Photocatalytic self-detoxification by coaxially electrospun fiber containing titanium dioxide nanoparticles

Abstract

Electrospun nanofibers containing titanium dioxide (TiO₂) were investigated as a self-detoxifying system consisting of polyacrylonitrile (PAN) and anatase TiO₂ nanoparticles. Fibers were prepared by uniaxial and coaxial electrospinning to study the effect of nanoparticle placement on the detoxification activities of the photocatalyst. Coaxially spun fibers had the particles selectively placed in the sheath layer by electrospinning pure PAN solution and PAN/TiO₂ solution as the core and sheath layer, respectively. Using scanning electron microscopy, X-ray microanalysis and X-ray photoelectron spectroscopy, it was confirmed that the coaxial approach resulted in the location of nanoparticles near the surface of the fibers compared to the uniform distribution obtained for uniaxial fibers. Photocatalytic activity of the fibers under ultraviolet irradiation was demonstrated by the degradation of aldicarb, as measured by high-performance liquid chromatography. In terms of degradation kinetics, the distribution density of TiO₂ nanoparticles in the fiber surface region significantly affected the initial degradation rate, while the final decomposition amounts after 3 h did not differ significantly.

Keywords

coaxial electrospinning self-decontamination core/sheath

Photocatalysis, a partial oxidation of alkanes and olefinic hydrocarbons, was introduced and developed in 1970;¹ these reactions took place at ambient temperature under ultraviolet (UV) irradiation. The nature of the reaction medium is heterogeneous, being comprised of at least two phases: the solid (catalyst) and a fluid reagent (gas or liquid). Current research and development activities use the application of photocatalysis as the basis for environmentally friendly technologies. Childs and Ollis² demonstrated that the hydroxyl radicals produced during the sequence of light-induced redox reactions were responsible for the oxidative degradation of organic pollutants present in water and air with titania as a photocatalyst. Within the past 30 years, semiconductor photocatalysis has been successfully used in the removal of over 1200 different organic toxicants in various media.³

Titanium dioxide (TiO₂) as a photocatalyst has been investigated for almost four decades.⁴ Photo-oxidation, one of the unique features of this metal oxide, is a mechanism suggested to define the driving force of strong oxidation by such metal-based inorganic catalysts.⁵ TiO₂ has been studied in various forms, such as nanoparticles,⁶ clusters,⁷ encapsulated particles,⁸ thin films,⁹ aerogels¹⁰ and nanofibers,^11,12 considering various applications, such as highly efficient photocatalysis, solar energy conversion and self-cleaning ingredients. In this study, TiO₂ nanoparticles were employed to form a photocatalytic nanofiber based on a polymeric substrate.

Coaxial electrospinning provides the technology to produce many different morphologies and nanofiber structures that were previously unattainable through simple monoaxial electrospinning.¹³ Introduced around 2003,¹⁴^–¹⁶ it uses two different fluids flowing through concentric spinnerettes to generate nanofibers with a core-sheath structure.¹⁷ Previous studies have indicated that while monoaxial nanofibers exhibit the capability to support catalytic nanoparticles and prevent their aggregation, if these nanoparticles are located at the center of the nanofiber there is a significant mass transfer limitation for the reactant to reach the catalytic particles, thereby making any catalyst at the center virtually unavailable.¹⁸ Coaxial electrospinning can be used to tune the catalyst location in the shell or surface region of the nanofiber. In this work, the TiO₂ nanoparticle was used as a self-decontaminating catalyst in both monoaxial and coaxial electrospun nanofibers to study any photocatalytic differences due to nanoparticle distributions: randomly distributed nanoparticles by monoaxial electrospinning versus sheath side-embedded ones by coaxial electrospinning.

Experimental details

Materials

Polyacrylonitrile (PAN) (Mn ∼ 150 kDa, Poly Science Inc., Warrington, PA), N,N-dimethylformamide (98%, Fluka, Milwaukee, WI) and aldicarb (2-methyl-2(methylthio)propanal o-[(methylamino)-carbonyl] oxime, purity 99%, FW: 190.26, Chem Service, West Chester, PA) were purchased commercially and anatase TiO₂ was provided by Samsung Cheil Industries (Seoul, South Korea).

