Protolytic Properties of Dyes Embedded in Gelatin Films

O filme de gelatina endurecida fixo em uma base transparente de triacetilcelulose pode ser utilizado como um suporte conveniente para indicadores ácido-base. Azul de bromofenol, verde de bromocresol, eosina, etil-eosina, n-decil-eosina, n-decil-fluoresceína, vermelho neutro e verde de malaquita foram incorporados nos filmes de gelatina por extração em meio aquoso sob condições apropriadas. A absorção de luz dos filmes dopados com corante exibe uma resposta ao pH de soluções tampão aquosas nas quais foram imersos. Valores reprodutíveis das constantes aparentes de dissociação dos indicadores, pKa , foram calculados pelo tratamento convencional dos valores de pH do tampão aquoso e também para a água absorvida nos filmes de gelatina. Os experimentos com pH abaixo e acima do ponto isoelétrico da gelatina (em torno de pH 5) resultaram em valores de pKa app semelhantes àqueles em meios micelares de tensoativos catiônicos e aniônicos, respectivamente. Os filmes de gelatina modificados por corantes podem ser recomendados para uso em sensores óticos e dispositivos similares.


Introduction
This paper is devoted to the influence of hardened gelatin media on the spectral and acid-base properties of immobilized xanthene, sulfonephthalein, and some other dyes.
Such modifications, or medium effects, are most studied and well documented for acid-base indicators. 8,9In this case, the influence of the microenvironment results in the shift of the indicator equilibrium as compared to that in aqueous media.The main reasons are the local polarity and the local electrical charge.The latter often (but not always) manifests itself in the decrease or increase in the 'apparent' pK a value, pK a app , in the case of positive and negative local charges, respectively. 8,9In addition, the penetration of indicator dyes into micro-aggregates, both charged and uncharged, leads to the displacements of the absorption bands of the conjugate species of the acid-base pair.
The protolytic and spectral effects observed in the above systems were explained by preferable fixation of the neutral forms of indicators on the surface and in the body of the uncharged polymer, 5,6,19 by electrostatic interaction of ionic indicator species with the net charge of the gel, 6,7 by conformations of dye molecules, 3 aggregation, 16 and also in terms of hydrophobic interactions 20 and hydrogen bond formation. 6][23][24][25][26] Gelatin is actually a polydisperse mixture of low-molecular polypeptides; 22 the molecules are pronouncedly anisotropic and asymmetric.In fact, the polymer contains a voluminous net of charges and some amount of free ions.Gelatin is a typical ampholyte; the averaged isoelectric point, pH ip , of the so-called alkaline gelatin is within the pH range of 4.8 to 5.1. 27wing to the long-chained molecules, cohered by a limited number of cross-linkages, the elasticity of gelatin films resembles that of rubber.Based on such polymeric arrays, it becomes possible to create matrix systems with rather uniform distribution of sterically available trapped molecules.This provides favorable conditions for promoting various chemical processes with participation of immobilized reactants. 22he aim of this study was to consider the behavior of dyes embedded in hardened gelatin films, wetted by water.For more objective examination, a set of indicator dyes with various charge types and different hydrophobicity was selected.
9][30][31][32][33] The dyes of such type are also utilized in sensor devices 31,34,35 and for holographic record. 32e studied visible-spectroscopically the protolytic properties of four xanthene dyes: eosin, ethyl eosin, n-decyleosin, and n-decylfluorescein, two sulfonephthaleins: bromophenol blue and bromocresol green, a triphenylmethane dye malachite green, and an azine dye neutral red in the films, by measuring their transmittance after immersing them into aqueous buffer solutions of different pH.
Attempts were made to sense the alterations of the electrostatic properties of gelatin in the above films along with variation of bulk water acidity.Also, the possibility of utilization of such dyed films in creating non-expandable test-materials for pH monitoring of aquatic media was estimated.

