Gaseous mercury behaviour in the presence of functionalized styrene–divinylbenzene copolymers

Materials

A Chloromethylated styrene-divinylbenzene copolymer supplied by Purolite Romania (S-6.7%DVB, %Cl=14.22, GF=4.01 mmol Cl/gram. copolymer) was used as starting material together with a poly(styrene-co-divinybenzene) grafted with aldehyde groups synthesized in our laboratory [12, 16]. The percentage of divinyl benzene in the copolymers is the same for both cases.

Synthesis of aminophosphonate grafted on copolymer (AMINOPHOS)

The synthesis of the aminophosphonate functionalized on styrene-6.7%divinylbenzene was performed according to a method described previously [11]. A mixture of 5 g of benzaldehyde grafted onto a styrene-divinylbenzene copolymer with 2.93 mmols –CHO /gram of copolymer and 50 mL THF (tetrahydrofuran) was kept under stirring for 2 h at room temperature to allow the copolymer beads to swell then phosphite and amine were added. The –CHO group: diethylphosphite: propylamine molar ratio was 1:1:1. The final mixture was maintained under stirring for 20 h at a temperature of 60 °C. After cooling, the polymer beads were separated by filtration, washed with ethanol (3 × 20 mL), 1,2-dichloromethane (3 × 20 mL) and ethyl ether (3 × 20 mL), and dried at 50 °C for 24 h.

One-pot synthesis of hydroxyphosphonic acid grafted on styrene-divinylbenzene copolymer (PHOS)

The synthesis of the hydroxyphosphonic acid functionalized on styrene –6.7%divinylbenzene was performed according to a method described previously [12]. A mixture of 5 g of aldehyde grafted onto styrene-divinylbenzene copolymer, phosphorous acid, K₃PO₄ as catalyst and 50 mL tetrahydrofuran was maintained under stirring for 35 h at 55 °C. The benzaldehyde group: phosphorous acid: K₃PO₄ molar ratio was 1:1.5:0.016. After cooling, the polymer beads were separated by filtration, washed with distilled water, methanol, acetone and diethyl ether and dried at 50 °C for 24 h.

Elemental mercury removal in gas phase

The experimental lab scale device used for this purpose consisted of a glass reactor into which the solid was placed to form a fixed bed. The sorbent bed was prepared by mixing 20 mg of copolymer with 60 mg of sand. The temperature was kept at 150 °C. Elemental mercury (Hg⁰) in gas phase was obtained from a permeation tube. The concentration of Hg in gas phase was 100 μg m^–3. A gas stream (500 mL/min) was used to transport the mercury species through the sorbent. Two gaseous atmospheres were employed. A synthetic gas mixture containing compounds obtained from a coal oxy-combustion process (4% O₂, 0.1% NO, 0.1% SO₂, 12% H₂O, 100 ppm NO₂, 25 ppm HCl, 20% N₂, 64% CO₂) was passed through the reactor. For interpretation and comparison purposes an air atmosphere was also tested. Oxy-fuel combustion is one of the emerging CO₂ capture and storage technologies for coal-fired power generation. In oxy-fuel technology, coal is burned with a mixture of oxygen and recirculated flue gas, instead of air. The main components in the flue gas are CO₂, N₂, H₂O and O₂.

Breakdown curves were recorded using a continuous mercury monitor (VM-3000). The possible oxidation of gaseous mercury was evaluated by capturing the oxidized mercury in an adsorption resin located prior to the continuous mercury analyzer. For this purpose a Dowex^® 1 × 8 chloride form resin was placed at the exit of the reactor. Dowex is an ion exchanger designed specifically for the selective extraction of mercury (II) species. It was analyzed at the end of the retention experiments by means of an automatic mercury analyser (AMA 254) to determine the amount of mercury that had been oxidized.

After the mercury removal experiments, mercury speciation by thermal desorption (TD) of the solid samples was carried out. A continuous mercury analyzer (RA-915) coupled to a furnace (PYRO-915), both from Lumex, were used for the TD procedure [17]. The temperature of the equipment was continuously monitored by means of a thermocouple. The parameters used for the selective desorption were previously optimised. The mercury compounds desorbed from the sample were conducted through the chambers in a stream of air, at a flow rate of 1 l min^–1, the heating rate from room temperature to 750 °C was 50 °C min^–1. About 100 mg of sample was used for each thermo desorption analysis. The mercury species were identified according to temperature they were released from the sample and by comparison with the desorption temperatures characteristic of several reference mercury compounds collected from the literature [17].

Synthesis of AMINOPHOS and PHOS and characterization

A chloromethylated styrene-divinylbenzene copolymer (14.22% Cl) was used as the starting material in all the syntheses. An intermediate styrene-divinylbenzene coplymers grafted with aldehyde groups was synthesized in our laboratory [12, 16].

The copolymers functionalized with aminophosphonate groups were synthesized according to Scheme 1:

Scheme 1

The synthesis of the AMINOPHOS sample.

The copolymers functionalized with phosphonic acid groups were synthesized according to Scheme 2:

Scheme 2

The synthesis of the PHOS sample.

The fraction of repeat units functionalized with aminophosphonate and α-hydroxyphosphonic acid groups was determined by using the statistical structure of the repetitive unit of functionalized copolymers [1, 2] presented in Scheme 3 (a, b).

Scheme 3

Statistical structures of the repeat unit of the functionalized copolymers (a) aminophosphonate groups, (b) α-hydroxyphosphonic acid groups.