Electrospinning

Solutions of 10 weight percent PAN (MW = 150 kDa) in N,N-dimethylformamide and various weight percent titania nanoparticles were prepared and heated at 100°C for 24 h. Monoaxial fibers were generated using a syringe pump (Harvard Apparatus PHD 2000 Infusion, Holliston, MA) flowing at 0.03 ml/min through a 22-gauge needle (Hamilton N722, Reno, NV). The electric field was generated by a 15 kV electric charge and a 15 cm tip to collector plate distance. Coaxial electrospun fiber samples were generated using concentric needles, inner of 24 gauge and outer of 18 gauge, with an inner/outer flow rate ratio of 2/3. The amount of components in each solution and the spinning conditions are listed in Table 1

Table 1.

Uniaxial and coaxial electrospinning of polyacrylonitrile/titanium dioxide (PAN/TiO₂) solutions

Spinning type	Sample code	PAN/TiO₂ (w/w)	Solution flow rate (ml/min)	Needle diameter (mm, I.D.)
Uniaxial	U-0	1/0	0.03	0.41
	U-1	1/0.2	0.03	0.41
	U-2	1/0.5	0.03	0.41
Coaxial	C-1	Core: 1/0	Core: 0.010	Core: 0.31
		Shell: 1/0.3	Shell: 0.015	Shell: 0.84
	C-2	Core: 1/0	Core: 0.010	Core: 0.31
		Shell: 1/0.5	Shell: 0.015	Shell: 0.84

Fiber characterization

For scanning electron microscopical analysis, electrospun fibers were mounted on aluminum microscopy stubs using carbon tape. The specimens were coated with gold-palladium (Au-Pd) for 30 s using an Edwards Auto 306 High Vacuum Evaporator (Edwards High Vacuum International, Wilmington, MA). Fiber morphology and TiO₂ particle distribution were observed using a Field Emission Scanning Electron Microscope (FESEM) – Hitachi 4500 (Tokyo, Japan). Backscattered electron (BSE) imaging and subsequent energy-dispersive X-ray (EDX) analyses were conducted using a Scanning Electron Microscope – JEOL model XA-8900R superprobe (JEOL Ltd., Tokyo, Japan) equipped with a Tracor Northern Flextran Series-II Energy Dispersive X-ray Analyzer (Middletown, WI). All the electron microscopy images were obtained with an accelerating voltage of 10 KeV. The specimen current in BSE imaging was 5.0 nA. It was necessary to use EDX analysis to identify TiO₂ in the fiber mat, putting the X-ray spot probe on a location of interest for 30 s with respect to the energy at 4.5 KeV of Ti. Analysis of these data allowed comparisons of TiO₂ nanoparticles at different locations on the fiber.

Sample preparation for the electron microscopy analysis of the fiber cross-section involved embedding the electrospun fiber mat. Epo-ﬁx embedding resins A (1232-R) and B (1232-H), obtained from Electron Microscopy Sciences (Hatﬁeld, PA), were mixed together in a mass ratio of 5:1 and transferred to silicone rubber molds (EMS, Hatﬁeld, PA), where fibers were aligned and fixed with two pins at both ends in the resin matrix. After curing in an oven at 70°C for 15 h, the embedded fiber was then sectioned vertically in a transverse direction at room temperature, using an ultra-microtome with a diamond knife, into slices 60–80 nm thick.

Elemental analysis was performed by X-ray photoelectron spectroscopy (XPS) (Surface Science Instrument, Model SSX-100, Mountain View, CA) with operating pressure <2 × 10^–9 Torr and monochromatic AlK_α X-rays at 1486.6 eV. Photoelectrons were collected at an angle of 0° from the surface normal of samples, and the analysis depth was 27 nm from the surface. The area by X-ray beam spot was about 1 mm diameter on the nanofibrous mats film. Survey scans were conducted with pass energy of 150 V.

Photodegradation of aldicarb solution

Photo-oxidation experiments were conducted with the electrospun PAN nanofibers containing varying amounts of TiO₂. A solution for the decontamination test was made with 2 mM aldicarb, a carbamate pesticide chemical, in high-performance liquid chromatography (HPLC)-grade water. A fiber mat 3 (±0.01) mg was submerged in each 5 ml aqueous aldicarb solution. After sonication in an ultrasonic bath for 5 min to distribute the fibers uniformly and remove air bubbles from the solution, test tubes containing solution contaminated with the electrospun fibrous web were placed in an UV chamber and exposed to UV radiation for 1, 2 and 3 h. The chamber had eight fixed UV lamps (350 nm wavelength, 4 watt, 3 inch tall each) on the wall side and a rotating sample holder with 2.5 cm distance between the sample and the lamp.

