Adsorptive Removal of Phosphate from Water by Ammonia ... - J-Stage

0 downloads 0 Views 2MB Size Report
PAN increase from 7 m2/g to 1000 m2/g or more, becoming activated carbon fiber (ACF), whereas helium treatment resulted in the maximum surface area of ...

J. Fiber Sci. Technol., 72(11), 237-243 (2016) doi 10.2115/fiberst.2016-0035 © 2016 The Society of Fiber Science and Technology, Japan

【Transaction】

Adsorptive Removal of Phosphate from Water by Ammonia Gas Activated Polyacrylonitrile Fiber Yuki Yamazaki*1, Tanita Gettongsong*2, Masahiro Mikawa*3, Yoshimasa Amano*3,4, and Motoi Machida*3,4,# *1

Faculty of Engineering, Chiba University, 1-33, Yayoi-cho, Inage-ku, Chiba 263-8522, Japan *2 Department of Marine Science, Chulalongkorn University, Bangkok 10330, Thailand *3 Graduate School of Engineering, Chiba University, 1-33, Yayoi-cho, Inage-ku, Chiba 263-8522, Japan *4 Safety and Health Organization, Chiba University, 1-33, Yayoi-cho, Inage-ku, Chiba 263-8522, Japan

Abstract: Commercially available oxidized black colored polyacrylonitrile (PAN) fiber was heat treated above 900 ̊C under helium or ammonia gas flow. Ammonia gas treatment made specific surface area of the oxidized PAN increase from 7 m2/g to 1000 m2/g or more, becoming activated carbon fiber (ACF), whereas helium treatment resulted in the maximum surface area of only 60 m2/g. Adsorption of phosphate improved from 0.022 mmol/g for oxidized PAN to 0.057 mmol/g for the helium treatments at pH range of 5-6. In case of ammonia treatment, the adsorption amount of phosphate attained 0.056 to 0.17 mmol/g in maximum, depending on the treating temperature ranging from 925 to 975 ̊C. The adsorption sites for negatively charged phosphate could be estimated to be positively charged quaternary nitrogen species generated on PAN fiber surface by the heat treatment. Langmuir adsorption affinity of PAN-ACF was derived from isotherms to be 0.6 L/mmol or more; moderate adsorption affinity was exhibited. (Received May 11, 2016; Accepted September 14, 2016)

1. Introduction

such as silica [5] and ferric sludge [6], respectively. However, there have been much less studies of

Phosphate-phosphorus and ammonium/nitrite/

phosphate removal by activated carbons [7] including

nitrate-nitrogen are essential for increasing food

in our previous study [8]. Phosphates in aqueous

production. However, excess amount of them has

solution are present as anionic species, thereby

been discharged to ecosystem as fertilizer and

positively charged surface has an advantage to

through livestock manure leading to red tide in closed

attract

and semi-closed water and sea area [1] and also

functional groups are oxygen such as carbonyl and

resulting in overgrowth of algae in lake [2]. Although

quinone as weak basic sites and nitrogen such as

nitrogen containing fertilizer come from industrial

quaternary and aliphatic amine as strong basic sites.

scale

In

ammonia

synthesis,

phosphorous

can

be

this

anionic

study,

substances.

introduction

Positively

of

basic

charged

nitrogen

obtained only from natural phosphate rock, which is

functionalities onto carbon surface was attempted

unevenly distributed to the limited area of Morocco in

with high temperature heat and ammonia treatments

the world [3]. Development of recovery technique of

of oxidized polyacrylonitrile (PAN) fiber which had

phosphate seems to be required in near future.

nitrogen and oxygen in the polymer structure. Using

Formation of hydroxyapatite from phosphate in

the

water is one of the promising techniques to recover

phosphate from aqueous solution was carried out to

phosphate. Adsorptive removal of phosphate is also a

inspect what kinds of species were estimated to be

candidate technique. A lot of adsorbents have been

generated and/or transformed on the PAN surface.

obtained

materials,

adsorptive

removal

examined to uptake phosphate from water [4] and most of them are ceramic and metal oxide materials # corresponding authors: Tel/fax: +81 43 290 3559

E-mail: [email protected] (M. Machida)

Journal of Fiber Science and Technology (JFST), Vol.72, No. 11 (2016)

237

of

2. Materials and methods

respectively, in which OG stands for outgassing so that surface oxygen of oxidized PAN (PYR) will be

All chemicals were reagent grade, purchased

principally

removed

by

the

helium

treatment.

from Kanto Chemical Co., Inc., and used without

Likewise ammonia treated PYR samples at 925-975 ̊C

further purification. Oxidized polyacrylonitrile (PAN)

are designated as PYR 925 AG, PYR 950 AG and PYR

fiber felt was obtained from Toho Tenax Co., Ltd.

