influence of chloride on the formation of iron … 32/32-3-167.pdf · influence of chloride on the...

Clays and Clay Minerals, Vol. 32, No. 3, 167-174, 1984.

INFLUENCE OF CHLORIDE ON THE FORMATION OF IRON OXIDES FROM Fe(II) CHLORIDE. I. EFFECT OF [C1]/[Fe] ON THE FORMATION OF MAGNETITE

R. M. TAYLOR

CSIRO Division of Soils, Private Bag 2, Glen Osmond South Australia 5064, Australia

Abstract--The formation of fine-grained magnetite (t0.1 #m) at pH 7 and 25~ from aeration of Fe(II) chloride solutions is presented. The magnetite converted at 105~ to maghemite with poorly developed superstructure lines. Under the experimental conditions employed, as the initial [C1]/[Fe] ratio was increased from the stoichiometric value of 2, the final product contained increasing amounts oflepidocrocite. The degree ofcrystallinity of this phase, as measured by the width at half height of the 020 X-ray diffraction peak, also increased with this ratio. The hydrolysis rate (base consumption to maintain pH) showed a plateau whose position and extent changed with the initial [C1]/[Fe] ratio. Through this plateau region the Eh decreased to a minimum value the position of which was directly related to the [C1]/[Fe] value. The formation of lepidocrocite rather than magnetite is likely due to the high [C1] where there would be increased difficulty for neighboring OH to condense and eliminate HE0. The formation of Fe-O-Fe bonds in this condensation would be impeded by C1 substitution for OH either in the first formed green rust stage or during its oxidation. Key Words--Akaganeite, Chloride, Green rust, Iron, Lepidocrocite, Maghemite, Magnetite, Synthesis.

I N T R O D U C T I O N

In the formation of pedogenic iron oxides the immediate environment in the zone of precipitation con- trois the direction of the reaction and the composition of the products. Many workers have investigated the effects of pH, oxidation rate, temperature, and ionic environment on the mineral phases formed, but often some of these parameters were controlled outside the range that would normally be encountered in soils. Taylor and Schwertmann (1974) showed that during the oxidation of an Fe(II) chloride solution around pH 7 and 20~ lepidocrocite (3,-FeOOH) with different a m o u n t s of goethite ( a -FeOOH) and ferr ihydr i te (FesHOs'4H20) were formed. The addition of some initial Fe(III) or an increase in the temperature or Fe(II) concentration favored the formation of maghemite ('y-Fe203). Similar results were found by Hamada and Kuma (1976) who concluded that the upper limit of temperature for obtaining a pure lepidocrocite phase decreased with increased pH and initial [Fe(II)]. Using high flow rates of CO2-free air (2 liter/min), they produced a pure ferromagnetic phase in the pH range 8- 9.5 from a chloride solution, higher pH favoring goethite and lower pH favoring lepidocrocite or mixtures of lepidocrocite and goethite.

Us ing different exper imenta l parameters m a n y workers have found different end products for the oxidation of Fe(II) solutions. Misawa et al. (1974) found that lepidocrocite formed under slightly acid conditions with goethite being favored as the pH decreased. They maintained, however, that chloride was necessary for the formation of akaganeite 03-FeOOH). Kiyama

Copyright �9 1984, The Clay Minerals Society

and Takada (1972) found that in the oxidation of FeC12 solutions at 70~ in which the pH was allowed to decrease during hydrolysis, goethite, lepidocrocite, and akaganeite formed, the proportion of the akaganeite increasing with higher [C1]/[Fe] ratios. Under the same experimental conditions, they obtained only goethite during the oxidation of FeSO4 solutions, suggesting a definite effect of anion on the reactions. In contrast, Taylor and Schwertmann (1978) showed that at low oxidation rates lepidocrocite formed as the dominant phase during the hydrolysis and oxidation of an Fe(II) sulfate solution around neutral pH and at ambient temperatures.

The many different phases formed during the oxidation of Fe(II) systems suggest that the oxidation rate, which can be influenced by pH, temperature, and other factors, is a dominant factor in determining the hydrolysis products. The rate of oxidation of an Fe(II) solution is to a first order reaction with regards to both [Fe] and dissolved oxygen concentration, and a second order reaction with regards to pH (Stumm and Lee, 1961). The ionic environment can affect the oxidation rate (Stumm and Lee, 1961) and so influence the hydrolysis products in a oxidizing Fe(II) system (Schwert- mann, 1959; Krause and Borkowska, 1963; Taylor and Schwertmann, 1978).

