Distribution of Picric Acid into Various Diluents

Abstract

Extraction constants, defined as 𝐾_ex=[HPic]_o/[H⁺][Pic⁻], for picric acid (HPic) into ten diluents were spectrophotometrically determined at 298 K, where the subscript “o” refers to an organic (o) phase which is composed of the diluent. Their values for the HPic extraction into benzene and chloroform were fairly consistent with those reported before. From the 𝐾_ex values, distribution constants (𝐾_D,HPic) of HPic into the o phases were estimated using the thermodynamic relation 𝐾_ex=𝐾_D,HPic𝐾_HPic, where 𝐾_HPic corresponds to a protonation constant of Pic⁻ in water. Thereby, contributions of functional groups, such as –Cl and –CH₃, involved in the diluents molecules to 𝐾_D,HPic were discussed. The same discussion was also applied for the distribution of the ion pair, composed of CdPic₂ and 18-crown-6 ether, into the diluents.

1. Introduction

Picric acid (HPic) has been widely used for the source of pairing anions in solvent extraction [1–5], application to selective sensors [6], sequential injection analysis with it [7], its separation science [8, 9], theoretical study for its partition [10], its distribution into ionic liquids [11], and so on. Also, many authors have studied extraction and separation of many metal ions by crown compounds from water (w) into various diluents [1–3, 12–17]. In particular, the extraction data of HPic are important for studying the extraction of easily hydrolyzed metal ions, such as Ag⁺ and Pb²⁺, by the compounds into the diluents [1, 2, 18]. However, few systematic studies of the HPic extraction have been reported [5].

In the present paper, we systematically determined at 298 K extraction constants (𝐾_ex/mol⁻¹ dm³) of HPic into ten diluents having lower polarities, such as 𝑚-Xylene (mX), toluene (TE), bromobenzene (BBz), chlorobenzene (CBz), and 1-chlrorobutane (CBu). These diluents were selected on the basis of the careful consideration by Takeda et al. [3]. From the 𝐾_ex values, distribution constants (𝐾_D,HA) of HPic between the w and organic (o) phases or the diluents were also calculated. Moreover, contributions of functional groups constituting the diluents molecules to 𝐾_D,HA were discussed. The same discussion was applied for the distribution of the ion pair of CdPic₂ with 18-crown-6 ether (18C6) into the diluents.

2. Materials and Methods

2.1. Materials

Picric acid (99.5%, HPic·𝑥H₂O) was purchased from Wako Pure Chemical Industries, Japan. Sodium hydroxide (97%, Kanto Chemical Co., Ltd., Japan) was used without further purification. Its purity was determined by acid-base titration with potassium hydrogen phthalate (guaranteed pure reagent: GR, 99.9 to 100.2%, Kanto). The chemicals HPic·𝑥H₂O were dissolved in water and then its concentration was determined by acid-base titration with an aqueous NaOH solution of 0.1 mol dm⁻³. Commercially available diluents (all GR grades, Wako and Kanto) were washed three times with water and kept under the conditions saturated with water. Other chemicals were of GR grade. A tap water was distilled once with the still of the stainless steel and then was purified by passing through the Autopure system (Yamato/Millipore, type WT101 UV). This water was employed for preparation of all aqueous solutions.

2.2. Extraction Procedures and Data Handling

Aqueous HPic solutions of (0.1₅–3.4) × 10⁻³mol dm⁻³ and diluents were mixed at equivalent volume (12 cm³) in glass tubes of about 30 cm³, and then these stoppered glass tubes were vigorously shaken by hand for 1 minute. The tubes were agitated for 2 h by a shaker with water-bath thermostated at 298 K and then, in order to separate the two phases, centrifuged for 7 minutes. The pH values of the w phases separated were measured at 298 K: pH readings were in the range of 2.57 to 4.10. A Horiba pH/ion meter (type F-23) equipped with a glass electrode (Horiba, type 9615-10D) was employed for the pH measurements. While the o phases or the diluents separated were transferred into other glass tubes and the aqueous NaOH solutions of 0.1 mol dm⁻³ were added to them, volume ratios (𝑉_o/𝑉) of the separated o phases against the added aqueous solutions with NaOH were 1/1, 1/2, 1/4, 1/5, 1/8, 1/10, 1/12, or 1/25. As similar to the above procedure, the amounts of HPic in the o phases were back-extracted into the aqueous NaOH solutions. Using a 1 cm quartz cell, concentrations of Pic⁻ in these aqueous solutions were determined spectrophotometrically at 355.0 nm and 298 K, where a spectrophotometer used was of a Hitachi U-2001 type. The operation of the back-extraction was repeated in some cases.

