Removal of gaseous sulfur and phosphorus compounds by carbon-coated porous magnesium oxide composites

Under dry conditions, the sorption capacities of the MgO/C composite with a low carbon content of 6.39 wt% MgO/C-1; 67.8 mg/g for DMMP and 35.3 mg/g for 2-CEES were higher than those of

Removal of gaseous sulfur and phosphorus compounds by carbon-coated

porous magnesium oxide composites

Anh-Tuan Vua,b, Keon Hoa, Chang-Ha Leea,⇑

a

Department of Chemical and Biomolecular Engineering, Yonsei University, Seoul, Republic of Korea

b

School of Chemical Engineering, Hanoi University of Science and Technology, Hanoi, Vietnam

h i g h l i g h t s

Carbon-coated MgO composites were

synthesized using an aerogel method

MgO/C composites had a high surface

area of 723 m2/g

The sorption capacity of DMMP and

2-CEES was higher than those of MgO

and AC

The composite sorbed DMMP almost

twice more than 2-CEES in dry

condition

Carbon layer on MgO protected active

catalytic and sorption sites from H2O

molecules

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:

Received 1 June 2015

Received in revised form 31 July 2015

Accepted 1 August 2015

Available online 28 August 2015

Keywords:

MgO composite

Carbon

Aerogel

Sorption

2-CEES

DMMP

a b s t r a c t

Carbon-coated porous magnesium oxide (MgO/C) composites were synthesized using an aerogel route for removal of dimethyl methylphosphonate (DMMP) and 2-chloroethyl ethyl sulfide (2-CEES) in dry and wet conditions The sorption capacities of the as-prepared samples for DMMP (0.23lg/mL) and 2-CEES (0.26lg/mL) were evaluated by breakthrough experiments in nitrogen under ambient conditions MgO/C composites exhibited a decrease in surface area with carbon content (648–723 m2/g), but had a higher surface area than MgO Under dry conditions, the sorption capacities of the MgO/C composite with

a low carbon content of 6.39 wt% (MgO/C-1; 67.8 mg/g for DMMP and 35.3 mg/g for 2-CEES) were higher than those of pure MgO and activated carbon (AC) The sorption capacity of MgO/C composites decreased with an increase in carbon content and became even lower than those of MgO and AC Under humid conditions, the sorption capacities and breakthrough time of pure MgO decreased significantly and became lower than that of AC In contrast, the sorption capacities of the MgO/C-1 composite for DMMP and 2-CEES under humid conditions remained at about 91 and 86% of those measured under dry conditions, and were higher than those of AC In addition, the MgO/C composite remained reactive toward 2-CEES even under humid conditions MgO/C composites were more stable than MgO under humid conditions because of the presence of carbon-coated shells

Ó 2015 Elsevier B.V All rights reserved

1 Introduction

Removal of hazardous chemicals from the environment is a

crit-ical issue from both biologcrit-ical and environmental standpoints

[1,2] Environmental regulations for emissions from industries and acceptable levels of human exposure are continuously being adjusted and made more stringent

The most abundant hazardous components can be classified into two categories based on their source: natural and anthro-pogenic hazardous materials Anthroanthro-pogenic pollutants generally originate from combustion, chemical reactions, or from the

http://dx.doi.org/10.1016/j.cej.2015.08.083

1385-8947/Ó 2015 Elsevier B.V All rights reserved.

⇑Corresponding author Tel.: +82 02 2123 2762; fax: +82 02 312 6401.

E-mail address: leech@yonsei.ac.kr (C.-H Lee).

Chemical Engineering Journal

j o u r n a l h o m e p a g e : w w w e l s e v i e r c o m / l o c a t e / c e j