The oxidation products of Fe(II) solutions are important because it is in this valence state that Fe is commonly mobilized during weathering and solubi- lized under the Eh-pH regime of some natural envi- ronments. Taylor and Schwertmann (1974) showed that lepidocrocite and maghemite formed together during

167

168 Taylor Clays and Clay Minerals

the oxidation of an Fe(II) chloride solution at pH 7, but in nature these two minerals are not found together, except when there has been an obvious thermal transformation oflepidocrocite to maghemite. Although the chloride system does not approximate to the normal natural soil environment, and small concentrations of Cl- and SO42- have little effect in a bicarbonate system, which is more realistic in terms of soil conditions and weathering (Stumm and Lee, 1961), further investi- gation of the oxidation-hydrolysis of the chloride system was undertaken to elucidate how some of the environmental factors influence the final products. In these experiments the influence of variations in the initial [C1]/[Fe] ratio on the composition of the oxidation products was examined, and an explanation is offered for the effects produced by the changing [C1] on the reaction rates.

EXPERIMENTAL

The Fe(II) chloride solutions were oxidized in an 80- m m high, parallel-sided glass weighing bottle with an internal diameter of 37 mm. The effective solution/air interface was 8.5 cm 2 after allowing for the area of electrodes and inserts. During the reaction the solutions were stirred by the triangular blade of a Radi- ometer TTA3 titration assembly at the lowest rotation rate (2300 rpm). The reaction vessel was fitted with Pt and calomel electrodes to measure Eh, a glass electrode for pH, a titrant-delivery tube, a delivery tube to allow N2 to flOW above the solution, and another tube to bubble N 2 or air into the solution. All inserts were immersed to a constant depth, and the orifice (0.55 m m diameter) of the air-delivery tube was maintained at a constant orientation with respect to the stirrer blade. The capillary-bore, titrant-delivery tube had a rounded end to minimize localized precipitation and resultant blocking of the exit orifice. These apparently minor details are important because of their influence on the oxidation rate. For example, a nonparallel-sided vessel presents a changing solution/air interfacial area which alters the oxidation rate at the solution surface, and very fast stirring produces turbulence or vortices which render the controlled rate of air bubbling mean- ingiess. The efficiency of stirring is influenced by the depth of immersion of the electrodes and other tubes, and the solution of 02 from the air bubbles is affected by the orientation of the bubble outlet with respect to the stirrer.

A stock solution was made by dissolving 31.3 g Merck AR FeC12.4H20 in 200 ml of N2-saturated H20. The solution was filtered to remove B-FeOOH which often forms as an oxidation product when FeC12.4H20 crys- tals are stored. Then 2.08 g of Fe powder and 7.5 ml of concentrated HC1 were added, and the solution was boiled until the added Fe was dissolved. The solution was again filtered, made up to 500 ml with N2-saturated H20, and then stored in well-stoppered and completely

filled 100-ml volumetric flasks at 5~ Prior to use, the solution was transferred to a series of completely filled 10-ml volumetric cylinders which were stoppered and also stored at 5~ to minimize oxidation. In the ma- jority of the experiments 6.76 ml of Fe(II) chloride solution (0.355 M Fe, 0.72 M C1) was added to 43.4 ml of N2-saturated distilled water into which N2 was bubbling and above which 50 ml /min of N2 also flowed at a constant rate. The pH was adjusted to 7 with 2.5 M NaOH, and the temperature was controlled at either 25 ~ or 0~ When pH 7 was attained, the N2 bubbling into the solution was replaced by 2 ml /min of air, controlled by a peristaltic pump (CO2 was not re- moved). Variations in the [CI]/[Fe] ratio were made by the addition of NaC1 to the original 43.24 ml of N2- saturated water.

When pH 7 was attained and the bubbling of air commenced, a recorder measured the subsequent ad- ditions of NaOH required to maintain this pH during the hydrolysis. The changes in the Eh were also con- tinuously recorded. When hydrolysis was completed and the Eh had reached equilibrium, the oxidation was terminated and the precipitate was recovered by cen- trifugation, washed, and dried from acetone under a heat lamp for a few minutes. The product was then examined by powder X-ray diffraction (XRD) (CoKa radiation) utilizing a monochromator. The integrated peak intensities were measured with a planimeter, and the widths at half height of the diffraction peaks (WHH) were measured from the recorded diffractograms.

