The influence of nitrogen groups introduced onto activated carbons by high- or low-temperature NH 3 treatment on SO 2 sorption capacity

Abstract

Activated carbons modified with ammonia were used as sorbents of SO₂. The SO₂ uptake was dependent on both texture and type of surface groups introduced onto the studied materials. The most important parameter was the size of micropores, which positively influenced the amount adsorbed. Differing chemical properties had a mixed influence. The oxygen-containing groups negatively affected SO₂ amount adsorbed. N-containing moieties had a positive influence, while the amount of introduced nitrogen was not the major parameter where SO₂ uptake was concerned, possibly due to the differences in the type of introduced surface species.

Keywords

Activated carbon ammoxidation ammonia X-ray Photoelectron Spectroscopy (XPS)

Introduction

The removal of sulfur dioxide from combustion exhaust is of particular interest because of detrimental effects of SO₂ on human health and environment. SO₂ influences additionally the efficiency of DeNOx processes, as it can deactivate the catalysts used in SCR-NH₃ (Cheng et al., 2010; Jangjou et al., 2016; Olsson et al., 2016; Samojeden et al., 2015; Shi et al., 2015; Xi et al., 2014; Zhang et al., 2014). Typically, flue gases from power stations are cleaned via adsorption or absorption on sorbents using calcium carbonates or calcium oxides/hydroxides, with appropriate installations placed downstream of electrostatic precipitators. On the other hand, the methods using activated carbon (AC) were proposed (The European IPPC Bureau—Large Combustion Plants—BREF, n.d.). Adsorptive removal of SO₂ is a promising technique as it does not produce waste and offers the possibility to regenerate and reuse the spent adsorbent. Due to their good stability under low-temperature conditions, carbon-based materials, especially ACs and activated carbon fibers (ACF), are the most widely proposed adsorbents for SO₂ recovery (DeBarr et al., 1997; Wang et al., 2016; Yan et al., 2013).

In carbon structure, both textural properties and the main functionalities, such as carboxyl, carbonyl, phenols, lactones, and quinones, can influence the uptake of pollutants and thus impact adsorption efficiency (Boudou, 2003; Guo et al., 2015; Karousos et al., 2016; Li et al., 2015; Mangun et al., 1997, 2001a, 2001b; Samojeden and Grzybek, 2016; Severa et al., 2015). This has led to increased interest in seeking or improving modification methods, which could result in enhancing the adsorptive potential of ACs. Among the studied treatment methods was oxidation leading to the improvement of SO₂ sorption and lately there has been a growing interest in nitrogenation, because the latter can increase the basicity of AC surface which should improve the uptake of acidic molecules (Qiao et al., 2002).

There is no agreement in literature, however, which of the factors—texture or surface chemistry of carbonaceous materials—plays a dominant role in increasing the amount of adsorbed SO₂. Grzyb et al. (2009) tested ACs obtained by steam activation of chars produced from polyacrylonitrile and its blends with coal-tar pitch. SO₂ adsorption capacity of these materials was directly proportional to their micropore volume. On the other hand, Mangun et al. (2001b) showed that the adsorption capacity for dry SO₂ was dependent on both the pore size and the amount of basic groups incorporated into the carbonaceous structure. Boudou (2003) studied ammonia-treated commercial AC fibers and their SO₂ removal activity. The amount of adsorbed SO₂ was related to the pore structure of the ACFs modified by the introduction of nitrogen functional groups. Carbonaceous materials with low pore volume and surface area had low sulfur removal capacity. Ammoxidation at 600℃ slightly increased the porosity parameters and hardly improved the removal performance. Modification of ACF by ammonia/steam at 800℃ extended the AC microporosity and considerably improved the sulfur retention activity. Arenillas et al. (2005a, 2005b) claimed that although the amount of nitrogen incorporated into the adsorbent was important for adsorption of acidic molecules, the type of final nitrogen functionalities was more significant. Similarly, Drage et al. (2007) and Thote et al. (2010) claimed that N-containing groups played the main role in processes of SO₂ sorption on carbonaceous materials.

