1

EFFECT OF POLYACRYLAMIDES ON THE PHYSICAL PROPERTIES OF SOME LIGHT-

TEXTURED SOILS

A thesis submitted in fulfillment of the requirements for the degree of

Master of Applied Science

Shane Phillips BSc, Grad Dip Sci, MAgSc

The University of Adelaide School of Earth & Environmental Sciences

Waite Campus

August 2007

2

Summary

The work presented in this thesis represents a combination of laboratory and field

experiments designed to explain some field observations I made on some coarse

sands in South Australia in 1999: that much of the irrigation water and nutrients

applied to the sands under drip-irrigation simply passed through the root zone leaving

the crops water-stressed shortly after irrigation events. There was clearly only

minimal lateral spread of the water in these coarse sands. However, by applying

small concentrations of polyacrylamide or ‘PAM’ in the irrigation water, the crops

seemed to do better. Furthermore the timing of irrigation events appeared to be more

flexible on the PAM-treated properties. I found this intriguing and saw an

opportunity to increase the lateral spread of water in the root zone and thereby

increasing the stored volume and residence time of water. By retaining more water in

the rootzone, there was potential to save a considerable amount of costly irrigation

water, and also improve crop production and quality. With encouragement from my

then employer (Elders Pty Ltd) and from Ciba Specialty Chemicals Australia, I

undertook to explore my findings in greater detail and to attempt to explain them

based on some ‘hard’ (experimental) evidence.

Increasing the lateral spread of water under drippers in coarse-textured soils requires

water to be retained in the root zone for longer periods during irrigation, but the

practical methods for doing this are limited to:

• Altering the pore size distribution to create a finer average range of pore

sizes, either by compaction or by stabilization of smaller pores using organic

matter or additions of clay.

• Reducing the wettability of the soil so that less water can be taken in and the

soil never becomes saturated. (This of course risks surface runoff and

suboptimal placement of irrigation water).

• Altering the physical properties of irrigation water (eg. viscosity, surface

tension) so that it interacts differently with soil pores and moves through

them more slowly.

3

The aim of the research was therefore to evaluate the potential for some

commercially available PAMs to reduce hydraulic conductivity and to increase water

retention on some drip-irrigated coarse sandy soils of South Australia and Victoria.

I worked with two commonly available anionic polyacrylamides, designated PAM-

1011 and PAM-135, and measured ponded infiltration in laboratory columns of

seven different sandy soils from around South Australia and Victoria. I varied the

concentration of the polymer within the range typically used in the field (0, 1 and 10

ppm for PAM-1011; 0, 2.5 and 25 ppm for PAM-135) and I also varied the quality of

the irrigation water used to mix the PAM solutions in terms of salinity and sodicity

(distilled water, 10 and 20 mmol(+) salt/L, using either sodium chloride, calcium

chloride, or both). I measured the effects of PAM on pore-size distribution of one of

the sands (by the water retention characteristic), on water repellence of the soils (by

measuring water droplet penetration times), and the kinematic viscosity of the PAM

solutions at various concentrations with various qualities of irrigation water. I also set

up transparent cases of sand to observe infiltration and wetting behaviour of the PAM

solution. Finally, with some understanding of how the physical and chemical

properties of the PAMs, I conducted a field trial to measure the soil water matric

potential at various depths and locations around drippers in vine rows receiving PAM

in the irrigation water.

Laboratory findings

The polymer PAM-1011 significantly reduced the steady-state infiltration rate in all

sands, and it did this with relatively modest concentrations (< 10 ppm). The

polyacrylamide PAM-135 was not effective for this purpose, which indicated that the

chemical properties of the polymer (not investigated here) influenced its physical

behaviour. Further work with PAM-135 was therefore discontinued in favour of

PAM-1011.

The effectiveness of PAM-1011 in reducing steady-state infiltration rates was related

to changes in the properties of the irrigating solution caused by PAM-1011 rather

than by a change in the properties of the soils to which it is applied. For example,

PAM-1011 had only minimal (if any) influence on the pore size distribution (water

4

retention) of a coarse sandy soil and had no significant impact on water repellence

(wettability) of another sandy soil. It did, however, have a large impact on the

kinematic viscosity of the irrigating solution, and the more PAM-1011 that was

dissolved, the more viscous the solutions became.

The effectiveness of PAM-1011 in reducing steady-state infiltration rates was

reduced in salty irrigation water, and there was evidence to suggest that cation-effects

may have been involved. When PAM-1011 was dissolved in distilled water,

infiltration rates were reduced by the greatest amount. When PAM-1011 was

dissolved in salty water containing the monovalent cation, sodium, infiltration rates

were not reduced as much; furthermore, if the solvent water contained the divalent

cation, calcium, PAM-1011 was even less effective than in sodium-rich water. Thus

electrolytes affected the physical conformation of PAM-1011 solutions, altering

viscosity. To overcome the salt-water effects, higher concentrations of PAM-1011

needed to be used.

The cation-effects were primarily related to the way each cation interacted with the

polymer to alter its kinematic viscosity. PAM-1011 in distilled water had the greatest

viscosity, while PAM-1011 in sodium-rich water had a lower viscosity, and PAM-

1011 in calcium-rich water had the lowest viscosity. A practical implication from this

is that irrigators using salty waters will need to dissolve more PAM-1011 in their

water-sources to increase the viscosity and thus gain the retarding effects of the

polymer on infiltration rates. The data suggest that the amount of polymer required to

overcome the salt effects is about 10 ppm PAM-1011. Rates as low as 1 ppm can be

used when irrigators have access to high-quality water with < 10 mmol(+) salt/L

present. Visual observations of the wetting fronts during infiltration showed that

irrigation water containing PAM-1011 at between 1 and 10 ppm reduced the depth of

percolation and increased its lateral spread in coarse sands.

Field study

The field work was largely unsuccessful because shortly after the treatments were

applied, a 1-in-100 year hailstorm struck that completely wiped out the vegetation on

5

the vines in the study. I spent most of the season simply trying to keep the vines alive

and to recover some of the leaf area for future years.

Overall, however, this work identified the ability of PAM-1011 to reduce water

movement through the root zone of coarse sands, and demonstrates the potential to

conserve a great deal of water – a significant move toward higher water- and

nutrient-use efficiencies on the coarser textured soils in the Murray-Darling Basin.

6

Declaration I declare that this thesis contains no material that has been accepted for the award of any other degree or diploma in any university and to the best of my knowledge and belief contains no material previously published or written by another person, except where due reference is made in the text. I give consent to this copy of my thesis, when deposited in the University library, being available for loan and photocopying. Signed _____________________________________ Date ___08/08/2007___

7

Acknowledgements

As a part time external student the journey to completion has been a long and slow one. The journey itself would have been impossible without the support of many people who have provided me with invaluable support over the time frame. To Ciba Specialty Chemicals and in particular Andrew McHugh, John Bellwood and Eric Hoftler, I would like to thank them for the endless support over the journey. No question was ever too trivial to be answered and the help provided over the project was fantastic and greatly appreciated. To Tandou Ltd, the ability to access an employer who not only allows further studies but, actively encourages personal development was invaluable. Organisations such as this are great ones to be involved with. To Cameron Grant and Rob Murray, I am unsure if two better supervisors exist. The positive criticism and continual encouragement along the way was always appreciated. To write on the impact that these two people have had on me, words would probably be inadequate. However from a farming perspective I am a better grower than I was before meeting both Cam and Rob. The thesis was as much about applied mentoring and as such I am eternally grateful for their input into my development. And to my wife who will no longer have to cringe when people ask her what I am interested in, the journey has come to a close.

8

Contents Summary 2 Declaration 6 Acknowledgments 7 List of Figures 10 List of Tables 12 CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW 1.1 Introduction. 13 1.2 Literature Review. 14

1.2.1 Organic polymers: their effects on soil physical & hydrological properties. 14 1.2.2 Polyacrylamides and their properties. 18 1.2.3. Influence of PAM on soil physical properties. 24 1.2.4. Impact of PAM on plant growth and development. 27 1.2.5 The influence of PAM on soil microorganisms. 28

1.3 Conclusion 30 CHAPTER 2: USE OF ANIONIC PAMs TO REDUCE INFILTRATION RATES IN SOME COARSE SANDY SOILS 2.1 Introduction. 33 2.2 PAM-polymers and their properties. 33 2.3 Coarse sandy soils. 34 2.4 Quality of irrigation water. 35 2.5 Infiltration equipment and procedures. 35 2.6 Results and discussion. 2.6.1 Overall effectiveness of two polyacrylamides (PAM-1011 and PAM-135) at reducing steady-state infiltration rate in three sands. 38 2.6.2 Effectiveness of PAM-1011 at reducing steady-state infiltration rate: influence of particle size. 41 2.6.3 Effectiveness of PAM-1011 at reducing steady-state infiltration rate in a range of different sandy soils. 43 2.7 Conclusions. 47 CHAPTER 3: EVALUATION OF WHY PAM REDUCED STEADY- STATE INFILTRATION RATES IN COARSE SANDS 3.1 Introduction. 49 3.2 Effect of PAM on water repellence in sands. 50 3.2.1 Introduction. 50 3.2.2 Materials and methods. 51 3.2.3 Results and discussion. 51 3.2.4 Conclusions. 53 3.3 Effect of PAM on pore-size distribution of sand from Overland Corner as shown by water retention. 53 3.3.1 Introduction. 53 3.3.2 Materials and methods. 54 3.3.3 Results and discussion. 54 3.3.4 Conclusions. 55 3.4 Kinematic viscosity of solutions of PAM-1011. 56

9

3.4.1 Introduction. 56 3.4.2 Materials and methods. 56 3.4.3 Results and discussion. 57 3.4.4 Conclusions. 58 3.5 Overall conclusions on why PAM-1011 reduced infiltration rates. 58 CHAPTER 4: VISUAL ASSESSMENT OF THE INFLUENCE OF PAM ON WATER MOVEMENT AND DISTRIBUTION THROUGH SANDS 4.1 Introduction. 59 4.2 Materials and methods. 59 4.3 Results and discussion. 60 4.4 Conclusions. 63 CHAPTER 5: FIELD STUDIES OF PAM-APPLICATION IN BURIED- DRIP IRRIGATION SYSTEMS ON SANDY SOILS 5.1 Introduction. 65 5.2 Materials and methods. 66 5.3 Results and discussion. 68 5.4 Conclusions. 71 CHAPTER 6: GENERAL DISCUSSION 6.1 Summary of findings. 72 6.2 Issues raised by this work and opportunities for research 73 REFERENCES 75 APPENDIX 82

10

Figures

1.1 Some commercially available anionic polymers in different forms, showing relative molecular weights and percentage of anionic acrylamide. 17 1.2 A schematic of the repeating structure of polyacrylamide (PAM), Showing acrylic acid, CH2CHCOO ± H, combined with the acryl- amide molecule, CH2CHCONH2. 21 2.1a Illustration of how infiltration volumes were corrected. The intercept of each line from a least-squares regression represents the volume of water that formed the shallow ‘pond’ over each column of soil. 37 2.1b Illustration of infiltration (mm) calculated from Figure 2.1a and plotted as a function of time (s) 37 2.1c Illustration of how the three sets of infiltration data from Figure 2.1b were re-plotted as a function of the square-root of time, and fitted to Philip’s polynomial. 38 2.2a Mean steady-state infiltration rates for 2 polymers at all concentrations relative to water of all qualities. 38 2.2b Mean steady-steady infiltration rates from Figure 2.2a, showing the differences between the three soils used in the preliminary study. 39 2.3a Cumulative particle size distributions for the first three soils examined (left) and the second group of soils (right) used in this study 40 2.3b Geometric mean diameters of the 7 soils used in this study 40 2.4 Effect of concentration of PAM-135 on steady-state infiltration rate of three sandy soils. 41 2.5 Effect of mean particle size on effectiveness of PAM-1011 at three different concentrations to reduce steady-state infiltration rate. 42 2.6 Percent reduction in effectiveness of PAM-1011 with particle size. 42 2.7 Effect of application rate of PAM-1011 on reducing steady-state infiltration rate in a range of sandy soils having different particle size distributions. 43 2.8 Overall mean steady-state infiltration rate of PAM-1011 in seven soils as affected by salt concentration in the water used to prepare the solutions. 44 2.9a Effect of 1 ppm PAM-1011 in irrigation water of varying salinity and sodicity on steady-state infiltration rate in three sandy soils. 45 2.9b Effect of 10 ppm PAM-1011 in irrigation water of varying salinity and sodicity on steady-state infiltration rate in three sandy soils. 45 2.10a Effect of 1 ppm PAM-1011 in irrigation water of varying salinity and sodicity on steady-state infiltration rate in Viognier- and Pardo- surface- and subsoils. 46 2.10b Effect of 10 ppm PAM-1011 in irrigation water of varying salinity and sodicity on steady-state infiltration rate in Viognier- and Pardo- surface- and subsoils. 47

11

3.1a Mean steady-state infiltration rate of water as a function of particle size for various sieved sand fractions (triangles) and natural sands from the field (circles). 49 3.1b Mean steady-state infiltration rate as a function of same particle sizes shown in Figure 3.1a for water treated with PAM-1011 at 1 and 10 ppm. 49 3.2 Water retention curves for Overland Corner with and without addition of PAM-1011. 54 3.3 Water retention curves for Overland Corner with and without pre- treatment with PAM-1011 55 3.4 Effect of PAM-1011 and solvent cation on solution kinematic viscosity 57 4.1a Photo of sketches on prespex showing wetting fronts for Water, Water + 1ppm PAM-1011, and Water + 10 ppm PAM-1011 after 30 minutes. 61 4.1b Photo of sketches on prespex showing wetting fronts for Na-water, Na-water + 1ppm PAM-1011, and Na-water + 10 ppm PAM-1011 after 30 minutes. 61 4.1c Photo of sketches on prespex showing wetting fronts for Ca-water, Ca-water + 1ppm PAM-1011, and Ca-water + 10 ppm PAM-1011 after 30 minutes. 61 4.2 Effect of PAM-1011 and sodium or calcium in the water on the maxi- mum depth to which dripping-liquid could penetrate Vigonier sand after 30 minutes. 62 4.3 Effect of PAM-1011 and sodium or calcium in the water on the maxi- mum lateral spread of dripping-liquid in Vigonier sand after 30 min. 63 5.1 Inline fertilizer injector installed in buried drip-line (December 2003) 67 5.2 Soil water matric potential as measured by tensiometers at 30, 60 and 90 cm below the soil surface in the Merlot-section during the 2003/04 growing season. 69 5.3 Soil water matric potential as measured by tensiometers at 30, 60 and 90 cm below the soil surface in the Viognier-section during the 2003/04 growing season. 70 5.4 Visual observations in the areas surrounding the tensiometers. 71

12

Tables 1.1 Examples of types of commercial polyacrylamides commonly used. 17 1.2 Typical industrial PAMs produced by Ciba Specialty Chemicals and the properties used to describe them. 22 2.1 Properties of the anionic PAM polymers used in this study. 34 2.2 PAM concentrations prepared by diluting stock solutions to obtain Equivalent concentrations of the monomer. 34 2.3 Sands used in this study and some of their properties. 35 2.4 Concentrations of sodium and calcium in the stock solutions. 35 2.5 Percentage of sample mass less than the arithmetic mean particle diameter for the soils used in the infiltration work. 39 3.1 Solutions used as droplets to determine the WDPT for each sample. 51 3.2 Classes of water repellence proposed by Lal and Shukla (20004). 51 3.3a Water Droplet Penetration Times (seconds) for solutions of different salt concentration for some surface soils used in this study 52 3.3b Droplet Penetration Times (seconds) for solutions of 1 ppm PAM-1011 of different salt concentration for some soils used in this study. 52 3.3c Droplet Penetration Times (seconds) for solutions of 10 ppm PAM-1011 of different salt concentration for some surface soils used in this study. 52 3.4 Droplet Penetration Times (seconds) for solutions of 1000 ppm PAM-1011 in distilled water for some soils used in this study. 53 3.5 Pre-treatment of samples for measurement of water retention. 54 3.6 Kinematic viscosity of irrigating solutions as influenced by PAM-1011 concentration and type of salt in the solvent. 57 4.1 Solutions used to observe wetting behaviour used to observe wetting behaviour in Viogneir (0-5 cm) sand. 60 5.1 Soil survey profile descriptions of the two sections of the vineyard chosen to apply PAM-1011 and monitor soil water status 66 5.2 Schedule of irrigation events with 1 ppm PAM-1011 during the 2003/04 growing season. Tick marks indicate irrigation occurred on the date shown; dash-marks indicate no irrigation occurred. 67

Appendix

Raw infiltration data (on CD)

13

CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW

1.1 Introduction

The world is faced with a rapidly increasing population plus an overall decline in

arable land and high-quality water for irrigation. The pressure to produce more food

with less land and poorer quality water is becoming a reality for farmers in many

parts of the world (Hillel 1998). Furthermore, many agricultural soils are light-

textured and do not retain water against drainage. Irrigation- (and rain-) water thus

infiltrates quickly and percolates past the root zone to depth. This wastes valuable

water and can lead to groundwater contamination and rising water tables (Beresford

et al. 2001). Responsible land managers therefore want to conserve soil and water

resources to achieve maximum yield and quality, while at the same time minimizing

damage to the environment.

