To prevent nanoparticle agglomeration, various strategies have been explored to stabilize various metal nanoparticles. To stabilize nanoscale iron oxides, various stabilizers have been found to be eff ective, including thiols [ 23 ], carboxylic acids [ 24 ], silica [ 25 ], surfactants [ 26 ], polymers, including some water soluble polysaccharides [ 27–29 ], copolymers of acrylic acids [ 30–31 ], styrenesulfonic acids [ 30 ], vinylsulfonic acid [ 30 ], and long- chain alcohols [ 32 ]. However, for environmental uses, not all these stabilizers are applicable to stabilizing ZVI nanoparticles. For example, thiols and carboxylic acids may be reduced by ZVI, some polymers may not function properly in water [ 32 ], and some stabilizers themselves are either environmentally harmful or cost-prohibitive.
Stabilization of ZVI nanoparticles was also explored recently. Mallouk and coworkers employed carbon nanoparticles and poly(acrylic acid) (PAA) for stabilizing and/or delivering iron-based nanoparticles [ 33–34 ]. Lowry et al. [ 35 ] prepared sorptive nanoparticles using so-called block copolymer shells consisting of a hydrophobic inner shell surrounded by a hydrophilic outer shell for dechlorination of DNAPLs. Bhattacharyya et al. [ 36 ] synthesized iron nanoparticles with the aid of cellulose acetate membrane supports and observed improved reaction rates for TCE degradation. Using a micro-emulsion method, Li et al. [ 37 ] synthesized iron nanoparticles that degraded TCE 2.6 times faster than non-stabilized particles. Sun and Zhang [ 38 ] reported that polyvinyl alcohols can reduce the size of ZVI nanoparticles from 60 to 7.9 nm.
Savage_Ch20.indd 283
Savage_Ch20.indd 283 11/11/2008 3:51:25 PM11/11/2008 3:51:25 PM
Zhao and coworkers developed a particle stabilization strategy that employs low-cost, food-grade polysaccharides (starch or cellulose) as stabilizers [ 16–17 , 39–40 ]. These polyhydroxylated and/or polycarboxylated macromolecules possess some novel features, which may prove highly useful for stabilizing ZVI nanoparticles. First, they can serve as molecular level capsules to shield formed nanoparticles from agglomeration. Second, they are much cheaper than virtually all other stabilizers tested so far. Third, they are environmentally benign and biodegradable. And fourth, they are mobile in soils, and hence, suitable for injection uses.
Polysaccharides are the most abundant natural biopolymers, and are composed of interconnected glucose and xylose subunits. Starch and cellulose are the most abundant polysaccharide members. Starch was recently used in surface modifi cation of superparamagnetic nanoparticles [ 41– 42 ]. Under the protection of starch, Ag nanoparticles remained suspended in water for months [ 42 ]. Recently, water-soluble starches were also used to disperse Au nanoparticles in water [ 43 ], and used as a morphology-directing agent for tellurium nanowires [ 44 ].
Similar to starch, cellulose also consists of anhydroglucose subunits, which however, are joined by so-called beta linkages to form a linear chain structure.
Although it is linear and nominally thermoplastic, native cellulose is not water- soluble. However, cellulose (and starch) can be easily converted to water-soluble derivatives with desired features. Numerous modifi ed celluloses are commercially available, many of which may serve as novel stabilizers for ZVI nanoparticles.
Carboxymethyl cellulose (CMC) was used as a stabilizer in the preparation of superparamagnetic iron [ 45 ] and Ag nanoparticles [ 46 ]. The NaCMC-stabilized Ag(0) nanoparticles remained highly stable in water after 2 months. Zhang et al. stabilized Se nanoparticles with modifi ed polysaccharides, including chitosan, konjac glucomannan, acacia gum, and CMC [ 47 ]. Chitosan was also reported to be an effi cient stabilizer for Ag, Au, Pd, and Pt nanoparticles [ 48–49 ].
Moreover, heparin, dextran, and dextrin were used as stabilizing and reducing agents for Au and Ag nanoparticles [ 50–51 ]. More recently, sodium alginate was used for preparing stabilized Au nanoparticles [ 52 ].
For environmental remediation uses, a stabilizer should possess some critical attributes: (1) it can eff ectively facilitate dispersion, that is, prevent agglomeration, of nanoparticles; (2) it must not cause any harmful environmental eff ect; (3) it will not signifi cantly alter the conductivity of soils; and (4) it must be cost-eff ective.
He et al. [ 16–17 , 39–40 ] demonstrated that starch, cellulose, and their derivatives can serve as eff ective stabilizers to yield ZVI nanoparticles suitable for in situ destruction of chlorinated hydrocarbons. Figure 20.1 depicts the schematic of this modifi ed water-based procedure for synthesizing stabilized nanoparticles. The preparation is conducted in fl asks attached to a vacuum line. In Step 1, a stabilizer solution is prepared at pH 6.7–6.9. In Step 2, an Fe 3+ /Fe 2+ stock solution is added to the stabilizer solution to yield a solution with desired Fe and stabilizer concentrations. In this step, the molar ratio of
Savage_Ch20.indd 284
Savage_Ch20.indd 284 11/11/2008 3:51:26 PM11/11/2008 3:51:26 PM
stabilizer to Fe and type of a stabilizer can be varied to yield nanoparticles of various particle sizes. In Step 3, Fe 3+ /Fe 2+ is reduced to Fe(0) using approximately 1.5 times stoichiometric amounts of borohydride (NaBH 4 ). To ensure effi cient use of the reducing agent and to preserve the reactivity of the resultant Fe(0) nanoparticles, the reactor system should be operated under anaerobic conditions. Ferric iron is reduced by borohydride through the reaction shown in Equation 20.1:
Fe(H2O)63+ + 3BH4– + 3H2O → Fe0 + 10 1–2H2 + 3B(OH)3 (20.1) The resultant ZVI nanoparticles can be either used as a mono-metallic agent, or in Step 4, loaded with trace amounts (approximately 0.1 percent of Fe, w/w) of a second metal (e.g. Pd) as a catalyst to yield Fe–Pd bimetallic nanoparticles. It was shown that the addition of trace amounts of Pd substantially enhanced the dechlorination reactivity.
