^{1}

^{*}

^{1}

^{2}

**Hydrolytic equilibria of Tm (III) in KOH solutions were studied at 25°****C****. A spectrophotometry with m-cresol purple and 2-naphthol as pH indicators was used at an ionic strength of not more than 0.0005. The results indicate that in freshly prepared solutions at pH ranging between 6 and 10 Tm is present as ****, ****, **** and ****. The stepwise stability constants of hydroxide complexes calculated at zero ionic strength were obtained as coefficient of linear regression equations from the graph of optical densities of the indicators in Tm solutions at varying pH.**

Recently, the studies of rare earth elements (REE) complexing have been considerably intensified due to evaluating the fate of the REE compounds in the environment [1-3]. Producing and accumulating of significant quantities of REE during nuclear fission in uranium and Plutonium reactors are a potential source of their formation. The lanthanides complexes are among the most important compounds in natural waters regarding predominant anions [

According to the review of the published literature, thulium appears to be the least researched element amongst lanthanides due to an exclusively complex technology of its production and very high prices. The published experimental data on thulium hydrolytic behavior at 25˚C are very limited, and being obtained under different experimental conditions, these are often incomparable with each other [7-10]. For example, in [_{2}O_{3} solutions at the ionic strength of 0.3, is reported as 5.78 log.unit The corresponding stability constant of the isotope ^{170}Tm, assessed by chelating organic ligands extraction together with radio-chemical labelling, is 9.6 log.unit and differs by nearly 4 log.unit [

We studied hydrolythic equilibria in Tm^{3+} solutions with a possible participation of higher order hydroxoforms besides of. To minimize the experimental errors caused by extrapolation to the zero ionic strength we studied hydrolytic reactions at minimal ionic strengths.

Spectrophotometric pH measurement of the solutions containing variable TmCl_{3} concentrations and constant concentrations of the acid-alkaline indicators and KOH were carried out. An increase in Tm concentration resulted in a decrease of the absorption of the deprotonized form of the indicator due to an increase in the protons quantity as per following reaction:

The pH values were calculated from the measuring absorption densities of the indicator using tabulated values of the ionization constants [

Reactions of hydroxocomplexes formation in solutions of trivalent metal ions

are characterized by stepwise ^{о}K_{n} and total ^{о}β_{n} stability constants expressed at zero ionic strength by the equations:

Both indicators are weak organic acids (HA) that react with a strong inorganic base KOH:

where HA and A^{−} are the protonated and deprotonated forms of the indicator, respectively.

If the equilibrium constant K_{BHA} for reaction (5)

is known, the equilibrium concentrations [HA], [А^{−}] and consequently [OH^{−}] and pH of the solution under study can be evaluated from the spectra.

The reaction between potassium hydroxide and the indicator (being a weak acid) neutralizes its HA part into KA salt (5). Then the solution denoted as number 1, containing the acid and KOH without Tm, has buffering properties. Therefore, adding the increasing of protons concentration would be equivalent to titration of alkaliscent solutions by a strong acid. The protons would react with А^{−} increasing the concentration of the weak acid HA. If the buffering capacity of the Tm-containing solutions is insufficient, the remainder of protons would neutralize OH^{−}-ions in the solution (i). The equation for protons created in the hydrolytic reactions can be written as:

If hydrolysis is treated as splitting off a proton from a water molecule in the hydrate shell of the rare earth ion, the number of produced protons would be equal to the number of hydroxide ions bound into complexes:

and therefore the ligand number can be calculated from Equation (9), where C_{Tm} is the analytical concentration of Tm. The index i is omitted for simplicity in the Equations (8) and (9).

The calculation algorithm was based on the stepwise approach and consisted of the following stages:

1) The activity coefficients were initially assumed to be equal to 1.

2) HA, A^{−} and consequently OH^{−} equilibrium concentrations were calculated using optical densities.

3) n values were calculated using Equation (8).^{}

4) values were expressed from (9) and calculated as the linear lest square method parameters.

5) Concentration of hydroxocomplexes and activity coefficients were calculated using the values obtained. Activity coefficients were evaluated by the Debye-Hü ckel equation in the second approximation.

6) The program was returned to Step 2, until all the calculated values became constant according to the preset accuracy.

The complex forms, and , interlinked by the constants and, were assumed to be present in solutions containing mcresol purple. They were calculated by the regressive equations:

where and are the activity coefficients for uniand trivalent ions.

complexes were suggested in 2-naphthol-containing solutions within the measured pH

interval. The stability constants were evaluated as per following equation:

The absorption spectra of pH-indicators solutions containing “analytical grade” TmCl_{3} (sourced from the Novosibirsk plant of chemical reagents) have been measured. The solutions for spectroscopy were prepared from 0.01 M TmCl_{3} aqueous stock solution with pH 5.84. The Tm concentration was controlled spectroscopically with an arsenazo [

The spectra of solutions were measured in closed quartz cells with 5 cm optical length using UV VIS spectrophotometer Specord M40. The experimental error was evaluated using the law of errors propagation. Its value depended on the accuracy of solution preparation and photometric measurement and did not exceed 1.2% at 25˚C. The accuracy of experimental evaluation of the stability constants was calculated as standard deviations of the linear regression parameters.