To some extent, water use efficiency (i.e. dry matter yield per mm water added) can

be improved through good agronomic practices (e.g. inter-row cropping with deep-

rooted species alongside shallow-rooted crops (Lefroy et al. 2001) – in this way,

water and nutrients missed by the shallow-rooted crops are intercepted lower in the

profile by the deep-rooted species). However, where water must be retained for plant

use in the surface soil layers, such agronomic practices are of little value. Here, we

require techniques that reduce percolation and retain water for plant uptake, the

options for which are fairly limited, including:

1. modify the pore-size distribution of the soil such that large pores become

smaller and thus retain water against gravity;

2. modify the properties of the irrigation water (e.g. surface tension, contact

angle, viscosity) such that it behaves in a more viscous manner and flows

through the soil more slowly;

3. modify both the pore-size distribution and the properties of the water.

There is abundant evidence in the agricultural and horticultural literature to suggest

that organic polymers, of which there are many, are capable of modifying the

properties of soils, particularly poorly-structured soils of medium texture to which

14

they are added (Sojka et al. 1998). Little work has been done, however, with

polymers on lighter-textured soils, particularly the coarse sandy soils of southeastern

Australia. Of particular interest to the author was some practical success found in

1999 by applying small concentrations of polyacrylamide (PAM) in the irrigation

water used in some horticultural operations on very coarse sandy soils in South

Australia. Little evidence was available, however, to explain anecdotal reports that

PAM was successful at conserving water and improving crop yields. In the author’s

experience, PAM-applications sometimes led to 100% increases in potato yields

applying 50% as much water compared to standard practices. This project was thus

conceived in the year 2000 with the desire to understand the factors responsible for

the large yield and water-use differences found on coarse sandy soils with and

without the application of PAM in irrigation water.

Experiments were undertaken based on the supposition that PAM might be capable

of altering both the physical properties of the soil as well as those of the irrigation

water to which it was applied. The nature of these experiments will be outlined in

separate chapters following a review of the literature on organic polymers and their

effects on soil and water.

1.2 Literature Review

1.2.1 Organic polymers: their effects on soil physical & hydrological properties.

Organic polymers such as the polyacrylamides (PAMs) and polyvinyl alcohols

(PVAs) have been widely used over the past 40 years to stabilise soils, control

erosion (Abu-Zreig 2006; Ben-Hur 1994, Sojka et al. 2004, 2007) and to increase

water and nutrient infiltration and retention (Ajwa & Trout 2006; Sivapalan 2003).

For example, Williams et al. (1967) showed that the wet-aggregate stability of a fine

sandy loam was greatly increased after treatment with PVA. Finch (1973) showed

that application of PVA improved soil structure by converting single-grained

structures into a more aggregated form. He also found that PVA was most effective

on loamy textured soils. Wallace et al. (1986) found that polymers created 100%

water-stable aggregates compared with only 38% in an untreated control soil, and

Aase et al. (1998) found PAM reduced runoff and erosion in sprinkler-irrigated

laboratory soils.

15

The utility of organic polymers is based on their long-chain molecular structures,

which contain repeating units held together by strong covalent bonds (Young 1981).

There are hundreds or even thousands of PAM-molecular formulations, and each one

PAMs are more suitable than others for stabilising different types of soil. The water-

soluble polymer, PAM, for example, is used largely to flocculate colloids and thicken

colloidal suspensions (Barvenik 1994). By contrast, the organic polymer,

polyacrylonitrile, is only slightly water soluble (Billmeyer 1971) and is thus more

suitable for non-aqueous applications.

Polymers can readily adsorb onto solid surfaces (Letey 1994; Lu et al. 2002). Each

polymer group or polymeric ion can have many groups or segments that can be

potentially adsorbed, the groups being essentially free of mutual interactions (Stumm

1992). The extent of polymer adsorption generally increases with increasing

molecular weight. The number and type of functional groups within the polymeric

structure also influence the extent of adsorption (Stumm 1992).

The widespread agricultural use of polymers has been somewhat restricted over the

past 40 years due to their high cost and application rates required (Sojka & Lentz

1994). For example, Finch (1973) found that up to 8-10 kg PVA/100m2 (0.8 -1 tonne

PVA/ha) were required for clay soils, which, at the time, was prohibitively expensive

($10-15/kg prior to 1980 – A McHugh, Ciba Specialty Chemicals, pers. comm.) for

most applications (Seybold 1993). Furthermore the elaborate application procedures

involved in the early 1970s made it relatively difficult to apply uniform

concentrations with available technologies at the time (Green & Stott 2001).

However, with higher purity, more efficacious polymers, better methods of

production and application for organic polymers, there is now considerably greater

scope to use polymers in horticultural ventures, where the costs can be more easily

justified1 (Nadler et al. 1994). This means we may now be in a better position to

consider different application rates for different soil types or for different qualities of

irrigation water (Ajwa & Trout 2006). The likelihood of an economic polymer-

1 It should be noted that in the 43 months prior to August 2007, prices for polymers have increased significantly due to the rising cost of oil-based raw materials such as acrylic acid, acrylonitrile, methyl acetate and propylene, all of which have more than doubled in price since 2002 (A McHugh, Ciba Specialty Chemicals, pers. comm.).

16

formulation is greatly increased where minimal industrial refinement is required or

where suitable organic polymers are produced as waste products from other industrial

processes.

Of particular interest to this study is the potential use of various synthetic organic

polymers known as the polyacrylamides (PAMs). These have many practical

applications, including their use in relatively small concentrations for municipal

water clarification, flooding agents in petroleum recovery, soil stabilisation, paper

thickening- and strengthening-agents (Finch 1973). PAMs were originally used in

World War II to allow rapid construction of roads and runways under wet conditions

(Wilson & Crisp 1975). Subsequent development of PAM-technology found its way

into the USA agricultural industries in the 1950’s for example, to enhance the

stability of tilled agricultural soils (Azzam 1980; Cary & Evans 1974; Shainberg &

Levy 1994), and as a coating agent in the formulation of biopesticides and fertilizers

(Burgess 1998; Sharma 1988). Approximately half a million hectares of irrigated

farmland in the USA currently use various PAMs for erosion control, infiltration

enhancement and improvement of runoff-water quality (Bicerano 1994; Lentz &

Sojka 1994; Sojka & Entry 2000). This is particularly important for reducing loads of

nutrients, pesticides, microorganisms, weed-seeds (and thus biological oxygen

demand) in water bodies that receive runoff (Grula et al. 1994).

Because the acronym ‘PAM’ is widespread in the literature with little description of

the properties of the particular ‘PAMs’ being used, their effects on soils are highly

variable. Such data are of little value without a detailed description of the

characteristics of the product being used. As a minimum, the number of acrylamide

monomers that are combined to form the polyacrylamide chain, the intrinsic

viscosity, the ionic content, plus the nature of any modification to the sub-units need

to be reported if the results are to be useful to future readers. Unfortunately, this sort

of information is difficult to obtain (due to commercial confidentiality) and reported

results therefore are of varying utility. Examples of the sort of limited information

available for commercial PAMs and other polymers are shown in Table 1.1 (A.

McHugh, Ciba Specialty Chemicals, pers. comm.).

17

There is a huge range in the properties of the commercially available polymers, as

shown by the wide scatter of product-numbers shown in Figure 1.1, which illustrates

the ion content of the polymers (i.e. percentage by weight of anionic acrylamide) and

their molecular weights. It is thus possible to obtain commercial PAMs that vary

enormously in their physical and chemical properties. The two PAMs used in the

experiments reported in this thesis are circled in red in Figure 1.1.

Table 1.1 Examples of types of commercial polyacrylamides commonly used.

Groups of polymers produced by Ciba Specialty Chemicals

Range of solid content by weight (%)

Poly-dadmacs 20-40 Poly-amines 50

DCD Polymers 50-55 Mannich Polymers 6-20

Other polymers 15

Figure 1.1. Some commercially available anionic polymers in different forms, showing relative molecular weights and percentage of anionic acrylamide. The same range of products is available for cationic and uncharged polymers (A. McHugh, Ciba Specialty Chemicals, pers. comm.).

10 7030 50 90

Anionic Value

Mo

lecu

lar

Wei

gh

t

351

800HP

1017

139

155

156

336342

338

345356

358

366

525

611

919

1011

10

24

A-600

A-661

A-662

A-663

A-665

A-HP22

A-W5

A-W14

A-W23

A-W50

5250L

5250H

X110

X125 X135X145

90L

110L

120L

Commercially available form

Gel

Bead

LDP

Inverse Emulsion

LT20

LT25

LT27

LT27AGA-HP21

(A-661W)

A-HP20DP1-8468

10 7030 50 90

Anionic Value

Mo

lecu

lar

Wei

gh

t

351

800HP

1017

139

155

156

336342

338

345356

358

366

525

611

919

1011

10

24

A-600

A-661

A-662

A-663

A-665

A-HP22

A-W5

A-W14

A-W23

A-W50

5250L

5250H

X110

X125 X135X145

90L

110L

120L

Commercially available form

Gel

Bead

LDP

Inverse Emulsion

LT20

LT25

LT27

LT27AGA-HP21

(A-661W)

A-HP20DP1-8468

18

An understanding of the physical and chemical properties of PAMs is crucial in

deciding how best to make use of them in any given situation. The following review

of the relevant chemistry of organic polymers and polymer formulation (particularly

in relation to PAM) explains how aqueous PAMs can bond with (and alter the

properties of) granular materials such as soils.

1.2.2 Polyacrylamides (PAMs) and their properties

Structures and properties of PAM

The formulation of polymers uses the industrial process known as polymerisation,

wherein large numbers of small molecules (monomers) are joined together to form

larger molecules of varying length. The process of polymerisation is the single most

important reaction within the alkene group, for which polymerisation occurs by

exchanging bonds within the molecule. The addition of 50% sulphuric acid at 100 C,

for example, creates an intermediate carbo-cation, which can react with the new

alkene to form a new tertiary carbo-cation. In the absence of suitable nucleophilic

compounds to react with the intermediate carbo-cations, reaction with the alkene is

the only option possible. Isobutylene is another example that can be reacted with a

small quantity of boron trifluoride at low temperatures to produce a high molecular-

weight polymer. Production of carbo-cations can only proceed in the presence of

water. The intermediate salt (BF3OH- ion) has low nucleophilicity and the t-butyl

cation is free to react with isobutylene to start the cationic polymerisation

(Streitwieser & Heathcock 1985). PAMs result from the polymerisation of

acrylamide molecules.

The degree of polymerization refers to the average number of monomer units per

polymer molecule (Brown & Lemay 1981) and is expressed in units of mole / mole –

examples of this are referred to below. Polymerisation often involves selective

bonding by exposure of a monomer (e.g. alkene) to an acid (e.g. sulphuric acid) and

then heating (or not heating) the mixture, depending upon the nature and type of

polymer being produced (Streitwieser & Heathcock 1985). Polymerisation of

monomers is initiated by free radicals (molecules containing one or more pairs of

unpaired electrons), which are quite reactive (Cowie and Arrighi 2007). An effective

initiator is a molecule that will undergo homolytic fission into a radical of greater

19

reactivity than the monomer radical. A radical then bonds with one of the carbons on

the polymer chain after uncoupling from the molecule, thus creating another free

radical, which repeats the uncoupling/coupling process and thus extends the length of

the polymer chain. The number of times a propagation cycle repeats itself determines

the final chain length (Brown 1988), which is achieved only after the free radicals are

brought into contact and coupled with another (Brown & Lemay 1981).

The length of the polymer chain can be determined precisely by introducing catalysts

that initiate or terminate free radicals. The choice of catalyst can also influence the

characteristics of the polymer. For example, polyethylene, which is generated from

polymerisation of ethylene, can be a pliable material or a much stiffer substance

depending on the nature of the catalyst used in production. Polymerisation can

produce branched-chain or linear-chain polymers, and cross-linkages can occur

between groups from separate chains to produce a three dimensional network. Cross-

linking restricts the relative mobility of the chains and thus impacts on the physical

properties of the polymer (Streitwieser & Heathcock 1985). Polyethylene and

polyacrylamide (PAM) are typical products of polymerization. The remainder of this

review will focus on the formulation and properties of PAM.

Many polymers are resistant to chemical attack and are stable when subjected to

mechanical deformation (Cowie and Arrighi 2007). However, very few polymers can

withstand the effects of extreme temperature or ultra violet radiation, and the PAMs

used in this study all degrade with time when exposed to sunlight.

The formation of free radicals from non-radical compounds is called chain initiation.

Cowie and Arrighi (2007) identified chain initiators in three categories: free-radical

initiators, cationic initiators and anionic initiators. The choice of initiator-type

depends largely on the nature of the R1- and R2-groups in the monomer and their

effect on the double carbon bond. This arises from the way the alkene bond reacts

differently to initiator species. An effective initiator is a molecule that will readily

undergo homolytic fission into radicals of greater reactivity than the monomer-

radical. In the polymerisation of the alkenes in the presence of peroxide, chain-

20

initiation is by thermal cleavage of the O-O bond in the peroxide molecule to

generate two alkoxy radicals.

Chain propagation is the reaction of a radical and a molecule to give another new

radical. Chain propagation occurs repeatedly such that the new radical reacts with a

molecule to produce another new radical and so on. The number of chain-

propagations is called the chain length, and this is terminated by the destruction of

radicals as they come into contact, such that their unpaired spins may couple and

prevent further reaction (Brown & Lemay 1981). When polymers are formed by free

radical reactions, conditions must be controlled so that the desired polymer chain

length is achieved.

Given the plethora of possible polymer reactions, it is considered difficult in an

industrial setting to create and recreate the precise conditions to ensure chain

termination and thus reproducible formulations during polymerization. As such it is

common for industrial polymers of a given product to have variable chain lengths

and thus variable physical and chemical properties. At a commercial level it is

unlikely such variances in chain length would be of great importance, but for

research purposes, it could be crucial. For this reason, it is important to select a

source of polymer for research taken from a single industrial batch and that its

characteristics be evaluated and reported in detail.

Properties of aqueous PAM solutions

PAMs are synthesized in four main structural groups: cationic PAMs, neutral PAMs,

amphoteric PAMs, and anionic PAMs (Seybold 1993). The anionic groups are most

common in environmental and agricultural applications, while the others are rarely

used because of their relatively high environmental toxicity (Sojka & Surapaneni

2000). For example, some cationic PAMs have sufficiently low LC50s to be of

concern to aquatic organisms. Cationic polymers are also less able to stabilize soil

particles at similar concentrations to anionic and neutral polymers (Lentz et al. 1993).

The nature of different PAMs varies according to the chain length and the number

and kinds of functional-groups substituted along the chain. The amide group on a

21

PAM-molecule, for example, can be replaced with functional groups containing

sodium ions or protons that dissociate in water to provide negatively charged sites

(Sojka & Lentz 1996). PAMs are formulated from the carboxylic group, 2-propenoic

acid (acrylic acid, CH2CHCOOH) and acrylimide (CH2CHCONH2). Acrylic acid

provides the anionic charge while the acrylamide monomer provides the ‘backbone’

of the polymer (Figure 1.2). PAMs are considered to be co-polymers because they are

formed from two types of monomer units (Cowie & Arrighi 2007; Sojka &

Surapaneni 2000).

Because the carboxyl group contains three polar covalent bonds, carboxylic acids are

polar compounds. They are capable of interacting with water molecules by hydrogen-

bonding between the carboxyl oxygen and the hydroxyl groups. The carboxyl group

consists of two distinct parts: a hydrophilic carboxyl group (which increases water

solubility), and a non-polar, hydrophobic, hydrocarbon chain (which decreases water

solubility). The size of the hydrocarbon chain relative to the size of the hydrophilic

carrier thus dictates how water soluble the polymer is (Brown 1988) and explains

why high-molecular weight PAMs are so difficult to dissolve in water.

Figure 1.2 A schematic of the repeating structure of polyacrylimide (PAM), showing acrylic acid, CH2CHCOO ± H, combined with the acrylamide molecule, CH2CHCONH2.

Variable lengths and molecular weights of PAM impart variations in its chemical and

physical properties. For example, Thomas (1964) reported PAM lengths up to 6

million monomer units, with the solutions of greater molecular weight having greater

viscosities. Charge-density also varies with chain length and the proportion of

sodium acrylate substituted for the amide group. Within each charge-density group,

molecular weights vary according to the length and makeup of the PAM molecule.

