-
4) b3
e a

rl
Ho
0


Fig. 17 -






- 72 -


120

100

80

60

40

20


0



Hr
M>


0 1 2 3 4 5 6 7 8 9 10 11
Polymer Dosage (pg/l)(10)2

The Effect of Initial Clay Concentration on the Destabilization
of Kaolinite Clay Suspensions with Cationic Polymer No. 6.
Initial Clay Concentrations:-- ----14.9 mg/1;
29.8 mg/; --- -- 73.2 mg/1.


I






1 I I I I I I I I I
0 1 2 3 4 5 6 7 8 9 10 11
Polymer Dosage (Ag/1)(10)2


















Polymer Dosage (pg/1)(10)2
/






---- .......-.....- -




0 1 2 3 4 5 6 7 8 9 10 13
Polymer Dosage ( g/l)(10)2


*-v

*4


'0
0
$.4 e-.
%,M
00


5"
UH
'0'
H
$4 .-


Fig. 18 -







- 73-


0 4
o >

.4,
.-
+0 av


Polymer Dosage (pg/1)(10)'


Fig. 19 The Destabilization of a Montmorillonite Clay Suspension
with Cationio Polymer No. 6. Initial Clay Concentration
34.4 mg/1.


0 1 2 3 4 5
Polymer Dosage (pg/1)(l)3
















0 1 2 3 4 5
Polymer Dosage (pg/1)(10)3










II I I ,


f.


4.4*-
02


00







- 74-


f------


I I I I
) 1 2 3 4 /
Polymer Dosage (pg/1)(10)3


1 2 3 4
Polymer Dosage (tg/1)(10)3


2 3 4
Polymer Dosage (tg/l)(10)


Fig. 20 -


The Destabilization of a Montmorillonite Clay Suspension
with Cationic Polymer No. 6. Initial Clay Concentration
66.8 mg/1.


0
o ^
M


o ^.
141


-*O


00
r-4
Ob

40.






- 75 -


+2

V +1- 0


I4 --



0 1 2 3 4 5 6 7
Polymer Dosage (pg/1)(10)3
.? ?


120


o o
S80

60


40
*r4


SI oI I I I
0 1 2 3 4 5 6 7




44


2
*B




0 tIIII
o 1 2 3 4 5 6 7
Polymer Dosage (pg/1)(10)'

Fig. 21 The Destabilization of a Montmorillonite Clay Suspension
with Cationio Polymer No. 6. Initial Clay Concentration
144 mg/i.






- 76 -


+1*C
+.I


0w
C ^
4V1 .
9i3


2 3 4 5
Polymer Dosage (&g/1)(1O)3


The Effect of Initial Clay Concentration on the Destabilization
of Montmorillonite Clay Suspensions with Cationic Polymer No. 6.
Initial Clay.Concentrations s 34.4 ag/l
66.8 mg/1; -- 144 mg/1.


0 1 2 3 4 5 6 7
Polymer Dosage (jig/1)(10)3















0 1 2 3 4 5 6 7
Polymer Dosage (ig/1)(lo)3


120

100


4)


*0

00

ri

Ij 7


Fig. 22 -







- 77 -


Table 4. However, in spite of the similar shapes of the adsorption,

residual turbidity, and electrophoretic mobility curves for the inter-

actions of cationic polymer No. 4 and cationic polymer No. 6 with the

same clays at corresponding clay concentrations, an examination of the

various experimental parameters at the optimum polymer dosage for each

polymer-clay system does reveal some differences which are due presumably

to the difference in the average molecular weight of the two polymers.

First, cationic polymer No. 6, which has a number average molecular weight

of less than 50 (10 3 consistently exhibits a lower optimum polymer dos-

age for a given clay and initial clay concentration than cationic polymer

No. 4, which has a number average molecular weight greater than 50 (10)3

(Figures 23 and 24). Second, the width of the zone of successful clay

destabilization appears to be wider for the interaction of cationic poly-

mer No. 4 with the kaolinite clay suspensions (Figure 10) than for the

similar interactions of polymer No. 6 (Figure 18). Warkentin and

Miller43 have noted a similar molecular weight effect for montmorillonite

polyacrylic acid systems. They attributed this behavior to the more

complete saturation of the adsorption sites on the clay particle by the

shorter chain polymer than by the longer chain polymer. However, in the

present study no significant difference in the width of the zone of

successful clay destabilization was found for the interaction of either

cationic polymer with the montmorillonite clay suspensions (Figures 14

and 22). This could be due to the mutual coagulation tendencies of the

montmorillonite clay which could have masked any small effect caused by

the difference in polymer molecular weight. However, the maximum degree







- 78 -


I I A I


0 10 20 30
Clay


I I


40 50 60 70 80 90 100
Concentration (mg/1)


Fig. 23 Stoichiometry of the Destabilization of Kaolinite Clay
Suspensions with Cationic Polymers ( 0 Polymer No. 4;
+ Polymer No. 6).


150

140

130

120

110

100


0
w




to
Q0
1$4
m'-


4a


10

0


___~_____







- 79 -


o



o
(0

5


0 20 40 60 80 100 120 140 160
Clay Concentration (mg/1)


Fig. 24 Stoichiometry of the Destabilization of Montmorillonite Clay
Suspensions with Cationic Polymers ( O Polymer No. 4;
+ Polymer No. 6).







- 80 -


of turbidity removal afforded by the two cationic polymers for both the

kaolinite and montmorillonite clays does not appear to be significantly

different.