975 AG, in which AG means ammonia gas treatment.

Activated PAN fiber (FE 200) was purchased from

Additionally, commercial activated PAN fiber of FE

Toho Chemical Engineering & Construction Co., Ltd.

200 was employed as a reference material. FE 200

2.1 Oxidized PAN fiber and preparation of adsorbents

was also treated in helium flow ranging from 925 to

In order to produce carbon fiber from white

975 ̊C and ammonia gas flow at 950 ̊C as well as PYR

polyacrylonitrile (PAN) raw resin, PAN was oxidized

series. They are referred to as FE 925 OG, FE 950 OG,

at 200 ̊C or higher for 3-4 days to be stabilized for the

FE 975 OG and FE 950 AG, respectively.

first step. This step is extremely critical and special

2.2 Characterization of prepared adsorbents

attention has to be paid to prevent the fiber from

The pH of the point of zero charge (pHpzc) of the

melting and shrinking that will thoroughly change its

fiber samples was measured by the pH drift method

morphology as obtained in our previous study [9]. Liao

using NaCl solutions, for which pH was modified by

et al. also examined the air oxidation of PAN resin and

0.1 M HCl or 0.1 M NaOH [11]. Thirty milligram of

proposed the formation of ladder structure of pyridine

adsorbent was mixed with 15 mL acidic or basic

rings at this stabilized stage [10]. In this study,

solution in which pHpzc was determined when the

commercially available felt shaped oxidized black

initial solution pH without fiber sample is equal to the

PAN fiber was purchased from Toho Tenax Co., Ltd.

final pH after sufficient agitation of the mixture of

The commercial name of the oxidized PAN fiber was

solution and fiber sample to attain equilibrium state.

PYROMEX, hereafter referred to as PYR.

Elemental composition of C, H and N for the fiber

PYR was further treated in quarts tube at 925 ̊C,

samples was determined by elemental analyzer

950 ̊C and 975 ̊C under helium or ammonia gas flow.

(Perkin-Elmer, PE 2400 II). Textural properties of the

A 1.5 g portion of PYR was put into the quarts tube of

samples were obtained from N2 adsorption-desorption

25 mm inner diameter and heated up to the desired

isotherms at ­196 ̊C observed by Beckman Coulter

temperatures using horizontal electric furnace with

Surface Analyzer (SA 3100). Specific surface area

the rate of increasing temperature of 10 ̊C/min at the

(SBET) was calculated using B.E.T. method and total

flow rate of 150 mL/min for helium, intending to

pore volume (Vtotal) was obtained from adsorbed N2

effectively remove ammonia formed on PAN surface

volume at relative pressure (Ps/P0) at 0.99. Pore width

in the heat treatments, and the flow rate of 100 mL/

(Wpore) was calculated from SBET and Vtotal assuming slit

min in case of ammonia gas. For the helium flow, the

shaped structure with Wpore = 2 Vtotal/SBET, because the

final temperature was held for 10 min, whereas the

obtained carbon fiber sample could be assumed to be

quarts tube was allowed to cool down immediately on

composed of incomplete graphene sheets.

attaining the final temperature for the ammonia

2.3 Adsorption of phosphate onto carbon fiber

treatment, because the decomposition of the obtained

derived from oxidized PAN

sample was progressed in our preliminary study in the ammonia flow.

