Water-based strippable coatings containing bentonite clay for heavy metal surface decontamination

1 Introduction

Humanity has not always taken into consideration the ecological implications of its actions, but the gradual degradation of the environment and the impact on human health are important warning signals. The use of compounds containing heavy metals in many fields, such as: chemical industry, extractive industry, defense and security industry, processes of fabrication and loading of explosives and ammunition, has caused serious contamination problems. For example, high concentrations of heavy metals have been found in shooting ranges or in explosives and ammunition factories, produced by the decomposition of compounds such as mercury fulminate, lead azide, lead styphnate or other explosives included in the composition of primers and detonators. Therefore, the removal of these contaminants represents a real problem for which low cost and efficient solutions must be considered.

Various methods for the removal of heavy metals are described in the literature including chemical, electrochemical or mechanical methods, but their application is limited by the type of contamination, costs, duration of decontamination process and other factors. In this context, surface decontamination is very difficult to achieve due to the contaminant location within the material pores and cracks, which makes removal more challenging. Many decontamination methods for surfaces have been described in the literature: aspiration (Almeida et al., 2013), abrasion of surface layer (Bossart and Blair, 2003), washing with water or with solvents (Kohli, 2013), foams (Kohli, 2013; Sutton et al., 2007), gels (Sutton et al., 2007) or biomass (Lesmana et al., 2009), etc. In the case of surfaces contaminated with heavy metals, strippable coatings (Gray et al., 2001; Barakat, 2011; Gray and Bergbreiter, 1997; Parra et al., 2009; Tan et al., 2009; Rao and Lal, 2004) seem to display many advantages in comparison with other methods due to ease of application, pores/cracks infiltration, contaminant entrapment and removal, low-waste generation after decontamination, low-costs, wide range of application, and compatibility with different types of surfaces (metal, polymer, glass, wood).

The principle of this method consists in applying initially a polymeric solution over the contaminated surface. Then, the decontamination agent contained in this solution reacts with the contaminant, which is drawn and fixed into the polymer matrix. The dried coating can be finally exfoliated and the surface remains decontaminated (Gray et al., 2001).

Polyvinyl alcohol is one of the most used polymers for this purpose, due to its many properties that recommend it: film forming ability (Jia et al., 2014), non-toxic water-soluble polymer (it does not involve risks during use), non-carcinogenic (DeMerlis and Schoneker, 2003; NTP Technical Report, 1998) and biodegradable (Chiellini et al., 2003).

The polymeric solutions used for strippable decontamination coatings also contain a compound that chemically binds to the contaminant. Depending on the pollutant type, many decontamination agents are described in the literature: ethylenediaminetetraacetic acid disodium salt (EDTA) (Gray et al., 2001; John and Jordan, 2005; Elliott and Brown, 1989), nitrilotriacetic acid (Elliott and Brown, 1989), N,N-bis(carboxymethyl)-L-glutamic acid (Kolodynska, 2011), 2,3-dimercaptosuccinic acid, monoisoamyl dimercaptosuccinic acid, 2,3-dimercapto-1-propanesulfonic acid, trans cyclohexyl-1,2-diamino-tetra-acetic acid, diethylene triaminepentaacetic acid, D-penicillamine, N-acetyl-L-cysteine, thiopronine (2-mercaptopropionyl glycine), thiron (4,5-Dihydroxy-1,3-benzenedisulfonic acid disodium salt monohydrate), deferoxamine, deferiprone (3-hydroxy-1,2-dimethylpyridin-4(1H)-one) (Domingo, 1998). Crown ethers are also known for their capacity to strongly bind some heavy metals or radioactive ions, depending on their size (Duman and Ayranci, 2010; Budianta, 2011).

Another alternative to remove heavy metals or radioactive ions is to use adsorbents such as clays, which have a high capacity for contaminants adsorption (Budianta, 2011; Inel et al., 1998; El-Mofty et al., 2008; Do and Park, 2011; Eba et al., 2010; Ijagbemi et al., 2009). For example, attapulgite is used as adsorbent for Cu (Wang and Wang, 2010) or Pb (Zhang et al., 2011) ions, vermiculite for Cs (Skorobogatov and Eremin, 2012), bentonite or montmorillonite for Cd, Co, Pb, Ni, Cu, Zn, Mn, Cr, Fe (Bidry et al., 2013) and Hg (Green-Ruiz, 2009). Bentonite is recommended for decontamination because it has negative charged layers (Petr Praus and Motakova, 2012) and high cationic exchange capacities (80–150 m_equiv/100 g) (Vega et al., 2005), which allows Na⁺, Ca²⁺, or K⁺ ions to be easily exchanged with Me²⁺ ions.

Although the application of strippable coatings as decontamination method has been proved, there are still aspects that need to be solved and optimized, such as: the solvents utilized (surface-friendly and non-toxic), increasing the degree of decontamination, and obtaining biodegradable materials (reducing waste materials).