Chemicals
Photographic films for offset printing manufactured by AGFA, with thickness of gelatin layer around 20 µm, fixed on the transparent triacetylcellulose support 36 were used in this work.The samples used were colorless and transparent.Previously, the silver halides have been completely removed from the films using the solution of the developer G 101c, containing hydroquinone CAS # 123-31-9, and rapid fixer G 333c from Agfa Graphics NV (Belgium).Also, gelatin of technical quality was used for preparation of gelatin solutions.Bromophenol blue, bromocresol green, malachite green, and neutral red were from Shostka Chemical Plant, Ukraine, eosin and ethyleosin were from the collection of the Department of Physical Chemistry of Kharkov V. N. Karazin National University, the samples of n-decylfluorescein and n-decyleosin were kindly put at our disposal by Dr. V. I. Alekseeva, Research Institute of Organic Intermediates and Dyes, Moscow, Russia.Sodium n-dodecyl sulfate (SDS, 99% purity) was used as commercially obtained.Photographic gelatin was used for preparing its aqueous solutions.
Stock solutions of indicators with concentrations 10 -3 to 10 -2 mol L -1 (except n-decyleosin, which initial concentration was one or two orders of magnitude lower) were prepared by dissolution of exactly weighed amounts of sulfonephthaleins, malachite green, and neutral red in water; in the case of xanthene dyes, ethanol with mass fraction 95.6% was used as solvent.All working solutions were prepared by dilution of stock solutions using distilled water.
Acetic acid, sodium acetate, Na 2 HPO 4 , NaH 2 PO 4 , glycine, HCl, and NaOH were used to establish the required pH values.All acids and salts were of reagent grade.Sodium hydroxide stock solution was prepared using carbonate-free saturated alkali solution, was kept protected from the atmospheric CO 2 and standardized by adipic acid. 37The values pH < 3.5 were produced by HCl solutions, pH around 5 by the acetate buffer, from 7 to 8 by phosphate buffers, and from 8.5 to 10.5 by (glycine + NaOH) mixtures.Alkaline media (pH > 11) was made using NaOH stock solution diluted by CO 2 -free water.

Apparatus
Absorption spectra of dye solutions and dye-containing gelatin films were measured using KFK-3 (Russia) apparatus against reference solutions or film, containing all components except the indicator.The pH values of working solutions were determined using the glass electrode ESL-63-07 (Russia) in a cell with liquid junction (aqueous saturated KCl solution) with the Ag/AgCl reference electrode EVL-1 M3 (Russia).The measurements were performed with the potentiometer P 307 (Russia) at 25.0 °C ± 0.1 according to the compensation scheme; the pH meter-millivoltmeter pH-121 was used as a nil-instrument.The repeatability did not exceed ± 0.3 mV.

Techniques Modification of gelatin films
A volume of 25 mL of indicator solution placed into the Petri dish and films with size 2.5×3.5 cm were dipped in the solution at room temperature, without stirring; reagents concentration and pH values in solutions are shown in Table 1.The films were pulled out of solution with tweezers; washed with distilled water, acidified with HCl to the necessary pH value and dried at air for 1 h.Thus prepared films were stored at room temperature in the dark; the dye concentrations in the films were 1-2 orders of magnitude higher as compared with those in the initial aquatic phases.

The determination of the apparent ionization constants in the two-phase system: water/gelatin film
Films with an immobilized indicator were dipped into the solution with a given pH value for 1 min and dried at air for 1 h at 18-22 o C before spectrophotometric measurements.
The acid-base equilibrium in the two-phase system can be described as follows: - Here the -H -B and -B denote the species located in the gelatin film (the charges are omitted for simplicity).The indices of the apparent constants were determined by the below formula: [9][10][11][12] (2) The pH value refers to the bulk (aqueous) phase, A B and A HB are absorbances of the film at the given wavelength after complete conversion into the corresponding form, and A is absorbance at the current pH.The ionic strength of the bulk phase (I) was as a rule 0.01 to 0.05 mol L -1 .
For calculation of the pK a app values, the results of 3-5 independent series were used; the experiments were repeated beginning from the preparation of the films and immobilization of the dyes.Finally, 20-30 pK a app values were utilized to obtain the averaged constant.The results obtained using different wavelengths (l max ± 10 nm) coincide.The changes of the pH values of the bulk solutions, resulting from the dipping procedure, are negligible.
During the dipping process, the aqueous phase stayed colorless within the working range of pH.The latter is limited by the conditions of dye extraction (Table 1).The color changes were reversible; for instance, the properties of bromophenol blue-containing films did not change after 30 to 50 soakings with solutions with different pH within the working acidity range.The air-dried films maintain their color during 1 year.