As a stock solution, potassium dihydrogen

After cooling down to room

phosphate (KH2PO4) was dissolved in pure water to

temperature, ammonia treated samples were heated

meet with 3.0 mmol/L. Thirty milligram of prepared

again to 300 ̊C under helium flow and held for 30 min

PAN fiber was dosed into the 15 mL KH2PO4 solution

to completely desorb ammonia remaining on the fiber

in Erlenmeyer flask. The flask was agitated at

samples, while the helium treated PYR was removed

100 rpm and 25 ̊C for 24 hours in order to reach a

from

room

state of adsorption equilibrium. The equilibrium

temperature. All treated samples were washed with

solution was pipetted out and diluted to desired ratio

warm water at 65-70 ̊C for 2 days or more using

with pure water and a molybdenum blue colorimetric

Soxhlet extractor until the pH of extraction water no

method with UV-Vis spectrometer (Shimadzu, UV-

longer changed and dried in oven at 110 ̊C. RYR

2550) was used for the determination of phosphate

treated in helium flow at 925 ̊C, 950 ̊C and 975 ̊C are

concentration. The adsorption amount of phosphate

named as PYR 925 OG, PYR 950 OG and PYR 975 OG,

was calculated by,

238

the

quarts

tube

after

cooling

to

Journal of Fiber Science and Technology (JFST), Vol.72, No. 11 (2016)

Q e = (C0 − Ce ) ×

was also observed for ammonia gas treatment of

v , w

(1)

FE 200 at 950 ̊C; SBET was considerably increased from 680 m2/g to 1190 m2/g. These results clearly

where Qe was equilibrium amount of phosphate on

indicated

adsorbent, C0 and Ce were the initial and the

carbonaceous materials [12] as well as steam and CO2

equilibrium concentrations of phosphate, v and m

activation. Carbonization of oxidized PAN fiber (PYR)

were volume of solution and amount of adsorbent

forming graphene sheets seems to be progressed by

dosage, respectively.

heat treatment under both the helium and the

that

ammonia

gas

could

activate

Influence of solution pH on adsorption amount of

ammonia gas flow, because carbon composition was

phosphate was also examined. Both phosphate species

increased but hydrogen was decreased. In contrast,

and surface charge of adsorbents will be significantly

since FE 200 itself was already activated PAN fiber,

changed by altering solution pH resulting in great

carbon

changes in adsorption capacity [8]. By varying the

significantly vary. The pHpzc value was also increased

initial

with

concentration

of

phosphate,

adsorption

and

hydrogen

increasing

surface

compositions area

and

did

not

progressing

isotherms were drawn as well and adsorption

carbonization, revealing that π-electron density was

parameters were calculated assuming Freundlich and

gradually increasing as graphite sheets were growing,

Langmuir isotherms. Desorption of phosphate from

because π-electrons on graphite layers could attract

the adsorbent was investigated by adding 20 mL of

protons in aqueous solution leading to the rise in pHpzc.

0.1 M HCl solution in order to inspect the possibility

For the development of porosity with rise in surface

for reuse of the adsorbent. All these adsorption

area by ammonia gas treatments, detailed mechanism

experiments mentioned above was made by the same

for the ammonia gas activation is not clear for the

procedure described at the beginning in this section.

present. 3.2 Adsorption capacity of phosphate Fig. 1 shows adsorption amount of phosphate (Qe)

3. Results and discussion

and equilibrium solution pH (pHe) for prepared 3.1 Properties of prepared samples

samples,

all

of

them

were

derived

from

In Table 1 was shown textural and surface

polyacrylonitrile resin. The adsorption amount of PYR

properties of prepared polyacrylonitrile (PAN) fibers.

was increased from 0.02 mmol/g to 0.06-0.17 mmol/g

Specific surface area (SBET) of oxidized PAN fiber

by ammonia gas treatment as displayed for PRY 925-

2

2

(PYR) rose from just 7 m /g to over 1000 m /g by

975 AG, whereas only 0-0.06 mmol/g could be attained

ammonia gas treatment having only micropore of the

for helium gas treatment as PYR 925-975 OG. The

uniform pore width of 1.1 nm, whereas only a slight

difference between AG and OG series for PYR could

increase could be achieved by outgassing. Activation

be attributed to the difference in the specific surface

Table 1

Sample

BET specific surface area (SBET ), 2

Textural and surface properties of polyacrylonitrile (PAN) fibers

Total pore volume (Vtotal),

Pore width (Wpore),

pHpzc

Elemental composition, wt-%

mL/g nm C RYR 7 5.0 59.6 PYR925OG 49 0.036 1.5 6.1 82.1 PYR950OG 12 6.0 84.7 PYR975OG 60 0.043 1.4 6.0 83.5 PYR925AG 1260 0.711 1.1 7.5 79.2 PYR950AG 1030 0.576 1.1 7.8 77.9 PYR975AG 1210 0.685 1.1 7.5 81.9 FE200 680 0.362 1.1 6.9 76.0 FE925OG 460 0.250 1.1 7.4 76.6 FE950OG 490 0.270 1.1 7.6 74.7 FE975OG 330 0.210 1.3 7.4 76.6 FE950AG 1190 0.656 1.1 8.8 71.2 * calculated by balance, Pore width; calculated from S BET and Vtotal assuming slit shaped pore m /g

