Tải bản đầy đủ - 0 (trang)
III. Chemical Methods of Functional Group Analysis

III. Chemical Methods of Functional Group Analysis

Tải bản đầy đủ - 0trang

STRUCTURAL CHEMISTRY OF SOIL HUMIC SUBSTANCES



343



they obtained 2.96 meq./g. by KOH, 2.54 by NaOH, 1.83 by Ba acetate,

and 0.69 by K acetate, The differences were explained in part by the

formation of inner complex compounds as well as normal salts. Steric

considerations were also thought to be important.

2. Reduction with Diborane

Martin et al. (1963) measured the H2 produced by the reaction of

diborane in inert solutions (e.g., tetrahydrofuran) with the H of COOH

and OH groups as an estimate of the total active hydrogen of soil organic

preparations. Independence of pK values was cited as one of the

advantages of diborane as a method of determining functional groups.

In addition, the small molecular size of diborane reduces the problem of

steric hindrance that appears to be a problem in the case of other

reagents ( Broadbent and Bradford, 1952). The carbonyl group of the

carboxyl function along with other carbonyl groups is readily reduced

to the corresponding alcohol without the production of Ha. On the

other hand, esters and lactones are reduced only very slowly. The active

hydrogen determined by this method for a Podzol BH organic preparation (11.6 meq./g.) was in satisfactory agreement with values of

total acidity obtained by the Ba( OH)e method (11.9 meq./g.).

3. Potentiometric Titration in Nonuqueous Solvents



Wright and Schnitzer ( 1960) titrated potentiometrically organic

preparations dissolved or suspended in pyridine, dimethyl formamide,

and ethylenediamine with sodium aminoethoxide. The values obtained

were comparable with those obtained by other methods. The curves

had one inflection point that corresponded approximately to the free

carboxyl groups. Dimethyl formamide gave only one inflection point with

a polyfunctional compound, salicylic acid, so it apparently was not a

satisfactory solvent medium for organic matter titrations. Although

pyridine and ethylenediamine both showed two inflections for salicylic

acid, only ethylenediamine gave reasonably satisfactory curves for

titrations with organic matter preparations.

Van Dijk ( 1980) compared potentiometric, conductometric, and

high frequency titrations of humic acid preparations in a variety of

aqueous and nonaqueous media. He concluded that the high frequency

titration of humic acids in dimethyl formamide using sodium isopropylate

as the titrant gave good results for a certain type of acid grouping,

presumably carboxyl. However, the method gave far less satisfactory

results for determination of the weaker acidic groups, such as phenolic

hydroxyls. It was suggested that these latter groups could probably best

be determined by measuring the total acidity by conductometric titration



344



G . T. FELBECK, JR.



with Ba( 0H)z and then subtracting the content of strong acid groups to

obtain an estimate of the content of weak acid groups.

Pommer and Breger ( 1960) carried out discontinuous potentiometric

titrations for periods up to 52 days on a humic acid extracted from peat.

They observed that the equivalent weight of this humic acid increased

from about 119 to about 183 in a 52-day period. This increase in

equivalent weight was coincident with a loss of carbonyl groups as

determined by infrared spectroscopy. Several mechanisms were proposed

which might account for these observations. Among those suggested

were the formation of hemiacetals, the keto-enol equilibrium, and the

aldol condensation process.

B. CARBOXYL

GROUPS

1. Ion Exchange Methods

For the reasons pointed out earlier under the discussion of total

acidity, the method of ion exchange gives ambiguous results when an

attempt is made to apply it to the direct differentiation of COOH groups

from phenolic OH groups. However, an indirect measure of the COOH

content can be obtained by subtracting the phenolic groups as determined by nonsaponifiable methyl content on methylation from total

acidity as determined by ion exchange.