RESULTS

During the oxidation at 250C the solution and suspension changed from light green -* dark blue-green -~ yellow-brown similar to the color changes described by Hamada and Kuma (1976) in their preparation of lepidocrocite. In the present work, however, the final suspension was a brownish-black. At the lowest [CI]/ [Fe] ratio used (2.19), the end precipitate was ferromagnetic, and the XRD pattern gave spacings between those of magnetite and maghemite (d(220) = 2.956/~). A trace of lepidocrocite was also detected in the final product. Prolonged drying of the sample in air at 105~ caused the sample to change from black to red-brown, a color change that was accompanied by an observable shift in the XRD pattern towards the spacings of maghemite and the appearance of broad superstructure lines (Figure 1). The line shifts were accompanied by a decrease in diffraction intensity of some lines and an increase in their WHH. Taylor and Schwertmann (1974) synthesized maghemite from an Fe(II) chloride system and suggested that magnetite was possibly the first phase in the formation of the fine-grained maghemites. The transformations of the ferromagnetic precipitate from black to red-brown and the shift in the XRD spacings in the present work support this earlier suggestion. The ease of oxidation of this magnetite to maghemite agrees

Vol. 32, No. 3, 1984 Effect of [C1]/[Fe] on magnetite formation 169

P e a k

I n t e n s i t y

I Superstructure lines af ter oxidation a1105 "C for 2 hr

I [ I I i I i i i i r i v I i i i i 44 40 3 6 32 28 24 lO 16 12

~ O e g r e e s 2 e - C o Ko~radiation

Figure 1. X-ray powder diffration pattern of magnetite and its oxidation product (maghemite) after 1 hr at 105~

Area (020) Lepidocrocite Peak (cm 2)

i o

/ , , 2 4 6 a

[CI]/[Fe] m o l a r r a t i o

iI

Area (220) ,o Maghemite

Peak (cm z)

Figure 3. Variation in X-ray powder diffraction peak areas of the oxidation products of 0.048 M Fe(]I) solutions with increasing [C1]/[Fe] values.

with FarreU (1972) who noted that fine-grained magnetite (~70 mVg) is unstable and spontaneously oxidizes at room temperature to maghemite when exposed t o oxygen.

During the oxidation of the original magnetic precipitate to the more maghemitic phase, the integrated intensity of the 220 XRD peak did not change, although the 111 reflection weakened. Likewise, the WHH of the 220 peak did not change on oxidation. If the precipitate was a mixture of several phases interme- diate between magnetite and maghemite, oxidation should have caused a reduction in the WHH as the precipitate moved towards a single phase. The precipitates were heated under a lamp at about 100~ for 1 hr prior to their quantitative XRD analysis to achieve maximum and uniform oxidation at low temperatures.

I 0 0

%

Total bose so consumption

60

40

20

i-cl:]: 203 / 3.a ..-'.Ii~.2 "/ .... . , , .

............ .:~./i

..-.I/y

./,7 / I I I I

100 200 300 400

Time (minutes)

Figure 2. Base consumption rate to maintain pH at 7 during the oxidation of 0.048 M Fe(II) solutions at 25~ with varying [C1]/[Fe] ratios.

With increasing [C1]/[Fe] ratios, the reaction times increased (Figure 2), and lepidocrocite formed in increasing amounts. These data support an apparent in- verse relationship between the decrease in area of the 220 maghemite line and the increase in area of the 020 lepidocrocite peak (Figure 3). The actual relationship between the integrated area of these peaks and the concentration ratio was not determined because it would be applicable only under the experimental conditions used. Moreover, a large error could be expected from so few samples. Although the results of only four experiments are shown here, about l0 series of experiments with [Cl]/[Fe] ratios of 2.2 and 9.1 in each series were carried out using slightly different parameters. In each series increasing the [C1]/[Fe] ratios reduced the value of the magnetite/lepidocrocite ratio in the product. The maximum amount of lepidocrocite formed (sample GL82-211, Table l) constituted about 28% of sample. The peak areas of each phase can be regarded as being proportional to the percentage of the phase present because the mass absorption coefficient of the

Table 1. Variation in peak areas and corrected widths at half height of lepidocrocite and magnetite X-ray diffraction peaks with increasing [C1].