unsecured effluent of toxic materials Furthermore, a considerable

amount of anthropogenic chemicals containing sulfur and

phos-phorus such as mustard, sarin, and soman agents are produced

highly persistent in the environment and critically harm humans

even at low concentrations

Many efforts have been made both to reduce pollution and to

eliminate toxic materials from the environment Adsorptive

removal of toxic components from the atmosphere is one of the

Among various candidate solid materials, metal oxides have

been reported to be effective materials for adsorption and

decom-position of toxic and persistent chemicals due to their high surface

areas, large number of highly reactive edges, corner defect sites,

unusual lattice planes, high surface-to-volume ratio, and

reusabil-ity [5–11] Particularly, magnesium oxide (MgO) materials have

been the focus of attention for decontamination of toxic chemicals

in recent years because well-designed MgO materials have a high

sorption capacity as well as effective decomposition ability

Many studies have focused on methods to synthesize MgOs

with high surface area, pore volume, and small crystal size to

improve sorption capacity and reactivity for toxic chemicals

[12–15] However, sorption capacity and reactivity of MgO are

sig-nificantly reduced in the presence of water, because of the strong

performance of hazardous chemicals on adsorbents under dry

con-ditions does not guarantee effectiveness in practice, because these

chemicals exist in the environment with a certain level of

humid-ity Even though many solid materials have been reported to have

higher efficiency under dry conditions than activated carbon, their

performance is often equivalent or worse than that of activated

carbon under humid conditions Therefore, selection of solid

materials for the removal of hazardous chemicals under humid

conditions has been limited to strong hydrophilic adsorbents It

would be highly desirable to design MgO-based composites

with hydrophobic surfaces that show high removal capacity for

hazardous chemicals under humid conditions

Carbon-coated metal oxides have been produced using various

synthesis methods for various applications A carbon-coated metal

oxide was developed to improve the stability, electric conductivity,

metal oxide with hydrophobic carbon could minimize the problem

addition, various carbon-coated metal oxides have been developed,

such as carbon-coated ZnO and CaO prepared by poly vinyl alcohol

composites, various synthesis methods have been suggested,

including chemical vapor deposition (CVD), pyrolysis of a

magne-sium hydroxide aerogel modified with resorcinol, precipitation

A chemical weapon agent (CWA) is a chemical substance whose

toxic properties are used to kill, injure, or incapacitate soldiers (and

sometimes civilians) Therefore, the decontamination of chemical

warfare agents is arguably the most challenging issue facing

militaries around the world Furthermore, due to possible terror

threats, protecting civilians from CWAs has become increasingly

important for many governments

Sulfur mustard (SM), commonly known as mustard gas, is a

class of related cytotoxic and vesicant CWAs, which can cause large

known antidote or specific treatment against SM exposure, and

the current therapy is largely supportive Like some other nerve

CWAs, sarin attacks the nervous system by interfering with

is used as a simulant for toxic phosphorus compounds such as sarin

Therefore, development of effective materials for removing DMMP and 2-CEES could help protect the environment and humans from refractory hazardous chemicals as well as CWAs

In this study, carbon-coated magnesium oxide (MgO/C) com-posites were synthesized via an aerogel route to remove efficiently CWAs As representative surrogates of CWA, dimethyl methylphos-phonate and 2-chloroethyl ethyl sulfide were selected The sorption capacities of the MgO/C composites were measured by breakthrough experiments in gas phase under ambient dry and

respectively) The removal efficiency of the MgO/C composites were evaluated and compared with those of pure MgO and activated carbon (AC)

2 Experimental section 2.1 Materials

The following materials were used in this study: toluene (Aldrich, USA, 99.9%), a magnesium methoxide solution in metha-nol (Aldrich, 7.82%), 2-CEES (Aldrich, 98%), DMMP (Aldrich, 97%), glucose (Aldrich, 99%), activated carbon (Calgon Filtrasorb 300,

(Dae-Deok Gas, 99.999%), and air (Dae-Deok Gas, 99.999%) All chemicals and solvents were used without further purification

2.2 Preparation of composites MgO/C composites were developed using an aerogel procedure

In a typical experiment, which was conducted at room tempera-ture, a mixture of toluene (100 mL) and magnesium methoxide solution (20 mL) was placed in a glass reactor with a stirrer Another solution was prepared by dissolving the desired amount

of glucose in 2.0 mL of distilled water and this solution was slowly added using a syringe to prepare the mixture The addition of the glucose solution led to a white cloudy precipitate, but the solution became clear after a few minutes To minimize evaporation of organic solvent, the glass reactor top was covered with aluminum foil The mixture was stirred vigorously overnight at room temper-ature to allow completion of hydrolysis

Subsequently, the hydrolysis gel was put into a high-pressure autoclave reactor The gel was first flushed and then pressurized

and this temperature was remained for 10 min Solvent vapors in the reactor were quickly vented to the atmosphere and flushed

for 12 h to remove residual organic solvents Hydrated powder (before calcination) was obtained and denoted HY-MgO/C The final step was calcination Calcination has been shown to improve the textural properties and sorption capacity of MgO

[12,29] In this study, the hydrated powder was calcined in a furnace under vacuum using the following steps to produce mesoporous MgO with a high surface area: (1) ramping from room

and soaking at 500°C for 5 h The magnesium oxide/carbon

In this study, to investigate the effect of the amount of carbon in

the MgO/C composites on their textural and physical properties as

well as sorption capacities, three different magnesium oxide/

carbon composites were prepared: molar ratios of magnesium

methoxide to glucose were 1:0.05, 1:0.09 and 1:0.18 The

as-prepared composites were denoted MgO/C-1, MgO/C-2, and

MgO/C-3, respectively Activated carbon (AC) and pure MgO

(prepared using the same procedure to composite) were used for

comparison purposes

2.3 Characterization

X-ray diffraction patterns (XRD) of as-synthesized samples

were recorded using an X-ray spectrometer (Ultima IV) using Cu

Morphology and size of the samples were observed by

transmis-sion electron microcopy (TEM, JEM-2010), scanning electron

microscopy (SEM, Hitachi 4700), and energy-dispersive X-ray

spec-troscopy (EDS, JSM-7100F) Textural properties were measured by

instru-ment (Autosorb iQ, version 3.0 analyzer) This instruinstru-ment was also

used to measure the adsorption isotherms of water vapor for the

MgO and MgO/C samples Fourier transform infrared spectroscopy

(FT-IR, VERTEX 70) was recorded in the wave range between 4000

2.4 Dynamic breakthrough experiments for removal of DMMP and

2-CEES

A schematic diagram of the dynamic breakthrough system is

glass column), packed with as-prepared material, was used to

mea-sure the sorption capacities of the synthesized materials for DMMP

and 2-CEES at ambient DMMP and 2-CEES vapors were generated

experimental vapor concentration and prevent vapor condensa-tion Gas flows were controlled by two mass flow controllers (MFCs) To confirm the homogeneous phase, the generated feed gas was passed through a mixing tank Before the breakthrough experiments, the concentration of feed gas was confirmed by an on-line gas chromatograph with a flame photometric detector (GC; Agilent 6890N) through a by-pass line of the sorption column And the on-line GC was used to monitor the concentrations of DMMP and 2-CEES at the outlet of the sorption column Addition-ally, the components of outlet gas samples were analyzed by a GC-MS (Agilent 7890A-5977A)