These Pic⁻ concentrations which were determined at 355.0 nm due to the absorption of Pic⁻ were assumed to equal the concentrations, [HPic]_o, of HPic in the o phases. Here, the subscript “o” denotes the o phase. From these values and total concentrations, [HPic]_t, of HPic, distribution ratios, 𝐷_Pic, for Pic⁻ were calculated, assuming that [HPic]_o≫[Pic⁻]_o+[(HPic)₂]_o [19] and then using the equation 𝐷_Pic=[HPic]_o/([HPic]_t−[HPic]_o) (see Section 3.1).

As a software for the analysis of the plots of log𝐷_Pic versus pH (see Figure 1), a KaleidaGraph (ver. 3.501, Hulinks Inc., Tokyo) was used.

Figure 1 

Plots of log𝐷Picversus pH for six diluents. Circle: DCE; triangle: DCM; full circle: CBu; diamond: BBz; inverted triangle: Bz; square: mX. A broken line is the regression one for the DCE system based on (2): log𝐷Pic=2.489--pH--log(1+3.86×10−pH) at 𝑅=0.998.

3. Results and Discussion

3.1. A Concise Theoretical Treatment of Extraction Equilibria

We assumed here the following equilibria: H⁺+A⁻⇌HA and HA⇌HA_o for the overall extraction equilibrium of which the constant was expressed as 𝐾_ex=[HA]_o/[H⁺][A⁻]. For this extraction system [9, 10, 19], the distribution ratio was defined as 𝐷_A=[HA]_o[A⁻]+[]=𝐾HA^exH⁺1+𝐾_HAH⁺. (1) Then, taking logarithms of both sides, the following equation was given: log𝐷_A=log𝐾_ex−pH−log1+𝐾_HAH⁺=log𝐾_ex−pH−log1+𝐾_HA10^−pH. (2) Assuming that the 𝐾_HA values are a constant in a given experimental pH range (see Section 2.2), then we can immediately obtain the log𝐾_ex value from the plot of log𝐷_A versus pH. Actually, the 𝐾_HA values which were averaged for all the values of ionic strength (𝐼) were employed for the analysis of the plots, where 𝐼=(1/2)([H⁺]+[A⁻])≈[H⁺]. Additionally, the log𝐾_D,HA value can be estimated from the thermodynamic relation: log𝐾_ex=log𝐾_D,HA+log𝐾_HA. Also, (2) can approximate to log𝐷_A≈log𝐾_ex − pH in the present experimental ranges of pH (see Section 2.2).

3.2. Determination of 𝐾_ex and 𝐾_D,HPic

Figure 1 shows examples for plots of log𝐷_A versus pH at A⁻ = Pic⁻. Thus, the plots yielded linear relationships between log𝐷_A and pH. The same is also true of the HPic extraction systems with the other four diluents. The extraction data thus obtained are summarized in Table 1. The log𝐾_ex values of the extraction into benzene, Bz, and chloroform, CF, were in accord with those [1, 2] reported before by Takeda et al. Although the measuring temperature was not described, the log𝐾_ex value (=2.28 [10]) estimated previously for the dichloromethane, DCM, system was also close to our value (=2.462). From the log𝐾_HPic values in Table 1, one can see easily that differences among the present 𝐼 values, the average 𝐼 ones, have no objection to comparisons among the 𝐾_ex values or component equilibrium constants, except for the cyclohexane, cHex, system [8].

Table 1 

Extraction constants for HPic into various diluents and their component equilibrium constants (𝐾_HA, 𝐾_D,HA) at 298 K.

3.3. Extraction Abilities of the Diluents for HPic

According to the relation log𝐾_ex=log𝐾_D,HA+log𝐾_HA, the extraction ability of the diluents employed for HPic is obviously controlled by the log𝐾_D,HA values (see Table 1). This is supported by the following facts. A plot of log𝐾_ex versus log𝐾_D,HPic for the ten diluents gave a straight line with the slope of 1.00₁ and the intercept of 0.58₄ at the correlation coefficient (𝑅) of 0.999. This intercept was also in good agreement with the log𝐾_HPic values, listed in Table 1, within the experimental errors of ±0.01.