22

Furthermore, the ratio of acrylamide to acrylic acid can be altered during manufacture

and this varies the anionic charge of each polymer chain. The two monomers then

react to produce a long chain of repeating acrylamide and acrylic acid units.

PAMs can be characterized by their molecular weights but are more commonly

described in terms of their intrinsic viscosities and ionic content. Intrinsic viscosity,

[η], gives an indication of the volume occupied by a single non-ionised molecule

(units = dL/g), and this influences its physical behaviour. The ionic content, IC,

describes the amount of charge on the polymer, and is expressed as the proportion by

weight of the formulation comprising sodium acrylate. Typical intrinsic viscosities

and ionic contents for a range of PAMs is shown in Table 1.2.

Table 1.2. Typical industrial PAMs produced by Ciba Specialty Chemicals and the properties used to describe them (A. McHugh, Ciba Specialty Chemicals pers. comm.).

PAM name Intrinsic viscosity, [η]

(dL/g) Ionic content (% sodium acrylate

on weight/weight basis) PAM-01 18 0 PAM-02 18 18

Magnafloc 1011 20 25 PAM-03 25 50 PAM-04 16 100

The usefulness of intrinsic viscosity as a measure of polymer molecular weight was

first recognised by Staudlinger in 1930 (cited in Billmeyer, 1971). The efflux-time

for a specified volume of a polymer solution to flow through a capillary tube is

measured and compared to that for the solvent (Billmeyer 1971). The size of the

dissolved polymer molecule can be (roughly) estimated from its intrinsic viscosity,

but the viscosity is affected by other properties of the polymer including its chain-

structure and the degree of branching within the chain – thus intrinsic viscosity is

only indirectly correlated with molecular weight.

Measuring the viscosity of PAM is not particularly easy – it varies with temperature

and concentration, and for concentrations > 400 ppm viscosity decreases with

increasing agitation or mechanical stress during pumping (Bjorneberg 1998). For

example, the viscosity of a PAM solution (2400 ppm) was reduced 15-20% when

passed through a centrifugal pump, because PAM molecules break and shear off

23

during pumping. Such reductions in viscosity reduce the effectiveness of PAM in soil

erosion control (Bjorneberg 1998). Of interest in the current study is the possible use

of less disruptive (diaphragm) irrigation pumps, which are less abrasive than

centrifugal pumps, and could thus maintain PAM-viscosities at their maximum levels

in the field.

Oliver and Ross (1964 p.257) described the nature of the various molecular

interactions that give organic polymers such strong adsorptive properties:

(a) London type dispersion forces resulting from induced-dipole/induced-dipole

and multipolar attractions,

(b) Induction forces brought about by the operation of a surface electric field on

induced- or permanent-dipoles of resident molecules,

(c) Charge transfer between the adsorbed molecules and other surfaces, resulting

in a no-bond resonance state,

(d) Dative bonding resulting from a chemical reaction between the adsorbate and

surface atoms.

Molecular interactions (a) and (b) result in physical adsorption, while the molecular

interaction (d) leads to chemi-sorption. Interaction (c) can be either physical or

chemical adsorption. When applied to soils and water, the behaviour of PAM may be

influenced by each of the molecular interactions described above. The size of the

PAM molecule, for example, influences the degree of London-type forces operating,

while soil water salinity and the composition of those salts influence the degree of

induction forces involved. Furthermore, it is not unrealistic to expect that PAM

applied in water to the soils should be influenced by the pH, EC and chemical

composition of the soil and water (Green et al. 2000). The retention of a polymer in

granular media depends on its diffusion coefficient and thus the hydrodynamic

diameter and molecular weight of the polymer.

While this study will concentrate primarily on the PAM-polymers, there are other

important synthetic polymers. Polyvinyl acetate (PVA), for example, was the first

totally synthetic colloid prepared by polymerising vinyl acetate in 1924 (Finch 1973).

PVA is a polymerised ester rather than a monomer (Finch 1973). In its purest form,

24

PVA is an explosive, flammable white hydrophilic powdery solid. The amount of

water adsorbed per unit volume of PVA is proportional to the external vapour

pressure of water up to 50% relative humidity. The coefficient of adsorption is

equivalent to about 13 moles water per hundred moles CH2CHOH (Pritchard 1970).

PVA dissolves more rapidly at higher temperatures, and its viscosity varies with salt

and pH of the solvent. For example, the viscosity of aqueous PVA increases with

salinity up to 6% NaOH, after which it decreases along with the solubility of PVA.

The extent of hydrogen bonding between water and PVA is significantly reduced in

the presence of NaOH, so the polymer tends to bond more effectively to itself and is

thus ‘salted out’ of solution. The viscosity is also increased by adding urea to

aqueous PVA (Pritchard 1970). As with many polymers, the solubility of PVA varies

with the degree of polymerisation and of hydrolysis. To control the degree of

polymerisation, a technique known as ‘emulsion polymerisation’ can be undertaken

using solutions of hydrophilic monomers. Acrylic acid or acrylamide can be

emulsified in a continuous oil phase using an appropriate water-in-oil emulsifier

(Billmeyer, 1971).

1.2.3 Influence of PAM on soil physical properties

Infiltration, runoff and erosion

Flood irrigation, which is still widely practiced around the world, causes a number of

problems, including runoff and soil erosion at the end of the irrigation furrows

(Bicerano 1994, Lentz et al. 1992). Ross et al. (1996) reported topsoil losses of 5- 50

t/ha/year on erodible soils in the northwestern USA. Eroded particles carry nutrients

and pesticides, and are regarded as major contributors to pollution of surface water

(Hayes et al. 2005; Sojka and Lentz, 1996). Much work has gone into changing the

properties of the soil using PAM and other polymers to increase infiltration and make

the soil less erodible (eg. Azzam 1980; Sojka and Lentz 1994). In some cases, PAMs

can even reduce wind erosion by stabilizing particles into larger aggregates, which

cannot be moved by wind (Chamberlain 1988).

The greatest erosion control is generally obtained by applying high-molecular-weight

PAMs at relatively low concentrations in the advancing furrow stream, rather than as

a dry product on the soil surface. Sojka and Lentz (1996) found that PAM applied at

25

10 ppm (18% anionic form) in furrow irrigation increased wet aggregate stability

(from 54 to 80% in 1993, and from 63 to 84% in 1994), increased infiltration by up

to 50%, and reduced runoff by up to 94%. Similarly, PAM erosion-trials in northern

Australia between 1995 and 1997 found that soils treated with only 1.2, 2 or 3.6 ppm

PAM greatly increased infiltration and reduced erosion (Schiller and Edmunds 1997).

The effects were greatest on coarse-textured soils, however, and work conducted on

heavy clay soils in the Darling Downs of Queensland showed only small increases in

infiltration. Sediment runoff and turbidity of all ‘tail waters’ (i.e. water leaving

irrigation furrows) were greatly reduced from between 0.4 – 1.8 g/L (control) down

to < 0.1 g/L where 3.6 ppm PAM was applied, and pesticide residues (eg.

oxyfluorfen) were completely eliminated (Schiller and Edmunds 1997). Levy et al.

(1992) applied two different PAMs in irrigation systems at 5, 10 and 20 ppm during

three consecutive irrigations, then applied water alone for two subsequent irrigations.

Both PAMs at all concentrations increased the stability and resistance of soil

aggregates to erosion. Ross et al. (1996) also found greater infiltration (10%) and

reduced sediment runoff (96%) using high-molecular-weight PAMs on a highly

erodible Portneuf silt loam in Idaho, USA. Orts et al. (2000) applied biopolymers at

rates between 5 and 120 ppm to reduce suspended solids by more than 80%. In fact,

applications of only 5 ppm reduced runoff by >90%. Becker (1997) reported on

research from Kimberly, Idaho that PAM applied at 5-10 ppm along with gypsum to

increase infiltration and completely eliminate rill erosion. Trout et al. (1995) reported

that 0.7 kg /ha of a high molecular weight PAM reduced surface sealing, increased

infiltration by 30%, and reduced furrow-induced erosion by 85-99%.

Many other studies have produced similar information with regard to reducing

irrigation-induced soil erosion. The low operating costs of furrow irrigation make it a

likely system to remain in place until either environmental or economic opportunities

allow change to proceed. In these systems the use of PAM offers growers a solution

to minimise issues involving irrigation-induced soil erosion. In a world where

economic and environmental issues are increasingly important, the cost of using

PAM in furrow based systems may offer irrigators a relatively economical option to

reduce erosion and increase water infiltration compared with the cost of installing

pressurized irrigation systems (Sojka et al. 2007).

26

Soil sodicity, structure and mechanical strength.

Increasing salt levels in irrigation water across the world pose problems for rural and

city dwellers alike – in Australia, this is most obvious in the Murray-Darling Basin.

Declining water quality threatens agricultural production, urban roads, buildings and

other infrastructure. As salinity increases, strategies need to be developed to enable

irrigated agricultural production to continue without major impacts on soils and

crops. The use of PAM offers an interesting opportunity in this battle.

The use of highly sodic water for irrigation can cause surface crusting (Cary & Evans

1974), reduce infiltration and hydraulic conductivity, as well as other problems with

soil structure. Gardiner and Shainberg (1996) used PAM at concentrations of 10, 25

and 40 ppm plus gypsum to overcome the effects of sodium when wastewater was

applied to the different soils over a period of eight weeks. PAM invariably increased

infiltration and hydraulic conductivity, and did this more effectively than gypsum.

Aly & Letey (1990) investigated the effectiveness of an anionic PAM (40J) and

cationic guar (T-4141) in increasing aggregate stability and flocculation of sandy

loam and sandy clay loam soils. Both PAM products increased flocculation over the

control for soils with sodium adsorption ratios (SAR) between 1 and 5, but the

anionic PAM (40J) was more effective than the cationic guar (T-4141) at low

concentrations. However only the anionic PAM (40J) produced a measurable

increase in flocculation in soil with SAR values as high as 15. They concluded that

the effects of polymers on one soil property such as flocculation could not be

extrapolated to the effects on another soil property such as aggregate stability or

rupture stress. Aly & Letey (1989) also showed that PAM ameliorated soil hardness

with water having an electrical conductivity (EC) of between 0.05 & 0.7 dS/m.

Malik et al. (1991) proposed that soil cracks could be used as drainage pathways if

these cracks could be stabilised with PAM. Samples of a montmorillonitic soil with

exchangeable sodium percentages (ESP) of 8 and 25 were packed in soil columns

then wetted and dried to create cracks. The soils were ponded with solutions of

anionic PAM at concentrations: 0, 25, 75 and 200 ppm, then allowed to drain and

dry. The PAM-treated soils all had significantly greater hydraulic conductivities and

27

salt removal. Zahow and Amrhein (1992) studied the impact of PAM and gypsum on

sodic soils and found that PAM alone increased hydraulic conductivity on soils with

ESP < 15. When both PAM and gypsum were used, hydraulic conductivity increased

from 0.0 to 0.28 mm/h in a soil with ESP = 32. When gypsum alone was used,

hydraulic conductivity increased to only 0.063 mm/h.

Levy et al. (1995) added small amounts of anionic PAM in dry form to two different

montmorillonitic soils with high ESP. Runoff was reduced from the soil of low ESP

(< 4) but less so for the soil having high ESP = 12. When PAM was added to the

irrigation water, however, it was more effective in the higher ESP soils. It was

possible that the reduced runoff was directly related to the viscous properties of the

PAM solutions, which slowed down the flow of water over the soil surfaces. The

possible change in viscous behaviour was hypothesized to be important in the work

reported later in this thesis.

Levy and Rapp (1999) added synthetic polymers to crusting soils and found that

seedling emergence was improved. The objective of the study was to evaluate the

mechanism by which polymer addition contributes to the reduction in crust

mechanical strength. Outcomes from this experiment revealed that for any given

moisture content, crust strength did not vary significantly between the control and the

polymer treatments. However, moisture content for a given cumulative drying time

was higher in the polymer treatments than in the control. This result indicated that the

polymer application delayed crust formation and maintained a crust with a lower

mechanical strength for a longer period of time. The application of low molecular

weight PAM at 40 kg/ha emerged as the most effective treatment for maintaining

high moisture content in the upper soil layer. The ability of the polymers to increase

aggregate stability at the soil surface and prevent clay dispersion is beneficial in

crust-susceptible soils exposed to overhead irrigation or rainfall.

1.2.4 Impact of PAM on plant growth and development

Crop yield

PAMs can increase crop yields in some circumstances, particularly where soil

erosion can be prevented. Levy et al. (1991) found that vegetative growth of

28

commercially grown cotton was inversely correlated with the amount of runoff in

irrigated fields. Where PAM was applied at 20 kg/ha, runoff was significantly

reduced on both vertisols and loesses, and yields were greater compared with

untreated soils. Stern et al. (1992) found that runoff from sprinkler-irrigated wheat

crops was reduced from 36% of total irrigation to only 1.4% when irrigation water

was treated with PAM plus phosphogypsum. (Where phosphogypsum alone was

used, runoff was 13% of total irrigation). The PAM treatments also gave significantly

higher grain yield and greater crop water use efficiency than the control plots. Pryor

(1988) also found that a cross-linked PAM applied at 15 kg/ha as a side-dressing

increased yield of processing tomatoes by 30%, and indicated the potential of PAM

to save water in drought-prone areas, or on sandy soils, or where water is highly

priced or in short supply.

Crop nutrition and soil nutrient levels

Leaching of nutrients through the soil profile can contribute to ground water

contamination and runoff from furrows can lead to nutrient losses from paddocks.

The application of PAM to soil can reduce these problems and improve crop

nutrition. For example, Wallace et al. (1986) grew wheat and tomatoes in soils to

which anionic polymers were applied at rates exceeding those required for soil

stabilisation. For an application rate of 1% PAM, vegetative growth rate increased in

both wheat and tomatoes, they both took up less phosphate and silicon, and the wheat

plants took up less manganese and boron. For an application rate of 5%, uptake of

some of the macro-element cations was also depressed.

Assessment of furrow runoff reveals water containing organic matter, sediments and

nutrients. The addition of PAM with the water was seen to markedly reduce furrow

runoff losses of sediment, orthophosphate, total phosphorus and chemical oxygen

demand. The use of PAM did not appear to influence nitrate runoff (Lentz et al.

1998a). PAM applied at 10 mg/L during the furrow advance had 5-7 times lower

phosphate loads in the runoff than the control areas (Lentz et al. 1998b).

Water-absorbent PAM-gels can reduce leaching losses of nitrogen in soilless media.

Bres & Watson (1993) incorporated two different PAMs (HydroSource and Agri-gel)

29

into the growing medium at 1, 2 and 3 g/L. Water retention by the growth medium, of

course, increased linearly with gel application rate. Nitrate and ammonium

concentrations in the plants were higher for the 3 g/L PAM treatments, and total

foliar nitrogen concentrations in the tomato leaves were significantly higher in the

HydroSource PAM than in either the control or Agri-gel treatments.

Abraham & Rajasekharan-Pillai (1990) coated urea fertilizer with N,N’methylene

bis-acrylamide cross-linked with different PAMs, which improved the slow-release

characteristics of the urea. The rate of release of nitrogen depended on the type of

PAM used in the coating, and clearly indicated the utility of PAMs for reducing

leaching losses and groundwater recharge, and for increasing availability and uptake

of nitrogen by plants.

Seedling emergence and survival

Huttermann et al. (1999) applied water absorbent PAM to sandy soils at rates ranging

from 0.4 to 4% when planting pine seedlings. The soil water content was invariably

higher where more PAM was added, and appeared to increase the water holding

capacity of the soil from that of a sand-textured material to that of a loam or silty clay

material. Under drought conditions, seedlings in the PAM-treated soils exhibited a

threefold increase in shoot and root growth relative to the controls. This has

significant implications for improving the success of tree-planting operations in

drought-prone Australia – the use of PAMs could lead to quicker establishment and

thus earlier return on investment. Little work has been done, however, to evaluate

PAMs under Australian conditions to consider how they can be used to modify soil-

water relationships.

1.2.5 The influence of PAM on soil microorganisms

Immobilization of microorganisms

Sojka and Entry (2000) noted that PAM-treated irrigation water accumulated algal

tufts in the tail waters at the end of the furrows whereas untreated furrows did not.

Active fungi were reduced in tail water and microbial numbers in the soil were

generally greater along the furrow length, which suggested that PAM-applications

reduced the mobility and outflow of all classes of microorganisms observed.

30

Tucker et al. (1998) used PAM gels in column reactors to immobilize intact cells of

Desulfovibrio desulfricans and thus remove uranium and molybdenum contaminants

from water. The use of PAMs in combination with various bacteria have great

potential to control pollutants in waste water systems.