Taking into account these contradictory statements, this work aimed at comparing the relative importance of several parameters, both structural and chemical, for SO₂ removal on ACs. In order to fulfill this goal, AC was treated with ammonia at high temperature or ammonia–oxygen mixture at low temperature. As described by Grzybek et al. (2007, 2008), Klinik et al. (2011), Samojeden (2014), Samojeden and Grzybek (2016), and Samojeden et al. (2008, 2015), these modification procedures allowed to vary textural parameters and introduce different types and amounts of nitrogen groups.

Experimental

Nitrogen species were introduced into AC N (Gryfskand Hajnówka) by low- or high-temperature treatment with ammonia. The details of the applied modification methods are fully described in our previous works (Grzybek et al., 2007, 2008; Klinik et al., 2011; Samojeden and Grzybek, 2016). In short, the starting AC (further designated “C”) was modified by the following procedures:

High-temperature ones:

Nonoxidized activated carbon (C) was treated with NH₃ at 800℃ (2 h, 1 ml/min) (C/N800) and cooled under:

Ar flow (60 ml/min) to room temperature (C/N800/Ar), or

NH₃ flow (60 ml/min) to room temperature (C/N800/N);

AC oxidized with HNO₃ at 90℃ (C90) was treated with NH₃ at 800℃ (2 h, 1 ml/min) (C90/N800) and cooled under Ar flow (60 ml/min) to room temperature (C90/N800/Ar)

A low-temperature procedure:

Ammoxidation (NH₃–air mixture in 1:3 volume ratio) at 350℃ (2 h, 1 ml/min) of nonoxidized AC (C/NH₃–air) and oxidized AC (C90/NH₃–air).

The textural parameters were obtained from Ar sorption isotherms measured at liquid nitrogen temperature using a standard volumetric equipment (Samojeden, 2014). Specific surface area was calculated according to BET method. Micropore volume and constant B were determined by Dubinin–Radushkevich equation. Mesopore volume was obtained from BJH method (Samojeden, 2014). Sorption of sulfur dioxide was determined using standard volumetric equipment at 0℃ in the range of relative pressure p/p₀ = 0–0.3. Before the measurement the sample was degassed at 120℃ to vacuum of 10⁻⁵Pa.

Results and discussion

The characterization data of the modified ACs are summarized in Table 1.

Table 1.

Characterization of the studied samples (according to Samojeden (2014)).

Sample	C/NH₃–air	C90/NH₃–air	C/N800/Ar	C90/N800/Ar	C/N800/N
Textural data
S_BET (m²/g)	936	907	974	1212	974
V_mic (cm³/g)	0.338	0.287	0.381	0.413	0.354
V_mes (cm³/g)	0.031	0.079	0.054	0.102	0.025
B·10⁶ (J mol⁻³ K⁻¹)²	1.03	1.18	1.36	1.35	1.18
Elemental (bulk) analysis
N (wt%)	4.7	7.3	0.96	1.43	1.26
N/C molar ratio	0.059	0.086	0.015	0.014	0.015
Surface analysis
N/C (at%/at%)	0.052	0.065	Not measured	0.014	0.012
Type of N surface groups (binding energy BE for N1s)	398.7^a	398.8^a;		406.9^a	405.8^a
Type of N surface groups (binding energy BE for N1s)		400.6^a		401.8^a	399.2^a

The fitting of the N 1s peaks gave the following binding energies: 398.7–399.2 (pyridinic); 400.6 (pyridinic and/or pyrrolic); 401.8 (nitrogen substituted in aromatic graphene structures—quaternary nitrogen); 405.8–406.9 (chemisorbed nitrogen oxides) (Grzybek et al., 2004; Pietrzak, 2009); BE of 398-402 may also be interpreted as –NH₂ group (Grzybek et al., 2004; Luo et al., 2013; Shen et al., 2007).