In this context, the present work deals with the elaboration and testing of a new composition for the polymeric solution employed as strippable coating for the removal of heavy metals from contaminated surfaces. The new idea presented within this paper consists in the employment of EDTA together with bentonite as trapping agents for the heavy metal, therefore coupling within the same composition complexation and adsorption mechanisms for the contaminant removal. The concentration influence of each component of the polymeric solution, i.e. polyvinyl alcohol, glycerin, EDTA and clay upon the most important properties of both solution and strippable coating is investigated, including decontamination ability. A copper surface contaminated with mercury was chosen for the decontamination experiments, since this heavy metal has been intensively used in primary explosives (in primers or detonators), while the most commonly used material for ammunition manufacture is copper. However, these coatings may also be applied on other types of surfaces.

2 Materials and methods

3 Results and discussion

The polymer solutions viscosity is an important parameter affecting the selection of both application method and type of the surface to be decontaminated. For example, solutions with higher viscosities are suitable to be applied by brushing while solutions with lower viscosities can be applied by spraying. A too high viscosity may also affect the mobility of the molecules, and as a consequence, the rate of heavy metal absorption, as well as the ability of the solution to enter into the pores and cracks of the surface.

Therefore, the first step of the study consisted in the investigation of the influence of each component upon the viscosity of the aqueous solutions. Besides PVA as film-forming component, the effect of the following additives was investigated: (a) glycerol, which is added as a plasticizer to improve the elasticity of the strippable coating, and (b) EDTA and bentonite which should bring and retain the heavy metal ions inside the coating by complexation and adsorption, respectively. Fig. 1 shows that increasing concentrations in PVA₉₈ or additive led to higher viscosities in all cases. Newtonian solutions resulted in the case of PVA alone and after the addition of EDTA or glycerol, while the PVA/bentonite solutions displayed a shear-thinning character, more pronounced at higher bentonite concentrations.

Close

Figure 1 The effect of PVA₉₈ – additive concentration upon the viscosity of their aqueous mixtures: (a) PVA₉₈; (b) glycerol (PVA₉₈ 10%); (c) EDTA (PVA₉₈ 10%); (d) bentonite (PVA₉₈ 10%).

Export to PPT

The effect of the additives upon the viscosity of the PVA aqueous solution may be explained through the physical interactions taking place between polymer and additives, for example hydrogen bonds. From this point of view, all the additives employed may form multiple hydrogen bonds per molecule with PVA, thus connecting the polymer macromolecules through physical forces. As a consequence, an apparent enhancement of the polymer molecular weight occurs, leading to an increase of viscosity. It seems that EDTA is the most efficient additive from this point of view, as it displays the strongest effect, very likely due to the stronger hydrogen bonds between COOH and OH groups, in comparison with the OH—OH interaction (Deruiter, 2005). Similar effects were noticed in the case of PVA₈₆-additive aqueous solutions.

The rate of coating formation, i.e. rate of solvent evaporation, is another important characteristic of such compositions. In practice, the evaporation rate of the decontamination solutions depends not only on temperature but also on many other factors (relative humidity, evaporation surface, etc.) (Wong et al., 2004). For these tests, several parameters have been maintained constant (evaporation surface, humidity and temperature) in order to compare the influence of the components on the evaporation rate. For each temperature and composition, triplicate measurements have been performed. The average values for the evaporation rate (measured at 25 °C, 30 °C and 35 °C), for all concentrations are displayed in Table 1.

The evaporation rate increased with temperature, as expected, and decreased with increasing additive concentration, presumably due to several factors: (a) higher viscosity; (b) hydrophilic character of the additives; (c) increasing number of intermolecular hydrogen bonds that leads to higher boiling points as a consequence of the higher amount of energy necessary to break them. The largest decrease was noticed for bentonite because of its high moisture retention capacity (Lim et al., 2013).

The thermal behavior of the polymer-additive films was investigated by DTA measurements. The thermograms of the PVA₉₈ films, similar to those of the PVA₈₆ films (results not shown) revealed three endothermic peaks (Fig. 2). For the strippable coatings containing PVA₉₈ and glycerol the first endothermic peak was observed between 70 and 170 °C and was ascribed to the crystallinity changes in the structure of the polymer (Moshin, 2011).

Close

Figure 2 DTA thermograms for PVA₉₈ – additive films.

Export to PPT

The next peak between 175 and 240 °C was associated with the PVA₉₈ crystalline domain melting (T_m), while the last peak is assigned to the decomposition process (T_d) beginning at 275 °C. Due to its role of plasticizer for PVA₉₈, glycerol leads to a shift of both T_m and T_d toward lower values (Table 2 – Exp. 2, 3).

