Microdistribution of copper-carbonate and iron oxide nanoparticles in treated wood
- First Online:
- Cite this article as:
- Matsunaga, H., Kiguchi, M. & Evans, P.D. J Nanopart Res (2009) 11: 1087. doi:10.1007/s11051-008-9512-y
- 412 Views
Aqueous dispersions of copper-carbonate nanoparticles and microparticles have just begun to be exploited commercially for the preservative treatment of wood. The success of the new systems will depend, in part, on the uniform distribution of the preservative in wood and the ability of copper to penetrate cell walls. We examined the distribution of copper in wood treated with a nano-Cu preservative. Copper particles are not uniformly distributed in treated wood, but they accumulate in voids that act as the flow paths for liquids in wood. Particles are deposited on, but not within cell walls. Nevertheless, elemental copper is present within cell walls, but at a lower level than that in wood treated with a conventional wood preservative. These findings suggest that nano-Cu preservatives are able to deliver bioactive components into wood cell walls even though the majority of copper particles are too large to penetrate the cell wall’s nanocapillary network.
Nanoparticles of copper-carbonate and iron oxide in aqueous systems have just begun to be exploited commercially for the preservative treatment of wood (Leach and Zhang 2004, 2005; Chen et al. 2006). In this application, preservatives containing nanoparticles compete with aqueous wood preservatives that contain dissolved or complexed copper compounds (Preston 2000; Evans 2003). Seventy-nine thousand tons of copper salts and oxides were used in North America in 2004 for the manufacture of wood preservatives (Vlosky 2006). Hence, wood preservatives have the potential to become one of the largest end uses for nanoparticles. The successful exploitation of nanoparticles in wood preservatives will depend on the ability of the new treatments to match the performance of current aqueous systems. Previous studies have demonstrated that the effectiveness of wood preservatives is related to their uniform distribution in wood and capacity of bioactive components such as copper to penetrate cell walls and react with cellulose, hemicellulose and lignin, the polymeric structural components of wood (Arsenault 1973; Hulme and Butcher 1977, Drysdale et al. 1980; Zhang and Kamdem 2000, Matsunaga et al. 2000, 2001, 2002, 2004). Lignified wood cell walls contain a nanocapillary network that is 1–10 nm in size (Kerr and Goring 1975; Flournoy et al. 1991; Fujino and Itoh 1998), but it is not known whether cell walls can be penetrated by nanoparticles. Furthermore, preservatives containing nanoparticles are so new that there have been no previous studies of their microdistribution in treated wood. The possibility exists, however, that their microdistribution will differ from that of wood treated with conventional aqueous preservatives because some of the particles may not be able to penetrate wood cell walls, unlike copper ions in conventional aqueous preservatives.
In this study we compared the microdistribution and deposition characteristics of wood treated with a preservative containing copper-carbonate and iron oxide nanoparticles and microparticles with that of wood treated with the conventional copper-based wood preservative, ACQ (alkaline copper quaternary), which accounted for approximately half of the total amount of copper-based preservatives used in North America in 2004 (Vlosky 2006). Conventional scanning electron microscopy in combination with energy dispersive analysis of X-rays (EDX) was used to examine and map the deposition of copper nanoparticles and microparticles within the porous microstructure of wood. The level of copper in the cell walls of thick-walled latewood fibres was quantified using point analysis of X-rays. High-resolution mapping of particles required the use of a field-emission-scanning electron microscope (FE-SEM) in combination with EDX and drift-correction software. Our findings are discussed in terms of their relevance to the performance and environmental impact of wood treated with preservatives containing nanoparticles.
Materials and methods
Preparation of wood samples
Four air-dried Southern pine (Pinus sp.) boards, 38 mm (thickness) × 140 mm (width) × 1500 mm (length) in size, that had been commercially pressure treated with an aqueous dispersion of copper carbonate and iron oxide particles, hereafter referred to as nano-Cu preservative, were purchased commercially. The sizes of the copper particles in the nano-Cu preservative have a Gaussian distribution from 1 to 25,000 nm, and a mean of 190 nm (Leach and Zhang 2004, 2005). A similar number of Southern pine boards that had been treated with the conventional amine-modified aqueous water-borne wood preservative ACQ were purchased at the same time. A strip measuring 38 × 140 × 10 mm was sawn from the centre of each board. Small specimens measuring 10 × 10 × 1.5 mm, and located 5 or 17 mm from the radial and tangential surfaces of the treated boards were cut from each board. Hence the total number of specimens that was prepared was 16 (4 boards × 2 treatments × 2 specimens per treatment). These specimens were either split using a single-edged razor blade to expose their radial surfaces, or cut in the dry condition using a sledge microtome to reveal their transverse surfaces.