The binding of dyes by the gelatin films
The contact time of colorless films with indicator solutions, necessary for dye immobilization (Table 1), The water solutions contained ethanol (mass fraction 8%).For ethyleosin and n-decyleosin, the cationic species H 2 R + appear only in strongly acidic media, 11 not studied here.
was determined using the saturation curves.The latter were obtained by plotting the absorbance vs. dipping time.Under proper conditions, both acid and basic forms of indicator dyes studied can be extracted into the films.Owing to their hydrophobicity, the long-chained lipophilic dyes n-decylfluorescein and n-decyleosin always tend to be embedded into the films.The indicators neutral red and malachite green are preferably extracted within the pH range above the isoelectric point of gelatin (pH > 5) in the form of cations HR + and R + , respectively (Figure 1), while anionic species HR -and R 2-of bromophenol blue, bromophenol green (Figure 2), and eosin as well as R -ions of ethyleosin (Figure 3) are readily bound at pH < 5. Similar regularities were observed by binding of eosin Y with bovine serum albumin. 14The dye does not bind with protein in alkaline medium (pH 9), because the macromolecule has mainly negative charge due to ionization of amino acid residues, while the dye is in the form of dianion R 2-.On the contrary, the positively charged surface of protein is able to bind the anionic form of the indicator rose bengal. 15Anions of p-nitrophenol, bromocresol purple, chlorophenol red, bromophenol blue, and bromocresol green can be bound both by neutral or positively charged surface of n-dodecyldimethylamineoxide micelles. 18ome specific peculiarities were observed with the dye neutral red.While the positively charged red form HR + was successfully inserted in the gelatin films in pure water, at pH 6-7, the dye immobilized in the yellow form R at pH > 10 became red (HR + ) under immersion in acid solutions with pH < 5 and was not washed out, though the grid charge of gelatin gel is positive under such conditions.It might be supposed that in this case hydrophobic interactions and hydrogen bond formation overcome the electrostatic repulsion.Vol.22, No. 5, 2011   Interestingly, analogous effect was observed by studying the sorption and desorption of anionic dyes reactive purple 5, acid blue 74 and direct red 28 on the chitin gel. 38Under the conditions favorable for sorption (pH 5.8) strong retention of indicators was provided by electrostatic interaction of sulfonate groups of the dyes with protonated amides of the gel (pK a,monomer = 7.6).However, no desorption of indicator direct red was observed in alkaline medium at pH 10, where the grid charge of the gel is negative. 38he positions of the absorption maxima of the airdried films in the visible region prove the non-aqueous microenvironment of the dye species, bound by the gelatin macromolecules.

Spectral characteristics of indicators immobilized in the gelatin gel
The absorption bands of acid and basic forms of all the indicators under study, immobilized in gelatin film, are shifted as compared with those in aqueous solutions.The electronic absorption spectra of the conjugated forms of indicators in films and in solutions are exemplified in Figures 4 and 5.
Hypsochromic shifts of absorption maxima were observed for HR -species of bromophenol blue and bromocresol green, HR form of ethyleosin and HR + ion of neutral red, while the batochromic shifts were registered for basic forms of all indicators and acid form of malachite green (R + ) and molecular species of n-decylfluorescein (HR) (Table 2).
For the above anionic dyes analogous effects were observed in micellar solutions of cationic surfactants, e.g., N-cetylpyridinium chloride and in microemulsions based on these surfactants. 9,11,39,400][11][12] Batochromic shift of 10 nm was observed for basic forms of eosin, ethyleosin, methyleosin, and uranin (disodium salt of fluorescein), captured in gelatin layer of photographic plates manufactured by Agfa (Agfa Gevaert graphic gelatin film, 61101508). 7The alterations of l max give evidence for interaction of dyes with the gelatin gel. 14,15The closeness of the l max values in gelatin films and in micellar solutions of ionic surfactants allows to expect that the microenvironment of dyes in two media is similar. 33,35The maxima of the spectra of neutral molecular forms of ethyleosin and n-decyleosin are expressed less distinctly (Figure 4).The neutral form of eosin (H 2 R), captured in the films, is not colored.This indicates the shift of tautomeric equilibrium of this neutral form from the quinonoid, which is colored like that of ethyleosin (Figure 4, curves 2, 4) toward the colorless lactone (Figure 6). 39,40t is worth to point out that disappearance of fluorescence was observed for xanthenes during injection of it in gelatin film.It can be caused both by concentration quenching and dimerization of the dyes in the gelatin matrix. 33However, we have not registered alterations of the absorption spectra typical for dimer formation.

