Citation
High pressure liquid chromatography and chemical characterization of extractable soil organic matter

Material Information

Title:
High pressure liquid chromatography and chemical characterization of extractable soil organic matter
Creator:
Loeppert, Richard Henry, 1944-
Publication Date:
Language:
English
Physical Description:
xii, 89 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Adsorption ( jstor )
Electrolytes ( jstor )
Elution ( jstor )
Gels ( jstor )
Molecular weight ( jstor )
pH ( jstor )
Soil organic matter ( jstor )
Soils ( jstor )
Solutes ( jstor )
Solvents ( jstor )
Dissertations, Academic -- Soil Science -- UF
Humus ( fast )
Liquid chromatography ( fast )
Soil Science thesis Ph. D
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Includes bibliographical references (leaves 83-88).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Richard Henry Loeppert, Jr.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
This item is presumed in the public domain according to the terms of the Retrospective Dissertation Scanning (RDS) policy, which may be viewed at http://ufdc.ufl.edu/AA00007596/00001. The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
020466161 ( ALEPH )
02686700 ( OCLC )

Downloads

This item has the following downloads:


Full Text














HIGH-PRESSURE LIQUID CHROMATOGRAPHY AND CHEMICAL CHARACTERIZATION
OF EXTRACTABLE SOIL ORGANIC MATTER
















By

RICHARD HENRY LOEPPERT, JR.
















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






UNIVERSITY OF FLORIDA

1976



































To my parents,

Richard and Adeline Loeppert


















ACKNOWLEDGMENTS




The author expresses sincere appreciation to Dr. J. G. A. Fiskell,

chairman, and Dr. B. G. Volk, cochairman, of the supervisory committee,

for their guidance, encouragement, and assistance during the progress

of this investigation. Appreciations are also extended to Dr. D. H.

Hubbell, Dr. N. Gammon, and Dr. W. S. Brey for their interest and

participation on the supervisory committee and review of manuscript.

Special appreciations are extended to Dr. L. W. Zelazny and Dr.

M. A. Battiste for important discussions and inspiration provided during

early stages of the investigation. A sincere thanks is extended to

faculty, staff, and students in the Soil Science Department for the many

stimulating discussions which served as the basis for the evolution

of this study.

A very special thank you is extended to Ms. Carolyn Beale and Mr.

Jerry Osbrach for assistance in the laboratory and to Ms. Ann Barry

for typing portions of the original manuscript. The author pays a

special tribute to Ms. Nancy McDavid for the very professional typing

and careful review of the manuscript and to Ms. Helen Huseman for

final preparation and drafting of several of the figures.

The author expresses his sincere gratitude to Dr. C. F. Eno,

chairman of the Soil Science Department at the University of Florida,




iii










and to Dr. D. W. Beardsley and Dr. D. H. Myhre, Center Directors, at

the Agricultural Research and Education Center, Belle Glade, for pro-

viding the research assistantship which has enabled the author to

pursue his doctoral program.

The author deeply appreciates the continuing encouragement and

assistance given him by his parents throughout the course of his studies

and the special upbringing which has encouraged the author to search,

to question, and to approach problems with an open mind.













































iv


















TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS. . .... . . . .... iii

LIST OF TABLES . ... . . . . vi

LIST OF FIGURES. . . ... . . .. .viii

ABSTRACT . . . . ... . . .. x

INTRODUCTION . . . . ... . . 1

LITERATURE REVIEW. . . . . ... . .. 4
Soil Organic Matter . . . .. .. 4
Extraction of Soil Organic Matter . . . 6
Gel Permeation Chromatography . . . 8
General Information. . . . . 8
Theory and Nomenclature. . . . 8
Packing Materials. . . . 10

MATERIALS AND METHODS. ..... . . . 16
Sample Pretreatment and Extraction. . . ... 16
Solubility Studies. . . . .... 19
Analytical Determinations . . . . 19
High Pressure Liquid Chromatography . . ... 21

RESULTS AND DISCUSSION . . . . ... . 25
Chemical Characteristics of Extractable Organic Matter. .. 25
Solubility Characteristics of Extractable Organic Matter. 31
High Pressure Liquid Chromatography . . ... 38
Porous Silica Packing Materials. . . .. 38
Polystyrene-divinylbenzene (DVB) . . .. 63

CONCLUSIONS. . . . . ... . . 78
Extraction and Fractionation. . . . .. 78
Solubility Properties . . . . ... .79
Liquid Chromatography . . . . ... 80

LITERATURE CITED . . . ... . 83

BIOGRAPHICAL SKETCH. . . . . ... . 89



v

















LIST OF TABLES

Table Page

1 Parameters of column packing materials. . ... 23

2 Yields of extractable soil organic matter ...... 26

3 Elemental and functional group concentrations of
extractable soil organic matter . . ... 27

4 Titratable acidity of extractable soil organic matter 28

5 Solubility of extractable soil organic matter in
selected solvents at 0.1% concentration . ... 32

6 Solubility of extractable soil organic matter as
influenced by saturating cation and solvent . .. 34

7 Solubility of fulvic acid in aqueous salt solutions 35

8 Peak elution volumes of extractable soil organic matter
on Porasil A with selected solvents . ... 44

9 Peak elution volumes of extractable soil organic matter
on Porasil AX with selected solvents. . ... 45

10 Peak elution volumes of extractable soil organic matter
on CPG-250 with selected solvents . . .. 46

11 Peak elution volumes of organic standards on Porasil A
with selected solvents. . . .. ... 49

12 Peak elution volumes of organic standards on Porasil AX
with selected solvents. . . . ... 50

13 Peak elution volumes of organic standards on CPG-250
with selected solvents. . . . ... 51

14 Peak elution volumes of cation-saturated fulvic acid on
Porasil A, Porasil AX, and CPC-:ln with water as
eluting solvent . . ... . 56




vi










LIST OF TABLES (continued)

Table Page

15 Peak elution volumes of selected organic acid standards
on Porasil A with Na SO4 solutions. . . ... 60

16 Peak elution volumes of selected organic acid standards
on Porasil AX with Na2SO4 solutions . . .. 61

17 Peak elution volumesof low molecular weight standards
o 0
eluted on 100 A Poragel with THF and DMF and on 100 A
p-Styragel with THF as the eluting solvent. . .. .70

18 Molecular weight estimates of soil humic fractions based
on elution of polystyrene standards on p-Styragel with
THF as the eluting solvent. . . . .. 77











































vii

















LIST OF FIGURES

Figure Page

1 Extraction scheme. . . . . ... 17

2 Infrared patterns. . . . . ... 30

3 Effect of flow rate on column efficiency, N, of Porasil
A, Porasil AX, and CPG-250 analytical columns. ... 39

4 Effect of sample size on column efficiency, N, of
Porasil A, Porasil AX,and CPG-250 analytical columns 41

5 Molecular weight calibration curves of 1 m x 0.318 cm
OD Porasil AX and CPG analytical columns obtained by
elution of polystyrene standards with THF. . .. 43

6 Elution patterns of the 2-propanol-Soxhlet extract of
DMF-extractable material on Porasil A and Porasil AX
with selected solvents . . . ... 47

7 Elution patterns of fulvic acid on Porasil A and Porasil
AX with selected solvents. . . . ... 47

8 Effect of excess neutral electrolyte on elution of Na-
saturated fulvic acid on Porasil A . . .. 58

9 Effect of excess neutral electrolyte on elution of Na-
saturated fulvic acid on Porasil AX. . . ... 58

10 Effect of excess neutral electrolyte on elution of Na-
saturated 1,2,4,5-tetracarboxybenzene on Porasil A 59

11 Effect of excess neutral electrolyte on elution of Na-
saturated 1,2,4,5-tetracarboxybenzene on Porasil AX. 59

12 Effect of flow rate on column efficiency, N, of 100 A
Poragel and 100 X U-Styragel preparative columns with
THF as the eluting solvent . . . ... 64

13 Effect of sampleosize on column efficiency, N, of 100 A
Poragel and 100 A w-Styragel preparative columns with
THF as the eluting solvent . . . ... 66




viii










LIST OF FIGURES (continued)

Figure Page

14 Molecular weight calibration curve of p-Styragel (2 f
x 0.954 cm OD 100 A + 2 f x 0.954 cm OD 500 A) obtained
by elution of polystyrene standards with THF . .. 67

15 Elution of Soxhlet extracts of NaOH-extractable soil
organic matter on 100 A p-Styragel with THF. . ... 74

16 Elution of Soxhlet extracts of DMF-extractable soil
organic matter on 100 A p-Styragel with THF. . ... 75














































ix











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

HIGH-PRESSURE LIQUID CHROMATOGRAPHY AND CHEMICAL CHARACTERIZATION
OF EXTRACTABLE SOIL ORGANIC MATTER

By

Richard Henry Loeppert, Jr.

December, 1976

Chairman: Dr. John G. A. Fiskell
Cochairman: Dr. B. G. Volk
Major Department: Soil Science

The objective of this investigation was to evaluate the use of

high-pressure liquid chromatography and a series of new packing materi-

als (Porasil-silica gel, Corning controlled pore glass, and polystyrene-

divinylbenzene) for the molecular size fractionation of extractable soil

organic matter. The effects of packing material, solvent, saturating-

cation, concentration of excess electrolyte, and pH on solute-gel-

solvent interactions were investigated. Preliminary experiments were

performed to investigate the behavior of soil humic compounds in organic

solvents and to select solvents which would be suitable for extraction

and fractionation of the soil humic complex.

The soil used was Terra Ceia muck, a Typic Medisaprist. Organic

matter was extracted from the soil by separate treatment with 0.5 N

NaOH and dimenthylformamide (DMF). The NaOH-extractable material was

separated into humic acid and fulvic acid fractions. In addition, both

the NaOH- and DMF-extractable materials were further fractionated with

a Soxhlet extraction scheme. The ash content of all samples was

lowered by dialysis to less than 0.5%.


x











The C and H content of the fractions decreased and the 0 and COOH

contents and total acidity increased according to the following order:

hexane-Soxhlet, benzene-Soxhlet, ethylacetate-Soxhlet, acetone-Soxhlet,

2-propanol-Soxhlet, methanol-Soxhlet, NaOH extract-humic acid-DMF

extract, fulvic acid.

Fulvic acid and organic acid standards prepared in the H-, Na-,

N(CH3)4-, and N(C H9)4-saturated forms were excluded from the pores

of Porasil packing material when water was used as the eluting solvent.

Acetone-, 2-propanol-, and methanol-extractable soil organic matter

and organic acid standards were predominantly excluded from the pores

when methanol or DMF was used as the eluting solvent and predominantly

adsorbed when tetrahydrofuran (THF) or acetone was used. Exclusion

phenomena were evident in the presence of organic solvents with sig-

nificant basic character and may be attributed to electrostatic repul-

sion of negatively charged organic matter by negatively charged sites

on the silica surface. Organic solutes with significant basic charac-

ter were adsorbed.

In the presence of 0.05 N excess neutral electrolyte, cation-

saturated fulvic acid and organic acid standards entered the porous

gel matrix due to suppression of charge and/or decreased electrical

double-layer thickness of the negatively charged solute molecules and

the negatively charged silica surface. As electrolyte concentration

was increased, however, adsorption phenomena became more prevalent due

to precipitation at the surface and/or direct interaction between

active sites on the silica surface and oxygen-containing functional




xi











groups of the organic solute. Deactivation of the silica surface

(Porasil X) resulted in reduction but not elimination of adsorption

and electrostatic exclusion phenomena.

Elution of soil humic compounds and low molecular weight standards

on polystyrene-divinylbenzene (DVB) was strongly influenced by the

solvent. Each of the low-molecular weight organic solutes entered

the porous gel matrix when eluted with THF, however, several hydrophobic

aromatic compounds (e.g. benzene, toluene, and anthracene) were ad-

sorbed. Soil humic compounds apparently readily entered the polystyrene-

DVB gel matrix when THF was used. When eluted with DMF, however, soil

humic compounds and low molecular weight organic acid standards were

either totally or partially excluded from the gel matrix. This phe-

nomenon was attributed to the effect of solvent on dissociation of

solute molecules and to a possible ion-inclusion effect.

None of the gels investigated was chemically inert, and each

apparently interacted with the soil humic material. Interactions

could be minimized, however, by proper selection of gel, solvent, and

concentration of excess electrolyte.

Molecular weights of acetone-, 2-propanol-, and methanol-Soxhlet

fractions were estimated to be 500 to 800, based on comparison of

elution patterns of soil humic fractions with those of polystyrene

standards in THF.












xii


















INTRODUCTION



Soil is an important international resource which serves as the

source of the majority of the world's food supply and as a major sink

for all man-made and natural products in the environment. Organic

matter, due to its reactive nature, has a large influence on soil

properties. In order to understand the chemical properties of soil

organic matter and the exact role of humic fractions in the soil, it

is essential that the scientist understand the chemical structures

involved. Many years of research have given structural clues, however,

due to the complexity of soil organic matter, the science is still in

its infancy.

One approach to the structural problem has been to initially

fractionate and simplify the humic material prior to further analyti-

cal investigations.

Since the development of the polydextran gels, there has been

considerable interest in gel filtration for molecular-size fractiona-

tion and characterization of soil organic matter (Swift and Posner,

1971). Scientists have used this method to isolate molecular-size

fractions and to obtain estimates of molecular weight of humic sub-

stances in soils and natural water.

During the past decade, rapid advances have been made in liquid

chromatography. These advances have been due primarily to the


1






2




development of the high-pressure liquid chromatograph which, in turn,

has made possible the use of small-diameter packing materials and

high efficiency columns. For example, with the new W-packing ma-

terials, column efficiencies as high as 5000 theoretical plates per

foot are commonly attained. The rapid rise in the application of

high-pressure liquid chromatography is readily apparent following

a quick glance through any recent issue of Analytical Chemistry.

Attempts at molecular-size fractionation of soil humic materials

have been complicated by the fact that no gel material, including

Sephadex, is completely inert. Therefore, separations may be adversely

affected by gel-solute and gel-solvent interactions which would lead

to misleading results. Also, the humic molecule is a strongly reactive

solute which has a strong tendency to interact with other solute mole-

cules and solvent molecules. For this reason, the fractionation of soil

humic compounds is affected by solute-solute and solute-solvent

interactions. Each of the above-mentioned interactions is strongly

influenced by packing material, solvent, saturating cation, concen-

tration of excess electrolyte, and pH.

The objectives of this work were to investigate the use of high-

pressure liquid chromatography and a series of new packing materials

for the size fractionation of soil organic matter extracts and to

investigate the effect of packing material, solvent, saturating-

cation, concentration of excess electrolyte, and pH on solute-solute,

gel-solute, solvent-solute, and gel-solvent interactions which would

influence size separations. Since several of the new packing materials










were compatible with organic solvents, but not with aqueous solvents,

we were also interested in the behavior of soil humic compounds in

organic solvents and in selection of organic solvents which were

suitable solvent media for extraction and preliminary fractionation

of the soil humic complex.


















LITERATURE REVIEW



Soil Organic Matter


For discussions of our current knowledge of the chemical makeup

of soil organic matter the reader is referred to texts by Kononova

(1966) and Schnitzer and Khan (1972) and a review by Hurst and Burges

(1967). Some of the important characteristics and properties which

are relevant to the discussion in the text are summarized in the next

few paragraphs.

Hurst and Burges (1967) suggested that humic acids are polycon-

densates of monomers immediately available in a particular microarea

of the soil and do not appear to have integrity of structure and the

rigid chemical configuietion of many other macromolecules due to the

complexity and heterogeneous nature of the system in which they are

formed. According to Kononova (1966), possibly no two humus molecules

would have exactly the same structure.

Elemental and functional group compositions for representative

humic and fulvic acids have been tabulated by Schnitzer and Khan (1972).

The most striking features of these tabulations are the relatively high

oxygen contents, low nitrogen and sulfur contents, high carbon to

hydrogen ratios, high total acidity, and high carboxyl and phenolic

hydroxyl contents. Comparisons of humic and fulvic acid (Schnitzer and



4






5




Khan, 1972) showed that fulvic acid had lower carboxyl contents, higher

oxygen contents, and higher total acidity and carboxyl content.

Infrared spectra of soil humic compounds show broad absorption

bands (Schnitzer and Khan, 1972). The majority of spectra did not
-l
show absorption bands in the 600-900 cm1 region, and,therefore, did

not demonstrate the presence of aromatic protons. Likewise, nuclear

magnetic resonance (NMR) spectra of methylated fulvic acid did not

indicate the presence of aromatic protons (Schnitzer and Skinner,

1968).

Schnitzer and Khan (1972) determined molecular weights of 1684

and 669 for humic acid and fulvic acid, respectively, by the freezing

point depression method; however, molecular weights as high as 100,000,

or greater, have been determined by other methods (Schnitzer and Khan,

1972).

Electron spin resonance (ESR) spectra of soil humic substances

indicate the presence of a high concentration of free radicals with

unpaired electrons. Possible sources include semiquinone polymer,

hydroxyquinone, or condensed polynuclear hydrocarbons (Steelink and

Tollin, 1967).

Various degradation procedures have been used to separate com-

plex molecules into monomeric components. Kumada and Suzuki (1961)

and Cheshire et al. (1967) identified polycyclic aromatic compounds

following alkaline permaganate oxidation. Hansen and Schnitzer (1969),

on the contrary, obtained no polycyclic aromatic compounds but did

obtain aliphatic carboxylic acids and all possible benzene carboxylic





6




acids except benzoic acid. Ortiz de Serra and Schnitzer (1973) iso-

lated and identified a number of phenolic acids. Hansen and Schnitzer

(1967) used nitric acid oxidation and identified a seriesof nitro-

phenols and aliphatic dicarboxylic, phenolic, and benzenecarboxylic

acids.

Based on degradative and nondegradative studies, Schnitzer has

proposed a structure for fulvic acid consisting of phenolic and benzene-

carboxylic acids joined by hydrogen bonds to form a polymeric matrix of

considerable stability (Schnitzer and Khan, 1972). Numerous additional

structures for humic and fulvic acids have been suggested (Burges,

Hurst, and Walkden, 1964; Flaig, Beutelspacher, and Reitz, 1975;

Haworth, 1971). The complexity and diversity of these structures

demonstrate the probable complexity of the total soil humic complex.



Extraction of Soil Organic Matter


Procedures for extraction of soil organic matter were reviewed

by Mortenson (1965) and Stevenson (1965). Dilute aqueous NaOH is

the most commonly used extractant of soil organic matter. Sodium

hydroxide produces high yields of extractable organic matter; however,

its use has been severely criticized due to chemical alterations

which may occur in alkaline conditions (Bremner and Lees, 1949;

Bremner, 1956; Choudhri and Stevenson, 1957). Bremner (1950) observed

that 02 was adsorbed from the atmosphere by alkaline soil suspensions.

Tinsley and Salam (1961) suggested that condensation reactions between

amino compounds and aldehydes or phenolic compounds may result in

formation of humin-type compounds during NaOH extraction.






7



Loeppert and Volk (1974) investigated the yields and properties

of extractable organic matter solubilized from Terra Ceia muck (Typic

Medisaprist) by a series of organic and inorganic extracting solvents.

Quantity of organic material extracted with dimethylformamide (DMF),

pyridine, and methanol were substantially increased when the soil

ash content was lowered by dialysis prior to extraction. Yields ob-

tained with less polar extractants (acetonitrile, chloroform, acetone,

and benzene) were not highly influenced by ash content. Choudhri and

Stevenson (1957) and Bremner and Lees (1949) were able to significantly

increase extraction yields with NaOH by pretreating the soil with 0.1

N HC1 to lower the ash content and remove exchangeable cations. Ex-

traction yields from Pahokee muck (Typic Medisaprist) obtained by

single 24-hour extractions with 0.5 N NaOH (10:1 extractant:soil ratio)

were increased from 23 to 40% following pretreatment with 0.1 N HC1

(Snow, Loeppert, and Volk, 1974). Similarly, extraction yields of

Terra Ceia muck were increased from 38% to 49% following treatment with

0.1 N HC1 (Loeppert and Volk, 1974). When DMF was used as the extract-

ing solvent, extraction yields were increased from 0.5% to 25% follow-

ing pretreatment with 0.1 N HC1. In these studies, it was observed

that the organic material extracted by DMF was similar in functional

group and elemental analyses to that extracted by 0.5 N NaOH, and

that H20 extracted a material with properties similar to those of

fulvic acid. Loeppert and Volk (1974) concluded from the high yields

obtained with DMF following pretreatment with 0.1 N HC1, the relative

mildness of DMF, and the similarity in properties of DMF- and 0.5






8



N NaOH-extractable materials that DMF may be an excellent solvent

for structural studies of the humic fraction. In general, organic

solvents employed to extract soil organic matter have had limited use

because yields were low and because a more specific fraction may be

extracted than with NaOH. Organic solvents which have been investi-

gated include acetylacetone (Halstead, Anderson, and Scott, 1966),

anhydrous formic acid (Parsons and Tinsley, 1960), pyridine (Kessler,

Friedel, and Sharkey, 1970), methanol (McIver, 1962), acetone-H20-

HC1 (Porter, 1967), aqueous THF (Salfeld, 1964), and EDTA (Schnitzer,

Shearer, and Wright, 1959).



Gel Permeation Chromatography


General Information


The practice of high-pressure liquid chromatography is covered in

a text by Kirkland (1974) and reviews by Zweig and Sherma (1974) and

Gaylor, James,and Weetall (1976). Gel-permeation chromatography is a

form of liquid chromatography in which molecules are separated accord-

ing to size. The larger molecules are excluded from all or a portion

of the solvent-filled pores of the packing material due to their physi-

cal size. On the other hand, a nonreactive small molecule may freely

enter the pores of the packing material.


Theory and Nomenclature


For discussions of chromatographic theory, the reader is referred

to the text by Giddings (1965) and review articles by Bly (1970),

Karger (1971), and Bombough (1971).






9



Column parameters are defined in terms of VO, VT, and column

efficiency, N. The total pore volume of the column, VT, is determined

by the elution volume of a nonreactive low molecular weight material

which freely enters the solvent-filled pores of the packing material.

The interstitial volume of the column, VO, is determined by the elu-

tion volume of a nonreactive high-molecular weight material which is

completely excluded from the pores of the packing material. Both V0

and VT are determined with standard compounds. In practice, the condi-

tion of absolute nonreactivity between solute and packing would probably

never be attained, since there is no such thing as a completely inert

gel network (Freeman, 1973). Likewise, there is no such thing as an

entirely inert solute in liquid chromatography. However, by proper

selection of solute, column parameters V and VT can be determined

with a high degree of accuracy.

Column efficiency is expressed by the theoretical plate count,

N, which is determined with the equation,



N = 16 where



VE is the elution volume of the solute, and w is the peak width at

the baseline. Column efficiency is influenced by particle diameter of

the column, linear velocity of the solvent, and how well the column

is packed (Karger, 1971; Dark and Limpert, 1973). Karger (1971) pre-

sents an excellent discussion of factors affecting resolution and

column efficiency.





10




Packing Materials


The various packing materials currently available have been sum-

marized by Dark and Limpert (1973), Kirkland (1974), and Laub (1974).

They are generally divided into three major classes: (i) rigid gels

or glasses, (ii) semirigid gels, and (iii) nonrigid gels. The gels

comprising the first group, the rigid gels, are composed of porous

silica and are suitable for high-pressure liquid chromatography. The

semirigid gels (e.g polystyrene-divinylbenzene) are highly cross-linked

organic polymers, will not distort under pressure, and are, therefore,

suitable for high-pressure liquid chromatography. The nonrigid gels

(e.g Sephadex G-gels) are lightly cross-linked organic polymers and

not suitable for high-pressure liquid chromatography since they will

distort under pressure, resulting in an altered pore structure.

The properties of the two porous silica packing materials used

in this study, Porasil and Corning controlled pore glass (CPG), have

been examined by Cooper and Barrall (1973) and Cooper, Bruzzone, Cain,

and Barrall (1971), respectively. Porasil has a higher pore volume than

CPG (Cooper and Barrall, 1973) and therefore has higher VT to VO ratios.

Electron microscope examination has shown that Porasil has a more

heterogeneous pore structure than the CPG packings. Cooper and co-

workers have shown both packing materials to be effective for macro-

molecular separation. Cooper and Barrall (1973) concluded that the

heterogeneous pore structure of Porasil results in useful separations

over a wider range of molecular sizes than CPG. They also concluded

that the heterogeneous pore structure of Porasil precludes its use






11




for studying theoretical proposals relating polymer elution character-

istics to pore size dimensions.

A third type of rigid gel, Bioglass, is manufactured by a pro-

cess similar to the Corning glasses; however, it has an intentionally

broad pore size distribution (Cooper and Bruzzone, 1973).

The rigid gels exhibit severe adsorption properties (Cooper,

Johnson, and Porter, 1973; Dark and Limpert, 1973; Spatorico, 1975)

which are attributed to OH groups on the surface and Lewis acid sites

present from the manufacturing process. Adsorption effects may be

reduced through deactivation of OH groups; however, Lewis acid sites

are not deactivated by these procedures (Dark and Limpert, 1973).

Deactivation procedures include chemical treatment with polyethylene

oxide (Hiatt et al., 1971; Hawk, Cameron, and Dufault, 1972) or diethylene

glycol (LePage, Beau, and DeVries, 1968) and permanent deactivation by

silyation with hexamethyldisilazane (Cooper and Johnson, 1969) and

trimethylchlorosilane (Unger et al., 1974).

Commercial Porasil packing material is chemically deactivated

by adsorbed polyethylene oxide and distributed under the trade name

Porasil X.

Essentially all solvents are compatible with the porous silicas

and glasses, except alkaline solvents which will dissolve the silica

(Dark and Limpert, 1973). Spatorico and Beyer (1975) observed strong

adsorption to the porous glass of polymers containing cationic groups,

and found that treatment of the glass with polyethylene oxide, or

surfactants, was not successful in eliminating adsorption.





12




Loeppert and Volk (1976) investigated the use of HPLC for mole-

cular size fractionation of soil humic fractions on Porasil and Porasil

X and observed adsorption and electrostatic exclusion phenomena which

were highly dependent on solvent, saturating-cation, and concentration

of excess neutral electrolyte.

The polystyrene-divinylbenzene (DVB) gels, i.e. Poragel and

Styragel, are widely used in the polymer and petroleum industries

(Gaylor, James, and Weetall, 1976). Styragel and Poragel are not com-

patible with aqueous solvents, acetone, or alcohols (Dark and Limpert,

1973) and exhibit a high sensitivity to solvent polarity. Changes in

solvent may result in significant changes in the amount of solvation

and swelling of the gel matrix and altered pore-size distributions

of the gel. Therefore, it is usually necessary to pack the gel as a

slurry in the same solvent which is to be used as the eluting solvent.

Edwards and Ng (1968) studied the elution of model compounds on

polystyrene-DVB gels and observed an apparent adsorption of aromatic

compounds to the gel matrix. Adsorption of compounds on polystyrene-

DVB usually caused pronounced tailing (Bergmann, Duffy, and Stevenson,

1971). Cogswell, McKay, and Latham (1971) separated the acidic con-

centrate of petroleum distillate, using methylene chloride as solvent,

into four spectroscopically definable fractions and suggested molecular

association of the more acidic fractions in this solvent.

Sephadex has been widely used in studies of soil organic matter.

Although Sephadex is a different type of gel than the materials used

in these studies, a close examination of the material is in order






13



since previous results may be useful in developing separation schemes

and interpreting results with the newer gels. The reader is referred

to an excellent review by Swift and Posner (1971).

Sephadex G-gels have been shown to strongly adsorb aromatic com-

pounds (Gelotte, 1960), heterocyclic compounds (Demetriou et al.,

1968), and phenolic compounds (Sommers, 1966; Brook and Housley,

1969). Gel-phenol affinity is related to the ether bonds in the cross-

linking group rather than to the polysaccharide (dextran) component

of the gel matrix (Determann and Walter, 1968). As degree of cross-

linking of the gel was increased, affinity of phenol for the gel was

also increased. Brook and Munday (1970) suggested that benzene

derivatives are adsorbed onto hydroxyether cross-linking by H-bonds

and that interaction of Sephadex dextran gels with monosubstituted

phenols, anilines, and benzoic acids operates through hydroxyl, amino,

and carboxylic groups, respectively. Gelotte (1960) observed that the

Sephadex bed material contained a small amount of ionized groups,

probably COOH groups, at concentrations of approximately 10 meq per

gram of dry Sephadex. Aromatic amino acids were adsorbed to the bed

material, basic amino acids were strongly adsorbed, and acidic amino

acids were partially excluded from the gel. Demetriou et al. (1968)

likewise found that aromatic compounds with COOH substituents were

excluded from the gel beads when distilled water was used as the

eluting solvent. The same compounds were adsorbed when columns were

eluted with acid-salt solutions. Similar results were observed

during the elution of soil humic acids (Posner, 1966). Humic acid






14




was excluded from the gel matrix when eluted with water, due to the

charge effect. Considerable adsorption was evident when dilute

electrolyte was used as the eluent. It was concluded by Posner that

it is not possible to select a concentration of electrolyte which

would completely eliminate adsorption and exclusion effects.

In studies of lignosulfonate, Forss and Stenlund (1975) concluded

that elution of this material on Sephadex may be infuenced by the

following factors: (i) polyelectrolyte expansion, (ii) ion exclusion,

(iii) ion inclusion, and (iv) steric exclusion. The ion inclusion

effect results from interaction of charged sites in the macroions which

are excluded from the gel with charged sites in more permeable macro-

ions. Elution of the lignosulfonate preparation with excess electro-

lyte resulted in reduced exclusion of the sample due to reduction in

both the ion exclusion effect and the ion inclusion effect. In a study

of simple electrolytes on Sephadex (Neddermeyer and Rogers, 1968), it

was observed that peaks were badly skewed, with diffuse front and sharp

trailing edges. These skewed peaks were attributed to the ion exclusion

effect produced by ionic solutes and fixed charges in the gel.

Swift and Posner (1971) noted that when humic samples were

eluted with water and the concentration of solute was decreased, a

greater percentage of sample moved into the excluded or near excluded

region. This phenomenon was explained on the basis of double-layer-

theory and decreased suppression of charge at lower concentrations.

Gel-solute interactions were categorized according to coulombic forces,

caused by charged sites on gel and solute, and adsorption, caused by

hydrophilic interaction. Coulombic interactions were most prevalent






15



when distilled water was used as eluent and were reduced by adding

electrolyte. Swift and Posner suggested that fractionation based

solely on molecular weight can be achieved by using alkaline buffers

containing large amino cations.

















MATERIALS AND METHODS



Sample Pretreatment and Extraction


The soil selected for study was the surface horizon (0-25 cm) of

Terra Ceia muck, a Typic Medisaprist (Volk and Schnitzer, 1973). The

salt content was lowered using a dialysis technique employed by Khan

(1971) and refined by Loeppert and Volk (1974). Soil was poured as

a slurry into one dialysis bag, and cation-exchange resin (Amberlite

IR 120) was poured into a second dialysis bag. Both bags were placed

in 0.1 N HC1 and dialysis was continued until ash contents were lowered

to less than 1.0%. Recharged resin was placed daily in the dialysis

chamber. Following dialysis, samples were air-dried.

The extraction procedure is outlined in Fig. 1. Soil extracts

were obtained by 24-hour treatments with the appropriate extractant

(10:1 extractant:soil ratio). Extracts were centrifuged for 2 hours

at 16,300 x gravity (G), filtered through Whatman #42 filter paper,

and purified as described below.

Initial studies were performed to determine the yields and

properties of soil organic matter extracted from the Terra Ceia

muck surface horizon with selected solvents. The experimental pro-

cedures and results of these studies are reported elsewhere (Loeppert

and Volk, 1974; Snow, Loeppert, and Volk, 1974).

16





17






TERRA CEIA MUCK



H DMBF NaOH
EXTRACT EXTRACT EXTRACT



tHUMIC FULVIC
ACID ACID



SOXHLET SCHEME


HEXANE


BENZEIIH E








2-PROP N OL


I ETH T H A C e










Fig. 1 Extraction scheme






18



The DMF-extractable material was evaporated to dryness under vacuum

at 40C with a rotary evaporator and suspended in deionized water. The

sample was transferred to dialysis bags and dialyzed against deionized

water for 6 hours with frequent changing of the external dialysis solu-

tion, against 0.1 N HC1 in the presence of strong-acid ion-exchange

resin (Loeppert and Volk, 1974) for 48 hours, and against deionized

water until the external dialysis solution gave a negative test for

Cl This procedure was repeated until a sample with constant nitrogen

content and less than 0.5% ash was obtained. The sample was lyophilized

and stored at OC.

The water extract was treated similarly to the DMF extract except

the initial dialysis with water was omitted. The dialysis procedure

was repeated until the sample contained less than 0.5% ash.

The NaOH extract was acidified to pH 7.0 with 6.0 N HC1 and con-

centrated under vacuum at 30C with a rotary evaporator. The sample was

purified, lyophilized, and stored using the same procedure as with the

DMF extract.

A separate fraction of the NaOH extract was separated into humic

and fulvic acid fractions by adjusting the pH to 2.0 and purified ac-

cording to the procedure outlined by Stevenson (1965). Samples were

further purified by the dialysis procedure to an ash content less than

0.5%, lyophilized, and stored at OC.

Soxhlet fractions were obtained according to the scheme outlined

in Fig. 1 by successive 48-hour extractions with each solvent in the

series. Extracts were concentrated under vacuum at room temperature,






19




dried under a dry-nitrogen jet, and redissolved in the extracting sol-

vent at room temperature. Samples were concentrated to 30-ml volume

and following, addition of 30 ml of H20, were reconcentrated to 30-ml

volume. The reconcentration procedure was repeated several times, and

the aqueous suspensions were transferred to dialysis bags for purifica-

tion to less than 0.5% ash. Purified samples were lyophilized and

stored at OC.



Solubility Studies


The solubility behavior of extractable organic matter and each of

the Soxhlet fractions was determined by placing 2.00 mg of the organic

material in 2 ml of the appropriate solvent. Following agitation the

mixture was visually observed to determine whether the sample was in-

soluble, partially soluble or completely soluble. Where appropriate,

the pH of the sample suspension was adjusted to the desired level by

addition of acid or base which contained the same counter ion as the

excess electrolyte.



Analytical Determinations


Carbon and H were determined with the Coleman C-H analyzer, N by

the micro-Kjeldahl method, and S using the Leco induction furnace.

Oxygen was determined by difference with the assumption that C, H, N,

S,and 0 were the only elemental constituents of the extractable organic

matter. Total acidity was determined by addition of excess Ba(OH)2

and back-titration of unreacted Ba(OH)2 with HC1 to pH 9.8 (Schnitzer





20




and Gupta, 1965; Schnitzer and Khan, 1972). Total COOH-group concen-

tration was determined by addition of excess Ca(C2H302)2 and back-

titration of excess C2H302 with HC1 to pH 8.4 (Schnitzer and Gupta,

1965; Schnitzer and Khan, 1972). Total OH was determined by an

acetylation procedure (Brooks, Durie, and Sternhell, 1957) and total

C=O by the oximation method (Fritz, Yamamura, and Bradford, 1959).

Phenolic OH group concentration was calculated as the difference between

total acidity and COOH-group concentration, and alcoholic OH concentra-

tion was estimated by subtracting phenolic OH concentration from total

OH concentration. All elemental and functional group analyses were cor-

rected for moisture content and ash content.

Moisture content was determined by heating a preweighed combustion

boat with a known quantity of material to IIOC for 24 hours and weigh-

ing. Ash content was determined by heating a preweighed combustion boat

with a known quantity of organic material to 700C for 4 hours and weigh-

ing. All weights, yields, and concentrations are reported on a moisture-

and ash-corrected organic matter basis.

Infrared patterns were obtained on a Perkin-Elmer model 127

spectrophotometer. The KBr pellets were prepared by mixing 0.8 mg of

organic material, previously dried 24 hours over P205 in a vacuum

desiccator, with 200 mg of dried KBr in a Wig-L-Bug amalgamator. The

mixture was dried an additional 12 hours in a vacuum desiccator over

P205 prior to obtaining the spectrum.

Continuous-and stepped-potentiometric titrations were performed on

10 mg of organic material placed in the appropriate solvent. Titrant






21




was added using a Radiometer autoburette ABU 13 with TTT 50 titrator

module. Potentials were determined with a Radiometer PHM 64 pH meter

and recorded using a Radiometer REC strip chart recorder with REA 160

titrigraph module. A Radiometer combination electrode with porous plug

liquid junction was used for all potentiometric determinations. The

calomel cell was filled with saturated KCI in H20 for aqueous titra-

tions or with saturated KC1 in CH OH for titrations in nonaqueous

media (Loeppert, Zelazny, and Volk, 1976).



High Pressure Liquid Chromatography


The Waters' ALC 202 liquid chromatograph equipped with the Model

6000 solvent delivery system and 401 differential refractometer and

UV detectors was used for all separations. Samples were injected

through the U6K universal liquid chromatograph injector. A solvent

flow rate of 0.5 ml/minute were used, unless otherwise specified.

Column effluent was monitored by differential refractive index or by

UV absorption at 254 or 280 nm, and recorded continuously on a Perkin

Elmer 201 strip-chart recorder.

Samples were dissolved in the eluting solvent to give a 0.1% con-

centration (W/V) and centrifuged at 35,000 x G for 20 min. Injection

volume was 10 pl.

Solvents used in the chromatographic studies were deionized

water and spectroquality methanol, 2-propanol, t-butanol, tetrahydrofuran

(THF), dimethylformamide (DMF), ethylacetate, acetone, and chloroform.






22



Salt solutions in water and methanol were prepared by adding the de-

sired quantity of 2.0 N NaOH, tetramethylammonium hydroxide, or

tetrabutylammonium hydroxide to the solvent and adjusting to the

appropriate pH with HC1, HNO3, H2SO4, or H3PO4. Halide salts were

avoided where possible, since these salts may result in corrosion of

the stainless steel surfaces of the liquid chromatographic pump, con-

necting lines, and columns. In the few cases where halide salts were

employed, the pH of the eluting solvent was maintained above 7.0, and

the chromatographic system was flushed with copious quantities of de-

ionized water following use of the halide salt.

Characteristics of the packing materials used in this study

are summarized in Table 1. The silica and porous glass packing ma-

terials were dry-packed into stainless steel columns (0.318 cm OD x

2 f or 0.318 cm OD x 1 m). The semirigid gels were slurry packed in

the same solvent as the eluting solvent. Poragel columns (stainless

steel, 0.954 cm OD x 2 f) packed in THF were used as obtained from the

manufacturer. Poragel in DMF was slurry packed into 0.954 cm OD x 2 f

stainless steel columns. The p-Styragel columns (stainless steel,

0.954 cm OD x 1 f) packed in THF were used as obtained from the manu-

facturer. When used in series, columns were connected with U-shaped

0.009 in. ID tubing.

A series of experiments were performed to investigate the elu-

tion behavior of extractable soil organic matter on the gel permeation

packing materials: (i) determination of optimum operating conditions,











Table 1. Parameters of column packing materials

Approximate
b c d
Packing material Description Column V V N molecular weight
sizea T working range
cm ml ml

Porasil A porous silica 0.318 1.82 3.35 360 1,000-60,000

Porasil C 1.76 3.36 340 1,000-250,000
Porasil E 1.62 3.42 290 1,000-2,000,000
Porasil AX porous silica 1.84 3.32 340 1,000-60,000
Porasil CX deactivated with 1.88 3.42 280 1,000-250,000
polyethylene oxide
Porasil EX 1.67 3.49 230 1,000-2,000,000
CPG 40 crushed glass 1.60 3.22 1,020 500-10,000
CPG 250 1.54 3.31 660 5,000-100,000
Poragel 100 A polystyrene-DVB 0.954 10.8 21.3 800 500-20,000
Poragel 500 A 10.6 20.8 800 1,000-100,000

p-Styragel 100 A 5.2 10.1 5,600 100-3,000
p-Styragel 500 A 4.6 9.8 5,000 100-10,000

aAll columns are one meter long except the Poragel and p-Styragel columns which are 0.610 and 0.305 meters,
respectively. Column size indicates the outside column diameter.
bDetermined by elution of 1,600,000 molecular weight polystyrene standard with THF.
c
Determined by elution of acetone or benzene with THF.
Column efficiency expressed in theoretical plates.





24




(ii) determination of column parameters, (iii) elution behavior of soil

organic extracts and organic standards with selected solvents, (iv)

elution behavior of soil organic extracts and organic standards as

influenced by saturating cations, and (v) elution behavior of soil

organic extracts and organic standards as influenced by excess electro-

lyte.

















RESULTS AND DISCUSSION



Chemical Characteristics of Extractable Organic Matter


Extraction yields of 25% and 49% were obtained by single treat-

ments of the surface horizon of Terra Ceia muck with DMF and 0.5 N NaOH,

respectively, following pretreatment with HC1 to lower the ash content

(Table 2). The Soxhlet solvents were able to solubilize 29.6%, 35.0%,

and 9.4% of the NaOH-extractable material, DMF-extractable material, and

Terra Ceia muck, respectively. Only minor portions of the materials were

extracted by hexane and benzene. The major portions were in the ace-

tone, 2-propanol, and methanol.fractions. Each of these fractions, how-

ever, represent only a minor portion of the total soil organic matter.

Obvious and important differences exist in the elemental and func-

tional group concentrations of the extractable organic matter (Table 3).

Comparisons of the Soxhlet fractions indicate that C and H content

decreases, O and N content increases, COOH content increases, and total

acidity increases according to the solvent order: hexane, benzene,

ethylacetate, acetone, 2-propanol, and methanol. Potentiometric titra-

tions of the Soxhlet fractions in DMF indicated that titratable acidity

increased according to the same solvent order (Table 4).





25













Table 2. Yields of extractable soil organic matter

Terra Ceia
NaOH-extractable material DMF-extractable material muck
Soil organic Extract Soil organic Extract Soil organic
Extracting solvent basis basis basis basis basis
----------------------------------------%---------------------------------------

NaOH 49.0 100.0 -- 49.0

DMF -- -- 25.0 100.0 25.0

Hexane 0.1 0.3 0.1 0.2 0.1

Benzene 0.1 0.1 0.1 0.1 0.5

Ethylacetate 1.0 2.1 1.1 4.2 0.7

Acetone 1.4 2.9 2.1 8.4 1.2

2-Propanol 4.4 8.9 1.7 6.8 1.6

Methanol 7.6 15.3 3.8 15.3 5.3

Total Soxhlet 14.6 29.6 8.9 35.0 9.4

Soxhlet residue 34.4 70.4 16.1 65.0 90.6











Table 3. Elemental and functional group concentrations of extractable soil organic matter

Total Phenolic Alcoholic
Sample C H N S O acidity COOH C=O OH OH
---------- weight %--------- ---------------- meq/g ---------------------
DMF extract 56.2 5.7 4.1 0.6 33.4 7.2 3.5 2.6 3.7 2.5
Hexane-Soxhlet 81.2 9.1 0.1 0.1 9.5 -- -- --
Benzene-Soxhlet 77.7 8.8 0.2 0.1 13.2 -- -
Ethylacetate-Soxhlet 74.2 8.2 0.5 0.1 17.0 -- -
Acetone-Soxhlet 69.3 8.0 0.6 0.1 22.0 3.9 1.9 1.9 2.0 1.3
2-Propanol-Soxhlet 67.5 7.7 1.0 0.3 23.5 4.1 2.0 2.1 2.1 1.4
Methanol-Soxhlet 62.4 6.5 1.4 0.4 29.3 4.5 2.3 2.0 2.2 1.5
Residue 56.9 5.7 3.7 0.6 33.1 7.5 4.0 2.6 3.5 2.3

Humic acid 55.7 5.6 4.0 0.8 33.9 7.4 3.8 2.6 3.6 2.4
Fulvic acid 47.3 4.4 2.7 0.6 45.0 10.1 5.5 2.4 4.6 3.0

Water extract 50.2 4.9 2.6 0.6 41.7 8.7 4.4 2.8 4.3 2.9

NaOH extract 55.6 5.4 3.7 0.7 34.6 7.4 3.7 2.8 3.7 2.5
Hexane-Soxhlet 79.9 9.0 0.1 0.1 10.9 -- -- -- -
Benzene-Soxhlet 77.2 8.8 0.3 0.1 13.6 -
Ethylacetate-Soxhlet 75.4 8.6 0.5 0.1 15.4
Acetone-Soxhlet 69.1 8.1 0.7 0.2 21.9 3.7 1.9 1.7 1.8 1.2
2-Propanol-Soxhlet 68.4 7.4 1.1 0.3 22.8 4.3 2.2 2.1 2.1 1.4
Methanol-Soxhlet 61.7 6.9 1.4 0.3 29.7 4.6 2.4 2.2 2.2 1.4
Residue 56.1 5.5 3.9 0.7 33.8 7.5 3.8 2.8 3.7 2.5





28







Table 4. Titratable acidity of extractable soil organic matter



Solvent medium

Sample DMF H20



-----------meq/g-------------

DMF extract 5.9

Hexane-Soxhlet 1.4 --

Benzene-Soxhlet 1.9

Ethylacetate-Soxhlet 2.9

Acetone-Soxhlet 4.1 --

2-Propanol-Soxhlet 4.9 2.3

Methanol-Soxhlet 5.7 2.4

Residue 5.8 4.3


Humic acid 6.3 4.0

Fulvic acid 7.3 5.6


Water extract 6.9 5.3


NaOH extract 6.0 --

Hexane-Soxhlet 1.1

Benzene-Soxhlet 1.7

Ethylacetate-Soxhlet 2.5

Acetone-Soxhlet 4.2 --

2-Propanol-Soxhlet 4.8 2.0

Methanol-Soxhlet 5.3 2.6

Residue 5.7 3.9





29




Infrared patterns of the Soxhlet fractions (Fig. 2) substanti-

ated differences evident in the elemental and functional group analyses.

From the least polar to the most polar extracting solvents, Soxhlet

extracts showed progressively weaker C-H stretching vibrations in the

2900 cm- region. Absorption at 1450 cm1 due to the CH2 bending vibra-

tion also decreased according to the same solvent order. These adsorp-

tion bands of the infrared patterns, along with the C and H concentrations,

indicated that the less polar solvents extracted a material of greater

aliphatic character. Each of the Soxhlet fractions had strong C=O
-I -i
stretching bands at 1725 cm and 1630 cm resulting from COOH and
-i
COO-, respectively. The band at 1630 cm was considerably stronger

in the absorption spectra of materials extracted by the more polar
-l
solvents. The band at 700-750 cm evident in the absorption spectra

of hexane-, benzene-, and 2-propanol-extractable materials, was attributed

to aromatic C-H out-of-plane bending vibrations. The absence of bands

in this region of the spectrum indicates either the absence of aromatic

structure or the possibility of a completely substituted aromatic ring
-i
system. The weak absorption at 700-750 cm for the methanol-Soxhlet

extracts compared to the hexane-, benzene-, ethylacetate-Soxhlet extracts

may be attributed to a highly substituted aromatic ring system. Each

of the Soxhlet-extractable materials showed a strong absorption band
-i
at 3440 cm attributed to H-bonded OH groups and a weaker band at
-i
approximately 1250 cm- attributed to C-O stretching vibrations.

Absorption at both frequencies increased with increasing polarity of

the extracting solvent. The hexane- and benzene-Soxhlet fractions,
-i
especially, showed significantly weaker absorption in the 3440 cm

region than fractions extracted by the more polar solvents.







30















D1IF Extrac

NaOH Exlroct


1H Haon-Soihirst




B enrn.- Soxhs
n\Ety e \ctoe-S-SoohlI





SElyiocelotc-Sth-Sot

-Pr joeto e y ootosi
Acaloe- SeoxNl





fA 12-Propor-oIl -So&oi 2-Proponol-SoxhIsI



MhthcnSi-Seehst


S I i l I 1 1 1 I I l l I
4000 2000 1500 1000 SCO 700 3000 2000 1500 1000 800 700
FRE UENCY (cm') FREQUENCY (cmj1

Wavelength, Wavelength,










Fig. 2 Infrared patterns





31



DMF- and NaOH-extractable organic matter, and the humic acid

fraction had very similar elemental compositions and concentrations of

O-containing functional groups. The same observation was previously

made by Loeppert and Volk (1974) in comparisons of DMF-and NaOH-ex-

tractable materials. These similarities were corroborated by the infra-

red patterns (Fig. 2).

The fulvic acid fraction had lower C and H contents, higher O

content, lower N content, higher total acidity, and higher COOH content

than the humic acid and DMF-extractable materials. The higher acidity

of fulvic acid compared to the other materials was corroborated by the

significantly higher titratable acidity, as determined by potentiometric

titration in DMF.

In summary, C and H content decreased, aliphatic C-H decreased,

O and COOH contentsincreased, and total acidity increased according to

the following order of extractable organic matter: hexane-Soxhlet

extract, benzene-Soxhlet extract, ethylacetate-Soxhlet extract, acetone-

Soxhlet extract, 2-propanol-Soxhlet extract, methanol-Soxhlet extract,

NaOH extract--DMF extract--humic acid, fulvic acid.



Solubility Characteristics of Extractable Organic Matter


Solubility characteristics of extractable organic matter in

selected solvents and in salt solutions are summarized in Tables 5-7.

Humic acid, and DMF- and NaOH-extractable organic matter were completely

soluble at 0.1% concentration only in DMF, dimethylsulfoxide (DMSO), and





32





Table 5. Solubility of extractable soil organic matter in selected
solvents at 0.1% concentration


Solvent

Methyl- Methyl-
Ethyl- ethyl isobutyl
Sample Hexane Benzene acetate Acetone ketone ketone

DMF extract Ia I pa P P

Hexane-Soxhlet Sa S P P P P

Benzene-Soxhlet P S S S S S

Ethylacetate-Soxhlet I P S S S S

Acetone-Soxhlet I I S S S S

2-Propanol-Soxhlet I I P P P p

Methanol-Soxhlet I I I P I I

Residue I I I I


Humic acid I I P P P p

Fulvic acid I I I I I I


Water extract I I I I I I


NaOH extract I I P P p p

Hexane-Soxhlet S S P P P p

Benzene-Soxhlet P S S S S S

Ethylacetate-Soxhlet I P S S S S

Acetone-Soxhlet I I S S S S

2-Propanol-Soxhlet I I P P p p

Methanol-Soxhlet I I I P I I

Residue I I I I I I
S = soluble, P = partially soluble, I = insoluble





33





Table 5 (Extended)



Solvent


Ethyl THF t-Butanol 2-Propanol Methanol Pyridine DMF DMSO H20
ether2

I P P P P P S S P

P S P I I T P P I

S S P I I P S P I

P S S I I S S S I

P S S P P S S S I

I S S S S S S S I

I P P P S S S S P

I I I I P P S S P


I P P P P P S S P

I P P P S S S S S


I P P P P P S S S


I P P P P P S S P

P S P I I I P P I

S S P I I P S P I

P S S I I S S S I

P S S P P S S S I

I S S S S S S S I

I P P P S S S S P

I I I I P P S S P





34





Table 6. Solubility of extractable soil organic matter as influenced
by saturating cation and solvent



Solvent
Saturating
Sample cation H 0 CH OH 2-Propanol THF DMF DMSO


DMF extract H Pa P P P Sa S
Na S S S Ia P S
N(CH3)4 S S S P P S
N(C H ) S S S P P S

DMF extract
acetone-Soxhlet H I P P S S S
Na S S S P P P
N(CH ) S S S P P P
N(C H ) S S S P P S

DMF extract
2-propanol-Soxhlet H I S S S S S
Na S S S P P P
N(CH3)4 S S S P P P
N(C4H9)4 S S S P P S

DMF extract
methanol-Soxhlet H P S P P S S
Na S S S P P P
N(CH )4 S S S P P P
N(C H ) S S S P P S

Humic acid H P P P P S S
Na S S S P P P
N(CH S S S P P S
N(C H 9)4 P S S P S S

Fulvic acid H S S P P S S
Na S S S P P P
N(CH ) S S S P P S
N(C4H) S S S P S S



S = soluble, P = partially soluble, I = insoluble





35









Table 7. Solubility of fulvic acid in aqueous salt solutions



pH
Saturating Excess
cation electrolyte Concentration 2.0 4.0 6.0 8.0 10.0


Na Na SO 0.000 N Sa S S S S
0.001 N S S S S S
0.01 N S S S S S
0.05 N S pa p p S
0.10 N S P P P S
0.10 N S P P P S

K K2SO 0.000 N S S S S S
0.001 N S S S S S
0.01 N S S S S S
0.05 N S P P P S
0.10 N S P P P S

N(CH3)4 [N(CH3)4]2S4 0.000 N S S S S S
0.001 N S S S S S
0.01 N S S S S S
0.05 N S S S S S
0.10 N S P P P S

N(C4H9)4 [N(C H) 4]2SO 0.000 N S S S S S
0.001 N S S S S S
0.01 N S S S S S
0.05 N S S S S S
0.10 N S S S S S



aS = soluble, P = partially soluble





36




0.5 N NaOH. The DMF and DMSO both have significant basic character

(Talhoun and Mortland, 1968). The acidic organic material is highly

dissociated and dispersed in each of these solvents and, therefore, is

soluble. Each of the Soxhlet fractions and fulvic acid were also com-

pletely soluble at 0.1% concentration in DMF and DMSO.

Fulvic acid was soluble in methanol and water, in addition to DMF

and DMSO, but was not completely soluble in any of the other solvents.

The solubility of extractable organic matter and Soxhlet fractions was

influenced to a great extent by the saturating cation (Table 6). Ex-
+ ++ + +
change of H by Na K N(CH3 ), or C(C H9) resulted in increased

solubility of the organic solutes in water, methanol, or 2-propanol,

and decreased solubility in DMF. For example, at 0.1% concentration,

the H-saturated DMF- and NaOH-extractable materials were only partially

soluble in water, methanol, or 2-propanol; however, the salt-saturated

solutes were completely soluble. On the other hand, the H-saturated

2-propanol-Soxhlet fraction and fulvic acid were soluble at 0.1% concen-

tration in DMF whereas the salt-saturated material was only slightly

soluble. These solubility characteristics greatly limit the solvent-

electrolyte combinations which are applicable for exclusion chroma-

tography. The enhanced solubility of the cation-saturated samples in

the protic solvents (water, methanol, and 2-propanol) may be attributed

to acidic properties of these solvents (King, 1973) which promote

stabilization of the solute anion. The very weakly acidic dipolar

aprotic solvents (e.g. THF, DMF) would not stabilize the solute anion

to as great an extent as the protic solvents.