The main characteristics of the aminophosphonate groups and α-hydroxyphosphonic acid groups grafted onto a polymeric support by means of “one-pot” polymer-analogous reactions are given in Table 1. The notation used was: Ff, – aminophosphonate groups/α-hydroxyphosphonic acid groups; Fi, –CHO groups; r, fraction of divinylbenzen (DVB) units; x-y, fraction of styrene units bearing pendant –CH₂Cl remaining groups; z, fraction of styrene units bearing pendant – aminophosphonate groups/α-hydroxyphosphonic acid groups (Ff); %P – phosphorus percentage in the final copolymer; GF – functionalization degree [11, 12, 18]. The phosphorus content was determined according to the modified Schoninger method [18]. The phosphorus content of the functionalized copolymers was used in order to determine the fraction of styrene units bearing aminophosphonate/α-hydroxyphosphonic acid groups (Ff) and it was calculated using previously published equations [11, 12, 18].

Confirmation of the formation of aminophosphonate groups and α-hydroxyphosphonic acid groups on the styrene-6.7%divinylbenzene copolymer is obtained in the FTIR spectra (Fig. 1.) On the basis of: (i) an absorption band at around 1220 cm^–1 associated with the vibration of the P=O bond, (ii) a characteristic P-OH band appearing at around 1000 cm^–1 and (iii) a medium intensity band at 1600 cm^–1 assigned to the P(O)OH group.

Fig. 1

FTIR spectra, where: phosphonic acid groups grafted onto S-6.7%DVB (PHOS) -light gray; aminophosphonate groups grafted onto S-6.7%DVB (AMINOPHOS) – black.

The thermal behavior of the aminophosphonate and phosphonic acid resins (AMINOPHOS (a) and PHOS (b) samples) studied by thermogravimetric analysis in air, at a heating rate of 10 °C/min shows a characteristic “step weight” loss above 150 °C (Fig. 2). These samples can therefore be used for the removal of elemental mercury in gas phase.

Fig. 2

Thermograms corresponding to the AMINOPHOS and PHOS samples.

In the SEM image (Fig. 3) of the AMINOPHOS sample, macropores are clearly visible among the agglomerations of microspheres.

Fig. 3

SEM images of styrene-6.7%divinylbenzene copolymer functionalized with aminophosphonate groups (code- AMINOPHOS).

Mercury adsorption and oxidation tests

The mercury adsorption curves are shown in Fig. 4. The slope of the curve corresponding to PHOS is higher than the one obtained for the AMINOPHOS sample suggesting faster kinetic in this case. The Hg background level has not been reached for either solid indicating the possible oxidation of the mercury. The mercury retention capacities and oxidation rates are shown in Table 2. The retention capacities are lower than those obtained with the reference activated carbon (Filtracarb) [19] that is frequently used as sorbent specifically for mercury removal. However these values are comparable with the retention capacities of coal fly ashes [20] or biomass gasification chars [19] in an air atmosphere. This moderate retention capacity can be attributed to the absence of a carbonaceous matrix for mercury adsorption. However a significant degree of mercury oxidation was observed in both samples. The sample with higher retention capacity, AMINOPHOS, is the sample in which the maximum oxidation ratio was observed (Fig. 5). This suggests that oxidized mercury is most probable the captured species, as can was demonstrated subsequently in thermal desorption tests.

Fig. 4

Hg adsorption using modified S-6.7%DVB copolymers.

Fig. 5

Fate of Hg using S-6.7%DVB copolymers as Hg adsorbents for gaseous Hg⁰.

When the gaseous atmosphere contained reactive gases such as SO₂, NO₂ or HCl, (Oxy-fuel combustion atmosphere), the mercury retention capacity and oxidized mercury ratio decreased. These results can be attributed to the competition reactions of the gaseous components for the active sites. This assumption is supported by the low mercury oxidation ratio found in this atmosphere. No mercury was oxidized and therefore no mercury was captured on the surface of the solid. In addition, during the mercury retention experiments it was observed that the solid copolymers turned into a light brown color. It was confirmed that this change was due to the interaction between the solid and the gaseous atmosphere.

Thermal desorption profiles obtained for the post-retention samples are shown in Fig. 6. Similar desorption profiles were obtained for the AMINOPHOS sample in both atmospheres (Fig. 6a and c). A major peak at 148 °C and a minor peak at 250 °C are attributed to the presence of HgCl₂ and Hg(CN)₂ respectively. A simple peak with a maximum at 145 °C was observed for the PHOS sample in the air atmosphere (Fig. 6b) corresponding to the formation of HgCl₂. When this sample was tested under the oxyfuel combustion atmosphere two mercury species were identified by thermal desorption. Peaks at 160 °C and 450 °C corresponding to HgCl₂ and HgO, respectively, were observed (Fig. 6d). The formation of HgCl₂ is due to the presence of residual chloromethyl pendant groups resulting from the sample preparation, see Table 1 (where: x-y=0.05, the fraction of styrene units bearing pendant –CH₂Cl remaining groups). The formation of Hg(CN)₂ and HgO confirms the capture of oxidized mercury.

Fig. 6

Hg TPD profiles for mercury speciation in S-6.7%DVB copolymers. Continuous lines refer to tested samples in air and oxy-combustion atmospheres. Discontinuous lines refer to commercial mercury standards.

All these results suggest the involvement of amino groups in the interaction of gaseous mercury with the functionalized copolymers.