Cawdlan Min*alogist VoI. ld w.4046 (1976) SOME FAGTORSAFFECTING THE SYNTHESISOF GRYPTOCRYSTALLINE STRENGITEFROM AN AMORPHOUSPHOSPHATE COMPLEX N. D. WARRT canada centre for Inhnd waters, P.o. Box 5050,Burlington, ontario, canada. JAMFS R. KRAMER Deportment of Geology, McMaster University, Eamilton, Ontafio, Canada. ABsrR.Acr ments undertaken by Einsele (1938) showed that in the pH range betweetr Four experiments were initiated to examine the 5 and 8, signifi- effects of pH, Eh (dissolved oxygen), agitation and cant removal of inorganic phosphate from solu- coocentration of key reactants (Fe and p) on the tion by precipitation with amorphous iron oxide reaction of P with amorphous iron oxyhydroxide did occur. Subsequent studies (Williams er a/. under conditions encou.ntered in natural systems. l97la,b,c) have suggestedthat this reaction may pH and dissolved oxygen content are tle most im- be of considerable importance in natural wa- portant variables controlling initial p uptake in the ter system$. Mortimer (1971) found that the aqueous phase. The pH range (4-Z) is optimal for uptake and release of P in natural waters could uptake wher Fe greater is than 0.25 mg,zl and dis- be related to the oxidation state of the sedi- solved oxygen concentration is )l|Vo saturatiou. ments, Oxygen depletion resulted in increased Digestion of the resultant iron oxyhydroxide-phos- P concentrations phate under conditions of neutral pH, oxidizing Eh in the overlying water. and T: 100"C, produced a mixture of crypto- The thermodynamic properties of the iron crystalline strengite and phosphosiderite. This'end- phosphate system have been examined in detail product verifies the thermodynamic predictions of in a seriesof papers by Nriagu (I972a,b, 1974). Nriagu (1972b). He has summarized the available thermochenr- ical data from tle basic iron phosphateminerals REsurvr6 and has developed theoretical models of the solubilities and stabilities of these minerals in Quaue exp6rienoesont 6t6 entreprisesafin aqueous systems. d'examiner les effets du pH, (oxygdne du Eh dis- No complete sous), de I'agitation et de la experimental description of the concentration des phosphorus r6actifs cl6s (Fe et P) sur la r6action du p avec de reaction of with amorphous iron I'oxyhydroxyde ferreux amorphe dans des conditions oxides exists, although Patrick & Khalid (1974) rencontr6es dans des systdmes naturels. k pH et have provided good experimental information le contenu en oxygdne dissous sont les varianies les about five different soil types. plus importantes controllant la fixation du p ini- The effects of Eh and pH on the sorption of tial de la phase aqueuse. La port6e du pH (4-7) est P onto ferric oxyhydroxide motivated the work au maximum pur la fixation lorsque Fe est plus described here. These experiments de- grand were que 0.25 mg/l et que la concentratlon d'oxy- signed g€ne dissous est plus grande que l|Vo de satva- (i) to describe experimentally the effects of Ia digestion de I'oxyhydroxyde de phosphate !ion. Eh and pH in a simple aqueousFe-P ferreux dans des conditions de pH neutre, oxirdant sys- - Eh et ? 100'q a produit un m6lange de itren- tem, and gito c4ptocristalline et de phosphosiderite. Ce pro- (ii) to define the conditions of formation duit final conflrme les pr€dictions thermodynami- of a possible crystalline end-product of ques de Nriagu (1972b). this interaction. (Iraduit par le journal) ExPBRIMENTAL INrnonucrroN In all experiments the simplest possible chem- The reaction of phosphate ions with amor- ical systems were used so that extraneous chem- phous iron oxide has been suggested as a prob- ical species would not interfere with the pro- able inorganic control on tl.e availabilitv of cesses being investigated. The master variables phosphorus in water (Mortimer 1941). Eiperi- of Eh, pH, temperature and ionic concentra- tions of the elements involved were set at values approximating as possible rTo closely as those en- whom all correspondence should be sent. countered in the natural environment. 40 SOME FACTORS AFFECTING SYNTIIESIS OF STRENGITE 4T TABLEI. SUI'1ilARTOF3 EXPERIMEfiTSCARRIED OIII TO DEFINE POA-IRON OXIDE ASSOCIATION Exp€fl[Ent Purposeof Experirent KeyVariEbles Results NuIDer 0bserved I (l) To detemlne effect of pH on P04-3 uptake by mrphous ircn oxlde pH uptake dependenton pH. (ll) To detemlne the stablllty of the phosphateiron hydrcxlde cooplex pH Deflned pH range of stability of complex, at dlfferent DH's. ltttl To detemlne ilm of ructlon Tlm Reactlon coltplete t! ? ![i!