H 3.6 0.4 0.5 0.4 0.6 0.5 0.4 0.6 0.5 0.7 0.7 0.7

Journal of Fiber Science and Technology (JFST), Vol.72, No. 11 (2016)

N 20.9 10.7 10.1 10.6 9.1 8.9 6.4 5.6 4.4 5.7 5.9 7.3

O* 15.9 6.8 4.7 5.5 11.1 12.7 11.3 17.8 18.5 18.9 16.8 20.8

239

positively charged carbon surface is required and at the same time negatively charged sites should be avoided to effectively attract phosphate on the carbon surface. Carboxy groups can be easily introduced onto carbon surface [13] and always negatively charged at pH above 4 resulting in no phosphate adsorption even if positively charged lactone groups are present in the same surface [8]. For the attractive sites in the carbon surface, π-electrons on graphene sheet itself can attract protons forming slightly positively

charged

surface

that

may

attract

phosphate [14]. Heteroatoms such as oxygen and nitrogen play an important role on the carbon surface Fig. 1 Adsorption amount of phosphate on each prepared adsorbent and equilibrium solution pH (pHe). Adsorbent dosage is 30 mg in 15 mL solution at the initial KH2PO4 concentration of 3.0 mmol/L.

to form positively charged sites to capture phosphate. As oxygen functional groups, lactone groups belong to positively charged Lewis acids and carbonyl groups are Lewis bases that may accommodate protons also resulting in becoming positively charged sites. As for nitrogen on the carbon surface,

area (SBET) of 12-60 m2/g for OG series but 1030-1260 m2/g

quaternary nitrogen (N-Q) probably formed inside and

for AG series. In case of OG and AG treatments of FE

peripheral of the graphene sheet is obviously

200, adsorption amount of phosphate was also

positively charged sites [15] and must attract

increased from 0.01 mmol/g for no treatment to 0.07-

negatively charged phosphate, but pyridine-N-oxide

0.16 mmol/g for FE 925-975 OG and FE 950 AG.

(N-X) may not be preferential because oxygen atom

Although the results imply that specific surface area

next to quaternary nitrogen is negatively charged in

may be one of the essential parameters for the

N-X that will generate repulsive force to phosphate

enhancement of adsorption amount of phosphate onto

anion. The other candidate for nitrogen element on

the

the

carbon surface is aliphatic amines, because their pKa

adsorption amounts are not always proportional to

polyacrylonitrile

(PAN)

carbon

fiber,

values are greater than 9.0 [16] that accommodate

specific surface areas of the PAN fibers. For example,

protons

the specific surface areas of FE 925-975 OG were

Hamoudi et al. examined the aliphatic amine grafted

smaller than that of FE 200, whereas the adsorption

to silica materials and achieved to 0.78-1.21 mmol/g of

amount of phosphate of them was much greater than

phosphate

FE 200 as clearly represented in Fig. 1. Particularly

quaternary nitrogen (N-Q) together unpreferably with

FE 975 OG exhibited 2.9 times greater adsorption

pyridine-N-oxide (N-X) could be generated on PAN

amount of phosphate than PYR 925 AG even though

fiber surface by the transformation from pyrrole (N-5)

2

forming

positively

adsorption.