In each experiment, sorbent (50 mg) was packed into a water-jacket column reactor (inner diameter: 7 mm; length: 175 mm) Then, glass beads and glass wool were put into both ends of the column The column temperature was controlled by a water

into the sorption column The feed concentrations of DMMP and 2-CEES was calibrated before each breakthrough experiment and controlled within the range of ±3%

When sorbent particles are tested in a breakthrough experi-ment, differences in the pressure drop and bed porosity can result

in experimental deviations Therefore, it is important to evaluate materials using the same packing conditions as used in the

sample particles was packed into the column for each experiment The packing length occupied by the sorbent particles was also kept equal in every experiment to maintain the same packing density Because the reactor was a glass column, the packing length could

be monitored during experiments In addition, a low fixed flow rate was used to minimize errors caused by changes in the packing density (length) in breakthrough experiments even though the experiments took longer

Since the boiling point between 2-CEES and DMMP was differ-ent, it was not easy to carry out the breakthrough experiments under the same flow rate and concentration In the study, the breakthrough experiments were carried out by using a similar con-centration condition for each sorbate: DMMP (concon-centration;

RTD: Resistance Temperature Detector PG: Pressure Gauge

pressure during the breakthrough experiments was measured by

two electrical pressure gauges The pressure drop in all

experi-ments was 0.25 psi due to packing sorbent particles Breakthrough

saturation took a long time in each experiment due to dense

packing, the slow flow rate, and low concentrations of DMMP

and 2-CEES

The sorption capacities of DMMP and 2-CEES were calculated

using the following equation:

Sorption capacityðmg=gÞ ¼F t0 :5 Co

The sorption equilibrium rate as well as sorbate breakthrough

can be used to quantify the dynamic sorption properties of

to fit the experimental breakthrough curves:

ln C

Co C

¼ KYN t  t0 :5 KYN ð2Þ

sorbents

3 Results and discussion

3.1 Characterization

The wide-angle XRD diffraction patterns of MgO, AC, and MgO/C

the (2 0 0), (2 2 0), and (2 2 2) plans of MgO, respectively This

indi-cated that MgO had a cubic structure with a crystal lattice

param-eter of a = 4.21 Å (JCPDS No 75-0447) In addition, the relatively

broad and low intensity of diffraction peaks indicated a small

crys-tallite size of 4.1 nm approximated by using the Scherrer equation

crystal phase of commercial activated carbon (JCPDS No 82-1691)

XRD diffraction peaks as MgO, but no characteristic peaks

corre-sponding to carbon were detected, even though the carbon content

of the composites was as high as 17.86 wt% (MgO/C-3) We

deduced that the carbon in the composites was amorphous The

intensity of diffraction peaks increased slightly with carbon

con-tent in the MgO/C composites The crystallite size of the

compos-ites (2.8–3.1 nm) was smaller than that of MgO, as shown in

Table 1

The chemical composition of the composites was determined

(b), full-scale XPS spectra of MgO and MgO/C composites exhibited

very clear MgO features The photoelectron peaks at 49.8, 92.2, and

1304 eV corresponded to Mg 2p, 2s, and 1s, respectively, and the O

Activated carbon was obviously observed by the intense

photoelec-tron peak C 1s at 284.6 eV The peak was also observed with a

lower intensity than carbons in all the composites, indicating the

successful incorporation of carbon into MgO The carbon content

in the composites estimated by the C 1s peak area was 6.39,

10.68, and 17.86 wt% for MgO/C-1, MgO/C-2, and MgO/C-3,

increased with an increase in the amount of glucose used during the synthesis

sur-face area, pore volume, and the average pore size diameter of

MgO was classified as a type IV based on the IUPAC system In addition, MgO had a sharp type H3 hysteresis loop containing a steep region associated with closure of the hysteresis loop at the

with non-rigid aggregated particles forming slit-shape pores The isotherm of AC was classified as type I, implying a microporous

composites were type IV with a type H4 hysteresis loop at

of composites were smaller than that of MgO and showed a shift of the closure of hysteresis loop to a lower value The results associ-ated with narrow slit pores, including a pore in the micropore region Pore size distributions of composites were smaller than that of MgO and were comparatively narrower with an average pore diameter in the range of 3.43–3.84 nm

When glucose and water were added into magnesium methox-ide solution, the homogenous gels were formed because of the interaction of glucose with Mg(OH) polymer-like gels by hydrogen

2-Theta (degree)

110

100 002

AC

MgO MgO/C-1 MgO/C-2

MgO/C-3 222

220 200

(a)

0.0 5.0x10 5

1.0x10 6

1.5x10 6

2.0x106 2.5x106

Binding Energy (eV)

(b)

AC MgO

MgO/C-1 MgO/C-2 MgO/C-3

Fig 2 (a) XRD patterns and (b) XPS spectra of MgO, AC, MgO/C-1, MgO/C-2, and MgO/C-3 samples.

bonding or MgAO bonding[35,36] (Fig S1) In addition, glucose

surface of MgO during calcination The glucose on the surface of

MgO resulted in improving the surface area and porosity of

MgO/C composites exhibited a high surface area and pore volume

However, at high molar ratio of magnesium methoxide to glucose

(1:0.09 and 1:0.18), the precipitation was very significant and

the gels became less homogeneous The extra carbon in the

com-posites led to a decrease in BET surface area as follows: MgO/C-1

were smaller than those of AC and HY-MgO/C-1 (MgO/C-1 before calcination) BJH mesopore volumes of composites (1.14–1.36 cc/g) were smaller than that of MgO, but the SF micropore volumes (0.434–0.451 cc/g) were higher than that of MgO As mentioned previously, the composites had well-developed mesopores and some micropores Even though the micropore volumes of the com-posites were similar to that of AC, the average pore diameter of the composites was larger than that of AC In addition, the surface areas of the composites were much smaller than that of AC FT-IR spectra of MgO and MgO/C-1 as a representative