Maghemite (220) Lepidocrocite (020)

[Cl]/[Fe] Area WHH Area WHH Experiment (initial) (em 2) (*20) (cm 2) (~

GL82-210 2.03 11.2 0.18 0.2 -- GL82-215 3.81 10.45 0.19 3.0 0.195 GL82-212 5.59 8.7 0.155 7.6 0.155 GL82-211 9.16 8.25 0.17 8.6 0.145 GL82-2131 2.03 -- -- 31.4 0.28 GL82-2141 9.16 -- -- 30.3 0.18

- - = not measured. Experiments carried out at 0~


Figure 4. Electron micrographs of magnetite and magnetite + lepidocrocite formed from the oxidation of 0.048 M Fe(II) solutions with varying [CI]/[Fe] ratios. (a) GL82-210 magnetite (C1/Fe = 2.19); (b) GL82-212 magnetite + lepidocrocite (C1/ Fe = 3.81); (c) GL82-21 l magnetite + lepidocrocite (CI/Fe = 9.16); (d) GL82-213 lepidocrocite (0~ (Ct/Fe = 2.19); (e) GL82- 214 lepidocrocite (0~ (CI/Fe = 9.16).

two phases are approximately the same (47.7 for maghemite and 43.7 for lepidocrocite for CoKa radiation).

Electron micrographs (Figure 4) show that the magnetic phase was reasonably uniform in particle size. The WHHs of the 220 peaks of magnetite/maghemite, although varying slightly (Table 1), did not change sys- tematically with increasing [C1]/[Fe]. The 020 peak of lepidocrocite, however, became sharper with increasing [CI] (Table 1). To check this observation, lepidocrocite was synthesized at 0~ a temperature that fa- vors lepidocrocite over magnetite, at [C1]/[Fe] = 2 and 9. The higher ratio again caused a reduction in the WHH of the 020 reflection. It cannot be ascertained whether the increase in the time for hydrolysis caused by the increase in the [C1]/[Fe] ratio was solely re- sponsible for the associated increase in crystallinity (decrease in WHH). The two experiments carried out at 0~ took 1270 rain (no added chloride) and 2020 rain (1 g NaC1 added, [C1]/[Fe] = 9.1) respectively; these times are much longer than those associated with synthesis at 25~ (290 rain, no added C1; 430 min for 1 g added NaCI, [C1]/[Fe] = 9.1), whereas the WHH of the 020 reflection was broader for the lower temperature samples (Table 1).

Variations in the Eh-time plots with changes in the [C1]/[Fe] ratio (Figure 5) were also noted. Increasing values of the ratio depressed and delayed the m i n i m u m value of the Eh. This ability of foreign ions to alter Eh during the oxidation of an Fe(II) system has been noted in this laboratory for AP + and Ti 4+ where increasing cation concentration also increased the Eh of the system and altered the final oxidation products. With increasing [CI]/[Fe], the plateau duration and the proportion of the total hydrolysis represented by this plateau increased (Figure 2).

The initial fast rate of hydrolysis was associated with a decrease in Eh (Figure 6). In systems where excess CI was added, the Eh increased to a maximum and then decreased when the hydrolysis rate started to slow down approaching the plateau region. In earlier work using higher rates of air flow through the solution (Tay- lor and Schwertmann, 1974), only a reduction of rate was observed and no plateau (rate = 0) region.

The plateau indicates that there is no base requirement to maintain the pH. In fact, the pH showed a small rise from the controlled value of 7 to about 7.1 in this region. When further hydrolysis again required added base to maintain pH, an associated rise in Eh

Vol. 32, No. 3, 1984 Effect of [CI]/[Fe] on magnetite formation 171

Eh (mY)

300

200

IO0

-IO0

-200

CoO

[Fe]

lil | : ,

f ! l % : " "/~'" z.,~ ,~J

J \ X b, '~i ~- \.'/x".x. _ J : . ..%.~" ~ "\[" f~l

' 'M./"~"Z..,'~-."j .~,./"

i t . . . . . . .

2 . 0 3 ~" "

l i . I: I

I I I I I I I I 200 400 600 800

Time (minutes)

Figure 5. Variation of Eh with time during the oxidation of an 0.048 M Fe(II) chloride solution at pH 7 for different [C1]/ [Fe] ratios.

was noted which increased very rapidly when about 95% of the total base requirement had been reached. The slope of this fast rise appeared to be independent of the [C1]/[Fe] ratio (Figures 5 and 6). The precipitate was dark green to greenish-blue at the start of the plateau region but generally changed to yellow-brown beyond this region.

DISCUSSION

With fast oxidation of Fe(II) chloride solutions, the green rust phase, being very susceptible to oxidation, may transform to lepidocrocite as soon as it forms. Inasmuch as the formation and structural breakdown of the green rust phase during oxidation are important steps in the formation of lepidocrocite and magnetite, a further understanding of the processes involved is desirable. Some insight is gained from the results of Detournay et al. (1976) who oxidized FeC12 solutions, hydrolyzed to different degrees, with air and measured the subsequent variations in pH and Eh~ Their results showed that in the pH range 6-7 a plateau region ex- isted through which the pH was relatively constant and that at the lowest rate of aeration (1.5 ml /min air vs. 2 ml /min in the present experiments) the pH rose slightly during the plateau region. They noted that during the initial drop in pH and rise in Eh, crystalline

300 ~ ,~/ / r I , Blue-green Olive green to i i

to dGrk gr~'en yellow br

200 Eh ~ I

(mV) j I00 I

1 !