37




The presence of excess neutral salt affected the solubility of

fulvic acid in water (Table 7) and methanol. Some very interesting

trends were evident in these studies. The H-, Na-, K-, N(CH3)4-, and

N(C Hd4-saturated fulvic acid samples at 0.1% concentration were

soluble at pH 2.0 in all concentrations of excess neutral electrolyte
-3 -2
up to 0.1 N. In the presence of 10 N or 102 N excess salt, 0.1%

fulvic acid remained completely dissolved as the pH was increased suc-

cessively to pH 4.0, 6.0, 8.0, and 10.0. However, in the presence of
-2
5 x 10-2 N excess salt, K-saturated fulvic acid began to precipitate

at pH 4.0. The sample redissolved at pH 8.0 and was completely soluble

as the pH was increased to 10.0. Fulvic acid saturated with N(CH3)4

or N(C4H9+ remained completely dissolved as the pH was increased from
-2
2.0 to 10.0, in the presence of 5 x 10 N excess neutral salt. As the
-1
ionic strength was increased to 5 x 10 N, however, K-, Na-, N(CH )4-,

and N(C4H9)4-saturated fulvic acid precipitated as the solution pH ap-

proached 4.0 and redissolved as the pH approached 10.0. The precipita-

tion was greatest in the approximate range of pH 4 to pH 7 and may be

attributed to unfavorable conditions for the electrostatic dispersion

of molecular units. It is interesting to note that greatest precipita-

tion occurred within the pH range at which greatest neutralization of

acidic carboxyl groups would occur.

The precipitation phenomena in the presence of excess neutral salt

greatly limits the conditions which may be employed for gel permeation

separations of extractable organic matter.





38




High Pressure Liquid Chromatography


Porous Silica Packing Materials


Operating conditions. Column efficiencies of Porasil A and

Porasil AX packing materials were greatly influenced by solvent flow

rates. With 0.318-cm OD analytical columns, maximum column efficiencies

and minimum peak broadening were obtained at flow rates of approximately

0.1 ml/min.; however, column efficiencies were not significantly differ-

ent at flow rates between 0.1 and 0.6 ml per min. (Fig. 3). At flow

rates greater than 0.6 ml per min., peak broadening was increased and

column efficiencies were decreased. For this reason, it was concluded

that low flow rates should be maintained with the Porasil packing

materials.

It is interesting to observe that for the Porasil packing materials,

there was a slight increase in column efficiency as flow rate was de-

creased to 0.1 ml per min. (Fig. 3). These results may be compared with

those obtained with the CPG packing materials for which maximum column

efficiencies were obtained at flow rates of 0.4 ml per min. Decreases

in column efficiency were observed when flow rates were decreased or

increased from this value. As with the Porasil packing material, in-

creases in solvent flow rates above 1.0 ml per min. resulted in peak

broadening and significant decreases in column efficiency.

The different behavior of the Porasil and CPG packing materials

at low flow rates may be at least partially attributed to the more





39





700-



600 o PORASIL A
A PORASIL AX
z 10 CPG-250
.500
z
0
0 400



aL 300



---
I 200

O

" 100




1.0 2.0 3.0 4.0 5.0
FLOW RATE, mi/min



Fig. 3 Effect of flow rate on column efficiency, N, of Porasil
A. Porasil AX, and CPG-250 analytical columns






40



uniform pore structure of the controlled pore glass material which may

result in reduced blockage of large pores by small pores and less re-

stricted diffusion of solute molecules through the gel matrix. The

more uniform pore structure may allow the use of higher flow rates.

Cooper and Barrall (1973) suggested that "pooling" or solute restric-

tion in porous silica media, necessitates the use of low flow rates

with these materials.

Based on these studies, solvent flow rates of 0..5 ml per min.

were selected for all subsequent studies using the 0.318-cm OD columns

with the Porasil and CPG packing materials. Based on similar studies

with 0.954-cm OD preparative columns, solvent flow rates of 1.5 ml

per min. were used for all subsequent studies on these columns packed

with Porasil or CPG packing materials.

The effects of sample size on column efficiencies of Porasil and

CPG packing materials are summarized for the analytical columns (Fig.

4). In general, the maximum sample volumes were 20 il for the analyti-

cal columns and 100 ul for preparative columns. Larger sample volumes

resulted in increased peak broadening and reduced apparent column

efficiencies. Column efficiencies were not noticeably affected with

lower sample volumes.

Even though column efficiencies are reduced with large sample

volumes, column overloading may be helpful in obtaining preparative

fractions, especially when used in conjunction with recycle chroma-

tography (Bombaugh, 1971).




41








700



600
z


) 500
0
0
U
F 400
-J
o-
S300



200
0

S o PORASIL A
100 A PORASIL AX
0 CPG-250

I I 1 I I i 1 i i 1
0.100 0.200
SAMPLE SIZE, ml

Fig. 4 Effect of sample size on column efficiency, N, of
Porasil A. Porasil AX. and CPG-250 analytical columns





42




Column parameters. Elution characteristics of packed columns are

summarized in Table 1. Column parameters, V0 and V were determined

from the elution volumes for acetone or benzene and blue dextran 2,000

or 2,600,000 molecular weight polystyrene, respectively. In all cases

the column efficiencies, indicated by theoretical plate count, N, de-

creased with increasing internal pore size of the packing material.

For example, the column efficiencies of Porasil AX, CX, and EX were

340, 280, and 230 theoretical plates per meter, respectively.

Molecular weight calibration curves (Fig. 5) were obtained by

elution of polystyrene standards with THF. The approximate molecular

weight working ranges, as determined with the polystyrene standards

are summarized in Table 1. The working curves obtained with polystyrene

standards on the Porasil and the CPG packing materials were not linear

over the working range of the gels.

Effect of solvent. Peak elution volumes of extractable organic

matter on Porasil A, Porasil AX, and CPG-250 packing materials are

summarized in Tables 8-10, respectively. Elution patterns of the

2-propanol-extractable material and fulvic acid are shown in Figs. 6

and 7. Samples were completely soluble at 0.1% concentration in each

of the solvents shown. A portion of the fulvic acid sample (Fig. 7) was

eluted at VO, the elution volume of a nonreactive high molecular weight

solute, on the Porasil A column when methanol, water, or DMF was used

as the eluting solvent. The relative quantity of sample eluted at VO

increased according to the following solvent order: methanol < DMF < H20.

Likewise, portions of the acetone-,2-propanol- and methanol-Soxhlet




43










106-


05
-3 0
w
C 10 5



4
D 10
0
-J
I

_3

0 oPORASIL AX
oJ CPG-250

102





1.0 2.0 3.0 4.0
ELUTION VOLUME, ml



Fig. 5 Molecular weight calibration curves of 1 m x 0.318 cm
OD Porasil AX and CPG analytical columns obtained by
elution of polystyrene standards with THF





44






Table 8. Peak elution voluems of extractable soil organic matter on
Porasil A with selected solvents

Solvent
Sample H 0 CH OH 2-Propanol t-Butanol Acetone THF DMF

-----------------------ml--------------------------

DMF extract -- -- -- -- -- 1.75

Acetone-Soxhlet -- 1.61 1.57 1.65 ADb 3.29 1.83
(3.18)a (3.19) (3.30) (3.17)

2-Propanol-Soxhlet -- 1.58 1.55 -- 3.32 1.79
(3.17) (3.15)

Methanol-Soxhlet -- 1.61 -- 3.31 1.82
(3.19)


NaOH extract -- -- -- 1.78

Acetone-Soxhlet -- 1.63 1.58 1.62 AD 3.31 1.74
(3.17) (3.18) (3.32)

2-Propanol-Soxhlet -- 1.60 1.57 -- 3.32 1.75
(3.16) (3.13)

Methanol-Soxhlet -- 1.60 3.35 1.76
(3.19)


Humic acid -- -- -- -- -- -- 1.75

Fulvic acid 1.57 1.76


Parentheses ( ) indicate secondary peak.

AD = severe adsorption.





45






Table 9. Peak elution volumes of extractable soil organic matter on
Porasil AX with selected solvents

Solvent
Sample H20 CH3 OH 2-Propanol t-Butanol Acetone THF DMF


----------------------ml--------------------------

DMF extract -- -- --

Acetone-Soxhlet -- 1.82 3.35 3.42 AD 3.39
(3.30)a AD AD

2-Propanol-Soxhlet -- 1.75 AD -- 3.40
(3.28)

Methanol-Soxhlet -- 1.80
(3.31)


NaOH extract -- --

Acetone-Soxhlet -- 1.81 3.33 3.37 AD 3.35
(3.31) (AD) (AD)

2-Propanol-Soxhlet -- 1.79 AD -- 3.39
(3.32)

Methanol-Soxhlet -- 1.82
(3.27)


Humic acid --

Fulvic acid 1.68 1.73


Parentheses ( ) indicate secondary peak.

AD = severe adsorption.





46






Table 10. Peak elution volumes of extractable soil organic matter on
CPG-250 with selected solvents


Solvent
Sample H20 CH3OH 2-Propanol t-Butanol Acetone THF DMF


----------------------ml--------------------------

DMF extract -- -- -- -- 1.77

Acetone-Soxhlet -- 1.75 3.73 3.97 AD 3.87 1.82
ADa AD

2-Propanol-Soxhlet -- 1.70 3.71 -- 3.85 1.75

Methanol-Soxhlet -- 1.71 -- -- -- -- 1.77


NaOH extract -- -- -- -- -- -- 1.79

Acetone-Soxhlet -- 1.76 3.84 4.12 AD 3.81 1.81

2-Propanol-Soxhlet -- 1.71 3.67 -- 3.83 1.76

Methanol-Soxhlet -- 1.69 -- -- -- 1.79


Humic acid -- -- -- -- -- 1.76

Fulvic acid 1.71 1.74 -- -- 1.72



AD = severe adsorption





47




DMF ,


CL\
z CH30H "'



T111
J CH30H


c THF
a-

Li THF

---PORASIL A
-PORASIL AX

I Vo 2 3 VT 4
ELUTION VOLUME,ML

Fig. 6 Elution patterns of the 2-propanol-Soxhlet extract of
DMF-extractable material on Porasil A and Porasil AX
with selected solvents






H20


cH20


cnCH30H



S\ CH30H
or
0











Porasil AX with selected solvents
UJ ---PORASIL A

-PORASIL AX


Vo 2 3 VT 4
ELUTION VOLUME,ML

Fig. 7 Elution patterns of fulvic acid on Porasil A and
Porasil AX with selected solvents





48




extracts (Table 8;Figs. 6 and 7) were eluted at VO when methanol or

DMF was used as the eluting solvent. A relatively larger quantity

was eluted at VO with DMF than with methanol. With each of the above

solvents, the organic solute was eluted prior to VT. On the contrary,

when acetone, t-butanol, or THF was used as the eluting solvent with

the Porasil A column, a portion of the organic solute was eluted past

VT, indicating an adsorptive interaction with the silica packing

material.

Deactivation of the Porasil surface (Porasil X) resulted in re-

duced exclusion of organic solute from the gel matrix in water and

methanol compared to the activated material (Figs. 6 and 7; Table 9).

Adsorption was reduced on the Porasil AX compared with the Porasil A

packing material, although a small portion of the organic solute was

still eluted past VT with t-butanol, acetone, and THF on Porasil AX.

Elution volumes of organic standards on Porasil A, Porasil AX,

and CPG-250 are shown in Tables 11-13, respectively. On Porasil A,

several of the organic acid standards (1,2,4,5-tetracarboxybenzene and

1,3,5-tricarboxybenzene) were eluted at a solvent volume equivalent

to V when DMF, water, or methanol was used as the eluting solvent and

at a solvent volume slightly greater than VT when acetone, 2-propanol,

t-butanol or THF was used as the eluting solvent.

In water, methanol, and DMF, the more highly substituted aromatic

acids were in general eluted at a smaller solvent volume than the less

substituted acids. For example, elution volume increased according to





49









Table 11. Peak elution volumes of organic standards on Porasil A with
selected solvents



Solvent
Sample H20 CH OH 2-Propanol t-Butanol Acetone THF DMF

----------------------ml--------------------------

1,2,4,5-Tetracar-
boxybenzene 1.73 1.72 3.70 9.25 -- 3.55 2.36

1,3,5-Tricarboxy-
benzene 1.76 1.94 ADa 3.68 -- 3.54 3.60

3,5-Dihydroxyben-
zoic acid 2.08 4.87 3.33 3.47 3.43

Benzoic acid 2.37 5.29 3.51 3.59 3.58 3.44

Pyridine 5.20 AD AD 4.45 3.87 3.58

Aniline 3.44 4.74 4.04 3.55 3.62 3.41

Methanol -- 3.64 3.94 3.91 3.42 3.94

Ethylene glycol 3.40 4.45 4.76 3.97 3.45 3.46

Acetone 3.36 3.97 4.52 -- 3.70 3.46




AD = severe adsorption






50










Table 12. Peak elution volumes of organic standards on Porasil AX
with selected solvents



Solvent
Sample H20 CH3OH 2-Propanol t-Butanol Acetone THF

-------------------ml------------------

1,2,4,5-Tetracarboxy-
benzene 1.83 2.58 10.25 12.00 -- 3.48

1,2,5-Tricarboxy-
benzene 2.11 2.71 4.95 3.49 -- 3.39

3,5-Dihydroxybenzoic
acid 2.53 2.85 3.59 3.28 -- 3.46

Benzoic acid 2.39 2.97 3.77 3.40 3.49 3.34

Pyridine 11.06 3.42 3.67 ADa 3.84 3.60

Aniline 4.69 3.38 3.52 3.55 3.52 3.49

Methanol 3.48 -- 3.44 3.41 3.65 3.71

Ethylene glycol 3.56 3.43 3.52 3.56 3.65 3.40

Acetone 3.58 3.38 3.44 3.46 -- 3.40




AD = severe adsorption





51








Table 13. Peak elution volumes of organic standards on CPG-250 with
selected solvents



Solvent
Sample H20 CH30H 2-Propanol t-Butanol Acetone THF DMF


------------------------ml------------------------

1,2,4,5-Tetracarboxy-
benzene 1.71 1.74 4.12 10.03 3.91 2.44

1,3,5-Tricarboxyben-
zene 1.78 1.93 ADa 4.27 3.93 3.38

3,5-Dihydroxybenzoic
acid 2.18 5.03 4.06 -- 3.91 3.86

Benzoic acid 2.26 5.36 4.99 3.98 3.92 3.88

Pyridine 4.89 AD AD 5.02 4.34 4.17

Aniline 4.12 4.72 4.76 4.17 4.10 3.95

Methanol -- 3.98 4.27 4.26 3.87 4.31

Ethylene glycol 3.87 5.01 5.38 4.37 3.89 4.26

Acetone 3.82 4.34 4.87 -- 4.06 3.92




AD = severe adsorption






52




the following solute order: 1,2,4,5-tetracarboxybenzene < 1,3,5-

tricarboxybenzene < 3,5-dihydroxybenzoic acid < p-hydroxybenzoic acid

< benzoic acid. Several compounds with basic properties (e.g. aniline,

pyridine) were adsorbed and eluted past VT. Also, several other com-

pounds (e.g. glucose, ethylene glycol, and sucrose) were eluted past VT.

Each of the compounds which showed strong evidence of adsorption on

Porasil A with water, methanol, or DMF as eluting solvent contained an

amino group or an aliphatic OH.

In THF, aromatic acids and simple alcohols each showed evidence

of adsorptive interaction with Porasil A. Elution on Porasil AX, the

deactivated analog of Porasil A, resulted in reduced adsorption.

Elution patterns of organic acid standards on Porasil A and

Porasil AX showed interesting similarities to the elution patterns of

extractable soil organic matter. Based on the molecular weights of

tetracarboxybenzene and tricarboxybenzene and the working molecular

weight ranges of the gels suggested by the manufacturer, one would

expect that the solute would elute at, or slightly before, V Inspec-

tion of the patterns, however, shows that the acid standards were

completely excluded from the gel matrix and eluted at VO when water

was used as the eluting solvent. Deviations from the expected elution

behavior of a low molecular weight nonreactive solute may be attributed

to adsorption, electrostatic exclusion, or molecular association.

The exclusion of the acidic organic solute from Porasil A in

water, methanol, or DMF may be attributed to (i) the porous structure

of the gel, (ii) association of solute molecules, and/or (iii) electro-

static exclusion from the porous matrix. The first explanation is





53




unlikely since the nonreactive solute, acetone, produced a symmetrical

peak at VT with negligible skewing, indicative of free entrance into

the porous gel matrix. Association of the solute molecules in water,

methanol, and DMF would be questionable since aggregation of the mole-

cular units should be greatest in the least polar and/or least basic

solvent. Comparison of the individual solvents shows that water, DMF,

and methanol have stronger basic character and are considerably more

polar than THF with dielectric constants of 76.2, 36.7, 32.6, and 7.58,

respectively. Also, deactivation of the porous silica resulted in

increased elution volumes, which should not have been the case if

skewing was entirely due to aggregation of the solute molecules.

With the first two explanations above eliminated as probable

major causes of exclusion of the low molecular weight acidic solute,

the third explanation, electrostatic exclusion,deserves careful con-

sideration. The acidic functional groups of the organic solute would

be partially dissociated in water, methanol, or DMF due to the basic

character of each of these solvents. Water and methanol have basic

character which is attributed to the presence of the electron-donor

oxygen atom. The DMF molecule has two basic sites (Talhoun and Mort-

land, 1968), the electron-donor oxygen atom of the carbonyl group and

the nitrogen atom. The surface Si(OH) groups of the silica packing

material are weak acid sites. In water (Kirkland, 1971), methanol, or

DMF the surface sites may dissociate, due to the basic properties of

these solvents, resulting in a negatively charged silica surface.





54




Especially in water and methanol, the negative surface sites would be

stabilized as a result of the acidic properties of the solvent mole-

cules. The exclusion of acidic organic solute from the porous matrix

of Porasil A may therefore be at least partially attributed to electro-

static repulsion between the charged solute molecules and the charged

silica surface. In the absence of excess neutral salt, the silica

would have an expanded electrical double layer and the solute molecules

would exist with larger effective radii. Therefore, it is possible

that low molecular weight solutes may be completely excluded from the

porous gel matrix.

As mentioned previously, there was no evidence of adsorptive inter-

action between the silica surface and the acidic solute in water,

methanol, and DMF; however, in 2-propanol, t-butanol, acetone, ethylace-

tate, and THF there was evidence of adsorption. In the former solvents,

the greater negative charge densities of the solutes and the silica

surface may have resulted in less adsorptive interaction between the

negatively charged species. In acetone, ethylacetate, and THF, however,

the organic solute would be much less dissociated as a consequence of

the very weak or negligible basic properties of these solvents. Also,

the silica surface would be less highly dissociated. Therefore, there is

a more favorable condition for direct H-bonding interactions between the

silica surface and the solute molecules.

Deactivation of the silica surface with polyethylene glycol would

block the reactive sites (Dark and Limpert, 1973) and result in reduced

negative charge density of the silica surface in water and methanol.





55




Therefore, electrostatic exclusion of negatively charged solute was

reduced on Porasil AX compared to Porasil A. When acetone, ethylace-

tate, or THF was used as the eluting solvent on Porasil AX, only a small

quantity of acidic solute was eluted past VT. This phenomenon indicates

a reduction in adsorptive interaction between the silica packing

material and the acidic solute on Porasil AX compared to Porasil A.

Adsorption and electrostatic exclusion werereduced on Porasil

AX, but were not completely eliminated. The evidence of adsorption and

electrostatic exclusion interactions between the solute and the Porasil

AX demonstrated that the packing material was not completely deactivated.

Effect of saturating cation. Fulvic acid in which the acidic

functional groups were saturated to pH 7.0 with Na K N(CH3)4, or

N(C H9)4 were eluted at VO on Porasil A and CPG-250 when water was used

as the eluting solvent (Table 14). The cation-saturated samples were

also completely excluded from the gel matrix on deactivated Porasil AX.

Likewise, both the 1,2,4,5-tetracarboxybenzene and the 1,3,5-tricar-

boxybenzene in methanol and water were excluded from the Porasil A and

Porasil AX gel matrices.

The pronounced exclusion of Na-, K-, N(CH3)4-, and N(C4H9) -

saturated fulvic acid and organic acid standards from the Porasil A

gel matrix may be attributed to electrostatic repulsion of the nega-

tively charged solute molecule and the negatively charged sites on the

silica surface. Similar exclusion phenomena have been observed during

elution of cation-saturated fulvic acid with distilled water on

Sephadex (Swift and Posner, 1971).





56












Table 14. Peak elution volumes of cation-saturated fulvic acid on
Porasil A, Porasil AX, and CPG-250 with water as eluting
solvent



Column

Saturating Porasil Porasil CPG-
cation A AX 250
-----------------------ml----------------------



H 1.62 1.68 1.79


Na 1.61 1.69 1.76


N(CH3)4 1.62 1.70 1.78


N(C4H9) 4. 1.60 1.69 1.76





57




Electrostatic exclusion phenomena have been reported for elution

of acidic amino acids (Gelotte, 1960), aromatic acids (Demetriou et al.,

1968), inorganic ions (Neddermeyer and Rogers, 1968), and lignosulfonate

(Forss and Stenlund, 1973) on the Sephadex G-gels and was attributed to

electrostatic repulsion between fixed charges on both the gel and the

solute molecules.

Effect of excess electrolyte. Elution patterns of Na-saturated

fulvic acid in the presence of excess neutral salt on Porasil A and

Porasil AX (Figs. 8 and 9, respectively) indicated that electrolyte

resulted in reduction in the relative quantity of solute excluded from

the gel matrix. Presence of neutral salt also influenced elution pat-

terns of low-molecular weight organic acids on Porasil A and Porasil AX

(Figs. 10 and 11, Tables 15 and 16) and resulted in reduced exclusion

of solute from the gel matrix. Similar phenomena have been observed

with Sephadex during the elution of soil organic matter extracts (Swift

and Posner, 1971; Posner, 1963), acidicamino acids (Gelotte, 1960),

aromatic acids (Demetriou et al., 1968), and lignosulfonates (Forss

and Stenlund, 1973).

At salt concentractions above 0.01 N, significant quantities of

fulvic acid were eluted past VT. The excess electrolyte resulted in

suppression of negative charge and reduction in thickness of electrical

double layer of both the negatively charged gel surface and the organic

solute molecules. Also, excess salt would decrease the effective size

of solute anions due to reduction in thickness of the electrical double

layer. Therefore, solute anions would more easily enter the porous





58




H20


z 0.001 0
cn
Li
O


rr
UJ



a ;0.05 N





I Vo 2 3 VT 4
ELUTION VOLUME,ML


Fig. 8 Effect of excess neutral electrolyte on elution of
Na-saturated fulvic acid on Porasil A









Li
U) \ H20
z
0

r ) / 0.001 N


WKI ---\- \ o^-- -
0r
LI


0.01 N

C .05 N


I Vo 2 3 VT 4
ELUTION VOLUME,ML


Fig. 9 Effect of excess neutral electrolyte on elution of
Na-saturated fulvic acid on Porasil AX





59



HzO20





(n
z
o 0.001 N
a-



o 0.05 N
0






I Vo 2 3 VT 4
ELUTION VOLUME.ML


Fig. 10 Effect of excess neutral electrolyte on elution of
Na-saturated 1,2,4,5-tetracarboxybenzene on Porasil A




H20


mn 0.001 N
Z



0.01 N
0

o 0.05 N
V/
LiJ
Cr



I Vo 2 3 VT 4
ELUTION VOLUME,ML

Fig. 11 Effect of excess neutral electrolyte on elution of
Na-saturated 1,2,4,5-tetracarboxybenzene on Porasil AX





60













Table 15. Peak elution volumes of selected organic acid standards on
Porasil A with Na2SO4 solutions




Electrolyte concentration

Sample 0.000 N 0.001 N 0.01 N 0.05 N

--------------------ml-------------------



1,2,4,5-Tetracarboxy-
benzene 1.74 1.68 1.86 2.60


1,3,5-Tricarboxyben-
zene 1.75 1.68 1.91 2.64


3,5-Dihydroxybenzoic
acid 1.82 1.97 2.32 2.71


Benzoic acid 1.96 2.41 2.51 3.14





61











Table 16. Peak elution volumes of selected organic acid standards on
Porasil AX with Na2SO4 solutions



Electrolyte concentration

Sample 0.000 N 0.001 N 0.01 N 0.05 N

----------------------ml--------------------



1,2,4,5-Tetracarboxy-
benzene 1.87 2.31 3.03 3.21


1,3,5-Tricarboxyben-
zene 2.18 2.43 3.05 3.17


3,5-Dihydroxybenzoic
acid 2.21 2.80 3.19 3.28


Benzoic acid 2.41 3.09 3.40 3.48





62





gel matrix. As double-layer thickness decreased with resulting reduc-

tion in electrostatic exclusion, adsorption increased due to direct H-

bonding interactions of oxygen-containing groups on the organic solute

and Si(OH) sites on the packing material. Adsorption effects were re-

duced on Porasil X series of packing materials but were not completely

eliminated. This evidence indicated that the silica surface was not

completely deactivated and/or solute molecules were interacting directly

with the silica surface.

The solubility studies (Table 7) indicated that fulvic acid began

to precipitate at electrolyte concentrations above 0.01 N, at pH values

of 4.0 to 8.0. Therefore, the partial elution of fulvic acid past VT

at the higher electrolyte concentrations may be caused by precipitation

of fulvic acid in the column. For this reason, electrolyte concentra-

tions must be maintained at values low enough to preclude precipitation

of the solute.

Because of the nature of the silica surface, special precautions

must be observed. As mentioned previously, the silica surface acts as a

weak acid due to the presence of SiOH groups. In the presence of a

protic solvent with basic properties, such as water or methanol, these

acid sites will dissociate, leaving the silica surface with a net nega-

tive charge. As the pH of the solvent medium is increased, the dis-

sociation of surface sites and the negative charge density of the

silica surface is also increased. The negative charge density of

acidic solute molecules would also increase with increasing pH. There-

fore, exclusion of negatively charged solute from the negatively

charged packing material would increase with increasing pH.






63




At pH values above 7.0, the silica surface may be destroyed by

solubilization and formation of silicate. Therefore, alkaline condi-

tions must be avoided. Under alkaline conditions, the chemically

adsorbed deactivating agent is also readily stripped from the surface

of the Porasil X packing material. The manufacturer recommends that

use of several organic solvents, especially DMF, should be avoided

with deactivated Porasil. Such a solvent may readily strip the de-

activating agent from the silica surface.


Polystyrene-divinylbenzene (DVB)


Operating conditions. As with the porous glass packing materials,

column efficiencies of Poragel and o-Styragel packing materials were

greatly influenced by solvent flow rates. With 0.054-cm OD columns and

THF as the eluting solvent, minimum peak broadening and maximum column

efficiencies were obtained at flow rates of approximately 0.8 and 3.0

ml per min. for the 100 A Poragel and o-Styragel columns, respectively

(Fig. 12). At higher flow rates, peak broadening was increased and

column efficiencies were decreased. At lower flow rates, column effi-

ciencies were also appreciably lowered. This later effect was much

more evident with the polystyrene-DVB gels than with the porous glasses.

At low flow rates, diffusion of solute molecules apparently resulted

in decreased column efficiencies. With the p-Styragel columns, the

high column efficiency at the high flow rate was due to the small

particle size and uniform pore size of packing which permitted rapid





64








7000

ol00 A j-STYRAGEL(lf)
0
6000 100 A PORAGEL (2f)

2z
6,00


z
D 5000 -
0







< 3000

-\

W 2000
O
ui
1-
1000-




2.0 4.0 6.0 8.0 10.0
FLOW RATE, ml/min

0
Fig. 12 Effect of flow rate on column efficiency, N, of 100 A
Poragel and 100 A U-Styragel preparative columns with
THF as the eluting solvent





65




equilibrium of solute molecules between internal pore space and inter-

stitial pore space.

Based on these studies, solvent flow rates of 0.8 and 3.0 ml per

min. in 0.954-cm OD columns were used for Poragel and i-Styragel

columns, respectively, in all subsequent studies.

The effect of sample size on column efficiencies of 0.954-cm

diameter columns of Poragel and p-Styragel are shown in Fig. 13. In

general, the maximum sample volumes were 250 pl and 25 pl for the

Poragel and i-Styragel columns, respectively. Larger samples resulted

in increased peak broadening and reduced apparent column efficiencies.

Column parameters. Column parameters (Table 1), VO and V of the

packed columns were determined by elution of acetone or 2,600,000 molecu-

lar weight polystyrene, respectively. Column efficiencies, N, were
0 0
approximately 800 and 5,000 for the 100 A Poragel and 100 A p-Styragel,

respectively. Molecular weight calibration curves, obtained by elution

of polystyrene standards with THF gave working molecular weight ranges

of 500 to 20,000 and 100 to 3,000, respectively, for the above gels

(Fig. 14).

Several of the highly substituted organic acid standards deviated

from the polystyrene calibration curve; therefore, the polystyrene

standards are not suitable for accurate determination of molecular

weights of low molecular weight organic acids. Based on the above ob-

servation, it is doubtful that polystyrene standards would be suitable

standards for molecular weight determinations of soil humic compounds.





66









6000-


Z o100 A a-STYRAGEL(lf)
*5000 o
5 A100 A PORAGEL(2f)

0
0 4000-
LI,


S3000
-j\
o
F 2000

0
O
I 1000


1 I I I I 1 I It
0.100 0.200
SAMPLE SIZE, ml



Fig. 13 Effect of sample size on column efficiency, N, of 100 A
Poragel and 100 A p-Styragel preparative columns with
THF as the eluting solvent






67







6-

196,000
I3: 111,000
0i 5 ,34,500


20,500

-j
S4 9,800
0
Ld
.j 3,550
0 2,025

3-
600

O
0 '

J 1,24,5-tetracarboxybenzens an
anthraesne
vanillin
2 m-hydroxybsnzaldehyda
acetone





10 20 30 40 50
ELUTION VOLUME, mi


Fig. 14 Molecular weightocalibration curve of U-Styragel (2 f x
0.954 cm OD 100 A + 2 f x 0.954 cm OD 500 A) obtained
by elution of polystyrene standards with THF






68





The selection of suitable molecular weight standards for soil humic

compounds remains a problem.

Effect of solvent on elution of standard compounds. Comparison

of the elution volumes of acetone and 2,600,000 molecular weight poly-
0
styrene in THF and DMF in 100 A Poragel suggests that the Poragel is

poorly swelled in DMF (Table 1). This conclusion is based on the as-

sumption that acetone is a nonreactive solute and readily enters the

Poragel gel matrix and that the high molecular weight polystyrene

standard is completely excluded from the gel matrix. In DMF, the elu-

tion volumes of acetone and the 2,600,000 molecular weight polystyrene

are separated by 5.3 ml compared to THF in which the elution volumes

of low and high molecular weight standards are separated by 10.5 ml.

Therefore, in DMF the internal pore volume of the packing material is

33% of the total pore volume compared to the THF in which the internal

pore volume of the packing material is approximately 49% of the total

pore volume. The greater swelling of the polystyrene-DVB gel in THF

compared to DMF may be attributed to the less polar and greater hydro-

phobic character of the former solvent which would make it more com-

patible with the hydrophobic gel.

As a consequence of the different swelling properties of the gel

in the different solvents, it is essential that column parameters and

molecular weight distribution patterns be determined for the same

solvent which is to be used as the eluting solvent. Also, it is

essential that the column be packed in the same solvent which is to be

used as the eluting solvent. Changing of the eluting solvent in the







69




column may produce voids and result in increased peak widths and re-

duced column efficiencies.

Peak elution volumes of organic standards eluted with THF and
0
DMF on 100 A Poragel are summarized in Table 17. The initial observa-

tion is that several of the solute species are behaving differently

in the two eluting solvents.

In looking more closely at solute behavior in THF, it can be

observed that each of the low-molecular weight solutes were eluted

in the vicinity of or slightly after VT. Apparently, each of these

solutes readily entered the pores of the Poragel gel matrix. Only

benzene and anthracene were eluted noticeably past VT. Edwards and

Ng (1968) also observed the adsorption of some aromatic compounds by

polystyrene-DVB when eluted with THF. It is probably the aromatic

character of the polystyrene-DVB gel which resulted in adsorption of

benzene and anthracene. The aromatic acids were not noticeably eluted

past VT and were apparently not strongly adsorbed. The elution of these

compounds near VT gave strong indication that they readily entered the

polystyrene-DVB gel matrix.
0
Elution of standard compounds with THF on 100 A u-Styragel produced

very similar results. Only benzene, toluene, and anthracene were eluted

past the assumed value of V due to an apparent adsorptive interaction

with the gel matrix. Other standard compounds tested, e.g. simple

alcohols, aromatic acids, aromatic bases, and phenolic acids, were

eluted in the vicinity of VT and apparently readily entered the

polystyrene-DVB gel matrix.






70







Table 17. Peak elution volumes of low molecular weight standards
eluted on 100 A Poragel with THF and DMF and on 100 A
i-Styragel with THF as the eluting solvent



Column packing
0 O
100 A Poragel 100 A p-Styragel

Sample THF DMF THF

-------------------ml--------------------

1,2,4,5-Tetracarboxy-
benzene 20.81 12.93 10.27

1,3,5-Tricarboxyben-
zene 21.04 12.97 10.34

3,5-Dihydroxybenzoic
acid 21.19 13.04 10.36

Benzoic acid 21.23 13.28 10.49

Pyridine 22.13 18.12 10.56

Aniline 21.28 17.21 10.52

Methanol 21.24 16.18 10.53

Ethylene glycol 21.13 16.34 10.49

Acetone 21.26 16.47 10.53

Benzene 21.84 17.63 10.89

Anthracene 22.43 18.02 11.42






71



When DMF was used as the eluting solvent, several of the aromatic

acids were eluted prior to the assumed value of V The more highly

substituted aromatic acids (e.g. 1,2,4,5-tetracarboxybenzene and 1,3,5-

tricarboxybenzene) were eluted near the assumed value of V0 and were

apparently completely excluded from the gel matrix. Several of the

aromatic acids and phenolic acids showed two elution peaks which cor-

responded closely to the assumed values of V0 and VT. Each of the low

molecular weight compounds which were eluted noticeably before V con-

tained an acidic side group, COOH and/or phenolic OH.

Several compounds, e.g. benzene, toluene, and anthracene, were

eluted considerably past the assumed value of V Each of these com-

pounds was hydrophobic in nature and was structurally similar to

monomers of the gel polymer. Benzene, toluene, and anthracene were

more strongly adsorbed with DMF than with THF as the eluting solvent.

This effect was probably due to the more polar character of the DMF.

Neutral solutes (e.g. simple alcohols) and compounds with basic

properties (e.g. pyridine, aniline) were eluted at or slightly after the

assumed VT and apparently readily entered the polystyrene-DVB gel

matrix.

Two interesting points from the above observations are that (i)

hydrophobic solutes were more strongly adsorbed to the polystyrene-DVB

gel matrix when eluted with DMF than with THF, and (ii) acidic solutes

were noticeably excluded from the gel matrix when DMF was used as the

eluting solvent, but not when THF was used as the solvent. The first

point may be explained in terms of relative hydrophobic character






72



of the two solvents, as discussed previously. The second point is

elaborated upon below.

The acidic functional groups of an acidic organic solute would be

partially dissociated in DMF due to basic character of this solvent;

therefore, the solute molecules are likely to be highly dispersed.

The negatively charged solute molecules would have larger effective

radii than the neutral species; however, this phenomena should not

entirely account for the exclusion phenomena since the exclusion limit

of the gel, based on polystyrene standards, is approximately 50,000

molecular weight. In the porous silica and Sephadex gels, the exclusion

phenomena can be explained in terms of electrostatic repulsion from

negative charge sites in the gel matrix. In porous silica, the negative

charge results from dissociation of Si(OH) sites at silica surface.

In Sephadex the negative charge has been attributed to COOH impurities

in the gel matrix. On the other hand, the polystyrene-DVB gel should

exist as a neutral species. Therefore, we must search for an alternate

explanation to the exclusion phenomena. A possible explanation is the

ion inclusion effect suggested by Forss and Stenlund (1975) in studies

of lignosulfonate. They attributed this effect to the interaction of

charged sites on the ions entering the pores with other charged ions

outside of the pores. The net effect is electrostatic repulsion. Such

an effect would not entirely account for the apparent total exclusion

of low molecular weight solutes observed. Further work will be required

to determine the nature of this phenomena.





73





Effect of solvent on elution of soil humic compounds. Elution

patterns of soil humic fractions on 100 A I-Styragel with THF as the

eluting solvent are shown in Figs. 15 and 16. In THF, all humic frac-

tions were eluted between V0 and VT and apparently readily entered the

porous gel matrix. As mentioned previously, the polystyrene standards

are not suitable for accurate molecular weight determinations of soil

humic materials. These standards, however, do provide a guide for

measurement which is probably no less suitable than others commonly

used, such as proteins or polysaccharides. Molecular weight estimates

based on the polystyrene standards are summarized in Table 18. In all

cases the Soxhlet fractions were estimated to have peak molecular

weights less than 800. These fractions, however, represent only a

minor portion of the total NaOH- or DMF-extractable materials, 28 and

32%, respectively, and are likely to contain materials with lower peak

molecular weights than those of the NaOH- or DMF-extractable materials.

These later materials are not sufficiently soluble in THF to obtain a

molecular weight fractionation.

In DMF, the major portions of all humic fractions were eluted at

volumes corresponding to the assumed values of VO and were apparently

largely excluded from the gel matrix. In all cases, a minor portion

of the material was eluted at VT. Reinjection of fractions collected

at VO produced patterns similar to the original patterns with major

peaks corresponding closely to V and minor peaks at the assumed value

of VT. Reinjection of the sample eluted at VT during the original

fractionation also produced a fractionation pattern similar to the





74






Vo Vt
ACETONE





Ln PROPANOL


CL




o* METHANOL
0
0





NaOH
SOXTHLET



6 12 18 24 50 36 42 48


Fig. 15 Elution of Soxhlet extgacts of NaOH-extractable soil
organic matter on 100 A p-Styragel with THF




75





Vo Vt
ACETONE





w PROPANOL
or
Z






cr_ METHANOL
o






NaOH
SOXHLET


6 12 18 24 30 36 42 48
ELNMETHAN V, OL
NaOI-I
SOXHLET


6 12 18 24 50 36 42 48
ELUTION VOLUME,ML



Fig. 16 Elution of Soxhlet extracts of DMF-extractable soil
organic matter on 100 A u-Styragel with THF






76




original pattern with a major peak eluted at V0 and a minor peak at

VT. This technique produced strong evidence that the initial frac-

tionation pattern was not the result of a separation according to

molecular size, but instead was an artifact resulting from a complex

gel-solvent-solute interaction. The elution patterns of the soil

humic acid fractions were indeed similar to the patterns obtained from

elution of the low molecular weight aromatic acids.






77














Table 18. Molecular weight estimates of soil humic fractions based
on elution of polystyrene standards on p-Styragel with
THF as the eluting solvent



Estimated molecular weight

Sample NaOH extract DMF extract


Acetone-Soxhlet 560 660



2-Propanol-Soxhlet 720 740



Methanol-Soxhlet 760 740


















CONCLUSIONS



Extraction and Fractionation


Several of the dipolar aprotic solvents, i.e. DMF and DMSO, were

shown to be excellent solvents for the soil humic fraction. Functional

group, elemental, and IR analysis indicated that the DMF-extractable

soil organic matter was chemically similar to the material extracted

by 0.5 N NaOH. The dipolar aprotic solvent may,therefore, serve as an

excellent complementary solvent to NaOH for chemical studies of extractable

soil organic matter.

The Soxhlet fractionation scheme was used successfully to fraction-

ate the extractable soil organic matter into samples with distinct charac-

teristics. The Soxhlet extraction scheme of hexane, benzene, ethylacetate,

acetone, 2-propanol, and methanol was utilized to obtain materials with

progressively greater hydrophilic character, lower C and H contents,

greater N, S, and 0 contents, greater COOH content, and greater total

acidity. The Soxhlet solvents were able to extract 29.6 and 35.0% of

the NaOH- and DMF-extractable materials, respectively. Even though

these fractions represent a minor portion of the total extractable

material, they are likely to contain materials of simpler average composi-

tion and lower peak molecular weights than the NaOH- and DMF-extractable




78





79




material and are important since they are likely to contain monomers

which compose the polymeric structure of the humic complex. Therefore,

investigations of these fractions provide information which will aid

in understanding properties of the total humic complex.



Solubility Properties


Humic acid and NaOH- and DMF-extractable soil organic matter were

100% soluble at the 0.1% concentration in both DMF and DMSO. Fulvic

acid was completely soluble at the 0.1% concentration in both water

and methanol. Solubility of extractable soil organic matter was de-

creased in the dipolar aprotic solvents, e.g. DMF, THF, and acetone, and

increased in the protic solvents as acidic hydrogen was placed with

Na+, K, or N(CH3)4+.

The solubility of fulvic-acid in aqueous systems was influenced

by pH and concentration of excess neutral electrolyte. Fulvic acid was

completely soluble at the 0.1% concentration in aqueous systems with

concentrations of excess neutral electrolyte up to 0.1 N at pH values

less than 4.0 and greater than 8.0; however, at pH values from 4.0 to

8.0, fulvic acid partially precipitated with concentrations of

Na2 SO4 or K2SO4 greater than 0.05 N.

The number of useful chromatographic fractionation schemes is

greatly limited by the solubility characteristics of the solute; there-

fore, an understanding of these characteristics is an essential

prerequisite to the rapid screening of possible fractionation schemes.






80




Liquid Chromatography


None of the chromatographic gels investigated was completely

inert. Each gel apparently interacted with the soil humic material;

therefore, the elution patterns were not entirely attributable to a

molecular seiving phenomenon but to a combination of molecular seiving,

adsorption, and ionic exclusion phenomena.

Solvent and electrolyte effects were especially evident in studies

of Porasil and CPG packing materials. When H-, Na-, K-, or N(CH3)4-

saturated low molecular weight organic acid standards or fulvic acid

were eluted with H20, the solute molecules were partially or totally

excluded from the porous gel matrix. As electrolyte concentration was

increased, the acidic solute molecules more readily entered the porous

matrix; however, at electrolyte concentrations above 0.01 N, significant

quantities of fulvic acid were adsorbed and eluted past VT. These

phenomena were attributed to decreased thickness of the electrical

double layer and/or suppression of charge of the negatively charged

solute molecule and the negatively charged silicate surface. Adsorption

of fulvic acid at the higher electrolyte concentrations was attributed

to increased interaction between active sites at the silica surface

and oxygen- and nitrogen-containing functional groups of the organic

solute and also to the possible precipitation of fulvic acid caused by

the high counter ion concentration at the negatively charged silica

surface. Low molecular weight organic solutes with significant basic

properties, i.e. pyridine, were strongly adsorbed to Porasil and CPG






81




gels in aqueous systems. Since the fulvic acid sample contained

nitrogen, indicating the probable presence of basic sites, interac-

tions of these sites in the negatively charged solute molecule with the

negatively charged silicate surface would be greater in the presence

of excess neutral electrolyte.

When protic solvents, i.e. H20, methanol, and 2-propanol, or

dipolar aprotic solvents with significant basic properties, i.e. DMF,

were used to elute low molecular weight acidic solutes or extractable

soil organic matter, electrostatic exclusion phenomena predominated.

In dipolar aprotic solvents without significant basic character, i.e.

acetone and THF, adsorption phenomena predominated. With all solvent

and electrolyte systems examined, it was not possible to completely

eliminate both electrostatic exclusion and adsorption phenomena. De-

activation of the Porasil surface, Porasil X, did not completely eliminate

adsorption and exclusion interactions between the acidic organic solute

and the silica surface.

The polystyrene-DVB gels were compatible with a more limited range

of solvents than the silica gels. Also, due to the swelling properties

of the gel it was essential to pack the column with the same solvent

which was to be used as the eluting solvent. With the polystyrene-DVB

gels, elution patterns were dependent on the eluting solvent. In DMF,

low and high molecular weight acidic organic solutes were totally or

partially excluded from the gel matrix. In THF, low molecular weight

acid solutes apparently readily entered the porous gel matrix and were






82




eluted in the vicinity of V T. Of the compounds tested, only several

hydrophobic aromatic compounds were strongly adsorbed. It appears

that THF is a suitable solvent for elution of acidic solutes; however,

only the benzene-, ethylacetate-, acetone-, and 2-propanol-Soxhlet

fractions were suitably soluble in THF. Methylated fractions of the

humic acid and fulvic acid fractions would also be soluble in THF.

Molecular weights of acetone-, 2-propanol-, and methanol-Soxhlet

fractions were estimated to be 500 to 800, based on elution patterns

of soil humic fractions with those of polystyrene standards in THF.

Further research will be needed to corroborate, by other methods, the

molecular weight estimates obtained with gel permeation chromatography.

One such method would be vapor pressure osmometry.

Each of the two general groups of gelsevaluated in this study

showed evidence of adsorptive and/or electrostatic interaction with

extractable soil organic matter. Mode and extent of interaction were

highly dependent on the solvent medium. Because of these possible

interactions, special care must be observed in the interpretation of

gel permeation chromatography patterns.

High pressure liquid chromatography is a valuable new technique

because of the use of high efficiency columns and the short time re-

quired to obtain elution patterns. In addition to the application of

gel permeation chromatography, there is the unexplored potential

application of liquid-liquid partition chromatography and liquid-solid

adsorption chromatography to fractionation of extractable soil organic

matter. High pressure liquid chromatography will also provide a valu-

able tool for studies of soil organic-mineral-ionic interactions.


















LITERATURE CITED



Bergmann, J.G., L.I. Duffy, and R.B. Stevenson. 1971. Solvent effects
in gel permeation chromatography. Anal. Chem. 43:131-133.

Bly, D.D. 1970. Gel permeation chromatography. Science 168:527-533.

Bombaugh, K.J. 1971. The practice of gel permeation chromatography.
In J.J. Kirkland (ed.), Modern Practice of Liquid Chromatography.
Wiley-Interscience, New York.

Bombaugh, K.J., W.A. Dark, and J.N. Little. 1969. Fractionation of
alcohol on deactivated porous silica beads by gel permeation
chromatography. Anal. Chem. 41:1337-1339.

Bremner, J.M. 1950. Some observations on the oxidation of soil
organic matter in the presence of alkali. J. Soil Sci. 1:
198-204.

Bremner, J.M. 1956. Some soil organic matter problems. Soil and
Fertilizers 19:115-123.

Bremner, J.M. and J. Lees. 1949. Studies on soil organic matter.
II. The extraction of organic matter from soil by neutral
reagents. J. Agr. Sci. 39:274-279.

Brook, A.J.W. and S. Housley. 1969. The interaction of phenols with
Sephadex gels. J. Chromatogr. 41:200-204.

Brook, A.J.W. and K.C. Munday. 1970. The interaction of phenols,
aniline, and benzoic acids with Sephadex gels. J. Chromatogr.
47:1-8.

Brooks, J.D., R.A. Durie, and S. Sternhell. 1958. Chemistry of
brown coals. III. Pyrolytic reactions. Aust. J. Appl. Sci.
9:303-320.

Burges, N.A., H.M. Hurst, and B. Walkden. 1964. The phenolic con-
stituents of humic acid and their relation to the lignin of
the plant cover. Geochim. Cosmochim. Acta 28:1547-1554.




83






84




Cheshire, M.V., P.A. Cranwell, C.P. Falshow. A.J. Floyd, and R.D.
Haworth. 1967. Humic acid: 2. Structure of humic acid.
Tetrahedron 23:1669-1682.

Choudhri, M.B., and F.J. Stevenson. 1957. Chemical and physico-
chemical properties of soil humic colloids: III. Extraction
of organic matter from soils. Soil Sci. Soc. Amer. Proc. 21:
508-513.

Cogswell, T.E., J.F. McKay, and D.R. Latham. 1971. Gel chromato-
graphic separation of petroleum acids. Anal. Chem. 43:645-648.

Cooper, A.R. and E.M. Barrall. 1973. Gel permeation chromatography:
physical characterization and chromatographic properties of
Porasil. J. Appl. Polym. Sci. 17:1253-1268.

Cooper, A.R. and A.R. Bruzzone. 1973. Characterization and proper-
ties of macromolecules. III. Gel permeation chromatography:
the effect of temperature on the elution volume and the effi-
ciency of the separation process. J. Polym. Sci. A-2, 11:
1423-1434.

Cooper, A.R., A.R. Bruzzone, J.H. Cain, and E.M. Barrall. 1971.
Characterization and chromatographic properties of Corning porous
glasses. J. Appl. Polym. Sci. 15:571.

Cooper, A.R. and J.F. Johnson. 1969. Gel permeation chromatography:
the effect of treatment with hexamethyl disilazane on porous
glass packings. J. Appl. Polym. Sci. 13:1487-1492.

Cooper, A.R., J.F. Johnson, and R.S. Porter. 1973. Gel-permeation
chromatography: current status. American Laboratory. May
1973:12-24.

Dark, W.A. and R.J. Limpert. 1973. An evaluation of available pack-
ings for GPC. J. Chromatogr. Sci. 11:114-120.

Demetriou, J.A. et al. 1968. Gel filtration chromatography of
fluorescent phenolic and heterocyclic compounds. J. Chromatogr.
34:342-350.

Determann, H. and I. Walter. 1968. Source of aromatic affinity to
Sephadex dextran gels. Nature (Lond.) 219:604-605.

Edwards, G.D. and Q.Y. Ng. 1968. Elution behavior of model compounds
in gel permeation chromatography. J. Polym. Sci. Part C 21:
105-117.

Flaig, W., H. Beutelspacher, and E. Rietz. 1975. Chemical composi-
tion and physical properties of humic substances. In J.E.
Gieseking (ed.), Soil components. Springer-Verlag, New York.







85





Forss, K.G. and B.G. Stenlund. 1975. The influence of charged groups
in gel permeation chromatography on polyelectrolytes. J. Polym.
Sci. 42:951-963.

Freeman, D.H. 1973. The gels for liquid chromatography. J.
Chromatogr. Sci. 11:175-180.

Fritz, J.S., S.S. Yamamura, and E.C. Bradford. 1959. Determination
of carbonyl compounds. Anal. Chem. 31:260-263.

Gaylor, V.F., H.L. James, and H.H. Weetall. 1976. Gel and affinity
chromatography. Anal. Chem. 48:44R-51R.

Gelotte, B. 1960. Studies on gel filtration sorption of the bed
material Sephadex. J. Chromatogr. 3:330-342.

Giddings, J.C. 1965. Dynamics of chromatography: principles and
theory. Marcel Dekker, New York.

Halstead, R.L., G. Anderson, and N.M. Scott. 1966. Extraction of
organic matter from soils by means of ultrasonic dispersion
in aqueous acetylacetone. Nature 211:1430-1431.

Hansen, E.H. and M. Schnitzer. 1967. The alkaline permanganate
oxidation of Danish illuvial organic matter. Soil Sci. Soc.
Amer. Proc. 30:745-748.

Hansen, E.H. and M. Schnitzer. 1969. Zinc-dust distillation of soil
humic compounds. Fuel 48:41-46.

Hawk, G.L., J.A. Cameron, and L.B. Dufault. 1972. Chromatography of
biological materials on polyethylene glycol treated controlled
pore glass. Prep. Biochem. 2:193-203.

Haworth, R.D. 1971. The chemical nature of humic acid. Soil Sci.
111:71-79.

Hiatt, C.W. et al. 1971. Treatment of controlled-pore glass with
poly (ethylene oxide) to prevent adsorption of rabies virus.
J. Chromatogr. 56:362-364.

Hurst, H.M. and N.A. Burges. 1967. Lignin and humic acids. In
A.D. McLaren and G.H. Peterson (eds.), Soil biochemistry.
Marcel Dekker, New York.

Karger, B.L. 1971. The relationship of theory to practice in high
speed liquid chromatography. In J.J. Kirkland (ed.), Modern
practice of liquid chromatography. Wiley-Interscience,
New York.







86




Kessler, T., R.A. Friedel, and A.G. Sharkey. 1970. Ultrasonic solva-
tion of coal in quinoline and other solvents. Fuel 49:222-223.

Khan, S.U. 1971. Distribution and characteristics of organic matter
extracted from black solnetzic and black chenozemic soils of
Alberta: the humic acid fraction. Soil Sci. 112:401-409.

King, E.J. 1973. Acid-base behavior. In A.K. Covington and T.
Dickinson (eds.), Physical chemistry of organic solvent systems.
Plenum Press, London.

Kirkland, J.J. 1974. Modern practice of liquid chromatography.
Wiley-Interscience, New York.

Kononova, M.M. 1966. Soil organic matter. 2nd English ed. Pergamon
Press, New York.

Kumada, K. and A. Suzuki. 1961. Isolation of anthraquinones from
humus. Nature 191:415-416.

Laub, R.J. 1974. Packings for HPLC. Research and development.
July 1974:24-28.

LePage, M., R. Beau, and A.J. DeVries. 1968. Evaluation of analyti-
cal gel chromatography columns packed with porous silica beads.
J. Polym. Sci., Part C 2:119-130.

Loeppert, R.H., Jr., and B.G. Volk. 1974. Nature of organic matter
extracted from Terra Ceia muck with selected solvents. Soil
and Crop Sci. Soc. Fla. 33:160-164.

Loeppert, R.H., Jr., and B.G. Volk. 1976. Application of high pres-
sure liquid chromatography to studies of extractable soil
organic matter-porous silica packings. International Symposium
on Soil Organic Matter Studies, Braunschweig, Germany. In
Press).

Loeppert, R.H., Jr., L.W. Zelazny, and B.G. Volk. 1976. Acidic
properties of kaolinite in water and acetonitile. Soil Sci.
Soc. Amer. J. (In press).

McIver, R.D. 1962. Ultrasonics--a rapid method for removing soluble
organic matter from sediments. Geochim. Cosmochim. Acta 26:
343-345.

Mortenson, J.L. 1965. Partial extraction of organic matter. In
C.A. Black (ed.), Methods of soil analysis, Agromony Monograph
No. 9, Vol. 2. American Society of Agronomy. Madison,
Wisconsin.






87




Neddermeyer, P.A. and L.B. Rogers. 1968. Gel filtration behavior of
inorganic salts. Anal. Chem. 40:755-762.

Ortiz de Serra, M.I. and M. Schnitzer. 1973. The chemistry of humic
and fulvic acids extracted from Argentine soils. II. Perman-
ganate oxidation of methylated humic and fulvic acids. Soil
Biol. Biochem. 5:287-296.

Parsons, J.W. and J. Tinsley. 1960. Extraction of organic matter
with anhydrous formic acid. Soil Sci. Soc. Amer. Proc. 24:198-
201.

Porter, L.K. 1967. Factors affecting the solubility and possible
fractionation of organic colloids extracted from soil and leo-
nardite with an acetone-H 2-HC1 solvent. J. Agr. Food Chem.
15:807-811.