[38,39] And these bands could be also ascribed to vibrational

bonding In addition, it was reported that the synthesized metal oxides using metal alkoxides by an aerogel method contained a

once again implied that isolated carbon in amorphous phase was formed after thermal decomposition of glucose in the

TEM, SEM, and SEM/EDS images of MgO and MgO/C-1 samples

100–150 nm in size with a rough surface that formed due to aggre-gation of many small particles with about 4 nm in size, as shown in

Fig 4(b), consistent with the XRD results presented inTable 1 The

that amorphous carbon particles were well dispersed on the sur-face of the aggregated MgO particles and the particle were

10 nm in size In addition, the morphology of carbon could be seen in the TEM images of MgO/C-1 sample after etching using

5 nm were evident, but the particle sizes were similar to those of MgO/C-1 before etching process To further confirm the composi-tion and structure of the composites, SEM/EDS measurements were conducted on the MgO/C-1 sample EDS spectrum results

Table 1

Textural properties of MgO, MgO/C composites, and activated carbon.

area (m 2

/g)

BJH mesopore volume (cc/g)

SF microspore volume (cc/g)

Total pore volume (cc/g)

Average pore diameter (nm)

Crystallite size a

(nm)

wt% of carbon b

a

Estimated by (2 0 0) XRD diffraction peak of MgO.

b Result from XPS analysis.

0

200

400

600

800

1000

1200

Relative pressure (P/Po)

MgO/C-1 MgO/C-2 MgO/C-3 MgO AC

(a)

0 5 10 15 20 25 30 35 40

0

1

2

3

4

5

6

Pore diameter (nm)

(b)

MgO/C-1 MgO

873

1458 1637

3440

428

592

Fig 3 (a) N 2 isotherm curves (inset: pore size distribution) of AC, MgO, MgO/C-1,

MgO/C-2, and MgO/C-3 samples, (b) FTIR spectra of MgO and MgO/C-1 samples.

of the concentrated regions of specific elements; dispersion of

car-bon on the surface of MgO particles was the dominant finding This

glu-cose on the surface of MgO during calcination Based on these

results, it was concluded that MgO was well coated by carbon,

resulting in core-shell structures However, at high molar ratios

of magnesium methoxide to glucose, the carbon coated MgO

parti-cles could be mixed with extra carbons, which came from the

addi-tional amount of carbon source (glucose), as described above

3.2 Removal of DMMP

Sorption behavior of as-prepared samples was determined by

breakthrough curves; the curves showed the DMMP concentration

at the outlet of the sorbent column as a function of time

Break-through curves for DMMP on MgO, activated carbon, and MgO/C

The breakthrough time of DMMP on MgO (199 min) was longer

than that of the MgO/C-2 and MgO/C-3 composites (164 and

129 min, respectively) However, the saturation time of MgO

(514 min) was shorter than that of MgO/C-2 and MgO/C-3

compos-ites (589 and 534 min, respectively) Therefore, the breakthrough

shapes of MgO/C-2 and MgO/C-3 were a little wider than that of

MgO In contrast, although there was not a significant difference

in breakthrough time between MgO/C-1 and MgO, the saturation

time of MgO/C-1 was much longer than that of MgO, MgO/C-2,

DMMP on activated carbon was steepest, while the breakthrough

and saturation times were shortest for the other samples even

though the activated carbon had a much higher surface area than

smal-ler than the pore size diameters of MgO, MgO/C composites, and activated carbon, the DMMP molecules can penetrate into the pores and sorb on the surface of the samples And the sorption of

pore volume are the important factors that significantly contribute

to the sorption capacity of the sorbents In regard to the surface area and total pore volume, the sorption capacity of DMMP

(43.7 mg/g) > MgO (42.2 mg/g) > MgO/C-3 (34.5 mg/g) > activated carbon (30.4 mg/g) This clearly showed that sorption capacity decreased with an increase in carbon content, and that it could

be considerably improved by coating MgO with a small amount

of carbon (MgO/C-1)

The breakthrough curve shape as well as slope can be affected

acti-vated carbon and as-prepared composites had different break-through curve shapes from one other This can be explained by the different affinities and concentration propagations of DMMP

on each sorbent material It was reported that the molecular diam-eters of DMMP and 2-CEES were 0.57 and 0.69 nm, respectively

[43,46] Considering the pore sizes of AC and MgO inTable 1, the sorption affinity was a more important factor than mass transfer resistance The breakthrough curve slopes of all the composites were wider than those of MgO and AC, and became steeper with

an increase in carbon content This implied that the sorption affin-ity of DMMP on all the composites was weaker than those of MgO Fig 4 SEM images of (a) MgO and (c) MgO/C-1; TEM images (b) MgO, (d) MgO/C-1, and (e) MgO/C-1 after etching MgO; (f) EDS spectrum of MgO/C-1; (g) and (h) elemental maps of carbon and magnesium of MgO/C-1, respectively; (i) SEM/EDS image of MgO/C-1.