2. Decarboxylation

Many COOH-containing organic compounds yield C02 when heated

either in the dry state or in a suitable solvent. This method has been

used for the estimation of COOH in organic preparations in a Podzol

soil by Wright and Schnitzer (1960). In a comparative study with

known compounds, 98% of the theoretical value was obtained when

anthranilic acid was heated, but only 79.9%when citric acid was similarly

treated, Values obtained for soil organic preparations of unknown

structure were generally lower than those obtained by ion exchange

methods.

3. lodometric Methods

Although this method is primarily based on an ion exchange, the

I- released is oxidized to 12 by KI03, and then irreversibly taken up

by thiosulfate (Van Krevelen and Schuyer, 1957). Wright and Schnitzer

(1960) generally obtained higher values for COOH groups by this

method than they obtained by Ca acetate methods.

4. Esterification

Two methylating reagents, diazomethane and dimethyl sulfate, have

been applied to the study of functional groups in soil organic matter



STRUCTURAL CHEMISTRY OF SOIL HUMIC SUBSTANCES



345



preparations ( Broadbent and Bradford, 1952). Diazomethane is reported

to react with a variety of acidic hydrogen functions, including carboxyl,

phenolic hydroxyl, some keto-enol compounds, H attached to N adjacent

to a carbonyl group, and alcoholic groups adjacent to an acidifying group.

On the other hand, dimethyl sulfate will react with all the above groups

except COOH and in addition will react with some of the more weakly

acidic groupings that do not react with diazomethane.

In a methylation study of seven soils Broadbent and Bradford (1952,)

found that over half the average exchange capacity of the total organic

matter was due to COOH groups. This value was arrived at by two

methods. First, the reduction in exchange capacity by methylation with

dimethyl sulfate was determined to be 46*12%, indicating that approximately 54% of the exchange capacity was due to COOH. Secondly, the

organic material was exhaustively treated with diazomethane and then

saponified. By this method about 74*19% of the total methyl content

was released, which was taken as an indication of the COOH content,

assuming no other groups could be saponified after methylation with

diazomethane.

C. HYDROXYL

GROUPS

A variety of hydroxyl groups have been reported as possibly present

in soil organic matter: phenols, alcohols, enols, hydroxyquinones ( Scheffer and Ulrich, 1960; Martin et al., 1963),and pyrones (Felbeck, 1965).

Chemical methods used to determine all or part of the hydroxyl groups

include acetylation, etherification, reaction with dinitrofluorobenzene,

and nonaqueous titrations.

1. Total Hydroxyls by Acetylation

Acetic anhydride in pyridine at 90°C. has been used as a method

of estimating the total hydroxyls by esterification (Wright and Schnitzer,

1960; Martin et al., 1963). This reagent reacts with both primary and

secondary amines and sulfhydryl groups as well as alcohols and phenols

( DeWalt and Glenn, 1952). However, Wright and Schnitzer ( 1W)

estimated that the total level of these interfering materials would not

usually produce a positive error in the estimations of total hydroxyls in

excess of 1 meq./g. of organic matter.

In this method, the hydroxyl content of organic matter can be

measured either by saponifying the ester and measuring the acetic acid

produced, or by determining the unreacted acetic anhydride in the

original mixture. In comparing these two approaches, Martin et al. (1963)

found much higher values by the latter method than by the former. They

concluded that estimating hydroxyl content by the titrimetric determination of unreacted acetic anhydride gave unreliable results.



346



G . T. FELBECK, JR.



2. Phenolic Hydroxycyls

As with COOH groups, no clear-cut method is available which reacts

stoichiometrically with phenolic hydroxyl groups. Values obtained by

ion exchange or potentiometric titration methods have to be corrected

by subtracting values for the COOH as estimated by an independent

method, with all the errors attending such a procedure.

a. Ubaldini procedure. Wright and Schnitzer (1959b) have applied

a modified Ubaldini procedure to the estimation of phenolic hydroxyls

in soil organic matter preparations from a Podzol (Ubaldini and

Siniramed, 1933; Mukherjee et al., 1957). Basically this procedure involves refluxing the organic material with an excess of alcoholic KOH.