Posner, A.M. 1966. The humic acid extracted by various reagents
from a soil. I. Yield, inorganic components, and titration
curves. J. Soil Sci. 17:65-68.

Posner, A.M. 1963. Importance of electrolyte in the determination
of molecular weights by Sephadex gel filtration with special
reference to humic acid. Nature 198:1161-1163.

Salfeld, J.C. 1964. Fractionation of a humus preparation with aqueous
organic solvents. Landbauferschung Volkenrode 14:131-136.

Schnitzer, M. and V.C. Gupta. 1965. Determination of acidity in soil
organic matter. Soil Sci. Soc. Amer. Proc. 29:274-277.

Schnitzer, M. and S.U. Khan. 1972. Humic substances in the environ-
ment. Marcel Dekker, Inc., New York.

Schnitzer, M., D.A. Shearer, and J.R. Wright. 1959. A study in the
infrared of high molecular weight organic matter extracted by
various reagents from a podzolic B horizon. Soil Sci. 87:
252-257.

Schnitzer, M. and S.I.M. Skinner. 1968. Gel filtration of fulvic
acid, a soil humic compound. In Isotopes and radiation in soil
organic matter studies. International Atomic Energy Agency,
Vienna.

Snow, J.T., R.H. Loeppert, Jr., and B.G. Volk. 1973. Yields and
functional group characteristics of NaOl-extracted humic acid
from Pahokee muck. Soil and Crop Sci. Soc. Fla. Proc. 33:
165-167.






88





Sommers, T.C. 1966. Wine tannins--Isolation of condensed flavanoid
pigments by gel filtration. Nature (Lond.) 209:368-370.

Spatorico, A.L. 1975. Exclusion chromatography using controlled
porosity glass. I. Comparison with styrene gels. J. Appl.
Polym. Sci. 19:1601-1610.

Spatorico, A.L. and G.L. Beyer. 1975. Exclusion chromatography using
porous glass. II. Application to hydrophilic polymers. J.
Appl. Polym. Sci. 19:2933-2945.

Steelink, C. and G. Tollin. 1967. Free radicals in soil. In A.D.
McLaren and G.H. Peterson (eds.), Soil biochemistry. Marcel
Dekker, New York.

Stevenson, F.J. 1965. Gross chemical fractionation of organic matter.
In C.A. Black (ed.), Methods of soil analysis, Agronomy Monograph
No. 9, Vol. 2. American Society of Agronomy. Madison, Wisconsin.

Swift, R.S. and A.M. Posner. 1971. Gel chromatography of humic acid.
J. Soil Sci. 22:237-249.

Talhoun, S.A. and M.M. Mortland. 1966. Complexes of montmorillonite
with primary, secondary, and tertiary amides. I. Protonation
of amides on the surface of montmorillonite. Soil Sci. 102:
248-254.

Tinsley, J. and A. Salam. 1961. Extraction of soil organic matter
with aqueous solvents. Soils and Fert. 24:81-84.

Unger, K.K., R. Kern, M. Ninou, and K.F. Krebs. 1974. GPC with a
new type of silica packing material. J. Chromatogr. 99:435-443.

Volk, B.G. and M. Schnitzer. 1973. Chemical and spectroscopic methods
for assessing subsidence in Florida Histosols. Soil Sci. Soc.
Amer. Proc. 37:886-888.

Zweig, G. and J. Sherma. 1974. Chromatography. Anal. Chem. 46:73R-
94R.




Full Text
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Professor
X J
Y
Fiskell, Chairman
of Soil Science
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Associate Professor
Soil Science
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Associate Professor of Soil Science
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philsophy.
/N. Gammon
Professor of Soil Science


68
The selection of suitable molecular weight standards for soil humic
compounds remains a problem.
Effect of solvent on elution of standard compounds. Comparison
of the elution volumes of acetone and 2,600,000 molecular weight poly-
O
styrene in THF and DMF in 100 A Poragel suggests that the Poragel is
poorly swelled in DMF (Table 1). This conclusion is based on the as
sumption that acetone is a nonreactive solute and readily enters the
Poragel gel matrix and that the high molecular weight polystyrene
standard is completely excluded from the gel matrix. In DMF, the elu
tion volumes of acetone and the 2,600,000 molecular weight polystyrene
are separated by 5.3 ml compared to THF in which the elution volumes
of low and high molecular weight standards are separated by 10.5 ml.
Therefore, in DMF the internal pore volume of the packing material is
33% of the total pore volume compared to the THF in which the internal
pore volume of the packing material is approximately 49% of the total
pore volume. The greater swelling of the polystyrene-DVB gel in THF
compared to DMF may be attributed to the less polar and greater hydro-
phobic character of the former solvent which would make it more com
patible with the hydrophobic gel.
As a consequence of the different swelling properties of the gel
in the different solvents, it is essential that column parameters and
molecular weight distribution patterns be determined for the same
solvent which is to be used as the eluting solvent. Also, it is
essential that the column be packed in the same solvent which is to be
used as the eluting solvent. Changing of the eluting solvent in the


37
The presence of excess neutral salt affected the solubility of
fulvic acid in water (Table 7) and methanol. Some very interesting
trends were evident in these studies. The H-, Na-, K-, N(CH^)^-, and
NfC^H^-saturated fulvic acid samples at 0.1% concentration were
soluble at pH 2.0 in all concentrations of excess neutral electrolyte
-3 -2
up to 0.1 N^. In the presence of 10 N or 10 N excess salt, 0.1%
fulvic acid remained completely dissolved as the pH was increased suc
cessively to pH 4.0, 6.0, 8.0, and 10.0. However, in the presence of
-2
5 x 10 N excess salt, K-saturated fulvic acid began to precipitate
at pH 4.0. The sample redissolved at pH 8.0 and was completely soluble
f*
as the pH was increased to 10.0. Fulvic acid saturated with N(CH^)^
or N(CAH9^+ remained completely dissolved as the pH was increased from
-2
2.0 to 10.0, in the presence of 5 x 10 N excess neutral salt. As the
ionic strength was increased to 5 x 10 ^ N, however, K-, Na-, NCH^)^-,
and NCC^Hg^-saturated fulvic acid precipitated as the solution pH ap
proached 4.0 and redissolved as the pH approached 10.0. The precipita
tion was greatest in the approximate range of pH 4 to pH 7 and may be
attributed to unfavorable conditions for the electrostatic dispersion
of molecular units. It is interesting to note that greatest precipita
tion occurred within the pH range at which greatest neutralization of
acidic carboxyl groups would occur.
The precipitation phenomena in the presence of excess neutral salt
greatly limits the conditions which may be employed for gel permeation
separations of extractable organic matter.


13
since previous results may be useful in developing separation schemes
and interpreting results with the newer gels. The reader is referred
to an excellent review by Swift and Posner (1971) .
Sephadex G-gels have been shown to strongly adsorb aromatic com
pounds (Gelotte, 1960), heterocyclic compounds (Demetriou et al.,
1968), and phenolic compounds (Sommers, 1966; Brook and Housley,
1969). Gel-phenol affinity is related to the ether bonds in the cross-
linking group rather than to the polysaccharide (dextran) component
of the gel matrix (Determann and Walter, 1968). As degree of cross-
linking of the gel was increased, affinity of phenol for the gel was
also increased. Brook and Munday (1970) suggested that benzene
derivatives are adsorbed onto hydroxyether cross-linking by H-bonds
and that interaction of Sephadex dextran gels with monosubstituted
phenols, anilines, and benzoic acids operates through hydroxyl, amino,
and carboxylic groups, respectively. Gelotte (1960) observed that the
Sephadex bed material contained a small amount of ionized groups,
probably COOH groups, at concentrations of approximately 10 meq per
gram of dry Sephadex. Aromatic amino acids were adsorbed to the bed
material, basic amino acids were strongly adsorbed, and acidic amino
acids were partially excluded from the gel. Demetriou et al. (1968)
likewise found that aromatic compounds with COOH substituents were
excluded from the gel beads when distilled water was used as the
eluting solvent. The same compounds were adsorbed when columns were
eluted with acid-salt solutions. Similar results were observed
during the elution of soil humic acids (Posner, 1966). Humic acid


CONCLUSIONS
Extraction and Fractionation
Several of the dipolar aprotic solvents, i.e. DMF and DMSO, were
shown to be excellent solvents for the soil humic fraction. Functional
group, elemental, and IR analysis indicated that the DMF-extractable
soil organic matter was chemically similar to the material extracted
by 0.5 N NaOH. The dipolar aprotic solvent may,therefore, serve as an
excellent complementary solvent to NaOH for chemical studies of extractable
soil organic matter.
The Soxhlet fractionation scheme was used successfully to fraction
ate the extractable soil organic matter into samples with distinct charac
teristics. The Soxhlet extraction scheme of hexane, benzene, ethylacetate,
acetone, 2-propanol, and methanol was utilized to obtain materials with
progressively greater hydrophilic character, lower C and H contents,
greater N, S, and 0 contents, greater COOH content, and greater total
acidity. The Soxhlet solvents were able to extract 29.6 and 35.0% of
the NaOH- and DMF-extractable materials, respectively. Even though
these fractions represent a minor portion of the total extractable
material, they are likely to contain materials of simpler average composi
tion and lower peak molecular weights than the NaOH- and DMF-extractable
78


69
column may produce voids and result in increased peak widths and re
duced column efficiencies.
Peak elution volumes of organic standards eluted with THF and
O
DMF on 100 A Poragel are summarized in Table 17. The initial observa
tion is that several of the solute species are behaving differently
in the two eluting solvents.
In looking more closely at solute behavior in THF, it can be
observed that each of the low-molecular weight solutes were eluted
in the vicinity of or slightly after V Apparently, each of these
solutes readily entered the pores of the Poragel gel matrix. Only
benzene and anthracene were eluted noticeably past V Edwards and
Ng (1968) also observed the adsorption of some aromatic compounds by
polystyrene-DVB when eluted with THF. It is probably the aromatic
character of the polystyrene-DVB gel which resulted in adsorption of
benzene and anthracene. The aromatic acids were not noticeably eluted
past V^ and were apparently not strongly adsorbed. The elution of these
compounds near gave strong indication that they readily entered the
polystyrene-DVB gel matrix.
O
Elution of standard compounds with THF on 100 A u~Styragel produced
very similar results. Only benzene, toluene, and anthracene were eluted
past the assumed value of V due to an apparent adsorptive interaction
with the gel matrix. Other standard compounds tested, e.g. simple
alcohols, aromatic acids, aromatic bases, and phenolic acids, were
eluted in the vicinity of V^ and apparently readily entered the
polystyrene-DVB gel matrix.


INTRODUCTION
Soil is an important international resource which serves as the
source of the majority of the world's food supply and as a major sink
for all man-made and natural products in the environment. Organic
matter, due to its reactive nature, has a large influence on soil
properties. In order to understand the chemical properties of soil
organic matter and the exact role of humic fractions in the soil, it
is essential that the scientist understand the chemical structures
involved. Many years of research have given structural clues, however,
due to the complexity of soil organic matter, the science is still in
its infancy.
One approach to the structural problem has been to initially
fractionate and simplify the humic material prior to further analyti
cal investigations.
Since the development of the polydextran gels, there has been
considerable interest in gel filtration for molecular-size fractiona
tion and characterization of soil organic matter (Swift and Posner,
1971). Scientists have used this method to isolate molecular-size
fractions and to obtain estimates of molecular weight of humic sub
stances in soils and natural water.
During the past decade, rapid advances have been made in liquid
chromatography. These advances have been due primarily to the
1


19
dried under a dry-nitrogen jet, and redissolved in the extracting sol
vent at room temperature. Samples were concentrated to 30-ml volume
and following, addition of 30 ml of ^0, were reconcentrated to 30-ml
volume. The reconcentration procedure was repeated several times, and
the aqueous suspensions were transferred to dialysis bags for purifica
tion to less than 0.3% ash. Purified samples were lyophilized and
stored at 0C.
Solubility Studies
The solubility behavior of extractable organic matter and each of
the Soxhlet fractions was determined by placing 2.00 mg of the organic
material in 2 ml of the appropriate solvent. Following agitation the
mixture was visually observed to determine whether the sample was in
soluble, partially soluble or completely soluble. Where appropriate,
the pH of the sample suspension was adjusted to the desired level by
addition of acid or base which contained the same counter ion as the
excess electrolyte.
Analytical Determinations
Carbon and H were determined with the Coleman C-H analyzer, N by
the micro-Kjeldahl method, and S using the Leco induction furnace.
Oxygen was determined by difference with the assumption that C, H, N,
S,and 0 were the only elemental constituents of the extractable organic
matter. Total acidity was determined by addition of excess BaCOH^
and back-titration of unreacted Ba(0H) with HC1 to pH 9.8 (Schnitzer


LIST OF FIGURES (continued)
Figure Page
14 Molecular weight calibration curve of y-Styragel (2 f
x 0.954 cm OD 100 A + 2 f x 0.954 cm OD 500 A) obtained
by elution of polystyrene standards with THF 67
15 Elution of Soxhlet extracts of NaOH-extractable soil
O
organic matter on 100 A y-Styragel with THF 74
16 Elution of Soxhlet extracts of DMF-extractable soil
O
organic matter on 100 y-Styragel with THF 75
ix


35
Table 7. Solubility of fulvic acid in aqueous salt solutions
Saturating Excess
cation electrolyte
Concentration
2.0
4>

pH
6.0
00
o
10.0
Na
Na SO,
0.000 N
sa
s
s
s
s
0.001 N
s
s
s
s
s
0.01 N
s
s
s
s
s
0.05 N
s
Pa
p
p
s
0.10 N
s
p
p
p
s
K
K0SO,
l 4
0.000 N
s
s
s
s
s
0.001
s
s
s
s
s
0.01 N
s
s
s
s
s
0.05 N
s
p
p
p
s
0.10 N
s
p
p
p
s
n(ch3)4
[N(CH )4]2S04
0.000 N
s
s
s
s
s
0.001
s
s
s
s
s
0.01 N
s
s
s
s
s
' 0.05 N
s
s
s
s
s
0.10 N
s
p
p
p
s
(c4h9)4
[N(C4H9)4J2S4
0.000 N
s
s
s
s
s
0.001 N
s
s
s
s
s
0.01 N
s
s
s
s
s
0.05 N
s
s
s
s
s
0.10 N
s
s
s
s
s
aS = soluble, P = partially soluble


9
Column parameters are defined in terms of V^, V and column
efficiency, N. The total pore volume of the column, V is determined
by the elution volume of a nonreactive low molecular weight material
which freely enters the solvent-filled pores of the packing material.
The interstitial volume of the column, V^, is determined by the elu
tion volume of a nonreactive high-molecular weight material which is
completely excluded from the pores of the packing material. Both
and V are determined with standard compounds. In practice, the condi
tion of absolute nonreactivity between solute and packing would probably
never be attained, since there is no such thing as a completely inert
gel network (Freeman, 1973) Likewise, there is no such thing as an
entirely inert solute in liquid chromatography. However, by proper
selection of solute, column parameters and can be determined
with a high degree of accuracy.
Column efficiency is expressed by the theoretical plate count,
N, which is determined with the equation,
N =
where
V is the elution volume of the solute, and to is the peak width at
the baseline. Column efficiency is influenced by particle diameter of
the column, linear velocity of the solvent, and how well the column
is packed (Karger, 1971; Dark and Limpert, 1973). Larger (1971) pre
sents an excellent discussion of factors affecting resolution and
column efficiency.


61
Table 16. Peak elution volumes of selected organic acid standards on
Porasil AX with Na.SO. solutions
2 4
Electrolyte concentration
Sample
0.000 N 0.001 N 0.01 N 0.05 N
ml
1,2,4,5-Tetracarboxy-
benzene 1.87
1.3.5-Tricarboxyben-
zene 2.18
3.5-Dihydroxybenzoic
acid 2.21
Benzoic acid 2.41
2.31
3.03
3.21
2.43
3.05
3.17
2.80
3.19
3.28
3.09
3.40
3.48


53
unlikely since the nonreactive solute, acetone, produced a symmetrical
peak at V with negligible skewing, indicative of free entrance into
the porous gel matrix. Association of the solute molecules in water,
methanol, and DMF would be questionable since aggregation of the mole
cular units should be greatest in the least polar and/or least basic
solvent. Comparison of the individual solvents shows that water, DMF,
and methanol have stronger basic character and are considerably more
polar than THF with dielectric constants of 76.2, 36.7, 32.6, and 7.58,
respectively. Also, deactivation of the porous silica resulted in
increased elution volumes, which should not have been the case if
skewing was entirely due to aggregation of the solute molecules.
With the first two explanations above eliminated as probable
major causes of exclusion of the low molecular weight acidic solute,
the third explanation, electrostatic exclusion, deserves careful con
sideration. The acidic functional groups of the organic solute would
be partially dissociated in water, methanol, or DMF due to the basic
character of each of these solvents. Water and methanol have basic
character which is attributed to the presence of the electron-donor
oxygen atom. The DMF molecule has two basic sites (Talhoun and Mort-
land, 1968), the electron-donor oxygen atom of the carbonyl group and
the nitrogen atom. The surface Si(OH) groups of the silica packing
material are weak acid sites. In water (Kirkland, 1971), methanol, or
DMF the surface sites may dissociate, due to the basic properties of
these solvents, resulting in a negatively charged silica surface.


70
Table 17. Peak elution volumes of low molecular weight standards
eluted on 100 A Poragel with THF and DMF and on 100
y-Styragel with THF as the eluting solvent
Column packing
100 A Poragel 100 A p-Styragel
Sample THF DMF THF
ml
1,2,4,5-Tetracarboxy-
benzene
20.81
12.93
10.27
1,3,5-Tricarboxyben-
zene
21.04
12.97
10.34
3,5-Dihydroxybenzoic
acid
21.19
13.04
10.36
Benzoic acid
21.23
13.28
10.49
Pyridine
22.13
18.12
10.56
Aniline
21.28
17.21
10.52
Methanol
21.24
16.18
10.53
Ethylene glycol
21.13
16.34
10.49
Acetone
21.26
16.47
10.53
Benzene
21.84
17.63
10.89
Anthracene
22.43
18.02
11.42


groups of the organic solute. Deactivation of the silica surface
(Porasil X) resulted in reduction but not elimination of adsorption
and electrostatic exclusion phenomena.
Elution of soil humic compounds and low molecular weight standards
on polystyrene-divinylbenzene (DVB) was strongly influenced by the
solvent. Each of the low-molecular weight organic solutes entered
the porous gel matrix when eluted with THF, however, several hydrophobic
aromatic compounds (e.g. benzene, toluene, and anthracene) were ad
sorbed. Soil humic compounds apparently readily entered the polystyrene-
DVB gel matrix when THF was used. When eluted with DMF, however, soil
humic compounds and low molecular weight organic acid standards were
either totally or partially excluded from the gel matrix. This phe
nomenon was attributed to the effect of solvent on dissociation of
solute molecules and to a possible ion-inclusion effect.
None of the gels investigated was chemically inert, and each
apparently interacted with the soil humic material. Interactions
could be minimized, however, by proper selection of gel, solvent, and
concentration of excess electrolyte.
Molecular weights of acetone-, 2-propanol-, and methanol-Soxhlet
fractions were estimated to be 500 to 800, based on comparison of
elution patterns of soil humic fractions with those of polystyrene
standards in THF.
Xll


RESULTS AND DISCUSSION
Chemical Characteristics of Extractable Organic Matter
Extraction yields of 25% and 49% were obtained by single treat
ments of the surface horizon of Terra Ceia muck with DMF and 0,5 N NaOH,
respectively, following pretreatment with HC1 to lower the ash content
(Table 2). The Soxhlet solvents were able to solubilize 29.6%, 35.0%,
and 9.4% of the NaOH-extractable material, DMF-extractable material, and
Terra Ceia muck, respectively. Only minor portions of the materials were
extracted by hexane and benzene. The major portions were in the ace
tone, 2-propanol, and methanol fractions. Each of these fractions, how
ever, represent only a minor portion of the total soil organic matter.
Obvious and important differences exist in the elemental and func
tional group concentrations of the extractable organic matter (Table 3).
Comparisons of the Soxhlet fractions indicate that C and H content
decreases, 0 and N content increases, COOH content increases, and total
acidity increases according to the solvent order: hexane, benzene,
ethylacetate, acetone, 2-propanol, and methanol. Potentiometric titra
tions of the Soxhlet fractions in DMF indicated that titratable acidity
increased according to the same solvent order (Table 4).
25


60
Table 15. Peak elution volumes of selected organic acid standards on
Porasil A with Na,,S0. solutions
2 4
Electrolyte
concentration
Sample
0.000 N
0.001 N
0.01 N
0.05 N
ml
1,2,4,5-Tetracarboxy-
benzene
1.74
1.68
1.86
2.60
1,3,5-Tricarboxyben-
zene
1.75
1.68
1.91
2.64
3,5-Dihydroxybenzoic
acid
1.82
1.97
2.32
2.71
Benzoic acid
1.96
2.41
2.51
3.14


14
was excluded from the gel matrix when eluted with water, due to the
charge effect. Considerable adsorption was evident when dilute
electrolyte was used as the eluent. It was concluded by Posner that
it is not possible to select a concentration of electrolyte which
would completely eliminate adsorption and exclusion effects.
In studies of lignosulfonate, Forss and Stenlund (1975) concluded
that elution of this material on Sephadex may be infuenced by the
following factors: (i) polyelectrolyte expansion, (ii) ion exclusion,
(iii) ion inclusion, and (iv) steric exclusion. The ion inclusion
effect results from interaction of charged sites in the macroions which
are excluded from the gel with charged sites in more permeable macro
ions. Elution of the lignosulfonate preparation with excess electro
lyte resulted in reduced exclusion of the sample due to reduction in
both the ion exclusion effect and the ion inclusion effect. In a study
of simple electrolytes on Sephadex (Neddermeyer and Rogers, 1968), it
was observed that peaks were badly skewed, with diffuse front and sharp
trailing edges. These skewed peaks were attributed to the ion exclusion
effect produced by ionic solutes and fixed charges in the gel.
Swift and Posner (1971) noted that when humic samples were
eluted with water and the concentration of solute was decreased, a
greater percentage of sample moved into the excluded or near excluded
region. This phenomenon was explained on the basis of double-layer-
theory and decreased suppression of charge at lower concentrations.
Gel-solute interactions were categorized according to coulombic forces,
caused by charged sites on gel and solute, and adsorption, caused by
hydrophilic interaction. Coulombic interactions were most prevalent


38
High Pressure Liquid Chromatography
Porous Silica Packing Materials
Operating conditions. Column efficiencies of Porasil A and
Porasil AX packing materials were greatly influenced by solvent flow
rates. With 0.318-cm OD analytical columns, maximum column efficiencies
and minimum peak broadening were obtained at flow rates of approximately
0.1 ml/min.; however, column efficiencies were not significantly differ
ent at flow rates between 0.1 and 0.6 ml per min. (Fig. 3). At flow
rates greater than 0.6 ml per min., peak broadening was increased and
column efficiencies were decreased. For this reason, it was concluded
that low flow rates should be maintained with the Porasil packing
materials.
It is interesting to observe that for the Porasil packing materials,
there was a slight increase in column efficiency as flow rate was de
creased to 0.1 ml per min. (Fig. 3). These results may be compared with
those obtained with the CPG packing materials for which maximum column
efficiencies were obtained at flow rates of 0.4 ml per min. Decreases
in column efficiency were observed when flow rates were decreased or
increased from this value. As with the Porasil packing material, in
creases in solvent flow rates above 1.0 ml per min. resulted in peak
broadening and significant decreases in column efficiency.
The different behavior of the Porasil and CPG packing materials
at low flow rates may be at least partially attributed to the more


48
extracts (Table 8;Figs. 6 and 7) were eluted at when methanol or
DMF was used as the eluting solvent. A relatively larger quantity
was eluted at with DMF than with methanol. With each of the above
solvents, the organic solute was eluted prior to V^. On the contrary,
when acetone, t-butanol, or THF was used as the eluting solvent with
the Porasil A column, a portion of the organic solute was eluted past
V^, indicating an adsorptive interaction with the silica packing
material.
Deactivation of the Porasil surface (Porasil X) resulted in re
duced exclusion of organic solute from the gel matrix in water and
methanol compared to the activated material (Figs. 6 and 7; Table 9).
Adsorption was reduced on the Porasil AX compared with the Porasil A
packing material, although a small portion of the organic solute was
still eluted past with t-butanol, acetone, and THF on Porasil AX.
Elution volumes of organic standards on Porasil A, Porasil AX,
and CPG-250 are shown in Tables 11-13, respectively. On Porasil A,
several of the organic acid standards (1,2,4,5-tetracarboxybenzene and
1,3,5-tricarboxybenzene) were eluted at a solvent volume equivalent
to Vq when DMF, water, or methanol was used as the eluting solvent and
at a solvent volume slightly greater than when acetone, 2-propanol,
t-butanol or THF was used as the eluting solvent.
In water, methanol, and DMF, the more highly substituted aromatic
acids were in general eluted at a smaller solvent volume than the less
substituted acids. For example, elution volume increased according to


52
the following solute order: 1,2,4,5-tetrac.arboxybenzene < 1,3,5-
tricarboxybenzene < 3,5-dihydroxybenzoic acid < p-hydroxybenzoic acid
< benzoic acid. Several compounds with basic properties (e.g, aniline,
pyridine) were adsorbed and eluted past V Also, several other com
pounds (e.g. glucose, ethylene glycol, and sucrose) were eluted past V,^.
Each of the compounds which showed strong evidence of adsorption on
Porasil A with water, methanol, or DMF as eluting solvent contained an
amino group or an aliphatic OH.
In THF, aromatic acids and simple alcohols each showed evidence
of adsorptive interaction with Porasil A. Elution on Porasil AX, the
deactivated analog of Porasil A, resulted in reduced adsorption.
Elution patterns of organic acid standards on Porasil A and
Porasil AX showed interesting similarities to the elution patterns of
extractable soil organic matter. Based on the molecular weights of
tetracarboxybenzene and tricarboxybenzene and the working molecular
weight ranges of the gels suggested by the manufacturer, one would
expect that the solute would elute at, or slightly before, V Inspec
tion of the patterns, however, shows that the acid standards were
completely excluded from the gel matrix and eluted at when water
was used as the eluting solvent. Deviations from the expected elution
behavior of a low molecular weight nonreactive solute may be attributed
to adsorption, electrostatic exclusion, or molecular association.
The exclusion of the acidic organic solute from Porasil A in
water, methanol, or DMF may be attributed to (i) the porous structure
of the gel, (ii) association of solute molecules, and/or (iii) electro
static exclusion from the porous matrix. The first explanation is


86
Kessler, T., R.A. Friedel, and A.G. Sharkey. 1970. Ultrasonic solva
tion of coal in quinoline and other solvents. Fuel 49:222-223.
Khan, S.U. 1971. Distribution and characteristics of organic matter
extracted from black solnetzic and black chenozemic soils of
Alberta: the humic acid fraction. Soil Sci. 112:401-409.
King, E.J. 1973. Acid-base behavior. In A.K. Covington and T.
Dickinson (eds.), Physical chemistry of organic solvent systems.
Plenum Press, London.
Kirkland, J.J. 1974. Modern practice of liquid chromatography.
Wiley-Interscience, New York.
Kononova, M.M. 1966. Soil organic matter. 2nd English ed. Pergamon
Press, New York.
Kumada, K. and A. Suzuki. 1961. Isolation of anthraquinones from
humus. Nature 191:415-416.
Laub, R.J. 1974. Packings for HPLC. Research and development.
July 1974:24-28.
LePage, M., R. Beau, and A.J. DeVries. 1968. Evaluation of analyti
cal gel chromatography columns packed with porous silica beads.
J. Polym. Sci., Part C 2:119-130.
Loeppert, R.H., Jr., and B.G. Volk. 1974. Nature of organic matter
extracted from Terra Ceia muck with selected solvents. Soil
and Crop Sci. Soc. Fla. 33:160-164.
Loeppert, R.H., Jr., and B.G. Volk. 1976. Application of high pres
sure liquid chromatography to studies of extractable soil
organic matter-porous silica packings. International Symposium
on Soil Organic Matter Studies, Braunschweig, Germany. (In
Press).
Loeppert, R.H., Jr., L.W. Zelazny, and B.G. Volk. 1976. Acidic
properties of kaolinite in water and acetonitile. Soil Sci.
Soc. Amer. J. (In press).
Mclver, R.D. 1962. Ultrasonicsa rapid method for removing soluble
organic matter from sediments. Geochim. Cosmochim. Acta 26:
343-345.
Mortenson, J.L. 1965. Partial extraction of organic matter. In
C.A. Black (ed.), Methods of soil analysis, Agromony Monograph
No. 9, Vol. 2. American Society of Agronomy. Madison,
Wisconsin.


MATERIALS AND METHODS
Sample Pretreatment: and Extraction
The soil selected for study was the surface horizon (0-25 cm) of
Terra Ceia muck, a Typic Medisaprist (Volk and Schnitzer, 1973). The
salt content was lowered using a dialysis technique employed by Khan
(1971) and refined by Loeppert and Volk (1974). Soil was poured as
a slurry into one dialysis bag, and cation-exchange resin (Amberlite
IR 120) was poured into a second dialysis bag. Both bags were placed
in 0.1 N HC1 and dialysis was continued until ash contents were lowered
to less than 1.0%. Recharged resin was placed daily in the dialysis
chamber. Following dialysis, samples were air-dried.
The extraction procedure is outlined in Fig. 1. Soil extracts
were obtained by 24-hour treatments with the appropriate extractant
(10:1 extractant:soil ratio). Extracts were centrifuged for 2 hours
at 16,300 x gravity (G), filtered through Whatman #42 filter paper,
and purified as described below.
Initial studies were performed to determine the yields and
properties of soil organic matter extracted from the Terra Ceia
muck surface horizon with selected solvents. The experimental pro
cedures and results of these studies are reported elsewhere (Loeppert
and Volk, 1974; Snow, Loeppert, and Volk, 1974).
16


LIST OF FIGURES
Figure Page
1 Extraction scheme 17
2 Infrared patterns 30
3 Effect of flow rate on column efficiency, N, of Porasil
A, Porasil AX, and CPG-250 analytical columns 39
4 Effect of sample size on column efficiency, N, of
Porasil A, Porasil AX,and CPG-250 analytical columns . 41
5 Molecular weight calibration curves of 1 m x 0.318 cm
0D Porasil AX and CPG analytical columns obtained by
elution of polystyrene standards with THF 43
6 Elution patterns of the 2-propanol-Soxhlet extract of
DMF-extractable material on Porasil A and Porasil AX
with selected solvents 47
7 Elution patterns of fulvic acid on Porasil A and Porasil
AX with selected solvents 47
8 Effect of excess neutral electrolyte on elution of Na-
saturated fulvic acid on Porasil A 58
9 Effect of excess neutral electrolyte on elution of Na-
saturated fulvic acid on Porasil AX 58
10 Effect of excess neutral electrolyte on elution of Na-
saturated 1,2,4,5-tetracarboxybenzene on Porasil A . 59
11 Effect of excess neutral electrolyte on elution of Na-
saturated 1,2,4,5-tetracarboxybenzene on Porasil AX. . 59
O
12 Effect of flow rate on column efficiency, N, of 100 A
Poragel and 100 A p-Styragel preparative columns with
THF as the eluting solvent 64
13 Effect of sampleosize on column efficiency, N, of 100 A
Poragel and 100 A y-Styragel preparative columns with
THF as the eluting solvent 66
viii


HIGH-PRESSURE LIQUID CHROMATOGRAPHY AND CHEMICAL CHARACTERIZATION
OF EXTRACTABLE SOIL ORGANIC MATTER
By
RICHARD HENRY LOEPPERT, JR.
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1976


62
gel matrix. As double-layer thickness decreased with resulting reduc
tion in electrostatic exclusion, adsorption increased due to direct H-
bonding interactions of oxygen-containing groups on the organic solute
and Si(OH) sites on the packing material. Adsorption effects were re
duced on Porasil X series of packing materials but were not completely
eliminated. This evidence indicated that the silica surface was not
completely deactivated and/or solute molecules were interacting directly
with the silica surface.
The solubility studies (Table 7) indicated that fulvic acid began
to precipitate at electrolyte concentrations above 0.01 N_, at pH values
of 4.0 to 8.0. Therefore, the partial elution of fulvic acid past
at the higher electrolyte concentrations may be caused by precipitation
of fulvic acid in the column. For this reason, electrolyte concentra
tions must be maintained at values low enough to preclude precipitation
of the solute.
Because of the nature of the silica surface, special precautions
must be observed. As mentioned previously, the silica surface acts as a
weak acid due to the presence of SiOH groups. In the presence of a
protic solvent with basic properties, such as water or methanol, these
acid sites will dissociate, leaving the silica surface with a net nega
tive charge. As the pH of the solvent medium is increased, the dis
sociation of surface sites and the negative charge density of the
silica surface is also increased. The negative charge density of
acidic solute molecules would also increase with increasing pH. There
fore, exclusion of negatively charged solute from the negatively
charged packing material would increase with increasing pH.


28
Table 4. Titratable acidity of extractable soil organic matter
Sample
DMF
Solvent medium
h2o
meq/g
DMF extract
5.9

Hexane-Soxhlet
1.4

Benzene-Soxhlet
1.9

Ethylacetate-Soxhlet
2.9

Acetone-Soxhlet
4.1

2-Propanol-Soxhlet
4.9
2.3
Methanol-Soxhlet
5.7
2.4
Residue
5.8
4.3
Humic acid
6.3
4.0
Fulvic acid
7.3
5.6
Water extract
6.9
5.3
NaOH extract
6.0

Hexane-Soxhlet
1.1

Benzene-Soxhlet
1.7

Ethylacetate-Soxhlet
2.5

Acetone-Soxhlet
4.2

2-Propanol-Soxhlet
4.8
2.0
Methanol-Soxhlet
5.3
2.6
Residue
5.7
3.9


66
SAMPLE SIZE, ml
Fig. 13 Effect of sample size on column efficiency, N, of 100 A
Poragel and 100 A y-Styragel preparative columns with
THF as the eluting solvent


RECORDER RESPONSE
75
Fig. 16 Elution of Soxhlet extracts of DMF-extraetable soil
organic matter on 100 A y-Styragel with THF


71
When DMF was used as the eluting solvent, several of the aromatic
acids were eluted prior to the assumed value of V The more highly
substituted aromatic acids (e.g. 1,2,4,5-tetracarboxybenzene and 1,3,5-
tricarboxybenzene) were eluted near the assumed value of and were
apparently completely excluded from the gel matrix. Several of the
aromatic acids and phenolic acids showed two elution peaks which cor
responded closely to the assumed values of and V Each of the low
molecular weight compounds which were eluted noticeably before V con
tained an acidic side group, COOH and/or phenolic OH.
Several compounds, e.g. benzene, toluene, and anthracene, were
eluted considerably past the assumed value of V^. Each of these com
pounds was hydrophobic in nature and was structurally similar to
monomers of the gel polymer. Benzene, toluene, and anthracene were
more strongly adsorbed with DMF than with THF as the eluting solvent.
This effect was probably due to the more polar character of the DMF.
Neutral solutes (e.g. simple alcohols) and compounds with basic
properties (e.g. pyridine, aniline) were eluted at or slightly after the
assumed V and apparently readily entered the polystyrene-DVB gel
matrix.
Two interesting points from the above observations are that (i)
hydrophobic solutes were more strongly adsorbed to the polystyrene-DVB
gel matrix when eluted with DMF than with THF, and (ii) acidic solutes
were noticeably excluded from the gel matrix when DMF was used as the
eluting solvent, but not when THF was used as the solvent. The first
point may be explained in terms of relative hydrophobic character


41
Fig. 4 Effect of sample size on column efficiency, N, of
Porasil A. Porasil AX, and CPG-250 analytical columns


30
FREQUENCY (cm'1! FREQUENCY Icm-1)
Wavelength, Wavelength,
Fig. 2 Infrared patterns


44
Table 8. Peak elution voluems of extractable soil organic matter on
Porasil A
with
selected
solvents
Solvent
Sample
h2o
CH3OH 2-
-Propanol
t-Butanol Acetone
THF
DMF
_ml -
DMF extract





1.75
Acetone-Soxhlet

1.61
(3.18)3
1.57
(3.19)
1.65 ADb
(3.30)
3.29
1.83
(3.17)
2-Propanol-Soxhlet

1.58
(3.17)
1.55
(3.15)

3.32
1.79
Methanol-Soxhlet
__
1.61
(3.19)

3.31
1.82
NaOH extract





1.78
Acetone-Soxhlet

1.63
(3.17)
1.58
(3.18)
1.62 AD
(3.32)
3.31
1.74
2-Propanol-Soxhlet

1.60
(3.16)
1.57
(3.13)

3.32
1.75
Methanol-Soxhlet
1.60
(3.19)


3.35
1.76
Humic acid





1.75
Fulvic acid
1.57




1.76
Parentheses ( ) indicate secondary peak.
bAD = severe adsorption.


LOG MOLECULAR WEIGHT
43
ELUTION VOLUME, ml
Fig. 5 Molecular weight calibration curves of 1 m x 0.318 cm
OD Porasil AX and CPG analytical columns obtained by
elution of polystyrene standards with THF


80
Liquid Chromatography
None of the chromatographic gels investigated was completely
inert. Each gel apparently interacted with the soil humic material;
therefore, the elution patterns were not entirely attributable to a
molecular seiving phenomenon but to a combination of molecular seiving,
adsorption, and ionic exclusion phenomena.
Solvent and electrolyte effects were especially evident in studies
of Porasil and CPG packing materials. When H-, Na-, K-, or N(CH^)^-
saturated low molecular weight organic acid standards or fulvic acid
were eluted with H90, the solute molecules were partially or totally
excluded from the porous gel matrix. As electrolyte concentration was
increased, the acidic solute molecules more readily entered the porous
matrix; however, at electrolyte concentrations above 0.01 N, significant
quantities of fulvic acid were' adsorbed and eluted past V These
phenomena were attributed to decreased thickness of the electrical
double layer and/or suppression of charge of the negatively charged
solute molecule and the negatively charged silicate surface. Adsorption
of fulvic acid at the higher electrolyte concentrations was attributed
to increased interaction between active sites at the silica surface
and oxygen- and nitrogen-containing functional groups of the organic
solute and also to the possible precipitation of fulvic acid caused by
the high counter ion concentration at the negatively charged silica
surface. Low molecular weight organic solutes with significant basic
properties, i.e. pyridine, were strongly adsorbed to Porasil and CPG


Table 2. Yields of extractable soil organic matter
Terra Ceia
Extracting solvent
NaOH-extractable
material
DMF-extrac table
material
muck
Soil organic
basis
Extract
basis
Soil organic
basis
Extract
basis
Soil organic
basis
/
NaOH
49.0
100.0
49.0
DMF


25.0
100.0
25.0
Hexane
0.1
0.3
0.1
0.2
0.1
Benzene
0.1
0.1
0.1
0.1
0.5
Ethylacetate
1.0
2.1
1.1
4.2
0.7
Acetone
1.4
2.9
2.1
8.4
1.2
2-Propanol
4.4
8.9
1.7
6.8
1.6
Methanol
7.6
15.3
3.8
15.3
5.3
Total Soxhlet
14.6
29.6
8.9
35.0
9.4
Soxhlet residue
34.4
70.4
16.1
65.0
90.6


10
Packing Materials
The various packing materials currently available have been sum
marized by Dark and Limpert (1973), Kirkland (1974), and Laub (1974).
They are generally divided into three major classes: (i) rigid gels
or glasses, (ii) semirigid gels, and (iii) nonrigid gels. The gels
comprising the first group, the rigid gels, are composed of porous
silica and are suitable for high-pressure liquid chromatography. The
semirigid gels (e.g polystyrene-divinylbenzene) are highly cross-linked
organic polymers, will not distort under pressure, and are, therefore,
suitable for high-pressure liquid chromatography. The nonrigid gels
(e.g Sephadex G-gels) are lightly cross-linked organic polymers and
not suitable for high-pressure liquid chromatography since they will
distort under pressure, resulting in an altered pore structure.
The properties of the two porous silica packing materials used
in this study, Porasil and Corning controlled pore glass (CPG), have
been examined by Cooper and Barrall (1973) and Cooper, Bruzzone, Cain,
and Barrall (1971), respectively. Porasil has a higher pore volume than
CPG (Cooper and Barrall, 1973) and therefore has higher V to ratios.
Electron microscope examination has shown that Porasil has a more
heterogeneous pore structure than the CPG packings. Cooper and co
workers have shown both packing materials to be effective for macro-
molecular separation. Cooper and Barrall (1973) concluded that the
heterogeneous pore structure of Porasil results in useful separations
over a wider range of molecular sizes than CPG. They also concluded
that the heterogeneous pore structure of Porasil precludes its use


73
Effect of solvent: on elution of soil humic compounds. Elution
o
patterns of soil humic fractions on 100 A p-Styragel with THF as the
eluting solvent are shown in Figs. 15 and 16. In THF, all humic frac
tions were eluted between V and and apparently readily entered the
porous gel matrix. As mentioned previously, the polystyrene standards
are not suitable for accurate molecular weight determinations of soil
humic materials. These standards, however, do provide a guide for
measurement which is probably no less suitable than others commonly
used, such as proteins or polysaccharides. Molecular weight estimates
based on the polystyrene standards are summarized in Table 18. In all
cases the Soxhlet fractions were estimated to have peak molecular
weights less than 800. These fractions, however, represent only a
minor portion of the total NaOH- or DMF-extractable materials, 28 and
32%, respectively, and are likely to contain materials with lower peak
molecular weights than those of the NaOH- or DMF-extractable materials.
These later materials are not sufficiently soluble in THF to obtain a
molecular weight fractionation.
In DMF, the major portions of all humic fractions were eluted at
volumes corresponding to the assumed values of and were apparently
largely excluded from the gel matrix. In all cases, a minor portion
of the material was eluted at V Reinjection of fractions collected
at Vq produced patterns similar to the original patterns with major
peaks corresponding closely to and minor peaks at the assumed value
of Vr Reinjection of the sample eluted at during the original
fractionation also produced a fractionation pattern similar to the


8
.N NaOH-extractable materials that DMF may be an excellent solvent
for structural studies of the humic fraction. In general, organic
solvents employed to extract soil organic matter have had limited use
because yields were low and because a more specific fraction may be
extracted than with NaOH. Organic solvents which have been investi
gated include acetylacetone (Halstead, Anderson, and Scott, 1966),
anhydrous formic acid (Parsons and Tinsley, 1960), pyridine (Kessler,
Friedel, and Sharkey, 1970), methanol (Mclver, 1962), acetone-^O-
HC1 (Porter, 1967), aqueous THF (Salfeld, 1964), and EDTA (Schnitzer,
Shearer, and Wright, 1959).
Gel Permeation Chromatography
General Information
The practice of high-pressure liquid chromatography is covered in
a text by Kirkland (1974) and reviews by Zweig and Sherma (1974) and
Gaylor, James, and Weetall (1976). Gel-permeation chromatography is a
form of liquid chromatography in which molecules are separated accord
ing to size. The larger molecules are excluded from all or a portion
of the solvent-filled pores of the packing material due to their physi
cal size. On the other hand, a nonreactive small molecule may freely
enter the pores of the packing material.
Theory and Nomenclature
For discussions of chromatographic theory, the reader is referred
to the text by Giddings (1965) and review articles by Bly (1970),
Karger (1971), and Bombough (1971).


18
The DMF-extractable material was evaporated to dryness under vacuum
at 40C with a rotary evaporator and suspended in deionized water. The
sample was transferred to dialysis bags and dialyzed against deionized
water for 6 hours with frequent changing of the external dialysis solu
tion, against 0.1 N HC1 in the presence of strong-acid ion-exchange
resin (Loeppert and Volk, 1974) for 48 hours, and against deionized
water until the external dialysis solution gave a negative test for
Cl This procedure was repeated until a sample with constant nitrogen
content and less than 0.5% ash was obtained. The sample was lyophilized
and stored at 0C.
The water extract was treated similarly to the DMF extract except
the initial dialysis with water was omitted. The dialysis procedure
was repeated until the sample contained less than 0.5% ash.
The NaOH extract was acidified to pH 7.0 with 6.0 N HC1 and con
centrated under vacuum at 30C with a rotary evaporator. The sample was
purified, lyophilized, and stored using the same procedure as with the
DMF extract.
A separate fraction of the NaOH extract was separated into humic
and fulvic acid fractions by adjusting the pH to 2.0 and purified ac
cording to the procedure outlined by Stevenson (1965). Samples were
further purified by the dialysis procedure to an ash content less than
0.5%, lyophilized, and stored at 0C.
Soxhlet fractions were obtained according to the scheme outlined
in Fig. 1 by successive 48-hour extractions with each solvent in the
series. Extracts were concentrated under vacuum at room temperature,


34
Table 6. Solubility of extractable soil organic matter as influenced
by saturating cation and solvent
Sample
Saturating
cation
Solvent
H2
CH3OH
2-Propanol
THF
DMF
DMSO
DMF extract
H
Pa
p
P
P
sa
s
Na
s
s
S
I3
P
S
N(CH )
s
s
s
P
P
S
n(C4H9)4
s
s
s
P
P
S
DMF extract
acetone-Soxhlet
H
I
p
p
s
s
s
Na
s
s
s
p
P
P
N(CH )
s
s
s
p
P
P
w4
s
s
s
p
P
S
DMF extract
2-propanol-Soxhlet
H
I
s
s
s
s
S
Na
s
s
s
p
p
P
n(ch3)
s
s
s
p
p
P
s
s
s
p
p
S
DMF extract
methanol-Soxhlet
H
p
s
p
p
s
S
Na
s
s
s
p
p
p
N(CH )
s
s
s
p
p
p
VVi
s
s
s
p
p
s
Humic acid
H
p
p
p
p
s
s
Na
s
s
s
p
p
p
N(CH )
s
s
s
p
p
s
VV*,
p
s
s
p
s
s
Fulvic acid
H
s
s
p
p
s
s
Na
s
s
s
p
p
p
N(CH )
s
s
s
p
p
s
s
s
s
p
s
s
3S = soluble, P
partially soluble, I
insoluble


7
Loeppert and Volk (1974) investigated the yields and properties
of extractable organic matter solubilized from Terra Ceia muck (Typic
Medisaprist) by a series of organic and inorganic extracting solvents.
Quantity of organic material extracted with dimethylformamide (DMF),
pyridine, and methanol were substantially increased when the soil
ash content was lowered by dialysis prior to extraction. Yields ob
tained with less polar extractants (acetonitrile, chloroform, acetone,
and benzene) were not highly influenced by ash content. Choudhri and
Stevenson (1957) and Bremner and Lees (1949) were able to significantly
increase extraction yields with NaOH by pretreating the soil with 0.1
N HC1 to lower the ash content and remove exchangeable cations. Ex
traction yields from Pahokee muck (Typic Medisaprist) obtained by
single 24-hour extractions with 0.5 N NaOH (10:1 extractant:soil ratio)
were increased from 23 to 40% following pretreatment with 0.1 N HC1
(Snow, Loeppert, and Volk, 1974). Similarly, extraction yields of
Terra Ceia muck were increased from 38% to 49% following treatment with
0.1 11 HC1 (Loeppert and Volk, 1974). When DMF was used as the extract
ing solvent, extraction yields were increased from 0.5% to 25% follow
ing pretreatment with 0.1 N HC1. In these studies, it was observed
that the organic material extracted by DMF was similar in functional
group and elemental analyses to that extracted by 0.5 N NaOH, and
that H2O extracted a material with properties similar to those of
fulvic acid. Loeppert and Volk (1974) concluded from the high yields
obtained with DMF following pretreatment with 0.1 N HC1, the relative
mildness of DMF, and the similarity in properties of DMF- and 0.5


THEORETICAL PLATE COUNT, N
64
FLOW RATE, ml/min
Fig. 12 Effect of flow rate on column efficiency, N, of 100 A
O
Poragel and 100 A u-Styragel preparative columns with
THF as the eluting solvent


11
for studying theoretical proposals relating polymer elution character
istics to pore size dimensions.
A third type of rigid gel, Bioglass, is manufactured by a pro
cess similar to the Corning glasses; however, it has an intentionally
broad pore size distribution (Cooper and Bruzzone, 1973).
The rigid gels exhibit severe adsorption properties (Cooper,
Johnson, and Porter, 1973; Dark and Limpert, 1973; Spatorico, 1975)
which are attributed to OH groups on the surface and Lewis acid sites
present from the manufacturing process. Adsorption effects may be
reduced through deactivation of OH groups; however, Lewis acid sites
are not deactivated by these procedures (Dark and Limpert, 1973).
Deactivation procedures include chemical treatment with polyethylene
oxide (Hiatt et al. 1971; Hawk, Cameron, and Dufault, 1972) or diethylene
glycol (LePage, Beau, and DeVries, 1968) and permanent deactivation by
silyation with hexamethyldisilazane (Cooper and Johnson, 1969) and
trimethylchlorosilane (Unger et al., 1974).
Commercial Porasil packing material is chemically deactivated
by adsorbed polyethylene oxide and distributed under the trade name
Porasil X.
Essentially all solvents are compatible with the porous silicas
and glasses, except alkaline solvents which will dissolve the silica
(Dark and Limpert, 1973). Spatorico and Beyer (1975) observed strong
adsorption to the porous glass of polymers containing cationic groups,
and found that treatment of the glass with polyethylene oxide, or
surfactants, was not successful in eliminating adsorption.


5
Khan, 1972) showed that fulvic acid had lower carboxyl contents, higher
oxygen contents, and higher total acidity and carboxyl content.
Infrared spectra of soil humic compounds show broad absorption
bands (Schnitzer and Khan, 1972). The majority of spectra did not
show absorption bands in the 600-900 cm ^ region, and,therefore, did
not demonstrate the presence of aromatic protons. Likewise, nuclear
magnetic resonance (NMR) spectra of methylated fulvic acid did not
indicate the presence of aromatic protons (Schnitzer and Skinner,
1968).
Schnitzer and Khan (1972) determined molecular weights of 1684
and 669 for humic acid and fulvic acid, respectively, by the freezing
point depression method; however, molecular weights as high as 100,000,
or greater, have been determined by other methods (Schnitzer and Khan,
1972).
Electron spin resonance (ESR) spectra of soil humic substances
indicate the presence of a high concentration of free radicals with
unpaired electrons. Possible sources include semiquinone polymer,
hydroxyquinone, or condensed polynuclear hydrocarbons (Steelink and
Tollin, 1967).
Various degradation procedures have been used to separate com
plex molecules into monomeric components. Kumada and Suzuki (1961)
and Cheshire et al. (1967) identified polycyclic aromatic compounds
following alkaline permaganate oxidation. Hansen and Schnitzer (1969),
on the contrary, obtained no polycyclic aromatic compounds but did
obtain aliphatic carboxylic acids and all possible benzene carboxylic


54
Especially in water and methanol, the negative surface sites would be
stabilized as a result of the acidic properties of the solvent mole
cules. The exclusion of acidic organic solute from the porous matrix
of Porasil A may therefore be at least partially attributed to electro
static repulsion between the charged solute molecules and the charged
silica surface. In the absence of excess neutral salt, the silica
would have an expanded electrical double layer and the solute molecules
would exist with larger effective radii. Therefore, it is possible
that low molecular weight solutes may be completely excluded from the
porous gel matrix.
As mentioned previously, there was no evidence of adsorptive inter
action between the silica surface and the acidic solute in water,
methanol, and DMF; however, in 2-propanol, t-butanol, acetone, ethylace-
tate, and THF there was evidence of adsorption. In the former solvents,
the greater negative charge densities of the solutes and the silica
surface may have resulted in less adsorptive interaction between the
negatively charged species. In acetone, ethylacetate, and THF, however,
the organic solute would be much less dissociated as a consequence of
the very weak or negligible basic properties of these solvents. Also,
the silica surface would be less highly dissociated. Therefore, there is
a more favorable condition for direct H-bonding interactions between the
silica surface and the solute molecules.
Deactivation of the silica surface with polyethylene glycol would
block the reactive sites (Dark and Limpert, 1973) and result in reduced
negative charge density of the silica surface in water and methanol.


65
equilibrium of solute molecules between internal pore space and inter
stitial pore space.
Based on these studies, solvent flow rates of 0.8 and 3.0 ml per
min. in 0.954-cm OD columns were used for Poragel and y-Styragel
columns, respectively, in all subsequent studies.
The effect of sample size on column efficiencies of 0.954-cm
diameter columns of Poragel and y-Styragel are shown in Fig. 13. In
general, the maximum sample volumes were 250 yl and 25 yl for the
Poragel and y-Styragel columns, respectively. Larger samples resulted
in increased peak broadening and reduced apparent column efficiencies.
Column parameters. Column parameters (Table 1), and V^, of the
packed columns were determined by elution of acetone or 2,600,000 molecu
lar weight polystyrene, respectively. Column efficiencies, N, were
O o
approximately 800 and 5,000 for the 100 A Poragel and 100 A y-Styragel,
respectively. Molecular weight calibration curves, obtained by elution
of polystyrene standards with THF gave working molecular weight ranges
of 500 to 20,000 and 100 to 3,000, respectively, for the above gels
(Fig. 14).
Several of the highly substituted organic acid standards deviated
from the polystyrene calibration curve; therefore, the polystyrene
standards are not suitable for accurate determination of molecular
weights of low molecular weight organic acids. Based on the above ob
servation, it is doubtful that polystyrene standards would be suitable
standards for molecular weight determinations of soil humic compounds.


3
were compatible with organic solvents, but not with aqueous solvents,
we were also interested in the behavior of soil humic compounds in
organic solvents and in selection of organic solvents which were
suitable solvent media for extraction and preliminary fractionation
of the soil humic complex.