40 0 nsurnption I Eh

I

-~00 % . " / x ~k/L / / ' " ". /

-2oo ,oo Boo 3~176 4oo 5'oo Boo~ Time (minutes)

Figure 6. Relation between Eh and hydrolysis rate during the oxidation of an 0.048 M FeC12 solution at pH 7 ([C1]/ [Fe] = 9.16).

Bo %

Total base so consumption

green rust I formed, and that lepidocrocite first appeared at the beginning of the plateau region.

Detournay et ak (1976) suggested that the initial reactions involved the formation of green rust I where- in the Fe in the Fe(II)Fe(III) hydroxy-chloride compound was predominantly in the divalent state. During the second stage, extending to the start of the plateau region, this precipitate oxidized to about 55% of the total Fe, the maximum degree of oxidation at which the structure is still stable (Feitknecht and Keller, 1950). Oxidation of the green rust phase must start as soon as it precipitates for lepidocrocite to be identified at the beginning of the plateau region. Detournay et al. (1976) stated that the plateau is a region of lepidocrocite formation in which [C1] and [Fe(II)] increase as FeC12 is liberated from the structure. The slight increase in pH in the plateau region of these present experiments indicates that OH is also liberated and is possibly replaced by O to satisfy the increased valence demands as Fe(II) oxidizes to Fe(III). In the present experiments at constant pH, hydrolysis required a further addition of alkali beyond the plateau region. Either the hydrolysis had been continuous with the equivalent amount of OH necessary to maintain the pH being liberated in sufficient amounts from the transforming green rust, or hydrolysis was inhibited in the plateau region. The latter explanation appears likely because oxidation would be required for continuous hydrolysis at pH 7 and the oxidation of the precipitate would probably use most of the available dissolved 02 at this low air-flow rate. At higher air-flow rates, or different solution concentrations, the plateau region was not al- ways observed, presumably due to the available 02 being sufficient to support both green rust formation and oxidation.

Inasmuch as the pH was controlled in the present experiments, variations in the Eh only reflected relative changes in the ratio of the concentration of the Fe(II) and Fe(III) species in solution. Before the plateau was


reached, a continuous steady addi t ion of alkali was necessary to maintain pH in this region. Especially for samples with higher [C1]/[Fe] ratios, the Eh first decreased then rose to a max imum before it decreased again to its min imum value. The small amount of Fe(III) present in the original solution or that formed during the initial oxidation would produce this initial decrease in Eh when it reacted with Fe(II) to form the insoluble green rust phase. Continued oxidation altered the original green rust precipitate towards the unstable oxidized form. The increases in Eh during this stage (Fig- ures 5 and 6) indicate either a continued decrease in [Fe(II)] or an increase in the solubility of the Fe(III) of the green rust species present. This rise in Eh, however, probably resulted from the particular condit ions used in the experiments. In similar experiments carried out under slightly different conditions, where added NaC1 still changed the reaction rates and increased lepidocrocite formation, this rise in Eh was not observed or was less marked. The most characteristic feature of the Eh was its decrease towards a m in imum value at the onset of the plateau and the at tainment of this minimum near the end of this region (see Figure 6). De- tournay et al. (1976) explained this decrease in Eh to be due to the start of the l iberation of Fe(II) as chloride from the degrading green rust structure. At the ter- minat ion of the plateau region the Eh again rose, slowly at first and then rapidly, as the [Fe(II)] decreased due to the continued oxidat ion and hydrolysis.

The mechanism by which the increased [C1] altered the transformation path is not clear. According to Ber- nal et al. (1959) green rust I alters topotactically to maghemite by dehydration, or to lepidocrocite by oxidation. The transformation products in the oxidation of green rust I depend on the environmental condit ions and pretreatment. A sample ofhydroxy-chlor ide green rust was made by the method of Taylor and McKenzie (1980). When washed and quickly dr ied from diethyl ether, it t ransformed in air after 60 hr to poorly crystalline akaganeite and lepidocrocite, with akaganeite dominant . The dried material when immersed in water for 60 hr gave the same two products with increased amounts oflepidocrocite. When the green rust was not dried after washing and resuspended in distilled water and allowed to oxidize, magnetite with a trace of lepidocrocite formed. When, however, the washed and moist compound was left to oxidize in an NaC1 solution (~0.2 M), a much higher proport ion of lepidocrocite formed with the magnetite. When air was passed through a stirring suspension and the pH was maintained at 7, the lepidocrocite became the dominant phase formed in the NaC1 solution. These changes in the oxidation products with environmental condit ions are similar to those described by Taylor and McKenzie (1980) for similar green rust compounds.

