Geckeler K.E., Nishide H. (Eds.) Advanced Nanomaterials

Подождите немного. Документ загружается.

19.4 Carbon Nanotubes as Adsorbents 629

adsorption capacity for Cd(II) than did the pristine, as - grown CNTs. Subsequently,

the authors measured the physico - chemical properties of oxidized CNTs and evalu-

ated their Cd(II) adsorption capacity. The speciﬁ c surface area and pore - size dis-

tributions of the as - grown and oxidized CNTs were measured using nitrogen

adsorption, with the BET (Brunauer – Emmett – Teller) method. The functional

groups on oxidized CNTs were assessed quantitatively using Boehm ’ s titration

method [69] , and the zeta potentials of the as - grown and oxidized CNTs were also

evaluated. Based on these results, it was proposed that the Cd(II) adsorption capaci-

ties for the three types of oxidized CNT were increased due to functional groups

having been introduced by oxidation, compared to the as - grown, pristine CNTs.

The observed Cd(II) adsorption capacity of the as - grown CNTs reached only

1.1 mg g

− 1

, compared to values of 2.6, 5.1, and 11.0 mg g

− 1

for nanotubes treated

with H

, HNO

and KMnO

, respectively. The authors linked these results to

the increase in surface area observed following each chemical treatment. The data

obtained regarding the particle size distribution and suspensibility of these materi-

als indicated that oxidation with H

and KMnO

only partially broke up the

nanotubes, whilst oxidation with HNO

cut short completely most of the CNTs.

The observed dependence of the zeta potential of the as - grown and oxidized CNTs

on pH is shown graphically in Figure 19.10 . At the same pH value, the zeta potential

for the three types of oxidized CNTs followed the order H

< KMnO

< HNO

and suggests that the amounts of acid - functional groups increase following the

same order. The adsorption isotherms of Cd(II) also indicated that the functional

groups introduced by oxidation improved the ion - exchange capabilities of the CNTs

and thus led to corresponding increases in the Cd(II) adsorption capacities. A

removal efﬁ ciency close to 100% at a CNT dosage of 0.08 g 100 ml

− 1

was observed

for the KMnO

- oxidized CNTs, which suggested that this treatment represented an

effective means of improving the Cd(II) adsorption capacity.

19.4.1.4 Adsorption of Copper ( II )

Despite being one of the most widespread environmental contaminants, copper

is essential to human life and health, yet is potentially toxic in larger quantities.

In humans, the ingestion of relative large quantities of copper salts may cause

Figure 19.10 Zeta potential curves versus pH for pristine and

oxidized CNTs. Reproduced from Ref. [68] .

630 19 Carbon Nanotubes as Adsorbents for the Removal of Surface Water Contaminants

severe abdominal pain, vomiting, diarrhea, hemolysis, hepatic necrosis, hematuria,

proteinuria, hypotension, tachycardia, convulsions, coma, and death. The major

sources of copper in industrial efﬂ uents are metal cleaning and electroplating.

Wu [70] has evaluated the Cu(II) adsorption efﬁ ciency of pristine and chemically

modiﬁ ed CNTs, the latter being functionalized using HNO

and NaOCl. This

chemical treatment caused increases in both the pore volume and average pore

size of the CNTs, while the value of the isoelectric point was shown to decrease.

A comparison of the infrared spectra of the as - produced and modiﬁ ed CNTs

indicated that several functional groups had been generated on the surface of

modiﬁ ed CNTs. The zeta potential values of the as - produced and modiﬁ ed CNTs,

as shown in Figure 19.11 , indicated that the surface of the as - produced and HNO

modiﬁ ed CNTs was positively charged in solution, whereas the zeta potentials of

NaOCl - modiﬁ ed CNTs were all negative. These negatively charged surfaces would

electrostatically favor the adsorption of Cu(II), as was observed by Wu and cowork-

ers. Indeed, the Cu(II) adsorption capacity followed the order NaOCl - modiﬁ ed >

HNO

- modiﬁ ed > as - produced CNTs. These ﬁ ndings suggest that modifying the

surface of the CNTs may not only provide a more negatively charged and

hydrophilic surface but also generate a variety of functional groups, markedly

promoting the adsorption of Cu(II) onto the modiﬁ ed CNTs. The maximum

adsorption capacities observed in this study at different temperatures are sum-

marized in Table 19.3 .