LITERATURE REVIEW
Soil Organic Matter
For discussions of our current knowledge of the chemical makeup
of soil organic matter the reader is referred to texts by Kononova
(1966) and Schnitzer and Khan (1972) and a review by Hurst and Burges
(1967). Some of the important characteristics and properties which
are relevant to the discussion in the text are summarized in the next
few paragraphs.
Hurst and Burges (1967) suggested that humic acids are polycon
densates of monomers immediately available in a particular microarea
of the soil and do not appear to have integrity of structure and the
rigid chemical configuiction of many other macromolecules due to the
complexity and heterogeneous nature of the system in which they are
formed. According to Kononova (1966), possibly no two humus molecules
would have exactly the same structure.
Elemental and functional group compositions for representative
humic and fulvic acids have been tabulated by Schnitzer and Khan (1972).
The most striking features of these tabulations are the relatively high
oxygen contents, low nitrogen and sulfur contents, high carbon to
hydrogen ratios, high total acidity, and high carboxyl and phenolic
hydroxyl contents. Comparisons of humic and fulvic acid (Schnitzer and
4


17
Fig. 1 Extraction scheme


46
Table 10. Peak elution volumes of extractable soil organic matter on
CPG-250 with selected solvents
Solvent
Sample f^O CH^OH 2-Propanol t-Butanol Acetone THF DMF
ml
DMF extract




1.77
Acetone-Soxhlet
1.75
3.73
3.97
AD
3.87
1.82
ADa
AD
2-Propanol-Soxhlet
1.70
3.71


3.85
1.75
Methanol-Soxhlet
1.71




1.77
NaOH extract





1.79
Acetone-Soxhlet
1.76
3.84
4.12
AD
3.81
1.81
2-Propanol-Soxhlet
1.71
3.67


3.83
1.76
Methanol-Soxhlet
1.69




1.79
Humic acid





1.76
Fulvic acid
1.71 1.74
1.72
a
AD
severe adsorption


32
Table 5. Solubility of extractable soil organic matter in selected
solvents at 0.1% concentration
Sample
Solvent
Hexane
Benzene
Ethyl-
acetate
Acetone
Methyl-
ethyl
ketone
Methyl-
isobutyl
ketone
DMF extract
Ia
I
Pa
P
P
P
Hexane-Soxhlet
sa
S
P
P
P
P
Benzene-Soxhlet
P
S
S
S
S
S
Ethylacetate-Soxhlet
I
P
S
S
S
S
Acetone-Soxhlet
I
I
S
S
S
S
2-Propanol-Soxhlet
I
I
p
P
P
P
Methanol-Soxhlet
I
I
I
P
I
I
Residue
I
I
I
I
I
I
Humic acid
I
I
p
P
P
P
Fulvic acid
I
I
I
I
I
I
Water extract
I
I
I
I
I
I
NaOH extract
I
I
p
P
P
P
Hexane-Soxhlet
S
S
p
P
P
P
Benzene-Soxhlet
p
S
s
S
S
S
Ethylacetate-Soxhlet
I
P
s
S
S
S
Acetone-Soxhlet
I
I
s
S
S
S
2-Propanol-Soxhlet
1
I
p
P
P
p
Methanol-Soxhlet
I
I
I
P
I
I
Residue
I
I
I
I
I
I
S = soluble, P = partially soluble, I =
insoluble


87
Neddermeyer, P.A. and L.B. Rogers. 1968. Gel filtration behavior of
inorganic salts. Anal. Chem. 40:755-762.
Ortiz de Serra, M.I. and M. Schnitzer. 1973. The chemistry of humic
and fulvic acids extracted from Argentine soils. II. Perman
ganate oxidation of methylated humic and fulvic acids. Soil
Biol. Biochem. 5:287-296.
Parsons, J.W. and J. Tinsley. 1960. Extraction of organic matter
with anhydrous formic acid. Soil Sci. Soc. Amer. Proc. 24:198-
201.
Porter, L.K. 1967. Factors affecting the solubility and possible
fractionation of organic colloids extracted from soil and leo-
nardite with an acetone-H O-HCl solvent. J. Agr. Food Chem.
15:807-811.
Posner, A.M. 1966. The humic acid extracted by various reagents
from a soil. I. Yield, inorganic components, and titration
curves. J. Soil Sci. 17:65-68.
Posner, A.M. 1963. Importance of electrolyte in the determination
of molecular weights by Sephadex gel filtration with special
reference to humic acid. Nature 198:1161-1163.
Salfeld, J.C. 1964. Fractionation of a humus preparation with aqueous
organic solvents. Landbauferschung Volkenrode 14:131-136.
Schnitzer, M. and V.C. Gupta. 1965. Determination of acidity in soil
organic matter. Soil Sci. Soc. Amer. Proc. 29:274-277.
Schnitzer, M. and S.U. Khan. 1972. Humic substances in the environ
ment. Marcel Dekker, Inc., New York.
Schnitzer, M., D.A. Shearer, and J.R. Wright. 1959. A study in the
infrared of high molecular weight organic matter extracted by
various reagents from a podzolic B horizon. Soil Sci. 87:
252-257.
Schnitzer, M. and S.I.M. Skinner. 1968. Gel filtration of fulvic
acid, a soil humic compound, jin Isotopes and radiation in soil
organic matter studies. International Atomic Energy Agency,
Vienna.
Snow, J.T., R.H. Loeppert, Jr., and B.G. Volk. 1973. Yields and
functional group characteristics of NaOH-extracted humic acid
from Pahokee muck. Soil and Crop Sci. Soc. Fla. Proc. 33:
165-167.


15
when distilled water was used as eluent and were reduced by adding
electrolyte. Swift and Posner suggested that fractionation based
solely on molecular weight can be achieved by using alkaline buffers
containing large amino cations.


HIGH-PRESSURE LIQUID CHROMATOGRAPHY AND CHEMICAL CHARACTERIZATION
OF EXTRACTABLE SOIL ORGANIC MATTER
By
RICHARD HENRY LOEPPERT, JR.
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1976

To my parents,
Richard and Adeline Loeppert

ACKNOWLEDGMENTS
The author expresses sincere appreciation to Dr. J. G. A. Fiskell,
chairman, and Dr. B. G. Volk, cochairman, of the supervisory committee,
for their guidance, encouragement, and assistance during the progress
of this investigation. Appreciations are also extended to Dr. D. H.
Hubbell, Dr. N. Gammon, and Dr. W. S. Brey for their interest and
participation on the supervisory committee and review of manuscript.
Special appreciations are extended to Dr. L. W. Zelazny and Dr.
M. A. Battiste for important discussions and inspiration provided during
early stages of the investigation. A sincere thanks is extended to
faculty, staff, and students in the Soil Science Department for the many
stimulating discussions which served as the basis for the evolution
of this study.
A very special thank you is extended to Ms. Carolyn Beale and Mr.
Jerry Osbrach for assistance in the laboratory and to Ms. Ann Barry
for typing portions of the original manuscript. The author pays a
special tribute to Ms. Nancy McDavid for the very professional typing
and careful review of the manuscript and to Ms. Helen Huseman for
final preparation and drafting of several of the figures.
The author expresses his sincere gratitude to Dr. C. F. Eno,
chairman of the Soil Science Department at the University of Florida,
ii i

and to Dr. D. W. Beardsley and Dr. D. H. Myhre, Center Directors, at
the Agricultural Research and Education Center, Belle Glade, for pro
viding the research assistantship which has enabled the author to
pursue his doctoral program.
The author deeply appreciates the continuing encouragement and
assistance given him by his parents throughout the course of his studies
and the special upbringing which has encouraged the author to search,
to question, and to approach problems with an open mind.
IV

TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS iii
LIST OF TABLES vi
LIST OF FIGURES viii
ABSTRACT x
INTRODUCTION 1
LITERATURE REVIEW 4
Soil Organic Matter 4
Extraction of Soil Organic Matter 6
Gel Permeation Chromatography 8
General Information 8
Theory and Nomenclature 8
Packing Materials 10
MATERIALS AND METHODS 16
Sample Pretreatment and Extraction 16
Solubility Studies 19
Analytical Determinations 19
High Pressure Liquid Chromatography 21
RESULTS AND DISCUSSION 25
Chemical Characteristics of Extractable Organic Matter. . 25
Solubility Characteristics of Extractable Organic Matter. 31
High Pressure Liquid Chromatography 38
Porous Silica Packing Materials 38
Polystyrene-divinylbenzene (DVB) 63
CONCLUSIONS 78
Extraction and Fractionation 78
Solubility Properties 79
Liquid Chromatography 80
LITERATURE CITED 83
BIOGRAPHICAL SKETCH 89
v

Table
LIST OF TABLES
Page
1 Parameters of column packing materials 23
2 Yields of extractable soil organic matter 26
3 Elemental and functional group concentrations of
extractable soil organic matter 27
4 Titratable acidity of extractable soil organic matter 28
5 Solubility of extractable soil organic matter in
selected solvents at 0.1% concentration 32
6 Solubility of extractable soil organic matter as
influenced by saturating cation and solvent 34
7 Solubility of fulvic acid in aqueous salt solutions . 35
8 Peak elution volumes of extractable soil organic matter
on Porasil A with selected solvents 44
9 Peak elution volumes of extractable soil organic matter
on Porasil AX with selected solvents 45
10 Peak elution volumes of extractable soil organic matter
on CPG-250 with selected solvents 46
11 Peak elution volumes of organic standards on Porasil A
with selected solvents 49
12 Peak elution volumes of organic standards on Porasil AX
with selected solvents 50
13 Peak elution volumes of organic standards on CPG-250
with selected solvents 51
14 Peak elution volumes of cation-saturated fulvic acid on
Porasil A, Porasil AX, and CPC-l^n with water as
eluting solvent 56
vx

LIST OF TABLES (continued)
Table Page
15 Peak elution volumes of selected organic acid standards
on Porasil A with NaSO. solutions 60
2 4
16 Peak elution volumes of selected organic acid standards
on Porasil AX with Na^SO, solutions 61
2 4
17 Peak elution volumesof low molecular weight standards
eluted on 100 A Poragel with THF and DMF and on 100 A
p-Styragel with THF as the eluting solvent 70
18 Molecular weight estimates of soil humic fractions based
on elution of polystyrene standards on p-Styragel with
THF as the eluting solvent 77
vii

LIST OF FIGURES
Figure Page
1 Extraction scheme 17
2 Infrared patterns 30
3 Effect of flow rate on column efficiency, N, of Porasil
A, Porasil AX, and CPG-250 analytical columns 39
4 Effect of sample size on column efficiency, N, of
Porasil A, Porasil AX,and CPG-250 analytical columns . 41
5 Molecular weight calibration curves of 1 m x 0.318 cm
0D Porasil AX and CPG analytical columns obtained by
elution of polystyrene standards with THF 43
6 Elution patterns of the 2-propanol-Soxhlet extract of
DMF-extractable material on Porasil A and Porasil AX
with selected solvents 47
7 Elution patterns of fulvic acid on Porasil A and Porasil
AX with selected solvents 47
8 Effect of excess neutral electrolyte on elution of Na-
saturated fulvic acid on Porasil A 58
9 Effect of excess neutral electrolyte on elution of Na-
saturated fulvic acid on Porasil AX 58
10 Effect of excess neutral electrolyte on elution of Na-
saturated 1,2,4,5-tetracarboxybenzene on Porasil A . 59
11 Effect of excess neutral electrolyte on elution of Na-
saturated 1,2,4,5-tetracarboxybenzene on Porasil AX. . 59
O
12 Effect of flow rate on column efficiency, N, of 100 A
Poragel and 100 A p-Styragel preparative columns with
THF as the eluting solvent 64
13 Effect of sampleosize on column efficiency, N, of 100 A
Poragel and 100 A y-Styragel preparative columns with
THF as the eluting solvent 66
viii

LIST OF FIGURES (continued)
Figure Page
14 Molecular weight calibration curve of y-Styragel (2 f
x 0.954 cm OD 100 A + 2 f x 0.954 cm OD 500 A) obtained
by elution of polystyrene standards with THF 67
15 Elution of Soxhlet extracts of NaOH-extractable soil
O
organic matter on 100 A y-Styragel with THF 74
16 Elution of Soxhlet extracts of DMF-extractable soil
O
organic matter on 100 y-Styragel with THF 75
ix

Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
HIGH-PRESSURE LIQUID CHROMATOGRAPHY AND CHEMICAL CHARACTERIZATION
OF EXTRACTABLE SOIL ORGANIC MATTER
By
Richard Henry Loeppert, Jr.
December, 1976
Chairman: Dr. John G. A. Fiskell
Cochairman: Dr. B. G. Volk
Major Department: Soil Science
The objective of this investigation was to evaluate the use of
high-pressure liquid chromatography and a series of new packing materi
als (Porasil-silica gel, Corning controlled pore glass, and polystyrene-
divinylbenzene) for the molecular size fractionation of extractable soil
organic matter. The effects of packing material, solvent, saturating-
cation, concentration of excess electrolyte, and pH on solute-gel-
solvent interactions were investigated. Preliminary experiments were
performed to investigate the behavior of soil humic compounds in organic
solvents and to select solvents which would be suitable for extraction
and fractionation of the soil humic complex.
The soil used was Terra Ceia muck, a Typic Medisaprist. Organic
matter was extracted from the soil by separate treatment with 0.5 N
NaOH and dimenthylformamide (DMF). The NaOH-extractable material was
separated into humic acid and fulvic acid fractions. In addition, both
the NaOH- and DMF-extractable materials were further fractionated with
a Soxhlet extraction scheme. The ash content of all samples was
lowered by dialysis to less than 0.5%.
x

The C and H content of the fractions decreased and the 0 and COOH
contents and total acidity increased according to the following order:
hexane-Soxhlet, benzene-Soxhlet, ethylacetate-Soxhlet, acetone-Soxhlet,
2-propanol-Soxhlet, methanol-Soxhlet. NaOH extract-humic acid-DMF
extract, fulvic acid.
Fulvic acid and organic acid standards prepared in the H-, Na-,
N(CH_).-, and N(C,H),-saturated forms were excluded from the pores
J 4 4 9 4
of Porasil packing material when water was used as the eluting solvent.
Acetone-, 2-propanol-, and methanol-extractable soil organic matter
and organic acid standards were predominantly excluded from the pores
when methanol or DMF was used as the eluting solvent and predominantly
adsorbed when tetrahydrofuran (THF) or acetone was used. Exclusion
phenomena were evident in the presence of organic solvents with sig
nificant basic character and may be attributed to electrostatic repul
sion of negatively charged organic matter by negatively charged sites
on the silica surface. Organic solutes with significant basic charac
ter were adsorbed.
In the presence of 0.05 N excess neutral electrolyte, cation-
saturated fulvic acid and organic acid standards entered the porous
gel matrix due to suppression of charge and/or decreased electrical
double-layer thickness of the negatively charged solute molecules and
the negatively charged silica surface. As electrolyte concentration
was increased, however, adsorption phenomena became more prevalent due
to precipitation at the surface and/or direct interaction between
active sites on the silica surface and oxygen-containing functional
xi

groups of the organic solute. Deactivation of the silica surface
(Porasil X) resulted in reduction but not elimination of adsorption
and electrostatic exclusion phenomena.
Elution of soil humic compounds and low molecular weight standards
on polystyrene-divinylbenzene (DVB) was strongly influenced by the
solvent. Each of the low-molecular weight organic solutes entered
the porous gel matrix when eluted with THF, however, several hydrophobic
aromatic compounds (e.g. benzene, toluene, and anthracene) were ad
sorbed. Soil humic compounds apparently readily entered the polystyrene-
DVB gel matrix when THF was used. When eluted with DMF, however, soil
humic compounds and low molecular weight organic acid standards were
either totally or partially excluded from the gel matrix. This phe
nomenon was attributed to the effect of solvent on dissociation of
solute molecules and to a possible ion-inclusion effect.
None of the gels investigated was chemically inert, and each
apparently interacted with the soil humic material. Interactions
could be minimized, however, by proper selection of gel, solvent, and
concentration of excess electrolyte.
Molecular weights of acetone-, 2-propanol-, and methanol-Soxhlet
fractions were estimated to be 500 to 800, based on comparison of
elution patterns of soil humic fractions with those of polystyrene
standards in THF.
Xll

INTRODUCTION
Soil is an important international resource which serves as the
source of the majority of the world's food supply and as a major sink
for all man-made and natural products in the environment. Organic
matter, due to its reactive nature, has a large influence on soil
properties. In order to understand the chemical properties of soil
organic matter and the exact role of humic fractions in the soil, it
is essential that the scientist understand the chemical structures
involved. Many years of research have given structural clues, however,
due to the complexity of soil organic matter, the science is still in
its infancy.
One approach to the structural problem has been to initially
fractionate and simplify the humic material prior to further analyti
cal investigations.
Since the development of the polydextran gels, there has been
considerable interest in gel filtration for molecular-size fractiona
tion and characterization of soil organic matter (Swift and Posner,
1971). Scientists have used this method to isolate molecular-size
fractions and to obtain estimates of molecular weight of humic sub
stances in soils and natural water.
During the past decade, rapid advances have been made in liquid
chromatography. These advances have been due primarily to the
1

2
development of the high-pressure liquid chromatograph which, in turn,
has made possible the use of small-diameter packing materials and
high efficiency columns. For example, with the new p-packing ma
terials, column efficiencies as high as 5000 theoretical plates per
foot are commonly attained. The rapid rise in the application of
high-pressure liquid chromatography is readily apparent following
a quick glance through any recent issue of Analytical Chemistry.
Attempts at molecular-size fractionation of soil humic materials
have been complicated by the fact that no gel material, including
Sephadex, is completely inert. Therefore, separations may be adversely
affected by gel-solute and gel-solvent interactions which would lead
to misleading results. Also, the humic molecule is a strongly reactive
solute which has a strong tendency to interact with other solute mole
cules and solvent molecules. For this reason, the fractionation of soil
humic compounds is affected by solute-solute and solute-solvent
interactions. Each of the above-mentioned interactions is strongly
influenced by packing material, solvent, saturating cation, concen
tration of excess electrolyte, and pH.
The objectives of this work were to investigate the use of high-
pressure liquid chromatography and a series of new packing materials
for the size fractionation of soil organic matter extracts and to
investigate the effect of packing material, solvent, saturating-
cation, concentration of excess electrolyte, and pH on solute-solute,
gel-solute, solvent-solute, and gel-solvent interactions which would
influence size separations. Since several of the new packing materials

3
were compatible with organic solvents, but not with aqueous solvents,
we were also interested in the behavior of soil humic compounds in
organic solvents and in selection of organic solvents which were
suitable solvent media for extraction and preliminary fractionation
of the soil humic complex.

LITERATURE REVIEW
Soil Organic Matter
For discussions of our current knowledge of the chemical makeup
of soil organic matter the reader is referred to texts by Kononova
(1966) and Schnitzer and Khan (1972) and a review by Hurst and Burges
(1967). Some of the important characteristics and properties which
are relevant to the discussion in the text are summarized in the next
few paragraphs.
Hurst and Burges (1967) suggested that humic acids are polycon
densates of monomers immediately available in a particular microarea
of the soil and do not appear to have integrity of structure and the
rigid chemical configuiction of many other macromolecules due to the
complexity and heterogeneous nature of the system in which they are
formed. According to Kononova (1966), possibly no two humus molecules
would have exactly the same structure.
Elemental and functional group compositions for representative
humic and fulvic acids have been tabulated by Schnitzer and Khan (1972).
The most striking features of these tabulations are the relatively high
oxygen contents, low nitrogen and sulfur contents, high carbon to
hydrogen ratios, high total acidity, and high carboxyl and phenolic
hydroxyl contents. Comparisons of humic and fulvic acid (Schnitzer and
4

5
Khan, 1972) showed that fulvic acid had lower carboxyl contents, higher
oxygen contents, and higher total acidity and carboxyl content.
Infrared spectra of soil humic compounds show broad absorption
bands (Schnitzer and Khan, 1972). The majority of spectra did not
show absorption bands in the 600-900 cm ^ region, and,therefore, did
not demonstrate the presence of aromatic protons. Likewise, nuclear
magnetic resonance (NMR) spectra of methylated fulvic acid did not
indicate the presence of aromatic protons (Schnitzer and Skinner,
1968).
Schnitzer and Khan (1972) determined molecular weights of 1684
and 669 for humic acid and fulvic acid, respectively, by the freezing
point depression method; however, molecular weights as high as 100,000,
or greater, have been determined by other methods (Schnitzer and Khan,
1972).
Electron spin resonance (ESR) spectra of soil humic substances
indicate the presence of a high concentration of free radicals with
unpaired electrons. Possible sources include semiquinone polymer,
hydroxyquinone, or condensed polynuclear hydrocarbons (Steelink and
Tollin, 1967).
Various degradation procedures have been used to separate com
plex molecules into monomeric components. Kumada and Suzuki (1961)
and Cheshire et al. (1967) identified polycyclic aromatic compounds
following alkaline permaganate oxidation. Hansen and Schnitzer (1969),
on the contrary, obtained no polycyclic aromatic compounds but did
obtain aliphatic carboxylic acids and all possible benzene carboxylic

6
acids except benzoic acid. Ortiz de Serra and Schnitzer (1973) iso
lated and identified a number of phenolic acids. Hansen and Schnitzer
(1967) used nitric acid oxidation and identified a seriesnitro-
phenols and aliphatic dicarboxylic, phenolic, and benzenecarboxylic
acids. *
Based on degradative and nondegradative studies, Schnitzer has
proposed a structure for fulvic acid consisting of phenolic and benzene
carboxylic acids joined by hydrogen bonds to form a polymeric matrix of
considerable stability (Schnitzer and Khan, 1972). Numerous additional
structures for humic and fulvic acids have been suggested (Burges,
Hurst, and Walkden, 1964; Flaig, Beutelspacher, and Reitz, 1975;
Haworth, 1971). The complexity and diversity of these structures
demonstrate the probable complexity of the total soil humic complex.
Extraction of Soil Organic Matter
Procedures for extraction of soil organic matter were reviewed
by Mortenson (1965) and Stevenson (1965) Dilute aqueous NaOH is
the most commonly used extractant of soil organic matter. Sodium
hydroxide produces high yields of extractable organic matter; however,
its use has been severely criticized due to chemical alterations
which may occur in alkaline conditions (Bremner and Lees, 1949;
Bremner, 1956; Choudhri and Stevenson, 1957). Bremner (1950) observed
that 09 was adsorbed from the atmosphere by alkaline soil suspensions.
Tinsley and Salam (1961) suggested that condensation reactions between
amino compounds and aldehydes or phenolic compounds may result in
formation of humin-type compounds during NaOH extraction.

7
Loeppert and Volk (1974) investigated the yields and properties
of extractable organic matter solubilized from Terra Ceia muck (Typic
Medisaprist) by a series of organic and inorganic extracting solvents.
Quantity of organic material extracted with dimethylformamide (DMF),
pyridine, and methanol were substantially increased when the soil
ash content was lowered by dialysis prior to extraction. Yields ob
tained with less polar extractants (acetonitrile, chloroform, acetone,
and benzene) were not highly influenced by ash content. Choudhri and
Stevenson (1957) and Bremner and Lees (1949) were able to significantly
increase extraction yields with NaOH by pretreating the soil with 0.1
N HC1 to lower the ash content and remove exchangeable cations. Ex
traction yields from Pahokee muck (Typic Medisaprist) obtained by
single 24-hour extractions with 0.5 N NaOH (10:1 extractant:soil ratio)
were increased from 23 to 40% following pretreatment with 0.1 N HC1
(Snow, Loeppert, and Volk, 1974). Similarly, extraction yields of
Terra Ceia muck were increased from 38% to 49% following treatment with
0.1 11 HC1 (Loeppert and Volk, 1974). When DMF was used as the extract
ing solvent, extraction yields were increased from 0.5% to 25% follow
ing pretreatment with 0.1 N HC1. In these studies, it was observed
that the organic material extracted by DMF was similar in functional
group and elemental analyses to that extracted by 0.5 N NaOH, and
that H2O extracted a material with properties similar to those of
fulvic acid. Loeppert and Volk (1974) concluded from the high yields
obtained with DMF following pretreatment with 0.1 N HC1, the relative
mildness of DMF, and the similarity in properties of DMF- and 0.5

8
.N NaOH-extractable materials that DMF may be an excellent solvent
for structural studies of the humic fraction. In general, organic
solvents employed to extract soil organic matter have had limited use
because yields were low and because a more specific fraction may be
extracted than with NaOH. Organic solvents which have been investi
gated include acetylacetone (Halstead, Anderson, and Scott, 1966),
anhydrous formic acid (Parsons and Tinsley, 1960), pyridine (Kessler,
Friedel, and Sharkey, 1970), methanol (Mclver, 1962), acetone-^O-
HC1 (Porter, 1967), aqueous THF (Salfeld, 1964), and EDTA (Schnitzer,
Shearer, and Wright, 1959).
Gel Permeation Chromatography
General Information
The practice of high-pressure liquid chromatography is covered in
a text by Kirkland (1974) and reviews by Zweig and Sherma (1974) and
Gaylor, James, and Weetall (1976). Gel-permeation chromatography is a
form of liquid chromatography in which molecules are separated accord
ing to size. The larger molecules are excluded from all or a portion
of the solvent-filled pores of the packing material due to their physi
cal size. On the other hand, a nonreactive small molecule may freely
enter the pores of the packing material.
Theory and Nomenclature
For discussions of chromatographic theory, the reader is referred
to the text by Giddings (1965) and review articles by Bly (1970),
Karger (1971), and Bombough (1971).

9
Column parameters are defined in terms of V^, V and column
efficiency, N. The total pore volume of the column, V is determined
by the elution volume of a nonreactive low molecular weight material
which freely enters the solvent-filled pores of the packing material.
The interstitial volume of the column, V^, is determined by the elu
tion volume of a nonreactive high-molecular weight material which is
completely excluded from the pores of the packing material. Both
and V are determined with standard compounds. In practice, the condi
tion of absolute nonreactivity between solute and packing would probably
never be attained, since there is no such thing as a completely inert
gel network (Freeman, 1973) Likewise, there is no such thing as an
entirely inert solute in liquid chromatography. However, by proper
selection of solute, column parameters and can be determined
with a high degree of accuracy.
Column efficiency is expressed by the theoretical plate count,
N, which is determined with the equation,
N =
where
V is the elution volume of the solute, and to is the peak width at
the baseline. Column efficiency is influenced by particle diameter of
the column, linear velocity of the solvent, and how well the column
is packed (Karger, 1971; Dark and Limpert, 1973). Larger (1971) pre
sents an excellent discussion of factors affecting resolution and
column efficiency.

10
Packing Materials
The various packing materials currently available have been sum
marized by Dark and Limpert (1973), Kirkland (1974), and Laub (1974).
They are generally divided into three major classes: (i) rigid gels
or glasses, (ii) semirigid gels, and (iii) nonrigid gels. The gels
comprising the first group, the rigid gels, are composed of porous
silica and are suitable for high-pressure liquid chromatography. The
semirigid gels (e.g polystyrene-divinylbenzene) are highly cross-linked
organic polymers, will not distort under pressure, and are, therefore,
suitable for high-pressure liquid chromatography. The nonrigid gels
(e.g Sephadex G-gels) are lightly cross-linked organic polymers and
not suitable for high-pressure liquid chromatography since they will
distort under pressure, resulting in an altered pore structure.
The properties of the two porous silica packing materials used
in this study, Porasil and Corning controlled pore glass (CPG), have
been examined by Cooper and Barrall (1973) and Cooper, Bruzzone, Cain,
and Barrall (1971), respectively. Porasil has a higher pore volume than
CPG (Cooper and Barrall, 1973) and therefore has higher V to ratios.
Electron microscope examination has shown that Porasil has a more
heterogeneous pore structure than the CPG packings. Cooper and co
workers have shown both packing materials to be effective for macro-
molecular separation. Cooper and Barrall (1973) concluded that the
heterogeneous pore structure of Porasil results in useful separations
over a wider range of molecular sizes than CPG. They also concluded
that the heterogeneous pore structure of Porasil precludes its use

11
for studying theoretical proposals relating polymer elution character
istics to pore size dimensions.
A third type of rigid gel, Bioglass, is manufactured by a pro
cess similar to the Corning glasses; however, it has an intentionally
broad pore size distribution (Cooper and Bruzzone, 1973).
The rigid gels exhibit severe adsorption properties (Cooper,
Johnson, and Porter, 1973; Dark and Limpert, 1973; Spatorico, 1975)
which are attributed to OH groups on the surface and Lewis acid sites
present from the manufacturing process. Adsorption effects may be
reduced through deactivation of OH groups; however, Lewis acid sites
are not deactivated by these procedures (Dark and Limpert, 1973).
Deactivation procedures include chemical treatment with polyethylene
oxide (Hiatt et al. 1971; Hawk, Cameron, and Dufault, 1972) or diethylene
glycol (LePage, Beau, and DeVries, 1968) and permanent deactivation by
silyation with hexamethyldisilazane (Cooper and Johnson, 1969) and
trimethylchlorosilane (Unger et al., 1974).
Commercial Porasil packing material is chemically deactivated
by adsorbed polyethylene oxide and distributed under the trade name
Porasil X.
Essentially all solvents are compatible with the porous silicas
and glasses, except alkaline solvents which will dissolve the silica
(Dark and Limpert, 1973). Spatorico and Beyer (1975) observed strong
adsorption to the porous glass of polymers containing cationic groups,
and found that treatment of the glass with polyethylene oxide, or
surfactants, was not successful in eliminating adsorption.

12
Loeppert and Volk (1976) investigated the use of HPLC for mole
cular size fractionation of soil humic fractions on Porasil and Porasil
X and observed adsorption and electrostatic exclusion phenomena which
were highly dependent on solvent, saturating-cation, and concentration
of excess neutral electrolyte.
The polystyrene-divinylbenzene (DVB) gels, i.e. Poragel and
Styragel, are widely used in the polymer and petroleum industries
(Gaylor, James, and Weetall, 1976). Styragel and Poragel are not com
patible with aqueous solvents, acetone, or alcohols (Dark and Limpert,
1973) and exhibit a high sensitivity to solvent polarity. Changes in
solvent may result in significant changes in the amount of solvation
and swelling of the gel matrix and altered pore-size distributions
of the gel. Therefore, it is usually necessary to pack the gel as a
slurry in the same solvent which is to be used as the eluting solvent.
Edwards and Ng (1968) studied the elution of model compounds on
polystyrene-DVB gels and observed an apparent adsorption of aromatic
compounds to the gel matrix. Adsorption of compounds on polystyrene-
DVB usually caused pronounced tailing (Bergmann, Duffy, and Stevenson,
1971). Cogswell, McKay, and Latham (1971) separated the acidic con
centrate of petroleum distillate, using methylene chloride as solvent,
into four spectroscopically definable fractions and suggested molecular
association of the more acidic fractions in this solvent.
Sephadex has been widely used in studies of soil organic matter.
Although Sephadex is a different type of gel than the materials used
in these studies, a close examination of the material is in order

13
since previous results may be useful in developing separation schemes
and interpreting results with the newer gels. The reader is referred
to an excellent review by Swift and Posner (1971) .
Sephadex G-gels have been shown to strongly adsorb aromatic com
pounds (Gelotte, 1960), heterocyclic compounds (Demetriou et al.,
1968), and phenolic compounds (Sommers, 1966; Brook and Housley,
1969). Gel-phenol affinity is related to the ether bonds in the cross-
linking group rather than to the polysaccharide (dextran) component
of the gel matrix (Determann and Walter, 1968). As degree of cross-
linking of the gel was increased, affinity of phenol for the gel was
also increased. Brook and Munday (1970) suggested that benzene
derivatives are adsorbed onto hydroxyether cross-linking by H-bonds
and that interaction of Sephadex dextran gels with monosubstituted
phenols, anilines, and benzoic acids operates through hydroxyl, amino,
and carboxylic groups, respectively. Gelotte (1960) observed that the
Sephadex bed material contained a small amount of ionized groups,
probably COOH groups, at concentrations of approximately 10 meq per
gram of dry Sephadex. Aromatic amino acids were adsorbed to the bed
material, basic amino acids were strongly adsorbed, and acidic amino
acids were partially excluded from the gel. Demetriou et al. (1968)
likewise found that aromatic compounds with COOH substituents were
excluded from the gel beads when distilled water was used as the
eluting solvent. The same compounds were adsorbed when columns were
eluted with acid-salt solutions. Similar results were observed
during the elution of soil humic acids (Posner, 1966). Humic acid

14
was excluded from the gel matrix when eluted with water, due to the
charge effect. Considerable adsorption was evident when dilute
electrolyte was used as the eluent. It was concluded by Posner that
it is not possible to select a concentration of electrolyte which
would completely eliminate adsorption and exclusion effects.
In studies of lignosulfonate, Forss and Stenlund (1975) concluded
that elution of this material on Sephadex may be infuenced by the
following factors: (i) polyelectrolyte expansion, (ii) ion exclusion,
(iii) ion inclusion, and (iv) steric exclusion. The ion inclusion
effect results from interaction of charged sites in the macroions which
are excluded from the gel with charged sites in more permeable macro
ions. Elution of the lignosulfonate preparation with excess electro
lyte resulted in reduced exclusion of the sample due to reduction in
both the ion exclusion effect and the ion inclusion effect. In a study
of simple electrolytes on Sephadex (Neddermeyer and Rogers, 1968), it
was observed that peaks were badly skewed, with diffuse front and sharp
trailing edges. These skewed peaks were attributed to the ion exclusion
effect produced by ionic solutes and fixed charges in the gel.
Swift and Posner (1971) noted that when humic samples were
eluted with water and the concentration of solute was decreased, a
greater percentage of sample moved into the excluded or near excluded
region. This phenomenon was explained on the basis of double-layer-
theory and decreased suppression of charge at lower concentrations.
Gel-solute interactions were categorized according to coulombic forces,
caused by charged sites on gel and solute, and adsorption, caused by
hydrophilic interaction. Coulombic interactions were most prevalent

15
when distilled water was used as eluent and were reduced by adding
electrolyte. Swift and Posner suggested that fractionation based
solely on molecular weight can be achieved by using alkaline buffers
containing large amino cations.

MATERIALS AND METHODS
Sample Pretreatment: and Extraction
The soil selected for study was the surface horizon (0-25 cm) of
Terra Ceia muck, a Typic Medisaprist (Volk and Schnitzer, 1973). The
salt content was lowered using a dialysis technique employed by Khan
(1971) and refined by Loeppert and Volk (1974). Soil was poured as
a slurry into one dialysis bag, and cation-exchange resin (Amberlite
IR 120) was poured into a second dialysis bag. Both bags were placed
in 0.1 N HC1 and dialysis was continued until ash contents were lowered
to less than 1.0%. Recharged resin was placed daily in the dialysis
chamber. Following dialysis, samples were air-dried.
The extraction procedure is outlined in Fig. 1. Soil extracts
were obtained by 24-hour treatments with the appropriate extractant
(10:1 extractant:soil ratio). Extracts were centrifuged for 2 hours
at 16,300 x gravity (G), filtered through Whatman #42 filter paper,
and purified as described below.
Initial studies were performed to determine the yields and
properties of soil organic matter extracted from the Terra Ceia
muck surface horizon with selected solvents. The experimental pro
cedures and results of these studies are reported elsewhere (Loeppert
and Volk, 1974; Snow, Loeppert, and Volk, 1974).
16

17
Fig. 1 Extraction scheme

18
The DMF-extractable material was evaporated to dryness under vacuum
at 40C with a rotary evaporator and suspended in deionized water. The
sample was transferred to dialysis bags and dialyzed against deionized
water for 6 hours with frequent changing of the external dialysis solu
tion, against 0.1 N HC1 in the presence of strong-acid ion-exchange
resin (Loeppert and Volk, 1974) for 48 hours, and against deionized
water until the external dialysis solution gave a negative test for
Cl This procedure was repeated until a sample with constant nitrogen
content and less than 0.5% ash was obtained. The sample was lyophilized
and stored at 0C.
The water extract was treated similarly to the DMF extract except
the initial dialysis with water was omitted. The dialysis procedure
was repeated until the sample contained less than 0.5% ash.
The NaOH extract was acidified to pH 7.0 with 6.0 N HC1 and con
centrated under vacuum at 30C with a rotary evaporator. The sample was
purified, lyophilized, and stored using the same procedure as with the
DMF extract.
A separate fraction of the NaOH extract was separated into humic
and fulvic acid fractions by adjusting the pH to 2.0 and purified ac
cording to the procedure outlined by Stevenson (1965). Samples were
further purified by the dialysis procedure to an ash content less than
0.5%, lyophilized, and stored at 0C.
Soxhlet fractions were obtained according to the scheme outlined
in Fig. 1 by successive 48-hour extractions with each solvent in the
series. Extracts were concentrated under vacuum at room temperature,

19
dried under a dry-nitrogen jet, and redissolved in the extracting sol
vent at room temperature. Samples were concentrated to 30-ml volume
and following, addition of 30 ml of ^0, were reconcentrated to 30-ml
volume. The reconcentration procedure was repeated several times, and
the aqueous suspensions were transferred to dialysis bags for purifica
tion to less than 0.3% ash. Purified samples were lyophilized and
stored at 0C.
Solubility Studies
The solubility behavior of extractable organic matter and each of
the Soxhlet fractions was determined by placing 2.00 mg of the organic
material in 2 ml of the appropriate solvent. Following agitation the
mixture was visually observed to determine whether the sample was in
soluble, partially soluble or completely soluble. Where appropriate,
the pH of the sample suspension was adjusted to the desired level by
addition of acid or base which contained the same counter ion as the
excess electrolyte.
Analytical Determinations
Carbon and H were determined with the Coleman C-H analyzer, N by
the micro-Kjeldahl method, and S using the Leco induction furnace.
Oxygen was determined by difference with the assumption that C, H, N,
S,and 0 were the only elemental constituents of the extractable organic
matter. Total acidity was determined by addition of excess BaCOH^
and back-titration of unreacted Ba(0H) with HC1 to pH 9.8 (Schnitzer

20
and Gupta, 1965; Schnitzer and Khan, 1972). Total COOH-group concen
tration was determined by addition of excess CaiC^H^op,, and back-
titration of excess C.-,H^0o with HC1 to pH 8.4 (Schnitzer and Gupta,
1965; Schnitzer and Khan, 1972). Total OH was determined by an
acetylation procedure (Brooks, Durie, and Sternhell, 1957) and total
C0 by the oximation method (Fritz, Yamamura, and Bradford, 1959).
Phenolic OH group concentration was calculated as the difference between
total acidity and COOH-group concentration, and alcoholic OH concentra
tion was estimated by subtracting phenolic OH concentration from total
OH concentration. All elemental and functional group analyses were cor
rected for moisture content and ash content.
Moisture content was determined by heating a preweighed combustion
boat with a known quantity of material to HOC for 24 hours and weigh
ing. Ash content was determined by heating a preweighed combustion boat
with a known quantity of organic material to 700C for 4 hours and weigh
ing. All weights, yields, and concentrations are reported on a moisture-
and ash-corrected organic matter basis.
Infrared patterns were obtained on a Perkin-Elmer model 127
spectrophotometer. The KBr pellets were prepared by mixing 0.8 mg of
organic material, previously dried 24 hours over P2O5 in a vacuum
desiccator, with 200 mg of dried KBr in a Wig-L-Bug amalgamator. The
mixture was dried an additional 12 hours in a vacuum desiccator over
P^O^ prior to obtaining the spectrum.
Continuous-and stepped-potentiometric titrations were performed on
10 mg of organic material placed in the appropriate solvent. Titrant

21
was added using a Radiometer autoburette ABU 13 with TTT 50 titrator
module. Potentials were determined with a Radiometer PHM 64 pH meter
and recorded using a Radiometer REC strip chart recorder with REA 160
titrigraph module. A Radiometer combination electrode with porous plug
liquid junction was used for all potentiometric determinations. The
calomel cell was filled with saturated KC1 in H2O for aqueous titra
tions or with saturated KC1 in CH^OH for titrations in nonaqueous
media (Loeppert, Zelazny, and Volk, 1976).
High Pressure Liquid Chromatography
The Waters' ALC 202 liquid chromatograph equipped with the Model
6000 solvent delivery system and 401 differential refractometer and
UV detectors was used for all separations. Samples were injected
through the U6K universal liquid chromatograph injector. A solvent
flow rate of 0.5 ml/minute were used, unless otherwise specified.
Column effluent was monitored by differential refractive index or by
UV absorption at 254 or 280 nm, and recorded continuously on a Perkin
Elmer 201 strip-chart recorder.
Samples were dissolved in the eluting solvent to give a 0.1% con
centration (W/V) and centrifuged at 35,000 x G for 20 min. Injection
volume was 10 yl.
Solvents used in the chromatographic studies were deionized
water and spectroquality methanol, 2-propanol, t-butanol, tetrahydrofuran
(THF), dimethylformamide (DMF), ethylacetate, acetone, and chloroform.

22
Salt solutions in water and methanol were prepared by adding the de
sired quantity of 2.0 N NaOH, tetramethylammonium hydroxide, or
tetrabutylammonium hydroxide to the solvent and adjusting to the
appropriate pH with HC1, HNO^, H^SO^, or H^PO^. Halide salts were
avoided where possible, since these salts may result in corrosion of
the stainless steel surfaces of the liquid chromatographic pump, con
necting lines, and columns. In the few cases where halide salts were
employed, the pH of the eluting solvent was maintained above 7.0, and
the chromatographic system was flushed with copious quantities of de
ionized water following use of the halide salt.
Characteristics of the packing materials used in this study
are summarized in Table 1. The silica and porous glass packing ma
terials were dry-packed into stainless steel columns (0.318 cm OD x
2 f or 0.318 cm OD x 1 m). The semirigid gels were slurry packed in
the same solvent as the eluting solvent. Poragel columns (stainless
steel, 0.954 cm OD x 2 f) packed in THF were used as obtained from the
manufacturer. Poragel in DMF was slurry packed into 0.954 cm OD x 2 f
stainless steel columns. The p-Styragel columns (stainless steel,
0.954 cm OD x 1 f) packed in THF were used as obtained from the manu
facturer. When used in series, columns were connected with U-shaped
0.009 in. ID tubing.
A series of experiments were performed to investigate the elu
tion behavior of extractable soil organic matter on the gel permeation
packing materials: (i) determination of optimum operating conditions,

Table 1. Parameters of column packing materials
Packing material
b
c
d
Approximate
Description
Column
vo
VT
N
molecular weight
a
size
I
working range
cm
ml
ml
Porasil A
porous silica
0.318
1.82
3.35
360
1,000-60,000
Porasil C
f 1
11
1.76
3.36
340
1,000-250,000
Porasil E
f 1
It
1.62
3.42
290
1,000-2,000,000
Porasil AX
porous silica
II
1.84
3.32
340
1,000-60,000
Porasil CX
deactivated with
polyethylene oxide
It
1.88
3.42
280
1,000-250,000
Porasil EX
1!
11
1.67
3.49
230
1,000-2,000,000
CPG 40
crushed glass
If
1.60
3.22
1,020
500-10,000
CPG 250
II
It
1.54
3.31
660
5,000-100,000
Poragel 100 A
polystyrene-DVB
0.954
10.8
21.3
800
500-20,000
Poragel 500 A
II
II
10.6
20.8
800
1,000-100,000
p-Styragel 100 A
II
If
5.2
10.1
5,600
100-3,000
p-Styragel 500 A
II
11
4.6
9.8
5,000
100-10,000
aAll columns are one meter long except the Porage.1 and y-Styragel columns which are 0.610 and 0.305 meters,
respectively. Column size indicates the outside column diameter.
^Determined by elution of 1,600,000 molecular weight polystyrene standard with THF.
c
Determined by elution of acetone or benzene with THF.
^Column efficiency expressed in theoretical plates.

24
(ii) determination of column parameters, (iii) elution behavior of soil
organic extracts and organic standards with selected solvents, (iv)
elution behavior of soil organic extracts and organic standards as
influenced by saturating cations, and (v) elution behavior of soil
organic extracts and organic standards as influenced by excess electro
lyte .

RESULTS AND DISCUSSION
Chemical Characteristics of Extractable Organic Matter
Extraction yields of 25% and 49% were obtained by single treat
ments of the surface horizon of Terra Ceia muck with DMF and 0,5 N NaOH,
respectively, following pretreatment with HC1 to lower the ash content
(Table 2). The Soxhlet solvents were able to solubilize 29.6%, 35.0%,
and 9.4% of the NaOH-extractable material, DMF-extractable material, and
Terra Ceia muck, respectively. Only minor portions of the materials were
extracted by hexane and benzene. The major portions were in the ace
tone, 2-propanol, and methanol fractions. Each of these fractions, how
ever, represent only a minor portion of the total soil organic matter.
Obvious and important differences exist in the elemental and func
tional group concentrations of the extractable organic matter (Table 3).
Comparisons of the Soxhlet fractions indicate that C and H content
decreases, 0 and N content increases, COOH content increases, and total
acidity increases according to the solvent order: hexane, benzene,
ethylacetate, acetone, 2-propanol, and methanol. Potentiometric titra
tions of the Soxhlet fractions in DMF indicated that titratable acidity
increased according to the same solvent order (Table 4).
25

Table 2. Yields of extractable soil organic matter
Terra Ceia
Extracting solvent
NaOH-extractable
material
DMF-extrac table
material
muck
Soil organic
basis
Extract
basis
Soil organic
basis
Extract
basis
Soil organic
basis
/
NaOH
49.0
100.0
49.0
DMF


25.0
100.0
25.0
Hexane
0.1
0.3
0.1
0.2
0.1
Benzene
0.1
0.1
0.1
0.1
0.5
Ethylacetate
1.0
2.1
1.1
4.2
0.7
Acetone
1.4
2.9
2.1
8.4
1.2
2-Propanol
4.4
8.9
1.7
6.8
1.6
Methanol
7.6
15.3
3.8
15.3
5.3
Total Soxhlet
14.6
29.6
8.9
35.0
9.4
Soxhlet residue
34.4
70.4
16.1
65.0
90.6

Table 3. Elemental and functional group concentrations of extractable soil organic matter
Sample
DMF extract
Hexane-Soxhlet
Benzene-Soxhlet
Ethylacetate-Soxhlet
Acetone-Soxhlet
2-Propanol-Soxhlet
Methanol-Soxhlet
Residue
Humic acid
Fulvic acid
Water extract
NaOH extract
Hexane-Soxhlet
Benzene-Soxhlet
Ethylacetate-Soxhlet
Acetone-Soxhlet
2-Propanol-Soxhlet
Methanol-Soxhlet
Residue
Total Phenolic Alcoholic
C H N S 0 acidity COOH C=0 OH OH
weight % meq/g
56.2
5.
7
4.
1
0.
6
33.4
7.2
3.5
2.6
3.7
2.5
81.2
9.
1
0.
1
0.
1
9.5





77.7
8.
8
0.
2
0.
1
13.2





74.2
8.
2
0.
5
0.
1
17.0





69.3
8.
0
0.
6
0.
1
22.0
3.9
1.9
1.9
2.0
1.3
67.5
7.
7
1.
0
0.
3
23.5
4.1
2.0
2.1
2.1
1.4
62.4
6.
5
1.
4
0.
4
29.3
4.5
2.3
2.0
2.2
1.5
56.9
5.
7
3.
7
0.
6
33.1
7.5
4.0
2.6
3.5
2.3
55.7
5.
6
4.
0
0.
8
33.9
7.4
3.8
2.6
3.6
2.4
47.3
4.
4
2.
7
0.
6
45.0
10.1
5.5
2.4
4.6
3.0
50.2
4.
9
2.
6
0.
6
41.7
8.7
4.4
2.8
4.3
2.9
55.6
5.
4
3.
7
0.
7
34.6
7.4
3.7
2.8
3.7
2.5
79.9
9.
0
0.
1
0.
1
10.9





77.2
8.
8
0.
3
0.
1
13.6





75.4
8.
6
0.
5
0.
1
15.4





69.1
8.
1
0.
7
0.
2
21.9
3.7
1.9
1.7
1.8
1.2
68.4
7.
4
1.
1
0.
3
22.8
4.3
2.2
2.1
2.1
1.4
61.7
6.
9
1.
4
0.
3
29.7
4.6
2.4
2.2
2.2
1.4
56.1
5.
5
3.
9
0.
7
33.8
7.5
3.8
2.8
3.7
2.5

28
Table 4. Titratable acidity of extractable soil organic matter
Sample
DMF
Solvent medium
h2o
meq/g
DMF extract
5.9

Hexane-Soxhlet
1.4

Benzene-Soxhlet
1.9

Ethylacetate-Soxhlet
2.9

Acetone-Soxhlet
4.1

2-Propanol-Soxhlet
4.9
2.3
Methanol-Soxhlet
5.7
2.4
Residue
5.8
4.3
Humic acid
6.3
4.0
Fulvic acid
7.3
5.6
Water extract
6.9
5.3
NaOH extract
6.0

Hexane-Soxhlet
1.1

Benzene-Soxhlet
1.7

Ethylacetate-Soxhlet
2.5

Acetone-Soxhlet
4.2

2-Propanol-Soxhlet
4.8
2.0
Methanol-Soxhlet
5.3
2.6
Residue
5.7
3.9

29
Infrared patterns of the Soxhlet fractions (Fig. 2) substanti
ated differences evident in the elemental and functional group analyses.
From the least polar to the most polar extracting solvents, Soxhlet
extracts showed progressively weaker C-H stretching vibrations in the
2900 cm region. Absorption at 1450 cm ^ due to the bending vibra
tion also decreased according to the same solvent order. These adsorp
tion bands of the infrared patterns, along with the C and H concentrations,
indicated that the less polar solvents extracted a material of greater
aliphatic character. Each of the Soxhlet fractions had strong C=0
stretching bands at 1725 cm ^ and 1630 cm ^ resulting from COOH and
C00-, respectively. The band at 1630 cm ^ was considerably stronger
in the absorption spectra of materials extracted by the more polar
solvents. The band at 700-750 cm \ evident in the absorption spectra
of hexane-, benzene-, and 2-propanol-extractable materials, was attributed
to aromatic C-H out-of-plane bending vibrations. The absence of bands
in this region of the spectrum indicates either the absence of aromatic
structure or the possibility of a completely substituted aromatic ring
system. The weak absorption at 700-750 cm ^ for the methanol-Soxhlet
extracts compared to the hexane-, benzene-, ethylacetate-Soxhlet extracts
may be attributed to a highly substituted aromatic ring system. Each
of the Soxhlet-extractable materials showed a strong absorption band
at 3440 cm attributed to H-bonded OH groups and a weaker band at
approximately 1250 cm attributed to C-0 stretching vibrations.
Absorption at both frequencies increased with increasing polarity of
the extracting solvent. The hexane- and benzene-Soxhlet fractions,
especially, showed significantly weaker absorption in the 3440 cm ^
region than fractions extracted by the more polar solvents.

30
FREQUENCY (cm'1! FREQUENCY Icm-1)
Wavelength, Wavelength,
Fig. 2 Infrared patterns

31
DMF- and NaOH-extractable organic matter, and the humic acid
fraction had very similar elemental compositions and concentrations of
O-containing functional groups. The same observation was previously
made by Loeppert and Volk (1974) in comparisons of DMF-and NaOH-ex
tractable materials. These similarities were corroborated by the infra
red patterns (Fig. 2).
The fulvic acid fraction had lower C and H contents, higher 0
content, lower N content, higher total acidity, and higher COOH content
than the humic acid and DMF-extractable materials. The higher acidity
of fulvic acid compared to the other materials was corroborated by the
significantly higher titratable acidity, as determined by potentiometric
titration in DMF.
In summary, C and H content decreased, aliphatic C-H decreased,
0 and COOH contents increased, and total acidity increased according to
the following order of extractable organic matter: hexane-Soxhlet
extract, benzene-Soxhlet extract, ethylacetate-Soxhlet extract, acetone-
Soxhlet extract, 2-propanol-Soxhlet extract, methanol-Soxhlet extract,
NaOH extractDMF extracthumic acid, fulvic acid.
Solubility Characteristics of Extractable Organic Matter
Solubility characteristics of extractable organic matter in
selected solvents and in salt solutions are summarized in Tables 5-7.
Humic acid, and DMF- and NaOH-extractable organic matter were completely
soluble at 0.1% concentration only in DMF, dimethylsulfoxide (DMSO), and

32
Table 5. Solubility of extractable soil organic matter in selected
solvents at 0.1% concentration
Sample
Solvent
Hexane
Benzene
Ethyl-
acetate
Acetone
Methyl-
ethyl
ketone
Methyl-
isobutyl
ketone
DMF extract
Ia
I
Pa
P
P
P
Hexane-Soxhlet
sa
S
P
P
P
P
Benzene-Soxhlet
P
S
S
S
S
S
Ethylacetate-Soxhlet
I
P
S
S
S
S
Acetone-Soxhlet
I
I
S
S
S
S
2-Propanol-Soxhlet
I
I
p
P
P
P
Methanol-Soxhlet
I
I
I
P
I
I
Residue
I
I
I
I
I
I
Humic acid
I
I
p
P
P
P
Fulvic acid
I
I
I
I
I
I
Water extract
I
I
I
I
I
I
NaOH extract
I
I
p
P
P
P
Hexane-Soxhlet
S
S
p
P
P
P
Benzene-Soxhlet
p
S
s
S
S
S
Ethylacetate-Soxhlet
I
P
s
S
S
S
Acetone-Soxhlet
I
I
s
S
S
S
2-Propanol-Soxhlet
1
I
p
P
P
p
Methanol-Soxhlet
I
I
I
P
I
I
Residue
I
I
I
I
I
I
S = soluble, P = partially soluble, I =
insoluble

33
Table 5 (Extended)
Solvent
Ethyl t-Butanol 2-Propanol Methanol Pyridine DMF DMSO H0
ether 2
I
P
S
P
P
I
I
I
P
S
s
s
s
s
P
I
p
p
p
s
s
s
p
I
p
I
I
I
p
s
p
I
p
I
I
I
p
s
s
p
p
1
p
s
s
s
s
p
s
p
s
c
s
s
s
s
s
p
p
s
s
s
s
s
p
I
I
I
I
I
p
p
p
p
p
p
p
s
p
s
I
p
s
p
p
I
I
I
p
s
s
s
s
s
p
I
p
p
p
s
s
s
p
I
p
I
I
I
p
s
p
I
p
I
I
I
p
s
s
p
p
I
p
s
s
s
s
p
s
p
s
s
s
s
s
p
p
s
s
s
s
s
p
I
I
I
I
I
p
p

34
Table 6. Solubility of extractable soil organic matter as influenced
by saturating cation and solvent
Sample
Saturating
cation
Solvent
H2
CH3OH
2-Propanol
THF
DMF
DMSO
DMF extract
H
Pa
p
P
P
sa
s
Na
s
s
S
I3
P
S
N(CH )
s
s
s
P
P
S
n(C4H9)4
s
s
s
P
P
S
DMF extract
acetone-Soxhlet
H
I
p
p
s
s
s
Na
s
s
s
p
P
P
N(CH )
s
s
s
p
P
P
w4
s
s
s
p
P
S
DMF extract
2-propanol-Soxhlet
H
I
s
s
s
s
S
Na
s
s
s
p
p
P
n(ch3)
s
s
s
p
p
P
s
s
s
p
p
S
DMF extract
methanol-Soxhlet
H
p
s
p
p
s
S
Na
s
s
s
p
p
p
N(CH )
s
s
s
p
p
p
VVi
s
s
s
p
p
s
Humic acid
H
p
p
p
p
s
s
Na
s
s
s
p
p
p
N(CH )
s
s
s
p
p
s
VV*,
p
s
s
p
s
s
Fulvic acid
H
s
s
p
p
s
s
Na
s
s
s
p
p
p
N(CH )
s
s
s
p
p
s
s
s
s
p
s
s
3S = soluble, P
partially soluble, I
insoluble

35
Table 7. Solubility of fulvic acid in aqueous salt solutions
Saturating Excess
cation electrolyte
Concentration
2.0
4>

pH
6.0
00
o
10.0
Na
Na SO,
0.000 N
sa
s
s
s
s
0.001 N
s
s
s
s
s
0.01 N
s
s
s
s
s
0.05 N
s
Pa
p
p
s
0.10 N
s
p
p
p
s
K
K0SO,
l 4
0.000 N
s
s
s
s
s
0.001
s
s
s
s
s
0.01 N
s
s
s
s
s
0.05 N
s
p
p
p
s
0.10 N
s
p
p
p
s
n(ch3)4
[N(CH )4]2S04
0.000 N
s
s
s
s
s
0.001
s
s
s
s
s
0.01 N
s
s
s
s
s
' 0.05 N
s
s
s
s
s
0.10 N
s
p
p
p
s
(c4h9)4
[N(C4H9)4J2S4
0.000 N
s
s
s
s
s
0.001 N
s
s
s
s
s
0.01 N
s
s
s
s
s
0.05 N
s
s
s
s
s
0.10 N
s
s
s
s
s
aS = soluble, P = partially soluble

36
0.5 N NaOH. The DMF and DMSO both have significant basic character
(Talhoun and Mortland, 1968). The acidic organic material is highly
dissociated and dispersed in each of these solvents and, therefore, is
soluble. Each of the Soxhlet fractions and fulvic acid were also com
pletely soluble at 0.1% concentration in DMF and DMSO.
Fulvic acid was soluble in methanol and water, in addition to DMF
and DMSO, but was not completely soluble in any of the other solvents.
The solubility of extractable organic matter and Soxhlet fractions was
influenced to a great extent by the saturating cation (Table 6). Ex
change of H+ by Na+, K+, N(CH^)^+, or C(C^Hg)^+ resulted in increased
solubility of the organic solutes in water, methanol, or 2-propanol,
and decreased solubility in DMF. For example, at 0.1% concentration,
the H-saturated DMF- and NaOH-extractable materials were only partially
soluble in water, methanol, or 2-propanol; however, the salt-saturated
solutes were completely soluble. On the other hand, the H-saturated
2-propanol-Soxhlet fraction and fulvic acid were soluble at 0.1% concen
tration in DMF whereas the salt-saturated material was only slightly
soluble. These solubility characteristics greatly limit the solvent-
electrolyte combinations which are applicable for exclusion chroma
tography. The enhanced solubility of the cation-saturated samples in
the protic solvents (water, methanol, and 2-propanol) may be attributed
to acidic properties of these solvents (King, 1973) which promote
stabilization of the solute anion. The very weakly acidic dipolar
aprotic solvents (e.g. THF, DMF) would not stabilize the solute anion
to as great an extent as the protic solvents.