The various oxidation products formed under different condit ions and the increase in lepidocrocite crys-

tallinity with increasing [C1], as measured by the decrease in the W H H of the 020 reflection (Table 1), suggest that the formation of lepidocrocite may have been via Solution rather than the solid state transfor- mat ion implied by the term "topotact ic ." Even in solid state transformations, however, the composi t ion of the external environment can exert an influence by con- trolling diffusion of ions and possibly inhibiting certain reactions. Thus, in the washed and dried green rust I exposed to air, C1- may have been inhibited from freely diffusing out during the oxidation stage. Its retention, therefore, led to the formation o f akaganeite in which C1- is a necessary anion, being held in a " tunnel" structure. Chloride was likely able to diffuse out more freely when the green rust phase was not dried and was sub- jected to further oxidation under water; thus, akaganeite did not form. Furthermore, in an aqueous envi ronment with a high [C1], some of the surface OH may have been replaced by C1. This replacement would to some extent have l imited the el imination of water by the condensation of two adjacent OH and the subsequent formation of F e - O - F e bonds necessary for magnetite. To demonstrate this effect of environment and to show that the Cl- rather than the Na + influenced the transformation, a sample of green rust I was pre- pared in the absence of excess chloride. After washing, half of the precipitate was immediate ly immersed in distilled water and the other half in 1 N CaC12 solution. After 60 hr of air contact, and with periodic adjustment of the pH to 7, the sample in water t ransformed to magnetite, while that in the chloride solution formed some lepidocrocite as well.

The presence of a high [C1] during the initial format ion of the green rust may also have caused a higher degree of substitution of C1 for OH throughout the structure, thereby influencing the final oxidation products. This influence was shown for the oxidation products of the green rust formed in a system with a [C1]/ [Fe] ratio of 6.25. The washed precipitate aged in water to give magnetite and some lepidocrocite, whereas that aged in 1 M NaC1 gave increased amounts of lepidocrocite.

The hydroxy-carbonate green phase, expected to be common in soils (Taylor, 1980) generally oxidizes to goethite but can also transform to a magnetic phase (magnetite/maghemite), both minerals commonly oc- curring together in soils. A high chloride environment may instead cause some lepidocrocite to form, assum- ing that there is no overriding influence from other foreign ions. This suggestion is supported by a recent observation (Chittleborough et al., 1983) that lepidocrocite is present at higher concentrations in the pro- files of low-lying coastal soils where the [C1] is higher. The effect of chloride is to some extent also evidenced by the common occurrence oflepidocrocite in the phases produced in the rusting of Fe under marine conditions. Although [C1] is generally low in lepidocrocite

Vol. 32, No. 3, 1984 Effect of [C1]/[Fe] on magnetite formation 173

soils (U. Schwer tmann , Lehrs tuhl fiir Bodenkunde , Techn i sche Univers i t t i t Mt inchen , Freising, Wes t Ger- many , pr iva te c o m m u n i c a t i o n ) the [Fe(II)] would also be low because it would generally be prec ip i ta ted as an insoluble ca rbona te f rom CO2-saturated water. Taylor and S c h w e r t m a n n (1978) p roduced lepidocroci te f rom a carbona te sys tem but wi th different ox ida t ion rates to those descr ibed in the p resen t paper . Wi th such low [Fe(II)] (~0 .02 M), low [C1] m a y exer t an inf luence on the final product .

The influence o f [C1] on the fo rma t ion o f akaganeite m a y be l imi ted to the aerial ox ida t ion o f dry green rust I f o r m e d in a high [C1] e n v i r o n m e n t (Taylor a n d McKenz ie , 1980), which may account for the rari ty o f akaganei te in nature.

A C K N O W L E D G M E N T S

Dr. M. R a u p a c h and Mr. T. Cock o f the C S I R O Div i s ion o f Soils, Adela ide , are t hanked for valuable d iscuss ion and he lp wi th the e lec t ron microscopy, respectively.