After filtration and washing to remove the alkali, the residue is SUSpended in 85% alcohol and saturated with COP.The material is filtered

and washed, and the liquid phase is titrated against standard acid to

determine K2C03. Each mole of K+ released by C 0 2 saturation is

equivalent to a mole of phenolic hydroxyl.

Wright and Schnitzer (1959b) obtained values of approximately

3 meq./g. phenolic hydroxyl for each of their Podzol preparations.

Similar results (2.8 meq./g.) were obtained by Mukherjee et al. ( 1957)

in a study of humic acids from an Indian lignite. However, estimations

of total hydroxyl in the humic acid from lignite by acetylation gave the

substantially higher value of 4.6 meq./g., indicating a level of alcoholic

OH in this material of about 1.8 meq./g.

b. Etherification with dimethyl sulfate. Dimethyl sulfate reportedly

will react with all groups capable of reacting with diazomethane except

carboxyl. In addition, it apparently reacts with more weakly acidic

groups that do not react with diazomethane. This nonspecificity, plus

the possibility of side reactions in the strongly alkaline medium required,

casts doubt on the reliability of dimethyl sulfate as a measure of phenolic

hydroxyl groups (Broadbent and Bradford, 1952; Lewis and Broadbent,

1961).

3. Alcoholic Hydroxyls



The status of alcoholic hydroxyls is completely unresolved at present.

They have been reported to be present in lignite humic acid (Mukherjee

et al., 1957) and absent in humic acid from a Podzol BHhorizon (Martin

et al., 1963).Because the reactivity of alcoholic groups is lower than that

of phenolic groups, present methods of detecting the former in the

presence of the latter appear to be quite unreliable.



STRUCTURAL CHEMISTRY OF SOIL HUMIC SUBSTANCES



347



D. CARBONYL

GROUPS

Of the three major oxygen-containing functional groups, methods

available for determining carbonyl groups are the least satisfactory

(Scheffer and Ulrich, 1960; Dubach and Mehta, 1963). Among the

chemical reagents used for detecting this group are hydroxylamine and

sodium borohydride.

The reaction of hydroxylamine with vanillin gave satisfactory results

(Wright and Schnitzer, 1960), but the accuracy of this method when

applied to organic matter preparations is doubtful.

Martin et al. (1963) determined carbonyl groups by reactions with

NaBH4. They obtained the expected values with p-benzoquinone. With

seven fractions from a Podzol soil they obtained values ranging from

1.0 to 2.5 meq./g. which compared favorably with Wright and Schnitzer’s

(1980) values of 1.0 to 3.1 meq./g. from a Canadian Podzol using hydroxylamine.

Meyer obtained an unexpectedly high value for carbonyls (14

meq./g.) using NaBH4 (as cited by Martin et al., 1963).



E. REMARKS

The results obtained by chemical methods of functional group analysis

are rather difficult to interpret and often contradictory. A number of

reasons might account for this difEculty. Among those given by

Dubach and Mehta (1963) are: ( 1 ) divergent origins of humic substances and lack of adequate criteria of purity; (2) variable molecular

weights, leading to incomplete reactions, adsorption of reagents, and

undesirable fractionations during manipulations; and ( 3) proximity of

many and different groups, which influence both the reactivity of the

groups and the specificity of reagents used in their detection and measurement.

In contrast, Schnitzer and Gupta (1965) after an examination of

the Ca(0Ac)Z and Ba(0H)z methods for determining COOH groups

and “total acidity” concluded that functional group analysis is the

most important tool presently available for characterizing the reactivity

of purified organic matter extracts. They based their conclusion, in part,

on the observation that work on “model” compounds may be misleading

because the main structural features of humic and fulvic acids are

presently unknown.