Impact on culturable heterotrophs

Shoemake-Kay et al. (1998a,b) found significantly greater numbers of culturable

heterotrophs in PAM-treated areas of potato fields. Total soil nitrogen levels were

significantly higher in PAM-treated soils (~36.7 mg NO3/kg and 1.3 mg NH4/kg in

treated soils, versus 10.7 mg NO3/kg and 0.5 mg NH4/kg in untreated soils), although

the effects were highly crop- and site-specific. The same effects were not found for

dry beans, and the overall effects on microorganisms are variable and inconsistent

(Sojka et al. 2007).

Seybold (1993) claimed that PAM, which resists microbial decomposition in many

circumstances, does not pose a threat to higher organisms when used correctly. The

only concern is the residual monomer, which is known to be a neurotoxin in humans.

Aqueous PAM acts as a substrate for mould and is degraded when nutrients are

present. PAM can also be broken down by cultivation, sunlight and the mechanical

breakage of the monomer chain (Seybold 1993). Levy et al. (1992) suggested that

wetting and drying in soils may cause degradation of PAM and noted its reduced

efficacy with time in the field, as did Lentz et al. (1992; 1994). This is relevant to

conditions in the South Australian mallee, where drying and occasional wetting

dominates the rural landscape. The effects of wetting and drying on the longevity of

PAMs has not been investigated but would be important to understand in relation to

the timing and application rates required.

1.3 Conclusion

The literature illustrates the broad application of PAM worldwide for improving

water and nutrient status and structural stability of soils. PAM improves the water

and nutrient status of most medium-textured soils by holding nutrients and

31

supporting beneficial microbial activity in the soil – all these can improve crop

growth and yield.

Most of the published research, however, focuses on using PAM to increase

infiltration (eg. Ben-Hur 2006), whereas my own field observations suggest this is

not a problem on most of the very sandy soils in South Australia – indeed, quite the

opposite is often true. Water movement through coarse sands is so rapid that plants

do not have time to use the applied water and nutrients efficiently. Of particular

concern is the rapid expansion of drip irrigation onto coarse sandy soils in southern

Australia, where lateral spread of water is minimal and where water movement

occurs primarily in a saturated, vertical cylinder of soil directly under the dripper.

Plant response to drip irrigation on very sandy soils has been variable, and water- and

nutrient-use efficiencies have been very low. It was clear from preliminary

observations in the field that ways to reduce the movement of water through soils

were required such that deep percolation of water and nutrients could be stopped.

Very few studies have used low-concentrations of PAM on coarse-textured soils

specifically to reduce deep percolation under irrigation-drippers. Malik and Letey

(1992) showed that application of high-molecular weight PAMs at concentrations in

the range 25 to 400 ppm reduced the hydraulic conductivity of coarse and fine sands,

and attributed this to viscous effects. Falatah et al. (1999) ponded three different

PAMs at concentrations up to 50 ppm on calcareous sands packed in 60 cm-long

tubes, and found that the PAMs reduced the rate at which the wetting front advanced,

but did not influence the total cumulative infiltration. They attributed their findings to

greater viscosity of the solutions, but their viscosities were measured on highly

concentrated PAM-solutions (250,000 to 1,000,000 ppm), which far exceeded

practical considerations. Sivapalan (2006) applied ‘Alcosorb 400’ (a product of Ciba

Specialty Chemicals) at rates of up to 0.2% by weight of dry soil and found that it

increased the length of the irrigation cycle, stored more water and increased soybean

grain yield significantly. The application rates, however, were not considered to be

economical at field scale. As indicated above, very little attention has been placed on

using low concentrations of PAM to achieve significant reductions in infiltration in

sands.

32

Seminal data reported by Ross et al. (1996) showed that PAM increased the lateral

spread of irrigation water in furrow irrigation systems. This work inspired me to

implement some preliminary field trials by injecting PAM into drip irrigation

systems. Initial observations showed that the addition of a PAM at a concentration of

only 1ppm (which is well within the variable cost structures of most horticultural

operations) significantly altered the wetting pattern under the drippers. What was not

understood from these field observations was anything about the magnitude of the

changes in lateral spread, how long such effects could be expected to last, and what

the mechanisms behind the effects might be.

Greater lateral spread of water under drippers in coarse-textured soils requires that

water be retained in the root zone for longer periods of time during irrigation, but

there are few practical ways to retain water in the soil. The aim of the research

reported in this thesis was to evaluate the potential for some commercially available

PAMs to reduce hydraulic movement and to increase water retention on some coarse

sandy soils of South Australia and Victoria, where drip and center-pivot irrigation

systems use enormous quantities of water from the Murray-Darling Basin.

33

CHAPTER 2: USE OF ANIONIC PAMs TO REDUCE INFILTRATION RATES IN

SOME COARSE SANDY SOILS

2.1 Introduction

Without pre-judging the mechanisms involved, it was proposed that addition of

various anionic PAMs to irrigation water would increase the lateral spread of water

(cf. Ross et al. 1996) and that this would occur by reducing the infiltration rate and

thereby allowing a greater proportion of that water to move in directions other than

downward.

Furthermore, some unpublished data provided by H. Ajawa and T. Trout (pers.

Comm.. 2004) showed that infiltration rates and hydraulic conductivities in packed

soil columns decreased with increasing PAM concentration, but that this effect was

influenced by the nature of the water used to dissolve the PAM. Reductions in

infiltration were less pronounced for example when PAM was dissolved in water rich

in calcium sulphate whereas the full effects were displayed when PAM was dissolved

in either pure water or water rich in sodium chloride. While they offered no

explanation for the salt-effects, they attributed the lower hydraulic conductivities to

changes in the apparent viscosity of their PAM-solutions (which they did not

measure).

With these preliminary observations in mind, a series of laboratory column

experiments were conducted to measure infiltration rates in different sands using two

commercially available PAMs in waters of varying quality. Possible mechanisms

involved will be explored in Chapter 3 of this thesis.

2.2 PAM-polymers and their preparation

Two water-soluble PAMs available in powdered form through Ciba Specialty

Chemicals (PAM-135 and PAM-1011) were selected on the basis of their wide

commercial use in southern Australia, and their properties are shown in Table 2.1.

According to the manufacturer specifications, PAM-135 contained significantly less

polyacrylamide than PAM-1011. Stock solutions were made up in reverse-osmosis

(RO) water to produce equivalent quantities of the acrylamide monomer (viz. 2500

34

ppm for PAM-135, and 1000 ppm for PAM-1011). The stock solutions were diluted

when required to produce a range of PAM concentrations used in different

experiments (Table 2.2).

Table 2.1: Properties of the anionic PAM polymers used in this study.

Anionic product

Commercial source & cost per unit weight

Molecular weight

(g / mol)

% acrylate to acrylamide (mol / mol)

Intrinsic viscosity,

[η] (dL / g)

PAM-135 Ciba Specialty

Chemicals, $10/kg 12.5 x 106 35%/65% 18

PAM-1011 Ciba Specialty

Chemicals, $7/kg 17 x 106 20%/80% 22

Table 2.2: PAM concentrations prepared by diluting stock solutions to obtain equivalent concentrations of the monomer.

PAM-135 concentrations (ppm) PAM-1011 concentrations (ppm) 2.5 1 25 10

2.3 Coarse sandy soils

Sands from the Mallee region of South Australia (all Rudosols – Isbell 1996) were

collected from paddocks near Gurrai, Blanchetown, and Overland Corner. These

soils are used extensively for potato production under centre pivot irrigation. The

preference for light textured soils has arisen from the increasing market preference

for clean-washed potatoes with glossy skins.

Two acidic sands were also collected from 0-5 cm and 15-20 cm in a citrus orchard

of Mr Tony Pardo, near Iraak in the Sunraysia region of the north-western part of the

State of Victoria (hereafter referred to as ‘Pardo’).

Two alkaline sands were collected from 0-5 cm and 15-20 cm in the Viognier section

of a vineyard near Lake Cullularaine, Victoria owned by Tandou Wines Pty Ltd

(hereafter referred to as ‘Viognier’).

Finally, a range of screened sand fractions was obtained from Keough’s Ltd in

Adelaide, to evaluate any interactive effects on infiltration between pore size

35

distribution (inferred from mean particle size) and polymer concentrations. All soils

are listed in Table 2.3 along with selected properties.

Table 2.3: Soils (100% sand) used in this study and some of their properties.

Source of the sand Bulk density in

field (g/cm3) pH in 1:5

CaCl2 Organic carbon

content (%) Gurrai, SA 1.35 7.8 < 0.3 Blanchetown, SA 1.28 7.2 < 0.3 Overland Corner, SA 1.35 7.2 < 0.3 Pardo (0-5 cm) 1.35 6.2 0.71 Pardo (15-20 cm) - 4.7 0.22 Viognier (0-5 cm) 1.32 7.5 0.38 Viognier (15-20cm) - 7.8 0.29

2.4 Quality of irrigation water

Water quality on the irrigated sands of southern Australia varies according to the

availability of river-water versus bore water. The concentrations of sodium and

calcium in these sources range widely from virtually zero up to 20 mmol(+)/L in

either cation, so stock solutions were prepared as shown in Table 2.4.

Table 2.4: Concentrations of sodium and calcium in the stock solutions.

Stock solution Concentration of (+) in Na Concentration of (+) in Ca R.O. water 0 0

Na-10 10 mmol(+)/L 0 Na-20 20 mmol(+)/L 0 Ca-10 0 10 mmol(+)/L Ca-20 0 20 mmol(+)/L

Na-Ca SAR 4.6 Na-Ca SAR 6.3

40:60 ratio by volume, 10 mmol(+)/L Na + Ca 40:60 ratio by volume, 20 mmol(+)/L Na + Ca

2.5 Infiltration equipment and procedures

Laboratory infiltration studies were undertaken by packing triplicate samples of each

soil into transparent acrylic cylinders 180 mm long and 25 mm (inside) diameter. The

cylinders were mounted on retort stands under 50 mL burettes containing solutions of

36

varying PAMs, concentrations, and water qualities. A shallow pond of solution

approximately 1 mm thick was established and maintained over the soil surface in

each acrylic cylinder as timed-infiltration occurred. The volume of solution leaving

the burette was measured and the depth of the wetting front was monitored at the

same time until it approached the bottom of the cylinder. The volume of solution

sitting in the shallow pond was subtracted from the total volume delivered (Figure

2.1a)2 and then infiltration (mm) was calculated by dividing the volume infiltrated by

the cross-sectional area of soil in the cylinder.

Infiltration, I (mm), was plotted as a function of time (s) as shown in Figure 2.1b, and

the data were examined for evidence of sorptive behaviour. In most cases, sorptivity

was minimal and steady-state infiltration occurred almost immediately (relatively

straight lines). Nevertheless, all infiltration data were plotted as a function of the

square-root of time (s½) and fitted to Philip’s polynomial: I = C + S(t½) + A(t½)2 ,

where S is the sorptivity (mm s-½), A estimates the hydraulic conductivity in the

transmission zone (referred to as the steady-state infiltration rate, mm s-1), and C is

the I-intercept (mm). To treat all data consistently, the value of C was set equal to

zero, even though, in practice, it often takes a small positive or negative value. An

example of this three-stage process is shown in Figures 2.1a, b and c, and all the raw

infiltration data are stored on a CD in the back cover of this thesis.

Evaluation of two PAMs (Experimental design)

The initial infiltration trials were conducted on the first three soils shown in Table

2.3, using two different PAMs (Table 2.1) prepared at two typical concentrations

with respect to the acrylamide monomer (Table 2.2), each mixed using water of four

typical qualities with respect to sodium and calcium (Table 2.4). In total I used 2

PAMs, 3 PAM-concentrations, 4 water qualities, 3 soils and 3 replications. In the

figures shown below the vertical error lines represent ± standard errors of the mean

steady-state infiltration rate, 1n

s2

−.

2 The intercept of a plot of volume delivered (mL) against the downward progression of the wetting front (mm) was determined. In a few cases, the intercept seemed unrealistically large, so the entire data set was discarded.

37

Figure 2.1a. Illustration of how infiltration volumes were corrected. The intercept of each line from a least-squares regression represents the volume of water that formed the shallow ‘pond’ over each column of soil. For example, in the first of the three replicated examples shown above, 0.9 mL (actually 0.9257 mL) was subtracted from all Rep-1-values to obtain the corrected volumes of infiltration.

Figure 2.1b. Illustration of infiltration (mm) calculated from Figure 2.1a and plotted as a function of time (s).

Volume delivered by burette during advance of wetting front

(Example)

Volume1 = 0.3666Depth + 0.9257

Volume2 = 0.3889Depth + 5.3554

Volume3 = 0.3363Depth + 3.8811

0

5

10

15

20

25

30

35

40

45

50

0 25 50 75 100 125

Depth of wetting front (mm)

Vo

lum

e d

eliv

ered

(mL

)

Rep 1

Rep 2

Rep 3

Infiltration as a function of time(Example)

0

10

20

30

40

50

60

0 20 40 60 80 100 120 140

Time (s)

Infil

trat

ion

(mm

)

Rep 1

Rep 2

Rep 3

38

Figure 2.1c. Illustration of how the three sets of infiltration data from Figure 2.1b were re-plotted as a function of the square-root of time, and fitted to the Philip polynomial. In the examples shown here, the steady-state infiltration rates (A-values) are shown to be 0.4573, 0.2323 and 0.1413 mm/s for Reps 1, 2 and 3 respectively.

2.6 Results and discussion

2.6.1 Overall effectiveness of two polyacrylamides (PAM-1011 and PAM-135) at

reducing steady-state infiltration rate in three sands.

The overall mean steady-state infiltration rate for all three soils examined initially

(i.e. for soils from Gurrai, Blanchetown, and Overland Corner) was significantly

reduced by application of the polymer, PAM-1011 (approximately 30%), but not by

application of the polymer PAM-135 (Figure 2.2a).

Figure 2.2a. Mean steady-state infiltration rates for 2 polymers at all concentrations relative to water of all qualities. Vertical lines on bars are standard errors.

Infiltration as a function of square-root of time(Example)

I-1 = 0.7782(Time1/2) + 0.4573(Time1/2)2 + R2 = 0.9963

I-2 = 2.8448(Time1/2) + 0.2323(Time1/2)2 + R2 = 0.9661

I-3 = 2.2819(Time1/2) + 0.1413(Time1/2)2 R2 = 0.9807

0

10

20

30

40

50

60

70

80

0 2 4 6 8 10 12 14

Time1/2 (s1/2)

Infil

trat

ion

(mm

)

Rep 1

Rep 2

Rep 3

0.00

0.05

0.10

0.15

0.20

0.25

0.30

no-PAM PAM-1011 PAM-135

PAM-polymer in water

Mea

n s

tead

y-st

ate

infil

trat

ion

rat

e (m

m/s

)

39

As might be expected from soils with different properties, the steady-state infiltration

rates varied somewhat among the three soils (Figure 2.2b). The reasons for the

variation in infiltration behaviour, including differences in particle size distribution,

and differences in wettability (i.e. degree of hydrophobicity) are explored in Section

2.6.2. For now, I will simply report the particle size distribution as measured on

samples collected in the field at each site (Table 2.5 and Figures 2.3a and 2.3b).

Figure 2.2b. Mean steady-state infiltration rates from Figure 2.2a, for the three soils used in the preliminary study. Vertical lines on the bars are standard errors.

Table 2.5. Percentage of sample mass less than the arithmetic mean particle diameter for the soils used in the infiltration work.

Arithmetic mean diameter (mm) Soil name

< 0.053 < 0.072 < 0.17 < 0.38 < 0.75 < 1.5 < 2.0 Geometric mean diameter (mm)

Gurrai 0 1 3 60 94 99 100 0.280 Blanchetown 0 2 3 57 99 100 100 0.258 Overland Corner 0 2 7 53 97 100 100 0.273 Pardo 0-5cm 0 2 5 39 81 94 100 0.411 Pardo 15-20cm 0 1 3 48 90 96 100 0.340 Viognier 0-5cm 0 2 10 47 77 86 100 0.457 Viognier 15-20cm 0 2 7 46 89 96 100 0.347

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

no-PAM PAM-1011 PAM-135

Mea

n s

tead

y-st

ate

infi

ltra

tio

n r

ate

(mm

/s)

GurraiBlanchetownOverland Corner

40

Figure 2.3a. Cumulative particle size distributions for the first three soils examined (left) and the second group of soils (right) used in this study.

Figure 2.3b. Geometric mean diameters of the seven soils used in this study.

Application of PAM-135 seemed to have little or no impact on water movement

regardless of the concentration applied in this study (Figure 2.4). Further work with

this polymer was therefore discontinued. All subsequent results in this study refer

only to the polymer PAM-1011.