As it is seen from Table 1, the studied ACs differed in both texture and chemical surface properties, to an extent dependent on the used treatment method. Chemistry of ACs modified with ammonia at high temperature was affected especially by the last stage of the preparation procedure, i.e. cooling conditions. Both the initial oxidation and the last stage of treatment (carried out either in unreactive Ar or reactive NH₃ atmosphere) influenced specific surface area and/or porosity. The existence of oxygen surface species on the starting AC (C90) resulted in higher specific surface area than for untreated AC modified with the identical high-temperature NH₃ procedure (cp. 1212 m²/g for C90/N800/Ar and 974 m²/g for C/N800/Ar). On the other hand, the B constant from Dubinin–Radushkevich equation is the same for both sorbents. Taking into account that, as discussed in Choma et al. (1993), the higher value of B indicates higher average pore size, this indicates almost identical type of microporosity for C90/N800/Ar and C/N800/Ar. It should be mentioned, however, that the former sample contained somewhat more mesopores than the latter.

The application of reactive atmosphere (as compared to the application of Ar) during several hours of cooling from 800 to 100℃ did not influence the specific surface area but led to the decrease of both the size of micropores and a slight decrease in their amount (C/N800/Ar as compared to C/N800/N). The decrease in micropore size may have been connected to the formation of additional nitrogen species on the surface, as expressed by slightly higher nitrogen content of C/N800/N.

The chemical character of the surface of the samples treated by NH₃ at high temperature also differs depending both on the starting material (preoxidized or not) and the final stage (cooling). Although the amount of nitrogen is not too different, the type of surface N-species is, which is expressed by differences in N1s binding energy. While C90/N800/Ar contains nitrogen substituted in aromatic graphene structures, C/N800/N is characterized mainly by pyridinic N-species, although due to difficulties in interpretation of N1s BE values the presence of, additionally, –NH₂ groups cannot be excluded (Grzybek et al., 2008).

Ammoxidized samples (C/NH₃–air and C90/NH₃–air) have similar specific surface area but differ in the volume and size of micropores, as well as in the amount of introduced N and the type of formed N-moieties (pyridinic for nonoxidized sample; pyrrolic and pyridonic for preoxidized one; in both cases the additional presence of –NH₂ groups cannot be excluded).

Figure 1 shows adsorption isotherms for all studied samples. From the figure the following sequence of amount adsorbed at p/p₀ = 0.2 may be derived: C90/NH₃–air< C/NH₃–air<C90/N800/Ar≈C/N800/Ar<C/N800/N. SO₂ amount adsorbed a(SO₂) is not proportional to specific surface area S_BET and thus S_BET is not the dominating structural parameter in case the type of porosity and/or chemical character of the samples differ.

Figure 1.

SO₂ adsorption isotherms for the tested samples.

In order to discuss the relative importance of other textural parameters (volume of micropores and their diameter, as expressed by factor B), as well as the possible influence of N content and chemical character of the surface, Figure 2 was drawn presenting the SO₂ amount adsorbed per 1 m² for lower p/p₀ where it may be expected that both texture and chemical character of the surface may play a more important role. From Figure 2 it may be seen that when two samples prepared by an identical procedure but starting with differently pretreated AC (without or with initial preoxidation, C and C90, respectively) are concerned, preoxidation of AC has a negative effect on SO₂ adsorption capacity. This is illustrated by the fact that C90/N800/Ar has lower SO₂ uptake than C/N800/Ar and so does C90/NH₃–air when compared to C/NH₃–air. This difference arises from the used oxidative pretreatment which increased the number of acidic surface species, as described in Grzybek et al. (2008). As discussed by Figueiredo et al. (1999), acidic surface groups, except the phenolic ones, decompose to CO₂ in TPD. Therefore, TPD m/e = 44 profile may be used as a rough estimation of acidity of carbonaceous materials. Such comparison shows that for both above-mentioned pairs of samples, the concentration of acidic sites was approximately three times higher for those originating from oxidized AC than nonoxidized one. An additional proof is the comparison of SO₂ uptake for the starting materials, C and C90, presented in Figure 3.

Figure 2.

SO₂ adsorption isotherms for the studied samples.

Figure 3.

SO₂ adsorption isotherms for the starting carbonaceous materials: unoxidized C and oxidized C90.