- 2 A Effect of oxygenconc. on stabillty of complex Eh Conplexdlssolves f,hen dlssolved oxygen2!g 02lI. R rf orvoan h:;nteB svsten?hen dois comllr refom? Eh RefonB whenorJgen content 1-Ang0,ll. Cancomplex fom a stable lnsoluble mlneral? Eh, pH, . Crystalllne prcduct fom and follm the conc. P0a-J themdJmMiq pred{ctlons of l{rlagu' For exporlment 2A results rsad dlssolved orygon <2 mg. The phosphonrs and ton concentrations em- had its pH raised to 7 or 8 for about 10 sec- ployed in these experiments closely approximate onds, then quickly had it lowered to 1.5 with a those ensountered in anoxic hypolimnetic wa- pre<neasured amount of HCl. Finally, the pH ters. I-ake Erie hypolimnetic waters contain be- of each solution was raised quickly to a final tween 0.35-2.5mg/l of Fe and 90-275p,gPOn/l value of 4 and 5 respectively. The solutions @urns & Ross 1972). Total P varied between were transferred immediately to 250 ml centri- 216-240 pC PO4/l in these experiments,total Fe fuge tubes and spun at 268O rpm for two rrin- was maintained at 2.5 mg Fe/I. The key para- utes. Fifty-ml samples were decanted from the meters of Eh and pH werq investigated one at tubes and analyzed for phosphate. a time in order to keep the system simple and Experiment 2: TWo groups (A and B) of three deterministirc. solutions each wete prepared as in Experiment Except where specified, the pH of all solu- 1. tions was in the range of 6 to 8, and the Eh Each solution initially contained 2L6 pC was of the order of 370 mV. The mineralization POro/I, 2.5 mg Fe/l and had a pH between 6 experiments were carried out at a temperature and 8. of 90'C, in order to increase kinetic rates to The six solutions were then treated in a a value which would produce the reaction end- manner identical with those of Experiment No. products in a time short enough to be useful in 1, except that this time, after the formation of this study. All other experiments were run at the solid, they were not agitated, but were left room temperature, which varied between 2G undisturbed f.or 24 hours. At the end of the 23"C. 24 hours, the top 2O ml were decanted and Table 1 is a summary of the experiments, and atalyzed for phosphate.The Eh and pH of each the following are details of each: solution was then measured. Experiment 1: TVo groups of, seven solutions, The flasks were made airtight, and flushed each oontaining2.5 mg/l Fe and 216 p,EPO,o,/l with ultra-pure nitrogen for 24 hours. The were prepared as described in Appendix A. The stopperswere then removed, the Eh and pH of pH's of the first group of seven solutions were each solution was immediately recorded, and a raised from an initial value near 2.0, by the 40 ml aliquot was decanted and analyzed for slow addition of 1.0 N NaOH, to final values phosphate.This completed the work with group of 3.0, 4.0, 5.O, . 10.O, care being taken to A. ensure that the final recorded value of pH was The three solutions of group B were left not exceeded at any time during the NaOH stoppered for a further 48 hours, but air con- addition. tact was available through the presence of two For the second group of seven solutions, the 4mmdiameter holes in each stopper. After this pH was raised quickly to about 7 long enough 48 hour exposure to air, a further 40 ml sam- to form a solid; then the pH was quickly lowered ple was decanted from each solution and anal- to about 2. T\e pH of each beaker was again yzed for Eh, pH and phosphate. slowly increased 1o a final value of 3.A, 4.O, Experiment J: About 0.4 g of the amorphous 5.0, 10.0, as in the first group. ferric oxide were placed in a sealed glass bottle In both sets of experiments, once the final containing 25O ml solution of D.D. water hav- pH value had been attained, the solutions were ing a pH of between 6.0 and 7.3. Varying con- stirred for about 20 minutes. They were then centrations 3.9 x lOsM to 3.9 x 10rM) of covered and left to stand for 24 hours, at which KHPOa were added in liquid and solid form as time the top 25 ml were decanted and analyzed per Table 1. The mixture was then digested at for phosphate. 90oC for varying periods of time up to 20 Two additional solutions were prepared. Each days. 42 THE CANADIAN MINERALOGIST given in Table 2. Almost all P is removed from, solution in the pH range 5.5-7.5 when no solid is present initially. Outside this pH range, con- siderable P remains in solution. The results from the second set of solutions. .e= 75 of Experiment No. 1 (rilith initial iron oxide present) are also shown Figure In pH a) in 1.