Apparent ionization constants of indicators in two-phase system: water /gelatin film
In order to determine the pK a app values, the absorbance of the dye-doped films was plotted against the pH values of aqueous solutions, where films were dipped into.The   In SDS micellar solutions, I = 0.05 mol L -1 (NaCl).c The neutral species HR of ethyleosin is poorly soluble in water; at 8% mass ethanol, at pH around 0.5, the spectrum with l max = 492 nm may reflect the appearance of the traces of R -anions.d 8% mass ethanol.e The pK a w is equated to that of ethyleosin, pK a w = 1.9.g The complete HR -spectrum was not singled out.h In the literature 10 the DpK a varies from 2.7 to 2.4 in 2 to 10 mass fraction % of SDS solutions, without supporting electrolyte.i The pK a w is equated to that of ethylfluorescein, pK a w = 6.31. 9dependences of the fraction of deprotonated form vs. pH are represented in Figure 7.
According to recent data, the content of water in hardened gelatin, also manufactured using the photographic films, is around 40 moles per kg; after heating up to 90 o C for 1 h near 80% H 2 O was removed. 41It is reasonable to consider this fraction of water as forming an internal aqueous medium within the gelatin film, while the rest H 2 O molecules are strongly bound by the macromolecules.Our experimental data, given below, are treated assuming that the pH value of the internal water, being constant within the whole massive of the film, is determined by the acidity of the solution, where the film is dipped in.Thus, the internal pH is conventionally equated to the bulk (external) buffer pH value, equation 2. The difference in the pH value of aqueous buffer solution and that in the microenvironment of the indicator dye is contributed into the pK a app value, causing its deviation from the true pK a value into the macromolecular phase.Naturally, the pH variation results in the alteration of the gird charge of the macromolecules owing to acid-base interactions of amino acid residues; the counterions are located in the internal water.
The numerical data are compiled in Table 2.In such systems (see Introduction), the following expression should be used for the pK a app value: Here pK a w is the thermodynamic pK a in water, Y is the local electrostatic potential, g B and g HB are activity coefficients of transfer of corresponding species from water to the gelatin phase, R is the gas constant and T is absolute temperature.The index of the so-called intrinsic constant, pK a i , equals to the sum pK a w + log(g B /g HB ).Equation 3 allows to explain the medium effects, DpK a , i.e., the deviations of pK app from pK a w .Indeed, the g B /g HB quantity is close to unity and, hence, pK a i → pK a w , then the Y value governs the medium effect.However, in the interfacial layers of surfactant micelles, phospholipid liposomes, microemulsion, etc., the pK a i value as a rule differs from pK a w , thus indicating the 'non-aqueous' character of the microenvironments. 9he net charge of gelatin films, and hence, the electrostatic potential value, is changing along with variation of pH due to the ampholytic nature of gelatin.Therefore, according to equation 3, the pK a app should vary depending on the working pH region.
The pK a app and DpK a of the dyes determined at pH values below the isoelectric point (pH ip ca. 5) resemble to more or less extent those determined in micellar solutions of cationic surfactants, 9,40 cationic surfactant-based microemulsions, 11 and cationic calixarenes. 43r the dye neutral red, the color transition occurs at pH values well above the pH ip .Hence, the pK a app value (Table 2) is close to that determined in aqueous SDS micellar solutions. 10The binding of this indicator to an anionic polyelectrolyte also results in a substantial increase in pK a app : in poly(sodium styrene sulphonate) aqueous solutions, the DpK a = 1.7 value was registered. 43he results obtained with malachite green are of especial interest.At high and medium pH in aqueous media, this cationic indicator slowly converts into the colorless carbinol (R + ROH + H + ). 43,445][46][47][48] while the non-equilibrium color transition corresponds to pH 11.5-14.0. 49Indeed, we determined the 'observed' value pK a w = 11.94 ± 0.11 in aqueous diluted NaOH solutions by registering the absorbance within 1 min after mixing. 50Contrary to aqueous systems, in the gelatin films no color change in time was observed.The pK a app = 12.44 should be considered as a true value, and the DpK a = 5.4 value for malachite green, together with the DpK a = 2.3 value for neutral red, proves the negative net charge of the gelatin structures bearing the indicator dyes.
The case of eosin is more complicated due to dibasic character of this acid, with two overlapping ionization steps.Indeed, in the pH-dependences of absorbance at proper wavelengths, this is clearly seen due to existence of two inflection points (see Supplementary Information).For calculation of the pK a1 and pK a2 values, the program CLINP 53 was used.The data of three separate experimental series were processed, using independently prepared gelatin films with the immobilized dye.
The increase in the pK a2 value of eosin, as compared with the value in aqueous solutions, gives evidence for the decrease in the positive net charge of gelatin along with approaching to the pH ip value.Here, the contribution of the last item in equation 3 is small, and pK a app → pK a i .In turn, the latter is always higher than pK a w for the second ionization step of eosin on going from water to micelles of non-ionic surfactants 54 and solutions of non-ionic polymers. 55n another paper, 56 the character of the dependence of absorbance at 552 nm (eosin concentration of 1.5 × 10 -4 mol L -1 ) on pH in water with gelatine mass fraction 0.05% addition indicates that pK a2 > 4.
The pH-induced color changes of the films can be easily observed by the naked eye (Table 3).Note, that these dyes, except the most hydrophobic ones (neutral red, n-decyleosin, and n-decylfluorescein), release from the films dipped into aquatic systems with "unfavorable" pH (see above).Hence, their species are unable to hold in the internal water if not fixed at the gelatin macromolecules.
For more precise monitoring of the acidity of solutions, for example, in sensor devices, the spectroscopic measurements can be used.The change of the gird charge of gelatin along with pH variation manifests itself most distinctly in the case of n-decylfluorescein (Figure 8).
The corresponding color transition varies from pH 2 to 10. Actually, this is in accordance with the pK a app values (4.9-5.5) and in cationic and anionic surfactant micelles respectively, depending on the bulk ionic strength. 9lso, the interfering of the first step, i.e., the dissociation of the cation H 2 R + , cannot be excluded; the corresponding pK a app values in two types of ionic micelles are (0.8-1.3) and (4.0-5.2) respectively. 9Interestingly, such behavior of n-decylfluorescein was also observed in the Langmuir-Blodgett carboxylic acid-based films. 33,35In contrast, for a similar dye, 2,7-n-dihexyl-n-octadecylfluorescein embedded into a polyurethane-based film, the transition from HR to R -occurs within a normal pH range; pK a app = 8.5. 57These results are in agreement with some other works, where the change of the interfacial charge and Y values, caused by acid-base interactions of interfacial functional groups, manifests itself in the alteration of the pK a app values of the bound indicator dyes. 12,18,28ith some other indicators, the pK a app drift was also registered, however, within a narrow pH region and not so distinct.Probably, the long hydrophobic hydrocarbon chain, fixing the dye in a definite constant position in the gelatin microenvironment, causes the peculiar behavior of n-decylfluorescein (in the case of n-decyleosin, such expressed pK a app alteration was not observed, because the complete ionization of this indicator occurs already in the acidic pH region, much lower than pH ip ).
The increase in the ionic strength of the aqueous buffer solutions, where the films were dipped in, from 0.05 to 0.50 mol L -1 (NaCl) displayed no distinct influence on the pK a app values of n-decylfluorescein.