Particle deposition and micro-scale mapping
Uncoated specimens were attached to aluminium stubs using double-sided tape and examined using a Hitachi S-2600N variable pressure SEM (VP-SEM) equipped with a XOne energy dispersive X-ray detector (EDX). Specimens were examined at a pressure of 10 Pa, using an accelerating voltage of 15 kV, an illuminating current of 0.3 nA, a working distance of 15 mm, and take-off and tilt angles of 20º and 0º, respectively. The contrast between copper nanoparticles and wood was greatest when images were formed from backscattered electrons and therefore these were used in preference to secondary electron images when examining the deposition of nanoparticles in the porous structure of wood. A second set of treated specimens was coated with carbon and examined using a high-vacuum Jeol JSM-5600LV SEM equipped with a Jeol JED-2140 EDX. This microscope was used to map the distribution of copper and iron in treated wood samples, and employed an accelerating voltage of 15 kV, an illuminating current of 1.5 nA, a working distance of 20 mm, take off and tilt angles of 30º and 0º, respectively, and dead and dwell (digital map) times of 15–20% and 0.2 ms/point, respectively. The elemental peaks detected were Fe-Kα and Cu-Kα using window widths of 6.33–6.48 keV and 7.96–8.13 keV, respectively.
Specimens containing prepared transverse and radial faces were attached to aluminium stubs and coated with iridium. A Hitachi S4800 FE-SEM equipped with an Apollo EDX system was used to map the distribution of copper at high resolution and differentiate copper from iron particles. Images were viewed in backscattered electron mode and a YAG detector and/or upper detector were used to capture backscattered electrons. The operating conditions were as follows: accelerating voltage, 1, 5 or 15 kV; illuminating current of 0.7 nA; working distance, 15 mm; take off angle, 30º; tilt angle, 0º; dwell time (digital map), 0.2 ms/point; pixels, 256 × 200. The elemental peaks detected were Fe-Lα and Cu-Lα using window widths of 0.650–0.740 keV and 0.880–0.970 keV, respectively. Drift correction system and software were used to help prepare sharp high resolutions maps of copper and iron particles.
Quantitative analysis of copper concentration in wood samples
The concentration of elemental copper in the latewood of wood samples subjected to EDX analyses was subsequently analyzed using inductively coupled plasma (ICP) atomic emission spectroscopy. Latewood samples from the eight boards and two specimens were digested separately in a mixture of HNO3/H2O2 as described by Haraguchi (1992). The resulting solution was filtered through a 0.45-μm disk filter unit. The copper content of the resulting filtrate was analyzed using a Perkin-Elmer Optima 4300DV ICP atomic emission spectrometer operating at a wavelength of 324.752 nm. Copper concentration was expressed as CuO/dry weight of sample (g/kg).
Levels of copper in cell walls of latewood fibres
Statistical analysis of copper in cell walls
Statistical analysis of the levels of copper in the cell wall layers of latewood fibres in wood treated with the nano-Cu preservative or ACQ was performed using analysis of variance for a balanced hierarchical design. The analysis examined the effect of preservative type, treatment depth, cell wall layers and random factors (variation between boards, specimens and cells) on the level of copper in wood cell walls. We also performed a second analysis of variance using concentration of copper in latewood as a covariate, because it is reasonable to assume that levels of copper in the cell walls of latewood fibres, obtained using EDX, will be related to the overall concentration of copper in latewood. Statistical computation was performed using Genstat 5 (Genstat 2000). Before the final analysis, diagnostic checks were performed to determine whether data conformed to the underlying assumptions of analysis of variance, i.e. normality with constant variance. Significant results (p < 0.05) are presented graphically and least significant difference (error) bars (p < 0.05) on graphs can be used to compare differences between individual means.
Results and discussion
Deposition of nanoparticles in the porous microstructure of wood
High-resolution mapping of copper and iron
Point analysis of copper in cell walls of latewood fibres
Clearly, our findings indicated that copper was present in the cell walls of latewood fibres in wood treated with the nano-Cu preservative. Cell walls in wood contain a nanocapillary network that can accommodate silver particles that are 80 nm in size (Rudman 1966). Particles of this size, however, were generated in vivo from ionic silver by impregnating wood with silver nitrate, followed by a second solution that precipitated silver in cell walls (Rudman 1966). Hence, this previous research does not demonstrate that nanoparticles of this size can penetrate wood cell walls. For the latter to occur, they would need to be small enough to pass through the wood cell walls nanocapillary network. Estimates of the size of the capillaries in this network vary from 1.2 nm for wood that has previously been dried to a maximum of 10 nm in wet (never dried) ‘green’ wood (Tarkow et al. 1966; Stone and Scallan 1968; Kerr and Goring 1975; Grethlein 1985; Flournoy et al. 1991). The wood that was analyzed here was dried before preservative treatment and, therefore, we may assume that the sizes of the nanocapillaries were at the lower end of previous estimates. Flournoy et al. (1991) used a series of molecular probes with diameters ranging from 0.4 to 61 nm to determine the accessibility and diameters of the cell wall’s nanocapillary network in wood that had been dried. Their results indicated that molecules larger than 1.2 nm were excluded from 80% of the cell wall. Nevertheless, it is possible that penetration of cell walls by nanoparticles smaller than 1 nm, which was well beyond the resolution of the FE-SEM used here, could account for the presence of some copper in cell walls of treated wood. Alternatively, the copper could have penetrated cell walls as Cu2+ ions, which have a diameter of ~0.087 nm (Martišius and Giraitis 2006). Numerous studies have shown that wood cell walls are accessible to ionic copper, which can then bind to wood’s molecular components (Petty and Preston 1968; Chou et al. 1973; Greaves 1974; Zhang and Kamdem 2000; Petrič et al. 2000; Matsunaga et al. 2004; Cao and Kamdem 2005). Copper carbonate, which was used in the nano-Cu wood preservative here, is insoluble in water. Nanoparticles of copper, however, are more readily solubilised than microparticles (Xia et al. 2006). Hence, it is possible that conversion of copper carbonate nanoparticles into Cu2+ ions could have produced sufficient ionic copper to account for the concentration and distribution of copper in cell walls of treated wood examined here. Further research, however, would be needed to confirm this, and also whether differences in the distribution and availability of copper in nano-Cu treated wood affects its resistance to attack by Basidiomycota, Ascomycota and Fungi Imperfecti, the fungi most commonly responsible for wood decay (Preston et al. 2008; Stirling et al. 2008).