3.4. For Interaction of the Diluents Molecules with HPic

The log𝐾_D,HA values were in the order cHex ≪ CBu < CF < oDCBz < BBz, CBz < Bz < DCM, TE ≤ DCE < mX (see Table 1 for these abbreviations). This order shows that an interaction with HPic becomes strengthened by going from cHex to mX. We divide here the order into (cHex ≪) oDCBz < BBz, CBz < Bz < TE < mX for aromatic diluents and CBu < CF < DCM ≤ DCE for aliphatic diluents and then will examine these orders in more detail.

The log𝐾_D,HA values were in the order Bz < TE < mX (Table 1). This fact indicates that substitution of –H by –CH₃ increases the interaction of the diluents molecules with HPic. Similar results were obtained for the substitution of –Cl or –Br by –CH₃: namely, CBz or BBz < TE and oDCBz < mX, although the position of the functional group in the latter case differs from each other. On the other hand, those of –Cl by –CH₂CH₂CH₃ and –CH₂CH₃ decreased the interaction with HPic. The corresponding examples were DCM (ClCH₂–Cl) ≫ CBu (ClCH₂–CH₂CH₂CH₃) and DCE (ClCH₂CH₂–Cl) ≫ CBu (ClCH₂CH₂–CH₂CH₃), respectively. Also, the substitution of –H by –Cl decreased their interaction. These examples were DCM > CF and Bz > CBz > oDCBz. Like the relation between Bz and CBz, the same was also true of that between Bz and BBz. The above results indicate that the functional groups constituting the diluents molecules strengthen the interaction with HPic in the order of –CH₂CH₂CH₃, –CH₂CH₃≪ –Cl, –Br < –H ≤ –CH₃.

The case of cHex ≪ Bz suggests that the substitution of –CH₂CH₂– by –CH=CH– strengthens the interaction with HPic, while, there was no difference in the interaction with HPic between –Cl and –Br and accordingly that in log𝐾_D,HA between CBz and BBz (Table 1).

3.5. Application for the Distribution of Ion Pairs with 18C6

The log𝐾_D,MLA2 values reported previously [20] at MLA₂ = Cd(18C6)Pic₂ were in the order Bz (−6.18) ≪ TE (−4.85) < CBu (−4.48) < mX (−4.13) ≪ CBz (−2.70) < CF (−2.62) < DCE (−2.53) ≤ DCM (−2.52). Comparing these values with the log𝐾_D,HPic ones for given diluents, we can easily see that Cd(18C6)Pic₂ more weakly (or more strongly) interacts with the diluents (or water) molecules than HPic does. The high polarity [20] of the Cd(18C6)Pic₂ structure is also supported by this comparison. Similarly, the order was divided into Bz ≪ TE < mX ≪ CBz for aromatic diluents and CBu ≪ CF < DCE ≤ DCM for aliphatic diluents.

The same analysis as that described in Section 3.4 gave the order of –H ≤ –CH₃ ≪ –Cl for the former, while it did that of –CH₂CH₂CH₃, –CH₂CH₃ ≪ –Cl ≤ –H for the latter. A comparison of these orders with that for the HPic distribution system indicates that the interaction of the aromatic diluents molecules with Cd(18C6)Pic₂ differs from that with HPic, while the interaction of the aliphatic diluents ones is essentially similar to it. Especially, it is suggested that the substitution of –H by –Cl in the aromatic diluents molecules or altering Bz into CBz strengthens the interaction with Cd(18C6)Pic₂, being contrary to the HPic system. The same is true of the substitution of –CH₃ by –Cl, altering TE into CBz.

4. Conclusions

The 𝐾_ex and 𝐾_D,HPic values for the HPic extraction into the ten low-polar diluents were systematically determined at 298 K. The magnitudes of the former constants were exclusively controlled by those of the latter ones from analyzing their thermodynamic relation with 𝐾_HPic. The contributions of the functional groups constituting the diluents molecules to 𝐾_D,HPic were clarified subsystematically. That is, it was demonstrated that the interaction of HPic with the diluents molecules is strengthened by introducing a methyl group in their molecules. Also, the contributions to 𝐾_{D,Cd(18C6)Pic2} were partly different from those to 𝐾_D,HPic. The authors cannot now explain essential meanings of such contributions.

Conflict of Interests

The authors do not have any possible conflict of interests with any trademarks mentioned in the paper. Of course, they do not receive any financial gains from the two companies at all.