Further, based on Figure 2, the relative influence of microporosity, as well as the amount and type of nitrogen surface species may be discussed. As seen from Figure 2, SO₂ uptake is higher for C/N800/N than C/N800/Ar. For these carbonaceous materials two important parameters differ: (i) the size of micropores, as illustrated by B factor, which is smaller for the former than the latter sample, and (ii) (slightly) the amount of N surface species. The dominating role of the number of introduced N-moieties may be thus excluded. If the introduced amount of N had the major role, then a(SO₂) for C/NH₃–air with 4.7 wt% of N should be several times higher than for C/N800/Ar with only 0.96 wt% of N. Thus, the narrower pores seem to play an essential role in the sorption of SO₂ on the tested materials, in case of similar micropore and mesopore volume and comparable acidity. The former part of this conclusion is in good agreement with Mangun et al. (2001b) who compared SO₂ sorption capacity for unoxidized ACs treated with ammonia at temperatures 500, 600,700 and 800℃ and reported that the small pore size provided enough driving force to fill most of its pore volume.

Ammoxidized samples C/NH₃–air and C90/NH₃–air show similar SO₂ capacity per 1 m². The nonoxidized sample has narrower pores, contains less nitrogen, and smaller number of acidic groups, as indicated by lower CO₂ desorption peak from TPD (Samojeden, 2014). Again it is obvious that the total content of N is less important than the porosity. It is, however, possible that the opposing effects, positive ones from texture and the negative one arising from acidic surface moieties, cancel each other to a certain extent. In order to exclude the impact of different micropore volume of these two samples, Figure 4 depicts SO₂ sorption capacity per 1 cm³ of micropores as a function of p/p₀.

Figure 4.

SO₂ sorption per volume of micropores for the studied materials.

Based on Figure 4 the relative influence of the type of microporosity, as well as the amount and type of nitrogen surface species may be considered.

Figure 4 illustrates again that the size of micropores has larger influence on SO₂ uptake than other, i.e. chemical, parameters (cp. C/NH₃–air versus C90/NH₃–air, and C/N800/Ar and C90/N800/Ar versus C/N/800/N). However, it should be noted that the shape of sorption isotherms is different for carbonaceous materials treated with NH₃ at high temperature and those ammoxidized at low temperature. This indicates somewhat different interaction with SO₂. As the main difference between these samples, apart from the amount of introduced N, which obviously plays a minor role, is the type of N-containing surface species, it may be concluded that the type of surface groups, though a secondary parameter, is also important. It may be dominant in case the applied carbonaceous materials have similar textural characteristics. This agrees with Díez et al. (2015) who reported that basicity formed by N-moieties introduction influenced the uptake of another acidic molecule (CO₂).

The above discussion stresses the fact that the modification of carbonaceous materials affects their sorption capacity toward SO₂ in a rather complicated way. The final effect is a superposition of the changes of both texture and chemical character of the surface. The most important parameter is the size of micropores, assuming there are no major differences in the chemical character of the surface. In case texture (size and volume of micropores) is similar, the type of surface groups starts playing an important role. The presented research does not, however, indicate which N-containing surface moieties would be of the most advantage for SO₂ removal process due to the complicated nature of such species formed on ACs. Thus, some more detailed studies are necessary, possibly on carbonaceous materials originating from nitrogen-containing polymers.

Conclusions

ACs prepared by modification with ammonia at high temperature or a mixture ammonia–air at low temperature and differing in textural and chemical properties were used as sorbents for SO₂ removal. The SO₂ uptake was dependent on both texture and type of surface groups on the studied materials. The most important parameter was the size of micropores, which positively influenced the amount adsorbed. Differing chemical properties had a mixed influence: (i) the oxygen-containing groups negatively affected a(SO₂) due to their acidic character; (ii) N-containing moieties had a positive influence, whose extent, however, was much lower than that of size of micropores; and (iii) the amount of introduced nitrogen was not the major parameter where SO₂ uptake was concerned, possibly due to the differences in the type of introduced surface groups.

Footnotes

Acknowledgements

First presented at the 15th Ukrainian–Polish Symposium on Theoretical and Experimental Studies of Interfacial Phenomena and their Technological Applications, Lviv, Ukraine, 12–15 September 2016.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was financially supported by Grant AGH (no. 11.11.210....).