Kim9 showed that a three-fold difference in the molecular weight

of polyvinyl pyridine had no significant effect on the destabilization

of knolinite clay suspensions. However, by increasing the cationic

activity of the polymer chain from 25 to 99 percent, the molecular

weight of the polymer being held constant, the optimum polymer dosage

was decreased seven-fold, and the degree of suspension destabilization

was increased slightly. According to Priesing, cationic polymer

adsorption is directly related to the number of cationic sites on the

polymer chain. Consequently, for a given concentration of clay, the

adsorption of a cationic polymer of controlled molecular weight should

decrease as the cationic activity of the polymer chain is increased.

The method by which the cationic polymer used in the present study

becomes attached to the clay surface is not definitely known, and

although one is tempted to assume that the forces binding the two are

merely electrostatic in nature, one cannot and should not rule out chem-

ical interactions which may occur between the polymer and clay surface.

Van Olphen10 indicates that when quaternary ammonium salts (R4N Cl") are

added to clay-water suspensions, the organic cation replaces the exchange

cations present on the clay surface. Perhaps the organic N cations in

the diallyldimethylammonium chloride polymer are adsorbed in a similar

manner. The work of Ruehrwein and Ward32 and Priesing6 also suggests

that cation exchange adsorption is the manner by which cationic polymers







- 81 -


are attached to clay surfaces.

The electrophoretic mobility data for the cationic polymer-clay

systems studied show that a substantial decrease in the initial mobility

of the clay particle occurred before maximum destabilization was real-

ized (Table 4). It is well known that the ability of cations to com-

press the double layer surrounding a clay particle depends both on the

charge and concentration of the particular cation. Consequently, it

would appear that a highly charged cationic polymer molecule should

have a pronounced effect on reducing the thickness of the double layer

surrounding a clay particle or in reducing the surface potential of the

clay particle depending on whether the polymer molecule is held in the

double layer or specifically adsorbed on the surface of the clay

particle. Either type of interaction will result in a reduction in the

closure between adjacent clay particles which favors interparticle bridg-

ing by extended segments of the adsorbed polymer molecules. However,

when the surface of the clay particle becomes covered with an excessive

number of polymer molecules interparticle bridging is reduced, or even

inhibited in the case of complete surface coverage. This may be due to

a decrease in the number of available bridging site as suggested by

La Mer,31 or to steric hindrance and/or electrostatic interactions be-

tween adsorbed polymer molecules, as suggested by Heller and Pugh.26

It has been shown in this investigation that for a given cationic

polymer-clay system a certain weight relationship must exist between

polymer and clay in order for optimum destabilization of the clay sus-

pension to occur. For polymer-clay weight ratios either greater than or







- 82 -


smaller than the optimum value, the destabilization of the clay suspen-

sion is not maximized. In addition, a stoichiometric relationship be-

tween the polymer dosage required for optimum clay destabilization and

the initial clay concentration of the suspension has been found to exist,

in the clay concentration range studied, for the kaolinite and montmor-

illonite clays and cationic polymers No. 4 and No. 6 (Figures 23 and 24).

However, the stoichiometric relationship is not so precise for the mont-

morillonite clay, possibly because of the self coagulating tendencies

exhibited by this clay. The validity of this stoichiometric relation-

ship for clay concentrations smaller than 15 mg/1 has not been verified.

In fact, the straight line plot for the kaolinite clay-cationic polymer

No. 6 system does not extrapolate to the origin. It is conceivable that

the presence of polyanion molecules such as polyphosphate or lignosul-

fonate could complex or precipitate a certain fraction of the cationic

polymer added to the system, thereby causing a noticeable deviation in

the linear relationship between optimum polymer dosage and clay concen-

tration in the low clay concentration range. Cohen et al. have shown

that the above mentioned polyanions do increase the cationic polymer

demand of a clay suspension.

In Chapter II, theoretical models for polymer adsorption were dis-

cussed, and an extensive review of the literature revealed that the ad-

sorption of polymer molecules onto various types of adsorbent surfaces

could be approximated, in many instances, by the Langmuir adsorption model.

Subsequently, it was found that the adsorption data for the interaction

of kaolinite and montmorillonite clays with cationic polymer No. 4 and







- 83 -


cationic polymer No. 6 fit the Langmuir adsorption model very well. These

adsorption data were plotted according to the linearized form of the

Langmuir adsorption equation


C/X a 1 C


where C = equilibrium concentration of polymer

X = weight of polymer adsorbed

M = weight of clay

b = saturation weight ratio (mg polymer)
( mg clay )

and are shwon in Figures 25 to 40. A least squares regression line was

fitted to each set of adsorption data, and the slopes of the lines were

determined mathematically. The reciprocal of the slope of each straight

line adsorption plot represents the maximum or saturation weight ratio

of polymer to clay. The saturation weight ratio (b value) for each

polymer-clay system is presented in Table 5.

Figures 28, 32, 36, and 40 are summary graphs of the Langmuir ad-

sorption isotherms for the kaolinite-cationic polymer No. 4, the mont-

morillonite-cationic polymer No. 4, the kaolinite-cationic polymer No. 6

and the montmorillonite-cationic polymer No. 6 interactions. From the

saturation weight ratios presented in Table 5, it is shown that, with

the exception of the montmorillonite-cationic polymer No. 6 interactions,

the saturation weight ratio of adsorbed polymer to clay decreases as the

initial clay concentration increases. This is regarded as support of

the interparticle bridging theory for polymer-clay interactions simply







- 84 -


60




50





40


UO


30-




20 0


100

10 0 0


0
0 1I I I I 1
0 1 2 3 4 5 6 7

Equilibrium Concentration
(C)(mg/1)(10)-1

Fig. 25 Langmuir Adsorption Isotherm. Cationic Polymer No. 4;
Kaolinite Clay Concentration 14.9 mg/l.