and AC because the pore sizes of the composites were still larger

than that of AC Since the difference in the pore sizes of all the

com-posites was small, the increased sorption affinity of the comcom-posites

stemmed from increased carbon content and the sorption affinity

approached that of AC even though the sorption capacity

decreased In addition, the Yoon–Nelson model could predict the

Together, the results indicated that MgO/C composites prepared using an aerogel method had improved BET surface area and micropore volume than MgO As carbon content in the composites increased, the BET surface area and BJH mesopore volume decreased Sorption capacity with an increase in carbon content was similar or smaller than that of MgO The contribution of improved surface area in the MgO/C composites to sorption capac-ity was limited MgO/C-1 composite had the longest breakthrough and saturation times as well as highest sorption capacity among the as-prepared composites MgO/C-1 was therefore selected for further evaluation under humid conditions

As mentioned previously, the fact that an as-prepared compos-ite has a high removal capacity for toxic chemicals under dry con-ditions compared to MgO and AC is not sufficient for its practical application, because toxic chemicals normally exist in humid con-ditions To evaluate the effect of water vapor on the sorption of DMMP, the removal efficiencies of MgO, MgO/C-1, and AC were

humidity of 30% (30% RH) The other conditions for these break-through experiments under humid conditions were the same as those used for experiments performed under dry conditions Breakthrough curves of DMMP on MgO, MgO/C-1, and activated

shapes of all test sorbents were similar to those obtained under dry conditions, but a reduction in the sorption capacity of MgO was clearly observed

The breakthrough and saturation times of MgO under humid conditions (44 and 309 min, respectively) were much shorter than those under dry conditions As a result, sorption capacity under humid conditions decreased significantly to 23.3 mg/g, corre-sponding to 56% of sorption capacity under dry conditions (Table 2) We ascribed this to the sorption of water vapor on active sites of MgO particles The effect of humidity on sorption was not

break-through and saturation times decreased slightly The relative decrease in sorption capacity (24.6 mg/g) was much smaller than

of AC became similar to that of MgO, but the breakthrough time was longer due to strong adsorption affinity under humid conditions

As expected from the AC results, the sorption of DMMP by carbon-coated MgO (MgO/C-1) was not significantly affected by the presence of water vapor The breakthrough and saturation times of DMMP for the MgO/C-1 sample were 179 and 744 min, respectively, showing very little decrease in comparison with dry conditions Correspondingly, the sorption capacity under humid conditions (61.5 mg/g) was 91% of that under dry conditions and approximately 2.5-fold higher than those of activated carbon and

0.0

0.2

0.4

0.6

0.8

1.0

(a)

MgO/C-1 MgO/C-2 MgO/C-3 MgO Activated carbon Model

Time (min)

0.0

0.2

0.4

0.6

0.8

1.0

MgO/C-1 MgO Activated Carbon Model

Time (min)

(b)

Fig 5 Comparison of breakthrough curves for DMMP sorption on sorption column

packed by (a) MgO, MgO/C-1, MgO/C-2, MgO/C-3 or activated carbon in dry

condition, and (b) MgO, MgO/C-1 or activated carbon in humid condition (30 %RH).

Table 2

Breakthrough and saturation times, and sorption capacity of DMMP in dry and humid condition (30 %RH).

Sample Condition t b (min) t s (min) t 0.5 (fit) t 0.5 (exp) K YN (min1) R 2

Sorption capacity (mg/g)

t b and t s are breakthrough and saturation times, respectively.

under humid conditions was also similar to that under dry

showed different behaviors for sorption of DMMP according to

the presence of water vapor The breakthrough and saturation

times as well as sorption capacity of MgO were significantly lower

than those obtained under dry conditions, while these values did

not differ much for the MgO/C-1 composite Therefore, a

carbon-coated MgO composite was successfully developed to remove

DMMP efficiently under both dry and humid conditions

To investigate the effect of water vapor on the crystalline

struc-ture of MgO and the MgO/C-1 composite, both sorbents were

ana-lyzed by XRD after DMMP sorption under humid conditions The

No significant changes in the peak intensity, full width at half

max-imum (FWHM), or peak position of (2 0 0), (2 2 0), and (2 2 2) crystal

plans of either sorbent was observed after DMMP sorption under

humid conditions MgO underwent substantial conversion to Mg

peaks This was due to the reaction of MgO with water In contrast,

MgO/C-1 composite This implied that it was difficult for MgO to

react with water molecules in the composite

The result is also supported by the water adsorption isotherms

water vapor was much stronger and larger than that of the MgO/C-1 composite This implied that the hydrophobicity of the carbon shell effectively protected the MgO crystals from water vapor

3.3 Removal of 2-CEES Identical breakthrough experiments as performed with DMMP

22.5 mL/min The breakthrough curves of 2-CEES on the MgO and

(a) Each sorbent had two breakthrough curves: a 2-CEES curve and a reacted product curve The composition of outlet gas from the breakthrough column for 2-CEES sorption was analyzed by a

detected at 6.7 and 12.2 min of retention times, respectively, as

(Fig S4) Breakthrough times, saturation times, and sorption

(d)

(c)

(b) (a)

222 220

200

Mg(OH)

2

Mg(OH)

2

2-Theta (degree) Fig 6 XRD patterns before and after DMMP sorption in humid condition,

respectively: (a) and (b) for MgO, and (c) and (d) for MgO/C-1.