Following the photoreaction, the specimens were centrifuged with a force of 1400 gravity (5 cm rotating radius at 5000 rpm) for 3 min; the supernatant was filtered through a disk-type syringe filter (Alltech Assoc. Inc., Deerfield, IL) with 25 mm diameter consisting of a 0.2 µm pore-size nylon membrane in order to remove particles and fibrils and then they were placed in a 2 ml HPLC vial. Photodegradation activities of the electrospun fibers were measured for three replicates by analyzing the concentration of aldicarb in each treatment solution of 1.5 ml using HPLC (Agilent 1100, Santa Clara, CA) with method conditions: 15°C, C18 column, 60% acetonitrile/40% water (pH 3 using H₃PO₄), 220/4 detector (DAD), flow rate 1 ml/min, with detection for 15 min.

Results and discussion

Fiber characterization

Fibers were electrospun from five PAN-TiO₂ solutions (Table 1). Both uniaxial and coaxial electrospun fibers had average diameters in the range of 0.8–0.9 µm, except C-2, while the higher TiO₂ content coaxially spun fiber showed 1.9 µm average diameter (Figure 1). In U-2 and C-2 (Figure 1(b) and (d)), some beaded fibers were observed as they had higher amounts of TiO₂ in the fibers. The morphology of coaxially electrospun fiber containing 33 wt% of TiO₂ (sample code: C-2) is shown in Figure 1(e). The fiber has an irregular surface and observable nanoparticles, which are embedded and exposed on the surface with diameters in the range of 20–50 nm. The areas with higher intensity of secondary electron scattering were investigated with X-ray microanalysis. This electron microprobe analysis confirmed that these regions on the fiber surface contained titanium (Ti) element with energy at 4.5 KeV, while there was no Ti observed for the background regions of the fiber surface. The XRD patterns shown in Figure 2 presented consistent characteristics of TiO₂ particles. The TiO₂ powder that we used exhibited typical peaks of anatase crystal structure (Figure 2(c)) that were also observed in electrospun PAN-TiO₂ fibers (Figure 2(a) and (b)).

Figure 1.

SEM images of electrospun PAN fiber containing TiO₂ nanoparticles; (a) U-1 with an average diameter 0.9 μm (range = 0.6 to 3.2 μm), (b) U-2 with an average diameter 0.8 μm (range = 0.4 to 1.5 μm), (c) C-1 with an average diameter 0.8 μm (range = 0.5 to 1.5 μm), (d) C-2 with an average diameter 1.9 μm (range = 0.4 to 17.6 μm), (e) Secondary electron image of C-2 single fiber, white scale bar = 10 μm and black scale bar = 0.5 μm.

Figure 2.

X-ray diffraction patterns (a) Electrospun fiber U-2; (b) Electrospun fiber U-1; (c) TiO₂ powder; (d) PAN powder.

X-ray microanalysis spot probes for cross-sectional specimens of fibers were also conducted (Figures 3 and 4). The results of cross-sectional X-ray microanalysis confirmed there is a structural difference between the uniaxial and coaxial electrospun fibers. For the uniaxially electrospun fiber, TiO₂ particles were uniformly dispersed (Figure 3). For coaxial spun fiber, Ti was observed at locations close to the surface (Figure 4; points 1, 2 and 5), while the locations near the center of the cross-section did not show any Ti signal (Figure 4; points 3 and 4).

Figure 3.

Secondary electron cross-section image and electron microprobe analysis of uniaxially electrospun PAN-TiO₂ fiber.

Figure 4.

Secondary electron cross-section image and electron microprobe analysis of coaxially electrospun PAN-TiO₂ fiber: surface region (sheath) at points 1, 2 and 5; center region (core) at points 3 and 4.

These results demonstrated that coaxial electrospinning formed a bi-component, core-sheath fiber structure with TiO₂ particles located in the sheath. In the Figures 3 and 4, the peaks of gold (Au) are due to the Au-Pd sputter coating on the fiber specimen used to reduce charging in the electron microscope. We note that fibers with very large diameter (∼10 µm) were selected for both cases to be able to probe spatial distribution of TiO₂ particles.