19.4.1.5 Adsorption of Zinc ( II )

Whilst zinc (II) is essential for human health, large amounts can be harmful. The

consequences of a relatively large intake of Zn(II) include lethargy, light - headed-

ness, ataxia, oropharyngeal cancer, gastric burns, epigastric tenderness, pharyn-

geal edema, hematemesis, and melena [71] . The suitability of CNTs (both MWNT

and SWNT) to adsorb Zn(II) from water was studied by Lu et al . [72] , whose data

showed the adsorption capacity of CNTs to be greatly improved following a speciﬁ c

Figure 19.11 Zeta potential curves versus pH for pristine

and oxidized CNTs. Reproduced from Ref. [70] .

19.4 Carbon Nanotubes as Adsorbents 631

Table 19.3 Summary of copper adsorption capacities ( q

) of

CNT s at different temperatures in terms of the adsorbed

mass of Cu(II) in solution (mg) per mass of nanotube (g).

Reproduced from Ref. [70] .

Temperature ( ° C)

Adsorption capacity ( q

) (mg g

− 1

)

As - produced CNTs HNO

- modiﬁ ed CNTs NaOCl - modiﬁ ed CNTs

7 6.39 12.46 44.64

17 7.87 13.10 45.87

27 8.25 13.87 47.39

37 9.34 15.11 49.02

47 10.17 16.04 51.81

chemical treatment that renders the CNTs more hydrophilic, and thus more effec-

tive in adsorbing Zn(II).

Lu and coworkers showed that the adsorption capacity of Zn(II) onto CNTs

increased in line with a rising pH of the solution over the range 1 to 8, was

maximal at pH 8 to 11, and then decreased at pH 12. A comparative study on the

adsorption of Zn(II) between CNTs and commercially available, powdered acti-

vated carbon, was also conducted. The maximum adsorption capacities for Zn(II)

observed were 43.66, 32.68, and 13.04 mg g

− 1

with SWNTs, MWNTs, and activated

carbon, respectively. The short contact time required to reach equilibrium, as well

as the high adsorption capacity, suggests that both SWNTs and MWNTs possess

a high potential for the removal of Zn(II) from water. The same group also sug-

gested that the higher adsorption capacity observed for SWNTs over MWNTs

might be due to the higher surface area observed for SWNTs (423 m

− 1

compared

to 297 m

− 1

observed for MWNTs), and also to the larger proportion of defects

present on the MWNTs (as observed using Raman spectroscopy).

However, an interesting result was observed for the case of activated carbon.

Although the surface areas of puriﬁ ed SWNTs and MWNTs were much lower than

that of activated carbon, the adsorption capacities of Zn(II) onto puriﬁ ed SWNTs

and MWNTs were much higher than was observed for activated carbon. This

superior adsorption capacity was attributed to a larger number of hydrophilic

groups present in the CNT walls.

Another study which focused on the adsorption kinetics and equilibrium of

Zn(II) adsorbed onto CNTs nanotubes has also been reported [73] . This thermo-

dynamic analysis revealed that the sorption of Zn(II) onto CNTs was endothermic

and spontaneous, and that the Zn(II) ions could easily be removed from the

surface sites of SWNTs and MWNTs by the action of a 0.1 M nitric acid solution.

Moreover, the original adsorption capacity was maintained after 10 cycles of this

sorption – desorption process. Such data suggest that both types of CNT material

could be reused through many cycles of water treatment and regeneration.

632 19 Carbon Nanotubes as Adsorbents for the Removal of Surface Water Contaminants

19.4.1.6 Adsorption of Nickel ( II )

Nickel is a toxic metal ion that is present in wastewaters. More than 40% of all

nickel produced is used in steel factories, in nickel batteries, and in the production

of some alloys, which leads to an increase in the Ni(II) burden on the ecosystem

and a deterioration in water quality. If ingested, Ni (II) is harmful, and may cause

vomiting, chest pain, and a shortness of breath [74] .