IV. Spectroscopy



Spectrographic methods are among the most useful of the nondestructive methods of examining unknown chemical compounds. The range

of spectra covers the region from the X-ray and ultraviolet through the



348



C. T. FELBECK, JR.



visible and infrared region to the area of the radiofrequencies, as exhibited by nuclear magnetic resonance. The usefulness of each of these

procedures is proportional to the purity of the compound being examined.

As with many procedures, spectroscopic data are more useful and more

easily interpreted when they are combined with other analytical and

structural data. Since humus compounds have never been prepared in a

“pure” state and separate laboratories seldom work on the same soil

organic matter preparations, correlation and interpretation of spectral

data have been rather chaotic. As extraction, fractionation, and purification procedures are improved, spectral data will become more useful.

A. ULTRAVIOLET

SPECTROSCOPY

The general application of ultraviolet spectral techniques in the area

of organic structural analysis has been reviewed by Brand and Scott

(1963). Kumada (1955, 1958) examined the ultraviolet and visible

absorption spectra of a number of soil and microbial extracts along with

several known compounds. The spectra of the soil extracts were rather

featureless, with the absorption increasing rather regularly as the wavelength was decreased. From a similarity in the spectra Kumada (1958)

suggested that anthraquinone might be one of the structural units in

the humic acid molecule, but this suggestion needs to be verified by

independent methods.



B. INFRARED

SPECTROSCOPY

In contrast to the rather featureless aspects of the UV spectra, the IR

spectra of humic preparations are noted for their complexity. Mortensen

and H h e s ( 1964) listed approximately twenty characteristic infrared

group frequencies found in organic matter preparations. Cole (1963)

has reviewed the application of IR spectroscopy to the general area of

structural elucidation of organic compounds.

Kumada and Aizawa (1958) examined humic acids from a group of

Japanese soils and concluded on the basis of the IR spectra that the

humic acid contained hydrogen-bonded OH, aromatic and aliphatic

CH groups, COOH, C=O, and C=C. The main structural entities

were thought to consist of various aromatic and aliphatic compounds

including phenols, quinones, ethers, and alcohols.

Johnston ( 1959) separated two fractions by curtain electrophoresis

from the acid-resistant fractions of humic acids from several soils. These

fractions, one a light-orange colored, highly fluorescent material and

the other dark colored and nonfluorescent, were examined by IR.

Elemental analyses were performed on the purest fraction of one of the

light-colored samples. The lighter-colored material had 6.70% CH30



STRUCTURAL CHEMISTRY OF SOIL HUMIC SUBSTANCES



349



and was thought to be aromatic in nature, Only a slight shoulder at

5 . 8 5 ~was observed, indicating that C=O groups might not be an

important part of this structure. For the dark-colored fractions IR bands

at 3.00, 3.45, and 5 . 8 5 ~were ascribed to OH, CH and C=O groups,

respectively.

The organic matter from the BH horizon of a Canadian Podzol was

extracted with a series of extractants by Schnitzer et al. (1959). They

examined the organic extracts before and after treatment with HF. The

IR spectra of the various extracts were very similar, indicating that the

nature of the extractant was not important for this soil so long as a

relatively high percentage of the organic matter could be extracted and

the extractant could then be removed from the organic fraction. From

the IR spectra hydroxyl and carboxyl groups were identified, but no

bands were assigned to C=O groups.

In an effort to identify more definitely the assignments of functional

groups to individual absorption bonds Orlov et al. (1962) examined

humic acids extracted from a variety of soil groups before and after

oxidation with alkaline KMn04, HN03, and H202. A summary of their

observations is presented in Table I.

From a study of the IR spectra Orlov et al. (1962) concluded that

humic acids from various soil groups are characterized by a common

structural pattern. They postulated the nucleus of humic acid as consisting of highly substituted benzene or possibly pyridine rings, a

considerable number of which were conjugated with C=C or C=O

bonds. The carboxyl groups were also thought to be conjugated as in

benzene carboxylic acids. The role of terminal CHQ or bridge CH,

groups was considered to be small, but the data on this are open to

question since paraffin oil was used as the supporting medium in the

IR studies.