Similarly, EDTA- and bentonite-containing films displayed a slight modification of these temperatures, but smaller than in the case of glycerol (Table 2 – Exp. 4–7). The decrease of T_m for PVA₉₈-additive films may be ascribed as well to the interactions existing among PVA₉₈ and the additive molecules.

The FT-IR analysis of the strippable coatings offered evidence about the interactions between the different components of each formulation (Fig. 3). The FT-IR spectra of the polymer-additive films displayed all the characteristic peaks of both components, some of them slightly shifted in comparison with the spectrum of the genuine compound due to their mutual interaction. Thus, in the case of PVA₉₈-glycerol films, the glycerol peak at 1039 cm⁻¹ (strong C—O stretching vibration of primary alcohol in glycerol) shifted to around 1044 cm⁻¹, while its intensity increased with glycerol concentration. Also, the vibrations in PVA₉₈ at around 3290 cm⁻¹ (ν_OH) and 1088 cm⁻¹ $({\mathrm{\nu }}_{\text{C—O}})$ shifted to higher values (3325 cm⁻¹ and 1096 cm⁻¹, respectively) due to the hydrogen bonds formed between PVA₉₈ and glycerol.

Close

Figure 3 FT-IR spectra of PVA₉₈ – additive films: (a) glycerol; (b) EDTA; (c) bentonite.

Export to PPT

For the PVA₉₈ – EDTA films, the specific peaks of the carboxylic acid groups were noticed at 3297 cm⁻¹ (ν_OH) and 1670 cm⁻¹ $({\mathrm{\nu }}_{\text{C⚌O}})$ , while the characteristic peak of $({\mathrm{\nu }}_{\text{C—N}})$ vibration was recorded at around 1019 cm⁻¹.

The spectra of PVA₉₈ – bentonite films displayed the Si—O stretching vibration at 1033–1037 cm⁻¹ shifted from 1015 cm⁻¹ in pure bentonite due to the formation of hydrogen bonds between the —OH groups of PVA and Si—O groups of bentonite (Atici, 2006). For the same reason, the 3290 cm⁻¹ peak in PVA₉₈ shifted to 3302 cm⁻¹ in the film.

The mechanical properties of the films are very important from the point of view of coating removal from the contaminated surfaces, i.e. the strippable coatings used for decontamination must be resistant and elastic. Therefore, the strippable coatings with PVA₉₈ and different concentrations of glycerol, EDTA and bentonite were subjected to tensile tests in order to observe how each component influences the mechanical properties of the obtained materials. The films that contain only PVA₉₈ and bentonite were brittle. Therefore, the PVA₉₈ + glycerol + EDTA + bentonite decontamination films were used to compare the results (Fig. 4 and Table 3).

Close

Figure 4 Stress–strain curves for PVA₉₈-additive films.

Export to PPT

The mechanical tests revealed that the modulus of elasticity was the highest in the case of PVA₉₈ films (Table 3 – Exp. 1) and decreased when EDTA or glycerol was present. The lowest modulus value was recorded for the film with 10% glycerol (Table 3 – Exp. 3), but increased as glycerol concentration decreased. The addition of bentonite determined an increase of the film modulus (Table 3 – Exp. 5). A similar dependence on composition was noticed for the maximum tensile stress. PVA₉₈ became less rigid with load increase, because the values of the modulus of elasticity for the second segment (0.3–0.4%) were smaller than the ones on the first segment (0.05–0.25%). This did not happen with the strippable coatings containing only glycerol because their modulus of elasticity was approximately equal on both segments.

The strippable coatings used for decontamination (Table 3 – Exp. 5) displayed a value for the elasticity modulus in-between PVA₉₈ and PVA₉₈-glycerol films, while the maximum tensile stress was comparable to the coating with 5% glycerol.

Comparing the results one should remark that the effect of glycerol is essential for obtaining the desired mechanical properties for these strippable coatings. Even if they were more resistant in the absence of glycerol, they would be too rigid and brittle to be useful for the purpose of decontamination.

According to the tensile measurements, one can say that each component modifies the mechanical properties of the strippable coatings, and therefore this aspect should be taken into account when decontamination solutions are prepared in order to obtain the desired characteristics of a strippable coating that can be easily peeled-off from the decontaminated surfaces, without cracking.

To estimate the decontamination capacity of strippable coatings of various compositions, HgCl₂ was deposited on the surface of some copper coupons. In order to “imitate” a long-term contamination, the contaminated coupons were heated in an oven at 700 °C for 60 min (Ebadian, 1998).