XPS analysis presented elemental contents in a region from the surface to 27 nm depth of the electrospun fibers. The atomic ratio of Ti/N (Table 2), a normalized relative intensity of Ti in PAN-TiO₂ fiber, showed that uniaxial nanofiber containing 33% TiO₂ had a significantly lower amount of titanium in the surface region than coaxial fiber with the same total TiO₂ content. The result is in good agreement with the cross-sectional morphologies of energy-dispersive spectroscopy in Figures 3 and 4. Furthermore, this method substantiates the overall characteristic that there are more TiO₂ particles existing on the surface of coaxial fiber than that of uniaxial fiber for the fiber mats (Table 2).

Table 2.

X-ray photoelectron spectroscopy (XPS) result for uniaxial and coaxial polyacrylonitrile/titanium dioxide (PAN/TiO₂) fibers containing 33 % TiO₂

Atom	Atomic %
Atom	Uniaxial	Coaxial
C	77	76
O	3	6
N	20	18
Ti (2p)	0.059	0.279
Ti/N	0.003	0.015

Photocatalytic degradation

In HPLC analysis, PAN-TiO₂ electrospun fiber exhibited photocatalytic degradation of aldicarb (Figure 5). The amount of aldicarb (I in Figure 6) with the retention time of 8.8 min decreased over time, while that of the other two products increased over the degradation time of 3 h. Two degradation products (II and III) were observed at retention times of 5.1 and 7.3 min (Figure 5 and Table 3) that are consistent with a previous study¹⁹ and the photo-oxidation pathway of aldicarb (Figure 6).

Figure 5.

HPLC Chromatogram Showing Changes in Concentration of Aldicarb and Oxidized Products by PAN-TiO₂ Fiber.

Figure 6.

Scheme for Photocatalytic Oxidation of Aldicarb.¹⁹.

Table 3.

Aldicarb and the oxidized derivatives

Compound	Common name	Compound in Figure 6	MW (g/mole)	Retention time in HPLC (min)
Propanal, 2-methyl-2(methylthio)-, O-[(methylamino)-carbonyl] oxime	Aldicarb	I	190	8.8
propanal, 2-methyl-2(methylsulfonyl)-, O-[(methylamino)-carbonyl] oxime	Aldicarb sulfone	II	222	7.3
2-propenal, 2-methyl-, O-[(methylamino) carbonyl] oxime	–	III	142	5.1

HPLC: high-performance liquid chromatography.

All fiber mats exhibited a decrease of aldicarb over time (Figure 7). In particular, it was observed that the fiber containing a higher content of TiO₂ degraded more aldicarb for both uniaxially (a) and coaxially (b) electrospun fibers. Photocatalytic properties of the TiO₂-containing electrospun fibers offer potential for application of these fibers for protective materials. A difference in degradation activity between the uniaxial and coaxial fibers was observed. The initial degradation rate for uniaxial fiber was lower than that for coaxial fiber. Degradation time to half of the initial mass (1.9 mg) was just about 1 h in coaxial fiber, while it was approximately 2 h in uniaxial fiber (Figure 8). In terms of degradation kinetics, the decomposition rates of aldicarb for the first hour were 0.69 mg/h by uniaxial and 0.95 mg/h by coaxial fiber, respectively, that is, coaxial fiber with TiO₂ located in the sheath degraded aldicarb faster than uniaxial fiber with the TiO₂ distributed through the fiber structure. This resulted from the diffusion lengths of aldicarb through the PAN matrix to TiO₂ and the probabilities of reaction between aldicarb and TiO₂. Thus, it is demonstrated that the location of TiO₂ close to the surface of the coaxial fiber influences the reaction condition and determines the decontamination rate. Since the reaction rate is important in protective materials, the core-sheath structure obtained by coaxial electrospinning offers the potential of providing enhanced self-decontamination properties.

Figure 7.

Photocatytic Degradation of Aldicarb by (a) Uniaxial and (b) Coaxial Electrospun Fiber Mats containing TiO₂ Nanoparticles.

Figure 8.

Photocatytic Degradation of Aldicarb by Uniaxial vs Coaxial Electrospun Fibers containing 33% TiO₂.