An examination of the IR spectra of a whole soil, a finely ground,

thick Chernozem, indicated that there were no substantial differences

between the whole soil and the humic acid. It was concluded that

extraction with alkali had little effect on the periphery of the molecule

and practically none at all on the nucleus.

In a Werent approach to the study of changes in IR spectra of

humic acids by treatment, Wagner and Stevenson (1965) examined

methylated and acetylated extracts from a Brunizem. The principal

observations of their study are included in Table I. They concluded that

humic acid contains isolated carboxyl groups leading to mixed anhydrides, adjacent carboxyl groups that when acetylated lead to 5-membered cyclic anhydrides on heating with KBr, and at least two kinds of

phenolic groups, i.e., those that could and those that could not easily be



Wavelength

(PI



TABLE I

Principal Effects of Various Treatments on the Infrared Absorption Spectra of Humic Acids

Band frequency

(cm.-1)

Proposed assignment

Effect of treatment



2.9-3.0

2.94



3380

3400



H-bonded OH

OH stretch



3.35

3.44

3.55

3.42



2985

2900

2820

2920



CH, and CH,

CH, and CH,

CH, and CH,

CH stretch of CH,



3.65

3.80-3.85

4.34.5

5.43



2740

2610

2260

1840



H-bonded OH

H-bonded OH

H-bonded OH



5.51

5.60



1815

1785



5.81



1720



5.9



1695



Higher C=O stretch of

cyclic anhydrides

C=O of m i x e d anhydrides



COOH



Referencea



Not changed on prolonged drying

Band reduced to 2500 to 3100

cm.-1 on methylation



-



Not present in original samples,

but appeared after methylation

Increases considerably on

oxidation



i



C=O stretch of phenolic

acetates and cyclic anhydrides



C=O stretch



E



Remained constant in intensity

on prolonged acetylation

Increased in intensity on

prolonged acetylation of

methylated and saponified

sample

( a ) Intensity increased on

methylation

( b ) Acetylation and

saponification of a previously

methylated sample reduced

intensity



b

b



b

a



6.0



1665



Olefinic C=C



6.18-6.26



1610



Conjugation of C=C in ring with

C=C or C=O of open chains,

also partly due to heterocyclic

compounds.



6.2



1613



C=C and COO-



6.25-6.29

6.44-6.48



1590

1550



Multinuclear aromatic C=C

Aromatic C=C



6.35



1575



Salts of COOH



6.8



1470



Aromatic C=C



Decreases somewhat after

oxidation with concentrated

KMn04



1



-



a



Clearly distinguishable after

oxidation



a



Appeared after titration to pH 9

with dilute NaOH

Band appears after KMnO, and

HNO, oxidation, but not after

H202 oxidation



1440



CH stretch of methyl



Not present in original samples

but appeared after

methylation



7.20



1390



Salts of COOH



Appeared on titration to pH 9

with dilute NaOH

Sharpened considerably after

methylation of phenolic OH.

No signscation alteration after

saponification



a



1280



C - 0 stretch



a



Remains after more vigorous

KMn04 oxidation



6.95



7.80



a



a



b



%

v)



a



F



1

K

b



3



b



21

!



2



s



B

b



References: a: Orlov et 02. (1962); b: Wagner and Stevenson (1965).



w



9



352



G. T. FELBECK, JR.



methylated with diazomethane. A possible explanation of the last

observation was that these phenolic hydroxyls were H-bonded with

quinone groups; however, bonds characteristic of quinone were not

detected on the IR spectra.

C. MAGNETIC

RESONANCE

SPECTROSCOPY

The published applications of magnetic resonance spectroscopy to

studies of soil humic substances to date have been in two groups, the

electron paramagnetic resonance ( EPR) spectroscopy of large polymers

extracted from the soil and the nuclear magnetic resonance (NMR)

spectroscopy of some of the degradation products of humic substances.