37
The presence of excess neutral salt affected the solubility of
fulvic acid in water (Table 7) and methanol. Some very interesting
trends were evident in these studies. The H-, Na-, K-, N(CH^)^-, and
NfC^H^-saturated fulvic acid samples at 0.1% concentration were
soluble at pH 2.0 in all concentrations of excess neutral electrolyte
-3 -2
up to 0.1 N^. In the presence of 10 N or 10 N excess salt, 0.1%
fulvic acid remained completely dissolved as the pH was increased suc
cessively to pH 4.0, 6.0, 8.0, and 10.0. However, in the presence of
-2
5 x 10 N excess salt, K-saturated fulvic acid began to precipitate
at pH 4.0. The sample redissolved at pH 8.0 and was completely soluble
f*
as the pH was increased to 10.0. Fulvic acid saturated with N(CH^)^
or N(CAH9^+ remained completely dissolved as the pH was increased from
-2
2.0 to 10.0, in the presence of 5 x 10 N excess neutral salt. As the
ionic strength was increased to 5 x 10 ^ N, however, K-, Na-, NCH^)^-,
and NCC^Hg^-saturated fulvic acid precipitated as the solution pH ap
proached 4.0 and redissolved as the pH approached 10.0. The precipita
tion was greatest in the approximate range of pH 4 to pH 7 and may be
attributed to unfavorable conditions for the electrostatic dispersion
of molecular units. It is interesting to note that greatest precipita
tion occurred within the pH range at which greatest neutralization of
acidic carboxyl groups would occur.
The precipitation phenomena in the presence of excess neutral salt
greatly limits the conditions which may be employed for gel permeation
separations of extractable organic matter.

38
High Pressure Liquid Chromatography
Porous Silica Packing Materials
Operating conditions. Column efficiencies of Porasil A and
Porasil AX packing materials were greatly influenced by solvent flow
rates. With 0.318-cm OD analytical columns, maximum column efficiencies
and minimum peak broadening were obtained at flow rates of approximately
0.1 ml/min.; however, column efficiencies were not significantly differ
ent at flow rates between 0.1 and 0.6 ml per min. (Fig. 3). At flow
rates greater than 0.6 ml per min., peak broadening was increased and
column efficiencies were decreased. For this reason, it was concluded
that low flow rates should be maintained with the Porasil packing
materials.
It is interesting to observe that for the Porasil packing materials,
there was a slight increase in column efficiency as flow rate was de
creased to 0.1 ml per min. (Fig. 3). These results may be compared with
those obtained with the CPG packing materials for which maximum column
efficiencies were obtained at flow rates of 0.4 ml per min. Decreases
in column efficiency were observed when flow rates were decreased or
increased from this value. As with the Porasil packing material, in
creases in solvent flow rates above 1.0 ml per min. resulted in peak
broadening and significant decreases in column efficiency.
The different behavior of the Porasil and CPG packing materials
at low flow rates may be at least partially attributed to the more

39
Fig. 3 Effect of flow rate on column efficiency, N, of Porasil
A, Porasil AX, and CPG-250 analytical columns

40
uniform pore structure of the controlled pore glass material which may
result in reduced blockage of large pores by small pores and less re
stricted diffusion of solute molecules through the gel matrix. The
more uniform pore structure may allow the use of higher flow rates.
Cooper and Barrall (1973) suggested that "pooling" or solute restric
tion in porous silica media, necessitates the use of low flow rates
with these materials.
Based on these studies, solvent flow rates of 0..5 ml per min.
were selected for all subsequent studies using the 0.318-cm OD columns
with the Porasil and CPG packing materials. Based on similar studies
with 0.954-cm OD preparative columns, solvent flow rates of 1.5 ml
per min. were used for all subsequent studies on these columns packed
with Porasil or CPG packing materials.
The effects of sample size on column efficiencies of Porasil and
CPG packing materials are summarized for the analytical columns (Fig.
4). In general, the maximum sample volumes were 20 pi for the analyti
cal columns and 100 pi for preparative columns. Larger sample volumes
resulted in increased peak broadening and reduced apparent column
efficiencies. Column efficiencies were not noticeably affected with
lower sample volumes.
Even though column efficiencies are reduced with large sample
volumes, column overloading may be helpful in obtaining preparative
fractions, especially when used in conjunction with recycle chroma
tography (Bombaugh, 1971).

41
Fig. 4 Effect of sample size on column efficiency, N, of
Porasil A. Porasil AX, and CPG-250 analytical columns

42
Column parameters. Elution characteristics of packed columns are
summarized in Table 1. Column parameters, and V were determined
from the elution volumes for acetone or benzene and blue dextran 2,000
or 2,600,000 molecular weight polystyrene, respectively. In all cases
the column efficiencies, indicated by theoretical plate count, N, de
creased with increasing internal pore size of the packing material.
For example, the column efficiencies of Porasil AX, CX, and EX were
340, 280, and 230 theoretical plates per meter, respectively.
Molecular weight calibration curves (Fig. 5) were obtained by
elution of polystyrene standards with THF. The approximate molecular
weight working ranges, as determined with the polystyrene standards
are summarized in Table 1. The working curves obtained with polystyrene
standards on the Porasil and the CPG packing materials were not linear
over the working range of the gels.
Effect of solvent. Peak elution volumes of extractable organic
matter on Porasil A, Porasil AX, and CPG-250 packing materials are
summarized in Tables 8-10, respectively. Elution patterns of the
2-propanol-extractable material and fulvic acid are shown in Figs. 6
and 7. Samples were completely soluble at 0.1% concentration in each
of the solvents shown. A portion of the fulvic acid sample (Fig. 7) was
eluted at V the elution volume of a nonreactive high molecular weight
solute, on the Porasil A column when methanol, water, or DMF was used
as the eluting solvent. The relative quantity of sample eluted at
increased according to the following solvent order: methanol < DMF < H20.
Likewise, portions of the acetone-, 2-propanol-, and methanol-Soxhlet

LOG MOLECULAR WEIGHT
43
ELUTION VOLUME, ml
Fig. 5 Molecular weight calibration curves of 1 m x 0.318 cm
OD Porasil AX and CPG analytical columns obtained by
elution of polystyrene standards with THF

44
Table 8. Peak elution voluems of extractable soil organic matter on
Porasil A
with
selected
solvents
Solvent
Sample
h2o
CH3OH 2-
-Propanol
t-Butanol Acetone
THF
DMF
_ml -
DMF extract





1.75
Acetone-Soxhlet

1.61
(3.18)3
1.57
(3.19)
1.65 ADb
(3.30)
3.29
1.83
(3.17)
2-Propanol-Soxhlet

1.58
(3.17)
1.55
(3.15)

3.32
1.79
Methanol-Soxhlet
__
1.61
(3.19)

3.31
1.82
NaOH extract





1.78
Acetone-Soxhlet

1.63
(3.17)
1.58
(3.18)
1.62 AD
(3.32)
3.31
1.74
2-Propanol-Soxhlet

1.60
(3.16)
1.57
(3.13)

3.32
1.75
Methanol-Soxhlet
1.60
(3.19)


3.35
1.76
Humic acid





1.75
Fulvic acid
1.57




1.76
Parentheses ( ) indicate secondary peak.
bAD = severe adsorption.

45
Table 9. Peak elution volumes of extractable soil organic matter on
Porasil AX with selected solvents
Solvent
Sample H^O CH.^OH 2-Propanol t-Butanol Acetone THF DMF
DMF extract
Acetone-Soxhlet
1.
,82
(3.
30)
2-Propanol-Soxhlet --
1.
,75
(3.
,28)
Methanol-Soxhlet
1,
,80
(3.
.31)
NaOH extract

Acetone-Soxhlet
1.
.81
(3.
.31)
2-Propanol-Soxhlet
1.
,79
(3.
,32)
Methanol-Soxhlet
1.
,82
(3.
.27)
Humic acid

Fulvic acid 1.68
1.
,73
ml
3.35
3.42
AD
3.39
ADb
AD
AD


3.40
3.33
3.37
AD
3.35
(AD)
(AD)
AD
3.39
Parentheses ( ) indicate secondary peak.
'
AD = severe adsorption.

46
Table 10. Peak elution volumes of extractable soil organic matter on
CPG-250 with selected solvents
Solvent
Sample f^O CH^OH 2-Propanol t-Butanol Acetone THF DMF
ml
DMF extract




1.77
Acetone-Soxhlet
1.75
3.73
3.97
AD
3.87
1.82
ADa
AD
2-Propanol-Soxhlet
1.70
3.71


3.85
1.75
Methanol-Soxhlet
1.71




1.77
NaOH extract





1.79
Acetone-Soxhlet
1.76
3.84
4.12
AD
3.81
1.81
2-Propanol-Soxhlet
1.71
3.67


3.83
1.76
Methanol-Soxhlet
1.69




1.79
Humic acid





1.76
Fulvic acid
1.71 1.74
1.72
a
AD
severe adsorption

47
ELUTION VOLUME,ml
Fig. 6 Elution patterns of the 2-propanol-Soxhlet extract of
DMF-extractable material on Porasil A and Porasil AX
with selected solvents
Fig. 7 Elution patterns of fulvic acid on Porasil A and
Porasil AX with selected solvents

48
extracts (Table 8;Figs. 6 and 7) were eluted at when methanol or
DMF was used as the eluting solvent. A relatively larger quantity
was eluted at with DMF than with methanol. With each of the above
solvents, the organic solute was eluted prior to V^. On the contrary,
when acetone, t-butanol, or THF was used as the eluting solvent with
the Porasil A column, a portion of the organic solute was eluted past
V^, indicating an adsorptive interaction with the silica packing
material.
Deactivation of the Porasil surface (Porasil X) resulted in re
duced exclusion of organic solute from the gel matrix in water and
methanol compared to the activated material (Figs. 6 and 7; Table 9).
Adsorption was reduced on the Porasil AX compared with the Porasil A
packing material, although a small portion of the organic solute was
still eluted past with t-butanol, acetone, and THF on Porasil AX.
Elution volumes of organic standards on Porasil A, Porasil AX,
and CPG-250 are shown in Tables 11-13, respectively. On Porasil A,
several of the organic acid standards (1,2,4,5-tetracarboxybenzene and
1,3,5-tricarboxybenzene) were eluted at a solvent volume equivalent
to Vq when DMF, water, or methanol was used as the eluting solvent and
at a solvent volume slightly greater than when acetone, 2-propanol,
t-butanol or THF was used as the eluting solvent.
In water, methanol, and DMF, the more highly substituted aromatic
acids were in general eluted at a smaller solvent volume than the less
substituted acids. For example, elution volume increased according to

49
Table 11. Peak elution volumes of organic standards on Porasil A with
selected solvents
Solvent
Sample H^O CH^OH 2-Propanol t-Butanol Acetone THF DMF
ml
1,2,4,5-Tetracar-
boxybenzene
1.73
1.72
3.70
9.25

3.55
2.36
1,3,5-Tricarboxy-
benzene
1.76
1.94
AD3
3.68

3.54
3.60
3,5-Dihydroxyben-
zoic acid
2.08
4.87
3.33

3.47
3.43
Benzoic acid
2.37
5.29
3.51
3.59
3.58
3.44
Pyridine
5.20
AD
AD
4.45
3.87
3.58
Aniline
3.44
4.74
4.04
3.55
3.62
3.41
Methanol

3.64
3.94
3.91
3.42
3.94
Ethylene glycol
3.40
4.45
4.76
3.97
3.45
3.46
Acetone
3.36
3.97
4.52
3.70
3.46
a
AD =
severe adsorption

50
Table 12. Peak elution volumes of organic standards on Porasil AX
with selected solvents
Solvent
Sample
Ho0
ch3oh
2-Propanol t-
-Butanol
Acetone
THF
ml
1,2,4,5-Tetracarboxy-
benzene
1.83
2.58
10.25
12.00

3.48
1,2,5-Tricarboxy-
benzene
2.11
2.71
4.95
3.49

3.39
3,5-Dihydroxybenzoic
acid
2.53
2.85
3.59
3.28

3.46
Benzoic acid
2.39
2.97
3.77
3.40
3.49
3.34
Pyridine
11.06
3.42
3.67
ADa
3.84
3.60
Aniline
4.69
3.38
3.52
3.55
3.52
3.49
Methanol
3.48

3.44
3.41
3.65
3.71
Ethylene glycol
3.56
3.43
3.52
3.56
3.65
3.40
Acetone
3.58
3.38
3.44
3.46

3.40
aAD = severe adsorption

51
Table 13. Peak elution volumes of organic standards on CPG-250 with
selected solvents
Solvent
Sample 1^0 CH^OH 2-Propanol t-Butanol Acetone THF DMF
ml
1,2,4,5-Tetracarboxy-
benzene
1.71
1.74
4.12
10.03
--
3.91
2.44
1,3,5-Tricarboxyben-
zene
1.78
1.93
ADa
4.27

3.93
3.38
3,5-Dihydroxybenzoic
acid
2.18
5.03
4.06

3.91
3.86
Benzoic acid
2.26
5.36
4.99
3.98
3.92
3.88
Pyridine
4.89
AD
AD
5.02
4.34
4.17
Aniline
4.12
4.72
4.76
4.17
4.10
3.95
Methanol

3.98
4.27
4.26
3.87
4.31
Ethylene glycol
3.87
5.01
5.38
4.37
3.89
4.26
Acetone
3.82
4.34
4.87
4.06
3.92
a
AD
severe adsorption

52
the following solute order: 1,2,4,5-tetrac.arboxybenzene < 1,3,5-
tricarboxybenzene < 3,5-dihydroxybenzoic acid < p-hydroxybenzoic acid
< benzoic acid. Several compounds with basic properties (e.g, aniline,
pyridine) were adsorbed and eluted past V Also, several other com
pounds (e.g. glucose, ethylene glycol, and sucrose) were eluted past V,^.
Each of the compounds which showed strong evidence of adsorption on
Porasil A with water, methanol, or DMF as eluting solvent contained an
amino group or an aliphatic OH.
In THF, aromatic acids and simple alcohols each showed evidence
of adsorptive interaction with Porasil A. Elution on Porasil AX, the
deactivated analog of Porasil A, resulted in reduced adsorption.
Elution patterns of organic acid standards on Porasil A and
Porasil AX showed interesting similarities to the elution patterns of
extractable soil organic matter. Based on the molecular weights of
tetracarboxybenzene and tricarboxybenzene and the working molecular
weight ranges of the gels suggested by the manufacturer, one would
expect that the solute would elute at, or slightly before, V Inspec
tion of the patterns, however, shows that the acid standards were
completely excluded from the gel matrix and eluted at when water
was used as the eluting solvent. Deviations from the expected elution
behavior of a low molecular weight nonreactive solute may be attributed
to adsorption, electrostatic exclusion, or molecular association.
The exclusion of the acidic organic solute from Porasil A in
water, methanol, or DMF may be attributed to (i) the porous structure
of the gel, (ii) association of solute molecules, and/or (iii) electro
static exclusion from the porous matrix. The first explanation is

53
unlikely since the nonreactive solute, acetone, produced a symmetrical
peak at V with negligible skewing, indicative of free entrance into
the porous gel matrix. Association of the solute molecules in water,
methanol, and DMF would be questionable since aggregation of the mole
cular units should be greatest in the least polar and/or least basic
solvent. Comparison of the individual solvents shows that water, DMF,
and methanol have stronger basic character and are considerably more
polar than THF with dielectric constants of 76.2, 36.7, 32.6, and 7.58,
respectively. Also, deactivation of the porous silica resulted in
increased elution volumes, which should not have been the case if
skewing was entirely due to aggregation of the solute molecules.
With the first two explanations above eliminated as probable
major causes of exclusion of the low molecular weight acidic solute,
the third explanation, electrostatic exclusion, deserves careful con
sideration. The acidic functional groups of the organic solute would
be partially dissociated in water, methanol, or DMF due to the basic
character of each of these solvents. Water and methanol have basic
character which is attributed to the presence of the electron-donor
oxygen atom. The DMF molecule has two basic sites (Talhoun and Mort-
land, 1968), the electron-donor oxygen atom of the carbonyl group and
the nitrogen atom. The surface Si(OH) groups of the silica packing
material are weak acid sites. In water (Kirkland, 1971), methanol, or
DMF the surface sites may dissociate, due to the basic properties of
these solvents, resulting in a negatively charged silica surface.

54
Especially in water and methanol, the negative surface sites would be
stabilized as a result of the acidic properties of the solvent mole
cules. The exclusion of acidic organic solute from the porous matrix
of Porasil A may therefore be at least partially attributed to electro
static repulsion between the charged solute molecules and the charged
silica surface. In the absence of excess neutral salt, the silica
would have an expanded electrical double layer and the solute molecules
would exist with larger effective radii. Therefore, it is possible
that low molecular weight solutes may be completely excluded from the
porous gel matrix.
As mentioned previously, there was no evidence of adsorptive inter
action between the silica surface and the acidic solute in water,
methanol, and DMF; however, in 2-propanol, t-butanol, acetone, ethylace-
tate, and THF there was evidence of adsorption. In the former solvents,
the greater negative charge densities of the solutes and the silica
surface may have resulted in less adsorptive interaction between the
negatively charged species. In acetone, ethylacetate, and THF, however,
the organic solute would be much less dissociated as a consequence of
the very weak or negligible basic properties of these solvents. Also,
the silica surface would be less highly dissociated. Therefore, there is
a more favorable condition for direct H-bonding interactions between the
silica surface and the solute molecules.
Deactivation of the silica surface with polyethylene glycol would
block the reactive sites (Dark and Limpert, 1973) and result in reduced
negative charge density of the silica surface in water and methanol.

55
Therefore, electrostatic exclusion of negatively charged solute was
reduced on Porasil AX compared to Porasil A. When acetone, ethylace-
tate, or THF was used as the eluting solvent on Porasil AX, only a small
quantity of acidic solute was eluted past V This phenomenon indicates
a reduction in adsorptive interaction between the silica packing
material and the acidic solute on Porasil AX compared to Porasil A.
Adsorption and electrostatic exclusion were reduced on Porasil
AX, but were not completely eliminated. The evidence of adsorption and
electrostatic exclusion interactions between the solute and the Porasil
AX demonstrated that the packing material was not completely deactivated.
Effect of saturating cation. Fulvic acid in which the acidic
functional groups were saturated to pH 7.0 with Na+, K+, NCCH^)^"*", or
N(C,H0),+ were eluted at on Porasil A and CPG-250 when water was used
4 9 4 0
as the eluting solvent (Table 14). The cation-saturated samples were
also completely excluded from the gel matrix on deactivated Porasil AX.
Likewise, both the 1,2,4,5-tetracarboxybenzene and the 1,3,5-tricar-
boxybenzene in methanol and water were excluded from the Porasil A and
Porasil AX gel matrices.
The pronounced exclusion of Na-, K-, N(CH),-, and N(C.H) -
3 4 4 9 4
saturated fulvic acid and organic acid standards from the Porasil A
gel matrix may be attributed to electrostatic repulsion of the nega
tively charged solute molecule and the negatively charged sites on the
silica surface. Similar exclusion phenomena have been observed during
elution of cation-saturated fulvic acid with distilled water on
Sephadex (Swift and Posner, 1971).

56
Table 14. Peak elution volumes of cation-saturated fulvic acid on
Porasil A, Porasil AX, and CPG-250 with water as eluting
solvent
Column
Saturating
cation
Porasi1
A
Porasil
AX
CPG-
250
m1
H
1.62
1.68
1.79
Na
1.61
1.69
1.76
n(ch3)4
1.62
1.70
1.78
Ah
1.60 '
1.69
1.76

57
Electrostatic exclusion phenomena have been reported for elution
of acidic amino acids (Gelotte, 1960), aromatic acids (Demetriou et al.,
1968), inorganic ions (Neddermeyer and Rogers, 1968), and lignosulfonate
(Forss and Stenlund, 1973) on the Sephadex G-gels and was attributed to
electrostatic repulsion between fixed charges on both the gel and the
solute molecules.
Effect of excess electrolyte. Elution patterns of Na-saturated
fulvic acid in the presence of excess neutral salt on Porasil A and
Porasil AX (Figs. 8 and 9, respectively) indicated that electrolyte
resulted in reduction in the relative quantity of solute excluded from
the gel matrix. Presence of neutral salt also influenced elution pat
terns of low-molecular weight organic acids on Porasil A and Porasil AX
(Figs. 10 and 11, Tables 15 and 16) and resulted in reduced exclusion
of solute from the gel matrix. Similar phenomena have been observed
with Sephadex during the elution of soil organic matter extracts (Swift
and Posner, 1971; Posner, 1963), acidic amino acids (Gelotte, 1960),
aromatic acids (Demetriou et al., 1968), and lignosulfonates (Forss
and Stenlund, 1973).
At salt concentractions above 0.01 N, significant quantities of
fulvic acid were eluted past The excess electrolyte resulted in
suppression of negative charge and reduction in thickness of electrical
double layer of both the negatively charged gel surface and the organic
solute molecules. Also, excess sal t would decrease the effective size
of solute anions due to reduction in thickness of the electrical double
layer. Therefore, solute anions would more easily enter the porous

58
Fig. 8
Effect of excess neutral electrolyte on elution of
Na-saturated fulvic acid on Porasil A
Fig. 9 Effect of excess neutral electrolyte on elution of
Na-saturated fulvic acid on Porasil AX

59
Fig. 10 Effect of excess neutral electrolyte on elution of
Na-saturated 1,2,4,5-tetracarboxybenzene on Porasil A
Fig. 11 Effect of excess neutral electrolyte on elution of
Na-saturated 1,2,4,5-tetracarboxybenzene on Porasil AX

60
Table 15. Peak elution volumes of selected organic acid standards on
Porasil A with Na,,S0. solutions
2 4
Electrolyte
concentration
Sample
0.000 N
0.001 N
0.01 N
0.05 N
ml
1,2,4,5-Tetracarboxy-
benzene
1.74
1.68
1.86
2.60
1,3,5-Tricarboxyben-
zene
1.75
1.68
1.91
2.64
3,5-Dihydroxybenzoic
acid
1.82
1.97
2.32
2.71
Benzoic acid
1.96
2.41
2.51
3.14

61
Table 16. Peak elution volumes of selected organic acid standards on
Porasil AX with Na.SO. solutions
2 4
Electrolyte concentration
Sample
0.000 N 0.001 N 0.01 N 0.05 N
ml
1,2,4,5-Tetracarboxy-
benzene 1.87
1.3.5-Tricarboxyben-
zene 2.18
3.5-Dihydroxybenzoic
acid 2.21
Benzoic acid 2.41
2.31
3.03
3.21
2.43
3.05
3.17
2.80
3.19
3.28
3.09
3.40
3.48

62
gel matrix. As double-layer thickness decreased with resulting reduc
tion in electrostatic exclusion, adsorption increased due to direct H-
bonding interactions of oxygen-containing groups on the organic solute
and Si(OH) sites on the packing material. Adsorption effects were re
duced on Porasil X series of packing materials but were not completely
eliminated. This evidence indicated that the silica surface was not
completely deactivated and/or solute molecules were interacting directly
with the silica surface.
The solubility studies (Table 7) indicated that fulvic acid began
to precipitate at electrolyte concentrations above 0.01 N_, at pH values
of 4.0 to 8.0. Therefore, the partial elution of fulvic acid past
at the higher electrolyte concentrations may be caused by precipitation
of fulvic acid in the column. For this reason, electrolyte concentra
tions must be maintained at values low enough to preclude precipitation
of the solute.
Because of the nature of the silica surface, special precautions
must be observed. As mentioned previously, the silica surface acts as a
weak acid due to the presence of SiOH groups. In the presence of a
protic solvent with basic properties, such as water or methanol, these
acid sites will dissociate, leaving the silica surface with a net nega
tive charge. As the pH of the solvent medium is increased, the dis
sociation of surface sites and the negative charge density of the
silica surface is also increased. The negative charge density of
acidic solute molecules would also increase with increasing pH. There
fore, exclusion of negatively charged solute from the negatively
charged packing material would increase with increasing pH.

63
At pH values above 7.0, the silica surface may be destroyed by
solubilization and formation of silicate. Therefore, alkaline condi
tions must be avoided. Under alkaline conditions, the chemically
adsorbed deactivating agent is also readily stripped from the surface
of the Porasil X packing material. The manufacturer recommends that
use of several organic solvents, especially DMF, should be avoided
with deactivated Porasil. Such a solvent may readily strip the de
activating agent from the silica surface.
Polystyrene-divinylbenzene (DVB)
Operating conditions. As with the porous glass packing materials,
column efficiencies of Poragel and y-Styragel packing materials were
greatly influenced by solvent flow rates. With 0.054-cm OD columns and
THF as the eluting solvent, minimum peak broadening and maximum column
efficiencies were obtained at flow rates of approximately 0.8 and 3.0
O
ml per min. for the 100 A Poragel and y-Styragel columns, respectively
(Fig. 12). At higher flow rates, peak broadening was increased and
column efficiencies were decreased. At lower flow rates, column effi
ciencies were also appreciably lowered. This later effect was much
more evident with the polystyrene-DVB gels than with the porous glasses.
At low flow rates, diffusion of solute molecules apparently resulted
in decreased column efficiencies. With the y-Styragel columns, the
high column efficiency at the high flow rate was due to the small
particle size and uniform pore size of packing which permitted rapid

THEORETICAL PLATE COUNT, N
64
FLOW RATE, ml/min
Fig. 12 Effect of flow rate on column efficiency, N, of 100 A
O
Poragel and 100 A u-Styragel preparative columns with
THF as the eluting solvent

65
equilibrium of solute molecules between internal pore space and inter
stitial pore space.
Based on these studies, solvent flow rates of 0.8 and 3.0 ml per
min. in 0.954-cm OD columns were used for Poragel and y-Styragel
columns, respectively, in all subsequent studies.
The effect of sample size on column efficiencies of 0.954-cm
diameter columns of Poragel and y-Styragel are shown in Fig. 13. In
general, the maximum sample volumes were 250 yl and 25 yl for the
Poragel and y-Styragel columns, respectively. Larger samples resulted
in increased peak broadening and reduced apparent column efficiencies.
Column parameters. Column parameters (Table 1), and V^, of the
packed columns were determined by elution of acetone or 2,600,000 molecu
lar weight polystyrene, respectively. Column efficiencies, N, were
O o
approximately 800 and 5,000 for the 100 A Poragel and 100 A y-Styragel,
respectively. Molecular weight calibration curves, obtained by elution
of polystyrene standards with THF gave working molecular weight ranges
of 500 to 20,000 and 100 to 3,000, respectively, for the above gels
(Fig. 14).
Several of the highly substituted organic acid standards deviated
from the polystyrene calibration curve; therefore, the polystyrene
standards are not suitable for accurate determination of molecular
weights of low molecular weight organic acids. Based on the above ob
servation, it is doubtful that polystyrene standards would be suitable
standards for molecular weight determinations of soil humic compounds.

66
SAMPLE SIZE, ml
Fig. 13 Effect of sample size on column efficiency, N, of 100 A
Poragel and 100 A y-Styragel preparative columns with
THF as the eluting solvent

67
lili I
10 20 30 40 50
ELUTION VOLUME, ml
Fig. 14 Molecular weightocalibration curve of U-St^tagel (2 f x
0.954 cm 0D 100 A + 2 f x 0.954 cm OD 500 A) obtained
by elution of polystyrene standards with THF

68
The selection of suitable molecular weight standards for soil humic
compounds remains a problem.
Effect of solvent on elution of standard compounds. Comparison
of the elution volumes of acetone and 2,600,000 molecular weight poly-
O
styrene in THF and DMF in 100 A Poragel suggests that the Poragel is
poorly swelled in DMF (Table 1). This conclusion is based on the as
sumption that acetone is a nonreactive solute and readily enters the
Poragel gel matrix and that the high molecular weight polystyrene
standard is completely excluded from the gel matrix. In DMF, the elu
tion volumes of acetone and the 2,600,000 molecular weight polystyrene
are separated by 5.3 ml compared to THF in which the elution volumes
of low and high molecular weight standards are separated by 10.5 ml.
Therefore, in DMF the internal pore volume of the packing material is
33% of the total pore volume compared to the THF in which the internal
pore volume of the packing material is approximately 49% of the total
pore volume. The greater swelling of the polystyrene-DVB gel in THF
compared to DMF may be attributed to the less polar and greater hydro-
phobic character of the former solvent which would make it more com
patible with the hydrophobic gel.
As a consequence of the different swelling properties of the gel
in the different solvents, it is essential that column parameters and
molecular weight distribution patterns be determined for the same
solvent which is to be used as the eluting solvent. Also, it is
essential that the column be packed in the same solvent which is to be
used as the eluting solvent. Changing of the eluting solvent in the

69
column may produce voids and result in increased peak widths and re
duced column efficiencies.
Peak elution volumes of organic standards eluted with THF and
O
DMF on 100 A Poragel are summarized in Table 17. The initial observa
tion is that several of the solute species are behaving differently
in the two eluting solvents.
In looking more closely at solute behavior in THF, it can be
observed that each of the low-molecular weight solutes were eluted
in the vicinity of or slightly after V Apparently, each of these
solutes readily entered the pores of the Poragel gel matrix. Only
benzene and anthracene were eluted noticeably past V Edwards and
Ng (1968) also observed the adsorption of some aromatic compounds by
polystyrene-DVB when eluted with THF. It is probably the aromatic
character of the polystyrene-DVB gel which resulted in adsorption of
benzene and anthracene. The aromatic acids were not noticeably eluted
past V^ and were apparently not strongly adsorbed. The elution of these
compounds near gave strong indication that they readily entered the
polystyrene-DVB gel matrix.
O
Elution of standard compounds with THF on 100 A u~Styragel produced
very similar results. Only benzene, toluene, and anthracene were eluted
past the assumed value of V due to an apparent adsorptive interaction
with the gel matrix. Other standard compounds tested, e.g. simple
alcohols, aromatic acids, aromatic bases, and phenolic acids, were
eluted in the vicinity of V^ and apparently readily entered the
polystyrene-DVB gel matrix.

70
Table 17. Peak elution volumes of low molecular weight standards
eluted on 100 A Poragel with THF and DMF and on 100
y-Styragel with THF as the eluting solvent
Column packing
100 A Poragel 100 A p-Styragel
Sample THF DMF THF
ml
1,2,4,5-Tetracarboxy-
benzene
20.81
12.93
10.27
1,3,5-Tricarboxyben-
zene
21.04
12.97
10.34
3,5-Dihydroxybenzoic
acid
21.19
13.04
10.36
Benzoic acid
21.23
13.28
10.49
Pyridine
22.13
18.12
10.56
Aniline
21.28
17.21
10.52
Methanol
21.24
16.18
10.53
Ethylene glycol
21.13
16.34
10.49
Acetone
21.26
16.47
10.53
Benzene
21.84
17.63
10.89
Anthracene
22.43
18.02
11.42

71
When DMF was used as the eluting solvent, several of the aromatic
acids were eluted prior to the assumed value of V The more highly
substituted aromatic acids (e.g. 1,2,4,5-tetracarboxybenzene and 1,3,5-
tricarboxybenzene) were eluted near the assumed value of and were
apparently completely excluded from the gel matrix. Several of the
aromatic acids and phenolic acids showed two elution peaks which cor
responded closely to the assumed values of and V Each of the low
molecular weight compounds which were eluted noticeably before V con
tained an acidic side group, COOH and/or phenolic OH.
Several compounds, e.g. benzene, toluene, and anthracene, were
eluted considerably past the assumed value of V^. Each of these com
pounds was hydrophobic in nature and was structurally similar to
monomers of the gel polymer. Benzene, toluene, and anthracene were
more strongly adsorbed with DMF than with THF as the eluting solvent.
This effect was probably due to the more polar character of the DMF.
Neutral solutes (e.g. simple alcohols) and compounds with basic
properties (e.g. pyridine, aniline) were eluted at or slightly after the
assumed V and apparently readily entered the polystyrene-DVB gel
matrix.
Two interesting points from the above observations are that (i)
hydrophobic solutes were more strongly adsorbed to the polystyrene-DVB
gel matrix when eluted with DMF than with THF, and (ii) acidic solutes
were noticeably excluded from the gel matrix when DMF was used as the
eluting solvent, but not when THF was used as the solvent. The first
point may be explained in terms of relative hydrophobic character

72
of the two solvents, as discussed previously. The second point is
elaborated upon below.
The acidic functional groups of an acidic organic solute would be
partially dissociated in DMF due to basic character of this solvent;
therefore, the solute molecules are likely to be highly dispersed.
The negatively charged solute molecules would have larger effective
radii than the neutral species; however, this phenomena should not
entirely account for the exclusion phenomena since the exclusion limit
of the gel, based on polystyrene standards, is approximately 50,000
molecular weight. In the porous silica and Sephadex gels, the exclusion
phenomena can be explained in terms of electrostatic repulsion from
negative charge sites in the gel matrix. In porous silica, the negative
charge results from dissociation of Si(OH) sites at silica surface.
In Sephadex the negative charge has been attributed to COOH impurities
in the gel matrix. On the other hand, the polystyrene-DVB gel should
exist as a neutral species. Therefore, we must search for an alternate
explanation to the exclusion phenomena. A possible explanation is the
ion inclusion effect suggested by Forss and Stenlund (1975) in studies
of lignosulfonate. They attributed this effect to the interaction of
charged sites on the ions entering the pores with other charged ions
outside of the pores. The net effect is electrostatic repulsion. Such
an effect would not entirely account for the apparent total exclusion
of low molecular weight solutes observed. Further work will be required
to determine the nature of this phenomena.

73
Effect of solvent: on elution of soil humic compounds. Elution
o
patterns of soil humic fractions on 100 A p-Styragel with THF as the
eluting solvent are shown in Figs. 15 and 16. In THF, all humic frac
tions were eluted between V and and apparently readily entered the
porous gel matrix. As mentioned previously, the polystyrene standards
are not suitable for accurate molecular weight determinations of soil
humic materials. These standards, however, do provide a guide for
measurement which is probably no less suitable than others commonly
used, such as proteins or polysaccharides. Molecular weight estimates
based on the polystyrene standards are summarized in Table 18. In all
cases the Soxhlet fractions were estimated to have peak molecular
weights less than 800. These fractions, however, represent only a
minor portion of the total NaOH- or DMF-extractable materials, 28 and
32%, respectively, and are likely to contain materials with lower peak
molecular weights than those of the NaOH- or DMF-extractable materials.
These later materials are not sufficiently soluble in THF to obtain a
molecular weight fractionation.
In DMF, the major portions of all humic fractions were eluted at
volumes corresponding to the assumed values of and were apparently
largely excluded from the gel matrix. In all cases, a minor portion
of the material was eluted at V Reinjection of fractions collected
at Vq produced patterns similar to the original patterns with major
peaks corresponding closely to and minor peaks at the assumed value
of Vr Reinjection of the sample eluted at during the original
fractionation also produced a fractionation pattern similar to the

RECORDER RESPONSE
74
Fig. 15 Elution of Soxhlet extracts of NaOH-extractable soil
organic matter on 100 A y-Styragel with THF

RECORDER RESPONSE
75
Fig. 16 Elution of Soxhlet extracts of DMF-extraetable soil
organic matter on 100 A y-Styragel with THF

76
original pattern with a major peak eluted at and a minor peak at
V This technique produced strong evidence that the initial frac
tionation pattern was not the result of a separation according to
molecular size, but instead was an artifact resulting from a complex
gel-solvent-solute interaction. The elution patterns of the soil
humic acid fractions were indeed similar to the patterns obtained from
elution of the low molecular weight aromatic acids.

77
Table 18. Molecular weight estimates of soil humic fractions based
on elution of polystyrene standards on y-Styragel with
THF as the eluting solvent
Sample
Estimated
molecular weight
NaOH extract
DMF extract
Acetone-Soxhlet
560
660
2-Propanol-Soxhlet
720
740
Methanol-Soxhlet
760
740

CONCLUSIONS
Extraction and Fractionation
Several of the dipolar aprotic solvents, i.e. DMF and DMSO, were
shown to be excellent solvents for the soil humic fraction. Functional
group, elemental, and IR analysis indicated that the DMF-extractable
soil organic matter was chemically similar to the material extracted
by 0.5 N NaOH. The dipolar aprotic solvent may,therefore, serve as an
excellent complementary solvent to NaOH for chemical studies of extractable
soil organic matter.
The Soxhlet fractionation scheme was used successfully to fraction
ate the extractable soil organic matter into samples with distinct charac
teristics. The Soxhlet extraction scheme of hexane, benzene, ethylacetate,
acetone, 2-propanol, and methanol was utilized to obtain materials with
progressively greater hydrophilic character, lower C and H contents,
greater N, S, and 0 contents, greater COOH content, and greater total
acidity. The Soxhlet solvents were able to extract 29.6 and 35.0% of
the NaOH- and DMF-extractable materials, respectively. Even though
these fractions represent a minor portion of the total extractable
material, they are likely to contain materials of simpler average composi
tion and lower peak molecular weights than the NaOH- and DMF-extractable
78

79
material and are important since they are likely to contain monomers
which compose the polymeric structure of the humic complex. Therefore,
investigations of these fractions provide information which will aid
in understanding properties of the total humic complex.
Solubility Properties
Humic acid and NaOH- and DMF-extractable soil organic matter were
100% soluble at the 0.1% concentration in both DMF and DMSO. Fulvic
acid was completely soluble at the 0.1% concentration in both water
and methanol. Solubility of extractable soil organic matter was de
creased in the dipolar aprotic solvents, e.g. DMF, THF, and acetone, and
increased in the protic solvents as acidic hydrogen was placed with
Na+, K+, or N(CH ),\
The solubility of fulvic- acid in aqueous systems was influenced
by pH and concentration of excess neutral electrolyte. Fulvic acid was
completely soluble at the 0.1% concentration in aqueous systems with
concentrations of excess neutral electrolyte up to 0.1 N at pH values
less than 4.0 and greater than 8.0; however, at pH values from 4.0 to
8.0, fulvic acid partially precipitated with concentrations of
Na^SO^ or t^SO^ greater than 0.05 N_.
The number of useful chromatographic fractionation schemes is
greatly limited by the solubility characteristics of the solute; there
fore, an understanding of these characteristics is an essential
prerequisite to the rapid screening of possible fractionation schemes.

80
Liquid Chromatography
None of the chromatographic gels investigated was completely
inert. Each gel apparently interacted with the soil humic material;
therefore, the elution patterns were not entirely attributable to a
molecular seiving phenomenon but to a combination of molecular seiving,
adsorption, and ionic exclusion phenomena.
Solvent and electrolyte effects were especially evident in studies
of Porasil and CPG packing materials. When H-, Na-, K-, or N(CH^)^-
saturated low molecular weight organic acid standards or fulvic acid
were eluted with H90, the solute molecules were partially or totally
excluded from the porous gel matrix. As electrolyte concentration was
increased, the acidic solute molecules more readily entered the porous
matrix; however, at electrolyte concentrations above 0.01 N, significant
quantities of fulvic acid were' adsorbed and eluted past V These
phenomena were attributed to decreased thickness of the electrical
double layer and/or suppression of charge of the negatively charged
solute molecule and the negatively charged silicate surface. Adsorption
of fulvic acid at the higher electrolyte concentrations was attributed
to increased interaction between active sites at the silica surface
and oxygen- and nitrogen-containing functional groups of the organic
solute and also to the possible precipitation of fulvic acid caused by
the high counter ion concentration at the negatively charged silica
surface. Low molecular weight organic solutes with significant basic
properties, i.e. pyridine, were strongly adsorbed to Porasil and CPG

81
gels in aqueous systems. Since the fulvic acid sample contained
nitrogen, indicating the probable presence of basic sites, interac
tions of these sites in the negatively charged solute molecule with the
negatively charged silicate surface would be greater in the presence
of excess neutral electrolyte.
When protic solvents, i.e. ^0, methanol, and 2-propanol, or
dipolar aprotic solvents with significant basic properties, i.e. DMF,
were used to elute low molecular weight acidic solutes or extractable
soil organic matter, electrostatic exclusion phenomena predominated.
In dipolar aprotic solvents without significant basic character, i.e.
acetone and THF, adsorption phenomena predominated. With all solvent
and electrolyte systems examined, it was not possible to completely
eliminate both electrostatic exclusion and adsorption phenomena. De
activation of the Porasil surface, Porasil X, did not completely eliminate
adsorption and exclusion interactions between the acidic organic solute
and the silica surface.
The polystyrene-DVB gels were compatible with a more limited range
of solvents than the silica gels. Also, due to the swelling properties
of the gel it was essential to pack the column with the same solvent
which was to be used as the eluting solvent. With the polystyrene-DVB
gels, elution patterns were dependent on the eluting solvent. In DMF,
low and high molecular weight acidic organic solutes were totally or
partially excluded from the gel matrix. In THF, low molecular weight
acid solutes apparently readily entered the porous gel matrix and were

82
eluted in the vicinity of V Of the compounds tested, only several
hydrophobic aromatic compounds were strongly adsorbed. It appears
that THF is a suitable solvent for elution of acidic solutes; however,
only the benzene-, ethylacetate-, acetone-, and 2-propanol-Soxhlet
fractions were suitably soluble in THF. Methylated fractions of the
humic acid and fulvic acid fractions would also be soluble in THF.
Molecular weights of acetone-, 2-propanol-, and methanol-Soxhlet
fractions were estimated to be 500 to 800, based on elution patterns
of soil humic fractions with those of polystyrene standards in THF.
Further research x^ill be needed to corroborate, by other methods, the
molecular weight estimates obtained with gel permeation chromatography.
One such method would be vapor pressure osmometry.
Each of the two general groups of gels evaluated in this study
showed evidence of adsorptive and/or electrostatic interaction with
extractable soil organic matter. Mode and extent of interaction were
highly dependent on the solvent medium. Because of these possible
interactions, special care must be observed in the interpretation of
gel permeation chromatography patterns.
High pressure liquid chromatography is a valuable new technique
because of the use of high efficiency columns and the short time re
quired to obtain elution patterns. In addition to the application of
gel permeation chromatography, there is the unexplored potential
application of liquid-liquid partition chromatography and liquid-solid
adsorption chromatography to fractionation of extractable soil organic
matter. High pressure liquid chromatography will also provide a valu
able tool for studies of soil organic-mineral-ionic interactions.

LITERATURE CITED
Bergmann, J.G., L.I. Duffy, and R.B. Stevenson. 1971. Solvent effects
in gel permeation chromatography. Anal. Chem. 43:131-133.
Bly, D.D. 1970. Gel permeation chromatography. Science 168:527-533.
Bombaugh, K.J. 1971. The practice of gel permeation chromatography.
In J.J. Kirkland (ed.), Modern Practice of Liquid Chromatography.
Wiley-Interscience, New York.
Bombaugh, K.J., W.A. Dark, and J.N. Little. 1969. Fractionation of
alcohol on deactivated porous silica beads by gel permeation
chromatography. Anal. Chem. 41:1337-1339.
Bremner, J.M. 1950. Some observations on the oxidation of soil
organic matter in the presence of alkali. J. Soil Sci. 1:
198-204.
Bremner, J.M. 1956. Some soil organic matter problems. Soil and
Fertilizers 19:115-123.
Bremner, J.M. and J. Lees. 1949. Studies on soil organic matter.
II. The extraction of organic matter from soil by neutral
reagents. J. Agr. Sci. 39:274-279.
Brook, A.J.W. and S. Housley. 1969. The interaction of phenols with
Sephadex gels. J. Chromatogr. 41:200-204.
Brook, A.J.W. and K.C. Munday. 1970. The interaction of phenols,
aniline, and benzoic acids with Sephadex gels. J. Chromatogr.
47:1-8.
Brooks, J.D., R.A. Durie, and S. Sternhell. 1958. Chemistry of
brown coals. III. Pyrolytic reactions. Aust. J. Appl. Sci.
9:303-320.
Burges, N.A., H.M. Hurst, and B. Walkden. 1964.
stituents of humic acid and their relation
the plant cover. Geochim. Cosmochim. Acta
The phenolic con-
to the lignin of
28:1547-1554.
83

84
Cheshire, M.V., P.A. Cranwel1, C.P. Falshow. A.J. Floyd, and R.D.
Haworth. 1967. Humic acid: 2. Structure of humic acid.
Tetrahedron 23:1669-1682.
Choudhri, M.B., and F.J. Stevenson. 1957. Chemical and physico
chemical properties of soil humic colloids: III. Extraction
of organic matter from soils. Soil Sci. Soc. Amer. Proc. 21:
508-513.
Cogswell, T.E., J.F. McKay, and D.R. Latham. 1971. Gel chromato
graphic separation of petroleum acids. Anal. Chem. 43:645-648.
Cooper, A.R. and E.M. Barrall. 1973. Gel permeation chromatography:
physical characterization and chromatographic properties of
Porasil. J. Appl. Polym. Sci. 17:1253-1268.
Cooper, A.R. and A.R. Bruzzone. 1973. Characterization and proper
ties of macromolecules. III. Gel permeation chromatography:
the effect of temperature on the elution volume and the effi
ciency of the separation process. J. Polym. Sci. A-2, 11:
1423-1434.
Cooper, A.R., A.R. Bruzzone, J.H. Cain, and E.M. Barrall. 1971.
Characterization and chromatographic properties of Corning porous
glasses. J. Appl. Polym. Sci. 15:571.
Cooper, A.R. and J.F. Johnson. 1969. Gel permeation chromatography:
the effect of treatment with hexamethyl disilazane on porous
glass packings. J. Appl. Polym. Sci. 13:1487-1492.
Cooper, A.R., J.F. Johnson, and R.S. Porter. 1973. Gel-permeation
chromatography: current status. American Laboratory. May
1973:12-24.
Dark, W.A. and R.J. Limpert. 1973. An evaluation of available pack
ings for GPC. J. Chromatogr. Sci. 11:114-120.
Demetriou, J.A. et al. 1968. Gel filtration chromatography of
fluorescent phenolic and heterocyclic compounds. J. Chromatogr.
34:342-350.
Determann, H. and I. Walter. 1968. Source of aromatic affinity to
Sephadex dextran gels. Nature (Lond.) 219:604-605.
Edwards, G.D. and Q.Y. Ng. 1968. Elution behavior of model compounds
in gel permeation chromatography. J. Polym. Sci. Part C 21:
105-117.
Flaig, W., H. Beutelspacher, and E. Rietz. 1975. Chemical composi
tion and physical properties of humic substances. In J.E.
Gieseking (ed.), Soil components. Springer-Verlag, New York.

85
Forss, K.G. and B.G. Stenlund. 1975. The influence of charged groups
in gel permeation chromatography on polyelectrolvtes. J. Polym.
Sci. 42:951-963.
Freeman, D.H. 1973. The gels for liquid chromatography. J.
Chromatogr. Sci. 11:175-180.
Fritz, J.S., S.S. Yamamura, and E.C. Bradford. 1959. Determination
of carbonyl compounds. Anal. Chem. 31:260-263.
Gaylor, V.F., H.L. James, and H.H. Weetall. 1976. Gel and affinity
chromatography. Anal. Chem. 48:44R-51R.
Gelotte, B. 1960. Studies on gel filtration sorption of the bed
material Sephadex. J. Chromatogr. 3:330-342.
Giddings, J.C. 1965. Dynamics of chromatography: principles and
theory. Marcel Dekker, New York.
Halstead, R.L., G. Anderson, and N.M. Scott. 1966. Extraction of
organic matter from soils by means of ultrasonic dispersion
in aqueous acetylacetone. Nature 211:1430-1431.
Hansen, E.H. and M. Schnitzer. 1967. The alkaline permanganate
oxidation of Danish illuvial organic matter. Soil Sci. Soc.
Amer. Proc. 30:745-748.
Hansen, E.H. and M. Schnitzer. 1969. Zinc-dust distillation of soil
humic compounds. Fuel 48:41-46.
Hawk, G.L., J.A. Cameron, and L.B. Dufault. 1972. Chromatography of
biological materials on polyethylene glycol treated controlled
pore glass. Prep. Biochem. 2:193-203.
Haworth, R.D. 1971. The chemical nature of humic acid. Soil Sci.
111:71-79.
Hiatt, C.W. et al. 1971. Treatment of controlled-pore glass with
poly (ethylene oxide) to prevent adsorption of rabies virus.
J. Chromatogr. 56:362-364.
Hurst, H.M. and N.A. Burges. 1967. Lignin and humic acids. In
A.D. McLaren and G.H. Peterson (eds.), Soil biochemistry.
Marcel Dekker, New York.
Karger, B.L. 1971. The relationship of theory to practice in high
speed liquid chromatography. In J.J. Kirkland (ed.), Modern
practice of liquid chromatography. Wiley-lnterscience,
New York.

86
Kessler, T., R.A. Friedel, and A.G. Sharkey. 1970. Ultrasonic solva
tion of coal in quinoline and other solvents. Fuel 49:222-223.
Khan, S.U. 1971. Distribution and characteristics of organic matter
extracted from black solnetzic and black chenozemic soils of
Alberta: the humic acid fraction. Soil Sci. 112:401-409.
King, E.J. 1973. Acid-base behavior. In A.K. Covington and T.
Dickinson (eds.), Physical chemistry of organic solvent systems.
Plenum Press, London.
Kirkland, J.J. 1974. Modern practice of liquid chromatography.
Wiley-Interscience, New York.
Kononova, M.M. 1966. Soil organic matter. 2nd English ed. Pergamon
Press, New York.
Kumada, K. and A. Suzuki. 1961. Isolation of anthraquinones from
humus. Nature 191:415-416.
Laub, R.J. 1974. Packings for HPLC. Research and development.
July 1974:24-28.
LePage, M., R. Beau, and A.J. DeVries. 1968. Evaluation of analyti
cal gel chromatography columns packed with porous silica beads.
J. Polym. Sci., Part C 2:119-130.
Loeppert, R.H., Jr., and B.G. Volk. 1974. Nature of organic matter
extracted from Terra Ceia muck with selected solvents. Soil
and Crop Sci. Soc. Fla. 33:160-164.
Loeppert, R.H., Jr., and B.G. Volk. 1976. Application of high pres
sure liquid chromatography to studies of extractable soil
organic matter-porous silica packings. International Symposium
on Soil Organic Matter Studies, Braunschweig, Germany. (In
Press).
Loeppert, R.H., Jr., L.W. Zelazny, and B.G. Volk. 1976. Acidic
properties of kaolinite in water and acetonitile. Soil Sci.
Soc. Amer. J. (In press).
Mclver, R.D. 1962. Ultrasonicsa rapid method for removing soluble
organic matter from sediments. Geochim. Cosmochim. Acta 26:
343-345.
Mortenson, J.L. 1965. Partial extraction of organic matter. In
C.A. Black (ed.), Methods of soil analysis, Agromony Monograph
No. 9, Vol. 2. American Society of Agronomy. Madison,
Wisconsin.

87
Neddermeyer, P.A. and L.B. Rogers. 1968. Gel filtration behavior of
inorganic salts. Anal. Chem. 40:755-762.
Ortiz de Serra, M.I. and M. Schnitzer. 1973. The chemistry of humic
and fulvic acids extracted from Argentine soils. II. Perman
ganate oxidation of methylated humic and fulvic acids. Soil
Biol. Biochem. 5:287-296.
Parsons, J.W. and J. Tinsley. 1960. Extraction of organic matter
with anhydrous formic acid. Soil Sci. Soc. Amer. Proc. 24:198-
201.
Porter, L.K. 1967. Factors affecting the solubility and possible
fractionation of organic colloids extracted from soil and leo-
nardite with an acetone-H O-HCl solvent. J. Agr. Food Chem.
15:807-811.
Posner, A.M. 1966. The humic acid extracted by various reagents
from a soil. I. Yield, inorganic components, and titration
curves. J. Soil Sci. 17:65-68.
Posner, A.M. 1963. Importance of electrolyte in the determination
of molecular weights by Sephadex gel filtration with special
reference to humic acid. Nature 198:1161-1163.
Salfeld, J.C. 1964. Fractionation of a humus preparation with aqueous
organic solvents. Landbauferschung Volkenrode 14:131-136.
Schnitzer, M. and V.C. Gupta. 1965. Determination of acidity in soil
organic matter. Soil Sci. Soc. Amer. Proc. 29:274-277.
Schnitzer, M. and S.U. Khan. 1972. Humic substances in the environ
ment. Marcel Dekker, Inc., New York.
Schnitzer, M., D.A. Shearer, and J.R. Wright. 1959. A study in the
infrared of high molecular weight organic matter extracted by
various reagents from a podzolic B horizon. Soil Sci. 87:
252-257.
Schnitzer, M. and S.I.M. Skinner. 1968. Gel filtration of fulvic
acid, a soil humic compound, jin Isotopes and radiation in soil
organic matter studies. International Atomic Energy Agency,
Vienna.
Snow, J.T., R.H. Loeppert, Jr., and B.G. Volk. 1973. Yields and
functional group characteristics of NaOH-extracted humic acid
from Pahokee muck. Soil and Crop Sci. Soc. Fla. Proc. 33:
165-167.

