Molecular weight distribution and chemical nature of organic matter in soil chronosequences

Abstract

A method of extracting a reasonable amount of organic carbon from the soil with minimum damage to the soil organic matter during extraction and purification was developed. This rapid extraction method was a modification of the method of Choudhri and Stevenson (1957). It employed a series of acid pre-treatments followed by repeated mild 0.1M Na₄P₂O₇ (pH 7.0) extractions and subsequent 0.5M NaOH extractions. The ash content of the extracted soil organic matter was reduced by using a series of treatments which included centrifuging, filtering, dialysis and ion exchange. Fractionation of purified organic matter extracts into nominal molecular weight fractions was accomplished by means of gel filtration.

The above method was used to extract organic matter from the soils (A horizons) of the Manawatu chronosequence. Between 84 and 68 per cent of the total solI organic carbon was released by the extractants (i.e. total 0.1M Na₄P₂O₇ (pH 7.0) + O.5M NaOH + 0.5M NaOH at 60° C). After purification these values were between 18 and 26 per cent. This indicates a large proportion of the material was lost during the purification. However, this would include mainly low molecular weight nonhumic organic matter.

Effects of acid pre-treatment and acid fractionation on the molecular weight distribution of 0.1M Na₄P₂O₇ (pH 7.0) organic matter extracts were investigated. Acid pre-treatment of the soil with 0.1M HCl followed by 0.1M HCl : 0.3M HF was found to enhance the polydispersion in the nominal molecular weight of the extracts. The same treatment resulted in significant increases in yield and reduction in ash content. For this reason, the organic matter released by the various pretreatments was recovered and combined with the 0.1M Na₄P₂O₇ (pH 7.0) extract, termed total 0.1M Na₄P₂O₇Na₄P₂O₇ (pH 7.0) extract. Prolonged standing of extracts in the acids led to acid-induced polymerization, resulting in a predominance of organic matter in the higher molecular weight ranges.

It was found that the fractionation of the total 0.1M Na₄P₂O₇ (pH 7.0) extracts by acid precipitation into humic and fulvic acids did not separate them according to molecular weight as commonly believed. Instead, fulvic acids from most soils were found to have similar nominal molecular weight distributions to those of their humic acid counterparts, with a large proportion of soil fulvic compounds in the >100,000 nominal molecular weight range.

The extraction method was adapted to handle large soil samples and large volumes of extracts. It was used to extract organic matter from the A horizons of soils from the Manawatu, Reef ton and Franz Josef chronosequences. The extracted organic matter was fractionated into five nominal molecular weight ranges (>200,000; 200,000-100,000; 100,000-50,000; 50,000-10,000; <10,000). With increasing soil development, the results show that the proportion of large molecules decreased, with a corresponding increase in the proportion of organic matter present in the intermediate molecular weight fractions. This was more evident in the molecular weight fractions obtained by the 0.5M NaOH extracts those from the mild total 0.1M Na₄P₂O₇ (pH 7.0) extracts. Differences in the molecular weight distribution of organic matter were also observed when there was a change from native to pasture vegetation in the Koputaroa soil, probably due to the nature of the plant residue (Kononova, 1966), and the rate of turnover of organic matter in these soils.

The distribution of C, N, P, S, and oxygen-containing functional groups in the soils and purified total 0.1M Na₄P₂O₇ (pH 7.0) and subsequent O.5M NaOH extracts were determined. They were found to increase with soil development in the soils (A horizons) of the Manawatu chronosequence. This trend was not always apparent in the soils (A horizons) of comparable ages in the Franz Josef and Heefton chronosequences, possibly - due to the more advanced state of development of the Westland soils. The effect of vegetation on C, N, P and S distribution appeared most evident in the Motuiti soil of the Manawatu, chronosequence. The exceptionally high amount of soil C and N and extractable C, N, P and S in this soil may be due to the presence of bracken fern.

More organic C was extracted from all the soils studied by the subsequent O.5M NaOH than by the total 0.1M Na₄P₂O₇ (pH 7.0) except the Koputaroa pasture soil. Using the large-scale extraction technique between 21 and 49 per cent of the total soil organic C was obtained in the two purified soil extracts. These are higher than those reported earlier (18 - 26%), using the small-scale extraction method, because in the large-scale method the 0.5M NaOH extract was repeated until no further soil organic matter was extracted.

Generally speaking, more P and S were present in the organic matter released by the total 0.1M Na₄P₂O₇ (pH 7.0) than by the subsequent 0.5M NaOH, whereas the reverse was true for N. However, this was not generally observed for the Franz. Josef soils, especially the oldest soil, Okarito, since in these soils, the more readily mineralizable forms of N and P removed by the total 0.1M Na₄P₂O₇ (pH 7.0) extractant were low.

Definite differences were observed in the chemical nature of the two molecular weight fractions (>50,000 and <50,000) of soil organic matter in soils from the three chronosequences. Lower ratios of N, P, S and oxygen-containing functional groups relative to C were found in the >50,000 molecular weight fraction than in the <50,000 molecular weight fraction. Further-more, the chemical nature of the >50,000 molecular weight fraction remained unchanged with soil age, whereas the chemical nature of the <50,000 molecular weight fraction changed with soil age, and appeared to be affected by vegetation.

These data suggested that there were two different fractions of organic matter in most soils; a large stable fraction, represented by the >50,000 molecular weight fraction, and a small active fraction represented by the <50,000 molecular weight fraction. The results strongly support a two-phase system of N, P and S cycles in soils. Furthermore, fertilizer P and S resulting from topdressing of pastures were found to be incorporated into the active pool of organic matter (i.e. <50,000 molecular weight fraction.