0

10

20

30

40

50

60

70

80

90

100

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00

Arithmetic mean particle diameter (mm)

% s

ampl

e <

giv

en s

ize

Gurrai

Blanchetown

Overland Corner

0

10

20

30

40

50

60

70

80

90

100

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00

Arithmetic mean particle diameter (mm)

% s

ampl

e <

giv

en s

ize

Viognier 0-5cm

Viognier 15-20cm

Pardo 0-5cm

Pardo 15-20cm

0.258 0.273 0.280

0.340 0.347

0.4110.457

0.00

0.10

0.20

0.30

0.40

0.50

Blanch

etow

n

Overla

nd C

orne

r

Gurra

i

Pardo

15-

20cm

Viognie

r 15-

20cm

Pardo

0-5

cm

Viognie

r 0-5

cm

Ge

om

etri

c M

ea

n D

iam

ete

r (m

m)

41

Figure 2.4. Effect of concentration of PAM-135 on steady-state infiltration rate of three sandy soils. Vertical lines on the bars are standard errors.

2.6.2 Effectiveness of PAM-1011 at reducing steady-state infiltration rate:

influence of particle size

The rather wide range of steady-state infiltration rates for the three different soils

used in Section 2.6.1 suggested that either non-wettability (water repellence) or

varying particle size distributions (or both) influenced the results. To determine the

extent to which the effectiveness of PAM-1011 may be limited by the particle size

distribution of sands to which it is applied, I measured ponded-infiltration

experiments on six different size-fractions of sand (0.1, 0.3, 0.4, 0.5, 1.0, 12.7 mm –

available commercially from Keough’s Sands, Adelaide) using PAM-1011 at three

different concentrations (0, 1, 10 ppm). Procedures were identical to those described

above, except that for the three coarsest fractions (0.5, 1, and 12.7 mm), infiltration

occurred so quickly that very few points could be measured before the 50-mL burette

emptied each time. In these cases, the slope of a straight line (fitted through the

available points) was calculated as the steady-state infiltration rate.

The steady-state infiltration rate increased with increasing mean size of sand particles

(Figure 2.5)3. The greatest reductions in infiltration rate occurred with a PAM-1011

3 This suggests particle size may not have been as important in the initial work described for the Gurrai, Blanchetown and Overland Corner samples above. The Gurrai soil was coarser than either Blanchetown or Overland Corner so its steady state infiltration rate should have been greater, yet it was only half that for Blanchetown and Overland Corner.

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0 ppm 2.5 ppm 25 ppm

Concentration of PAM-135

Mea

n s

tead

y-st

ate

infil

trat

ion

rat

e (m

m/s

)

Gurrai

Blanchetown

Overland Corner

42

application rate of only 1 ppm and little further reduction occurred with the higher

concentration (10 ppm). Reductions expressed as a proportion of the No-PAM

infiltration rate declined with increasing particle diameter regardless of PAM-

concentration (Figure 2.6) although reductions were still quite significant in the

largest fraction examined in this study. The persistent effect of the PAM even in the

12.7 mm gravel indicates that significant reductions in steady-state infiltration can be

achieved across a broad range of coarse sandy soils.

Figure 2.5. Effect of mean particle size on effectiveness of PAM-1011 at three different concentrations to reduce steady-state infiltration rate. Figure 2.6. Percent reduction in effectiveness of PAM-1011 with particle size.

%Reduction1ppm = -0.0646Ln(Size) + 0.4094

R2 = 0.3319

%Reduction10ppm = -0.0808Ln(Size) + 0.4917

R2 = 0.2781

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0.1 1.0 10.0 100.0Mean particle size (mm)

Per

cen

t red

uct

ion

in s

tead

y-st

ate

infil

trat

ion

rat

e 1 ppm

10 ppm

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

0.1 0.3 0.4 0.5 1.0 12.7

Mean particle size (mm)

Mea

n s

tead

y-st

ate

infil

trat

ion

rat

e (m

m/s

)

0 ppm (RO water)

1 ppm

10 ppm

43

2.6.3 Effectiveness of PAM-1011 at reducing steady-state infiltration rate in a

range of different sandy soils.

The work reported in Sections 2.6.1 and 2.6.2 suggested PAM-1011 can be highly

effective in reducing steady-state infiltration rate of various uniform sand fractions

sandy soils. This needed wider testing, however, and I was particularly interested in

the wetting behaviour of some subsoils having somewhat coarser particle size

distributions in the sand range. I thus chose soil samples from the surface and

subsurface layers of two commercial horticultural properties in northern Victoria (i.e.

Viognier and Pardo – see Tables 2.3 and 2.5). Procedures for measuring steady-state

infiltration as affected by PAM-1011 were identical to those described in Section 2.5,

except that a range of different salt concentrations in sodium, calcium and both salts

was used to prepare the solutions of PAM (Table 2.4). The different salts and salt

concentrations were included to reflect the wide range of water quality available to

irrigators on commercial properties.

Overall effect of PAM-1011

Almost invariably, the application of PAM-1011 at both 1 and 10 ppm reduced the

steady-state infiltration rates for all seven soils examined (Figure 2.7).

Figure 2.7. Effect of application rate of PAM-1011 on reducing steady-state infiltration rate in a range of sandy soils having different particle size distributions.

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0 ppm 1 ppm 10 ppm

Concentration of PAM-1011

Mea

n s

tead

y-st

ate

infil

trat

ion

rat

e (m

m/s

)

Blanchetown

Overland Corner

Gurrai

Pardo 0-5 cm

Pardo 15-20 cm

Viognier 0-5 cm

Viognier 15-20 cm

44

Effectiveness of PAM-1011: water quality – total salt content

As indicated above, the quality of water available to irrigators in South Australia

varies enormously, particularly in relation to the total salt concentration. The

question arose as to whether the effectiveness of polyacrylamides might vary

depending on the water source used to prepare the PAM solutions during irrigation. It

was hypothesised that higher salt concentrations might cause the polymer to

flocculate more significantly and thus be less effective in reducing infiltration rates.

Salt solutions were thus prepared at concentrations of 0, 10 and 20 mmol(+)/L in

sodium only, calcium only, and a mixture of both sodium and calcium. In this section

I present results on the overall effect of salt concentration; effects of individual salts

are evaluated separately.

The effectiveness of PAM-1011 in reducing steady-state infiltration rates diminished

in the presence of salt for all soils examined here (Figure 2.8). Infiltration rates either

remained constant or they increased with increasing salt concentration. The reduced

effectiveness of the polymer at higher salt concentrations could be interpreted as a

salt-induced flocculation effect, which made the PAM less viscous (i.e. made it

occupy a smaller volume per unit mass). However, salt x polymer x concentration

interactions may have occurred, and these are explored in the next section.

Figure 2.8. Overall mean steady-state infiltration rate of PAM-1011 in seven soils as affected by salt concentration in the water used to prepare the solutions.

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0 10 20

Water quality (salt concentration, mmol/L)

Mea

n s

tead

y-st

ate

infil

trat

ion

rat

e (m

m/s

)

Blanchetown

Overland Corner

Gurrai

Pardo 0-5 cm

Pardo 15-20 cm

Viognier 0-5 cm

Viognier 15-20 cm

45

Effectiveness of PAM-1011: water quality – type and concentration of salt

An overall comparison of Figures 2.9a and 2.9b for the three soils involved in the

initial experiments (Blanchetown, Overland Corner, and Gurrai) confirms that greater

concentrations of PAM-1011 reduced the steady-state infiltration rate in almost all

cases relative to the No-PAM control.

Figure 2.9a. Effect of 1 ppm PAM-1011 in irrigation water of varying salinity & sodicity on steady-state infiltration rate in three sandy soils.

Figure 2.9b. Effect of 10 ppm PAM-1011 in irrigation water of varying salinity & sodicity on steady-state infiltration rate in three sandy soils.

1 ppm PAM-1011

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

No PAM(control)

0 salt SAR 0 10 mmol(+)Ca/L SAR 0

10 mmol(+)Na-Ca/L SAR

4.2

10 mmol(+)Na/L SAR

infinite

20 mmol(+)Ca/L SAR 0

20 mmol(+)Na-Ca/L SAR

6.3

20 mmol(+)Na/L SAR

infinite

Ste

ady-

stat

e in

filt

rati

on

rat

e (m

m/s

)

Blanchetow n

Overland Corner

Gurrai

10 ppm PAM-1011

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

No PAM(control)

0 salt SAR 0 10 mmol(+)Ca/L SAR 0

10 mmol(+)Na-Ca/L SAR

4.2

10 mmol(+)Na/L SAR

infinite

20 mmol(+)Ca/L SAR 0

20 mmol(+)Na-Ca/L SAR

6.3

20 mmol(+)Na/L SAR

infinite

Ste

ady-

stat

e in

filt

rati

on

rat

e (m

m/s

)

Blanchetow n

Overland Corner

Gurrai

46

Secondly, modest salinity levels in the irrigation water (i.e. 10 mmol(+)/L in either

sodium or calcium or both salts) had no significant impact on the efficacy of PAM-

1011 to reduce steady-state infiltration rates relative to the No-PAM water. The

higher salinity water (i.e. 20 mmol(+)/L in either sodium or calcium or both salts),

however, significantly decreased the ability of PAM-1011 to reduce infiltration rates.

In fact, with the exception of the Gurrai soil, the infiltration rates for the 20

mmol(+)/L waters (regardless of their sodicity) were pretty much the same as the No-

PAM waters. This result was especially pronounced in the 1 ppm PAM treatments,

which suggests that when an irrigator uses very salty water, s/he will have to apply

PAM at higher rates to achieve the same results as those using low-salt water.

In the other four soils (Viognier surface and subsoil sands, Pardo surface and subsoil

sands) the trends were similar (Figures 2.10a and 2.10b), although the infiltration

experiments at high salinity (i.e. 20 mmol(+) salt/L) were not performed. Application

of PAM-1011 in water of modest salt concentration (i.e. 10 mmol(+) salt/L) had little

effect in most cases on infiltration rates, with one exception (Pardo subsoil when

calcium was the salt involved). For all soils, however, application of PAM-1011 at

10 ppm significantly reduced steady-state infiltration rates relative to those at 1 ppm

regardless of salt and salt concentration.

Figure 2.10a. Effect of 1 ppm PAM-1011 in irrigation water of varying salinity and sodicity on steady-state infiltration rate in Viognier- and Pardo- surface- and subsoils.

1 ppm PAM-1011

0.00

0.05

0.10

0.15

0.20

0.25

No PAM 1 ppm - 0 salt 1 ppm - 10 mmol(+) Na/L 1 ppm - 10 mmol(+) Ca/L

Mea

n s

tead

y-st

ate

infil

trat

ion

rat

e (m

m/s

)

Viognier (0-5 cm)Viognier (15-20 cm)Pardo (0-5 cm)Pardo (15-20 cm)

47

Figure 2.10b. Effect of 10 ppm PAM-1011 in irrigation water of varying quality (salinity and sodicity) on the steady-state infiltration rate in Viognier- and Pardo- surface- and subsoils.

2.7 Conclusions

To moderate the downward percolation of irrigation water into coarse sandy soils,

application of the polyacrylamide “PAM-1011” at rather low concentrations appears

sufficient to reduce steady-state infiltration rates in columns of sand used in

commercial horticultural production in South Australia and Victoria. Importantly, the

two polymers examined in this work varied in their efficacy for the purpose. By

comparison to PAM-1011, the product PAM-135 was completely ineffective at

similar concentrations of the monomer. In practical terms this means “PAM ain’t just

PAM” – myriad industrial-scale formulations are commercially available, so it is

crucial that the properties of any product be tested before using in the field. While

PAM-1011 and PAM-135 are both relatively high-molecular-weight anionic

polymers, Figure 1.1 shows PAM-1011 is produced as a gel of slightly higher

molecular weight and lower anionic content, whereas PAM-135 is produced as an

inverse emulsion of somewhat lower molecular weight and higher anionic content. It

is also possible that PAM-135 contained a surfactant, but I was unable to obtain this

10 ppm PAM-1011

0.00

0.05

0.10

0.15

0.20

0.25

No PAM 10 ppm - 0 salt 10 ppm - 10 mmol(+) Na/L 10 ppm - 10 mmol(+) Ca/L

Mea

n s

tead

y-st

ate

infi

ltra

tio

n r

ate

(mm

/s)

Viognier (0-5 cm)

Viognier (15-20 cm)

Pardo (0-5 cm)

Pardo (15-20 cm)

48

commercially confidential information from the supplier. The significance of these

differences in terms of the mechanisms causing attenuation of infiltration requires

further evaluation (Chapter 3).

PAM-1011 is effective at relatively modest concentrations in irrigation water.

Significant reductions in steady-state infiltration rate were achieved in some soils

using only 1 ppm - even greater reductions were achieved with 10 ppm. With very

uniform sand fractions, the effectiveness of PAM-1011 declines as particle size

increases but it even had a moderating impact in the gravel. This implies that higher

concentrations of PAM may be needed on very coarse sands.

The quality of irrigation water appears to have some influence on the effectiveness of

PAM-1011. If highly saline water is used, greater concentrations of PAM-1011 must

be dissolved in the water to produce the same result as that obtained using less saline

water. While the precise chemistry behind this was not investigated in this work, a

relevant hypothesis would be that the flocculating effect of salty water causes the

polymer to conform upon itself and thus make the solution less viscous. If this

hypothesis were shown to be true, measurement of the viscosity might act as a

practical tool to select suitable polymers for commercial application.

49

CHAPTER 3: EVALUATION OF WHY PAM REDUCED STEADY-STATE

INFILTRATION RATES IN COARSE SANDS 3.1 Introduction The work reported in Chapter 2 suggested the factors responsible for reducing steady-

state infiltration rates by PAM-1011 were not necessarily straightforward. This was

complicated by the fact that the infiltration behaviour of the natural sands collected

from the field differed significantly from that of sieved sand-fractions purchased

from Keough’s Ltd. The sieved sand-fractions behaved as one might expect: the

steady-state infiltration rate increased with particle size (and thus pore size). For the

natural sands, however, infiltration rate declined with increasing particle size across a

similar size range (Figure 3.1a), regardless of PAM application rate (Figure 3.1b).

Figure 3.1a Mean steady-state infiltration rate of water as a function of particle size for various sieved sand fractions (triangles) and natural sands from the field (circles).

Figure 3.1b Mean steady-state infiltration rate as a function of same particle sizes shown in Figure 3.1a for water treated with PAM-1011 at 1 and 10 ppm.

50

This observation suggested that particle size distribution alone (and by implication,

pore size distribution) did not control the infiltration behaviour entirely.

A range of factors was likely involved, one of which might have been the degree of

non-wetting due to water repellence – a problem widely known to occur in coarse

sands, causing poor infiltration, uneven wetting in the root zone, and serious soil

erosion (eg. van Dam et al. 1990). Another possible factor was that the PAM altered

the pore size distribution of some of the sands and thus their water-retention & -

transmission properties. Sivapalan (2006), for example, added a neutral acrylic

copolymer (Alcosorb 400) to a siliceous sand and found water contents at field

capacity increased by 23% and 95% for application rates of 0.03 and 0.07%

respectively. He also found that water contents were also higher at permanent wilting

point, which suggests that PAM generated smaller pores in the soil, which could

reduce the rate of water movement. It was also possible that the PAM simply

increased the kinematic viscosity of the infiltrating water without causing any change

in the pore-size distribution or the wettability of the soils. Finally a combination of

the factors outlined above may have been responsible. Experiments to evaluate these

possibilities are presented here.

3.2 Effect of PAM on water repellence in sands.

3.2.1 Introduction

Water repellence in Australian soils is generally caused by a waxy coating on the

sand grains, which makes them hydrophobic at the soil surface as well as at depth

(Ma’shum et al. 1988). It is particularly common in sandy soils after long dry periods

(De Bano 2003). Non-wetting under drip irrigation is a significant problem, which

prevents accurate control of water distribution at the soil surface as well as in the

rootzone, where water can follow preferential flow paths and completely bypass the

rootzone – leading to deep percolation of water and nutrients, and poor water use

efficiency by crops.

The behaviour of the sands described in Figures 3.1a and b would suggest that if

hydrophobicity were a significant problem, it should have caused the greatest

reductions in infiltration rates for the finer sands than for the coarser ones because De

51

Jonge et al. (1999) found that water repellence is generally worst in finer sands.

Clearly, this was not the case in Figures 3.1a and b. Nevertheless, the investigation

described here proceeded with a view to potentially discounting the non-wetting

phenomenon as a significant factor reducing infiltration rates with and without PAM.

3.2.2 Materials and methods

Five-gram samples of each soil (Blanchetown, Gurrai, Overland Corner, Pardo 0-

5cm, and Viognier 0-5 cm, plus a sample of extremely water repellent sand from

Western Flat, SA) were placed in Petri dishes and replicated drops of various

solutions (Table 3.1) were placed on the soil and the average time required for

complete absorption was recorded using a modified version of the Water Droplet

Penetration Time (WDPT) method (Doerr 1998). A classification of the severity of

water repellence according to this method is shown in Table 3.2.

Table 3.1 Solutions used as droplets to determine the WDPT for each sample.