0

100

200

300

400

500

600

700

800

900

Relative pressure (P/Po)

MgO/C-1

MgO

0.0 0.2 0.4 0.6 0.8

1.0

(a)

Time (min)

2-CEES/MgO product/MgO 2-CEES/MgO/C-1 product/MgO/C-1 model

0.0 0.2 0.4 0.6 0.8

1.0

(b)

Time (min)

2-CEES/MgO 2-CEES/MgO/C-1 product/MgO/C-1 model

Fig 8 Comparison of breakthrough curves for 2-CEES sorption on sorption column packed by (a) MgO and MgO/C-1 in dry condition, and (b) MgO and MgO/C-1 in

The breakthrough curve shapes of 2-CEES on MgO and the MgO/

and sorption capacity of 2-CEES on MgO was 202 min, 392 min,

and 33.7 mg/g, respectively The corresponding values for the

MgO/C-1 composite (222 min, 397 min, and 35.3 mg/g) were

to the results obtained for DMMP, the sorption rate constant was

faster, but the saturated sorption amount was smaller

2-CEES on MgO was greater than of the MgO/C-1 composite The

reacted product appeared before 2-CEES breakthrough in MgO,

but the reacted product took much longer to appear for the MgO/

C-1 sample due to penetration of 2-CEES and product molecules

into the carbon-coated shells of the composite particles As

remained on the surface of MgO and that these groups played a

could be explained by the following equation:

ðOHÞAMgOA þ CH2ClACHASACH2ACH3

! ClAMgOA þ CH2@CHASACH2ACH3þ H2O ð3Þ

The sorption and reaction of 2-CEES on MgO and MgO/C-1

(b) The breakthrough shape of 2-CEES on MgO was wider with a

sorption affinity of 2-CEES for MgO was significantly lower under

humid conditions than dry conditions The breakthrough and

satu-ration times of 2-CEES on MgO under humid conditions also

decreased steeply to 42 and 297 min, respectively As a result,

the sorption capacity was 22.2 mg/g and the change in sorption

behavior under humid conditions was the same as that obtained

product was observed owing to sorption of water molecules on

cat-alytic active sites of MgO

In contrast, the breakthrough shape of the MgO/C-1 composite

than that obtained under dry conditions The breakthrough and

saturation times of the MgO/C-1 composite were 152 and

324 min, respectively, representing a smaller decrease than seen

for MgO The sorption capacity of the composite under humid

con-ditions was about 86% of that under dry concon-ditions, and was much

MgO/C-1 composite, vinyl ethyl sulfide from the reaction of 2-CEES

with MgO was still detected, although levels of this product

con-centration were lower than that obtained under dry conditions

High sorption capacity and reactivity of 2-CEES on the MgO/C-1

composite under humid conditions confirmed again that the

car-bon in the MgO/C composite worked as a hydrophobic shell and

protected MgO from water sorption

4 Conclusions

Carbon-coated MgO composites were prepared via an aerogel

route with glucose as a carbon precursor to efficiently remove

2-CEES and DMMP MgO/C composites had higher surface areas and microspore volumes and lower mesopore volumes and crystal-lite sizes than MgO As the glucose amount in the synthesis step increased, the carbon content of the MgO/C composites increased and the surface area and mesopore volume decreased because too much amount of glucose addition led to forming carbon out

of the surface of MgO The MgO/C composite with 6.39 wt% carbon showed the highest sorption capacity for DMMP (67.8 mg/g at

as-synthesized composites under dry conditions The sorption capac-ities of MgO and the MgO/C composites were higher than that of

AC, even though the surface area of AC was the highest In addition, the sorption capacity of MgO/C composites decreased with an increase in carbon content MgO/C composites with more than

10 wt% carbon had lower sorption capacity than MgO, even though their surface areas were larger than that of MgO This implied that the contribution of the surface area to sorption capacity was lim-ited, but the sorption affinity played an important role in deter-mining sorption capacity

Under humid conditions, the sorption capacities of DMMP and 2-CEES on MgO decreased significantly Furthermore, MgO lost reactivity toward 2-CEES due to the sorption of water molecule

on catalytic active sites In contrast, carbon content of the compos-ite allowed effective protection of sorption and reaction of DMMP and 2-CEES from water vapor The sorption capacities of DMMP and 2-CEES on the MgO/C-1 sample under humid conditions were 61.5 mg/g and 30.4 mg/g, about 91% and 86% of those under dry conditions, respectively The carbon shell protected the sorbent composite from water vapor Because aerogel MgO with higher mesopore volume than AC can have a higher sorption affinity and capacity for CWA molecules than AC, a carbon thin layer coat-ing of MgO is the most promiscoat-ing way to produce the sorbents with higher sorption capacity and reactivity than AC in humid condition Acknowledgements

We would like to acknowledge the financial support from the R&D Convergence Program of MSIP (Ministry of Science, ICT and Future Planning) and NST (National Research Council of Science

& Technology) of Republic of Korea (CRC-14-1-KRICT)

Appendix A Supplementary material Supplementary data associated with this article can be found, in

References [1] E Barea, C Montoro, J.A.R Navarro, Toxic gas removal–metal–organic frameworks for the capture and degradation of toxic gases and vapours, Chem Soc Rev 43 (2014) 5419–5430

[2] Y Safa, H.N Bhatti, Adsorptive removal of direct textile dyes by low cost agricultural waste: application of factorial design analysis, Chem Eng J 167 (2011) 35–41

[3] Y.C Yang, J.A Baker, J.R Ward, Decontamination of chemical warfare agents,

Table 3

Breakthrough and saturation times, and sorption capacity of 2-CEES in dry and humid condition (39 %RH).

Sample Condition t b (min) t s (min) t 0.5 (fit) t 0.5 (exp) K YN (min1) R 2

Sorption capacity (mg/g)

t b and t s are breakthrough and saturation times, respectively.

t 0.5 (fit) and t 0.5 (exp) are the times for 50% sorbate breakthrough obtained from fitting and interpolation for experimental data, respectively.

