Department of Civil Engineering, The University of Michigan, Ann Arbor, Mich. 48104
(First received 2 October 1970; in revised form 20 November 1970)
Abstract-The interaction of iron(Ill) and uranyl ions with dissolved silica is studied by a spectrophotometric
technique. The interaction with iron is studied as a function of the degree of polymerization
of the silica. For uranyl ion, a complex with monomeric silica is inferred. An equilibrium constant
PQ1 – 2.0 for the equilibrium UO22+ + Si(OH)4 = UO2SiO(OH)31+ + H ÷ is measured.
IN THE only previous quantitative study of complex formation between aqueous
silicic acid and a metal ion, Weber and Stumm[1] found that the interaction with
the iron(Ill) ion could be described by the equation
Fe 3+ + Si(OH)4 = FeSiO(OH)3 z+ + H +, Q1 (1)
An equilibrium constant Q1 –0.57 was obtained in 0.1M NaCIO4 media at 25°C
by use of u.v. spectrophotometry. Attempts to reproduce this result in our laboratory,
with the aim of extending the technique to other metal ions, have met
with considerable difficulty. We will show here that much of the difficulty can
be traced to the presence of multimeric silica species, assumed not to be present
in the original work. Some data for the interaction of polymeric silica with iron(Ill)
is given. Application of the spectrophotometric technique to the study of a
silicato-uranyl complex is then presented.
All chemicals used were reagent grade. Iron(III), uranyl, and sodium perchlorate were obtained
from G. F. Smith and used without further treatment. A stock solution of 0.864M HCIO4 was used
in all experiments. Measurements with iron used a common stock solution 8 x 10-4M in Fe(CIO4)3
and 4.3 x 10-ZM in HCIO4. The analysis for iron is based on the absorbance of a dilute acidified solution
at 272 m/x [2]. Measurements with uranium used a common stock solution 0.106M in UO2(CIO4h
and 8.6 x 10-3M in HC104. The uranium analysis is based on the absorbance at 417 mtL of an aliquot
rendered 5 per cent in HC104. A molar absorptivity e = 7″ 1 for uranyl ion in perchloric acid media
can be inferred from literature data[3]. Baker and Adamson sodium silicate nonahydrate was used to
prepare a 0.94M stock solution. This solution, somewhat turbid when first prepared, was filtered

through a 0.45 m/z Gelman filter and stored in a Nalgene polyethylene bottle. Analysis for silica was
performed by the colorimetric silicomolybdate method [4].
For measurements with iron, five different stock solutions of silicic acid in various degrees of
polymerization were prepared. The formal concentration of each solution was 1″9 x 10-2M in terms
of monomeric silica. Solution 1 was prepared by first mixing 20 ml of 0.94M Na~SiO3 with 100 ml
of 1M NaOH, then adding the mixture with rapid stirring to an excess of HCIO4 which produced
a final pH -2. Although the resulting solution is unstable with respect to polymerization[5], polymerization
is quite slow at this pH [6]. The mixture was then diluted to 1 1. Solution 2 was prepared
by direct neutralization of 0-94M Na~SiO3 with excess HCIO~, again to pH -2. Solutions 3-5 were
prepared by first doing a partial neutralization of Na2SiO3 with HC104 to give a solution 3’8 x 10-2M
in silica and with pH 9-10. Under these conditions silicic acid polymerized to an appreciable extent
over a period of several hours. A single solution was prepared in this way; aliquots were then drawn
off at various times to be neutralized to pH – 2 and diluted to a silica concentration of 1.9 × 10-2M.
For measurements with uranyl ion, varying amounts of 1M Na2SiOa were premixed with IM
NaOH, then neutralized with HCIO4 to pH – 3.5. The stock solutions were 4.7-7.0 x 10-2M in silica
and were all 0.4M in sodium ion. The acid and basic solutions were chilled before mixing to minimize
the polymerization produced during mixing. The heat of neutralization raised the temperature to
20°C after mixing.
For a series of absorbance measurements, fixed aliquots of a silica stock solution and o1~ the uranyi
or iron stock solution were added to 100 ml volumetric flasks. The pH was adjusted by adding varying
amounts of HC104, and the ionic strength was adjusted with a NaC104 stock solution. The perchlorate
ion concentrations were fixed at 0″ IM for the iron solutions and 0.2M for the uranium
solutions. After dilution of 100 ml, the solutions were thermostatted in a 25°C temperature bath for
30 min before measurement of the absorbance.
Absorbance measurements were made with a Beckman Model DU spectrophotometer using 5 or
10 cm cells. The hydrogen ion concentrations were assessed for a given series with a fixed silica stock
solution by measuring the pH of the least acid solution and calculating [H +] for the other solutions
from the amount of perchloric acid added, pH measurements were made at 25°C with an Instrumentation
Laboratories, Inc. model 205 pH meter and a combination glass-calomel electrode.
Some light scattering and differential refractometry experiments were carried out to estimate the
molecular weight of polymerized silica solutions, and to study the effect of iron(III) ions on the rate
of polymerization of silica under the conditions of our equilibrium experiments. A Brice-Phoenix
Model 8P-2008-8 light scattering photometer and a Brice-Phoenix Model BP-2000-V differential
refractometer were used for these measurements. A wavelength of 436 m/~ was used for both types
of measurement. The weight-average molecular weight was estimated simply from the measured
turbidity of the silica stock solutions; there was no correction for extrapolation to zero concentration.
The weight fraction of silica is based on its weight as SiO2.