The use of nanoparticles for the preservation of wood is in its infancy and the commercial systems consist of a heterogeneous mix of nanoparticles and microparticles. The size of the particles in the preservatives varies from 1 nm to 25 μm (Leach and Zhang 2004, 2005). The patent literature mentions the desirability of having a more truncated size distribution from 5 nm to 10 μm or even more preferably from 50 nm to 1 μm (Leach and Zhang 2005). The exclusion of large micro-particles (>1 μm) from the preservatives is mentioned as being desirable because they can block flow paths and prevent the penetration and uniform distribution of the preservative in the wood (Leach and Zhang 2005). However, there is little other published information on the effect of particle size and its distribution on the characteristics of the nano-Cu treated wood or its performance. Clearly, there is great potential for further optimization of the systems to improve their efficacy. Reduction in particle size could help cell wall penetration as suggested by our findings, but the presence of larger particles, which could provide a reservoir of active compounds, might also be desirable, because wood preservatives are required to prevent decay for very long periods of time (>30 years). Over such periods of time, there is the potential for wood preservatives to leach into the environment (Waldron et al. 2005), and this has led to restrictions on the use of many wood preservatives (Evans 2003). Concern has also been expressed about the leaching of nanoparticles from materials, and the potential adverse environmental effects resulting from the accumulation of nanoparticles in groundwater or soil, and also their impacts on the health of people who are exposed to them (Borm et al. 2006). Colvin (2003) recommended a proactive approach to understanding the environmental impacts of nanomaterials. This recommendation is particularly germane to wood containing nanoparticles because, as mentioned above, treated wood could become a very significant end use for nanoparticles and, furthermore, there are manifold opportunities during the use of treated wood for the release of nanoparticles into the environment.
Aqueous dispersions of copper carbonate and iron oxide nanoparticles and microparticles impregnated into wood under pressure are deposited in the void spaces, which act as the main flow paths for liquids in wood. The particles were rectangular in shape and varied in size from ~50 to 700 nm. The size of the particles and their propensity to aggregate in pits that connect cellular elements in wood and on cell walls suggested that they were too big to penetrate wood cell walls. Furthermore, we were unable to observe nanoparticles in wood cell walls using an FE-SEM with a resolution of at least ~10–20 nm. Nevertheless, we detected elemental copper in cell walls of latewood fibres in wood treated with the nano-Cu preservative, but its concentration was lower than that in wood treated with the conventional wood preservative ACQ. This difference may be explained by the lower overall concentration of copper in the latewood of wood samples treated with the nano-Cu preservative compared to those treated with ACQ. In both cases, however, the concentration of copper was higher at the interface between cells than in the secondary wall, although the difference was more pronounced in samples treated with ACQ. It is possible that conversion of copper carbonate nanoparticles into ionic copper may explain the presence and micro-distribution of copper in cell walls of wood treated with the nano-Cu preservative. Such a mechanism for the delivery of copper into wood cell walls could be exploited more generally in nanoparticulate wood preservative systems by changing the size of the particles to obtain more controlled release of bioactive agents. This would be desirable because wood preservatives need to remain biologically active for long periods of time, while at the same time minimizing leaching of biocides into the environment.
The authors gratefully acknowledge the expert technical assistance of Derrick Horne (BioImaging Facility, UBC) and Bill Roth (Hitachi High Technologies America, Pleasanton, CA). We are very grateful for the financial support for this research provided by The OECD (Award of a Co-operative Research Program Fellowship to HM), Canadian Foundation for Innovation, BC Knowledge Development Fund and Ministry of Education, Culture, Sports, Science and Technology of Japan (Grant-in-Aid for Scientific Research no. 19780138). We would like to thank Dr. Kazuyuki Oda and Dr. Junji Matsumura for their suggestions and Dr. Yuko Ito for ICP analysis.