[4] N.A Khan, Z Hasan, S.H Jhung, Adsorptive removal of hazardous materials

using metal-organic frameworks (MOFs): a review, J Hazard Mater 244–245

(2013) 444–456

[5] A Kleinhammes, G.W Wagner, H Kulkarni, Y Jia, Q Zhang, L.-C Qin, Y Wu,

Decontamination of 2-chloroethyl ethylsulfide using titanate nanoscrolls,

Chem Phys Lett 411 (2005) 81–85

[6] R.M Narske, K.J Klabunde, S Fultz, Solvent effects on the heterogeneous

adsorption and reactions of (2-chloroethyl)ethyl sulfide on nanocrystalline

magnesium oxide, Langmuir 18 (2002) 4819–4825

[7] M.E Martin, R.M Narske, K.J Klabunde, Mesoporous metal oxides formed by

aggregation of nanocrystals Behavior of aluminum oxide and mixtures

with magnesium oxide in destructive adsorption of the chemical warfare

surrogate 2-chloroethylethyl sulfide, Microporous Mesoporous Mater 83

(2005) 47–50

[8] G.W Wagner, O.B Koper, E Lucas, S Decker, K.J Klabunde, Reactions of VX,

GD, and HD with nanosize CaO: autocatalytic dehydrohalogenation of HD, J.

Phys Chem B 104 (2000) 5118–5123

[9] K.J Klabunde, J Stark, O Koper, C Mohs, D.G Park, S Decker, Y Jiang, I.

Lagadic, D Zhang, Nanocrystals as stoichiometric reagents with unique surface

chemistry, J Phys Chem 100 (1996) 12142–12153

[10] B.W.L Jang, J.J Spivey, Catalytic hydrodesulfurization and hydrodechlorination

of chloroethyl ethyl sulfide, Catal Today 55 (2000) 3–10

[11] A.-T Vu, S Jiang, K Ho, J.B Lee, C.-H Lee, Mesoporous magnesium oxide and

its composites: preparation, characterization, and removal of 2-chloroethyl

ethyl sulfide, Chem Eng J 269 (2015) 82–93

[12] A.-T Vu, S Jiang, Y.-H Kim, C.-H Lee, Controlling the physical properties of

magnesium oxide using a calcination method in aerogel synthesis: its

application to enhanced sorption of a sulfur compound, Ind Eng Chem Res.

53 (2014) 13228–13235

[13] Y.-H Kim, V.A Tuan, M.-K Park, C.-H Lee, Sulfur removal from municipal gas

using magnesium oxides and a magnesium oxide/silicon dioxide composite,

Microporous Mesoporous Mater 197 (2014) 299–307

[14] Y Ding, G Zhang, H Wu, B Hai, L Wang, Y Qian, Nanoscale magnesium

hydroxide and magnesium oxide powders: control over size, shape, and

structure via hydrothermal synthesis, Chem Mater 13 (2001) 435–440

[15] G Moussavi, M Mahmoudi, Degradation and biodegradability improvement of

the reactive red 198 azo dye using catalytic ozonation with MgO nanocrystals,

Chem Eng J 152 (2009) 1–7

[16] X Liu, N Bi, C Feng, S.W Or, Y Sun, C Jin, W Li, F Xiao, Onion-like carbon

coated CuO nanocapsules: a highly reversible anode material for lithium ion

batteries, J Alloys Compd 587 (2014) 1–5

[17] A.F Bedilo, M.J Sigel, O.B Koper, M.S Melgunov, K.J Klabunde, Synthesis

3599–3604

[18] B Özkal, W Jiang, O Yamamoto, K Fuda, Z.-E Nakagawa, Preparation and

characterization of carbon-coated ZnO and CaO powders by pyrolysis of PVA, J.

Mater Sci 42 (2007) 983–988

[19] H Bae, T Ahmad, I Rhee, Y Chang, S.-U Jin, S Hong, Carbon-coated iron oxide

nanoparticles as contrast agents in magnetic resonance imaging, Nanoscale

Res Lett 7 (2012) 1–5

[20] W Lin, H Cheng, J Ming, Y Yu, F Zhao, Deactivation of Ni/TiO 2 catalyst in the

hydrogenation of nitrobenzene in water and improvement in its stability by

coating a layer of hydrophobic carbon, J Catal 291 (2012) 149–154

[21] M.S Mel’gunov, E.A Mel’gunova, V.I Zaikovskii, V.B Fenelonov, A.F Bedilo, K.J.

nanocrystalline MgO, Langmuir 19 (2003) 10426–10433

[22] Q Zhou, J.-W Yang, Y.-Z Wang, Y.-H Wu, D.-Z Wang, Preparation of

nano-MgO/carbon composites from sucrose-assisted synthesis for highly efficient

dehydrochlorination process, Mater Lett 62 (2008) 1887–1889

[23] L She, J Li, Y Wan, X Yao, B Tu, D Zhao, Synthesis of ordered mesoporous

MgO/carbon composites by a one-pot assembly of amphiphilic triblock

copolymers, J Mater Chem 21 (2011) 795–800

[24] S.C Stout, S.C Larsen, V.H Grassian, Adsorption, desorption and thermal

oxidation of 2-CEES on nanocrystalline zeolites, Microporous Mesoporous

Mater 100 (2007) 77–86

[25] L.L Chao, S Kriger, S Buckley, P Ng, S.G Mueller, Effects of low-level sarin and

cyclosarin exposure on hippocampal subfields in Gulf War Veterans,

NeuroToxicology 44 (2014) 263–269

[26] J Quenneville, R.S Taylor, A.C.T van Duin, Reactive molecular dynamics

studies of DMMP adsorption and reactivity on amorphous silica surfaces, J.