Figure 20.2 shows pictures of various iron particles. Whereas non-stabilized iron particles form agglomerates and precipitate in a few minutes, the starch- or cellulose-stabilized nanoparticles remain fully dispersible in water after a month. This observation suggests that the presence of a stabilizer can prevent ZVI nanoparticles from agglomeration. Fourier transform infrared spectroscopy (FTIR) study [ 17 ] revealed that the stabilization was facilitated by the sorption CMC molecules onto the surface of the ZVI nanoparticles.
Figure 20.3 compares transmission electron microscope (TEM) images of Fe–Pd nanoparticles prepared without a stabilizer (a) and with a soluble starch (b) or NaCMC (c) [ 16– 17 ]. Figure 20.3(a) shows that the non-stabilized Fe–Pd particles appeared as much bulkier dendritic aggregates. In contrast, the starch or NaCMC-stabilized nanoparticles appeared as discrete and much fi ner nanoparticles even after 1 day of aging. The mean particle size was 14.1 nm for starched particles and 8.1 nm for NaCMC-stabilized particles.
Evidently, the presence of the stabilizers prevented the agglomeration of the resultant iron particles.
Figure 20.1 An innovative procedure for synthesizing stabilized zero-valent iron (ZVI) nanoparticles.
Step 1. Solution with 0~1% (w/w) of a stabilizer ( )
Step 2. Fe3+ or Fe2+
complexes with stabilizer.
+ + +
+
+
+
Add electron donor, BH4-
Add Pd (0.1% of Fe)
Step 3. Formation of Fe(0)clusters coated with stabilizers.
Step 4. Formation of stabilized Fe-Pd nanoparticles.
Add Fe3+ or Fe2+
Savage_Ch20.indd 285
Savage_Ch20.indd 285 11/11/2008 3:51:26 PM11/11/2008 3:51:26 PM
Figure 20.4(a) shows that as the stabilizer-to-Fe molar ratio increases from 0.0025 to 0.003, the size distribution of the ZVI nanoparticles shifted substantially from primarily 37.2 and > 200 nm particles to primarily 22.8 nm particles. Further increase in CMC/Fe ratio resulted in further decrease in particle size and narrower size distribution although not as signifi cant. This observation indicates that ZVI nanoparticles of diff erent size and size distribution can be prepared by varying the stabilizer-to-Fe molar ratio.
Figure 20.4(b) shows that based on an equal CMC concentration of 0.2 percent (w/w), a CMC with greater M.W. results in smaller nanoparticles.
To test the eff ect of DS (degree of substitution), three types of CMC with the
(a) (b) (c) (d)
Figure 20.2 (a) Commercial (Fisher Scientifi c) iron powder (labeled as 100 nm);
(b) Fisher “nanoparticles” precipitate immediately in water; (c) 0.1g/L iron nanoparticles stabilized with 0.2 percent (w/w) of a starch prepared in our lab—nanoparticles remain fully dispersed in water after a month; and (d) 0.1g/L iron particles without a stabilizer prepared in our lab—non-stabilized iron particles agglomerate in minutes and precipitate in water.
Figure 20.3 Freshly prepared zero-valent iron (ZVI) particles (a) without a stabilizer, (b) with 0.2 percent (w/w) of a starch, or (c) stabilized with 0.2 percent (w/w) CMC90K and after 1 day of aging (i.e., storage in a sealed vial at 4 o C).
200 nm 200 nm
(a) (b) (c)
50 nm
Savage_Ch20.indd 286
Savage_Ch20.indd 286 11/11/2008 3:51:26 PM11/11/2008 3:51:26 PM
same M.W. of 90k but with diff erent DS (CMC90K with a DS of 0.7, EP-ML with a DS of 1.2, and 2-hydroxyethyl cellulose with no COO – groups) were compared. (The higher DS indicates more -OH groups in CMC are substituted by carboxymethyl groups). Figure 20.4(b) shows that EP-ML resulted in smaller nanoparticles (16.0 vs. 18.6 nm) than CMC90K.
CMC/Fe2+, mol/mol
0.002 0.003 0.004 0.005
Volume percentage, %
0 20 40 60 80 100
D < 50nm 50nm < D <200nm D > 200nm 37.2nm
22.8nm
19.6nm
Samples
1 2 3 4 5
Volume percentage, %
0 20 40 60 80 100
D < 50nm
50nm < D < 200nm D > 200nm
HP7A
CMC90k
CMC250k EP-ML 18.6nm
15.6nm 16.0nm
>200nm (a)
(b)
Figure 20.4 (a) Size distribution of zero-valent iron (ZVI) nanoparticles synthesized at diff erent carboxymethyl cellulose (CMC)/Fe 2+ ratios and at an initial Fe 2+ concentration of 1 g/L (stabilizer: CMC90k, temp: 22°C). (b) Size distribution of Fe nanoparticles stabilized with CMC of various molecular weight and degree of substitution (Fe: 0.1g/L, CMC90K: 0.2 percent (w/w), temp: 22°C).
Savage_Ch20.indd 287
Savage_Ch20.indd 287 11/11/2008 3:51:29 PM11/11/2008 3:51:29 PM