Thermodynamic characteristics of cotton dyeing with reactive dyes in non-aqueous media

Abstract Understanding the dyeing behaviors of reactive dyes onto cotton in non-aqueous media is helpful to facilitate the development of sustainable non-aqueous dyeing. Here, the thermodynamic characteristics of cotton dyeing with reactive dyes in the non-aqueous media of liquid paraffin (LP) and decamethylcyclopentasiloxane (D5), were investigated and compared to those in aqueous medium. The results showed that the adsorption isotherms of the reactive dyes in LP, D5, and water media accorded with the Freundlich model, indicating that the dyeing proceeded via the physical adsorption of multi-molecular layers. Furthermore, the dyeing enthalpy and dyeing entropy of cotton dyeing using reactive dyes in LP and D5 media were positive, which was significantly different from those in an aqueous medium. These properties help to understand and interpret the dyeing properties and behavior of reactive dyes onto cotton in LP and D5 media.


Introduction
Reactive dyes have received increasingly wide attention and have become the most important dyes for cotton dyeing (Blackburn & Burkinshaw, 2002;Pu et al., 2015); however, conventional cotton dyeing using reactive dyes shows a low dye uptake, requires large amounts of neutral salts, consumes large amounts of water, and produces large amounts of wastewater (Khatri et al., 2015;Liu et al., 2019). To solve the above problems, Liu and Wang developed decamethylcyclopentasiloxane (D5), a non-polar, non-aqueous medium for reactive dyes that cannot dissolve or are incompatible with water (Pei et al., 2019). The D5 non-aqueous dyeing of reactive dyes showed a high dye uptake (nearly 100%) and high fixation rate (higher than 90%), using a salt-free and water-saving process with low pollution production (Fu et al., 2016;Khatri et al., 2011). The LP non-aqueous medium dyeing developed by Shao's group retained the advantages of D5 non-aqueous medium dyeing and was environmentally friendly and safer for humans An et al., 2021;Fan et al., 2021;Pei et al., 2017). Therefore, it holds tremendous potential to revolutionize dyeing technology in the textile industry.
While the potential of such dyeing methods has been extensively demonstrated, it remains challenging to perform theoretical studies of dyeing to clarify the dyeing behaviors of reactive dyes in non-aqueous media (Bhavsar et al., 2017;Khatri et al., 2011;Ma et al., 2016). In this work, we report the thermodynamic characteristics of cotton dyeing with a reactive dye in D5 and LP media compared to those obtained in an aqueous medium. The dyeing affinities of the reactive dyes in D5 and LP media were much higher than those in the aqueous medium. The dyeing enthalpy and dyeing entropy of the reactive dyeing in D5 or LP medium were positive, which was significantly different from the values obtained in the aqueous medium. As a result, the dyeing behaviors of reactive dyes in the D5 and LP media could be reasonably explained and understood. Sustainable non-aqueous reactive dyeing is expected to revolutionize dyeing in the textile industry (Horii & Kannan, 2008;Sawada & Ueda, 2007).
The LP and D5 were purchased from Mike Chemicals & Instruments Co., Ltd., China. Anhydrous sodium carbonate and anhydrous sodium sulfate were purchased from Gaojing Fine Chemicals Reagent Co., Ltd., China. The chemical reagents were of analytical grade and used without further purification.

Preparation of dyeing solutions and the dyeing processes
The D5 or LP medium dyeing solutions with dye concentration of 0.005, 0.010, 0.015, 0.020, 0.025, 0.030, 0035, 0.040, 0.045, 0.050 g/L were prepared. The following steps were used to prepare the 0.01 g/L reactive dye solution: The reactive dye (0.002 g) was mixed with 0.1 ml of deionized water with stirring, and then 200 ml of D5 or LP was added with stirring. The dyeing solution was then ultrasonically dispersed for 5 min. The method was also used to prepare the other dyeing solutions with different dye concentrations.
The cotton fabric was dried to a constant weight in an oven at 80 C and stored in a sealed container before dyeing. Ten parts of 0.200 g cotton fabric were cut from predried cotton fabric, immersed in 30 g/L Na 2 CO 3 solution for 30 min, and squeezed until the liquid carrying rate was 120%. Next, it was put into the dyeing solution prepared above, heated to the specified temperature (40, 60, or 80 C), and then dyed with a dyeing machine (YP11BB10, Shanghai Chain-Lin Automation Equipment Co., Ltd., China). At a specific temperature, the solution ratio of 1: 1000 (cotton and non-aqueous medium) was maintained for 3 h to simulate an infinite dyeing bath. After dyeing was terminated, the remaining dye in the non-aqueous dye solution was extracted with an appropriate amount of ultrapure water and poured into a volumetric flask. It was brought to volume using deionized water, and the absorbance of the dyeing residue was determined by an UV-Vis spectrophotometer (Lambda35, Perkin Elmer Co. Ltd, USA) at the maximum absorption wavelength( k max of C.I. Reactive Red 195 is 542 nm, k max of C.I. Reactive Yellow 176 is 423 nm, and k max of C.I. Reactive Blue 194 is 600 nm).
The dyeing process in the aqueous medium was the same as that of conventional dyeing (Zhao, 2009). The cotton fabric mass, dye concentration, liquor ratio, dyeing temperature, and dyeing time in the aqueous medium were the same as those used in the non-aqueous media, except 40 g/L of anhydrous sodium sulfate was added. After dyeing, the absorbance of the residual dyeing solution (A r ) was measured.