Conclusions

PAN/TiO₂ electrospun fibers using uniaxial and coaxial methods were investigated as self-detoxifying materials, comparing the detoxification activities of the photocatalyst in different fiber morphologies. It was demonstrated that the coaxial approach resulted in a bi-component core-sheath fiber structure with TiO₂ particles located in the sheath, while overall distribution of TiO₂ particles was obtained for uniaxial fiber. Photocatalytic activity of the fibers under UV irradiations exhibited degradation of aldicarb. The TiO₂ nanoparticle surface functions as a photocatalyst producing OH^· radicals that are strong oxidizing agents,⁴^–¹² resulting in degradation of aldicarb, as shown in Figure 6.¹⁹ In terms of degradation kinetics, higher distribution density of TiO₂ particles in the sheath region resulted in a higher initial degradation rate. According to the results of this study, a larger amount and closer distribution of catalytic powders to the fiber surface reduce the detoxification time. Increasing surface areas of fibers by reducing the diameter or creating pores and channels are options that can be used to increase the rate and amount of photodegradation. In addition, a surface etching process has been shown to increase degradation toxins on fibers containing metal oxides.²⁰ The effective degradation activity of the coaxial electrospun fibers showed potential for application of these fibers for self-decontaminating materials, such as protective clothing and filter media.

Footnotes

Funding

This work was funded though grants from the National Textile Center and partly by the College of Human Ecology and the American Association of Textile Chemists and Colorists. This work was performed in part at the Cornell Center for Materials Research Shared Experimental Facilities, supported through the NSF MRSEC program (DMR-0079992).

References

Formenti

Juillet

Teichner

. Photo-oxydation Ménagée de Parafffines et Oléfines sur Anatase à Température Ambiante. Cr Acad Sci C Chim 1970; 270: 138–138.

Childs

Ollis

. Is photocatalysis catalytic. J Catal 1980; 66: 383–390.

Blake DM. Bibliography of work on heterogeneous photocatalytic removal of hazardous compounds from water and air. NREL/TP-430-22197, 1997. Golden, CO: National Renewable Energy Laboratory.

Fujishima

Honda

. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972; 238: 37–38.

Linsebigler

Yates

. Photocatalysis on TiO₂ surfaces: principles, mechanisms, and selected results. Chem Rev 1995; 95: 735–758.

Liqiang

Xiaojun

Baifu

. The preparation and characterization of La doped TiO₂ nanoparticles and their photocatalytic activity. J Solid State Chem 2004; 177: 3375–3382.

Lee

. Low-temperature fabrication and photocatalytic activity of clustered TiO₂ particles formed on glass fibers. J Cryst Growth 2005; 280: 612–619.

Shchukin

Ustinovich

Sviridov

. Photocatalytic microreactors based on TiO₂-modified polyelectrolyte multilayer capsules. Photochem Photobiol Sci 2003; 2: 975–977.

Masuda

Sugiyama

Seo

. Deposition mechanism of anatase TiO₂ on self-assembled monolayers from an aqueous solution. Chem Mater 2003; 15: 2469–2476.

10.

Baia

Peter

Cosoveanu

. Synthesis and nanostructural characterization of TiO₂ aerogels for photovoltaic devices. Thin Solid Films 2006; 511/512: 512–516.

11.

Kobayashi

Hanabusa

Suzuki

. Preparation of TiO₂ fiber in a sol–gel system containing organogelator. Chem Lett 1999; 10: 1077–1077.

12.

Lee

Kim

Choi

S-S

. Preparation of SiO₂/TiO₂ composite fibers by sol-gel reaction and electrospinning. Mater Lett 2007; 61: 889–893.

13.

Xia

. Direct fabrication of composite and ceramic hollow nanofibers by electrospinning. Nano Lett 2004; 4: 933–938.

14.

Loscertales

Barrero

Marquez

. Electrically forced coaxial nanojets for one-step hollow nanofiber design. J Am Chem Soc 2004; 126: 5376–5376.

15.

Sun

Zussman

Yarin

. Compound core–shell polymer nanofibers by co-electrospinning. Adv Mater 2003; 15: 1929–1929.

16.

Fridrikh

Rutledge

. Production of submicron diameter fibers from difficult-to-process materials by two-fluid electrospinning. Adv Mater 2004; 16: 1562–1562.

17.

McCann

Xia

. Electrospinning of nanofibers with core-sheath, hollow, or porous structures. J. Mater Chem 2005; 15: 735–738.

18.

Kalra

Lee

Park

. Confined assembly of asymmetric block copolymer nanofibers via multi-axial jet electrospinning. Small 2009; 5: 2323–2323.

19.

Dixit

Tewari

Obendorf

. Fungal growth inhibition of regenerated cellulose nanofibrous membranes containing Quillaja saponin. J Chromatogr A 2009; 1216: 6394–6394.

20.

Lange

Obendorf

. Effect of plasma etching on destructive adsorption properties of polypropylene fibers containing magnesium oxide nanoparticles. Arch Environ Contam Toxicol 2012; 62: 185–194.