RO water 10 mmol(+)Na/L 10 mmol(+) Ca/L Control

1 ppm PAM-1011 10 ppm PAM-1011

Control 1 ppm PAM-1011 10 ppm PAM-1011

Control 1 ppm PAM-1011 10 ppm PAM-1011

Table 3.2 Classes of water repellence proposed by Lal and Shukla (2004).

Water Droplet Penetration Time (s)

Water repellence class Water repellence

description <5 0 Non-repellent

5-60 1 Slightly repellent 60-600 2 Strongly repellent

600-3600 3 Severely repellent >3600 4 Extremely repellent

3.2.3 Results and Discussion

The Droplet Penetration Time for water (distilled, 10 mmol(+)Na/L, 10

mmol(+)Ca/L) on each soil sample is shown in Table 3.3a in comparison with that

for Western Flat sand. The Droplet Penetration Times for solutions of 1 and 10 ppm

PAM-1011 in distilled water, 10 mmol(+) Na/L, and 10 mmol(+) Ca/L on each soil

sample are shown in Tables 3.3b and 3.3c. Clearly, in comparison with the Droplet

Penetration Times for Western Flat Sand, none of the soils used in my study could be

52

considered to be water repellent. My soils were all in Class “0” (Non-repellent) – the

Gurrai soil is the only one that even approached Class “1” (Slightly repellent) in Lal

and Shukla’s (2004) classification, and this only when solutions of 10 ppm PAM-

1011 were used as the penetrating droplet. Water quality had no significant impact on

this results. Only an unrealistically high concentration of 1000 ppm PAM-1011 had

any significant impact on wettability of the soils (Table 3.4).

Table 3.3a Water Droplet Penetration Times (seconds) for solutions of different salt concentration for some surface soils used in this study.

Water Soil

Distilled water 10 mmol(+) Na / L 10 mmol(+) Ca / L Gurrai 1.5 1 1 Blanchetown < 1 < 1 < 1 Overland Corner < 1 < 1 < 1 Pardo (0-5 cm) < 1 < 1 < 1 Viognier (0-5 cm) < 1 < 1 < 1 Western Flat Sand 1095 - - Table 3.3b Droplet Penetration Times (seconds) for solutions of 1 ppm PAM-1011 of different salt concentration for some surface soils used in this study.

1 ppm PAM-1011 Soil

Distilled water 10 mmol(+) Na / L 10 mmol(+) Ca / L Gurrai 1.5 1 1 Blanchetown < 1 < 1 < 1 Overland Corner < 1 < 1 < 1 Pardo (0-5 cm) < 1 < 1 < 1 Viognier (0-5 cm) < 1 < 1 < 1 Western Flat Sand 945 - - Table 3.3c Droplet Penetration Times (seconds) for solutions of 10 ppm PAM-1011 of different salt concentration for some surface soils used in this study.

10 ppm PAM-1011 Soil

Distilled water 10 mmol(+) Na / L 10 mmol(+) Ca / L Gurrai 4.5 4 2 Blanchetown 1 1 < 1 Overland Corner 1.5 1.5 < 1 Pardo (0-5 cm) 1 1 1 Viognier (0-5 cm) 1 1 1 Western Flat Sand 925 - -

53

Table 3.4. Droplet Penetration Times (seconds) for solutions of 1000 ppm PAM-1011 in distilled water for some surface soils used in this study.

Soil Droplet Penetration Time (seconds) Overland Corner >2100

Blanchtown >2100 Gurrai 740

Viognier >2100 Pardo >2100

3.2.4 Conclusions

Only very high concentrations of PAM-1011 (i.e. 1000 ppm) had any serious impact

on wettability of the soils used in this study – droplets of water with 1 and 10 ppm

PAM-1011 regardless of water quality had virtually no effect. This suggests that

contact angle and surface tension effects played a negligible or only a very small part

in the reductions in steady-state infiltration rates reported in Chapter 2.

Had experiments been conducted using a range of PAM-1011 concentrations

between 10 and 1000 ppm (say between 10 and 100 ppm), some effects might have

been found. However, 10 ppm is pretty close to the upper limit of economic

relevance for most irrigated crops, so I saw little benefit in pursuing experiments to

resolve this question.

3.3 Effect of PAM on pore-size distribution of sand from Overland Corner as

shown by water retention

3.3.1 Introduction

As indicated above, the work of Sivapalan (2006) suggested that addition of an

anionic PAM increased the quantity of water held in very small pores (i.e. water

retention increased by the same amount between matric suctions of 10 and 1500 kPa

and thus did not increase the amount of plant available water). On this basis it was

felt that results obtained from one of the finer sands used in this study were likely to

be indicative of what would happen on most other sands. Thus only one soil – the

Overland Corner sample – was used in this investigation.

54

3.3.2 Materials and Methods

Samples of Overland Corner sand were dried and treated with the solutions shown in

Table 3.5, which are typical of what this soil would be exposed to in the field.

Duplicate samples were placed in rings on one of several ceramic pressure plates and

wetted with the appropriate solutions. Water retention after one month was measured

in pressure chambers set at the following pressures: 10, 30, 45, 80, 100, 1500 kPa.

Table 3.5 Pre-treatment of samples for measurement of water retention. Pre-treatment Distilled water only � water retention 10 ppm PAM-1011 in distilled water � water retention 10 ppm PAM-1011 in distilled water, dried, re-treat with PAM � water retention 10 ppm PAM-1011 in 20 mmol(+)Na / L � water retention 10 ppm PAM-1011 in 20 mmol(+)Na/ L, dried, re-treat with PAM � water retention

3.3.3 Results and discussion

The average effect on water retention of adding PAM-1011 is shown in Figure 3.2.

Variability was low (vertical bars, where visible, represent standard deviations of the

mean water content), but the only significant difference is shown at 10 kPa suction.

Figure 3.2 Water retention curves for Overland Corner with and without addition of PAM-1011.

0.015

0.025

0.035

0.045

0.055

0.065

0.075

10 100 1000 10000

Matric suction (kPa)

Gra

vim

etric

wat

er c

onte

nt (g

/g)

Water only

PAM-1011 added

55

Examining the water retention data according to whether or not the samples were

pre-treated with PAM-1011also shows very little difference except at 10 kPa suction

(Figure 3.3). No differences in water retention were found at higher suctions.

Figure 3.3 Water retention curves for Overland Corner with and without pre-treatment with PAM-1011.

The results confirm that water retention in this sand (and other coarse sands) is

minimal – at ‘field capacity’ (10 kPa) the gravimetric water content is only between

5.5 and 7.0 g/g, and that most of the water is depleted by a suction of only 45 kPa.

They also highlight the perennial problem of having to irrigate coarse sandy soils

almost continuously to keep the matric suction about 50 kPa and avoid problems

with water stress in most horticultural crops.

3.3.4 Conclusions

The addition of PAM appears to have little impact on any of the water retention

characteristics of this fine sandy soil, except perhaps at the very wet end, where it

reduced water retention by a small amount (approximately 0.01 g/g). It is clear from

this work, that PAM-1011 played only a minor role, if any, in reducing the steady-

state infiltration rates observed in Chapter 2, at least for the Overland Corner soil. It

0.015

0.025

0.035

0.045

0.055

0.065

0.075

0.085

10 100 1000 10000Matric suction (kPa)

Gra

vim

etric

wat

er c

onte

nt (

g/g)

Water only

Pre-wet with PAM-1011

Direct application of PAM-1011

56

is possible that PAM-1011 might have altered the pore-size distribution of the other

soils, but only for the coarser Pardo and Viognier soils – the Gurrai and Blanchetown

soils had very similar particle size distributions to the one from Overland Corner

(see Table 2.5 and Figure 2.3b).

The fact that PAM did not increase soil water retention while field experience shows

higher water contents using PAMs indicates that other factors controlling water

infiltration and distribution need to be considered.

3.4 Kinematic viscosity of solutions of PAM-1011

3.4.1 Introduction

The lack of any major effects of PAM on either wettability or pore-size distribution

of soils suggested that PAM probably causes minimal changes to the properties of the

soils to which it is applied. In light of the work of Malik and Letey (1992) and of

Falatah et al. (1999), it is more likely that the properties of the irrigating fluid itself

change when PAM is added. The most likely property to change is the viscosity. This

study evaluates the effect of PAM-1011 concentration and quality of water in which

it is mixed on the kinematic viscosity of the irrigating solutions.

3.4.2 Materials and methods

The kinematic viscosity, η, of solutions of PAM-1011 at 0, 1 and 10 ppm mixed in

distilled water, 10 mmol(+)Na/L and 10 mmol(+)Ca/L was measured using an

Ubbelohde Suspended Level Capillary Viscometer4. The measurement is based on

Poiseulle’s law, which by rearrangement yields:

η = LQ8

PR4π,

where R is the radius of the capillary (0.001 m), L is the length of the capillary (0.3

m), Q is the volumetric flow rate through the capillary (m3/s), and P is the pressure

head forcing the liquid through the capillary (in this case, the hydrostatic pressure of

the solutions, m). The viscometer reported results in units of cStokes, which are

converted to SI units (m2/s) by multiplying x 10-6. Solutions were made up and

4 The viscometer, loaned to me by Ciba Specialty Chemicals, was a Cannon BS/IP/LS-1 (Product code 9724-F10).

57

equilibrated at 25 C in a water bath prior to using the viscometer. The efflux-time for

each PAM-solution was measured in comparison with that for its solvent (water or

salt solution).

3.4.3 Results and discussion

The addition of PAM-1011 increased the kinematic viscosity of all the solutions in

comparison to distilled water, regardless of the quality of the solvent used (Table

3.6). The extent of the increase in kinematic viscosity due to PAM-1011, however,

was diminished somewhat by the presence of both sodium and calcium salts (Figure

3.4). This is largely due to ionic effects on the polymer chain; as predicted in Section

2.6, the divalent cation, calcium, had the greatest inhibitory effect on the viscosity,

probably because it caused the polymer to conform upon itself and thereby occupy

less volume per unit mass. The monovalent salt reduced the viscosity to a lesser

extent because its interactions with the polymer chain are weaker.

Kinematic viscosity, η, Distilled water 10 mmol(+) Na/L 10 mmol(+) Ca/L

PAM-1011 concentration

(ppm) cStokes m2 s-1 cStokes m2 s-1 cStokes m2 s-1

0 0.920 9.20 x 10-7 0.918 9.18

x 10-7 0.917 9.17 x 10-7

1 0.944 9.44 x 10-7 0.922 9.22

x 10-7 0.921 9.21 x 10-7

10 1.119 1.119 x 10-6 0.986 9.86

x 10-7 0.944 9.44 x 10-7

Table 3.6 Kinematic viscosity of irrigating solutions as influenced by PAM-1011 concentration and type of salt in the solvent.

Figure 3.4 Effect of PAM-1011 and solvent cation on solution kinematic viscosity.

0.80

0.85

0.90

0.95

1.00

1.05

1.10

1.15

0 5 10Concentration of PAM-1011

Kin

emat

ic v

isco

sity

(cS

toke

) Distilled water10 mmol(+) Na/L10 mmol(+) Ca/L

58

3.4.4 Conclusions

The addition of increasing concentrations of PAM-1011 to water increases its

kinematic viscosity. The extent of this increase is tempered by the presence of

soluble salts, and the order of the effects is consistent with the ionic effects caused by

monovalent versus divalent cations – the interference by calcium was greater than

that caused by sodium. This would appear to be the most significant effect of all the

factors considered in the investigations outlined here in Chapter 3.

3.5 Overall conclusions on why PAM-1011 reduced infiltration rates

Although the work outlined in this Chapter does not present an unequivocal

explanation of Figures 3.1a and b, it is possible among the six natural soils examined

in this work (whose geometric mean diameters were all within 0.2 mm of one

another) that variable particle-shape and inter-particle packing accounted for the

unusual hydrodynamic behaviour shown. Given the variation in the mean steady state

infiltration rates across this range, it might simply be coincidental that the linear

regression through the points showed a negative slope with increasing geometric

mean diameter. Much ado about nothing perhaps?

Changes in water repellence and pore size distribution due to addition of PAM-1011

to coarse-textured soils played only minor roles in moderating steady-state

infiltration rates observed in Chapter 2. By far the greatest PAM-effect on water

movement was the increase in viscosity of the water to which it was added. The

fluids having greater viscosity were simply unable to move through the soil pores at

the same rate that water could. Of particular relevance in explaining why poor quality

irrigation water diminished the effects of PAM-1011 to reduce infiltration rates (cf.

Figures 2. 9a and b) was the effect of sodium and calcium on viscosity (Figure 3.4).

The calcium-PAM-1011 solutions were not as viscous as the sodium-PAM-1011

solutions. This indicates that when using lower quality irrigation water (i.e. high

salinity), a higher concentration of PAM-1011 is required to gain its full benefit of

higher viscosities that it creates.

59

CHAPTER 4: VISUAL ASSESSMENT OF THE INFLUENCE OF PAM ON

WATER MOVEMENT AND DISTRIBUTION THROUGH SANDS

4.1 Introduction

The work outlined in Chapters 2 and 3 indicated that by treating irrigation water with

PAM-1011 it became more viscous and thus able to reduce steady-state infiltration

rates in sandy soils. The fate of the irrigation water in terms of its lateral distribution,

however, was not investigated, and has particular relevance to the rapid conversion

from furrow-irrigation to drip-irrigation on sandy soils across South Australia and

Victoria. The question arose as to whether a measurable change in the lateral spread

of irrigation water from a point source could be observed after application of PAM-

1011 to the water. Greater lateral spread within the root-zone would allow greater

uptake of water by plants and therefore less opportunity for deep percolation below

the root-zone.

It was hypothesized here that smaller infiltration rates would allow greater lateral

spread of point-source irrigation water. Such a phenomenon would explain why

greater water use efficiencies have been observed in the field when using PAMs on

coarse textured soils (cf. Section 1.1).

A simple, visual approach was taken in this work, inspired by the early time-lapse

photography used by W.H. Gardner (1962), whereby the vertical and lateral

movement of irrigation water in soils was monitored through transparent, Perspex

sheets over time.

4.2 Materials and methods

Three pairs of rigidly-held transparent Perspex sheets (60 cm deep, 55 cm wide and 4

cm thick) were clamped together on retort stands in the laboratory and filled with

equal masses of the Viognier sand (0-5 cm). Each apparatus was marked on the side

with a scale in centimetres, then placed under a commercial dripper attached to a

water pump calibrated to deliver 2 L/h from the centre-top of the soil surface5. The

5 2 L/h is typical of drip irrigations now used for many perennial crops in South Australia

60

solutions used in these experiments were the same as those used in Chapter 2, and are

shown in Table 4.1.

Table 4.1 Solutions used to observe wetting behaviour in Viognier (0-5cm) sand.

Irrigation solutions Distilled water Distilled water + 1 ppm PAM-1011 Distilled water + 10 ppm PAM-1011 10 mmol(+) Na/L 10 mmol(+) Na/L + 1 ppm PAM-1011 10 mmol(+) Na/L + 10 ppm PAM-1011 10 mmol(+) Ca/L 10 mmol(+) Ca/L + 1 ppm PAM-1011 10 mmol(+) Ca/L + 10 ppm PAM-1011

Each of the three apparatuses was assigned to one of the solvents (either distilled

water, 10 mmol(+) Na/L or 10 mmol(+) Ca/L), which meant that all experiments for

a given solution were conducted sequentially in the same apparatus. The wetted sand,

of course, was replaced with an equal quantity of dry sand each time. Lines were

drawn onto the Perspex sheets using a wax-crayon to identify the extent of wetting

(maximum depth and width) at the end of the 30-minute period (or at various

intervals over the 30-minute period). The experiment was then repeated with a

different solution (Figures 4.1a, b, c). The position of all the marked wetting fronts

was then photographed. The average maximum length and width of wetting was then

measured and plotted as a function of time. While this approach produced only two-

dimensional results and while no replication of the work was conducted, I considered

the results to be indicative of field expectations.

4.3 Results and discussion

The movement of water to the extreme edges of the apparatus after about 30minutes

governed the length of time the experiments could run. While a wider model could

give a better impression of water movement over longer periods, difficulties with

controlling the flexibility of the Perspex sheeting would have generated more

problems than would have been solved.

61

Figure 4.1a Photo of sketches on pers- pex showing wetting fronts for Water, Water + 1 ppm PAM-1011, and Water + 10 ppm PAM-1011 after 30 min.

Figure 4.1b Photo of sketches on pers- pex showing wetting fronts for Na-water, Na-water + 1 ppm PAM-1011, and Na-water + 10 ppm PAM-1011 after 30 min.