The pK a app values of indicators in aqueous solutions of gelatin
Unusual behavior of n-decylfluorescein in the gelatin film impelled us to examine the acid-base equilibrium of the dye in gelatin solutions.Solutions with gelatin mass fraction 1% were used; further increase in the concentration resulted in high turbidity, which hinders the spectrophotometric measurements.The "titration curve" of n-decylfluorescein in gelatin solutions is of common type and thus differs essentially from that in gelatin films (Figure 8).The pK a app = 8.17 value (Table 4) is close to that in micelles of anionic surfactants.This value is markedly  .96± 0.04 b,h a In micellar solutions of N-cetylpyridinium chloride, I = 0.05 mol L -1 (KCl).b in SDS micellar solutions, I = 0.05 mol L -1 (NaCl).c In the literature 10 the DpK a varies from 2.7 to 2.4 in 2 to 10 mass % SDS solutions, without supporting electrolyte.d See Table 2. e 8 mass % ethanol.f In micellar solutions of cetyltrimethylammonium chloride, I = 4.0 mol L -1 (KCl).g For ethyleosin in N-cetylpyridinium chloride micellar solutions, I = 0.05 mol L -1 (KCl), pK a app = 0.5. 39h I = 0.05 mol L -1 (NaCl). 9

Table 3 .Figure 8 .
Figure8.Dependence of n-decylfluorescein absorbance at 515 nm in gelatin films on pH of buffer solutions where they were soaked in (1) and at 500 nm in gelatin solutions with mass fraction 1% on pH of the bulk phase (2).I = 0.05 mol L -1 at pH ≥ 1.3.

Table 1 .
Conditions of dye immobilization into the gelatin films

Table 2 .
The l max /nm and pK a app values of indicators and medium effects (DpK a = pK a app -pK a

Table 4 .
The l max /nm and pK a app values of indicators in water and in gelatin solution (ω = 1%), I = 0.05 mol L -1 (NaCl + buffer components)