R E F E R E N C E S

Bernal, J. D., Dasgupta, D. R., and Mackay, A. L. (1959) The oxides and hydroxides of iron and their structural in- terrelationships: Clay Min. Bull. 4, 15-30.

Chittleborough, D. J., Walker, P. H., and Oades, J .M. (1983) Textural differentiation in chronosequences from eastern Australia. Part 1. Description, chemical properties and mi- cromorphology of soils: Geoderma. 32 (in press).

Detournay, J., Derie, R., and Ghodsi, M. (1976) Etude de l'oxydation par aeration de Fe(OH)2 en milieu chlorure: Z. Anorg. Allg. Chemie 427, 265-273.

Farrell, D. M. (1972) Infrared absorption in the oxidation

of magnetite to maghemite and hematite: Can. Mines Br. lnvest. Rep. IR 72-118, 44 pp.

Feitknecht, W. and Keller, G. (1950) Uber Hydroxyde und basische Salze des 2 wertigen Eisens und deren dunkelgriine Oxydationsprodukte: Z. Anorg. Chemic 262, 61-68.

Hamada, S. and Kuma, K. (1976) Preparation ofa,-FeOOH by aerial oxidation of iron (II) chloride solution: Bull. Chem. Soc. Japan 49, 3695-3696.

Kiyama, M. and Takada, T. (1972) Iron compounds formed by aerial oxidation of ferrous salt solutions: Bull. Chem. Soc. Japan 45, 1923-1924.

Krause, A. and Borkowska, A. (1963) Der Einfluss yon Fremdenanionen und die Struktur der Oxydationspro- dukte: Z. Anorg. Allg. Chemie 326, 216-224.

Misawa, T., Hashimoto, K., and Shimodaira, S. (1974) The mechanism of formation of iron oxide and oxyhydroxides in aqueous solutions at room temperature: Corros. Sci. 14, 131-149.

Schwertmann, U. (1959) Mineralogische und chemische Untersuchungen an Eisenoxyden in Bidden und Sedimen- ten: Neues Jb. Miner. 93, 67-86.

Stumm, W. and Lee, G. F. (1961) Oxygenation of ferrous iron: Ind. Eng. Chem. 15, 143-146.

Taylor, R. M. (1980) Format ion and propert ies of Fe(II)Fe(III) hydroxy-carbonate and its possible significance in soil formation: Clay Miner. 15, 369-382.

Taylor, R. M. and McKenzie, R. M. (1980) The influence of aluminum on iron oxides. VI. The formation of Fe(II)- AI(III) hydroxy-chlorides, -sulfates, and -carbonates as new members of the pyroaurite group and their significance in soils: Clays & Clay Minerals 28, 178-t87.

Taylor, R. M. and Schwertmann, U. (1974) Maghemite in soils and its origin. II. Maghemite synthesis at ambient temperature and pH 7: Clay Miner. 10, 299-310.

Taylor, R. M. and Schwertmann, U. (1978) The influence of A1 on iron oxides. I. The influence of A1 on Fe oxide formation from the Fe(II) system: Clays & Clay Minerals 26, 373-383.

(Received 16 April 1983; accepted 7 August 1983)

PeamMe--1Ipe~cTa]3~eHo o6paaoBaHne MeJIKo3epHHCTOFO MarneTHTa (~ 0, i #m) npH pH = 7 n 2 5~ HyTeM aapatt~tn pacTaopoB Fe(II) xnoprt/Ia. MarHeTnT npeBpamancn npa 105~ a MareMHT C He~opaaa~ITraiMH .rlkIHnaMIt cynepCTpyKTypbL B ycnoan~x aToro aKcnepnMenTa, c yaennqeHneM uaqaJibnoro OTHOI/JeHH~I [CI]/ [Fe] OT cTexrIoMeTpaqecKoro 3naqeHria 2, KoneqnbI~ npo/lyKx co~ep~an yae.~rI~uaa~omneca KostaqecTaa zennnorpoKnTa. CTeuens xpacTa~an~nocTn aTO~ d~aabi, nsiepeHuan no mnpnne na nozoanne aSlCOTSl Jinnnn 020 penTrenoacro~ uopomroao~ ~Hqbpar~nn, Tar~Ke yaenHq~aa~ac~ c yaeJmaeHneM aToro OTnO- mennn. CKOpOCT~ ra)IpoJmaa (pacxo/x melioqH )IJ/~ no~epmanrm pH) noxaaa.na nJmTO, noJxozcenae a paaMep KOTOpOro ~aMennzncb c ae~IrIqnno~ OTnomennn [C1]/[Fe]. Ha npoTn~KenHn aToro pa~ona (nJmTO), Beanqnria Eh yMeribma~acs ~o MrmnMa.a9noro 3naqenan a onpeae~eHnoM MecTe, noJ~o~Keaae KOTOpOro nenocpea- cTaenno 3aBncnao OT ae.rIl~ql~rI/~I OTnomeHnn [C1]/[Fe]. O6paaoBaaae c~opee nen.aoKpo~Ta, qeM MarneTnTa ~lB~l.rIOCb, Bepo~ITHO, cJIeRCTBHeM BblCOKOFO 3naqenan [C1]. B aTOM cJ~ynae !,IMelOTC~I yBeJm~fennble TpyR- HOCTn ~rm Ko~eHcat~nn coce~Hx OH n acKJno~enn~ H20, qTO cnoco6cTByeT o6paaoaanmo ca~arI Fe -O- Fe acJ~eZCTBne aaMemenn~ OH ,onaM. C1 Jm6O a CTaZ~ o6paaoBannofl a Haqaae 3eJ/eHofI pz(aaqnrlbl, ~n6o a Teqemie ee oK,cJ~ennz. [E.G.]

