Water vapor adsorption by nanostructured polyphase compositions based on the solid component of welding aerosol

Introduction

Our earlier studies (Rakitskaya et al., 2015a, 2015b, 2016) showed that the nanoparticles constituent of the solid component of welding aerosols formed in electric arc plasma in the process of steel welding by electrodes of Ukrainian manufacture can be used as catalysts active in the reaction of ozone decomposition.

It is well known that the activity of oxide and metal catalysts as well as catalysts formed as a result of anchoring metal complexes on various supports in the process of air purification from gaseous toxicants (CO, SO₂ and O₃) substantially depends on the water content in them. In some cases, the activity of the catalysts can be lost because of the water adsorption from the gas–air mixture: water molecules are able to block surface catalytic sites and/or change chemical properties of catalyst surfaces (Rakitskaya and Ennan, 2012). The blockage of the active sites is caused by the competition between water and other gaseous molecules for active sites of catalysts (Shaw et al., 1995; Stenger et al., 1993; Tortorelli et al., 2014). The information about the features of water vapor adsorption on catalyst surfaces is necessary for optimizing conditions of catalyst application in air purification plants, ventilating systems, and personal respiratory protection devices for workers breathing in both gaseous toxicants and nanostructured aerosols. For instance, the influence of water adsorption on HEPA filters used in respirators intended for respiratory protection from aerodispersed nanomaterials was studied in detail (Gupta et al., 1993; Joubert et al., 2010; Mocho and Ouf, 2011; Ribeyre et al., 2014). Consequences of water adsorption are of special interest in the case of catalysts used in personal respiratory protection devices developed for workers exposed to the action of welding aerosol consisted of gaseous (including ozone) and solid components.

From the literature analysis, it can be deduced that systematic investigations of water vapor adsorption were realized only for atmospheric aerosols such as carbonaceous (carbon black) (Popovicheva et al., 2008) and silica-alumina (clinoptilolite, bentonite) (Hatch et al., 2014) ones.

Since the consequences of the water vapor adsorption on catalysts based on simple and complex oxides intended for ozone decomposition are not discussed in literature, the aim of the work was to investigate the water vapor adsorption by both freshly formed and modified SCWA samples.

Experimental

In the study, SCWA fractions with the aerodynamic diameter no more than 1 µm contained in the welding fume formed as a result of steel arc welding by electrodes with rutile (ANO-4 (ISO 2560 E432R 21)), rutile-aluminosilicate (ANO-21 (ISO 2560 E432RC 11), and carbonate-fluorite (TsL-11 (ISO E19.9NbB20) and UONI 13/55 (ISO 2560 E434B20)) coatings were used. The following welding conditions were met: U = 33 V, I = 140–150A and a welding speed of 4.5 mm/s. Such conditions are required to attain the maximum level of SCWA emission (Rakitskaya et al., 2016). Freshly formed SCWA samples were denoted as F-SCWA-ANO-4, F-SCWA-ANO-21, F-SCWA-TsL, and F-SCWA-UONI.

Modified SCWA samples were prepared as follows: 20 ml of distilled water were added to 1 g of SCWA and the suspension obtained was kept at 60℃ for 2 h at continuous stirring. After that, the particulate was filtered and dried at 110℃ till constant weight. The modified SCWA samples were denoted as M-SCWA-ANO-4, M-SCWA-ANO-21, M-SCWA-TsL, and M-SCWA-UONI.

The SCWA samples were investigated by X-ray diffraction phase analysis on a Siemens D500 diffractometer (CuK_α radiation, λ = 1.54178 Å) with a secondary beam graphite monochromator. The phases were identified based on ICDD (International Centre for Diffraction Data) PDF-1 databases provided as a part of the Siemens D500 diffractometer software.

Infrared analysis was carried out using a Perkin Elmer FT-IR spectrometer with resolution of 4 cm⁻¹; 1 mg of the material under study was compressed with 200 mg of KBr.