Figure 4.1c Photo of sketches on pers- pex showing wetting fronts for Ca-water, Ca-water + 1 ppm PAM-1011, and Ca-water + 10 ppm PAM-1011 after 30 min.

a)

b)

c)

62

Figure 4.2 shows the effect of PAM-1011 and of water quality on the maximum

depth of wetting that occurred during 30 minutes. The depth to which water could

penetrate was reduced considerably as the concentration of PAM-1011 in the water

increased. Importantly, the presence of sodium or calcium alone in the water had no

impact on the maximum wetting depth, but in combination with PAM-1011,

particularly at the higher concentration, sodium reduced the effectiveness of the PAM

and allowed the water to penetrate deeper. The presence of calcium allowed even

deeper penetration of the water.

Figure 4.2. Effect of PAM-1011 and sodium or calcium in the water on the maximum depth to which dripping-liquid could penetrate Viognier sand after 30 minutes.

Figure 4.3 shows the effect of PAM-1011 and water quality on the maximum lateral

spread of water that occurred over 30 minutes. The amount of lateral spread was least

in the three solutions without PAM-1011 and it declined with increasing

concentrations of PAM-1011 – this, of course, was because the experiments were

stopped at 30 minutes. Had the experiments proceeded over a longer period of time

(i.e. if I had a wider apparatus) greater lateral spread would have occurred.

0

5

10

15

20

25

30

0 1 2 3 4 5 6 7 8 9 10

Concentration of PAM-1011 (ppm)

Max

imum

wet

ting

dept

h in

30

min

(cm

)

Distilled water

Na-water

Ca-water

63

Figure 4.3. Effect of PAM-1011 and sodium or calcium in the water on the maximum lateral spread of dripping-liquid in Viognier sand after 30 minutes.

More important was the finding that the presence of sodium with 1 ppm PAM-1011

reduced the lateral spread compared to 1 ppm PAM in distilled water, and that the

presence of calcium with 1 ppm PAM-1011 reduced lateral spread even more. The

effects were less pronounced at 10 ppm PAM-1011, which is in line with the

expectation that higher concentrations of PAM-1011 would be required to counter

the moderating effects of salty irrigation water. In fact, the 10 ppm PAM-1011

solutions tended to pond on the soil surface when applied at 2 L/h, which suggests

irrigation rates would need to be minimized if using PAM at this concentration. In

any case, lateral spread would certainly be encouraged under this scenario.

4.4 Conclusions

Despite the limitations of the two-dimensional model and the narrow apparatus used

in this work, it is clear from the visual observations taken that PAM-1011 reduces

deep percolation of irrigation water and probably increases its lateral spread in sandy

soils.

30

35

40

45

50

0 1 2 3 4 5 6 7 8 9 10

Concentration of PAM-1011 (ppm)

Max

imum

late

ral s

prea

d in

30

min

(cm

) Distilled water

Na-water

Ca-water

64

The presence of salt (regardless of the main cation – sodium or calcium) reduces the

effectiveness of PAM-1011 to limit deep percolation and increase the lateral spread

of irrigation water. This is consistent with all the infiltration data presented in

Chapter 2 and is well explained by the changes in kinematic viscosity outlined in

Chapter 3.

Ponding on the soil surface, which I observed with the 10 ppm PAM-1011

treatments, indicates that such behavioiur under field drippers would increase the

diameter of the wetted area at the soil surface and almost certainly lead to greater

lateral flow of water in the root zone. It follows that during a typical irrigation cycle

there would be less deep percolation and wastage of water using PAM-1011,

particularly at higher concentrations. For growers in light textured soils where saline

irrigation water is being used, higher PAM concentrations would be desirable

anyhow, so the higher concentrations examined in this work have practical relevance

in the field.

65

CHAPTER 5: FIELD STUDIES OF PAM-APPLICATION IN BURIED-DRIP

IRRIGATION SYSTEMS ON SANDY SOILS

5.1 Introduction

Infiltration studies outlined in Chapter 2 indicated that application of PAM-1011 at

modest rates significantly reduced steady-state downward infiltration rate.

Subsequent work reported in Chapter 3 confirmed that the effects were largely

associated with increases in the viscosity of the irrigation water. The reductions in

infiltration rate at higher concentrations of PAM-1011 occurred regardless of water

quality, so it was considered worthwhile to test the findings in the field. I was

particularly encouraged by the ability of PAM-1011 to reduce deep percolation and

apparently increase the lateral spread of water under drippers (Chapter 4). If it were

possible to observe similar trends under field conditions in commercial horticultural

operations, the findings would have significant practical applications.

I chose for my field work a vineyard at Lake Cullularaine, Victoria, which was

developed by Tandou Wineries Ltd using an irrigation system with buried (i.e.

subsurface) drip-lines. In this particular vineyard the dripper-outlets were not

centered uniformly on the vines and many parts of the vineyard received uneven

water applications due to sloping rows, tall, narrow banks, etc. [This vineyard was

planted after the subsurface drippers were installed at a depth of 20 cm and approximately 30 cm

from the planned location of the vine-rows. This was done to reduce installation costs (by installing

two drip-lines at the same time) and to reduce disturbance of the young vines as well as the dripper

tape when trellises were established. Unfortunately the vines were subsequently planted such that two

drip-lines serviced one vine-row while the next vine-row missed out!] Under these

circumstances, there was evidence to suggest that the uneven watering might be

having detrimental effects on vine-development and fruit quality. The owners were

therefore keen to increase the degree of lateral spread of irrigation water from the

drippers by whatever means possible. Testing the use of PAM-1011 for this purpose

thus seemed ideal.

The approach taken to evaluate the effectiveness of PAM-1011 in producing more

uniform water distribution relative to the vines was to observe soil matric suctions in

66

the rootzone at various depths from the soil surface along the vine rows. It was

expected that the increased lateral spread promoted by PAM-1011 would wet the soil

to a greater distance from the vine but keep it drier at depth.

5.2 Materials and methods

The Viognier section (1A) and the Merlot section (7) of the vineyard were chosen for

application of PAM-1011 and installation of the soil-water monitoring equipment

(nearby some of the original soil-survey points examined on a 75m x 75m grid by

Yandilla Park Services during June 1996). Descriptions of both soil profiles (Table

5.1) indicate their duplex nature and thus the potential utility of the PAM in reducing

steady-state infiltration rates.

Table 5.1 Soil survey profile descriptions of the two sections of the vineyard chosen to apply PAM-1011 and monitor soil water status. Section of the vineyard Depth (cm) Texture

0 – 30 Sandy loam 30 – 50 Class IIIA sandy loam 50 – 70 Class IIIB sandy clay loam 70 – 110 Class IIIA sandy clay loam

Viognier section

110 – 140 Sandy clay loam 0 – 65 Sandy loam 65 – 130 Class IIIA sandy loam Merlot section 130 – 170 Class IIIA sandy clay loam

PAM-1011 was injected into the irrigation lines using in-line fertigation units (white

screw-cap shown in Figure 5.1), which were constructed to withstand variable

operating pressures (10-15psi within the dripline) and to be isolated from the

vineyard irrigation system. The fertigation units were made from class-9 PVC

sections available from the hardware store, into which the pre-mixed solutions of

PAM could be added and injected into the irrigation line readily. Irrigation events

occurred at each site according to the schedule outlined in Table 5.2. Minor

variations in water application rates occurred between the two sections of the

vineyard because they were controlled from different pressure systems. (The

Viognier section received 1.1 mm/h while the Merlot section received 0.88 mm/h).

To ensure the two sections received approximately the same application rates of

67

PAM during a typical 4-hour irrigation cycle (1 ppm PAM-1011), the Viognier

section received 1.1g of pre-mixed PAM-1011 solution per irrigation cycle, while the

Merlot received only 1.0 g of pre-mixed PAM-1011 solution per cycle.

Figure 5.1. Inline fertiliser injector installed in buried drip-line (December 2003).

Table 5.2 Schedule of irrigation events with 1 ppm PAM-1011 during the 2003/04 growing season. Tick-marks indicate irrigation occurred on the date shown; dash-marks indicate no irrigation occurred. The shaded bar at 17/11/2003 marks the dividing point in the growing season after which a severe hail storm (19/11/2003) damaged all vines, particularly in Viognier section, which was completely defoliated. Date of irrigation & application of PAM-1011 Viognier section Merlot section

17/10/2003 - √ 23/10/2003 - √ 10/11/2003 - √ 17/11/2003 - √√√√ 27/11/2003 √ √ 10/12/2003 √ √ 18/12/2003 √ √ 29/12/2003 √ √ 07/01/2004 √ √ 12/01/2004 √ √ 22/01/2004 - √ 27/01/2004 √ √ 06/02/2004 √ - 09/02/2004 √ - 16/02/2004 √ -

Tensiometers were installed between vine rows in the vicinity of the PAM-injectors

at depths of 30 cm, 60 cm and 90 cm below the soil surface. This approach enabled a

68

PAM-treated part of a row to be evaluated independently from the rest of the vine

row.

5.3 Results and discussion

Unfortunately, a 1 in 100 year hailstorm occurred on November 19th 2003, which

completely defoliated the Viognier section and severely damaged the Merlot section

as well. This meant there was little significant (and highly variable) water use during

a significant part of the 2003/04 growing season (latter half of Nov 2003 through Feb

2004). Nevertheless, irrigation with and without PAM-1011 continued throughout

the growing season to help the vines recover from the hail damage.

Figures 5.2 shows the soil water matric potential over the entire 2003/04 growing

season at 30, 60 and 90 cm below the soil surface for the Merlot-section vines

irrigated with and without 1 ppm PAM-1011. [The soil water matric potentials for

the Viognier-section, even though they are difficult to interpret because of the hail-

damage, are shown in Figure 5.3 for the sake of completeness, but are not discussed].

Figure 5.2 indicates that throughout most of the growing season the PAM-treated soil

in the Merlot section (green dots connected by blue line) was generally wetter at all

depths than the soil irrigated with water alone (red triangles connected by red line). If

the highly viscous PAM-solution prevented rapid movement of water through the

root-zone (as expected from the laboratory work outlined in Chapters 2 and 3) the

higher matric potentials that I measured at 30 cm below the soil surface is consistent

with this behaviour. It is not consistent, however, with the higher matric potentials

found in the PAM-treated soil at 60 and 90 cm below the surface. It was expected the

soil irrigated with water alone would be drier in the top 30 cm (which it was) and

wetter at depth, which it wasn’t. The only consistent and understandable trend in the

data shown here is that the variability in the soil water matric head was significantly

lower in the PAM-treated soil. The matric potentials in the soil treated with water

alone fluctuated widely and presumably related to the rapid movement of water

through the sandy soil immediately following the weekly irrigation events. Again,

however, a clear explanation of this may not be possible in light of the massive

defoliation caused by the hailstorm.

69

Figure 5.2. Soil water matric potential as measured by tensiometers at 30, 60 and 90 cm below the soil surface in the Merlot-section during the 2003/04growing season.

Merlot section - 30 cm below surface

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

20/09/03 18/10/03 15/11/03 13/12/03 10/01/04 7/02/04

Date of measurement

Soi

l wat

er m

atric

pot

entia

l (kP

a)

PAM-1011

Water alone

Merlot section - 60 cm below surface

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

20/09/03 18/10/03 15/11/03 13/12/03 10/01/04 7/02/04

Date of measurement

Soi

l wat

er m

atric

pot

entia

l (kP

a)

Merlot section - 90 cm below surface

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

20/09/03 18/10/03 15/11/03 13/12/03 10/01/04 7/02/04

Date of measurement

Soi

l wat

er m

atric

pot

entia

l (kP

a)

70

Figure 5.3. Soil water matric potential as measured by tensiometers at 30, 60 and 90 cm below the soil surface in the Viognier-section during the 2003/04growing season.

Viognier section - 30 cm below surface

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

20/09/03 18/10/03 15/11/03 13/12/03 10/01/04 7/02/04

Date of measurement

Soi

l wat

er m

atric

pot

entia

l (kP

a)

PAM-1011

Water alone

Viognier section - 60 cm below surface

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

20/09/03 18/10/03 15/11/03 13/12/03 10/01/04 7/02/04Date of measurement

Soi

l wat

er m

atric

pot

entia

l (kP

a)

Viognier section - 90 cm below surface

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

20/09/03 18/10/03 15/11/03 13/12/03 10/01/04 7/02/04

Date of measurement

Soi

l wat

er m

atric

pot

entia

l (kP

a)

71

Despite the inconclusive soil matric potential results shown above, visual

observations in the vicinity of the tensiometers (Figures 5.4) showed that the PAM-

treated areas of both the Merlot and Viognier sections were relatively moist

compared to the control (no-PAM) sections, at least near the soil surface. The wetter,

PAM-treated sections contained a fair amount of green vegetation (mainly weeds:

Cenchrus pauciflorus, Tribulus terrestis and Chondrilla juncea), while the control

areas had no living vegetation at all. This suggests a higher level of weed control

might be required when using PAM.

Figure 5.4 Visual observations in the areas surrounding the tensiometers. One of the PAM-treated regions is shown on the left, and one of the control regions is shown on the right.

5.4 Conclusions

The field work attempted in this project was seriously compromised by the virtual

destruction of the canopies in both sections of vineyard only three weeks after the

first PAM-solutions were inserted in the weekly irrigation cycles. Nevertheless, there

were indications that the PAM-treated soil was somewhat wetter near the soil surface

than the soil irrigated with water alone (prolific weed growth in the soil wherever

PAM-1011 was applied).

The question as to how much lateral spread of water occurs in the field using PAM-

1011 has yet to be answered definitively. It was completely out of the question to

repeat my field work at the time, but repetition would be worthwhile in the future on

a number of horticultural crops, with careful control of the relative placement of

vines, drippers and water-monitoring sensors.

72

CHAPTER 6: GENERAL DISCUSSION

6.1 Summary of findings

The work in this thesis indicates that PAM-1011 can significantly reduce steady-state

infiltration rates in a range of relatively coarse sandy soils, and it does this with

relatively modest concentrations (< 10 ppm). The other polyacrylamide, PAM-135,

was not effective for this purpose, which indicated that the chemical properties of the

polymer have a big impact on its physical behaviour. The lack of an effect by PAM-

135 suggests it may contain a surfactant (not be verifiable with the equipment

available for this study).

The effectiveness of PAM-1011 in reducing steady-state infiltration rates seems to be

dominated by changes in the properties of the irrigating solution caused by PAM-

1011 and less influenced by the properties of the sands to which it is applied. For

example, PAM-1011 had only minimal (if any) influence on the water retention of a

coarse sandy soil and had no significant impact on water repellence (wettability) of

another sandy soil. It does, however, have a large impact on the kinematic viscosity

of the irrigating solution, and the more PAM-1011 that is dissolved, the more viscous

the solution becomes.

The effectiveness of PAM-1011 in reducing steady-state infiltration rates appears to

be reduced in salty irrigation water, and there was evidence to suggest that cation-

effects may be involved. When PAM-1011 was dissolved in distilled water,

infiltration rates were reduced by the greatest amount. When PAM-1011 was

dissolved in salty water containing the monovalent cation, sodium, infiltration rates

were not reduced as much; furthermore, if the solvent water contained the divalent

cation, calcium, PAM-1011 was even less effective than in sodium-rich water. To

overcome the salt-water effects, higher concentrations of PAM-1011 needed to be

used.

The cation-effects were primarily related to the way each cation interacted with the

polymer to alter its kinematic viscosity. PAM-1011 in distilled water had the greatest

73

viscosity, while PAM-1011 in sodium-rich water had a lower viscosity, and PAM-

1011 in calcium-rich water had the lowest viscosity. A practical implication is that

irrigators using salty waters need to dissolve more PAM-1011 in their water-sources

to increase the viscosity and thus gain the retarding effects of the polymer on

infiltration rates. The amount of polymer required to overcome the salt effects is

probably in the order of 10 ppm PAM-1011. Rates as low as 1 ppm can be used when

irrigators have access to high-quality water with < 10 mmol(+) salt/L present.

Visual observations (modeled after the work of Gardner, 1962) suggest that irrigation

water containing PAM-1011 at concentrations between 1 and 10 ppm reduces the

depth of percolation and increases its lateral spread in coarse sands. While the field

work for this thesis was inconclusive, the laboratory work points to the practical

advantage PAM-1011 allows less water to move through the root zone – this implies

less water is wasted and less water is required – a significant move toward higher

water- and nutrient-use efficiencies on the coarser textured soils in the Murray-

Darling Basin, and probably reduced groundwater contamination with leached

nutrients. Importantly, this study highlights the fact that the criteria for selecting

PAM to reduce infiltration and seepage to the water table in coarse sands are nearly

opposite to those to increase infiltration on finer textured soils. This is important

considering the overwhelming focus on the use of PAMs for erosion control.

6.2 Issues raised by this work and opportunities for research

This study raised as many questions and issues to be investigated as it addressed.

Eight of these are outlined briefly below.

• The reduced rates of infiltration caused by PAM-1011 in the laboratory

suggest that less water would be required in the field for equal growth and

that the time between irrigation cycles on coarse sands could be extended or

at least be more flexible. It would be useful to determine precisely how much

less water is used in the field for a typical range of crops and precisely how

much longer the irrigation cycle could be extended under a wide range of

evaporative demand.