Resiimee--Es wird die Bildung yon feinkSrnigem Magnetit (~0,1 ~m) bei pH 7 und 25~ durch Durch- liiftung yon Fe(II)-ChloridlSsungen beschrieben. Der Magnetit wandelte sich bei 105~ in Maghemit mit schlecht entwickelten 0berstrukturlinien urn. Unter den gegebenen experimentellen Bedingungen enthielt das Endprodukt--da das ursprtingliche [C1]/[Fe]-Verhttltnis h0her war als der stSchiometrische Wert von 2--zunehmende Gehalte an Lepidokrokit. Der Kristallinittitsgrad dieser Phase nahm ebenfalls mit zu- nehmendem Verhtiltnis zu, wie aus der Breite bei halber H6he des 020 RSntgendiffraktionspeaks her- vorgeht. Die Hydrolysegeschwindigkeit zeigte ein Plateau, dessen Lage und Ausmal3 sich mit dem ur- spriinglichen [C1]/[Fe]-Verhtiltnis vertinderte. Innerhalb dieses Plateaus nahm der Eh-Wert auf einen Mindestwert ab, dessen Lage direkt mit dem [C1]/[Fe]-Wert zusammenhing. Die Bildung yon Lepidokrokit vor Magnetit ist wahrscheinlich vom hohen [C1]-Wert abhtingig, wodurch es fiir benachbarte (OH)- Gruppen schwieriger ist zu kondensieren und H20 auszuschalten. Eine derartige Reaktion kann Fe -O- Fe-Bindungen hervorrufen, was aufder Substitution yon C1 ftir OH, entweder im zuerst gebildeten griinen Rost oder w~ihrend seiner Oxidation, beruht. [U.W.]


R6sum6--On prSsente la formation de magn6tite h fins grains (~ 0.1 /am) au pH 7 et ~ 25~ ~ partir de l'a6ration de solutions de chloride Fe(II). La magnetite s'est convertie ~ 105~ en magh6mite avec des lignes superstructurales pauvrement developp6es. Sous les conditions exp6rimentales employ6es, le produit final contenait de plus en plus grandes quantit6s de 16pidocrite ~ fur et ~t mesure que la proportion initiale [C1]/[Fe] a 6t6 augment6e ~ partir de la valeur stoichiom6trique 2. Le degr6 de cristallinit6 de cette phase, mesur6 ~ la largeur de la demi-hauteur du sommet de diffraction des rayons-X 020, a 6galement augment6 proportionnellement h cette augmentation. Le taux d'hydrolyse (consomrnation de base pour maintenir le pH) a montr8 un plateau dont la position et l '6tendue ont chang6 scion la proportion initiale [Cl]/[Fe]. A travers eette r6gion de plateau, l 'Eh a diminu6 gt une valeur min imum dont la position 6tait directement apparent6e ~ la valeur [C1]/[Fe]. La formation de 16pidocrite plut6t que de magn6tite est sans doute due au haut [C1] 6u il y avait d'avantage de diificult6 pour l 'OH avoisinant de se condenser et d'61iminer H20. Une telle r6action provoquerait la formation de liens Fe-O-Fe fi cause de la substitution de C1 pour OH soit dans l'6tape de rouiUe verte form6e d'abord, ou pendant son oxidation. [D.J.]

influence of chloride on the formation of iron … 32/32-3-167.pdf · influence of chloride on the...

Documents