Water vapor adsorption was studied using a thermostat at 294.0 ± 0.2 K vacuum setup with a conventional McBain silica-spring balance. The adsorbate (bidistilled water) was doubly degassed by its freezing-unfreezing at the liquid-nitrogen temperature and further evacuation. Samples weighing (1.0–2.0) × 10⁻⁴ kg at first were dried at 383 K for 2 h. Their evacuation was carried out by a fore pump and an oil-vapour vacuum pump till constant weight. The residual pressure of 0.013Pa was monitored by a VIT-2M ionization-thermocouple vacuum meter. Water vapour puffing was performed till the constant weight attainment. The partial pressure of air was measured using a U-tube mercury manometer with an accuracy of ± 2.6 Pa. The time of equilibrium achievement was 24 h. Both a change in the sample weight caused by sorption and a difference of U-tube mercury manometer level were measured by a KM-6 cathetometer. Its accuracy was within 2%.

Results and discussion

Overall performance of SCWAs

The data obtained with the help of X-ray phase analysis and IR spectroscopy of SCWA samples obtained as a result of welding by ANO-4, UONI, and TsL electrodes, both freshly formed and further modified, were published earlier (Rakitskaya et al., 2015a, 2015b, 2016). As was found, the SCWA samples were polyphase and contained magnetite (the predominant phase in SCWA-ANO-4) as well as manganochromite, goethite, hematite, and water-soluble compounds, i.e. carbonates, fluorides, and chromates of alkaline metals. Here, we introduce the data characterizing a phase composition of previously not investigated SCWA-ANO-21. X-ray diffraction patterns presented in Figure 1 show that both F- and M-SCWA-ANO-21 are crystalline and contain magnetite, Fe₃O₄, [JCPDS 19-0629], as a predominant phase characterized by the following values of parameters d (Å): 2.968; 2.530; 2.098; 1.714; 1.615; 1.484; 1.280 for F-SCWA-ANO-21 and 2.970; 2.532; 2.099; 1.714; 1.615; 1.484; 1.279 for M-SCWA-ANO-21. For both samples, F-SCWA-ANO-21 and M-SCWA-ANO-21, a (311) reflection (2θ = 35.456°, d = 2.530 Å and 2θ = 35.421°, d = 2.532 Å) characteristic of Fe₃O₄ phase is most intense. Taking into consideration the (311) reflection parameters and using Scherrer equation, sizes of magnetite nanoparticles for F- and M-SCWA-ANO-21, were calculated. They were 37 and 34 nm, respectively. For other SCWAs, sizes of magnetite nanoparticles were varied from 15 to 68 nm (Rakitskaya et al., 2015a). IR spectra of SCWA-UONI and SCWA-TsL are rather complicated because of their phase irregularity and structural features of some phases (Rakitskaya et al., 2015a, 2015b). Whereas IR spectra of SCWA-ANO-4 (Rakitskaya et al., 2015a) and SCWA-ANO-21 (Figure 2), both freshly formed and modified, are very similar: their bands in the range of 1000–1100 cm⁻¹ are characteristic of stretching vibrations of an Fe–OH band in spinels and the bands in the low-frequency region are characteristic of stretching vibrations of an Fe–O band in a spinel structure. Some differences between freshly formed and modified SCWA samples are observed in the range of stretching and deformation vibrations of water molecules. For instance, for F-SCWA-ANO-21, stretching vibrations of OH groups show themselves as a complex band centered at 3409 cm⁻¹ with shoulders at 3457 and 3451 cm⁻¹ and their deformation vibrations take the form of a complex band centered at 1637 cm⁻¹ with a shoulder at 1616 cm⁻¹; for M-SCWA-ANO-21, a wide band at 3436 cm⁻¹ and a weak band at 1635 cm⁻¹ are observed in the above mentioned regions. For F-SCWA-ANO-4 and M-SCWA-ANO-4, the bands corresponding to stretching and deformation bands of OH groups are located identically. OPEN IN VIEWER

Figure 1.

XDR patterns of freshly formed (a) and modified (b) SCWA-ANO-21 samples.

OPEN IN VIEWER

Figure 2.

IR spectra of freshly formed (1) and modified (2) SCWA-ANO-21 samples.

Isotherms of water vapor ad/desorption

Figure 3 shows isotherms of water vapor ad/desorption by all freshly formed and modified SCWA samples under study. The isotherm profiles for all SCWA samples are similar: adsorption values increase slightly at P/P_s < 0.6 being evidence of a weak interaction between water molecules and nanoparticle surfaces; an abrupt growth in adsorption values at P/P_s > 0.6 is evidence of water vapor condensation in adsorbent macropores. In contrast to our results, isotherms of water vapor adsorption obtained for transition metal oxides, Cr₂O₃, and TiO₂, (Harju et al., 2005) demonstrate a smooth increase in adsorption values at P/P_s < 0.7. OPEN IN VIEWER

Figure 3.