The sulfonaphthalein indicator m-cresol purple (mCP) was used by Clayton and Byrne for surface and deep-water spectrophotometric pH measurement of sea water [_{2}I, HI^{−} and I^{2−}. In the visible range within the interval of the measured pH values mCP’s spectrum is represented by distinct intensive bands corresponding to its protonated (HI^{−},) and deprotonated (, ,) forms. According to our preliminary measurements, the extinction values of HI^{−} were ranging from 0 to 408 within the 15,400 to 17,300 cm^{−1} interval. That allowed to disregard HI^{−} absorption in our measuring of absorption of the I^{2}^{−} band. This indicator was also used by the authors of [

The chemical equilibrium between two forms of mCP: I^{2}^{−} + H^{+}^{ }HI^{−} is described by stepwise formation constant:

The value of pK_{a} ionization at 25˚C and an ionic strength 0.7 is equal to 8.146 according to [_{a} for the zero ionic strength.

_{3} solutions. They are characterized by distinct isobestic point at 20,500 cm^{-1}. The initial analytical concentrations of KOH and mCP in this experiment were constant and equal to 2.6 × 10^{−4} and 3.9 × 10^{−5}, respectively. The concentration of TmCl_{3} in series of 10 solutions varied from_{ }0.0 to 5.0 × 10^{−5} М. The pH values were estimated as negative logarithm of hydrogen ions activity and were calculated by comparing of I^{2−} absorption in solutions under study (i) with the absorption in solution (0) with KOH concentration of 0.01 M, were all the indicator is in the I^{2−} form.

The published data confirms that mCP forms complexes with Fe (III) [^{2−} had to be decreased with increasing of metal concentration without any isobestic point. Should the Tm hydroxocomplexes be also formed, the isobestic point would be displaced vertically down. The pattern of spectra in

^{−}^{1} interval, and their standard deviations.

Unfortunately, we could not find in literature the experimental data on the values of _{ }for Tm. However these are available for erbium [_{4} medium (under conditions excluding the presence of CO_{2}) are as follows: = 7.7, = 13.5, = 18.9, log^{o}β_{4} = 19.2. Taking into account the difference of experimental conditions one can consider that the value = 15.09 for thulium which we have obtained at the zero ionic strength satisfactorily agrees with the results of [

The ionization constant of 2-naphthol K_{a} is determined by the equation:

where NapOH and NapO‾ are the protonated and deprotonated forms of 2-naphthol, respectively. The values of рК_{а} were defined spectrophotometrically up to 400˚C in [_{3} which depend upon pH. The absorption band in the interval 26,000 - 30,000 cm^{−}^{1}^{ }belongs to NapO^{−}. ^{o}K_{3} and their standard deviations. The measurements were carried out in the range from 27,200 to 28,800 cmˉ^{1} in nine points of the spectra. The initial analytical concentration KOH and 2-naphthol were constant and equal to 3.0 × 10ˉ^{4} and 1.0 × 10ˉ^{4} М respectively. The TmCl_{3} concentration varied from 0.0 M to 4.5 × 10ˉ^{5} М in the series of 10 solutions. The β_{3} values_{ }are published by Fatin-Rouge and Bünzliin [_{ }at a constant_{ }ionic strength supported by the adding 0.1 М NaCl. However, only complex forms Tm^{3+} and were suggested within pH range of 8.5 - 11. The measured value of ^{*}β_{3}-21.14 could be considered as agreeing with our results (19.35) if the difference in the experimental conditions is accounted for.

The number of bound hydroxide ions as well as thermodynamic values of the stability constants of the first three

^{о}K_{n} and their standard deviations (sd) in solutions of m-cresol purple measured at ν = 16,400 cm^{−}^{1}.

^{о}K_{n} and their standard deviations (sd) in solutions of 2-naphthol measured at ν = 27,600 cm^{−}^{1}.

Tm (III) hydroxo complexes have been evaluated by the indicator spectrophotometric method in the absence of polymer forms, side reactions and hydroxides precipitate at 25˚C at minimal ionic strength. Although the research was initially intended to form a basis for estimating stability of hydroxide, carbonic and mixed hydroxocarbonate forms of REE at elevated temperatures, the employed methodology at near-room temperatures has proved to be advantageous over the direct potentiometeric measurements Our methodology utilizes tabulated values of ionization constants and extinction coefficients of pH indicators that removes the need for electrode calibration during each measurement. The employed spectroscopic technique allows carry out direct observations of the hydrolythic behavior of elements in solutions.

The work was supported by the Russian Branch for Basic Research, project No. 11-05-00662-а.