Phys Chem C 114 (2010) 18894–18902

[27] C.W Kanyi, D.C Doetschman, J.T Schulte, Nucleophilic chemistry of X-type Faujasite zeolites with 2-chloroethyl ethyl sulfide (CEES), a simulant of common mustard gas, Microporous Mesoporous Mater 124 (2009) 232–235 [28] J.A Arcibar-Orozco, T.J Bandosz, Visible light enhanced removal of a sulfur mustard gas surrogate from a vapor phase on novel hydrous ferric oxide/graphite oxide composites, J Mater Chem A 3 (2015) 220–231 [29] A.-T Vu, Y Park, P.R Jeon, C.-H Lee, Mesoporous MgO sorbent promoted with KNO 3 for CO 2 capture at intermediate temperatures, Chem Eng J 258 (2014) 254–264

[30] S.-H Lim, E.-J Woo, H Lee, C.-H Lee, Synthesis of magnetite–mesoporous silica composites as adsorbents for desulfurization from natural gas, Appl Catal B 85 (2008) 71–76

[31] J.-M Kwon, J.-H Moon, Y.-S Bae, D.-G Lee, H.-C Sohn, C.-H Lee, Adsorptive desulfurization and denitrogenation of refinery fuels using mesoporous silica adsorbents, ChemSusChem 1 (2008) 307–309

[32] A Koriakin, Y.-H Kim, C.-H Lee, Adsorptive desulfurization of natural gas using lithium-modified mesoporous silica, Ind Eng Chem Res 51 (2012) 14489–14495

[33] Y.H Yoon, J.H Nelson, Application of gas adsorption kinetics – II A theoretical model for respirator cartridge service life and its practical applications, Am Indian Hygiene Assoc J 45 (1984) 517–524

[34] M.-H Oh, K.-W Kim, S Choi, R.K Singh, Preparation of monodispersed magnesia coated silica particles through a surface-induced precipitation method, Powder Technol 204 (2010) 154–158

[35] J Liu, T.-H Bae, O Esekhile, S Nair, C Jones, W Koros, Formation of Mg(OH) 2 nanowhiskers on LTA zeolite surfaces using a sol–gel method, J Sol–Gel Sci Technol 60 (2011) 189–197

[36] C Trionfetti, I.V Babich, K Seshan, L Lefferts, Formation of high surface area Li/MgO—efficient catalyst for the oxidative dehydrogenation/cracking of propane, Appl Catal A 310 (2006) 105–113

[37] J.W Lee, L.C Thomas, S.J Schmidt, Can the thermodynamic melting temperature of sucrose, glucose, and fructose be measured using rapid-scanning differential rapid-scanning calorimetry (DSC), J Agric Food Chem 59 (2011) 3306–3310

[38] Y Diao, W.P Walawender, C.M Sorensen, K.J Klabunde, T Ricker, Hydrolysis

of magnesium methoxide Effects of toluene on gel structure and gel chemistry, Chem Mater 14 (2002) 362–368

[39] J Zhou, S Yang, J Yu, Facile fabrication of mesoporous MgO microspheres and their enhanced adsorption performance for phosphate from aqueous solutions, Colloids Surf A 379 (2011) 102–108

[40] R Trujillano, E Rico, M.A Vicente, M Herrero, V Rives, Microwave radiation and mechanical grinding as new ways for preparation of saponite-like materials, Appl Clay Sci 48 (2010) 32–38

[41] D Hassouna, C Hedia, T Fathi, Mg(OH) 2 nanorods synthesized by A facile hydrothermal method in the presence of CTAB, Nano-Micro Lett 3 (2011) 153–159

[42] X Wu, H Cao, G Yin, J Yin, Y Lu, B Li, MgCO 3 3H 2 O and MgO complex nanostructures: controllable biomimetic fabrication and physical chemical properties, Phys Chem Chem Phys 13 (2011) 5047–5052

[43] X Lu, V Nguyen, X Zeng, B.J Elliott, D.L Gin, Selective rejection of a water-soluble nerve agent stimulant using a nanoporous lyotropic liquid crystal–butyl rubber vapor barrier material: evidence for a molecular size-discrimination mechanism, J Membr Sci 318 (2008) 397–404

[44] Y.X Li, O Koper, M Atteya, K.J Klabunde, Adsorption and decomposition of organophosphorus compounds on nanoscale metal oxide particles In situ

GC-MS studies of pulsed microreactions over magnesium oxide, Chem Mater 4 (1992) 323–330

[45] A Ryzhikov, V Hulea, D Tichit, C Leroi, D Anglerot, B Coq, P Trens, Methyl mercaptan and carbonyl sulfide traces removal through adsorption and catalysis on zeolites and layered double hydroxides, Appl Catal A 397 (2011) 218–224

[46] A Roy, A.K Srivastava, B Singh, D Shah, T.H Mahato, A Srivastava, Kinetics of degradation of sulfur mustard and sarin simulants on HKUST-1 metal organic framework, Dalton Trans 41 (2012) 12346–12348

[47] D.B Mawhinney, J.A Rossin, K Gerhart, J.T Yates, Adsorption and reaction

of 2-chloroethylethyl sulfide with Al 2 O 3 surfaces, Langmuir 15 (1999) 4789–4795

[48] D.A Giannakoudakis, J.A Arcibar-Orozco, T.J Bandosz, Key role of terminal hydroxyl groups and visible light in the reactive adsorption/catalytic conversion of mustard gas surrogate on zinc (hydr)oxides, Appl Catal B 174–175 (2015) 96–104