Lu et al. [75] have analyzed the effect of CNT mass, agitation speed, initial Ni(II)

concentration, and solution ionic strength on the Ni(II) adsorption capacity of

CNTs. The effects of agitation speed and solution ionic strength on Ni(II) sorption

by oxidized CNTs are shown in Figures 19.12 and 19.13 . The adsorption capacity

of both SWNTs and MWNTs was shown to increase as the agitation speed

increased, but to decrease when the ionic strength of the solution increased. The

same group also reported that SWNTs showed a better performance for Ni(II)

adsorption than did MWNTs. A similar conclusion was reached by Chen et al . [58] ,

when studying the adsorption of Ni(II) onto oxidized MWNTs as a function of

contact time, pH, ionic strength, MWNT amount, and temperature. The results

showed that Ni(II) adsorption onto MWNTs was heavily dependent on the pH and,

to a lesser extent, the ionic strength. Kinetic data showed the adsorption process

to achieve equilibrium within 40 min, and that the process followed a pseudo

second - order rate equation. The adsorption data ﬁ tted the Langmuir model and,

together with thermodynamic data, indicated the spontaneous and endothermic

nature of the process. The results of a desorption study showed that Ni(II) adsorbed

onto oxidized MWNTs could easily be desorbed at pH < 2. The authors proposed

that ion exchange might be the predominant mechanism for Ni(II) adsorption on

oxidized MWNTs.

Figure 19.12 Effect of agitation speed on Ni(II) adsorption by

oxidized CNTs. Reproduced from Ref. [75] .

19.4 Carbon Nanotubes as Adsorbents 633

19.4.1.7 Competitive Adsorption of Heavy Metals Ions

Whilst most reports on heavy metal adsorption using CNTs has focused on a single

metal ion, Li and colleagues [76] were the ﬁ rst to conduct a study on the competitive

adsorption of Pb(II), Cu(II) and Cd(II) onto HNO

- treated MWNTs. These studies

showed the afﬁ nity order of these metal ions towards CNTs to vary in the order:

Pb(II) > Cu(II) > Cd(II) (Figure 19.14 ). The Langmuir adsorption model repre-

sented the experimental data for Pb(II) and Cu(II) well, but did not provide a good

ﬁ t for the adsorption data of Cd(II). It was also observed that, at a low pH, the

adsorption percentages were negligible, but that at pH values between 1.8 and 6.0

the proportions of Pb(II) and Cu(II) increased sharply, almost attaining values of

100%, while only a small increase was noted for Cd(II).

Figure 19.13 Effect of solution ionic strength on Ni(II)

adsorption by oxidized CNTs. Reproduced from Ref. [75] .

Figure 19.14 Competitive adsorption data for three ions of

Pb(II), Cu(II) and Cd(II) onto CNTs at room temperature and

pH 5.0. Reproduced from Ref. [76] .

634 19 Carbon Nanotubes as Adsorbents for the Removal of Surface Water Contaminants

Another study, conducted by Hsieh et al . [77] , focused on the competitive adsorp-

tion of Pb(II), Cu(II) and Cd(II) by CNTs grown on microsized Al

particles.

The authors noted that the adsorption behavior of these metal ions on CNTs on

particles followed Langmuir ’ s adsorption model, with observed adsorption

capacities on the CNTs for Pb(II), Cu(II) and Cd(II) of 32, 18, and 8 mg g

− 1

, respec-

tively. These results conﬁ rmed that CNTs supported on Al

particles showed

potential for the removal of soluble heavy metals from aqueous solutions.

Peng and collaborators [78] conducted a similar study to determine the adsorp-

tion capacity of Pb(II) and Cu(II) on CNT – iron oxide composites. These ferric

composites offer the advantage of a continuous contaminant adsorption from a

liquid efﬂ uent whilst, when adsorption is complete, the adsorbent can be separated

from the liquid phase simply by using a magnet. Both, X - ray diffraction ( XRD )

and SEM studies indicated the presence of an entangled network of CNTs with

attached iron oxide nanoclusters. The adsorption isotherms obtained for Pb(II)

and Cu(II) adsorbed onto these magnetic composites are shown in Figure 19.15 .