88
Sommers, T.C. 1966. Wine tanninsIsolation of condensed flavanoid
pigments by gel filtration. Nature (Lond.) 209:368-370.
Spatorico, A.L. 1975. Exclusion chromatography using controlled
porosity glass. I. Comparison with styrene gels. J. Appl.
Polym. Sci. 19:1601-1610.
Spatorico, A.L. and G.L. Beyer. 1975. Exclusion chromatography using
porous glass. II. Application to hydrophilic polymers. J.
Appl. Polym. Sci. 19:2933-2945.
Steelink, C. and G. Tollin. 1967. Free radicals in soil. Jen. A.D.
McLaren and G.H. Peterson (eds.), Soil biochemistry. Marcel
Dekker, New York.
Stevenson, F.J. 1965. Gross chemical fractionation of organic matter.
In C.A. Black (ed.), Methods of soil analysis, Agronomy Monograph
No. 9, Vol. 2. American Society of Agronomy. Madison, Wisconsin
Swift, R.S. and A.M. Posner. 1971. Gel chromatography of humic acid.
J. Soil Sci. 22:237-249.
Talhoun, S.A. and M.M. Mortland. 1966. Complexes of montmorillonite
with primary, secondary, and tertiary amides. I. Protonation
of amides on the surface of montmorillonite. Soil Sci. 102:
248-254.
Tinsley, J. and A. Salam. 1961. Extraction of soil organic matter
with aqueous solvents. Soils and Fert. 24:81-84.
Unger, K.K., R. Kern, M. Ninou, and K.F. Krebs. 1974. GPC with a
new type of silica packing material. J. Chromatogr. 99:435-443.
Volk, B.G. and M. Schnitzer. 1973. Chemical and spectroscopic methods
for assessing subsidence in Florida Histosols. Soil Sci. Soc.
Amer. Proc. 37:886-888.
Zweig, G. and J. Sherma. 1974. Chromatography. Anal. Chem. 46:73R-
94R.

BIOGRAPHICAL SKETCH
Richard Henry Loeppert, Jr., was born September 26, 1944, in
Raleigh, North Carolina. He graduated from Needham B. Broughtan High
School in Raleigh, North Carolina, in June, 1962. In August, 1966,
he received his Bachelor of Science degree with a major in soil science
from North Carolina State University, Raleigh, North Carolina.
Following graduation, he was employed as Assistant County Agent
with the Florida Agriculture Extension Service in Jackson County,
Florida. He began his graduate studies at the University of Florida
in 1970 and received his Master of Science degree in soil science in
August, 1973. He is currently a candidate for the Ph.D. degree in the
Department of Soil Science, University of Florida.
He is a member of the American Society of Agronomy, the Soil
Science Society of America, the International Society of Soil Science,
and the Clay Minerals Society.
89

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Professor
X J
Y
Fiskell, Chairman
of Soil Science
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Associate Professor
Soil Science
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Associate Professor of Soil Science
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philsophy.
/N. Gammon
Professor of Soil Science

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
This dissertation was submitted to the Graduate Faculty of the College
of Agriculture and to the Graduate Council, and was accepted as par
tial fulfillment of the requirements for the degree of Doctor of
Philosophy.
December, 1976
Dean,/ pire
Dean, Graduate School



6
acids except benzoic acid. Ortiz de Serra and Schnitzer (1973) iso
lated and identified a number of phenolic acids. Hansen and Schnitzer
(1967) used nitric acid oxidation and identified a seriesnitro-
phenols and aliphatic dicarboxylic, phenolic, and benzenecarboxylic
acids. *
Based on degradative and nondegradative studies, Schnitzer has
proposed a structure for fulvic acid consisting of phenolic and benzene
carboxylic acids joined by hydrogen bonds to form a polymeric matrix of
considerable stability (Schnitzer and Khan, 1972). Numerous additional
structures for humic and fulvic acids have been suggested (Burges,
Hurst, and Walkden, 1964; Flaig, Beutelspacher, and Reitz, 1975;
Haworth, 1971). The complexity and diversity of these structures
demonstrate the probable complexity of the total soil humic complex.
Extraction of Soil Organic Matter
Procedures for extraction of soil organic matter were reviewed
by Mortenson (1965) and Stevenson (1965) Dilute aqueous NaOH is
the most commonly used extractant of soil organic matter. Sodium
hydroxide produces high yields of extractable organic matter; however,
its use has been severely criticized due to chemical alterations
which may occur in alkaline conditions (Bremner and Lees, 1949;
Bremner, 1956; Choudhri and Stevenson, 1957). Bremner (1950) observed
that 09 was adsorbed from the atmosphere by alkaline soil suspensions.
Tinsley and Salam (1961) suggested that condensation reactions between
amino compounds and aldehydes or phenolic compounds may result in
formation of humin-type compounds during NaOH extraction.


47
ELUTION VOLUME,ml
Fig. 6 Elution patterns of the 2-propanol-Soxhlet extract of
DMF-extractable material on Porasil A and Porasil AX
with selected solvents
Fig. 7 Elution patterns of fulvic acid on Porasil A and
Porasil AX with selected solvents


2
development of the high-pressure liquid chromatograph which, in turn,
has made possible the use of small-diameter packing materials and
high efficiency columns. For example, with the new p-packing ma
terials, column efficiencies as high as 5000 theoretical plates per
foot are commonly attained. The rapid rise in the application of
high-pressure liquid chromatography is readily apparent following
a quick glance through any recent issue of Analytical Chemistry.
Attempts at molecular-size fractionation of soil humic materials
have been complicated by the fact that no gel material, including
Sephadex, is completely inert. Therefore, separations may be adversely
affected by gel-solute and gel-solvent interactions which would lead
to misleading results. Also, the humic molecule is a strongly reactive
solute which has a strong tendency to interact with other solute mole
cules and solvent molecules. For this reason, the fractionation of soil
humic compounds is affected by solute-solute and solute-solvent
interactions. Each of the above-mentioned interactions is strongly
influenced by packing material, solvent, saturating cation, concen
tration of excess electrolyte, and pH.
The objectives of this work were to investigate the use of high-
pressure liquid chromatography and a series of new packing materials
for the size fractionation of soil organic matter extracts and to
investigate the effect of packing material, solvent, saturating-
cation, concentration of excess electrolyte, and pH on solute-solute,
gel-solute, solvent-solute, and gel-solvent interactions which would
influence size separations. Since several of the new packing materials


40
uniform pore structure of the controlled pore glass material which may
result in reduced blockage of large pores by small pores and less re
stricted diffusion of solute molecules through the gel matrix. The
more uniform pore structure may allow the use of higher flow rates.
Cooper and Barrall (1973) suggested that "pooling" or solute restric
tion in porous silica media, necessitates the use of low flow rates
with these materials.
Based on these studies, solvent flow rates of 0..5 ml per min.
were selected for all subsequent studies using the 0.318-cm OD columns
with the Porasil and CPG packing materials. Based on similar studies
with 0.954-cm OD preparative columns, solvent flow rates of 1.5 ml
per min. were used for all subsequent studies on these columns packed
with Porasil or CPG packing materials.
The effects of sample size on column efficiencies of Porasil and
CPG packing materials are summarized for the analytical columns (Fig.
4). In general, the maximum sample volumes were 20 pi for the analyti
cal columns and 100 pi for preparative columns. Larger sample volumes
resulted in increased peak broadening and reduced apparent column
efficiencies. Column efficiencies were not noticeably affected with
lower sample volumes.
Even though column efficiencies are reduced with large sample
volumes, column overloading may be helpful in obtaining preparative
fractions, especially when used in conjunction with recycle chroma
tography (Bombaugh, 1971).


58
Fig. 8
Effect of excess neutral electrolyte on elution of
Na-saturated fulvic acid on Porasil A
Fig. 9 Effect of excess neutral electrolyte on elution of
Na-saturated fulvic acid on Porasil AX


56
Table 14. Peak elution volumes of cation-saturated fulvic acid on
Porasil A, Porasil AX, and CPG-250 with water as eluting
solvent
Column
Saturating
cation
Porasi1
A
Porasil
AX
CPG-
250
m1
H
1.62
1.68
1.79
Na
1.61
1.69
1.76
n(ch3)4
1.62
1.70
1.78
Ah
1.60 '
1.69
1.76


LITERATURE CITED
Bergmann, J.G., L.I. Duffy, and R.B. Stevenson. 1971. Solvent effects
in gel permeation chromatography. Anal. Chem. 43:131-133.
Bly, D.D. 1970. Gel permeation chromatography. Science 168:527-533.
Bombaugh, K.J. 1971. The practice of gel permeation chromatography.
In J.J. Kirkland (ed.), Modern Practice of Liquid Chromatography.
Wiley-Interscience, New York.
Bombaugh, K.J., W.A. Dark, and J.N. Little. 1969. Fractionation of
alcohol on deactivated porous silica beads by gel permeation
chromatography. Anal. Chem. 41:1337-1339.
Bremner, J.M. 1950. Some observations on the oxidation of soil
organic matter in the presence of alkali. J. Soil Sci. 1:
198-204.
Bremner, J.M. 1956. Some soil organic matter problems. Soil and
Fertilizers 19:115-123.
Bremner, J.M. and J. Lees. 1949. Studies on soil organic matter.
II. The extraction of organic matter from soil by neutral
reagents. J. Agr. Sci. 39:274-279.
Brook, A.J.W. and S. Housley. 1969. The interaction of phenols with
Sephadex gels. J. Chromatogr. 41:200-204.
Brook, A.J.W. and K.C. Munday. 1970. The interaction of phenols,
aniline, and benzoic acids with Sephadex gels. J. Chromatogr.
47:1-8.
Brooks, J.D., R.A. Durie, and S. Sternhell. 1958. Chemistry of
brown coals. III. Pyrolytic reactions. Aust. J. Appl. Sci.
9:303-320.
Burges, N.A., H.M. Hurst, and B. Walkden. 1964.
stituents of humic acid and their relation
the plant cover. Geochim. Cosmochim. Acta
The phenolic con-
to the lignin of
28:1547-1554.
83


ACKNOWLEDGMENTS
The author expresses sincere appreciation to Dr. J. G. A. Fiskell,
chairman, and Dr. B. G. Volk, cochairman, of the supervisory committee,
for their guidance, encouragement, and assistance during the progress
of this investigation. Appreciations are also extended to Dr. D. H.
Hubbell, Dr. N. Gammon, and Dr. W. S. Brey for their interest and
participation on the supervisory committee and review of manuscript.
Special appreciations are extended to Dr. L. W. Zelazny and Dr.
M. A. Battiste for important discussions and inspiration provided during
early stages of the investigation. A sincere thanks is extended to
faculty, staff, and students in the Soil Science Department for the many
stimulating discussions which served as the basis for the evolution
of this study.
A very special thank you is extended to Ms. Carolyn Beale and Mr.
Jerry Osbrach for assistance in the laboratory and to Ms. Ann Barry
for typing portions of the original manuscript. The author pays a
special tribute to Ms. Nancy McDavid for the very professional typing
and careful review of the manuscript and to Ms. Helen Huseman for
final preparation and drafting of several of the figures.
The author expresses his sincere gratitude to Dr. C. F. Eno,
chairman of the Soil Science Department at the University of Florida,
ii i


24
(ii) determination of column parameters, (iii) elution behavior of soil
organic extracts and organic standards with selected solvents, (iv)
elution behavior of soil organic extracts and organic standards as
influenced by saturating cations, and (v) elution behavior of soil
organic extracts and organic standards as influenced by excess electro
lyte .


76
original pattern with a major peak eluted at and a minor peak at
V This technique produced strong evidence that the initial frac
tionation pattern was not the result of a separation according to
molecular size, but instead was an artifact resulting from a complex
gel-solvent-solute interaction. The elution patterns of the soil
humic acid fractions were indeed similar to the patterns obtained from
elution of the low molecular weight aromatic acids.


12
Loeppert and Volk (1976) investigated the use of HPLC for mole
cular size fractionation of soil humic fractions on Porasil and Porasil
X and observed adsorption and electrostatic exclusion phenomena which
were highly dependent on solvent, saturating-cation, and concentration
of excess neutral electrolyte.
The polystyrene-divinylbenzene (DVB) gels, i.e. Poragel and
Styragel, are widely used in the polymer and petroleum industries
(Gaylor, James, and Weetall, 1976). Styragel and Poragel are not com
patible with aqueous solvents, acetone, or alcohols (Dark and Limpert,
1973) and exhibit a high sensitivity to solvent polarity. Changes in
solvent may result in significant changes in the amount of solvation
and swelling of the gel matrix and altered pore-size distributions
of the gel. Therefore, it is usually necessary to pack the gel as a
slurry in the same solvent which is to be used as the eluting solvent.
Edwards and Ng (1968) studied the elution of model compounds on
polystyrene-DVB gels and observed an apparent adsorption of aromatic
compounds to the gel matrix. Adsorption of compounds on polystyrene-
DVB usually caused pronounced tailing (Bergmann, Duffy, and Stevenson,
1971). Cogswell, McKay, and Latham (1971) separated the acidic con
centrate of petroleum distillate, using methylene chloride as solvent,
into four spectroscopically definable fractions and suggested molecular
association of the more acidic fractions in this solvent.
Sephadex has been widely used in studies of soil organic matter.
Although Sephadex is a different type of gel than the materials used
in these studies, a close examination of the material is in order


The C and H content of the fractions decreased and the 0 and COOH
contents and total acidity increased according to the following order:
hexane-Soxhlet, benzene-Soxhlet, ethylacetate-Soxhlet, acetone-Soxhlet,
2-propanol-Soxhlet, methanol-Soxhlet. NaOH extract-humic acid-DMF
extract, fulvic acid.
Fulvic acid and organic acid standards prepared in the H-, Na-,
N(CH_).-, and N(C,H),-saturated forms were excluded from the pores
J 4 4 9 4
of Porasil packing material when water was used as the eluting solvent.
Acetone-, 2-propanol-, and methanol-extractable soil organic matter
and organic acid standards were predominantly excluded from the pores
when methanol or DMF was used as the eluting solvent and predominantly
adsorbed when tetrahydrofuran (THF) or acetone was used. Exclusion
phenomena were evident in the presence of organic solvents with sig
nificant basic character and may be attributed to electrostatic repul
sion of negatively charged organic matter by negatively charged sites
on the silica surface. Organic solutes with significant basic charac
ter were adsorbed.
In the presence of 0.05 N excess neutral electrolyte, cation-
saturated fulvic acid and organic acid standards entered the porous
gel matrix due to suppression of charge and/or decreased electrical
double-layer thickness of the negatively charged solute molecules and
the negatively charged silica surface. As electrolyte concentration
was increased, however, adsorption phenomena became more prevalent due
to precipitation at the surface and/or direct interaction between
active sites on the silica surface and oxygen-containing functional
xi


42
Column parameters. Elution characteristics of packed columns are
summarized in Table 1. Column parameters, and V were determined
from the elution volumes for acetone or benzene and blue dextran 2,000
or 2,600,000 molecular weight polystyrene, respectively. In all cases
the column efficiencies, indicated by theoretical plate count, N, de
creased with increasing internal pore size of the packing material.
For example, the column efficiencies of Porasil AX, CX, and EX were
340, 280, and 230 theoretical plates per meter, respectively.
Molecular weight calibration curves (Fig. 5) were obtained by
elution of polystyrene standards with THF. The approximate molecular
weight working ranges, as determined with the polystyrene standards
are summarized in Table 1. The working curves obtained with polystyrene
standards on the Porasil and the CPG packing materials were not linear
over the working range of the gels.
Effect of solvent. Peak elution volumes of extractable organic
matter on Porasil A, Porasil AX, and CPG-250 packing materials are
summarized in Tables 8-10, respectively. Elution patterns of the
2-propanol-extractable material and fulvic acid are shown in Figs. 6
and 7. Samples were completely soluble at 0.1% concentration in each
of the solvents shown. A portion of the fulvic acid sample (Fig. 7) was
eluted at V the elution volume of a nonreactive high molecular weight
solute, on the Porasil A column when methanol, water, or DMF was used
as the eluting solvent. The relative quantity of sample eluted at
increased according to the following solvent order: methanol < DMF < H20.
Likewise, portions of the acetone-, 2-propanol-, and methanol-Soxhlet


21
was added using a Radiometer autoburette ABU 13 with TTT 50 titrator
module. Potentials were determined with a Radiometer PHM 64 pH meter
and recorded using a Radiometer REC strip chart recorder with REA 160
titrigraph module. A Radiometer combination electrode with porous plug
liquid junction was used for all potentiometric determinations. The
calomel cell was filled with saturated KC1 in H2O for aqueous titra
tions or with saturated KC1 in CH^OH for titrations in nonaqueous
media (Loeppert, Zelazny, and Volk, 1976).
High Pressure Liquid Chromatography
The Waters' ALC 202 liquid chromatograph equipped with the Model
6000 solvent delivery system and 401 differential refractometer and
UV detectors was used for all separations. Samples were injected
through the U6K universal liquid chromatograph injector. A solvent
flow rate of 0.5 ml/minute were used, unless otherwise specified.
Column effluent was monitored by differential refractive index or by
UV absorption at 254 or 280 nm, and recorded continuously on a Perkin
Elmer 201 strip-chart recorder.
Samples were dissolved in the eluting solvent to give a 0.1% con
centration (W/V) and centrifuged at 35,000 x G for 20 min. Injection
volume was 10 yl.
Solvents used in the chromatographic studies were deionized
water and spectroquality methanol, 2-propanol, t-butanol, tetrahydrofuran
(THF), dimethylformamide (DMF), ethylacetate, acetone, and chloroform.


82
eluted in the vicinity of V Of the compounds tested, only several
hydrophobic aromatic compounds were strongly adsorbed. It appears
that THF is a suitable solvent for elution of acidic solutes; however,
only the benzene-, ethylacetate-, acetone-, and 2-propanol-Soxhlet
fractions were suitably soluble in THF. Methylated fractions of the
humic acid and fulvic acid fractions would also be soluble in THF.
Molecular weights of acetone-, 2-propanol-, and methanol-Soxhlet
fractions were estimated to be 500 to 800, based on elution patterns
of soil humic fractions with those of polystyrene standards in THF.
Further research x^ill be needed to corroborate, by other methods, the
molecular weight estimates obtained with gel permeation chromatography.
One such method would be vapor pressure osmometry.
Each of the two general groups of gels evaluated in this study
showed evidence of adsorptive and/or electrostatic interaction with
extractable soil organic matter. Mode and extent of interaction were
highly dependent on the solvent medium. Because of these possible
interactions, special care must be observed in the interpretation of
gel permeation chromatography patterns.
High pressure liquid chromatography is a valuable new technique
because of the use of high efficiency columns and the short time re
quired to obtain elution patterns. In addition to the application of
gel permeation chromatography, there is the unexplored potential
application of liquid-liquid partition chromatography and liquid-solid
adsorption chromatography to fractionation of extractable soil organic
matter. High pressure liquid chromatography will also provide a valu
able tool for studies of soil organic-mineral-ionic interactions.


33
Table 5 (Extended)
Solvent
Ethyl t-Butanol 2-Propanol Methanol Pyridine DMF DMSO H0
ether 2
I
P
S
P
P
I
I
I
P
S
s
s
s
s
P
I
p
p
p
s
s
s
p
I
p
I
I
I
p
s
p
I
p
I
I
I
p
s
s
p
p
1
p
s
s
s
s
p
s
p
s
c
s
s
s
s
s
p
p
s
s
s
s
s
p
I
I
I
I
I
p
p
p
p
p
p
p
s
p
s
I
p
s
p
p
I
I
I
p
s
s
s
s
s
p
I
p
p
p
s
s
s
p
I
p
I
I
I
p
s
p
I
p
I
I
I
p
s
s
p
p
I
p
s
s
s
s
p
s
p
s
s
s
s
s
p
p
s
s
s
s
s
p
I
I
I
I
I
p
p


Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
HIGH-PRESSURE LIQUID CHROMATOGRAPHY AND CHEMICAL CHARACTERIZATION
OF EXTRACTABLE SOIL ORGANIC MATTER
By
Richard Henry Loeppert, Jr.
December, 1976
Chairman: Dr. John G. A. Fiskell
Cochairman: Dr. B. G. Volk
Major Department: Soil Science
The objective of this investigation was to evaluate the use of
high-pressure liquid chromatography and a series of new packing materi
als (Porasil-silica gel, Corning controlled pore glass, and polystyrene-
divinylbenzene) for the molecular size fractionation of extractable soil
organic matter. The effects of packing material, solvent, saturating-
cation, concentration of excess electrolyte, and pH on solute-gel-
solvent interactions were investigated. Preliminary experiments were
performed to investigate the behavior of soil humic compounds in organic
solvents and to select solvents which would be suitable for extraction
and fractionation of the soil humic complex.
The soil used was Terra Ceia muck, a Typic Medisaprist. Organic
matter was extracted from the soil by separate treatment with 0.5 N
NaOH and dimenthylformamide (DMF). The NaOH-extractable material was
separated into humic acid and fulvic acid fractions. In addition, both
the NaOH- and DMF-extractable materials were further fractionated with
a Soxhlet extraction scheme. The ash content of all samples was
lowered by dialysis to less than 0.5%.
x


29
Infrared patterns of the Soxhlet fractions (Fig. 2) substanti
ated differences evident in the elemental and functional group analyses.
From the least polar to the most polar extracting solvents, Soxhlet
extracts showed progressively weaker C-H stretching vibrations in the
2900 cm region. Absorption at 1450 cm ^ due to the bending vibra
tion also decreased according to the same solvent order. These adsorp
tion bands of the infrared patterns, along with the C and H concentrations,
indicated that the less polar solvents extracted a material of greater
aliphatic character. Each of the Soxhlet fractions had strong C=0
stretching bands at 1725 cm ^ and 1630 cm ^ resulting from COOH and
C00-, respectively. The band at 1630 cm ^ was considerably stronger
in the absorption spectra of materials extracted by the more polar
solvents. The band at 700-750 cm \ evident in the absorption spectra
of hexane-, benzene-, and 2-propanol-extractable materials, was attributed
to aromatic C-H out-of-plane bending vibrations. The absence of bands
in this region of the spectrum indicates either the absence of aromatic
structure or the possibility of a completely substituted aromatic ring
system. The weak absorption at 700-750 cm ^ for the methanol-Soxhlet
extracts compared to the hexane-, benzene-, ethylacetate-Soxhlet extracts
may be attributed to a highly substituted aromatic ring system. Each
of the Soxhlet-extractable materials showed a strong absorption band
at 3440 cm attributed to H-bonded OH groups and a weaker band at
approximately 1250 cm attributed to C-0 stretching vibrations.
Absorption at both frequencies increased with increasing polarity of
the extracting solvent. The hexane- and benzene-Soxhlet fractions,
especially, showed significantly weaker absorption in the 3440 cm ^
region than fractions extracted by the more polar solvents.


20
and Gupta, 1965; Schnitzer and Khan, 1972). Total COOH-group concen
tration was determined by addition of excess CaiC^H^op,, and back-
titration of excess C.-,H^0o with HC1 to pH 8.4 (Schnitzer and Gupta,
1965; Schnitzer and Khan, 1972). Total OH was determined by an
acetylation procedure (Brooks, Durie, and Sternhell, 1957) and total
C0 by the oximation method (Fritz, Yamamura, and Bradford, 1959).
Phenolic OH group concentration was calculated as the difference between
total acidity and COOH-group concentration, and alcoholic OH concentra
tion was estimated by subtracting phenolic OH concentration from total
OH concentration. All elemental and functional group analyses were cor
rected for moisture content and ash content.
Moisture content was determined by heating a preweighed combustion
boat with a known quantity of material to HOC for 24 hours and weigh
ing. Ash content was determined by heating a preweighed combustion boat
with a known quantity of organic material to 700C for 4 hours and weigh
ing. All weights, yields, and concentrations are reported on a moisture-
and ash-corrected organic matter basis.
Infrared patterns were obtained on a Perkin-Elmer model 127
spectrophotometer. The KBr pellets were prepared by mixing 0.8 mg of
organic material, previously dried 24 hours over P2O5 in a vacuum
desiccator, with 200 mg of dried KBr in a Wig-L-Bug amalgamator. The
mixture was dried an additional 12 hours in a vacuum desiccator over
P^O^ prior to obtaining the spectrum.
Continuous-and stepped-potentiometric titrations were performed on
10 mg of organic material placed in the appropriate solvent. Titrant


81
gels in aqueous systems. Since the fulvic acid sample contained
nitrogen, indicating the probable presence of basic sites, interac
tions of these sites in the negatively charged solute molecule with the
negatively charged silicate surface would be greater in the presence
of excess neutral electrolyte.
When protic solvents, i.e. ^0, methanol, and 2-propanol, or
dipolar aprotic solvents with significant basic properties, i.e. DMF,
were used to elute low molecular weight acidic solutes or extractable
soil organic matter, electrostatic exclusion phenomena predominated.
In dipolar aprotic solvents without significant basic character, i.e.
acetone and THF, adsorption phenomena predominated. With all solvent
and electrolyte systems examined, it was not possible to completely
eliminate both electrostatic exclusion and adsorption phenomena. De
activation of the Porasil surface, Porasil X, did not completely eliminate
adsorption and exclusion interactions between the acidic organic solute
and the silica surface.
The polystyrene-DVB gels were compatible with a more limited range
of solvents than the silica gels. Also, due to the swelling properties
of the gel it was essential to pack the column with the same solvent
which was to be used as the eluting solvent. With the polystyrene-DVB
gels, elution patterns were dependent on the eluting solvent. In DMF,
low and high molecular weight acidic organic solutes were totally or
partially excluded from the gel matrix. In THF, low molecular weight
acid solutes apparently readily entered the porous gel matrix and were


45
Table 9. Peak elution volumes of extractable soil organic matter on
Porasil AX with selected solvents
Solvent
Sample H^O CH.^OH 2-Propanol t-Butanol Acetone THF DMF
DMF extract
Acetone-Soxhlet
1.
,82
(3.
30)
2-Propanol-Soxhlet --
1.
,75
(3.
,28)
Methanol-Soxhlet
1,
,80
(3.
.31)
NaOH extract

Acetone-Soxhlet
1.
.81
(3.
.31)
2-Propanol-Soxhlet
1.
,79
(3.
,32)
Methanol-Soxhlet
1.
,82
(3.
.27)
Humic acid

Fulvic acid 1.68
1.
,73
ml
3.35
3.42
AD
3.39
ADb
AD
AD


3.40
3.33
3.37
AD
3.35
(AD)
(AD)
AD
3.39
Parentheses ( ) indicate secondary peak.
'
AD = severe adsorption.


84
Cheshire, M.V., P.A. Cranwel1, C.P. Falshow. A.J. Floyd, and R.D.
Haworth. 1967. Humic acid: 2. Structure of humic acid.
Tetrahedron 23:1669-1682.
Choudhri, M.B., and F.J. Stevenson. 1957. Chemical and physico
chemical properties of soil humic colloids: III. Extraction
of organic matter from soils. Soil Sci. Soc. Amer. Proc. 21:
508-513.
Cogswell, T.E., J.F. McKay, and D.R. Latham. 1971. Gel chromato
graphic separation of petroleum acids. Anal. Chem. 43:645-648.
Cooper, A.R. and E.M. Barrall. 1973. Gel permeation chromatography:
physical characterization and chromatographic properties of
Porasil. J. Appl. Polym. Sci. 17:1253-1268.
Cooper, A.R. and A.R. Bruzzone. 1973. Characterization and proper
ties of macromolecules. III. Gel permeation chromatography:
the effect of temperature on the elution volume and the effi
ciency of the separation process. J. Polym. Sci. A-2, 11:
1423-1434.
Cooper, A.R., A.R. Bruzzone, J.H. Cain, and E.M. Barrall. 1971.
Characterization and chromatographic properties of Corning porous
glasses. J. Appl. Polym. Sci. 15:571.
Cooper, A.R. and J.F. Johnson. 1969. Gel permeation chromatography:
the effect of treatment with hexamethyl disilazane on porous
glass packings. J. Appl. Polym. Sci. 13:1487-1492.
Cooper, A.R., J.F. Johnson, and R.S. Porter. 1973. Gel-permeation
chromatography: current status. American Laboratory. May
1973:12-24.
Dark, W.A. and R.J. Limpert. 1973. An evaluation of available pack
ings for GPC. J. Chromatogr. Sci. 11:114-120.
Demetriou, J.A. et al. 1968. Gel filtration chromatography of
fluorescent phenolic and heterocyclic compounds. J. Chromatogr.
34:342-350.
Determann, H. and I. Walter. 1968. Source of aromatic affinity to
Sephadex dextran gels. Nature (Lond.) 219:604-605.
Edwards, G.D. and Q.Y. Ng. 1968. Elution behavior of model compounds
in gel permeation chromatography. J. Polym. Sci. Part C 21:
105-117.
Flaig, W., H. Beutelspacher, and E. Rietz. 1975. Chemical composi
tion and physical properties of humic substances. In J.E.
Gieseking (ed.), Soil components. Springer-Verlag, New York.


77
Table 18. Molecular weight estimates of soil humic fractions based
on elution of polystyrene standards on y-Styragel with
THF as the eluting solvent
Sample
Estimated
molecular weight
NaOH extract
DMF extract
Acetone-Soxhlet
560
660
2-Propanol-Soxhlet
720
740
Methanol-Soxhlet
760
740


49
Table 11. Peak elution volumes of organic standards on Porasil A with
selected solvents
Solvent
Sample H^O CH^OH 2-Propanol t-Butanol Acetone THF DMF
ml
1,2,4,5-Tetracar-
boxybenzene
1.73
1.72
3.70
9.25

3.55
2.36
1,3,5-Tricarboxy-
benzene
1.76
1.94
AD3
3.68

3.54
3.60
3,5-Dihydroxyben-
zoic acid
2.08
4.87
3.33

3.47
3.43
Benzoic acid
2.37
5.29
3.51
3.59
3.58
3.44
Pyridine
5.20
AD
AD
4.45
3.87
3.58
Aniline
3.44
4.74
4.04
3.55
3.62
3.41
Methanol

3.64
3.94
3.91
3.42
3.94
Ethylene glycol
3.40
4.45
4.76
3.97
3.45
3.46
Acetone
3.36
3.97
4.52
3.70
3.46
a
AD =
severe adsorption


22
Salt solutions in water and methanol were prepared by adding the de
sired quantity of 2.0 N NaOH, tetramethylammonium hydroxide, or
tetrabutylammonium hydroxide to the solvent and adjusting to the
appropriate pH with HC1, HNO^, H^SO^, or H^PO^. Halide salts were
avoided where possible, since these salts may result in corrosion of
the stainless steel surfaces of the liquid chromatographic pump, con
necting lines, and columns. In the few cases where halide salts were
employed, the pH of the eluting solvent was maintained above 7.0, and
the chromatographic system was flushed with copious quantities of de
ionized water following use of the halide salt.
Characteristics of the packing materials used in this study
are summarized in Table 1. The silica and porous glass packing ma
terials were dry-packed into stainless steel columns (0.318 cm OD x
2 f or 0.318 cm OD x 1 m). The semirigid gels were slurry packed in
the same solvent as the eluting solvent. Poragel columns (stainless
steel, 0.954 cm OD x 2 f) packed in THF were used as obtained from the
manufacturer. Poragel in DMF was slurry packed into 0.954 cm OD x 2 f
stainless steel columns. The p-Styragel columns (stainless steel,
0.954 cm OD x 1 f) packed in THF were used as obtained from the manu
facturer. When used in series, columns were connected with U-shaped
0.009 in. ID tubing.
A series of experiments were performed to investigate the elu
tion behavior of extractable soil organic matter on the gel permeation
packing materials: (i) determination of optimum operating conditions,


and to Dr. D. W. Beardsley and Dr. D. H. Myhre, Center Directors, at
the Agricultural Research and Education Center, Belle Glade, for pro
viding the research assistantship which has enabled the author to
pursue his doctoral program.
The author deeply appreciates the continuing encouragement and
assistance given him by his parents throughout the course of his studies
and the special upbringing which has encouraged the author to search,
to question, and to approach problems with an open mind.
IV


Table 1. Parameters of column packing materials
Packing material
b
c
d
Approximate
Description
Column
vo
VT
N
molecular weight
a
size
I
working range
cm
ml
ml
Porasil A
porous silica
0.318
1.82
3.35
360
1,000-60,000
Porasil C
f 1
11
1.76
3.36
340
1,000-250,000
Porasil E
f 1
It
1.62
3.42
290
1,000-2,000,000
Porasil AX
porous silica
II
1.84
3.32
340
1,000-60,000
Porasil CX
deactivated with
polyethylene oxide
It
1.88
3.42
280
1,000-250,000
Porasil EX
1!
11
1.67
3.49
230
1,000-2,000,000
CPG 40
crushed glass
If
1.60
3.22
1,020
500-10,000
CPG 250
II
It
1.54
3.31
660
5,000-100,000
Poragel 100 A
polystyrene-DVB
0.954
10.8
21.3
800
500-20,000
Poragel 500 A
II
II
10.6
20.8
800
1,000-100,000
p-Styragel 100 A
II
If
5.2
10.1
5,600
100-3,000
p-Styragel 500 A
II
11
4.6
9.8
5,000
100-10,000
aAll columns are one meter long except the Porage.1 and y-Styragel columns which are 0.610 and 0.305 meters,
respectively. Column size indicates the outside column diameter.
^Determined by elution of 1,600,000 molecular weight polystyrene standard with THF.
c
Determined by elution of acetone or benzene with THF.
^Column efficiency expressed in theoretical plates.


39
Fig. 3 Effect of flow rate on column efficiency, N, of Porasil
A, Porasil AX, and CPG-250 analytical columns


RECORDER RESPONSE
74
Fig. 15 Elution of Soxhlet extracts of NaOH-extractable soil
organic matter on 100 A y-Styragel with THF


BIOGRAPHICAL SKETCH
Richard Henry Loeppert, Jr., was born September 26, 1944, in
Raleigh, North Carolina. He graduated from Needham B. Broughtan High
School in Raleigh, North Carolina, in June, 1962. In August, 1966,
he received his Bachelor of Science degree with a major in soil science
from North Carolina State University, Raleigh, North Carolina.
Following graduation, he was employed as Assistant County Agent
with the Florida Agriculture Extension Service in Jackson County,
Florida. He began his graduate studies at the University of Florida
in 1970 and received his Master of Science degree in soil science in
August, 1973. He is currently a candidate for the Ph.D. degree in the
Department of Soil Science, University of Florida.
He is a member of the American Society of Agronomy, the Soil
Science Society of America, the International Society of Soil Science,
and the Clay Minerals Society.
89


59
Fig. 10 Effect of excess neutral electrolyte on elution of
Na-saturated 1,2,4,5-tetracarboxybenzene on Porasil A
Fig. 11 Effect of excess neutral electrolyte on elution of
Na-saturated 1,2,4,5-tetracarboxybenzene on Porasil AX


88
Sommers, T.C. 1966. Wine tanninsIsolation of condensed flavanoid
pigments by gel filtration. Nature (Lond.) 209:368-370.
Spatorico, A.L. 1975. Exclusion chromatography using controlled
porosity glass. I. Comparison with styrene gels. J. Appl.
Polym. Sci. 19:1601-1610.
Spatorico, A.L. and G.L. Beyer. 1975. Exclusion chromatography using
porous glass. II. Application to hydrophilic polymers. J.
Appl. Polym. Sci. 19:2933-2945.
Steelink, C. and G. Tollin. 1967. Free radicals in soil. Jen. A.D.
McLaren and G.H. Peterson (eds.), Soil biochemistry. Marcel
Dekker, New York.
Stevenson, F.J. 1965. Gross chemical fractionation of organic matter.
In C.A. Black (ed.), Methods of soil analysis, Agronomy Monograph
No. 9, Vol. 2. American Society of Agronomy. Madison, Wisconsin
Swift, R.S. and A.M. Posner. 1971. Gel chromatography of humic acid.
J. Soil Sci. 22:237-249.
Talhoun, S.A. and M.M. Mortland. 1966. Complexes of montmorillonite
with primary, secondary, and tertiary amides. I. Protonation
of amides on the surface of montmorillonite. Soil Sci. 102:
248-254.
Tinsley, J. and A. Salam. 1961. Extraction of soil organic matter
with aqueous solvents. Soils and Fert. 24:81-84.
Unger, K.K., R. Kern, M. Ninou, and K.F. Krebs. 1974. GPC with a
new type of silica packing material. J. Chromatogr. 99:435-443.
Volk, B.G. and M. Schnitzer. 1973. Chemical and spectroscopic methods
for assessing subsidence in Florida Histosols. Soil Sci. Soc.
Amer. Proc. 37:886-888.
Zweig, G. and J. Sherma. 1974. Chromatography. Anal. Chem. 46:73R-
94R.


LIST OF TABLES (continued)
Table Page
15 Peak elution volumes of selected organic acid standards
on Porasil A with NaSO. solutions 60
2 4
16 Peak elution volumes of selected organic acid standards
on Porasil AX with Na^SO, solutions 61
2 4
17 Peak elution volumesof low molecular weight standards
eluted on 100 A Poragel with THF and DMF and on 100 A
p-Styragel with THF as the eluting solvent 70
18 Molecular weight estimates of soil humic fractions based
on elution of polystyrene standards on p-Styragel with
THF as the eluting solvent 77
vii


63
At pH values above 7.0, the silica surface may be destroyed by
solubilization and formation of silicate. Therefore, alkaline condi
tions must be avoided. Under alkaline conditions, the chemically
adsorbed deactivating agent is also readily stripped from the surface
of the Porasil X packing material. The manufacturer recommends that
use of several organic solvents, especially DMF, should be avoided
with deactivated Porasil. Such a solvent may readily strip the de
activating agent from the silica surface.
Polystyrene-divinylbenzene (DVB)
Operating conditions. As with the porous glass packing materials,
column efficiencies of Poragel and y-Styragel packing materials were
greatly influenced by solvent flow rates. With 0.054-cm OD columns and
THF as the eluting solvent, minimum peak broadening and maximum column
efficiencies were obtained at flow rates of approximately 0.8 and 3.0
O
ml per min. for the 100 A Poragel and y-Styragel columns, respectively
(Fig. 12). At higher flow rates, peak broadening was increased and
column efficiencies were decreased. At lower flow rates, column effi
ciencies were also appreciably lowered. This later effect was much
more evident with the polystyrene-DVB gels than with the porous glasses.
At low flow rates, diffusion of solute molecules apparently resulted
in decreased column efficiencies. With the y-Styragel columns, the
high column efficiency at the high flow rate was due to the small
particle size and uniform pore size of packing which permitted rapid


72
of the two solvents, as discussed previously. The second point is
elaborated upon below.
The acidic functional groups of an acidic organic solute would be
partially dissociated in DMF due to basic character of this solvent;
therefore, the solute molecules are likely to be highly dispersed.
The negatively charged solute molecules would have larger effective
radii than the neutral species; however, this phenomena should not
entirely account for the exclusion phenomena since the exclusion limit
of the gel, based on polystyrene standards, is approximately 50,000
molecular weight. In the porous silica and Sephadex gels, the exclusion
phenomena can be explained in terms of electrostatic repulsion from
negative charge sites in the gel matrix. In porous silica, the negative
charge results from dissociation of Si(OH) sites at silica surface.
In Sephadex the negative charge has been attributed to COOH impurities
in the gel matrix. On the other hand, the polystyrene-DVB gel should
exist as a neutral species. Therefore, we must search for an alternate
explanation to the exclusion phenomena. A possible explanation is the
ion inclusion effect suggested by Forss and Stenlund (1975) in studies
of lignosulfonate. They attributed this effect to the interaction of
charged sites on the ions entering the pores with other charged ions
outside of the pores. The net effect is electrostatic repulsion. Such
an effect would not entirely account for the apparent total exclusion
of low molecular weight solutes observed. Further work will be required
to determine the nature of this phenomena.



PAGE 1

HIGH-PRESSURE LIQUID CHROMATOGRAPHY AND CHEMICAL CHARACTERIZATION OF EXTRACTABLE SOIL ORGANIC MATTER By RICHARD HENRY LOEPPERT, JR. A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1976

PAGE 2

To my parents, Richard and Adeline Loeppert

PAGE 3

ACKNOWLEDGMENTS The author expresses sincere appreciation to Dr. J. G. A. Fiskell, chairman, and Dr. B. G. Volk, cochairman of the supervisory coimnittee, for their guidance, encouragement, and assistance during the progress of this investigation. Appreciations are also extended to Dr. D. H. Hubbell, Dr. N. Gammon, and Dr. W. S. Brey for their interest and participation on the supervisory committee and review of manuscript. Special appreciations are extended to Dr. L. W. Zelazny and Dr. M. A. Battiste for important discussions and inspiration provided during early stages of the investigation. A sincere thanks is extended to faculty, staff, and students in the Soil Science Department for the many stimulating discussions which served as the basis for the evolution of this study. A very special thank you is extended to Ms. Carolyn Beale and Mr. Jerry Osbrach for assistance in the laboratory and to Ms. Ann Barry for typing portions of the original manuscript. The author pays a special tribute to Ms. Nancy McDavid for the very professional typing and careful review of the manuscript and to Ms. Helen Huseman for final preparation and drafting of several of the figures. The author expresses his sincere gratitude to Dr. C. F. Eno, chairman of the Soil Science Department at the University of Florida, ii i

PAGE 4

and to Dr. D. W. Beardsley and Dr. D. H. Myhre, Center Directors, at the Agricultural Research and Education Center, Belle Glade, for providing the research assistantship which has enabled the author to pursue his doctoral program. The author deeply appreciates the continuing encouragement and assistance given him by his parents throughout the course of his studies and the special upbringing which has encouraged the author to search, to question, and to approach problems with an open mind. iv

PAGE 5

TABLE OF CONTENTS Page ACKNOWLEDGMENTS iii LIST OF TABLES vi LIST OF FIGURES viii ABSTRACT x INTRODUCTION 1 LITERATURE REVIEW 4 Soil Organic Matter 4 Extraction of Soil Organic Matter 6 Gel Permeation Chromatography 8 General Information 8 Theory and Nomenclature 8 Packing Materials 10 MATERIALS AND METHODS 1& Sample Pretreatment and Extraction 16 Solubility Studies 19 Analytical Determinations 19 High Pressure Liquid Chromatography ... 21 RESULTS AND DISCUSSION 25 Chemical Characteristics of Extractable Organic Matter. 25 Solubility Characteristics of Extractable Organic Matter. 31 High Pressure Liquid Chromatography ..... 38 Porous Silica Packing Materials 38 Polystyrene-divinylbenzene (DVB) 63 CONCLUSIONS 78 Extraction and Fractionation 78 Solubility Properties 79 Liquid Chromatography 80 LITERATURE CITED 83 BIOGRAPHICAL SKETCH 89 V

PAGE 6

LIST OF TABLES Table Page 1 Parameters of column packing materials 23 2 Yields of extractable soil organic matter ....... 26 3 Elemental and futidtional group concentrations of extractable soil organic matter 27 4 Titratable acidity of extractable soil organic matter 28 5 Solubility of extractable soil organic matter in selected solvents at 0.1% concentration 32 6 Solubility of extractable soil organic matter as influenced by saturating cation and solvent ...... 34 7 Solubility of fulvic acid in aqueous salt solutions 35 8 Peak elution volumes of extractable soil organic matter on Porasil A with selected solvents 44 9 Peak elution volumes of extractable soil organic matter on Porasil AX with selected solvents 45 10 Peak elution volumes of extractable soil organic matter on CPG-250 with selected solvents 46 11 Peak elution volumes of organic standards on Porasil A with selected solvents 49 12 Peak elution volumes of organic standards on Porasil AX with selected solvents 50 13 Peak elution volumes of organic standards on CPG-250 with selected solvents 51 14 Peak elution volumes of cation-saturated fulvic acid on Porasil A, Porasil AX, and CP(;-;."-n with water as eluting solvent 56 vi

PAGE 7

LIST OF TABLES (continued) Table page 15 Peak elution volumes of selected organic acid standards on Porasil A with Na^SO, solutions. 60 16 Peak elut ion volumes of selected organic acid standards on Porasil AX with Na^SO^ solutions ..... 61 17 Peak elut ion volumesof low molecular weight standards eluted on 100 A Poragel with THF and DMF and on 100 A M-Styragel with THF as the eluting solvent 70 18 Molecular weight estimates of soil humic fractions based on elution of polystyrene standards on y-Styragel with THF as the eluting solvent 77 vii

PAGE 8

LIST OF FIGURES Figure Page 1 Extraction scheme 2.7 2 Infrared patterns 30 3 Effect of flow rate on column efficiency, N, of Porasil A, Porasil AX, and CPG-250 analytical columns 39 4 Effect of sample size on column efficiency, N, of Porasil A, Porasil AX, and CPG-250 analytical columns 41 5 Molecular weight calibration curves of 1 m x 0.318 cm OD Porasil AX and CPG analytical columns obtained by elution of polystyrene standards with THF. ....... 43 6 Elution patterns of the 2-prOpanol-Soxhlet extract of DMF-extractable material on Porasil A and Porasil AX with selected solvents ...... 47 7 Elution patterns of fulvic acid on Porasil A and Porasil AX with selected solvents. ..... 47 8 Effect of excess neutral electrolyte on elution of Nasaturated fulvic acid on Porasil A .......... 58 9 Effect of excess neutral electrolyte on elution of Na— saturated fulvic acid on Porasil AX 58 10 Effect of excess neutral electrolyte on elution of Nasaturated 1 2 4 5-tetracarboxybenzene on Porasil A 59 11 Effect of excess neutral electrolyte on elution of Nasaturated 1 2 4 5-tetracarboxybenzene on Porasil AX. 59 12 Effect of flow rate on column efficiency, N, of 100 A Poragel and 100 A p-Styragel preparative columns with THF as the eluting solvent 64 13 Effect of sample^size on column efficiency, N, of 100 A Poragel and 100 A y-Styragel preparative columns with THF as the eluting solvent 66 viii

PAGE 9

LIST OF FIGURES (continued) Figure Page 14 Molecular weight calibration curve of y-Styragel (2 f X 0.954 cm OD 100 A + 2 f x 0.954 cm 00 500 A) obtained by elution of polystyrene standards with THF 67 15 Elution of Soxhlet extracts of NaOH-extractable soil organic matter on 100 A y-Styragel with THF 74 16 Elution of Soxhlet extracts of DMF-extractable soil organic matter on 100 A vi-Styragel with THF 75 ix

PAGE 10

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy HIGH-PRESSURE LIQUID CHROMATOGRAPHY AND CHEMICAL CHARACTERIZATION OF EXTRACTABLE SOIL ORGANIC MATTER By Richard Henry Loeppert, Jr. December, 19 76 Chairman: Dr. John G. A. Fiskell Cochairman: Dr. B. G. Volk Major Department: Soil Science The objective of this investigation was to evaluate the use of high-pressure liquid chromatography and a series of new packing materials (Porasil-silica gel, Corning controlled pore glass, and polys tyrenedivinylbenzene) for the molecular size fractionation of extractable soil organic matter. The effects of packing material, solvent, saturatingcation, concentration of excess electrolyte, and pH on solute-gelsolvent interactions were investigated. Preliminary experiments were performed to investigate the behavior of soil humic compounds in organic solvents and to select solvents which would be suitable for extraction and fractionation of the soil humic complex. The soil used was Terra Ceia muck, a Typic Medisaprist. Organic matter was extracted from the soil by separate treatment with 0.5 N NaOH and dimenthylf ormamide (DMF) The NaOH-extractable material was separated into humic acid and fulvic acid fractions. In addition, both the NaOHand DMF-extractable materials were further fractionated with a Soxhlet extraction scheme. The ash content of all samples was lowered by dialysis to less than 0.5%.

PAGE 11

The C and H content of the fractions decreased and the 0 and COOH contents and total acidity increased according to the following order: hexane-Soxhlet benzene-Soxhlet ethylacetate-Soxhlet acetone-Soxhlet 2-propanol-Soxhlet methanol-Soxhlet NaOH extract-humic acid-DMF extract, fulvic acid. Fulvic acid and organic acid standards prepared in the H-, Na-, N(CH^)^-, and N (C^Hg) ^-saturated forms were excluded from the pores of Porasil packing material when water was used as the elating solvent. Acetone-, 2-propanol-, and methanol-extractable soil organic matter and organic acid standards were predominantly excluded from the pores when methanol or DMF was used as the eluting solvent and predominantly adsorbed when tetrahydrof uran (THF) or acetone was used. Exclusion phenomena were evident in the presence of organic solvents with significant basic character and may be attributed to electrostatic repulsion of negatively charged organic matter by negatively charged sites on the silica surface. Organic solutes with significant basic character were adsorbed. In the presence of 0.05 excess neutral electrolyte, cationsaturated fulvic acid and organic acid standards entered the porous gel matrix due to suppression of charge and/or decreased electrical double-layer thickness of the negatively charged solute molecules and the negatively charged silica surface. As electrolyte concentration was increased, however, adsorption phenomena became more prevalent due to precipitation at the surface and/or direct interaction between active sites on the silica surface and oxygen-containing functional xi

PAGE 12

groups of the organic solute. Deactivation of the silica surface (Porasil X) resulted in reduction but not elimination of adsorption and electrostatic exclusion phenomena. Elution of soil humic compounds and low molecular weight standards on polystyrene-divinylbenzene (DVB) was strongly influenced by the solvent. Each of the low-molecular weight organic solutes entered the porous gel matrix when eluted with THE, however, several hydrophobic aromatic compounds (e.g. benzene, toluene, and anthracene) were adsorbed. Soil humic compounds apparently readily entered the polystyreneDVB gel matrix when THE was used. When elated with DME, however, soil humic compounds and low molecular weight organic acid standards were either totally or partially excluded from the gel matrix. This phenomenon was attributed to the effect of solvent on dissociation of solute molecules and to a possible ion-inclusion effect. None of the gels investigated was chemically inert, and each apparently interacted with the soil humic material. Interactions could be minimized, however, by proper selection of gel, solvent, and concentration of excess electrolyte. Molecular weights of acetone-, 2-propanol-, and methanol-Soxhlet fractions were estimated to be 500 to 800, based on comparison of elution patterns of soil humic fractions with those of polystyrene standards in THE. xii

PAGE 13

INTRODUCTION Soil is an important international resource which serves as the source of the majority of the world's food supply and as a major sink for all man-made and natural products in the environment. Organic matter, due to its reactive nature, has a large influence on soil properties. In order to understand the chemical properties of soil organic matter and the exact role of humic fractions in the soil, it is essential that the scientist understand the chemical structures involved. Many years of research have given structural clues, however, due to the complexity of soil organic matter, the science is still in its infancy. One approach to the structural problem has been to initially fractionate and simplify the humic material prior to further analytical investigations. Since the development of the polydextran gels, there has been considerable interest in gel filtration for molecular-size fractionation and characterizatioti of soil organic matter (Swift and Posner, 1971). Scientists have used this method to isolate molecular-size fractions and to obtain estimates of molecular weight of humic substances in soils and natural water. During the past decade, rapid advances have been made in liquid chromatography. These advances have been due primarily to the

PAGE 14

2 development of the high-pressure liquid chromatograph which, in turn, has made possible the use of small-diameter packing materials and high efficiency columns. For example, with the new y-packing materials, column efficiencies as high as 5000 theoretical plates per foot are commonly attained. The rapid rise in the application of high-pressure liquid chromatography is readily apparent following a quick glance through any recent issue of Analytical Chemistry. Attempts at molecular-size fractionation of soil humic materials have been complicated by the fact that no gel material, including Sephadex, is completely inert. Therefore, separations may be adversely affected by gel-solute and gel-solvent interactions which would lead to misleading results. Also, the hxamic molecule is a strongly reactive solute which has a strong tendency to interact with other solute molecules and solvent molecules. For this reason, the fractionation of soil humic compounds is affected by solute-solute and solute-solvent interactions. Each of the above-mentioned interactions is strongly influenced by packing material, solvent, saturating cation, concentration of excess electrolyte, and pH. The objectives of this work were to investigate the use of highpressure liquid chromatography and a series of new packing materials for the size fractionation of soil organic matter extracts and to investigate the effect of packing material, solvent, saturatingcation, concentration of excess electrolyte, and pH on solute-solute, gel-solute, solvent-solute, and gel-solvent interactions which would influence size separations. Since several of the new packing materials

PAGE 15

3 were compatible with organic solvents, but not with aqueous solvents, we were also interested in the behavior of soil humic compounds in organic solvents and in selection of organic solvents which were suitable solvent media for extraction and preliminary fractionation of the soil humic complex.