In

charged

the

sites

present

[17].

study,

specific surface area of FE 975 OG (330 m /g) is 3.8

and pyridine (N-6). The N-Q sites are expected to

times smaller than PYR 925 AG (1260 m2/g) with

attract phosphate anions. From the reports of thermal

similar pore width (1.1-1.3 nm) of them. These results

treatments

indicated that surface chemical environment should

materials, the N-Q formation will be enhanced at

be considered to determine the difference in

higher

adsorption capacities of phosphate as long as some

transformations were reported for PAN fiber in

300 m2/g or higher specific surface area could be

which

obtained. In our previous study for the removal of

occurred at higher temperature than that of N-Q [15,

phosphate, adsorption capacity of 0.16 mmol/g could

21]. Based on the earlier studies in the literature, the

be achieved as well by wood-based activated carbon

increase in adsorption amount of phosphate with

outgassed in helium flow at 1000 ̊C [8]. Phosphate ions

increasing heating temperature could be attributed to

in aqueous solution are negatively charged in the

the transformation of quaternary nitrogen (N-Q) from

solution pH above 4.5, irrespective of total phosphate

N-5 and N-6 in the prepared activated PAN fiber

concentration examined in this study. Thereby

structures. Similar results were also obtained in our

240

of

nitrogen-containing

temperature the

formation

[ 18, of

Journal of Fiber Science and Technology (JFST), Vol.72, No. 11 (2016)

19,

carbonaceous 20 ] .

pyridine-N-oxide

Similar (N-X)

previous study using FE 400 activated PAN fiber in

specific surface area and total pore volume of PYR

which the maximum adsorption amount of phosphate

975 are greater than PYR 950. The difference also

could be observed at 950 ̊C outgassing [22]. Decrease

indicated that surface chemistry could play an

in adsorption amount of phosphate for PYR 975 AG

important role for the phosphate adsorption and

compared to PYR 950 AG might be caused by

around 950 ̊C might be the optimum temperature to

eliminating total N-Q species as nitrogen containing

maximize the attractive species as quaternary

gas from PAN fiber, because nitrogen content was

nitrogen (N-Q). Table 2 represented Freundlich and

decreased from 8.9 wt-% to 6.4 wt-% by increasing

Langmuir

ammonia treatment temperature from 950 ̊C to

regression data analysis. The dotted and solid lines in

975 ̊C. The estimation was supported by the fact that

Fig. 2 were drawn with Freundlich and Langmuir

decrease in adsorption of phosphate could be

equations represented in Eqs. (2) and (3), respectively,

observed for another activated PAN fiber (FE 400) as

using the parameters in Table 2,

parameters

obtained

by

the

linear

well by increasing outgassing temperature from Q e = K F Ce1/n ,

950 ̊C to 1000 ̊C [22]. On the other hand, adsorption

(2)

amount of phosphate on FE 975 OG was a little larger than that of FE 950 OG possibly corresponding to a

where Qe and Ce are equilibrium adsorption amount

slight increase in nitrogen content from 5.7 wt-% to

and equilibrium concentration of phosphate in mmol/

5.9 wt-% by outgassing in this case.

g and mmol/L, respectively. KF in mmol/g and n

3.3 Adsorption isotherms of ammonia treated PAN

(dimensionless) are Freundlich constants [23].

fiber Qe =

Fig. 2 represented adsorption isotherms of phosphate for PYR 950 AG and PYR 975 AG. The

K eC e Qmax , 1 + K eC e

(3)

adsorption amounts of phosphate (Qe) on PYR 950 are

where Qmax is the maximum adsorption amount of

always larger than those of PYR 975 at any

phosphate in mmol/g, and Ke is adsorption affinity of

equilibrium phosphate concentration (Ce), besides both

phosphate in L/mmol. For both PYR 950 AG and PYR 975 AG, the R2 values of coefficient of correlation are a little better for Freundlich isotherms than Langmuir isotherms as can be seen in Table 2, implying that heterogeneity of surface adsorption sites of phosphate. However as drawn in Fig. 2, Langmuir isotherms approaching saturation amounts are more realistic to represent

the

concentrations

experimental than

plots

Freundlich

at

higher

isotherms.

The

estimated Langmuir maximum adsorption capacities exhibited the same value of 0.28 mmol/g for the two adsorbents in Table 2, but adsorption affinity of PYR 950 AG is 1.6 times larger than PYR 975 AG, also implying that the amount of attractive surface species