The maximum adsorption capacities for Pb(II) and Cu(II) in the concentration

range studied were 105.67 and 45.12 mg g

− 1

. After adsorption, a magnetic separa-

tion process was carried out using a permanent magnet, providing a 98% recovery

of the Pb(II) and Cu(II) ion mass adsorbed.

19.4.2

Adsorption of Other Inorganic Elements

In addition to heavy metal ions, other common inorganic pollutants in drinking

water include ionic forms of ﬂ uoride, arsenate, and americium - 243(III). Although

ﬂ uoride is added via drinking water, it is often present in surface waters due to

Figure 19.15 Adsorption data for Pb(II) and Cu(II) onto

magnetic CNTs composites (pH = 5.0, T = 20 ° C). Reproduced

from Ref. [78] .

19.4 Carbon Nanotubes as Adsorbents 635

natural erosion or to discharges from fertilizers and aluminum factories. The

ingestion of ﬂ uoride can cause bone disease and mottled teeth in children. Arse-

nate is occasionally present due to erosion of natural deposits, or it may be released

into surface waters via the runoffs from orchards or from glass and electronics

production wastes. Arsenate ingestion causes skin damage and circulatory prob-

lems; it is also a well - known carcinogen [2] . Americium - 243(III) contributes sig-

niﬁ cantly to the radiotoxicity of nuclear waste, and may be released into the

environment during nuclear waste storage, processing, or disposal. Exposure to

small traces of americium - 243(III) increases the risk of cancer.

19.4.2.1 Adsorption of Fluoride

The acceptable ﬂ uoride concentration in drinking water is generally in the range

of 0.5 to 1.5 mg l

− 1

[79] . Higher concentrations will affect the metabolism of

calcium and phosphorus in the human body, and lead to dental and bone ﬂ uorosis

[80] . Many methods have been adopted to remove excess ﬂ uoride from drinking

water, the most common approach being adsorption with activated alumina, which

has a good adsorption capacity and selectivity for ﬂ uoride. Unfortunately, the

optimum capacity for ﬂ uoride removal in alumina occurs only at pH values below

6.0, which strongly limits the practical applications of this material [81] .

Li and collaborators [52] have reported that amorphous Al

supported on CNTs

represents a major candidate for ﬂ uoride adsorption from water. These authors

used CNTs as a supports for Al

, and showed the composite to have a high poten-

tial for removing ﬂ uoride from drinking water. Based on their adsorption isotherms

(Figures 19.16 and 19.17 ), Li and coworkers found that Al

/CNT composites

showed a high ﬂ uoride adsorption capacity over a pH range from 5.0 to 9.0. They

also found the adsorption capacity for the Al

/CNT composite to be about 13.5 -

fold higher than that of activated carbon, fourfold higher than for γ - alumina, and

also higher than that of the commercial polymeric resin, IRA - 410. This broad range

of pH values and high adsorption capacities observed for the Al

/CNT composite

makes this material very attractive for ﬂ uoride removal from water.

Figure 19.16 Effect of solution pH on ﬂ uoride adsorption

onto an Al

/CNT composite. Reproduced from Ref. [52] .

636 19 Carbon Nanotubes as Adsorbents for the Removal of Surface Water Contaminants

A year later, the same group presented improved results using an array of

aligned MWNTs [10] . For this, the authors studied the kinetics of the ﬂ uoride

adsorption process, and the effect of pH on ﬂ uoride adsorption capacity. The

kinetic data indicated that the ﬂ uoride adsorption rate was rapid during the ﬁ rst