• The effects of sodium and calcium in irrigation water clearly reduce the

viscosity of PAM-1011, but sodium and calcium are not the only cations

74

present in irrigation water. The presence of magnesium, iron and other cations

may generate interactive effects on the viscosity, and these would be worth

investigating to determine whether some water sources are worse than others

– and thus to be avoided when using PAM-1011.

• In coarse sandy soils, water tends to finger its way down the soil profile, so

water contents are rather variable in the root zone. Does PAM-1011 reduce

the variability by retaining the water in contact with the soil for longer? Some

lysimeter studies might address this question.

• The cost of PAM is undoubtedly going to increase as the cost of crude oil and

natural gas increase over time. In this work, PAM was applied in every

irrigation event, but this frequency may not be necessary. It will be important

to determine how long the effects of PAM-1011 last and whether it could be

used in alternate irrigation events. I am aware that ultra-violet light can

degrade the polymer chain, so experiments with subsurface and surface

irrigation could sort out this question.

• The utility of PAM-1011 appears to cover a broad range of sands but higher

concentrations are required to reduce infiltration rates as the particle size

distribution becomes coarser. The question arises as to what the practical

limitations are in relation to the coarsest and finest textures for which PAM-

1011 can produce a reduction in infiltration rate and an increase in the lateral

spread of irrigation water. The work using Keough’s sands suggests the limits

are quite broad, at least at the coarser end.

• Reduced water inputs would have significant implications for the placement

and timing of nutrients by fertigation. Less frequent irrigation events may

imply the use of more concentrated fertilizers, and the implications of this for

water- and nutrient-use efficiencies need to be sorted out.

• The benefits of adding PAM-1011 to irrigation water need to be weighed

against the practical limitations of mixing large quantities of the polymer and

distributing it along a complex system of pipes and hoses.

• It may be a bit of a ‘long-shot’ but where greater water retention occurs near

the soil surface due to PAM-1011, there may be implications for the

proliferation of potentially pathogenic microbial populations in the soil.

75

REFERENCES

Aase JK, Bjourneborg DL, Sojka RE 1998. Sprinkler irrigation runoff and erosion

control with polyacrylamide – Laboratory tests. Soil Science Society of America

Journal 62, 1681-1686.

Abraham J, Rajasekharan-Pillai VN 1990. N,N’–methylene bisacrylamide

crosslinked polyacrylamide for controlled release urea fertiliser formulations.

Communications in Soil Science & Plant Analysis 26, 3231-3241.

Abu-Zreig M 2006. Control of rainfall-induced soil erosion with various types of

polyacrylamide. Journal of Soils & Sediments 6, 137-144.

Ajwa HA, Trout TJ 2006. Polyacrylamide and water quality effects on infiltration in

sandy loam soils. Soil Science Society of America Journal 70, 643-650.

Aly S, Letey J 1989. The effect of two polymers and water qualities on dry cohesive

strength of three soils. Soil Science Society America of Journal 53, 255-259.

Aly S, Letey J 1990. Physical properties of sodium treated soil as affected by two

polymers. Soil Science Society of America Journal 54, 504-510.

Azzam R 1980. Agricultural polymers, polyacrylamide preparation, application and

prospects in soil conditioning. Communications in Soil Science & Plant Analysis

11, 767-834.

Barvenik FW 1994. Polyacrylamide characteristics related to soil applications. Soil

Science 158, 235-243.

Becker H 1997. Stopping erosion with gypsum and PAM. Agricultural Research

45(September), 19-21.

Ben-Hur M 1994. Runoff, erosion, and polymer application in moving-sprinkler

irrigation. Soil Science 158, 283-290.

Ben-Hur M 2006. Using synthetic polymers as soil conditioners to control runoff and

soil loss in arid and semi-arid regions – a review. Australian Journal of Soil

Research 44, 191-204.

Beresford Q, Bekle H, Phillips H, Mulcock J 2001. The salinity crisis: landscapes,

communities and politics. University of Western Australia Press. Crawley, WA.

ISBN: 1 876268 60 3

Bicerano J 1994. Predicting key polymer properties to reduce erosion in irrigated

soil. Soil Science 158, 255-266.

76

Billmeyer FW 1971. Textbook of Polymer Science 2nd Edition. John Wiley and Sons

Inc. New York.

Bjorneberg DL 1998. Temperature, concentration, and pumping effects on PAM

viscosity. American Society of Agricultural Engineers 41, 1651-1655.

Bres W, Weston LA 1993. Influence of gel additives on nitrate, ammonium and

water retention and tomato growth in a soilless medium. HortSci 28, 1005-1007.

Brown TL, Lemay EH 1981. Chemistry: The Central Science 2nd Edition. Prentice

Hall International Inc. New York.

Brown WH 1988. Introduction to Organic Chemistry 4th Edition. Brooks Cole

Publishing Company. New York.

Burgess HD 1998. Formulation of Microbial Biopesticides. Kluwer Academic

Publishers, London.

Cary JW, Evans DD (editors) 1974. Soil crusts. University of Arizona Agricultural

Research Station Technical Bulletin No.214.

Chamberlain P 1988. Polymers as soil conditioners. Chemistry in Britain, February,

144-146.

Cowie JMG Arrighi V 2007. Polymers: Chemistry and Physics of Modern Materials

3rd edition. CRC Press. Boca Raton, FL.

De Bano LF 2003. Historical overview of soil water repellency. pp.3-24 In CJ

Ritsema, LW Dekker (eds) Soil Water Repellency – Occurrence, Consequences

and Amelioration. Elsevier. Amsterdam.

De Jonge LW, Jacobsen OH, Møldrup P 1999. Soil water repellency: effects of water

content, temperature, and particle size. Soil Science Society of America Journal,

63, 437–442.

Doerr SH 1998. On standardizing the ‘Water Drop Penetration Time’ and the

‘Molarity of an Ethanol Droplet’ techniques to classify soil hydrophobicity: A

case study using medium textured soils. Earth Surface Processes and Landforms

23, 663-668.

Falatah, AM, Al-Omran AM, Shalaby AA, Mursi MM 1999. Infiltration in a

calcareous sandy soil as affected by water-soluble polymers. Arid Soil Research

& Rehabilitation 13, 61-73.

Finch CA 1973. Polyvinyl Alcohol – Properties and Applications. John Wiley and

Sons Ltd., New York.

77

Gardiner D, Shainberg I 1996. Effects of PAM on infiltration of high sodium water.

Abstract of final report on Project 94R-02 to the USDA Cooperative State

Research, Education, and Extension Service. (International Arid Lands

Consortium URL: http://alic.arid.arizona.edu/ialc/ialc4.asp?proj=94R-02 )

Gardner WH 1962. How water moves in the soil. Part 1 – the basic concept; Part 2 –

In the field. Crops & Soils Magazine, pp.7-9 Oct issue; pp9-11 Nov issue, 1962.

Green VS, Stott DE 2001. Polyacrylamide: A review of the use, effectiveness, and

cost of a soil erosion control amendment. Pp.384-389 in DE Stott, RH Hohtar &

GC Steinhardt (eds). Sustaining the Global Farm – selected papers from 10th

International Soil Conservation Organization, held 24-29 May 1999. Purdue

University & USDA-ARS. West Lafayette, IN.

Green VS, Stott DE, Norton LD, Graveel JG 2000. Polyacrylamide molecular weight

and charge effects on infiltration under simulated rainfall. Soil Science Society of

America Journal 64, 1786-1791.

Grula MM, Huang M-L, Sewell G 1994. Interactions of certain polyacrylamides with

soil bacteria. Soil Science 158, 291-300.

Hayes SA, McLaughlin RA, Osmond DL 2005. Polyacrylamide use for erosion and

turbidity control on construction sites. Journal Soil & Water Conservation 60,

193-199.

Hillel D 1998. Environmental soil physics. Academic Press. New York.

Huttermann A, Zommorodi M, Reise K 1999. Addition of hydrogels to soil for

prolonging the survival of Pinus halepensis seedlings subjected to drought. Soil

& Tillage Research 50, 295-304.

Isbell RF 1996. The Australian Soil Classification CSIRO Australia. Collingwood.

Lal R, Shukla MK 2004. Principles of Soil Physics. Marcel Dekker. New York.

Lefroy EC, Stirzaker RJ, Pate JS 2001. The influence of tagasaste (Chamaecytisus

proliferus Link.) trees on the water balance of an alley cropping system on deep

sand in south-western Australia. Australian Journal Agricultural Research 52,

235-246.

Lentz RD, Shainberg I, Sojka RE, Carter DL 1992. Preventing irrigation furrow

erosion with small applications of polymers. Soil Science Society of America

Journal 56, 1926-1932.

78

Lentz RD, Sojka RE 1994. Field results using polyacrylamide to manage furrow

erosion and infiltration. Soil Science 158, 274-281.

Lentz RD, Sojka RE, Carter DL 1993. Influence of polymer charge type and density

on polyacrylamide ameliorated irrigated furrow erosion. pp.159-168. In:

Proceedings 24th Annual International Erosion Control Association Conference.

23-26 Feb., 1993, Indianapolis, IN.

Lentz RD, Sojka RE, Robbins CW 1998a. Reducing soil and nutrient losses from

furrow irrigated fields with polymer application. Advances in GeoEcology 31,

1233-1238.

Lentz RD, Sojka RE, Robbins CW 1998b. Reducing phosphorus losses from surface

irrigated fields: Emerging polymer technology. Journal of Environmental

Quality 27, 305-312.

Letey J 1994. Adsorption and desorption of polymers on soil. Soil Science 158, 244-

248.

Levy GJ, Ben-Hur M, Agassi M 1991. The effect of polyacrylamide on runoff,

erosion, and cotton yield from fields irrigated with moving sprinkler systems.

Irrigation Science 12, 55-60.

Levy GJ, Levin J, Gai M, Ben-Hur M, Shainberg I 1992. Polymer effects on

infiltration and soil erosion during consecutive simulated sprinkler irrigations.

Soil Science Society of America Journal 56, 902-907.

Levy GJ, Levin J, Shainberg I 1995. Polymer effects on runoff and soil erosion from

sodic soils. Irrigation Science 16, 9-14.

Levy GJ, Rapp I 1999. Polymer effects on surface mechanical strength of a crusting

loess soil. Australian Journal of Soil Research 37, 91-102.

Lu JH, Wu L, Letey J 2002. Effects of soil and water properties on anionic

polyacrylamide sorption. Soil Science Society of America Journal 66, 578-584.

Malik M, Letey J 1992. Pore-size-dependent apparent viscosity for organic solutes in

saturated porous media. Soil Science Society of America Journal 56, 1032-1035.

Malik M, Amrhein C, Letey J 1991. Polyacrylamide to improve water flow and salt

removal in a high shrink swell soil. Soil Science Society of America Journal 55,

1664-1667.

79

Ma'shum M, Tate ME, Jones GP, Oades JM 1988. Extraction and characterization of

water repellent materials from Australian soils. Journal of Soil Science 39, 99-

110.

Nadler A, Magaritz M, Leib L 1994. PAM application techniques and mobility in

soil. Soil Science 158, 249-254.

Oliver JP, Ross S 1964. On Physical Adsorption. Interscience Publishers. John Wiley

& Sons Inc. New York.

Orts WJ, Sojka RE, Glenn GM 2000. Bioploymer additives to reduce erosion

induced soil losses during irrigation. Industrial Crops & Products 11, 19-29.

Pritchard JG 1970. Poly(vinyl alcohol) – Basic properties and uses. McDonald & Co.

London.

Pryor A 1988. Water-absorbing polymers have been shown to improve yields in

processing tomatoes. California Farm (Oct), 2.

Ross CW, Sojka RE, Lentz RD 1996. Polyacrylamide as a tool for controlling

sediment runoff and improving infiltration under furrow irrigation. Proc.

Australian New Zealand Nat. Soils Conf. Vol. 2 Oral Papers, 229-230.

Schiller C, Edmunds E 1997. Increasing irrigation efficiency and reducing pesticide

runoff with polyacrylamides. Unpublished report for Ozchem Pty Ltd, Cytec

Australia Pty Ltd.

Seybold C 1993. Polyacrylamide: soil conditioning and environmental fate.

Communications in Soil Science & Plant Analysis 25, 2171-2185.

Shainberg I, Levy GJ 1994. Organic polymers and soil sealing in cultivated soils. Soil

Science 158, 267-274.

Sharma PK 1988. Interaction of FeSO4 with polyacrylamide and urea formaldehyde

in aggregating light-textured soils. Soil Science 146, 185-191.

Shoemake-Kay JL, Watwood ME, Lentz RD, Sojka RE 1998a. Polyacrylamide as an

organic nitrogen source for soil microorganisms with potential effects on

inorganic soil nitrogen in agricultural soil. Soil Biology & Biochemistry 30,

1045-1052.

Shoemake-Kay JL, Watwood ME, Lentz RD, Sojka RE 1998b. Polyacrylamide as a

substrate for microbial amidase in culture and soil. Soil Biology & Biochemistry

30, 1647-1654.

80

Sivapalan S 2003. Polyacrylamides: their potential benefits for agriculture. Irrigation

Australia 18, 23-26.

Sivapalan S 2006. Benefits of treating a sandy soil with a crosslinked-type

polyacrylamide. Australian Journal of Experimental Agriculture 46, 579-584.

Sojka RE, Bjorneberg DL, Entry JA, Lentz RD, Orts WJ 2007. Polyacrylamide in

agriculture and environmental land management. Advances in Agronomy 92, 75-

162.

Sojka RE, Entry JA 2000. Influence of polyacrylamide application to soil on

movement of microorganisms in runoff water. Environmental Pollution 108,

405-412.

Sojka RE, Lentz RD 1994. Time for yet another look at soil conditioners. Soil

Science 158, 233- 234.

Sojka RE, Lentz RD 1996. A PAM primer: A brief history of PAM and PAM related

issues. Symp. Managing Irrigation Induced Erosion and Infiltration with

Polyacrylamide May 6-8 , Twin Falls ID, University of Idaho, Misc. Pub.

101/96, 11-20.

Sojka RE, Lentz RD, Westermann DT 1998. Water and erosion management with

multiple applications of polyacrylamide in furrow irrigation. Soil Science Society

of America Journal 62, 1672-1680.

Sojka RE, Surapaneni A 2000. Potential use of PAM in Australian agriculture to

improve off and on site environmental impacts and infiltration management.

LWRRDC Project Number UNE 39 Polyacrylamides in Agriculture. Final report

June 2000 (unpublished).

Stern R, Van Der Merwe AJ, Laker MC, Shainberg I 1992. Effect of soil surface

treatments on runoff and wheat yields under irrigation. Agronomy Journal 84,

114-119.

Streitwieser A, Heathcock CH 1985. Introduction to Organic Chemistry 3rd Edition.

Macmillan Publishing Co. Inc., New York.

Stumm W 1992. Chemistry of the Solid-Water Interface. John Wiley & Sons Inc.

New York.

Thomas WM 1964. Acrylamide polymers. In HF Mark, NG Gaylord, NM Bikales

(eds) Encyclopedia of Polymer Science & Technology: Plastics, Resins,

Rubbers, Fibers. Interscience Publications, New York.

81

Trout TJ, Sojka RE, Lentz RD 1995. Polyacrylamide effect on furrow erosion and

infiltration. American Society of Agricultural Engineering 38, 761-765.

Tucker MD, Barton LL, Thomson BM 1998. Removal of U and Mo from water by

immobilised Desulfovibrio desulfricans in column reactors. Biotechnology &

Bioengineering 60, 88-96.

van Dam JC, Hendrickx JMH, van Ommen HC, Bannink MH, van Genuchten MTh,

Dekker LW 1990. Water and solute movement in a coarse-textured water-

repellent field soil. Journal of Hydrology (Amsterdam) 120, 359-379.

Wallace A, Wallace GA, Abouzamzam AM 1986. Effects of excess levels of a

polymer as a soil conditioner on yields and mineral nutrition of plants. Soil

Science 141, 377-380.

Williams BG, Greenland DJ, Quirk JP 1967. The effect of polyvinyl alcohol on the

nitrogen surface area and pore structure of soils. Australian Journal of Soil

Research 5, 77-83.

Wilson AD, Crisp S 1975. Rigid highly carboxylated ionic polymers. pp. 208-257, In

L Holiday (ed). Ionic Polymers. Chapman and Hall, New York.

Young RJ 1981. Introduction to polymers. Chapman and Hall, New York.

Zahow MF, Amrhein C 1992. Reclamation of a saline sodic soil using synthetic

polymers and gypsum. Soil Science Society of America Journal 56, 1257-1260.

82

APPENDIX 1

Raw infiltration data used to prepare Chapter 2 of thesis (see attached Compact Disc)