References

Arenillas

Drage

Smith

(2005) CO₂ removal potential of carbons prepared by co-pyrolysis of sugar and nitrogen containing compounds. Journal of Analytical and Applied Pyrolysis 74(1–2): 298–306.

Arenillas

Rubiera

Parra

(2005) Surface modification of low cost carbons for their application in the environmental protection. Applied Surface Science 252(3): 619–624.

Boudou

(2003) Surface chemistry of a viscose-based activated carbon cloth modified by treatment with ammonia and steam. Carbon 41(10): 1955–1963.

Cheng

Lambert

Kim

(2010) The different impacts of SO₂ and SO₃ on Cu/zeolite SCR catalysts. Catalysis Today 151(3–4): 266–270.

Choma J, Jaroniec M, Burakiewicz-Mortka W, et al. (1993) Wpływ struktury porowatej we¸gla aktywnego na adsorpcje¸ benzenu z roztworœodnych (Influence of the porous structure of activated carbon on benzene adsorption from aqueos solution). Ochrona Środowiska 3(50): 45–48.

DeBarr

Lizzio

Daley

(1997) Adsorption of SO₂ on bituminous coal char and activated carbon fiber. Energy and Fuels 11(18): 267–271.

Díez

Álvarez

Granda

(2015) N-enriched ACF from coal-based pitch blended with urea-based resin for CO2 capture. Microporous and Mesoporous Materials 201(C): 10–16.

Drage

Arenillas

Smith

(2007) Preparation of carbon dioxide adsorbents from the chemical activation of urea-formaldehyde and melamine-formaldehyde resins. Fuel 86(1–2): 22–31.

Figueiredo

Pereira

Freitas

(1999) Modification of the surface chemistry of activated carbons. Carbon 37: 1379–1389.

10.

Grzyb

Albiniak

Broniek

(2009) SO₂ adsorptive properties of activated carbons prepared from polyacrylonitrile and its blends with coal-tar pitch. Microporous and Mesoporous Materials 118(1–3): 163–168.

11.

Grzybek

Klinik

Samojeden

(2007) Badanie węgla aktywnego modyfikowanego związkami azotu i manganu w reakcji redukcji tlenku azotu amoniakiem (The investigation of active carbon modified with nitrogen and manganese compounds in the reaction of reduction of nitric oxide with ammonia). Gospodarka Surowcami Mineralnymi 23(3/3): 133–142.

12.

Grzybek

Klinik

Samojeden

(2008) Nitrogen-promoted active carbons as DeNOx catalysts. 1. The influence of modification parameters on the structure and catalytic properties. Catalysis Today 137(2–4): 228–234.

13.

Grzybek

Rogóż

Papp

(2004) The interaction of NO with active carbons promoted with transition metal oxides/hydroxides. Catalysis Today 90(1–2): 61–68.

14.

Guo

Zhu

(2015) Investigation of SO₂ and NO adsorption species on activated carbon and the mechanism of NO promotion effect on SO₂. Fuel 143: 536–542.

15.

Jangjou

Wang

Kumar

(2016) SO₂ poisoning of the NH₃-SCR reaction over Cu-SAPO-34: effect of ammonium sulfate versus other S-containing species. ACS Catalysis 6(10): 6612–6622.

16.

Karousos

Labropoulos

Sapalidis

(2016) Nanoporous ceramic supported ionic liquid membranes for CO₂ and SO₂ removal from flue gas. Chemical Engineering Journal 313: 777–790.

17.

Klinik

Samojeden

Grzybek

(2011) Nitrogen promoted activated carbons as DeNOx catalysts. 2. The influence of water on the catalytic performance. Catalysis Today 176(1): 303–308.

18.

Guo

Zhu

(2015) Adsorption and desorption of SO₂, NO and chlorobenzene on activated carbon. Journal of Environmental Sciences (China) 43: 128–135.

19.

Luo

Chi

Wang

(2013) Electrochemical surface modification of carbon mesh anode to improve the performance of air-cathode microbial fuel cells. Bioprocess and Biosystems Engineering 36(12): 1889–1896.

20.