Measurement and calculation of D f and D S
According to the measured A r , the corresponding concentration (C r ) (g/L) of the extracted residual dye solution was obtained from the standard working curve. Moreover, the amount of residual dye (q r ) (g) and the amount of dye on the dyed fiber (q f ) (g) were calculated by Equations (1) and (2), respectively. The dye concentration on the dyed fiber (D f ) (mg/g) and the dye concentration in the dyeing solution after dyeing (D s ) (mg/L) were calculated by Equations (3) and (4), respectively (Valko, 1957).
where q 0 is the initial amount of dye in the dye solution (g), V r is the volume of the extracted residual dye solution (L), m is the mass of fiber (g), and V s is the volume of the dyeing solution (L).
where R is the gas constant (8.314 JÁK À1 Ámol À1 ); T is the absolute temperature, K; D f and D s are the concentrations of dye on the fiber and in the dye solution at dyeing equilibrium, respectively; ÀDl 1 and ÀDl 2 are the standard affinities at temperatures T 1 and T 2 , respectively.
According to the relevant literature , the staining affinity (ÀDl o ) has a linear relationship with T, the staining heat (DH ) changes with temperature, and the staining entropy (DS ) is obtained from the slope of a linear fit. -Dl and DH were obtained using Equations (5) and (6), respectively.

Adsorption isotherms
It can be seen from Figure 1 that the D5 and LP media systems were more conducive to dyeing cotton fibers with reactive dyes than the conventional aqueous medium, and this phenomenon was more evident at higher temperatures. At higher temperatures, more dye was hydrolyzed in the aqueous medium, resulting in slow dyeing. In contrast, the D5 and LP media systems contained only a small amount of water, which inhibited dye hydrolysis. The same trend was obtained using C.I. Reactive Yellow 176 and C.I. Reactive Blue 194 (Figures S1 and S2).
There are three main types of adsorption isotherms: Nernst, Langmuir, and Freundlich (Briggs & Bull, 1922;Kongkachuichay et al., 2002). In this article, the above three adsorption isotherms of dyes in the non-aqueous media were studied.

Nernst adsorption isotherms
The Nernst adsorption model describes dyes with an affinity for the fiber, where the dye dissolves in the fiber. When dyeing equilibrium is reached, the ratio of the dye concentration on the fiber (D f ) to the dye concentration in the dyeing solution (D s ) is constant, which means that D f increases proportionally with D s until saturation is reached. This fully conforms to the distribution law described by Equation (8) (Zhao, 2009).
In the formula, D f and D s are the dye concentrations on the fiber and in the dye solution after dyeing, respectively; K is the proportional constant. Figures 2, S3, and S4 show that D f is not proportional to D s. Furthermore, see Tables 1, S1, and S2, the Nernst fitting parameter R 2 of reactive dyes in three media is low, about 0.95. In conclusion, the adsorption of reactive dyes onto cotton is not belong to the Nernst adsorption.

Langmuir adsorption isotherms
The localized adsorption of dyes generally occurs via chemical adsorption, which can be described by the Langmuir adsorption isotherm. All dye-seats on the fiber have the same dye adsorption ability, and they do not interfere with each other. One dye-seat can only adsorb one dye molecule via monolayer adsorption. When dye molecules have occupied all dye seats, dye saturation is reached. The adsorption equation (Zhao, 2009) is as follows: Where S f is the saturation value of dyeing, and K is a constant.
Figures 3, S5, and S6 indicate that 1/D f does not linearly increase upon increasing 1/D s , and Tables 2, S3, and S4 show that the Langmuir fitting parameters (R 2 ) of the reactive dyes onto cotton fabrics in different media were mostly around 0.95. Hence, the adsorption of reactive dyes onto cotton in non-aqueous media and aqueous media was not accurately described by the Langmuir adsorption model.