PAGE 16

LITERATURE REVIEW Soil Organic Matter For discussions of our current knowledge of the chemical makeup Of soil organic matter the reader is referred to texts by Kononova (1966) and Schnitzer and Khan (1972) and a review by Hurst and Burges (1967) Some of the important characteristics and properties which are relevant to the discussion in the text are summarized in the next few paragraphs. Hurst and Burges (1967) suggested that humic acids are polycondensates of monomers immediately available in a particular microarea of the soil and do not appear to have integrity of structure and the rigid chemical conf iguistion of many other macromolecules due to the complexity and heterogeneous nature of the system in which they are formed. According to Kononova (1966), possibly no two humus molecules would have exactly the same structure. Elemental and functional group compositions for representative humic and fulvic acids have been tabulated by Schnitzer and Khan (1972), the most striking features of these tabulations are the relatively high oxygen contents, low nitrogen and sulfur contents, high carbon to hydrogen ratios, high total acidity, and high carboxyl and phenolic hydroxyl contents. Comparisons of humic and fulvic acid (Schnitzer and 4

PAGE 17

5 Khan, 1972) showed that fulvic acid had lower carboxyl contents, higher oxygen contents, and higher total acidity and carboxyl content. Infrared spectra of soil humic compounds show broad absorption bands (Schnitzer and Khan, 1972). The majority 0f spectra did not show absorption bands in the 600-900 cm~^ region, and, therefore, did not demonstrate the presence of aromatic protons. Likewise, nuclear magnetic resonance (WIR) spectra of methylated fulvic acid did not indicate the presence of aromatic protons (Schnitzer and Skinner, 1968) Schnitzer and Khan (1972) determined molecular weights of 1684 and 669 for humic acid and fulvic acid, respectively, by the freezing point depression method; however, molecular weights as high as 100,000, or greater, have been determined by other methods (Schnitzer and Khan, 1972) Electron spin resonance (ESR) spectra of soil humic substances indicate the presence of a high concentration of free radicals with unpaired electrons. Possible sources include semiquinone polymer, hydroxyqtilnone, or condensed polynuclear hydrocarbons (Steelink and Tollin, 1967). Various degradation procedures have been used to separate complex molecules into monomeric components. Kumada and Suzuki (1961) and Cheshire et al. (1967) identified polycyclic aromatic compounds following alkaline permaganate oxidation. Hansen and Schnitzer (1969), on the contrary, obtained no polycyclic aromatic compounds but did obtain aliphatic carboxylic acids and all possible benzene carboxylic

PAGE 18

6 acids except benzoic acid. Ortiz de Serra and Schnitzer (1973) isolated and identified a number of phenolic acids. Hansen and Schnitzer (1967) used nitric acid oxidation and identified a series^ nitrophenols and aliphatic dicarboxylic phenolic, and benzenecarboxylic acids. # Based on degradative and nondegradative studies, Schnitzer has proposed a structure for fulvic acid consisting of phenolic and benzenecarboxylic acids joined by hydrogen bonds to form a polymeric matrix of considerable stability (Schnitzer and Khan, 1972). Numerous additional structures for humic and fulvic acids have been suggested (Burges, Hurst, and Walkden, 1964; Flaig, Beutelspacher and Reitz, 1975; Haworth, 1971). The complexity and diversity of these structures demonstrate the probable complexity of the total soil humic complex. Extraction of Soil Organic Matter Procedures for extraction of soil organic matter were reviewed by Mortenson (1965) and Stevenson (1965). Dilute aqueous NaOH is the most commonly used extractant of soil organic matter. Sodium hydroxide produces high yields of extractable organic matter; however, its use has been severely criticized due to chemical alterations which may occur in alkaline conditions (Bremner and Lees, 1949; Bremner, 1956; Choudhri and Stevenson, 1957). Bremner (1950) observed that O2 was adsorbed from the atmosphere by alkaline soil suspensions. Tinsley and Salam (1961) suggested that condensation reactions between amino compounds and aldehydes or phenolic compounds may result in formation of humin-type compounds during NaOH extraction.

PAGE 19

7 Loeppert and Volk (1974) investigated the yields and properties of extractable organic matter solubilized from Terra Ceia muck (Typic Medisaprist) by a series of organic and inorganic extracting solvents. Quantity of Organic material extracted with dimethylf ormamide (DMF) pyridine, and methanol were substantially increased when the soil ash content was lowered by dialysis prior to extraction. Yields obtained with less polar extractants (acetonitrile chloroform, acetone, and benzene) were not highly influenced by ash content. Choudhri and Stevenson (1957) and Bremner and Lees (1949) were able to significantly increase extraction yields with NaOH by pretreating the soil with 0.1 N HCl to lower the ash content and remove exchangeable cations. Extraction yields from Pahokee muck (Typic Medisaprist) obtained by single 24-hour extractions with 0.5 N NaOH (10:1 extractant : soil ratio) were increased from 23 to 40% following pretreatment with 0.1 N HCl (Snow, Loeppert, and Volk, 1974). Similarly, extraction yields of Terra Ceia muck were increased from 38% to 49% following treatment with 0.1 N HCl (Loeppert and Volk, 1974). When DMF was used as the extracting solvent, extraction yields were increased from 0.5% to 25% following pretreatment with 0.1 N HCl. In these studies, it was observed that the organic material extracted by DMF was similar in functional group and elemental analyses to that extracted by 0.5 N NaOH, and that H2O extracted a material with properties similar to those of fulvic acid. Loeppert and Volk (1974) concluded from the high yields obtained with DMF following pretreatment with 0.1 N HCl, the relative mildness of DMF, and the similarity in properties of DMFand 0.5

PAGE 20

8 N NaOH-extractable materials that DMF may be an excellent solvent for structural studies of the humic fraction. In general, organic solvents employed to extract soil organic matter have had limited use because yields were low and because a more specific fraction may be extracted than with NaOH. Organic solvents which have been investigated include acetylacetone (Halstead, Anderson, and Scott, 1966), anhydrous formic acid (Parsons and Tinsley, 1960), pyridine (Kessler, Friedel, and Sharkey, 1970), methanol (Mclver, 1962), acetone-H^OHCl (Porter, 1967), aqueous THF (Salfeld, 1964), and EDTA (Schnitzer, Shearer, and Wright, 1959). Gel Permeation Chromatography General Information The practice of high-pressure liquid chromatography is covered in a text by Kirkland (1974) and reviews by Zweig and Sherma (1974) and Gaylor, James, and Weetall (1976). Gel-permeation chromatography is a form of liquid chromatography in which molecules are separated according to size. The larger molecules are excluded from all or a portion of the solvent-filled pores of the packing material due to their physical size. On the other hand, a nonreactive small molecule may freely enter the pores of the packing material. Theory and N o menclature For discussions of chromatographic theory, the reader is referred to the text by Giddings (1965) and review articles by Biy (1970), Karger (1971), and Borabough (1971).

PAGE 21

9 Column parameters are defined in terms of V^, V^, and column efficiency, N. The total pore volume of the column, V^, is determined by the elution volume of a nonreactive low molecular weight material which freely enters the solvent-filled pores of the packing material. The interstitial volume of the column, V^, is determined by the elution volume of a nonreactive high-molecular weight material which is completely excluded from the pores of the packing material. Both andV^are determined with standard compounds. In practice, the condition of absolute nonreactivity between solute and packing would probably never be attained, since there is no such thing as a completely inert gel network (Freeman, 1973). Likewise, there is no such thing as an entirely inert solute in liquid chromatography. However, by proper selection of solute, column parameters and can be determined with a high degree of accuracy. Column efficiency is expressed by the theoretical plate count, N, which is determined with the equation, is the elution volume of the solute, and 0)^ is the peak width at the baseline. Column efficiency is influenced by particle diameter of the column, linear velocity of the solvent, and how well the column is packed (Karger, 1971; Dark and Limpert, 1973). Karger (1971) presents an excellent discussion of factors affecting resolution and column efficiency.

PAGE 22

10 Packing Materials The various packing materials currently available have been summarized by Dark and Limpert (1973), Kirkland (1974), and Laub (1974). They are generally divided into three major classes: (i) rigid gels or glasses, (ii) semirigid gels, and (iii) nonrigid gels. The gels comprising the first group, the rigid gels, are composed of porous silica and are suitable for high-pressure liquid chromatography. The semirigid gels (e.g. polystyrene-divinylbenzene) are highly cross-linked organic polymers, will not distort under pressure, and are, therefore, suitable for high-pressure liquid chromatography. The nonrigid gels (e.g Sephadex G-gels) are lightly cross-linked organic polymers and not suitable for high-pressure liquid chromatography since they will distort under pressure, resulting in an altered pore structure. The properties of the two porous silica packing materials used in this study, Porasil and Corning controlled pore glass (CPG) have been examined by Cooper and Barrall (1973) and Cooper, Bruzzone, Cain, and Barrall (1971), respectively. Porasil has a higher pore volume than CPG (Cooper and Barrall, 197 3) and therefore has higher to ratios. Electron microscope examination has shown that Porasil has a more heterogeneous pore structure than the CPG packings. Cooper and coworkers have shown both packing materials to be effective for macromolecular separation. Cooper and Barrall (1973) concluded that the heterogeneous pore structure of Porasil results in useful separations over a wider range of molecular sizes than CPG. They also concluded that the heterogeneous pore structure of Porasil precludes its use

PAGE 23

11 for studying theoretical proposals relating polymer elution characteristics to pore size dimensions. A third type of rigid gel, Bioglass, is manufactured by a process similar to the Corning glasses; however, it has an intentionally broad pore size distribution (Cooper and Bruzzone, 1973). The rigid gels exhibit severe adsorption properties (Cooper, Johnson, and Porter, 1973; Dark and Limpert, 1973; Spatorico, 1975) which are attributed to OH groups on the surface and Lewis acid sites present from the manufacturing process. Adsorption effects may be reduced through deactivation of OH groups; however, Lewis acid sites are not deactivated by these procedures (Dark and Limpert, 1973). Deactivation procedures include chemical treatment with polyethylene oxide (Hiatt et al. 1971; Hawk, Cameron, and Dufault, 1972) or diethylene glycol (LePage, Beau, and DeVries, 1968) and permanent deactivation by silyation with hexamethyldisilazane (Cooper and Johnson, 1969) and trimethylchlorosilane (Unger et al., 1974). Commercial Porasil packing material is chemically deactivated by adsorbed polyethylene oxide and distributed under the trade name Porasil X. Essentially all solvents are compatible with the porous silicas and glasses, except alkaline Solvents which will dissolve the silica (Dark and Limpert, 1973). Spatorico and Beyer (1975) observed strong adsorption to the porous glass of polymers containing cationic groups, and found that treatment of the glaSs with polyethylene oxide, or surfactants, was not successful in eliminating adsorption.

PAGE 24

12 Loeppert and Volk (1976) investigated the use of HPLC for molecular size fractionation of soil humic fractions on Porasil and Porasil X and observed adsorption and electrostatic exclusion phenomena which were highly dependent on solvent, saturating-cation, and concentration of excess neutral electrolyte. The polystyrene-divinylbenzene (DVB) gels, i.e. Poragel and Styragel, are widely used in the polymer and petroleum industries (Gaylor, James, and Weetall, 1976). Styragel and Poragel are not compatible with aqueous solvents, acetone, or alcohols (Dark and Limpert, 1973) and exhibit a high sensitivity to solvent polarity. Changes in solvent may result in significant changes in the amount of solvation and swelling of the gel matrix and altered pore-size distributions of the gel. Therefore, it is usually necessary to pack the gel as a slurry in the same solvent which is to be used as the eluting solvent. Edwards and Ng (1968) studied the elution of model compounds on polystyrene-DVB gels and observed an apparent adsorption of aromatic compounds to the gel matrix. Adsorption of compounds on polystyreneDVB usually caused pronounced tailing (Bergmann, Duffy, and Stevenson, 1971). Cogswell, McKay, and Latham (1971) separated the acidic concentrate of petroleum distillate, using methylene chloride as solvent, into four spectroscopically definable fractions and suggested molacular association of the more acidic fractions in this solvent. Sephadex has been widely used in studies of soil organic matter. Although Sephadex is a different type of gel than the materials used in these studies, a close examination of the material is in order

PAGE 25

13 since previous results may be useful in developing separation schemes and interpreting results with the newer gels. The reader is referred to an excellent review by Swift and Posner (1971) Sephadex G-gels have been shown to strongly adsorb aromatic compounds (Gelotte, 1960), heterocyclic compounds (Demetriou et al 1968) and phenolic compounds (Sommers, 1966; Brook and Housley, 1969) Gel-phenol affinity is related to the ether bonds in the crosslinking group rather than to the polysaccharide (dextran) component of the gel matrix (Determann and Walter, 1968). As degree of crosslinking of the gel was increased, affinity of phenol for the gel was also increased. Brook and Munday (1970) suggested that benzene derivatives are adsorbed onto hydroxyether cross-linking by H-bonds and that interaction of Sephadex dextran gels with monosubstituted phenols, anilines, and benzoic acids operates through hydroxyl, amino, and carboxylic groups, respectively. Gelotte (1960) observed that the Sephadex bed material contained a small amount of ionized groups, probably COOH groups, at concentrations of approximately 10 meq per gram of dry Sephadex. Aromatic amino acids were adsorbed to the bed material, basic amino acids were strongly adsorbed, and acidic amino acids were partially excluded from the gel. Demetriou et al. (1968) likewise found that aromatic compounds with COOH substituents were excluded from the gel beads when distilled water was used as the eluting solvent. the same compounds were adsorbed when columns were eluted with acid-salt solutions. Similar results were observed during the elution of soil humic acids (Posner, 1966). Humic acid

PAGE 26

14 was excluded from the gel matrix when eluted with water, due to the charge effect. Considerable adsorption was evident when dilute electrolyte was used as the eluent. It was concluded by Posner that it is not possible to select a concentration of electrolyte which would completely eliminate adsorption and exclusion effects. In studies of lignosulf onate Forss and Stenlund (1975) concluded that elution of this material on Sephadex may be infuenced by the following factors: (i) polyelectroly te expansion, (ii) ion exclusion, (iii) ion inclusion, and (iv) steric exclusion. The ion inclusion effect results from interaction of charged sites in the macroions which are excluded from the gel with charged sites in more permeable macroions. Elution of the lignosulf onate preparation with excess electrolyte resulted in reduced exclusion of the sample due to reduction in both the ion exclusion effect and the ion inclusion effect. In a study of simple electrolytes on Sephadex (Neddermeyer and Rogers, 1968), it was observed that peaks were badly skewed, with diffuse front and sharp trailing edges. These skewed peaks were attributed to the ion exclusion effect produced by ionic solutes and fixed charges in the gel. Swift and Posner (1971) noted that when humic samples were eluted with water and the concentration of solute was decreased, a greater percentage of sample moved into the excluded or near excluded region. This phenomenon was explained on the basis of double-layertheory and decreased suppression of charge at lower concentrations. Gel-solute interactions were categorized according to coulombic forces, caused by charged sites on gel and solute, and adsorption, caused by hydrophilic interaction. Coulombic interactions were most prevalent

PAGE 27

when distilled water was used as eluent and were reduced by adding electrolyte. Swift and Posner suggested that fractionation based solely on molecular weight can be achieved by using alkaline buffers containing large amino cations.

PAGE 28

MATERIALS AND METHODS Sample Pretreatment and Extraction The soil selected for study was the surface horizon (0-25 cm) of Terra Ceia muck, a Typic Medisaprist (Volk and Schnitzer, 1973). The salt content was lowered using a dialysis technique employed by Khan (1971) and refined by Loeppert and Volk (1974). Soil was poured as a slurry into one dialysis bag, and cation-exchange resin (Amberlite IR 120) was poured into a second dialysis bag. Both bags were placed in 0.1 N HCl and dialysis was continued until ash contents were lowered to less than 1.0%. Recharged resin was placed daily in the dialysis chamber. Following dialysis, samples were air-dried. The extraction procedure is outlined in Fig. 1. Soil extracts were obtained by 24-hour treatments with the appropriate extractant (10:1 extractant :soil ratio). Extracts were centrifuged for 2 hours at 16,300 X gravity (G) filtered through Whatman #42 filter paper, and purified as described below. Initial studies were performed to determine the yields and properties of soil organic matter extracted from the Terra Ceia muck surface horizon with selected solvents. The experimental procedures and results of these studies are reported elsewhere (Loeppert and Volk, 1974; Snow, Loeppert, and Volk, 1974). 16

PAGE 29

17 EXTRACT DMF EXTRACT IIUMIC FULV IC ACID ACID SOXHLET SCHEME (lEXANE B E N Z E fi E V ETHYL ACE TATE ACETOIIE 2-PROPANOL METHANOL Fig. 1 Extraction scheme

PAGE 30

18 The DMF-extractable material was evaporated to dryness under vacuum at AOC with a rotary evaporator and suspended in deionized water. The sample was transferred to dialysis bags and dialyzed against deionized water for 6 hours with frequent changing of the external dialysis solution, against 0.1 N HCl in the presence of strong-acid ion-exchange resin (Loeppert and Volk, 1974) for A8 hours, and against deionized water until the external dialysis solution gave a negative test for CI This procedure was repeated until a sample with constant nitrogen content and less than 0.5% ash was obtained. The sample was lyophilized and stored at OG. The water extract was treated similarly to the DMF extract except the initial dialysis with water was omitted. The dialysis procedure was repeated until the sample contained less than 0.5% ash. The NaOH extract was acidified to pH 7.0 with 6.0 N HCl and concentrated under vacuum at 30C with a rotary evaporator. The sample was purified, lyophilized, and stored using the same procedure as with the DMF extract. A separate fraction of the NaOH extract was separated into humic and fulvic acid fractions by adjusting the pH to 2.0 and purified according to the procedure outlined by Stevenson (1965). Samples were further purified by the dialysis procedure to an ash content less than 0.5%, lyophilized, and stored at OC. Soxhlet fractions were obtained according to the scheme outlined in Fig. 1 by successive 48-hour extractions with each solvent in the series. Extracts were concentrated under vacuum at room temperature.

PAGE 31

dried under a dry-nitrogen jet, and redissolved in the extracting solvent at room temperature. Samples were concentrated to 30-ml volume and following, addition of 3Q ml of H2O, were reconcentrated to 30-ml volume. The reconcentration procedure was repeated several times, and the aqueous suspensions were transferred to dialysis bags for purification to less than 0.5% ash. Purified samples were lyophilized and stored at OC. Solubility Studies The solubility behavior of extractable organic matter and each of the Soxhlet fractions was determined by placing 2.00 mg of the organic material in 2 ml of the appropriate solvent. Following agitation the mixture was visually observed to determine whether the sample was insoluble, partially soluble or completely soluble. Where appropriate, the pH of the sample suspension was adjusted to the desired level by addition of acid or base which contained the same counter Ion as the excess electrolyte. Analytical Determinations Carbon and H were determined with the Coleman C-H analyzer, N by the micro-Kjeldahl method, and S using the Leco induction furnace. Oxygen was determined by difference with the assumption that C, H, N, S.and 0 were the only elemental constituents of the extractable organic matter. Total acidity was determined by addition of excess Ba(0H)2 and back-titration of unreacted BaCOH)^ with HCl to pH 9.8 (Schnitzer

PAGE 32

20 and Gupta, 1965; Schnitzer and Khan, 1972). Total COOH-group concentration was determined by addition of excess Ca(C2H^02)2 and backtitration of excess C2H^02~ with HCl to pH 8.4 (Schnitzer and Gupta, 1965; Schnitzer and Khan, 1972). Total OH was determined by an acetylation procedure (Brooks, Durie, and Sternhell, 1957) and total C=0 by the oximation method (Fritz, Yamamura, and Bradford, 1959). Phenolic OH group concentration was calculated as the difference between total acidity and COOH-group concentration, and alcoholic OH concentration was estimated by subtracting phenolic OH concentration from total OH concentration. All elemental and functional group analyses were corrected for moisture content and ash content. Moisture content was determined by heating a preweighed combustion boat with a known quantity of material to llOC for 24 hours and weighing. Ash content was determined by heating a preweighed combustion boat with a known quantity of organic material to 700C for 4 hours and weighing. All weights, yields, and concentrations are reported on a moistureand ash-corrected organic matter basis. Infrared patterns were obtained on a Perkin-Elmer model 127 spectrophotometer. The KBr pellets were prepared by mixing 0.8 mg of organic material, previously dried 24 hours over P^O^ in a vacuum desiccator, with 200 mg of dried KBr in a Wig-L-Bug amalgamator. The mixture was dried an additional 12 hours in a vacuum desiccator over ^2*^5 P^^^ ^ obtaining the spectrum. Continuousand stepped-potentiometric titrations were performed on 10 mg of organic material placed in the appropriate solvent. Titrant

PAGE 33

21 was added using a Radiometer autoburette ABO 13 With TTT 50 titrator module. Potentials were determined with a Radiometer PHM 64 pH meter and recorded using a Radiometer REC strip chart recorder with REA 160 titrigraph module. A Radiometer combination electrode with porous plug liquid junction was used for all potentiometric determinations. The calomel cell was filled with saturated KCl in for aqueous titrations or with saturated KCl in CH^OH for titrations in nonaqueous media (Loeppert, Zelazny, and Volk, 1976). High Pressure Liquid Chromatography The Waters' ALC 202 liquid chromatograph equipped with the Model 6000 solvent delivery system and 401 differential ref ractometer and UV detectors was used for all separations. Samples were injected through the U6K universal liquid chromatograph injector. A solvent flow rate of 0.5 ml/minute were used, unless otherwise specified. Column effluent was monitored by differential refractive index or by UV absorption at 254 or 280 nm, and recorded continuously on a Perkin Elmer 201 strip-chart recorder. Samples were dissolved in the eluting solvent to give a 0.1% concentration (W/V) and centrifuged at 35,000 x G for 20 min. Injection volume was 10 yl. Solvents used in the chromatographic studies were deionized water and spectroquality methanol, 2-propanol, t-butanol tetrahydrof uran (THE), dimethylformamide (DMF) ethy lacetate acetone, and chloroform.

PAGE 34

22 Salt solutions in water and methanol were prepared by adding the desired quantity of 2.0 N NaOH, tetramethy lammonium hydroxide, or tetrabutylammonium hydroxide to the solvent and adjusting to the appropriate pH with HCl, HNO^, H^SO^, or H^PO^ Halide salts were avoided where possible, since these salts may result in corrosion of the stainless steel surfaces of the liquid chromatographic pump, connecting lines, and columns. In the few cases where halide salts were employed, the pH of the eluting solvent was maintained above 7.0, and the chromatographic system was flushed with copious quantities of deionized water following use of the halide salt. Characteristics of the packing materials used in this study are summarized in Table 1. The silica and porous glass packing materials were dry-packed into stainless steel columns (0.318 cm CD x 2 f or 0.318 cm CD X 1 m) The semirigid gels were slurry packed in the same solvent as the eluting solvent. Poragel columns (stainless steel, 0.954 cm OD x 2 f) packed in THF were used as obtained from the manufacturer. Poragel in DMF was slurry packed into 0.954 cm OD x 2 f stainless steel columns. The y-Styragel columns (stainless steel, 0.954 cm OD X 1 f) packed in THF were used as obtained from the manufacturer. When used in series, columns were connected with U-shaped 0.QQ9 in. ID tubing. A series of experiments were performed to investigate the elution behavior of extractable soil organic matter on the gel permeation packing materials: (i) determination of optimum operating conditions.

PAGE 35

to S-l CU 4J 0 0 4-) QJ 4: 0 0 E M 0) 0 0 0 0 0 0 CU -H M 0 0 0 0 0 0 in 4-1 0) d 0 0 0 0 0 0 0 0 0 CO 5 CO 0 0 0 0 0 0 0 cn (3 0 0 0 0 0 0 0 0 •H S-l 0 uo 0 0 0 0 0 0 0 0 V (Ti Oi, vO CN — 1 ^ 0 0 M d 1 1 1 1 1 1 0 1 0 1 0 t) 0 0 0 0 0 tH 0 CN 0 cn tH d a. u ^ 0 0 0 0 0 1 0 1 0 1 1 CO 0) s-l 0 0 0 0 0 0 0 0 0 0 C < .-1 0 0 0 0 0 0 0 S tH tH .H tH tH U~l in LO tH iH rH tH E • 0 CU s-l pt4 CO XI 0 0 c3 0 0 0 0^ e3 H z OJ iH tn CO iH 00 e iS u ro — 1 > e ro ro ro CO cn cn CN CN rH 0 to 0 iJ • to tH S-l (u a) 0) CN CN CO -i E S-l > E 1-H .H rH rH rH tH ,H tH rH tH >, CO iJ -H >^ 4-1 CO x) to 1 3 d rH S 0 T3 3 a. • d ^ dco 00 tu tu s-l 0 0 •H •H CO a U 4-1 0 0 tu o •H t-i tH T) a) tH 1 X to 0 x; CO i-i •H •H di c SO 4) Oi 0 0) 4-1 D. tx Cfl CO 4-1a) d •H c H CO rH TD tu 60 13 0 0 d d S-l CO, CO > tl) M a d O' W E u 3 ^ 3 •H X : rd — Z 0 -H v0 tu cn 0 0 4J U tn 4-1 rH r u XI r— 1 (U s-l s-l U QJ 3 CO CU tH to tu O Q 0 0 CO >, ^-1 >^ S-l N to u dl rH U tH 0) -H H to t) 0 0 4-] 1^ 0 0 14-1 G. tu u O E d d d Cu E 0 0 X cn tu 3 •H •H tu u C .H 4-) 4-1 0) 0 0 3 3 4-1 u tH rH 0 0) f— 1 0 < 0 < tu tu d e CO tu CO •H 0 0 CO • >^ H s-l S-l 0 < 0 < 0 0 XI XI 0 CO (L) rH in tn rH •r^ O. 4-1 0 0 d tu X) XI Uh CO ><; 0 0 tH tH E > tu tu C4-1 E < u W <; U W rH tn tu tu 3 -H d d tu 60 00 tH 4-) H •H tH CO r-l .H iH tH .H tH 0 ^ iH CO CO 0 u E E d d •H •H H •H H -tH 0 u-i 0) cu S-l S-l u tu Sj s-l E tu •H cn cn cn to CO CO , 0) tu 3 .H CO CO CO CO CO CO CO CO 4J u r-H to u 4-1 tH J3 CJ S-l s-l u S-l S-l s-l 0 S-l s-l cn CO rH tu tu tu 0 CO CO 0 0 0 0 0 0 (X, P-i 0 0 1 < S-l Q Q u H Pi D-i Ph Pu Cu a U o. Oi 33. to rO u X)

PAGE 36

24 (ii) determination of column parameters, (iii) elution behavior of soil organic extracts and organic standards with selected solvents, (iv) elution behavior of soil organic extracts and organic standards as influenced by saturating cations, and (v) elution behavior of soil organic extracts and organic standards as influenced by excess electrolyte.

PAGE 37

RESULTS AND DISCUSSION Chemical Characteristics of Extractable Organic Matter Extraction yields of 25% and 49% were obtained by single treatments of the surface horizon of Terra Ceia muck with DMF and 0.5 N NaOH, respectively, following pretreatment with HCl to lower the ash content (Table 2). The Soxhlet solvents were able to solubilize 29.6%, 35.0%, and 9.4% of the NaOH-extractable material, DMF-extractable material, and Terra Ceia muck, respectively. Only minor portions of the materials were extracted by hexane and benzene. The major portions were in the acetone, 2-propanol, and methanol • fractions Each of these fractions, however, represent only a minor portion of the total soil organic matter. Obvious and important differences exist in the elemental and functional group concentrations of the extractable organic matter (Table 3). Comparisons of the Soxhlet fractions indicate that C and H content decreases, 0 and N content increases, COOH content increases, and total acidity increases according to the solvent order: hexane, benzene, ethylacetate, acetone, 2-propanol, and methanol. Potentiometric titrations of the Soxhlet fractions in DMF indicated that titratable acidity increased according to the same solvent order (Table 4). 25

PAGE 38

26 ^1 OJ 4-J 4-J re e c re M u o o re 4-i u re M 4-J X -i cn tJ 4-) re QJ X 4J w re E QJ re 4-1 0 re •H u c 4J re X txO cn QJ •H 1 0 w re re •H z 0 00 c QJ > O m M C o O o o 0^ C o 00 0-) 0 0 0 0 GO ^£1 in 00 a^ t-H H 00 ^ O O 0-) 1-H rH a^ 0-1 0 0 Csl CM 00 rH CN 0 CO o re z Ch Q QJ D 4-1 -0 QJ QJ •H 4J -H CO re 4-1 ol xh re QJ c 1— 1 0 QJ U QJ re 0 CO 4-1 Q) c re c a c QJ C QJ rH 0 0 re 1— ( rH re N 4-1 u re rC X c QJ Pu 4J 4-1 X QJ QJ 4-1 U 1 0 0 P3 w < rsi H CO

PAGE 39

o H rH o O O o o •H rH O C 0) o o II u o o u rH 4-1 TO •U TD O -H H O TO u-i CNl • I M I I I I n J un o^ O ro O • I I I .... c^ I I I rH i^g c-j cN a> rH Lo in •III .... I I I ro . or^ ^ 00 • • • •HvCrOOOOrHrHCO -^rsl CN r^rHfOLn^-^rH<^cT^ COOOOOrHrHCO r^rH. 00 CM o in r-^ incjNoooooor^vDin in e rH 3 3 rH 4-J X Q) X rH 4-J 4-1 0 4-1 X QJ 4-1 QJ CO QJ X rH Q) rH 1 rH 0 X rH X QJ X CO X 4J X X 4_J X 1 0 u 4-1 X 0 TO 0 rH CO TO U o CO 4-J CO 1 u TO 1 QJ 1 C rH 4-J U 1 QJ 0 QJ TO 0 QJ X 4-1 QJ C TO C a. C 3 Q) X c QJ rH 0 0 TO QJ TO N 4J u X -H u X C X 0) CL, 4-1 m QJ :r: QJ QJ 4-J u QJ Q) 4-J o 3: pa w
PAGE 40

28 Table 4. Titratable acidity of extractable soil organic matter Solvent medium Sample DMF -meq / 1 DMF extract 5.9 Hexane-Soxhlet 1.4 — Benzene-Soxhlet 1.9 Ethylacetate-Soxhlet 2.9 Acetone-Soxhlet 4.1 2-Propanol-Soxhlet 4.9 2.3 Methanol-Soxhlet 5.7 2.4 Residue 5.8 4.3 Humic acid 6.3 4.0 Fulvic acid 7.3 5.6 Water extract 6.9 5.3 NaOH extract 6.0 Hexane-Soxhlet 1.1 Benzene-Soxhlet 1.7 Ethylacetate-Soxhlet 2.5 Acetone-Soxhlet 4.2 2-Propanol-Soxhlet 4.8 2.0 Methanol-Soxhlet 5.3 2.6 Residue 5.7 39

PAGE 41

29 Infrared patterns of the Soxhlet fractions (Fig. 2) substantiated differences evident in the elemental and functional group analyses. From the least polar to the most polar extracting solvents, Soxhlet extracts showed progressively weaker C-H stretching vibrations in the 2900 cm region. Absorption at 1450 cm' ^ due to the CH^ bending vibration also decreased according to the same solvent order. These adsorption bands of the infrared patterns, along with the C and H concentrations, indicated that the less polar solvents extracted a material of greater aliphatic character. Each of the Soxhlet fractions had strong C=0 stretching bands at 1725 cm and 1630 cm"""" resulting from COOH and C00-, respectively. The band at 1630 cm""*" was considerably stronger in the absorption spectra of materials extracted by the more polar solvents. The band at 700-750 cm"''', evident in the absorption spectra of hexane-, benzene-, and 2-propanol-extractable materials, was attributed to aromatic C-H out-of-plane bending vibrations. The absence of bands in this region of the spectrum indicates either the absence of aromatic structure or the possibility of a completely substituted aromatic ring system. The weak absorption at 700-75,0 ciii"^ for the methanol-Soxhlet extracts compared to the hexane-, benzene-, ethylacetate-Soxhlet extracts may be attributed to a highly substituted aromatic ring system. Each of the Soxhlet-extractable materials showed a strong absorption band at 3440 cm ^ attributed to H-bonded OH groups and a weaker band at approximately 1250 cm ^ attributed to C-0 stretching vibrations. Absorption at both frequencies increased with increasing polarity of the extracting solvent. The hexaneand benzene-Soxhlet fractions, especially, showed significantly weaker absorption in the 3440 cm""'" region than fractions extracted by the more polar solvents.

PAGE 42

30 Fig. 2 Infrared patterns

PAGE 43

31 DMFand NaOH-extractable organic matter, and the humic acid fraction had very similar elemental compositions and concentrations of 0-containing functional groups. The same observation was previously made by Loeppert and Volk (1974) in comparisons of DMFand NaOH-extractable materials. These similarities were corroborated by the infrared patterns (Fig. 2). The fulvic acid fraction had lower C and H contents, higher 0 content, lower N content, higher total acidity, and higher COOH content than the humic acid and DMF-extractable materials. The higher acidity of fulvic acid compared to the other materials was corroborated by the significantly higher titratable acidity, as determined by potentiometric titration in DMF. In summary, C and H content decreased, aliphatic C-H decreased, 0 and COOH contents increased, and total acidity increased according to the following order of extractable organic matter: hexane-Soxhlet extract, benzene-Soxhlet extract, ethylacetate-Soxhlet extract, acetoneSoxhlet extract, 2-propanol-Soxhlet extract, methanol-Soxhlet extract, NaOH extract— DMF extract— humic acid, fulvic acid. Solubility Characteristics of Extractable Organic Matter Solubility characteristics of extractable organic matter in selected solvents and in salt solutions are summarized in Tables 5-7. Humic acid, and DMFand NaOH-extractable organic matter were completely soluble at 0.1% concentration only in DMF, dimethylsulf oxide (DMSO) and

PAGE 44

32 Table 5. Solubility of extractable soil organic matter in selected solvents at 0.1% concentration Solvent Sample MethylMethylEthylethyl isobutyl HfeKane Benzen e acetate Acetone ketone ketone DMF extract l' Hexane-Soxhlet s' Benzene-Soxhlet P Ethylacetate-Soxhlet I Acetone-Soxhlet I 2-Propanol-Soxhlet I Methanol-Soxhlet I Residue I P P S S S P I I P P S S S P P 1 P P S S S P I I P P S S S P I I Humic acid Fulvic acid P I Water extract NaOH extract IIP Hexane-Soxhlet S S P Benzene-Soxhlet P S S Ethylacetate-Soxhlet IPS Acetone-Soxhlet IIS 2-Propanol-Soxhlet IIP Methanol-Soxhlet III Residue III P P S S S p p I p p S s s p I I p p s s s p I I S = soluble, P = partially soluble, I = insoluble

PAGE 45

33 Table 5 (Extended) Solvent Ethyl ether TUT? 1 rir t-Butanol 2-Propanol Methanol Pyridine DMF DM so H^O I p P P P P S S p P s P I I 1 P P I S s P I I P S P I P s S I I S S S I P s s P P S S s I I s s S S s S s I T i F ft P P S s S s p I I I I P p s s p I P p P P p s s p I P p P s s s s s I P p P p p s s s I P p P p p s s p P S p I I I p p I s s p I I p s p I p s s I 1 s s s I p s s P p s s s I I s s S s s s s I I p p P s s s s p I I I I p p s s p

PAGE 46

34 Table 6. Solubility of extractable soil organic matter as influenced by saturating cation and solvent Sample Saturating cation Solvent H^Q CH^OH 2-Propanol THF DMF DMSO DMF extract DMF extract acetone-Soxhlet DMF extract 2-propanol-Soxhlet DMF extract methanol-Soxhlet Humic acid Fulvic acid H Na N(CH ) H Na N(CH ) H Na N(CH ) H Na N(CH ) H Na N(CH ) H Na N(CH ) P S s s I s s s p s s s p s s p s s s s p s s s s s s s p s s s s s s s p s s s p s s s p s s s I p p s p p p s p p p p p p p p p p p p p p p s p p p s p p p s p p p s p p p s p p s s p p s s p p s s p p s s p p s s p s s s p s s S = soluble, P = partially soluble, I = insoluble

PAGE 47

35 Table 7. Solubility of fulvic acid in aqueous salt solutions Saturating cation Excess electrolyte pH Concentration 2.0 4.0 6.0 8.0 10.0 Na Na„SO. z 4 0.000 N 0.001 N 0.01 N O.03 W 0.10 N S S S s p s s s p s s s s K„SO. z 4 0.000 N 0.001 N 0.01 N 0.05 N 0.10 N S S S P P N(CH^)^ [N(CH^)^: 2^^ 0.000 N O.OQl M 0.01 N 0.05 N 0.10 N [N(C^Hg)^]2S0^ 0.000 N 0.001 N 0.01 N 0.05 N 0.10 N S = soluble, P = partially soluble

PAGE 48

36 0.5 N NaOH. The DMF and DMSO both have significant basic character (Talhoun and Mortland, 1968). The acidic organic material is highly dissociated and dispersed in each of these solvents and, therefore, is soluble. Each of the Soxhlet fractions and fulvic acid were also completely soluble at 0.1% concentration in DMF and DMSO. Fulvic acid was soluble in methanol and water, in addition to DMF and DMSO, but was not completely soluble in any of the other solvents. The solubility of exfcractable organic matter and Soxhlet fractions was influenced to a great extent by the saturating cation (Table 6). Exchange of by Na^, k"^, N(CH^)^^, or C(C^Hg)^^ resulted in increased solubility of the organic solutes in water, methanol, or 2-propanol, and decreased solubility in DMF. For example, at 0.1% concentration, the H-saturated DMFand NaOH-extractable materials were only partially soluble in water, methanol, or 2-propanol; however, the salt-saturated solutes were completely soluble. On the other hand, the H-saturated 2-propanol-Soxhlet fraction and fulvic acid were soluble at 0.1% concentration in DMF whereas the salt-saturated material was only slightly soluble. These solubility characteristics greatly limit the solventelectrolyte combinations which are applicable for exclusion chromatography. The enhanced solubility of the cation-saturated samples in the protic solvents (water, methanol, and 2-propanol) may be attributed to acidic properties of these solvents (King, 1973) which promote stabilization of the solute anion. The very weakly acidic dipolar aprotic solvents (e.g. THF, DMF) would not stabilize the solute anion to as great an extent as the protic solvents.

PAGE 49

37 The presence of excess neutral salt affected the solubility of fulvic acid in water (Tiible 7) and methanol. Some very interesting trends were evident in these studies. The H-, Na-, K-, N(CH ),-, and 3 4 ^^^4^9^4~^^^^^^'^'^'^ fulvic acid samples at 0.1% concentration were soluble at pH 2.0 in all concentrations of excess neutral electrolyte -3 -2 up to 0.1 N. In the presence of 10 N or 10 N excess salt, 0.1% fulvic acid remained completely dissolved as the pH was increased successively to pH 4.0, 6.0, 8.0, and 10. 0. However, in the presence of -2 5 X 10 N excess salt, K-saturated fulvic acid began to precipitate at pH 4.0. The sample redissolved at pH 8.0 and was completely soluble as the pH was increased to 10.0. Fulvic acid saturated with N(CH^)^"^ or N(C^Hg^"^ remained completely dissolved as the pH was increased from 2.0 to 10.0, in the presence of 5 x 10 N excess neutral salt. As the ionic strength was increased to 5 x 10~^ N, however, K-, Na-, N(CH ) -, 3 4 and N(C^Hg)^-saturated fulvic acid precipitated as the solution pH approached 4.0 and redissolved as the pH approached 10.0. The precipitation was greatest in the approximate range of pH 4 to pH 7 and may be attributed to unfavorable conditions for the electrostatic dispersion of molecular units. It is interesting to note that greatest precipitation occurred within the pH range at which greatest neutralization of acidic carboxyl groups would occur. The precipitation phenomena in the presence of excess neutral salt greatly limits the conditions which may be employed for gel permeation separations of extractable organic matter.

PAGE 50

38 High Pressure Liquid Chromatography Porous Silica Packing Materials Op6ratiftg conditions Column efficiencies of Porasil A and Porasil AX packing materials were greatly influenced by solvent flow rates. With 0.318-cm OD analytical columns, maximum column efficiencies and minimum peak broadening were obtained at flow rates of approximately 0.1 ml/min. ; however, column efficiencies were not significantly different at flow rates between 0.1 and 0.6 ml per min. (Fig. 3). At flow rates greater than 0.6 ml per min., peak broadening was increased and column efficiencies were decreased. For this reason, it was concluded that low flow rates should be maintained with the Porasil packing materials It is interesting to observe that for the Porasil packing materials, there was a slight increase in column efficiency as flow rate was decreased to 0.1 ml per min. (Fig. 3). These results may be compared with those obtained with the CPG packing materials for which maximum column efficiencies were obtained at flow rates of 0.4 ml per min. Decreases in column efficiency were observed when flow rates were decreased or increased from this value. As with the Porasil packing material, increases in solvent flow rates above 1.0 ml per min. resulted in peak broadening and significant decreases in column efficiency. The different behavior of the Porasil and CPG packing materials at low flow rates may be at least partially attributed to the more

PAGE 51

39 700 r 600 500 ~ 400 300 H 200 LiJ o PORASIL A A PORASIL AX CPG-250 1.0 2.0 3.0 4.0 FLOW RATE, ml/min 5.0 Fig. 3 Effect of flow rate on column efficiency, N, of Porasil A. Porasil AX, and CPG-250 analytical columns

PAGE 52

40 uniform pore structure of the controlled pore glass material which may result in reduced blockage of large pores by small pores and less restricted diffusion of solute molecules through the gel matrix. The more uniform pore structure may allow the use of higher flow rates. Cooper and Barrall (1973) suggested that "pooling" or solute restriction in porous silica media, necessitates the use of low flow rates with these materials. Based on these studies, solvent flow rates of 0..5 ml per min. were selected for all subsequent studies using the 0.318-cm OD columns with the Porasil and CPG packing materials. Based on similar studies with 0.954-cm OD preparative columns, solvent flow rates of 1.5 ml per min. were used for all subsequent studies on these columns packed with Porasil or CPG packing materials. The effects of sample size on column efficiencies of Porasil and CPG packing materials are summarized for the analytical columns (Fig. 4). In general, the maximum sample volumes were 20 yl for the analytical columns and 100 pi for preparative columns. Larger sample volumes resulted in increased peak broadening and reduced apparent column efficiencies. Column efficiencies were not noticeably affected with lower sample volumes. Even though column efficiencies are reduced with large sample volumes, column overloading may be helpful in obtaining preparative fractions, especially when used in conjunction with recycle chromatography (Bombaugh, 1971).

PAGE 53

41 700 600 i 500 o o \Z 400 -J CL < 300 o LU g 200 UJ H100 o PORASIL A A PORASIL AX D CPG -250 J I L_i_ i J L J 0.100 SAMPLE SIZE, ml 0.200 Fig. 4 Effect of sample size on column efficiency, N, of Porasil A. Porasil AX, and CPG-250 analytical columns

PAGE 54

42 Column paramete rs^. Elation characteristics of packed columns are summarized in Table 1. Column parameters, and V^, were determined from the elution volumes for acetone or benzene and blue dextran 2,000 or 2,600,000 molecular weight polystyrene, respectively. In all cases the column efficiencies, indicated by theoretical plate count, N, decreased with increasing internal pore size of the packing material. For example, the column efficiencies of Porasil AX, CX, and EX were 340, 280, and 230 theoretical plates per meter, respectively. Molecular weight calibration curves (Fig. 5) were obtained by elution of polystyrene standards with THF. The approximate molecular weight working ranges, as determined with the polystyrene standards are summarized in Table 1. The working curves obtained with polystyrene standards on the Porasil and the CPG packing materials were not linear over the working range of the gels. Effect of solve n_t. Peak elution volumes of extractable organic matter on Porasil A, Porasil AX, and CPG-250 packing materials are summarized in Tables 8-10, respectively. Elution patterns of the 2-propanol-extractable material and fulvic acid are shown in Figs. 6 and 7. Samples were completely soluble at 0.1% concentration in each of the solvents shown. A portion of the fulvic acid sample (Fig. 7) was eluted at V^, the elution volume of a nonreSctive high molecular weight solute, on the Porasil A column when methanol, water, or DMF was used as the eluting solvent. The relative quantity of sample eluted at increased according to the following solvent order: methanol < DMF < H 0. Likewise, portions of the acetone2-propanol-, and methanol-Soxhlet

PAGE 55

43 Fig. 5 Molecular weight calibration curves of 1 m x 0.318 cm OD Porasil AX and CPG analytical columns obtained by elution of polystyrene standards with THF

PAGE 56

44 Table 8. Peak elution voluems of extractable soil organic matter on Porasil A with selected solvents Solvent Sample hTo CH OH 2-Propanol t'-Butanol Acetone THF DMF extract Acetone-Soxhlet — 1.61 (3.18)^ 2-Propanol-Soxhlet — 1.58 (3.17) Methanol-Soxhlet — 1.61 (3.19) ml 1.75 1.57 1.65 AD^ 3.29 1.83 (3.19) (3.30) (3.17) 1.55 — — 3.32 1.79 (3.15) 3.31 1.82 NaOH extract Acetone-Soxhlet — 1.63 (3.17) 2-Propanol-Soxhlet — 1.60 (3.16) Methanol-Soxhlet — 1.60 (3.19) 1.78 1.58 1.62 AD 3.31 1.74 (3.18) (3.32) 1.57 — — 3.32 1.75 (3.13) 3.35 1.76 Humic acid — — — — — — 1.75 Fulvic acid 1.57 — — — — — 1.76 Parentheses ( ) indicate secondary peak. '^AD = severe adsorption.

PAGE 57

45 Table 9. Peak elation w
PAGE 58

46 Table 10. Peak elution volumes of extractable soil organic matter on CPG-250 with selected solvents Solvent Sample CH^OH 2-Propanol t-Butanol Acetone THF DMF u,l DMF extract — — — — — — 1.77 Acetone-Soxhlet ~ 1.75 3.73 3.97 AD 3.87 1.82 AD^ AD 2-Propanol-Soxhlet — 1.70 3.71 — — 3.85 1.75 Methanol-Soxhlet — 1.71 — — — — 1.77 NaOH extract — — — — — — 1,79 Acetone-Soxhlet ^1.76 3.84 4.12 AD 3.81 1.81 2-Propanol-Soxhlet — 1.71 3.67 — — 3.83 1.76 Methanol-Soxhlet — 1.69 — — — — 1.79 Hutnic acid — — — — — — 1.76 Fulvic acid 1.71 1.74 — — — — 1.72 ^AD = severe adsorption

PAGE 59

47 UJ IS) o m UJ cc LU Q q: o o lU DMF \ /' ) f 1 / I \ 1 \ 1 \ 1 \ 1 CH30H ^ / / N / ^ 1 _> CHsOH THF \J s THF — — — — — 1_ -PORASIL A \j -PORASIL AX 1 ,1 Vo 2 3 Vt ELUTION VOLUME, ML Fig. 6 Elution patterns of the 2-propanol-Soxhlet extract of DMF-extractable material on Porasil A and Porasil AX with selected solvents UJ CO z o CL (fi UJ cr c: UJ a cr o o U-l ir H20 \ \ /''' H20 V, \ ^ CH30H ^ '' I V / ^ V ( \ 1 I 1 1 1 V 1 t t 1 t 1 / CH30H — PORASIL A — PORASIL AX 1 Vo 2 3 Vt ELUTION VOLUME. ML Fig. 7 Elution patterns of fulvic acid on Porasil A and Porasil AX with selected solvents

PAGE 60

48 extracts (Table 8;Fi2s. 6 and 7) were eluted at when methanol or DMF was used as the eluting solvent. A relatively larger quantity was eluted at with DMF than with methanol. With each of the above solvents, the organic solute was eluted prior to V^. On the contrary, when acetone, t-butanol, or THF was used as the eluting solvent with the Porasil A column, a portion of the organic solute was eluted past V^, indicating an adsorptive interaction with the silica packing material Deactivation of the Porasil surface (Porasil X) resulted in reduced exclusion of organic solute from the gel matrix in water and methanol compared to the activated material (Figs. 6 and 7; Table 9). Adsorption was reduced on the Porasil AX compared with the Porasil A packing material, although a small portion of the organic solute was still eluted past with t-butanol, acetone, and THF on Porasil AX. Elution volumes of organic standards on Potasil A, Porasil AX, and CPG-250 are shown in Tables 11-13, respectively. On Porasil A, several of the organic acid standards (1 2 4 5-tetracarboxybenzene and 1,3,5-tricarboxybenzene) were eluted at a solvent volume equivalent to when DMF, water, or methanol was used as the eluting solvent and at a solvent volume slightly greater than when acetone, 2-propanol, t-butanol or THF toas used as the eluting solvent. In water, methanol, and DMF, the more highly substituted aromatic acids were in general eluted at a smaller solvent volume than the less substituted acids. For example, elution volume increased according to

PAGE 61

49 Table 11. Peak elution volumes of organic standards on Porasil A with selected solvents Solvent Sample hTo cIT^OH 2-Propanol t-Butanol Acetone THF DMF ml 1,2,4,5-Tetracarboxybenzene 1.73 1. 72 3. 70 9.25 3. 55 2.36 1,3, 5-Tricarboxybenzene 1.76 1. ,94 AD^ 3 68 3. 54 3.60 3 5-Dihydroxybenzoic acid 2. ,08 4.87 3.33 3. 47 3.43 Benzoic acid 2. .37 5.29 3.51 3. 59 3. 58 3.44 Pyridine 5. ,20 AD AD 4. 45 3. ,87 3.58 Aniline 3. .4.4 4.74 4.04 3. 55 3. .62 3.41 Methanol 3.64 3.94 3. 91 3. .42 3.94 Ethylene glycol 3, .40 4.45 4.76 3. 97 3. .45 3.46 Acetone 3, .36 3.97 4.52 3, .70 3.46 AD = severe adsorption

PAGE 62

50 Table 12. Peak elution volumes of organic standards on Porasil AX with selected solvents Solvent Sample H„0 CH^OH 'T^^'opanoT t^uTanol Acetone THF ml 1,2,4, 5-Tetracar boxybenzene 1 .83 2 .58 10 .25 12.00 3 .48 1,2, 5-Tricarboxybenzene 2 .11 2 .71 4, .95 3.49 3 .39 3, 5-Dihydroxybenzoic acid 2 .53 2 .85 3, .59 3.28 3 .46 Benzoic acid 2, .39 2, .97 3, .77 3.40 3.49 3 .34 Pyridine 11, .06 3, .42 3, .67 AD^ 3.84 3, .60 Aniline 4, .69 3. ,38 3. ,52 3.55 3.52 3, .49 Methanol 3. .48 3. ,44 3.41 3.65 3, .71 Ethylene glycol 3. ,56 3, .43 3. ,52 3.56 3.65 3, ,40 Acetone 3. ,58 3. 38 3. 44 3.46 3, .40 AD = severe adsorption

PAGE 63

51 Table 13. Peak elution volumes of organic standards on CPG-250 with selected solvents Solvent Sample H„Q CH-,OH 2-Propanol t-Butanol Acetone THF DMF ml 1,2,4, 5-Tetracarboxybenzene 1.71 1, ,74 4.12 10.03 3. 91 2. 44 1 3, 5-Tricarboxybenzene 1.78 1, .93 AD^ 4.27 3. ,93 3. 38 3, 5-Dihydroxybenzoic acid 2, .18 5.03 4.06 3, ,91 3. 86 Benzoic acid 2. ,26 5.36 4.99 3, ,98 3. ,92 3. 88 Pyridine 4, .89 AD AD 5. ,02 4. ,34 4. 17 Aniline 4, .12 4.72 4. 76 4. ,17 4. ,10 3. 95 Methanol 3.98 4.27 4. ,26 3. ,87 4. 31 Ethylene glycol 3. .87 5.01 5.38 4. ,37 3. ,89 4. 26 Acetone 3. 82 4.34 4.87 4. 06 3. 92 AD = severe adsorption

PAGE 64

52 the following solute order: 1 2 A 5-tetracarboxybenzene < 1,3,5tricarboxybenzene < 3, 5-dihydroxybenzoic acid < p-hydroxybenzoic acid < benzoic acid. Several compounds with basic properties (e.g. aniline, pyridine) were adsorbed and eluted past V^. Also, several other compounds (e.g. glucose, ethylene glycol, and sucrose) were eluted past V^. Each of the compounds which showed strong evidence of adsorption on Porasil A with water, methanol, or DMF as eluting solvent contained an amino group or an aliphatic OH. In THF, aromatic acids and simple alcohols each showed evidence of adsorptive interaction with Porasil A. Elution on Porasil AX, the deactivated analog of Porasil A, resulted in reduced adsorption. Elution patterns of organic acid standards on Porasil 4 and Porasil AX showed interesting similarities to the elution patterns of extractable soil organic matter. Based on the molecular weights of tetracarboxybenzene and tr icarboxybenzene and the working molecular weight ranges of the gels suggested by the manufacturer, one would expect that the solute would elute at, or slightly before, V^. Inspection of the patterns, however, shows that the acid standards were completely excluded from the gel matrix and eluted at when water was used as the eluting solvent. Deviations from the expected elution behavior of a low molecular weight nonreactive solute may be attributed to adsorption, electrostatic exclusion, or molecular association. The exclusion of the acidic organic solute from Porasil A in water, methanol, or DMF may be attributed to (i) the porous structure of the gel, (ii) association of solute molecules, and/or (iii) electrostatic exclusion from the porous matrix. The first explanation is

PAGE 65

53 unlikely since the nonreactive solute, acetone, produced a symmetrical peak at with negligible skewing, indicative of free entrance into the porous gel matrix. Association of the solute molecules in water, methanol, and DMF would be questionable since aggregation of the molecular units should be greatest in the least polar and/or least basic solvent. Comparison of the individual solvents shows that water, DMF, and methanol have stronger basic character and are considerably more polar than THF with dielectric constants of 76.2, 36.7, 32.6, and 7.58, respectively. Also, deactivation of the porous silica resulted in increased elution volumes, which should not have been the case if skewing was entirely due to aggregation of the solute molecules. With the first two explanations above eliminated as probable major causes of exclusion of the low molecular weight acidic solute, the third explanation, electrostatic exclusion, deserves careful consideration. The acidic functional groups of the organic solute would be partially dissociated in water, methanol, or DMF due to the basic character of each of these solvents. Water and methanol have basic character which is attributed to the presence of the electron-donor oxygen atom. The DMF molecule has two basic sites (Talhoun and Mortland, 1968), the electron-donor oxygen atom of the carbonyl group and the nitrogen atom. The surface Si(OH) groups of the silica packing material are weak acid sites. In water (Kirkland, 1971), methanol, or DMF the surface sites may dissociate, due to the basic properties of these solvents, resulting in a negatively charged silica surface.

PAGE 66

54 Especially in water and methanol, the negative surface sites would be stabilized as a result of the acidic properties of the solvent molecules. The exclusion of acidic organic solute from the porous matrix of Porasil A may therefore be at least partially attributed to electrostatic repulsion between the charged solute molecules and the charged silica surface. In the absence of excess neutral salt, the silica would have an expanded electrical double layer and the solute molecules would exist with larger effective radii. Therefore, it is possible that low molecular weight solutes may be completely excluded from the porous gel matrix. As mentioned previously, there was no evidence of adsorptive interaction between the silica surface and the acidic solute in water, methanol, and DMF; however, in 2-propanol, t-butanol, acetone, ethylacetate, and THF there was evidence of adsorption. In the former solvents, the greater negative charge densities of the solutes and the silica surface may have resulted in less adsorptive interaction between the negatively charged species. In acetone, ethylacetate, and THF, however, the organic solute would be much less dissociated as a consequence of the very weak or negligible basic properties of these solvents. Also, the silica surface would be less highly dissociated. Therefore, there is a more favorable condition for direct H-bonding interactions between the Silica surface and the solute molecules. Deactivation of the silica surface with polyethylene glycol would block the reactive sites (Dark and Limpert, 1973) and result in reduced negative charge density of the silica surface in water and methanol.