Mangun

Barr

Riha

(1997) Adsorption of SO₂ onto oxidized and heat treated activated carbon fibers (ACFS). Carbon 35(3): 411–417.

21.

Mangun

Benak

Economy

(2001a) Surface chemistry, pore sizes and adsorption properties of activated carbon fibers and precursors treated with ammonia. Carbon 39(12): 1809–1820.

22.

Mangun

DeBarr

Economy

(2001b) Adsorption of sulfur dioxide on ammonia-treated activated carbon fibers. Carbon 39(11): 1689–1696.

23.

Olsson

Wijayanti

Leistner

(2016) A kinetic model for sulfur poisoning and regeneration of Cu/SSZ-13 used for NH₃-SCR. Applied Catalysis B: Environmental 183: 394–406.

24.

Pietrzak

(2009) XPS study and physico-chemical properties of nitrogen-enriched microporous activated carbon from high volatile bituminous coal. Fuel 88(10): 1871–1877.

25.

Qiao

Korai

Mochida

(2002) Preparation of an activated carbon artifact: Oxidative modification of coconut shell-based carbon to improve the strength. Carbon 40(3): 351–358.

26.

Samojeden B (2014) Wpływ modyfikacji węgli aktywnych związkami azotu na ich właściwości fizykochemiczne (The effect of modification with nitrogen compounds on activated carbons to their physicochemical properties). In: Grzegorz SJ (ed.) Paliwa i Energia XXI Wieku. Kraków: Wydawnictwo Naukowe “Akapit”, pp.519–536.

27.

Samojeden

Grzybek

(2016) The influence of the promotion of N-modified activated carbon with iron on NO removal by NH₃-SCR (selective catalytic reduction). Energy 116: 1484–1491.

28.

Samojeden

Klinik

Grzybek

(2008) Charakterystyka modyfikowanych węgli aktywnych jako katalizatorów w reakcji DeNOx (The characterization of modified active carbons for DeNOx reaction). Gospodarka Surowcami Mineralnymi 24(3/3): 295–303.

29.

Samojeden

Motak

Grzybek

(2015) The influence of the modification of carbonaceous materials on their catalytic properties in SCR-NH₃. A short review. Comptes Rendus Chimie 18(10): 1049–1073.

30.

Severa

Bethune

Rocheleau

(2015) SO₂ sorption by activated carbon supported ionic liquids under simulated atmospheric conditions. Chemical Engineering Journal 265: 249–258.

31.

Shen

Huang

(2007) Thermo-physical properties of epoxy nanocomposites reinforced with amino-functionalized multi-walled carbon nanotubes. Composites Part A: Applied Science and Manufacturing 38(5): 1331–1336.

32.

Shi

Tan

(2015) Inhibitory effect of SO₂ on side reactions of NH₃-SCR over olivine. Catalysis Science and Technology 5(7): 3613–3623.

33.

The European IPPC Bureau—Large Combustion Plants—REF (n.d.). Available at: http://eippcb.jrc.ec.europa.eu/reference/BREF/LCP_FinalDraft_06_2016.pdf (accessed 5 January 2017).

34.

Thote

Iyer

Chatti

(2010) In situ nitrogen enriched carbon for carbon dioxide capture. Carbon 48(2): 396–402.

35.

Wang

Meng

Chen

(2016) Desulphurization performance and mechanism study by in situ DRIFTS of activated coke modified by oxidization. Industrial and Engineering Chemistry Research 55(13): 3790–3796.

36.

Ottinger

Liu

(2014) New insights into sulfur poisoning on a vanadia SCR catalyst under simulated diesel engine operating conditions. Applied Catalysis B: Environmental. 160–161(1): 1–9.

37.

Yan

Liu

Zhang

(2013) Activated semi-coke in SO₂ removal from flue gas: Selection of activation methodology and desulfurization mechanism study. Energy and Fuels 27(6): 3080–3089.

38.

Zhang

Wang

Liu

(2014) SO₂ poisoning impact on the NH₃-SCR reaction over a commercial Cu-SAPO-34 SCR catalyst. Applied Catalysis B: Environmental 156–157: 371–377.