Fig. 2 Adsorption isotherms of phosphate on PYR 950 AG (circle) and PYR 975 AG (triangle). Adsorbent dosage is 30 mg in 15 mL KH2PO4 solution. Dotted and Solid lines are respectively Freundlich and Langmuir model fitting using parameters given in Table 2. Table 2

on PYR 950 AG toward phosphate could be more than that of PYR 975 AG. 3.4 Influence

of

solution

pH

on

phosphate

adsorption Fig. 3 (a), (b) and (c) represents influence of

Freundlich and Langmuir parameters of ammonia activated PAN fibers

Freundlich isotherms Adsorbent PYR950AG PYR975AG

Langmuir isotherms 2

KF

n

R

0.10 0.075

1.7 1.4

0.9774 0.9534

Qmax, mmol/g

Ke, L/mmol

R

0.28 0.28

0.63 0.39

0.9429 0.8994

Journal of Fiber Science and Technology (JFST), Vol.72, No. 11 (2016)

2

241

dramatically switched from HPO42­ to H2PO4­ around pHe 7.2 with decrease in pHe value also accompanied by more positive charged surface at lower pHe. Based on the experimental results, H2PO4­ species would be preferable to adsorb onto PYR 925-975 AG series, but decrease in adsorption amount for lower pHe solution could

also

be

observed,

probably

caused

by

-

competitive adsorption with Cl anion increased when solution pHe shifted to acidic side by the modification with hydrochloric acid [24, 25]. 3.5 Regeneration of adsorbent for repeated use The result of repeated use performance of PYR 950 AG was displayed in Fig. 4. As displayed in desorption for the first-step, most of phosphate could be removed from adsorbents. The adsorption amount of the second-step was obviously declined compared to the first-step, implying that relatively strong adsorption sites for chloride ion compared to those for phosphate could be present on PYR 950 AG as well as competitive adsorption with high concentration of chloride ion remaining on the surface [24], because regeneration was attempted with 0.1 M HCl (pH 1.0) solution. In principle, not all chloride anion on the surface of PYR 950 AG cannot be replaced with phosphate ion, when excess amount of chloride anion are still present on the surface just after the treatment of 0.1 M HCl solution. Evidently the adsorption amount of third-step was as much as that

Fig. 3 Adsorption amounts of phosphate on PYR 925 AG (a), PYR 950 AG (b) and PYR 975 AG (c) as a function of equilibrium solution pH (pHe) together with speciation diagram of phosphate at total phosphate concentration of 3.0 mmol/L. equilibrium solution pH (pHe) on adsorption amounts of phosphate on PYR 925 AG, PYR 950 AG and PYR 975 AG,

respectively,

together

with

speciation

diagrams of phosphate. Although adsorption amounts of phosphate were significantly altered with varying equilibrium solution pH (pHe) for each fiber adsorbent, the maximum adsorption amount was observed for PYR 950 AG. The significant difference in adsorption amount for each fiber could be principally caused by changing phosphate species in the pHe range examined in this study; phosphate species would be

242

Fig. 4 Adsorption amount of phosphate on PYR950AG adsorbent as a function of the number of recycle times. Adsorption was carried out using mixture of 15 mL KH2PO4 solution at initial concentration of 3.0 mmol/L and 30 mg PYR 950 AG. Desorption was conducted by 0.1 M HCl solution to replace phosphate anion trapped on PYR 950 AG with chloride anion.

Journal of Fiber Science and Technology (JFST), Vol.72, No. 11 (2016)

of the second step, although limited to 75% of the first adsorption

amount.

Consequently

the

prepared

samples can be expected to be repeatedly applicable as phosphate adsorbents from the practical point of view.

5. S. Hamoudi, A. El-Nemr, M. Bouguerra, K. Belkacemi, Can. J. Chem. Eng., 90, 34‒40 (2012). 6. X. Song, Y. Pan, Q. Wu, Z. Cheng, W. Ma, Desalination, 280, 384‒390 (2011). 7. B. Keito, S. Tanada, T. Miyoshi, R. Yamasaki, N. Ohtani, T. Tamura, Jpn. J. Hyg., 42(3), 710‒720

4. Conclusions

(1987). 8. Y. Amano, Y. Misugi, M. Machida, Sep. Sci.

Based on the experimental result, summary of this study can be itemized below.

Technol., 47, 2348‒2357 (2012). 9. M. A. A. Zaini, Y. Amano, M. Machida, J. Hazard.