Freundlich adsorption isotherms
The unsaturated physical adsorption of dyes can usually be described by Freundlich adsorption isotherm, and the Equation(10) (Zhao, 2009) is as follows: where K is a constant, 0 < n < 1. It can be seen in Figures 4, S7, and S8 that lg[D] f increases linearly with the increase of lg[D] s , and Tables 3, S5 and S6 show that the fitting parameters R 2 are all around 0.99, and 0 < n < 1, indicating a high degree of fitting. So, it can be concluded that, the cotton dyeing with reactive dyes in D5, LP, and aqueous media conforms to the Freundlich adsorption isotherm.    Tables 4, S7, and S8.  Tables 4, S7, and S8 show that the dyeing affinity values in the D5, LP, and aqueous media were all positive, and the affinity in the aqueous medium was the lowest. Reactive    dyes are hydrophilic, and the dye uptake in the aqueous medium was much lower than that in the LP and D5 media. The higher dyeing affinity was caused by the higher chemical potential of the dye in the non-aqueous dyeing solutions. The dispersions containing reactive dyes, water, and non-aqueous medium were prepared and used with intensive mechanical agitation, so that enormous interfacial energy was generated between the droplets of dye liquid and the D5 or LP medium, greatly increasing the potential energy of the systems. Thus, the chemical potential of the dye in the dyeing dispersions (l s ) increased, while the chemical potential of the dye on the fiber (l f ) was almost same as that on the fiber dyed in an aqueous medium, resulting in an increased dyeing affinity according to the definition of affinity, ÀDl ¼ l s À l f , which promoted the dye molecules transferring to the fiber. This may explain why reactive dyes achieved almost 100% absorption in nonaqueous media without using neutral salts.
A higher dyeing temperature increased the affinity in the D5 or LP medium systems (Tables 4, S7, and S8), which was an opposite trend to the conventional dyeing system. This phenomenon can be explained by the positive dyeing entropy, as shown in Equation (7). The dyeing entropies of the cotton dyed with reactive dyes in the D5 or LP medium were positive (Tables 4, S7, and S8), revealing an increased system disorder due to the transfer of reactive dyes from the non-aqueous solutions to fibers. Generally, compared with the disorder degree of dyes in fibers, there should be a higher disorder degree of dyes in solution. Typically, the dyeing entropy is negative (Chen et al., 2015;Lee et al., 2019;Tam et al., 1997), but, the dyeing entropy values of cotton dyed with reactive dyes in the D5 or LP media were positive due to a change in the substances' physical states in the dyeing systems.
As shown in Figure 5, the hydrophobic binding of LP or D5 molecules and the cluster formed by the self-aggregation of water molecules in the system affected the uniformity of the dye distribution in the solution. This may account for the inherent uneven dyeing in non-aqueous media. In addition, this may also explain why mechanical stirring significantly improved the dyeing uniformity in non-aqueous media. Obviously, it is due to the weakened hydrophobic self-aggregation and water clusters under strong mechanical force action.
In general, dyeing enthalpy is negative, but it is positive in LP or D5 medium, as shown in Tables 4, S7, and S8, indicating that the dyeing with reactive dyes in the D5 or LP medium was an endothermic process, instead of an exothermic process, due to the changes in entropy. In the D5 or LP dyeing systems, the ice-like structures and the hydrophobic self-aggregation structures formed among the water molecules and the non-aqueous molecules, respectively.  During the dyeing, the ice-like structure "melted", and the self-aggregation structures "broke", due to the absorption of heat. Hence, the cotton dyeing with reactive dyes in LP or D5 medium was an endothermic process, and the dyeing enthalpy was positive.

Conclusions
The adsorption isotherms of reactive dyes onto cotton fabrics in LP, D5, and water conformed to the Freundlich adsorption model, a physical adsorption of multi-molecular layers. The dyeing affinity of reactive dyes in LP or D5 medium was much higher than that in aqueous medium due to the higher chemical potential of dyes in LP and D5 media. The dyeing affinity in aqueous medium decreased with increasing dyeing temperature as usual, while it did increase in the non-aqueous medium, due mainly to the positive dyeing entropy. The dyeing enthalpy and dyeing entropy of cotton dyed with reactive dyes in D5 and LP media were positive, which was significantly different from those in the aqueous medium.

Disclosure statement
No potential conflict of interest was reported by the authors.