60 min, and quickly reached an adsorption capacity of 3.0 mg g

− 1

, with equilibrium

achieved after 180 min. A mild dependence of the adsorption capacity on the pH

of the solution was also observed, with the highest adsorption capacity being

observed at pH 7, and reaching 4.5 mg g

− 1

at ﬂ uoride concentrations of 15 mg l

− 1

The adsorption isotherms obtained for this material, compared to those obtained

in γ - Al

, activated carbon, and a soil, are shown in Figure 19.18 . At low ﬂ uoride

Figure 19.17 Adsorption isotherms of ﬂ uoride on activated

carbon (AlC - 300), CNT, γ - Al

, a commercial resin (IRA - 410)

and Al

/CNT composite. The data for activated carbon and

IRA - 410 were ﬁ tted with a Langmuir isotherm, while the data

for CNT and Al

/CNTs were ﬁ tted with a Freundlich

isotherm. Reproduced from Ref. [52] .

Figure 19.18 Adsorption isotherm of ﬂ uoride on CNTs (at pH

7 and 25 ° C) compared with γ - Al

, soil, and activated

carbon. Reproduced from Ref. [10] .

19.4 Carbon Nanotubes as Adsorbents 637

initial concentrations ( < 1 mg l

− 1

), the nanotube material and alumina had the same

adsorption capacities, but at higher ﬂ uoride concentrations the ﬂ uoride adsorption

capacity of the CNTs was higher. These results again indicated the potential of

CNTs in ﬂ uoride removal.

19.4.2.2 Adsorption of Arsenic

Arsenic is required as a micronutrient for the human body, yet it may be carcino-

genic if consumed constantly. Thus, it is of great importance to remove arsenic

from water before it is used for drinking. Two main forms of arsenic are encoun-

tered in natural water, namely trivalent (As(III), arsenite) and the higher - oxidized

form, pentavalent (As(V), arsenate). Whilst either of these species can be found in

natural waters, As(III) is more common in groundwater, and As(V) is more

common in surface water [82] .

When a novel ceria – CNT composite was proposed as an alternative for the

removal of arsenate from water [83] , the results indicated that arsenate adsorption

on these materials was pH - dependent. The presence of Ca(II) and Mg(II) also

signiﬁ cantly enhanced the adsorption capacity. These very promising results

suggest that these materials might represent a promising adsorbent for drinking

water puriﬁ cation. The effect of pH on the adsorption of As(V) onto CeO

/CNT is

shown in Figure 19.19 .

These results indicate that the adsorption of As(V) is pH - dependent, and that

the pristine composite has a higher adsorption capacity than the chemically modi-

ﬁ ed nanotubes. The authors proposed that the dependence of adsorption on pH

was due to variations in the surface charge on the nanotube composites. This was

corroborated by zeta potential measurements. Moreover, the authors also meas-

ured the inﬂ uence of Ca(II) and Mg(II) on the adsorption capability of the CeO

CNT material. From the results shown in Figure 19.20 it is clear that both Ca(II)

and Mg(II) signiﬁ cantly enhance the adsorption capacity of the nanotube compos-

ite. In fact, an increase from 0 to 10 mg l

− 1

in the concentration of Ca(II) and Mg(II)

resulted in an almost one order of magnitude increase in the amount of As(V)

Figure 19.19 Effect of pH on As(V) adsorption by chemically

treated CNTs (

䊏

) and CeO

/CNT composite (

䉬

). Reproduced

from Ref. [83] .

638 19 Carbon Nanotubes as Adsorbents for the Removal of Surface Water Contaminants

adsorbed. The authors explained this observed increase in adsorption capacity by

proposing the formation of a ternary surface complex between calcium or magne-

sium, arsenate, and the ceria surface [84] .

19.4.2.3 Adsorption of Americium - 243 ( III )

Wang and collaborators [85] examined the use of MWNTs as adsorbents for the

radionuclide americium - 243 (III). The data obtained (see Figure 19.21 ) indicated

Figure 19.20 Effect of Ca(II) and Mg(II) on As(V) adsorption

(initial concentration of As(V) = 20 mg l

− 1

). Reproduced from

Ref. [83] .

Figure 19.21 Adsorption isotherms for

243

Am(III) onto

MWNTs in polyethylene tubes using Milli - Q water in the

presence of 0.1 M NaClO. The contact time was 4 days.

Reproduced from Ref. [85] .