1)Ammonia treatment of oxidized PAN fiber at the temperature 925−975 ̊C results in the formation

Mater., 180, 552‒560 (2010). 10. X. Liao, Y. Ding, L. Chen, W. Ye, J. Zhu, H. Fang, H. Hou, Chem. Commun., 51, 10127‒10130 (2015).

of activated carbon fiber (ACF). 2)Combined with earlier reports in the literature,

11. M. V. Lopez-Ramon, F. Stoeckli, C. Moreno-

ammonia treatment at 950 ̊C maximizes quater-

Castilla, F. Carrasco-Marin, Carbon, 37, 1215‒1221

nary nitrogen (N-Q) on the surface of PAN ACF adsorbing the maximum amount of phosphate

(1999). 12. R. Wang, Y. Amano, M. Machida, J. Anal. Appl. Pyrolysis, 104, 667‒674 (2013).

ions. 3)Adsorption of phosphate on PAN ACF is governed by surface charge, phosphate species and concentration of co-existing ions such as

13. M. Machida, S. Chensun, Y. Amano, F. Imazeki, Bull. Chem. Soc. Jpn., 88(1), 127‒132 (2015). 14. S. Sato, K. Yoshihara, K. Moriyama, M. Machida, H. Tatsumoto, Appl. Surf. Sci., 253(20), 8554‒8559

chloride and sulfate ions. 4)Repeated use of PAN ACF would be possible with hydrochloric acid to remove phosphate

(2007). 15. J. R. Pels, F. Kapteijn, J. A. Moulijn, Q. Zhu, K. M. Thomas, Carbon, 33(11), 1641‒1653 (1995).

trapped on the adsorbent.

16. J. H. Bitter, S. van Dommele, K. P. de Jong, Catal. Today, 150(1–2), 61‒66 (2010).

Acknowledgements

17. T. Iida, Y. Amano, M. Aikawa, M. Machida, J. This study was funded in part by the Japan

Environ. Chem., 23(2), 91‒94 (2013).

Society for the Promotion of Science (JSPS) under

18. S. R. Kelemen, M. L. Gorbaty, P. J. Kwiatek, T. H.

Grants-in-aid for Scientific Research (C) (No. 26340058).

Fletcher, M. Watt, M. S. Solum, R. J. Pugmire,

Gratitude is greatly extended to Ms. Shizuka

Energy Fuels, 12(1), 159‒173 (1998).

Ishibashi, Safety and Health Organization, Chiba University,

for

her

dedicated

support

in

the

19. T. N. Huan, T. Van Khai, Y. Kang, K. B. Shim, J. Mater. Chem., 22(29), 14756‒14762 (2012).

experiments. We also thank Prof. Dr. Fumio Imazeki,

20. G. Wu, N. H. Mack, W. Gao, S. Ma, R. Zhong, J. Han,

the head of Safety and Health Organization, Chiba

J. K. Baldwin, P. Zelenay, ACS Nano, 6(11), 9764‒

University, for his financial support on our study.

9776 (2012). ! 21. K. Stanczyk, R. Dziembaj, Z. Piwowarska, S.

References

Witkowski, Carbon, 33(10), 1383‒1392 (1995). 22. M. Machida, Y. Amano, F. Imazeki, TANSO, 270,

1. M. Suzumura, H. Kokubun, N. Arata, Mar. Pollut. Bull., 49, 496‒503 (2004). 2

241‒249 (2015). 23. R. Chitrakar, S. Tezuka, A. Sonoda, K. Sakane, K.

Y. Amano, Y. Sakai, T. Sekiya, K. Takeya, K. Taki,

Ooi, T. Hirotsu, J. Colloid Interface Sci., 298, 602‒

M. Machida, J. Environ. Sci., 22(11), 1666‒1673

608 (2006).

(2010).

24. S. Hamoudi, R. Saad, K. Belkacemi, Ind. Eng. Chem.

3. J. Cooper, R. Lombardi, D. Boardman, C. CarliellMarquet, Conservation and Recycling, 57, 78‒86

Res., 46, 8806‒8812 (2007). 25. T. Iida, Y. Amano, M. Machida, F. Imazeki, Chem. Pharm. Bull. 61(11), 1173‒1177 (2013).

(2011). 4. J.-W. Choi, S.-Y. Lee, K.-Y. Park, K.-B. Lee, D.-J. Kim, S.-H. Lee, Desalination, 266, 281‒285 (2011).

Journal of Fiber Science and Technology (JFST), Vol.72, No. 11 (2016)

243

Suggest Documents