PAGE 67

55 Therefore, electrostatic exclusion of negatively charged solute was reduced on Porasil AX compared to Porasil A. When acetone, ethylacetate, or THF was used as the eluting solvent on Porasil AX, only a small quantity of acidic solute was eluted past V^. This phenomenon indicates a reduction in adsorptive interaction between the silica packing material and the acidic solute on Porasil AX compared to Porasil A. Adsorption and electrostatic exclusion were reduced on Porasil AX, but were not completely eliminated. The evidence of adsorption and electrostatic exclusion interactions between the solute and the Porasil AX demonstrated that the packing material was not completely deactivated. Effect of saturating cation Fulvic acid in which the acidic functional groups were saturated to pH 7.0 with Na"*", k"*", NCCH^)^"*", or NCC^Hg)^"^ were eluted at on Porasil A and CPG-250 when water was used as the eluting solvent (Table 14). The cation-saturated samples were also completely excluded from the gel matrix on deactivated Porasil AX. Likewise, both the 1 2 4 5-tetracarboxybenzene and the 1 3 5-tricarboxybenzene in methanol and water were excluded from the Porasil A and Porasil AX gel matrices. The pronounced exclusion of Na-, K-, N(CH^),-, and N(C,H„) J '4 4 9 4 saturated fulvic acid and organic acid standards from the Porasil A gel matrix may be attributed to electrostatic repulsion of the negatively charged solute molecule and the negatively charged sites on the silica surface. Similar exclusion phenomena have been observed during elution of cation-saturated fulvic acid with distilled water on Sephadex (Swift and Posner, 1971).

PAGE 68

56 Table 14. Peak elutlon vdlumes of cation-saturated fulvic acid on Porasil A, Porasil AX, and CPG-250 with water as eluting solvent Column Saturating cation Porasi 1 A Porasil AX CPG250 -ml1.62 1.68 1.79 Na 1.61 1.69 1.76 N(CH^)^ 1.62 1.70 1.78 ^(^9)4 1.60 1.69 1.76

PAGE 69

57 Electrostatic exclusion phenomena have been reported for elution of acidic amino acids (Gelotte, 1960), aromatic acids (Demetriou et al., 1968), inorganic ions (Neddermeyer and Rogers, 1968), and lignosulf onate (Forss and Stenlund, 1973) on the Sephadex G-gels and was attributed to electrostatic repulsion between fixed charges on both the gel and the solute molecules. Effect of excess electrolyte Elution patterns of Na-saturated fulvic acid in the presence of excess neutral salt on Porasil A and Porasil AX (Figs. 8 and 9, respectively) indicated that electrolyte resulted in reduction in the relative quantity of solute excluded from the gel matrix. Presence of neutral salt also influenced elution patterns of low-molecular weight organic acids on Porasil A and Porasil AX (Figs. 10 and 11, Tables 15 and 16) and resulted in reduced exclusion of solute from the gel matrix. Similar phenomena have been observed with Sephadex during the elution of soil organic matter extracts (Swift and Posner, 1971; Posner, 1963), acidic amino acids (Gelotte, I960), aromatic acids (Demetriou et al., 1968), and lignosulf onates (Forss and Stenlund, 1973) At salt concentractions above 0.01 N^, significant quantities of fulvic acid were eluted past V^. The excess electrolyte resulted in suppression of negative charge and reduction in thickness of electrical double layer of both the negatively charged gel surface and the organic solute molecules. Also, excess sal t would decrease the effective size of solute anions due to reduction in thickness of the electrical double layer. Therefore, solute anions would more easily enter the porous

PAGE 70

58 O o r H20 1 1 1^ 0.001 y 1 0.01 N I 1 ""-"oos t^j ,1 J I ... I \ I Vo 2 3 Vt 4 ELUTION VOLUME. ML Fig. 8 Effect of excess neutral electrolyte on elution of Na-sa turated fulvic acid on Porasil A Fig. 9 Effect of excess neutral electrolyte on elution of Na-saturated fulvic acid on Porasil AX

PAGE 71

59 Fig. 10 Effect of excess neutral electrolyte on elution of Na-saturated 1 2 4 5-tetracarboxybenzene on Porasil A 1 Vo 2 3 Vt 4 ELUTION VOLUME. ML Fig. 11 Effect of excess neutral electrolyte on elution of Na-saturated 1 2 4 5tet racarboxybenzene on Porasil AX

PAGE 72

60 Table 15. Peak elution volumes of selected organic acid standards on Porasil A with Na„SO, solutions Z 4 Electrolyte concentration Sample 0.000 N 0.001 N 0.01 N 0.05 N ml 1,2,4, 5-Tetracarboxybenzene 1.74 1.68 1.86 2.60 1, 3, 5-Tricarboxybenzene 1.75 1.68 1.91 2.64 3, 5-Dihydroxybenzoic acid 1.82 1.97 2.32 2.71 Benzoic acid 1.96 2.41 2.51 3.14

PAGE 73

61 Table 16. Peak elution volumes of selected organic acid standards on Porasil AX with Na.SO, solutions Electrolyte concentration Sample 0.000 N 0.001 N 0,01 N 0.05 N ml • 1,2,4, 5-Tetracar boxybenzene 1.87 2.31 3.03 3.21 1, 3 5-Tricarboxybenzene 2.18 2.43 3.05 3.17 3, 5-Dihydroxybenzoic acid 2.21 2.80 3.19 3.28 Benzoic acid 2.41 3.09 3.40 3.48

PAGE 74

62 gel matrix. As double-layer thickness decreased with resulting reduction in electrostatic exclusion, adsorption increased due to direct Hbonding interactions of oxygen-containing groups on the organic solute and Si(OH) sites on the packing material. Adsorption effects were reduced on Porasil X series of packing materials but were not completely eliminated. This evidence indicated that the silica surface was not completely deactivated and/or solute molecules were interacting directly with the silica surface. The solubility studies (Table 7) indicated that fulvic acid began to precipitate at electrolyte concentrations above 0.01 N^, at pH values of 4.0 to 8.0. Therefore, the partial elution of fulvic acid past at the higher electrolyte concentrations may be caused by precipitation of fulvic acid in the column. For this reason, electrolyte concentrations must be maintained at values low enough to preclude precipitation of the solute. Because of the nature of the silica surface, special precautions must be observed. As mentioned previously, the silica surface acts as a weak acid due to the presence of SiOH groups. In the presence of a protic solvent with basic properties, such as water or methanol, these acid sites will dissociate, leaving the silica surface with a net negative charge. As the pH of the solvent medium is increased, the dissociation of surface sites and the negative charge density of the silica surface is also increased. The negative charge density of acidic solute molecules would also increase with increasing pH. Therefore, exclusion of negatively charged solute from the negatively charged packing material would increase with increasing pH.

PAGE 75

63 At pH values above 7.0, the silica surface may be destroyed by solubilization and formation of silicate. Therefore, alkaline conditions must be avoided. Under alkaline conditions, the chemically adsorbed deactivating agent is also readily stripped from the surface of the Porasil X packing material. The manufacturer recommends that use of several organic solvents, especially DMF, should be avoided with deactivated Porasil. Such a solvent may readily strip the deactivating agent from the silica surface. Polystyrene-diviny Ibenzene (DVB ) Operating conditions. As with the porous glass packing materials, column efficiencies of Poragel and y-Styragel packing materials were greatly influenced by solvent flow rates. With 0.05A-cm OD columns and THF as the eluting solvent, minimum peak broadening and maximum column efficiencies were obtained at flow rates of approximately 0.8 and 3 0 o ml per min. for the 100 A Poragel and y-Styragel columns, respectively (Fig. 12). At higher flow rates, peak broadening was increased and column efficiencies were decreased. At lower flow rates, column efficiencies were also appreciably lowered. This later effect was much more evident with the polystyrene-DVB gels than with the porous glasses. At low flow rates, diffusion of solute molecules apparently resulted in decreased column efficiencies. With the y-Styragel columns, the high column efficiency at the high flow rate was due to the small particle size and uniform pore size of packing which permitted rapid

PAGE 76

64 7000 r o|00 A ;i-STYRAGEL(If) A 100 A PORAGEL (Zf) 4.0 6.0 8,0 FLOV^ RATE, ml/min 10.0 Fig. 12 Effect of flow rate on column efficiency, N, of 100 A O Poragel and 100 A u-Styragel preparative columns with THF as the eluting solvent

PAGE 77

65 equilibrium of solute molecules between internal pore space and interstitial pore space. Based on these studies, solvent flow rates of 0.8 and 3.0 ml per min. in 0.95A-cra OD columns were used for Poragel and y-Styragel columns, respectively, in all subsequent studies. The effect of sample size on column efficiencies of 0.95A-cm diameter columns of Poragel and p-Styragel are shot^m in Fig. 13. In general, the maximum sample volumes were 250 yl and 25 yl for the Poragel and y-Styragel columns, respectively. Larger samples resulted in increased peak broadening and reduced apparent column efficiencies. Column parameters Column parameters (Table 1), and V^, of the packed columns were determined by elution of acetone or 2,600,000 molecular weight polystyrene, respectively. Column efficiencies, N, were o O approximately 800 and 5,000 for the 100 A Poragel and 100 A y-Styragel, respectively. Molecular weight calibration curves, obtained by elution of polystyrene standards with THF gave working molecular weight ranges of 500 to 20,000 and 100 to 3,000, respectively, for the above gels (Fig. 14). Several of the highly substituted organic acid standards deviated from the polystyrene calibration curve; therefore, the polystyrene standards are not suitable for accurate determination of molecular weights of low molecular weight organic acids. Based on the above observation, it is doubtful that polystyrene standards would be suitable standards for molecular weight determinations of soil humic compounds.

PAGE 78

6000 olOO A >i-STYRAGEL(lf) A|00 A P0RAGEL(2f) 0.100 SAMPLE SIZE, ml 0.200 Fig. 13 Effect of sample^size on column efficiency, N, of 100 A Poragel and 100 A iJ-Styragel preparative Columns with THF as the eluting solvent

PAGE 79

I t t I I 10 20 30 40 50 ELUTION VOLUME, ml Fig. 14 Molecular weight^calibration curve of U-St^ragel (2 f x 0.954 cm OD 100 A + 2 f X 0.954 cm OD 500 A) obtained by elution of polystyrene standards with THF

PAGE 80

68 The selection of suitable molecular weight standards for soil humic compounds remains a problem. Effect of solvent on elution of standard compounds. Cotnparison of the elution volumes of acetone and 2,600,000 molecular weight poly0 styrene in THF and DMF in 100 A Poragel suggests that the Poragel is poorly Swelled in DHF (Table 1) This conclusion is based on the assumption that acetone is a nonreactive solute and readily enters the Poragel gel matrix and that the high molecular weight polystyrene standard is completely excluded from the gel matrix. In DMF, the elution volumes of acetone and the 2,600,000 molecular weight polystyrene are separated by 5.3 ml compared to THF in which the elution volumes of low and high molecular weight standards are separated by 10.5 ml. Therefore, in DMF the internal pore volume of the packing material is 33% of the total pore volume compared to the THF in which the internal pore volume of the packing material is approximately 49% of the total pore volume. The greater swelling of the polystyrene-DVB gel in THF compared to DMF may be attributed to the less polar and greater hydrophobic character of the former solvent which would make it more compatible with the hydrophobic gel. As a consequence of the different swelling properties of the gel in the different solvents, it is essential that column parameters and molecular weight distribution patterns be determined for the same solvent which is to be used as the eluting solvent. Also, it is essential that the column be packed in the same solvent which is to be used as the eluting solvent. Changing of the eluting solvent in the

PAGE 81

69 column may produce voids and result in increased peak widths and reduced column efficiencies. Peak elution volumes of organic standards eluted with THF and o DMF on lOQ A Poragel are summarized in Table 17. The initial observation is that several of the solute species are behaving differently in the two eluting solvents. In looking more closely at solute behavior in THF, it can be observed that each of the low-molecular weight solutes were eluted in the vicinity of or slightly after V^. Apparently, each of these solutes readily entered the pores of the Poragel gel matrix. Only benzene and anthracene were eluted noticeably past V^. Edwards and Ng (1968) also observed the adsorption of some aromatic compounds by polystyrene-DVB when eluted with THF. It is probably the aromatic character of the polystyrene-DVB gel which resulted in adsorption of benzene and anthracene. The aromatic acids were not noticeably eluted past V^ and were apparently not strongly adsorbed. The elution of these compounds near V^ gave strong indication that they readily entered the polystyrene-DVB gel matrix. o Elution of standard compounds with THF on 100 A y-Styragel produced very similar results. Only benzene, toluene, and anthracene were eluted past the assumed value of V^, due to an apparent adsorptive interaction with the gel matrix. Other standard compounds tested, e.g. simple alcohols, aromatic acids, aromatic bases, and phenolic acids, were eluted in the vicinity of V^ and apparently readily entered the polystyrene-DVB gel matrix.

PAGE 82

70 Table 17. Peak elution volumes of low molecular weight standards eluted on 100 A Poragel with THF and DMF and on 100 A y-Styragel with THF as the eluting solvent Column packing 100 A Poragel 100 A p-Styragel Sample THF DMF THF -ml1,2,4, 5-Tetracar boxybenzene 20.81 12.93 10.27 1,3, 5-Tricarboxybenzene 21. OA 12.97 10.34 3 5-Dihydroxybenzoic acid 21.19 13.04 10.36 Benzoic acid 21.23 13.28 10.49 Pyridine 22.13 18.12 10.56 Aniline 21.28 17.21 10.52 Methanol 21.24 16.18 10.53 Ethylene glycol 21.13 16.34 10.49 Acetone 21.26 16.47 10.53 Benzene 21.84 17.63 10.89 Anthracene 22.43 18.02 11.42

PAGE 83

71 When DMF was used as the eluting solvent, several of the aromatic acids were eluted prior to the assumed value of V^. The more highly substituted aromatic acids (e.g. 1 2 4 5-tetracarboxybenzene and 1,3,5tricarboxybenzene) were eluted near the assumed value of and were apparently completely excluded from the gel matrix. Several of the aromatic acids and phenolic acids showed two elution peaks which corresponded closely to the assumed values of and V^. Each of the low molecular weight compounds which were eluted noticeably before contained an acidic side group, COOH and/or phenolic DH. Several compounds, e.g. benzene, toluene, and anthracene, were eluted considerably past the assumed value of V^. Each of these compounds was hydrophobic in nature and was structurally similar to monomers of the gel polymer. Benzene, toluene, and anthracene were more strongly adsorbed with DMF than with THF as the eluting solvent. This effect was probably due to the more polar character of the DMF. Neutral solutes (e.g. simple alcohols) and compounds with basic properties (e.g. pyridine, aniline) were eluted at or slightly after the assumed and apparently readily entered the polystyrene-DVB gel matrix. Two interesting points from the above observations are that (i) hydrophobic solutes were more strongly adsorbed to the polystyrene-DVB gel matrix when eluted with DMF than with THF, and (ii) acidic solutes were noticeably excluded from the gel matrix when DMF was used as the eluting solvent, but not when THF was used as the solvent. The first point may be explained in terms of relative hydrophobic character

PAGE 84

72 of the two solvents, as discussed previously. The second point is elaborated upon below. The acidic functional groups of an acidic organic solute would be partially dissociated in DMF due to basic character of this solvent; therefore, the solute molecules are likely to be highly dispersed. The negatively charged solute molecules would have larger effective radii than the neutral species; however, this phenomena should not entirely account for the exclusion phenomena since the exclusion limit of the gel, based on polystyrene standards, is approximately 50,000 molecular weight. In the porous silica and Sephadex gels, the exclusian phenomena can be explained in terms of electrostatic repulsion from negative charge sites in the gel matrix. In porous silica, the negative charge results from dissociation of Si(OH) sites at silica surface. In Sephadex the negative charge has been attributed to COOH impurities in the gel matrix. On the other hand, the polystyrene-DVB gel should exist as a neutral species. Therefore, we must search for an alternate explanation to the exclusion phenomena. A possible explanation is the ion inclusion effect suggested by Forss and Stenlund (1975) in studies of lignosulf onate They attributed this effect to the interaction of charged sites on the ions entering the pores with other charged ions outside of the pores. The net effect is electrostatic repulsion. Such an effect would not entirely account for the apparent total exclusion of low molecular weight solutes observed. Further work will be required to determine the nature of this phenomena.

PAGE 85

73 Effect of solvent on elution of soil humic compounds Elution patterns of soil humic fractions on 100 A u-Styragel with THF as the elating solvent are shown in Figs. 15 and 16. In THE, all humic fractions were eluted between and and apparently readily entered the porous gel matrix. As mentioned previously, the polystyrene standards are not suitable for accurate molecular weight determinations of soil humic materials. These standards, however, do provide a guide for measurement which is probably no less suitable than others commonly used, such as proteins or polysaccharides. Molecular weight estimates based on the polystyrene standards are summarized in Table 18. In all cases the Soxhlet fractions were estimated to have peak molecular weights less than 800. These fractions, however, represent only a minor portion of the total NaOHor DMF-extractable materials, 28 and 32%, respectively, and are likely to contain materials with lower peak molecular weights than those of the NaOHor DMF-extractable materials. These later materials are not sufficiently soluble in THF to obtain a molecular weight fractionation. In DMF, the major portions of all humic fractions were eluted at volumes corresponding to the assumed values of and were apparently largely excluded from the gel matrix. In all cases, a minor portion of the material was eluted at V^. Reinjection of fractions collected at Vq produced patterns similar to the original patterns with major peaks corresponding closely to and minor peaks at the assumed value of V^. Reinjection of the sample eluted at during the original fractionation also produced a fractionation pattern similar to the

PAGE 86

74 ACETONE PROPANOL METHANOL NaOH SOXHLET I I 12 18 24 30 36 ELUTION VOLUME, ML 48 Fig. 15 Elution of SoxhleL extjacts of NaOH-extractable soil organic matter on 100 A y-Styragel with THF

PAGE 87

75 CO o CL CO UJ q: q: UJ Q q: o o LU 01 ACETONE PROPANOL METHANOL NaOH SOXHLET 12 18 24 30 36 ELUTION VOLUME, ML Fig. 16 Elution of Soxhlet extracts of DMF-extractable soil organic matter on 100 A y-Styragel with THF

PAGE 88

76 original pattern with a major peak eluted at and a minor peak at V^. This technique produced strong evidence that the initial fractionation pattern was not the result of a separation according to molecular size, but instead was an artifact resulting from a complex gel-solvent-solute interaction. The elution patterns of the soil humic acid fractions were indeed similar to the patterns obtained from elution of the low molecular weight aromatic acids.

PAGE 89

77 Table 18. Molecular weight estimates of soil humic fractions based on elutlon of polystyrene standards on y-Styragel with THF as the eluting solvent Estimated molecular weight Sample NaOH extract UMF extract Acetone-Soxhlet 560 660 2-Propanol-Soxhlet 720 740 Methanol-Soxhlet 760 740

PAGE 90

CONCLUSIONS Extractio n and Fractionation Several of the dipolar aprotic solvents, i.e. DMF and DMSO, were shown to be excellent solvents for the soil humic fraction. Functional group, elemental, and IR analysis indicated that the DMF-extractable soil organic matter was chemically similar to the material extracted by 0.5 N NaOH. The dipolar aprotic solvent may, therefore serve as an excellent complementary solvent to NaOH for chemical studies of extractable soil organic matter. The Soxhlet fractionation scheme was used successfully to fractionate the extractable soil organic matter into samples with distinct characteristics. The Soxhlet extraction scheme of hexane, benzene, ethylacetate acetone, 2-propanol, and methanol was utilized to obtain materials with progressively greater hydrophilic character, lower C and H contents, greater N, S, and 0 contents, greater COOH content, and greater total acidity. The Soxhlet solvents were able to extract 29.6 and 35.0% of the NaOHand DMF-extractable materials, respectively. Even though these fractions represent a minor portion of the total extractable material, they are likely to contain materials of simpler average composition and lower peak molecular weights than the NaOHand DMF-extractable 78

PAGE 91

79 material and are important since they are likely to contain monomers which compose the polymeric structure of the humic complex. Therefore, investigations of these fractions provide information which will aid in understanding properties of the total humic complex. Solubility Properties Humic acid and NaOHand DMF-extractable soil organic matter were 100% soluble at the 0.1% concentration in both DMF and DMSO. Fulvic acid was completely soluble at the 0.1% concentration in both water and methanol. Solubility of extractable soil organic matter was decreased in the dipolar aprotic solvents, e.g. DMF, THF, and acetone, and increased in the protic solvents as acidic hydrogen was placed with Na"^, k"*", or NCCH^),"*". The solubility of fulvicacid in aqueous systems was influenced by pH and concentration of excess neutral electrolyte. Fulvic acid was completely soluble at the 0.1% concentration in aqueous systems with concentrations of excess neutral electrolyte up to 0.1 N at pH values less than 4.0 and greater than 8.0; however, at pH values from 4.0 to 8.0, fulvic acid partially precipitated with concentrations of Na^SO^ or K2S0^ greater than 0.05 N. The number of useful chromatographic fractionation schemes is greatly limited by the solubility characteristics of the solute; therefore, an understanding of these characteristics is an essential prerequisite to the rapid screening of possible fractionation schemes.

PAGE 92

80 Liqui d Chromatography None of the chromatographic gels Investigated was completely inert. Each gel apparently interacted with the soil humic material; therefore, the elution patterns were not entirely attributable to a molecular seiving phenomenon but to a combination of molecular seiving, adsorption, and ionic exclusion phenomena. Solvent and electrolyte effects were especially evident in studies of Porasil and CPG packing materials. When H-, Na-, or N(CH^)^saturated low molecular weight organic acid standards or fulvic acid were eluted with H2O, the solute molecules were partially or totally excluded from the porous gel matrix. As electrolyte concentration was increased, the acidic solute molecules more readily entered the porous matrix; however, at electrolyte concentrations above 0.01 N, significant quantities of fulvic acid were' adsorbed and eluted past V^. These phenomena were attributed to decreased thickness of the electrical double layer and/or suppression of charge of the negatively charged solute molecule and the negatively charged silicate surface. Adsorption of fulvic acid at the higher electrolyte concentrations was attributed to increased interaction between active sites at the silica surface and oxygenand nitrogen-containing functional groups of the organic solute and also to the possible precipitation of fulvic acid caused by the high counter ion concentration at the negatively charged silica surface. Low molecular weight organic solutes with significant basic properties, i.e. pyridine, were strongly adsorbed to Porasil and CPG

PAGE 93

81 gels in aqueous systems. Since the fulvic acid sample contained nitrogen, indicating the probable presence of basic sites, interactions of these sites in the negatively charged solute molecule with the negatively charged silicate surface would be greater in the presence of excess neutral electrolyte. When protic solvents, i.e. H2O, methanol, and 2-propanol, or dipolar aprotic solvents with significant basic properties, i.e. DMF, were used to elute low molecular weight acidic solutes or extractable soil organic matter, electrostatic exclusion phenomena predominated. In dipolar aprotic solvents without significant basic character, i.e. acetone and THF, adsorption phenomena predominated. With all solvent and electrolyte systems examined, it was not possible to completely eliminate both electrostatic exclusion and adsorption phenomena. Deactivation of the Porasil surface, Porasil X, did not completely eliminate adsorption and exclusion interactions between the acidic organic solute and the silica surface. The polystyrene-DVB gels were compatible with a more limited range of solvents than the silica gels. Also, due to the swelling properties of the gel it was essential to pack the column with the same solvent which was to be used as the eluting solvent. With the polystyrene-DVB gels, elution patterns were dependent on the eluting solvent. In DMF, low and high molecular weight acidic organic solutes were totally or partially excluded from the gel matrix. In THF, low molecular weight acid solutes apparently readily entered the porous gel matrix and were

PAGE 94

82 eluted in the vicinity of V^. Of the compounds tested, only several hydrophobic aromatic compounds were strongly adsorbed. It appears that THF is a suitable solvent for elution of acidic solutes; however, only the benzene-, ethylacetate-, acetone-, and 2-propanol-Soxhlet fractions were suitably soluble in THF. Methylated fractions of the humic acid and fulvic acid fractions would also be soluble in THF. Molecular weights of acetone-, 2-propanol-, and methanolSoxhlet fractions were estimated to be 500 to 800, based on elution patterns of soil humic fractions with those of polystyrene standards in THF. Further research will be needed to corroborate, by other methods, the molecular weight estimates obtained with gel permeation chromatography. One such method would be vapor pressure osmometry. Each of the two general groups of gels evaluated in this study showed evidence of adsorptive and/or electrostatic interaction with extractable soil organic matter. Mode and extent of interaction were highly dependent on the solvent medium. Because of these possible interactions, special care must be observed in the interpretation of gel permeation chromatography patterns. High pressure liquid chromatography is a valuable new technique because of the use of high efficiency columns and the short time required to obtain elution patterns. In addition to the application of gel permeation chromatography, there is the unexplored potential application of liquid-liquid partition chromatography and liquid-solid adsorption chromatography to fractionation of extractable soil organic matter. High pressure liquid chromatography will also provide a valuable tool for studies of soil organic-mineral-ionic interactions.

PAGE 95

LITERATURE CITED Bergmann, J.G., L.l. Duffy, and R.B. Stevenson. 1971. Solvent effects in gel permeation chromatography. Anal. Chem. 43:131-133. Bly, D.D. 1970. Gel permeation chromatography. Science 168:527-533. Bombaugh, K.J. 1971. The practice of gel permeation chromatography. In J.J. Kirkland (ed.). Modern Practice of Liquid Chromatography. Wiley-Interscience New York. Bombaugh, K.J., W.A. Dark, and J.N. Little. 1969. Fractionation of alcohol on deactivated porous silica beads by gel permeation chromatography. Anal. Chem, 41:1337-1339. Bremner, J.M. 1950. Borne observations on the oxidation of soil organic matter in the presence of alkali. J. Soil Sci. 1: 198-204. Bremner, J.M. 1956. Some soil organic matter problems. Soil and Fertilizers 19:115-123. Bremner, J.M. and J. Lees. 1949. Studies on soil organic matter. II. The extraction of organiq matter from soil by neutral reagents. J. Agr. Sci. 39:274-279. Brook, A.J.W. and S. Housley. 1969. The interaction of phenols with Sephadex gels. J. Chromatogr. 41:200-204. Brook, A.J.W. and K.C. Munday 1970. The interaction of phenols, aniline, and benzoic acids with Sephadex gels. J. Chromatogr. 47:1-8. Brooks, J.D., R.A. Durie, and S. Sternhell. 1958. Chemistry of brown coals. III. Pyrolytic reactions. Aust. J. Appl. Sci. 9:303-320. Burges, N.A. H.M. Hurst, and B. Walkden. 1964. The phenolic constituents of humic acid and their relation to the lignin of the plant cover. Geochim. Cosmochim. Acta 28:1547-1554. 83

PAGE 96

8A Cheshire, M.V., P. A. Cranwell, CP. Falshow, A.J. Floyd, and 6.D. Haworth. 1967. Hutaic acid: 2. Structure of humic acid. Tetrahedron 23:1669-1682. GhoUdhri, M.B., and F.J. Stevenson. 1957. Chemical and physicochemical properties of soil humic colloids: III. Extraction of organic matter from soils. Soil Sci. Soc. Amer. Proc. 21: 508-513. Cogswell, T.E., J.F. McKay, and D.R. Latham. 1971. Gel chromatographic separation of petroleum acids. Anal. Chem. 43:645-648. Cooper, A.R. and E.M. Barrall. 1973. Gel permeation chromatography: physical characterization and chromatographic properties of Porasil. J. Appl. Polym. Sci. 17:1253-1268. Cooper, A.R. and A.R. Bruzzone. 1973. Characterization and properties of macromolecules III. Gel permeation chromatography: the effect of temperature on the elution volume and the efficiency of the separation process. J. Polym. Sci. A-2, 11: 1423-1434. Cooper, A.R., A.R. Bruzzone, J.H. Gain, and E.M. Barrall. 1971. Characterization and chromatographic properties of Corning porous glasses. J. Appl. Polym. Sci. 15:571. Cooper, A.R. and J.F. Johnson. 1969. Gel permeation chromatography: the effect of treatment with hexamethyl disilazane on porous glass packings. J. Appl. Polym. Sci. 13:1487-1492. Cooper, A.R., J.F. Johnson, and R.S. Porter. 1973. Gel-permeation chromatography: current status. American Laboratory. May 1973:12-24. Dark, W.A. and R.J. Limpert. 1973. An evaluation of available packings for GPC. J. Chromatogr. Sci. 11:114-120. Demetriou, J. A. et al. 1968. Gel filtration chromatography of fluorescent phenolic and heterocyclic compounds. J. Chromatogr. 34:342-350. Determann, H. and I. Walter. 1968. Source of aromatic affinity to Sephadex dextran gels. Nature (Lond.) 219:604-605. Edwards, G.D. and Q.Y. Ng 1968. Elution behavior of model compounds in gel permeation chromatography. J. Polym. Sci. Part C 21: 105-117. Flaig, W. H. Beutelspacher and E. Rietz. 1975. Chemical composition and physical properties of humic substances. In J.E. Gieseking (ed.), Soil components. Springer-Verlag New York.

PAGE 97

85 Forss, K.G. and B.G. Stenlund. 1975. The influence of charged groups in gel permeation chromatography on polyelectroly tes J. Polym. Sci. 42:951-963. Freeman, D.H. 1973. The gels for liquid chromatography. J. Chromatogr. Sci. 11:175-180. Fritz, J.S., S.S. Yamamura, and E.G. Bradford. 1959. Determination of carbonyl compounds. Anal. Chem. 31:260-263. Gaylor, V.F., H.L, James, and H.H. Weetall. 1976. Gel and affinity chromatography. Anal. Chem. 48:44R-51R. Gelotte, B, 1960. Studies on gel filtration sorption of the bed material Sephadex. J. Ghromatogr. 3:330-342. Giddings, J.C. 1965. Dynamics of chromatography: principles and theory. Marcel Dekker New York. Halstead, R.L., G. Anderson, and S.M. Scott. 1966. Extraction of organic matter from soils by means of ultrasonic dispersion in aqueous acetylacetone Nature 211:1430-1431. Hansen, E.H. and M. Schnitzer. 1967. The alkaline permanganate oxidation of Danish illuvial organic matter. Soil Sci. Soc. Amer. Proc. 30:745-748. Hansen, E.H. and M. Schnitzer. 1969. Zinc-dust distillation of soil humic compounds. Fuel 48:41-46. Havjk, G.L., J. A. Cameron, and L.B. Dufault. 1972. Chromatography of biological materials on polyethylene glycol treated controlled pore glass. Prep. Biochem. 2:193-203. Haworth, R.D. 1971. The chemical nature of humic acid. Soil Sci. 111:71-79. Hiatt, C.W. et al. 1971. Treatment of controlled-pore glass with poly (ethylene oxide) to prevent adsorption of rabies virus. J. Chromatogr. 56:362-364. Hurst, H.M. and N.A. Burges 1967. Lignin and humic acids. In A.D. McLaren and G.H. Peterson (eds.). Soil biochemistry. Marcel Dekker, New York. Karger, B.L. 1971. The relationship of theory to practice in high speed liquid chromatography. Tn J.J. Kirkland (ed.), Modern practice of liquid chromatography. VJlley-Interscience New York.

PAGE 98

86 Kessler, T., R.A. Friedel, and A.G. Sharkey. 1970. Ultrasonic solvation of coal in quinoline and other solvents. Fuel 49:222-223. Khan, S.U. 1971. Distribution and characteristics of organic matter extracted from black solnetzic and black chenozemic soils of Alberta: the humic acid fraction. Soil Sci. 112:401-409. King, E.J. 1973. Acid-base behavior. In A.K. Covington and T, Dickinson (eds.). Physical chemistry of organic solvent systems. Plenum Press, London. Kirkland, J.J. 1974. Modern practice of liquid chromatography. Wiley-Interscience New York. Kononova, M.M. 1966. Soil organic matter. 2nd English ed Pergamon Press, New York. Kumada, K. and A. Suzuki. 1961. Isolation of anthraquinones from humuS. Nature 191:415-416. Laub, R.J. 1974. Packings for HPLC. Research and development. July 1974:24-28. LePage, M. R. Beau, and A.J. DeVries. 1968. Evaluation of analytical gel chromatography columns packed with porous silica beads. J. Polym. Sci., Part C 2:119-130. Loeppert, R.H. Jr., and B.C. Volk. 1974. Nature of organic matter extracted from terra Ceia ittu-ck with selected solvents. Soil and Crop Sci. Soc. Fla. 33:160-164. Loeppert, R.H., Jr., and B.C. Volk. 1976. Application of high pressure liquid chromatography to studies of extractable soil organic matter-porous silica packings. International Symposium on Soil Organic Matter Studies, Braunschweig, Germany. (In Press). Loeppert, R.H., Jr., L.W. Zelazny, and B.C. Volk. 1976. Acidic properties of kaolinite in water and acetonitile. Soil Sci. Soc. Amer. J. (In press). Mclver, R.D. 1962. Ultrasonics — a rapid method for removing soluble organic matter from sediments. Geochim. Cosmochim. Acta 26: 343-345. Mortenson, J.L. 1965. Partial extraction of organic matter. Iii C.A. Black (ed.). Methods of soil analysis, Agromony Monograph No. 9, Vol. 2. American Society of Agronomy. Madison, Wisconsin

PAGE 99

87 Neddermeyer, P. A. and L.B. Rogers. 1968. Gel filtration behavior of inorganic salts. Anal. Chem. 40:755-762. Ortiz de Serra, M.I. and M. Schnitzer. 1973. The chemistry of humic and fulvic acids extracted from Argentine soils. II. Permanganate oxidation of methylated humic and fulvic acids. Soil Biol. Biochem. 5:287-296. Parsons, J.W. and J. Tinsley. 1960. Extraction of organic matter with anhydrous formic acid. Soil Sci. Soc. Amer. Proc. 24il98201. Porter, L.K. 1967. Factors affecting the solubility and possible fractionation of organic colloids extracted from soil and leonardite with an acetone-H 0-HCl solvent. J. Agr. Food Chem. 15:807-811. Posner, A.M. 1966. The humic acid extracted by various reagents from a soil. I. Yield, inorganic components, and titration curves. J. Soil Sci. 17:65-68. Posner, A.M. 1963. Importance of electrolyte in the determination of molecular weights by Sephadex gel filtration with special reference to humic acid. Nature 198:1161-1163. Salfeld, J.C. 1964. Fractionation of a humus preparation with aqueous organic solvents. Landbauf erschung Volkenrode 14:131-136. Schnitzer, M. and V.C. Gupta. 1965. Determination of acidity in soil organic matter. Soil Sci. Soc. Amer. Proc. 29:274-277. Schnitzer, M. and S.U. Khan. 1972. Humic substances in the environment. Marcel Dekker, Inc., New York. Schnitzer, M., D.A. Shearer, and J.R. Wright. 1959. A study in the infrared of high molecular weight organic matter extracted by various reagents from a podzolic B horizon. Soil Sci. 87: 252-257. Schnitzer, M. and S.I.M. Skinner. 1968. Gel filtration of fulvic acid, a soil humic compound. Ln Isotopes and radiation in soil organic matter studies. International Atomic Energy Agency, Vienna. Snow, J.T., R.H. Loeppert, Jr., and B.G. Volk. 1973. Yields and functional group characteristics of NaOH-extracted humic acid from Pahokee muck. Soil and Crop Sci. Soc. Fla. Proc. 33: 165-167.

PAGE 100

88 Sommers, T.C. 1966. Wine tannins — Isolation of condensed flavanoid pigments by gel filtration. Nature (Load.) 2D$2 368-370. Spatorico, A.L. 1975. Exclusion chromatography using controlled porosity glass. I. Comparison with styrene gels. J. Appl. Polym. Sci. 19:1601-1610. Spatorico, A.L. and G.L. Beyer. 1975. Exclusion chromatography using porous glass. II. Application to hydrophilic polymers. J. Appl. Polym. Sci. 19:2933-2945. Steelink, C. and G. Tollin. 1967. Free radicals in soil. In A.B. McLaren and G.H. Peterson (eds.), Soil biochemistry. Marcel Dekker, New York. Stevenson, F.J. 1965. Gross chemical fractionation of organic matter. In C.A. Black (ed ) Methods of soil analysis. Agronomy Monograph No. 9, Vol. 2. American Society of Agronomy. Madison, Wisconsin Swift, R.S. and A.M. Posner. 1971. Gel chromatography of humic acid. J. Soil Sci. 22:237-249. Talhoun, S.A. and M.M. Mortland. 1966. Complexes of montmorillonite with primary, secondary, and tertiary amides. I. Protonation of amides on the surface of montmorillonite. Soil Sci. 102: 248-254. Tinsley, J. and A. Salam. 1961. Extraction of soil organic matter with aqueous solvents. Soils and Pert. 24:81-84. Unger, K.K. R. Kern, M. Ninou, and K.F. Krebs. 1974. GPC with a new type of silica packing material. J. Chromatogr. 99:435-443. Volk, E.G. and M. Schnitzer. 1973. Chemical and spectroscopic methods for assessing subsidence in Florida Histosols. Soil Sci. Soc Amer. Proc. 37:886-888. Zweig, G. and J. Sherma. 1974. Chromatography. Anal. Chem. 46:73R94R.

PAGE 101

BIOGRAPHICAL SKETCH Richard Henry Loeppert, Jr., was born September 26, 1944, in Raleigh, North Carolina. He graduated from Needham B. Broughtan High School in Raleigh, North Carolina, in June, 1962. In August, 1966, he received his Bachelor of Science degree with a major in soil science from North Carolina State University, Raleigh, North Carolina. Following graduation, he was employed as Assistant County Agent with the Florida Agriculture Extension Service in Jackson County, Florida. He began his graduate studies at the University of Florida in 1970 and received his Master of Science degree in soil science in August, 1973. He is currently a candidate for the Ph.D. degree in the Department of Spil Science, University of Florida. He is a member of the American Society of Agronomy, the Soil Science Society of America, the International Society of Soil Science, and the Clay Minerals Society. 89

PAGE 102

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. John G. A. Fiskell, Chairman Professor of Soil Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Bob G. Volk, Cochairman Associate Professor of Soil Science I certify that 1 have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Associate Professor of Soil Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philsophy. /N. Gammon / Professor of Soil Science

PAGE 103

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy December, 1976 Dean, Graduate School


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E1LBJR9C2_39RUU1 INGEST_TIME 2014-10-30T21:22:59Z PACKAGE AA00026355_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


Table 3. Elemental and functional group concentrations of extractable soil organic matter
Sample
DMF extract
Hexane-Soxhlet
Benzene-Soxhlet
Ethylacetate-Soxhlet
Acetone-Soxhlet
2-Propanol-Soxhlet
Methanol-Soxhlet
Residue
Humic acid
Fulvic acid
Water extract
NaOH extract
Hexane-Soxhlet
Benzene-Soxhlet
Ethylacetate-Soxhlet
Acetone-Soxhlet
2-Propanol-Soxhlet
Methanol-Soxhlet
Residue
Total Phenolic Alcoholic
C H N S 0 acidity COOH C=0 OH OH
weight % meq/g
56.2
5.
7
4.
1
0.
6
33.4
7.2
3.5
2.6
3.7
2.5
81.2
9.
1
0.
1
0.
1
9.5





77.7
8.
8
0.
2
0.
1
13.2





74.2
8.
2
0.
5
0.
1
17.0





69.3
8.
0
0.
6
0.
1
22.0
3.9
1.9
1.9
2.0
1.3
67.5
7.
7
1.
0
0.
3
23.5
4.1
2.0
2.1
2.1
1.4
62.4
6.
5
1.
4
0.
4
29.3
4.5
2.3
2.0
2.2
1.5
56.9
5.
7
3.
7
0.
6
33.1
7.5
4.0
2.6
3.5
2.3
55.7
5.
6
4.
0
0.
8
33.9
7.4
3.8
2.6
3.6
2.4
47.3
4.
4
2.
7
0.
6
45.0
10.1
5.5
2.4
4.6
3.0
50.2
4.
9
2.
6
0.
6
41.7
8.7
4.4
2.8
4.3
2.9
55.6
5.
4
3.
7
0.
7
34.6
7.4
3.7
2.8
3.7
2.5
79.9
9.
0
0.
1
0.
1
10.9





77.2
8.
8
0.
3
0.
1
13.6





75.4
8.
6
0.
5
0.
1
15.4





69.1
8.
1
0.
7
0.
2
21.9
3.7
1.9
1.7
1.8
1.2
68.4
7.
4
1.
1
0.
3
22.8
4.3
2.2
2.1
2.1
1.4
61.7
6.
9
1.
4
0.
3
29.7
4.6
2.4
2.2
2.2
1.4
56.1
5.
5
3.
9
0.
7
33.8
7.5
3.8
2.8
3.7
2.5


TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS iii
LIST OF TABLES vi
LIST OF FIGURES viii
ABSTRACT x
INTRODUCTION 1
LITERATURE REVIEW 4
Soil Organic Matter 4
Extraction of Soil Organic Matter 6
Gel Permeation Chromatography 8
General Information 8
Theory and Nomenclature 8
Packing Materials 10
MATERIALS AND METHODS 16
Sample Pretreatment and Extraction 16
Solubility Studies 19
Analytical Determinations 19
High Pressure Liquid Chromatography 21
RESULTS AND DISCUSSION 25
Chemical Characteristics of Extractable Organic Matter. . 25
Solubility Characteristics of Extractable Organic Matter. 31
High Pressure Liquid Chromatography 38
Porous Silica Packing Materials 38
Polystyrene-divinylbenzene (DVB) 63
CONCLUSIONS 78
Extraction and Fractionation 78
Solubility Properties 79
Liquid Chromatography 80
LITERATURE CITED 83
BIOGRAPHICAL SKETCH 89
v


79
material and are important since they are likely to contain monomers
which compose the polymeric structure of the humic complex. Therefore,
investigations of these fractions provide information which will aid
in understanding properties of the total humic complex.
Solubility Properties
Humic acid and NaOH- and DMF-extractable soil organic matter were
100% soluble at the 0.1% concentration in both DMF and DMSO. Fulvic
acid was completely soluble at the 0.1% concentration in both water
and methanol. Solubility of extractable soil organic matter was de
creased in the dipolar aprotic solvents, e.g. DMF, THF, and acetone, and
increased in the protic solvents as acidic hydrogen was placed with
Na+, K+, or N(CH ),\
The solubility of fulvic- acid in aqueous systems was influenced
by pH and concentration of excess neutral electrolyte. Fulvic acid was
completely soluble at the 0.1% concentration in aqueous systems with
concentrations of excess neutral electrolyte up to 0.1 N at pH values
less than 4.0 and greater than 8.0; however, at pH values from 4.0 to
8.0, fulvic acid partially precipitated with concentrations of
Na^SO^ or t^SO^ greater than 0.05 N_.
The number of useful chromatographic fractionation schemes is
greatly limited by the solubility characteristics of the solute; there
fore, an understanding of these characteristics is an essential
prerequisite to the rapid screening of possible fractionation schemes.


31
DMF- and NaOH-extractable organic matter, and the humic acid
fraction had very similar elemental compositions and concentrations of
O-containing functional groups. The same observation was previously
made by Loeppert and Volk (1974) in comparisons of DMF-and NaOH-ex
tractable materials. These similarities were corroborated by the infra
red patterns (Fig. 2).
The fulvic acid fraction had lower C and H contents, higher 0
content, lower N content, higher total acidity, and higher COOH content
than the humic acid and DMF-extractable materials. The higher acidity
of fulvic acid compared to the other materials was corroborated by the
significantly higher titratable acidity, as determined by potentiometric
titration in DMF.
In summary, C and H content decreased, aliphatic C-H decreased,
0 and COOH contents increased, and total acidity increased according to
the following order of extractable organic matter: hexane-Soxhlet
extract, benzene-Soxhlet extract, ethylacetate-Soxhlet extract, acetone-
Soxhlet extract, 2-propanol-Soxhlet extract, methanol-Soxhlet extract,
NaOH extractDMF extracthumic acid, fulvic acid.
Solubility Characteristics of Extractable Organic Matter
Solubility characteristics of extractable organic matter in
selected solvents and in salt solutions are summarized in Tables 5-7.
Humic acid, and DMF- and NaOH-extractable organic matter were completely
soluble at 0.1% concentration only in DMF, dimethylsulfoxide (DMSO), and


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
This dissertation was submitted to the Graduate Faculty of the College
of Agriculture and to the Graduate Council, and was accepted as par
tial fulfillment of the requirements for the degree of Doctor of
Philosophy.
December, 1976
Dean,/ pire
Dean, Graduate School


50
Table 12. Peak elution volumes of organic standards on Porasil AX
with selected solvents
Solvent
Sample
Ho0
ch3oh
2-Propanol t-
-Butanol
Acetone
THF
ml
1,2,4,5-Tetracarboxy-
benzene
1.83
2.58
10.25
12.00

3.48
1,2,5-Tricarboxy-
benzene
2.11
2.71
4.95
3.49

3.39
3,5-Dihydroxybenzoic
acid
2.53
2.85
3.59
3.28

3.46
Benzoic acid
2.39
2.97
3.77
3.40
3.49
3.34
Pyridine
11.06
3.42
3.67
ADa
3.84
3.60
Aniline
4.69
3.38
3.52
3.55
3.52
3.49
Methanol
3.48

3.44
3.41
3.65
3.71
Ethylene glycol
3.56
3.43
3.52
3.56
3.65
3.40
Acetone
3.58
3.38
3.44
3.46

3.40
aAD = severe adsorption


36
0.5 N NaOH. The DMF and DMSO both have significant basic character
(Talhoun and Mortland, 1968). The acidic organic material is highly
dissociated and dispersed in each of these solvents and, therefore, is
soluble. Each of the Soxhlet fractions and fulvic acid were also com
pletely soluble at 0.1% concentration in DMF and DMSO.
Fulvic acid was soluble in methanol and water, in addition to DMF
and DMSO, but was not completely soluble in any of the other solvents.
The solubility of extractable organic matter and Soxhlet fractions was
influenced to a great extent by the saturating cation (Table 6). Ex
change of H+ by Na+, K+, N(CH^)^+, or C(C^Hg)^+ resulted in increased
solubility of the organic solutes in water, methanol, or 2-propanol,
and decreased solubility in DMF. For example, at 0.1% concentration,
the H-saturated DMF- and NaOH-extractable materials were only partially
soluble in water, methanol, or 2-propanol; however, the salt-saturated
solutes were completely soluble. On the other hand, the H-saturated
2-propanol-Soxhlet fraction and fulvic acid were soluble at 0.1% concen
tration in DMF whereas the salt-saturated material was only slightly
soluble. These solubility characteristics greatly limit the solvent-
electrolyte combinations which are applicable for exclusion chroma
tography. The enhanced solubility of the cation-saturated samples in
the protic solvents (water, methanol, and 2-propanol) may be attributed
to acidic properties of these solvents (King, 1973) which promote
stabilization of the solute anion. The very weakly acidic dipolar
aprotic solvents (e.g. THF, DMF) would not stabilize the solute anion
to as great an extent as the protic solvents.


67
lili I
10 20 30 40 50
ELUTION VOLUME, ml
Fig. 14 Molecular weightocalibration curve of U-St^tagel (2 f x
0.954 cm 0D 100 A + 2 f x 0.954 cm OD 500 A) obtained
by elution of polystyrene standards with THF


Table
LIST OF TABLES
Page
1 Parameters of column packing materials 23
2 Yields of extractable soil organic matter 26
3 Elemental and functional group concentrations of
extractable soil organic matter 27
4 Titratable acidity of extractable soil organic matter 28
5 Solubility of extractable soil organic matter in
selected solvents at 0.1% concentration 32
6 Solubility of extractable soil organic matter as
influenced by saturating cation and solvent 34
7 Solubility of fulvic acid in aqueous salt solutions . 35
8 Peak elution volumes of extractable soil organic matter
on Porasil A with selected solvents 44
9 Peak elution volumes of extractable soil organic matter
on Porasil AX with selected solvents 45
10 Peak elution volumes of extractable soil organic matter
on CPG-250 with selected solvents 46
11 Peak elution volumes of organic standards on Porasil A
with selected solvents 49
12 Peak elution volumes of organic standards on Porasil AX
with selected solvents 50
13 Peak elution volumes of organic standards on CPG-250
with selected solvents 51
14 Peak elution volumes of cation-saturated fulvic acid on
Porasil A, Porasil AX, and CPC-l^n with water as
eluting solvent 56
vx


51
Table 13. Peak elution volumes of organic standards on CPG-250 with
selected solvents
Solvent
Sample 1^0 CH^OH 2-Propanol t-Butanol Acetone THF DMF
ml
1,2,4,5-Tetracarboxy-
benzene
1.71
1.74
4.12
10.03
--
3.91
2.44
1,3,5-Tricarboxyben-
zene
1.78
1.93
ADa
4.27

3.93
3.38
3,5-Dihydroxybenzoic
acid
2.18
5.03
4.06

3.91
3.86
Benzoic acid
2.26
5.36
4.99
3.98
3.92
3.88
Pyridine
4.89
AD
AD
5.02
4.34
4.17
Aniline
4.12
4.72
4.76
4.17
4.10
3.95
Methanol

3.98
4.27
4.26
3.87
4.31
Ethylene glycol
3.87
5.01
5.38
4.37
3.89
4.26
Acetone
3.82
4.34
4.87
4.06
3.92
a
AD
severe adsorption


85
Forss, K.G. and B.G. Stenlund. 1975. The influence of charged groups
in gel permeation chromatography on polyelectrolvtes. J. Polym.
Sci. 42:951-963.
Freeman, D.H. 1973. The gels for liquid chromatography. J.
Chromatogr. Sci. 11:175-180.
Fritz, J.S., S.S. Yamamura, and E.C. Bradford. 1959. Determination
of carbonyl compounds. Anal. Chem. 31:260-263.
Gaylor, V.F., H.L. James, and H.H. Weetall. 1976. Gel and affinity
chromatography. Anal. Chem. 48:44R-51R.
Gelotte, B. 1960. Studies on gel filtration sorption of the bed
material Sephadex. J. Chromatogr. 3:330-342.
Giddings, J.C. 1965. Dynamics of chromatography: principles and
theory. Marcel Dekker, New York.
Halstead, R.L., G. Anderson, and N.M. Scott. 1966. Extraction of
organic matter from soils by means of ultrasonic dispersion
in aqueous acetylacetone. Nature 211:1430-1431.
Hansen, E.H. and M. Schnitzer. 1967. The alkaline permanganate
oxidation of Danish illuvial organic matter. Soil Sci. Soc.
Amer. Proc. 30:745-748.
Hansen, E.H. and M. Schnitzer. 1969. Zinc-dust distillation of soil
humic compounds. Fuel 48:41-46.
Hawk, G.L., J.A. Cameron, and L.B. Dufault. 1972. Chromatography of
biological materials on polyethylene glycol treated controlled
pore glass. Prep. Biochem. 2:193-203.
Haworth, R.D. 1971. The chemical nature of humic acid. Soil Sci.
111:71-79.
Hiatt, C.W. et al. 1971. Treatment of controlled-pore glass with
poly (ethylene oxide) to prevent adsorption of rabies virus.
J. Chromatogr. 56:362-364.
Hurst, H.M. and N.A. Burges. 1967. Lignin and humic acids. In
A.D. McLaren and G.H. Peterson (eds.), Soil biochemistry.
Marcel Dekker, New York.
Karger, B.L. 1971. The relationship of theory to practice in high
speed liquid chromatography. In J.J. Kirkland (ed.), Modern
practice of liquid chromatography. Wiley-lnterscience,
New York.


To my parents,
Richard and Adeline Loeppert


55
Therefore, electrostatic exclusion of negatively charged solute was
reduced on Porasil AX compared to Porasil A. When acetone, ethylace-
tate, or THF was used as the eluting solvent on Porasil AX, only a small
quantity of acidic solute was eluted past V This phenomenon indicates
a reduction in adsorptive interaction between the silica packing
material and the acidic solute on Porasil AX compared to Porasil A.
Adsorption and electrostatic exclusion were reduced on Porasil
AX, but were not completely eliminated. The evidence of adsorption and
electrostatic exclusion interactions between the solute and the Porasil
AX demonstrated that the packing material was not completely deactivated.
Effect of saturating cation. Fulvic acid in which the acidic
functional groups were saturated to pH 7.0 with Na+, K+, NCCH^)^"*", or
N(C,H0),+ were eluted at on Porasil A and CPG-250 when water was used
4 9 4 0
as the eluting solvent (Table 14). The cation-saturated samples were
also completely excluded from the gel matrix on deactivated Porasil AX.
Likewise, both the 1,2,4,5-tetracarboxybenzene and the 1,3,5-tricar-
boxybenzene in methanol and water were excluded from the Porasil A and
Porasil AX gel matrices.
The pronounced exclusion of Na-, K-, N(CH),-, and N(C.H) -
3 4 4 9 4
saturated fulvic acid and organic acid standards from the Porasil A
gel matrix may be attributed to electrostatic repulsion of the nega
tively charged solute molecule and the negatively charged sites on the
silica surface. Similar exclusion phenomena have been observed during
elution of cation-saturated fulvic acid with distilled water on
Sephadex (Swift and Posner, 1971).


57
Electrostatic exclusion phenomena have been reported for elution
of acidic amino acids (Gelotte, 1960), aromatic acids (Demetriou et al.,
1968), inorganic ions (Neddermeyer and Rogers, 1968), and lignosulfonate
(Forss and Stenlund, 1973) on the Sephadex G-gels and was attributed to
electrostatic repulsion between fixed charges on both the gel and the
solute molecules.
Effect of excess electrolyte. Elution patterns of Na-saturated
fulvic acid in the presence of excess neutral salt on Porasil A and
Porasil AX (Figs. 8 and 9, respectively) indicated that electrolyte
resulted in reduction in the relative quantity of solute excluded from
the gel matrix. Presence of neutral salt also influenced elution pat
terns of low-molecular weight organic acids on Porasil A and Porasil AX
(Figs. 10 and 11, Tables 15 and 16) and resulted in reduced exclusion
of solute from the gel matrix. Similar phenomena have been observed
with Sephadex during the elution of soil organic matter extracts (Swift
and Posner, 1971; Posner, 1963), acidic amino acids (Gelotte, 1960),
aromatic acids (Demetriou et al., 1968), and lignosulfonates (Forss
and Stenlund, 1973).
At salt concentractions above 0.01 N, significant quantities of
fulvic acid were eluted past The excess electrolyte resulted in
suppression of negative charge and reduction in thickness of electrical
double layer of both the negatively charged gel surface and the organic
solute molecules. Also, excess sal t would decrease the effective size
of solute anions due to reduction in thickness of the electrical double
layer. Therefore, solute anions would more easily enter the porous