In order to assess the composition of the copper surfaces prior and after decontamination, SEM–EDX analysis was employed (Fig. 5). Fig. 5d and j reveals the presence of the HgCl₂ contaminant on the copper coupon surface as well as in the material defects. After the coupon surface was covered by the strippable polymer coating, only carbon and oxygen as the major constituents could be seen (Fig. 5e and k). Following the removal of the polymer coating (Fig. 5f and l), the absence of both contaminant agent and polymer was confirmed. Thus, the decontaminating effect of the strippable coating was easily demonstrated as well as its complete and facile removal from the contaminated copper coupons.

Close

Figure 5 SEM and EDX images at micrometric (a–f) and nanometric (g–l) scale for: (a, d, g, j)-copper coupon before decontamination; (b, e, h, k)-copper coupon with strippable coating on; (c, f, i, l)-copper coupon after the removal of the coating.

Export to PPT

SEM–EDX analysis also provided the information necessary for the calculation of the decontamination factor, by employing Eq. (1):

In order to establish the influence of each component on the decontamination degree, solutions of PVA₈₆ and PVA₉₉ containing different concentrations of glycerol, EDTA and bentonite were employed (Table 4). The results showed that glycerol did not influence decontamination (Table 4 – Experiments 1, 2, 12, 13), because it does not react with the contaminant agent, but only acts as a plasticizer.

EDTA employed as one of the decontamination agents displayed a major influence on the decontamination degree, as confirmed by the data in Table 4 – Exp. 3–6 and 14–17. The increase of EDTA concentration improved the decontamination factor, but after a certain concentration, DF started to decrease. This behavior may be explained by the viscosity increase determined by the higher EDTA concentration, which made the molecules to move slower through the solution and, consequently, decreased the complexation rate by EDTA. Also, a too high viscosity prevented the polymer solution to penetrate into the small pores and cracks on the coupon surface, and therefore HgCl₂ was not removed. According to the results from Table 4 – Exp.5, the decontamination solution with the composition: PVA₈₆ + 1% EDTA + 1% bentonite + 5% glycerol displayed the highest decontamination factor (DF = 76%). The addition of bentonite proved to be beneficial on the decontamination capacity as it is an excellent adsorbent for heavy metals (Talaat et al., 2011).

The concentration of bentonite had a major influence on solutions viscosity and increased the decontamination factor (Table 4 – Exp. 7–11 and 18–22), but after a certain concentration of bentonite, the DF begins to decrease, very likely due to the too high viscosity.

4 Conclusions

The present work demonstrates that PVA strippable coatings can be successfully used for the decontamination of surfaces that contain heavy metals. The novelty consists in using both complexation and adsorption mechanisms for the removal of the contaminant from a surface by employing EDTA together with bentonite as trapping agents for the heavy metal.

The method described within this work ensures a good decontamination factor for mercury salts even for deep contamination cases. These polymer aqueous solutions are non-toxic and non-carcinogenic and they can be safely handled and applied. The decontamination solutions may be used on any type of surface as the water employed as solvent does not affect the surfaces. They can be applied on the contaminated sites with a brush, in many layers, or they can be sprayed with a special device, depending on their viscosity.

The influence of the concentration of each component of the polymeric solution upon the most important properties of both solution and strippable coating were investigated in order to optimize the composition for improving the decontamination capacity.

The mobility of the decontamination agents is influenced by the viscosity of the polymer solutions; therefore, rheological measurements were useful in selecting the appropriate concentrations for these solutions. Bentonite is the only component that has a perceptible influence on the evaporation rate of water from the polymer solution, probably due to its high absorption and retention capacity of moisture. The formation of hydrogen bonds occurs between all the components of the decontamination solutions, as shown in FT-IR spectra, being all hydrophilic species.

Among all the components of the strippable coatings, glycerol is the one that influences the most the thermal behavior of the PVA films, being the only one that considerably shifts T_m and T_d to lower values.

According to tensile tests, bentonite acts also as strengthening agent, meanwhile glycerol having a noticeable plasticizer effect. The obtained coatings are resistant and can be easily peeled off, efficiently removing the contaminant this way. Another advantage of this method is that it is low-waste generator as the films can be packed and stored in sealable containers.

Decontamination tests showed that introducing bentonite in the polymer solution resulted in a significant increase of the decontamination factor, but the subsequent increase of the viscosity determines a maximum amount that can be added. The contaminant is entrapped in the coating by two mechanisms: complexation by EDTA and adsorption on bentonite and therefore it has more chances to be removed from the surfaces.

SEM–EDX analysis also revealed that the decontamination solution penetrates also the pores and cracks and when the coating is peeled off it also retains the contaminants that were present in these spaces.

Thus, the strippable coatings composition described within this work is easy to use, while its properties recommend it for the removal of heavy metal contaminants from any type of surface. This may represent a future solution for decontamination problems that have not been solved until now due to the cost or the toxicity of the solvents employed. The composition described here may find applications for the decontamination of the military shooting ranges, explosive and ammunition factories, but also for any other places where heavy metals are present, such as industry, and laboratories.