Basically EPR spectroscopy is a technique for measuring electron spin

resonance of large free radicals, whereas NMR spectroscopy is of most

use in determining the nature of the protons in smaller organic molecules.



I. Electron Paramagnetic Resonance

Rex (1960) appears to have been the first to investigate free radicals

in lignins and humic acids by means of their EPR spectra. He found

strong EPR signals ( 10l8 spins/g.) in HC1-catalyzed dioxane extracts

of wood, in pine needles that had already been attacked by fungi although they were collected directly from the tree, and in plant debris

which had collected in the soil as peat. Material extractable from the

peat with aqueous base had a sharp EPR signal, whereas the residue

was characterized by a broad signal. On the other hand, freshly cut

redwood ranging from sapwood to rotted heartwood (40 years old) in

addition to wood buried up to 600 years in damp soil (presumably,

in each case, the plant lignin had not been appreciably modiiied) showed

no, or only barely detectable, EPR signals. From these observations he

suggested that lignin does not normally occur in living plants, but that

it is the polymerization product between the semiquinone free radicals

of dehydrogenated or hydrolyzed plant tissue and other compounds

capable of reacting with them. These free radicals are capable of surviving throughout geological time ( lo8 years).

In a more intensive series of studies on the EPR of soil humic acids

(Steelink and Tollin, 1962; Tollin et al., 1963; Steelink, 1964), it was

shown that the EPR signal is an inherent part of the molecular structure

and not due to an occluded impurity, a surface absorption effect, or an

artifact created by the extraction process. The sodium salts of humic

acids had EPR signals substantially greater than the original product,

the signal strength returning to the original level on reacidification. Extensive acid hydrolysis and high temperature oxidation ( CuO NaOH,

N



N



+



STRUCTURAL CHEMISTRY OF SOIL HUMIC SUBSTANCES



353



17OoC.,3 hours) increased signal strength, whereas reduction with Na

metal or LiAlH4 caused only small decreases in spin concentration.

From these observations it was suggested that humic acid consists of

a polymeric network of semiquinone ( I ) and quinhydrone species ( II),

as shown in Fig. 1. The reason for this suggestion (Steelink and Tollin,

1962) was that an analysis of the EPR signal indicated that it could best

be interpreted in terms of existence of two separate absorbing species

whose spectra were superimposed. Since most investigators appear to

favor biosynthetic schemes based on the oxidative polymerization of

phenols from a variety of plant and soil sources, the semiquinone-quinhydrone hypothesis would satisfy these schemes plus the EPR observations

of acid stability and increase in signal with formation of Na salts.



(1)



(11)



FIG.1. Semiquinone ( I ) and quinhydrone (11) species proposed as sources of

electron paramagnetic resonance signals in humic acid. (After Steelink, 1964.)



In a comparison of the EPR signals of lignin and soil organic preparations (Steelink, 1964), it was observed that there was a regular increase in spin concentration in the sequence: Brauns native lignin (lowest

spin concentration), chemically or fungal degraded lignin, fulvic acid,

humic acid (highest spin concentration), the difference between the

degraded lignin and fulvic acid being small.



2. Nuclear Magnetic Resonance

The nature of NMR limits its usefulness to organic molecules that

have been fairly well defined by other techniques. Therefore, an NMR

examination of a large, poorly defined polymer would be relatively

fruitless. The published applications to date of NMR to soil humus studies

have been confined to aiding in the identification of specific fractions or

degradative products of soil humic substances.

Barton and Schnitzer (1963) examined methylated fractions of the

organic material from a Podzol BE1 horizon and observed the absence of

both olefinic and aromatic protons. Both methyl and methylene protons

were detected along with protons on methyl ethers and methyl esters

produced by the preliminary treatment.



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

III. Chemical Methods of Functional Group Analysis

Tải bản đầy đủ ngay(0 tr)

×