Isotherms of water vapor adsorption (○) and desorption (•) by SCWAs obtained as a result of welding by UONI (1, 5), TsL (2, 6), ANO-4 (3, 7), and ANO-21 (4, 8), freshly formed (1–4) and further modified (5–8).

After modification of all SCWA samples, the appearance of their adsorption isotherms did not change; however, the adsorption values decreased. It can be explained by dissolving of water-soluble phases such as fluorides, carbonates, and chromates previously contained in the freshly formed SCWA samples. The most considerable decrease observed for M-SCWA-UONI and M-SCWA-TsL was caused by the highest content of water-soluble phases in their freshly formed forms. A high affinity between the water-soluble phases and water molecules is the cause of high adsorption values. It was confirmed by Grekova et al. (2016) who increased the water adsorption capacity of Al₂O₃ by its modification with hygroscopic CaCl₂ or LiCl.

Figure 3 also shows that the capillary condensation hysteresis loops for all F-SCWA are closed at P/P_s = 0.25–0.35 and their forms are similar except for F-SCWA-UONI (curve 1). A stepwise profile of the hysteresis loop not only for F- but also for M-SCWA-UONI, according to (Ermolenko and Efros, 1971), can indicate the biporous nature of the adsorbent. Widths of hysteresis loops (Δa, mmol/g) depend on the SCWA nature and are more narrow for the modified SCWA samples. For instance, the least Δa value was found for M-SCWA-ANO-21 (curve 8); the adsorption and desorption branches practically coincide and the hysteresis loop is closed at P/P_s = 0.55.

The adsorption isotherms were analyzed using BET equation in its linear form

P / P s a ( 1 - P / P s ) = 1 a m · C + C - 1 a m · C · P / P s

(1)

where a is an adsorption value at equilibrium relative pressure, P/P _s , a _m is a monolayer volume, C is a constant characterizing an affinity between the given adsorbate and adsorbent.

Equation (1) is realized for all adsorption isotherms obtained by us up to P/P_s ≈ 0.5 with correlation coefficient R² of 0.988–0.999.

A specific surface area of the adsorbents was estimated by the use of the following equation

S sp = a m × N A × ω · 10 - 20 , m 2 / g

(2)

where a _m is a monolayer capacity, mmol/g, N _A is Avogadro constant (6.022 × 10²³ mol⁻¹), ω is an effective cross-sectional area of one adsorbate molecule, m² (10.8 × 10⁻²⁰ m² for water).

The parameters calculated using equations (1) and (2) are presented in Table 1. It should be noted that S_sp values estimated for freshly formed SCWA samples are doubtful because of the presence of hygroscopic impurities in them. The minimum values of a _m , a_∞, and S_sp are observed for F- and M-SCWA-ANO-21 samples.

OPEN IN VIEWER

Table 1.

Structure-adsorption parameters of freshly formed and modified SCWA samples.

SCWA sample	am, mmol/g	a ∞ , mmol/g at P/Ps ∼ 1	C	Ssp, m2/g
F-SCWA-ANO-21	0.31	2.57	124.04	20
M-SCWA-ANO-21	0.44	3.05	61.89	29
F-SCWA-ANO-4	0.82	9.10	356.79	53
M-SCWA-ANO-4	0.62	6.08	41.55	40
F-SCWA-UONI	0.76	24.3	15.89	49
M-SCWA-UONI	0.83	13.6	56.12	54
F-SCWA-TsL	2.15	22.8	6.34	140
M-SCWA-TsL	1.15	9.60	8.99	75

Conclusion

SCWA samples both formed as a result of steel welding by some electrodes commonly used in Ukraine and those further modified with hot water were investigated by X-ray phase analysis, IR spectroscopy, and water vapor ad/desorption. Isotherms of water vapor ad/desorption, obtained for the first time, were described by BET equation up to P/P_s ≈ 0.5. All parameters estimated based on BET equation substantially depend on chemical and phase compositions of the SCWA. Values of both a monolayer capacity and a specific surface area of the SCWA samples, both freshly formed and modified, increase in the order SCWA-ANO-21 < SCWA-UONI < SCWA-ANO-4 < SCWA-TsL .