Introduction

Biodegradable and organic components of construction waste such as timber, bamboo, plastic and packaging waste are considered as non-inert substances that require landfill disposal (HK EPD 2010; Hossain et al. 2016). These non-inert materials accounted for about 23% of construction waste in Hong Kong (Lu and Yuan 2010), of which 346 t of timber waste was daily disposed of at landfills while timber recycling was merely 17 t each day (HK EPD 2015). Landfill disposal of timber waste is undoubtedly a non-sustainable practice that results in depletion of landfill space and substantial emission of greenhouse gases (Tsang et al. 2007; Hattab et al. 2014; Wang et al. 2016a, 2017). Besides, the management of plastic waste also presents a challenging task. About 2015 t of plastic (20.6% of total municipal solid waste) was disposed of daily, while only 4.8% of waste plastic was transported to mainland China for recycling, and there have been increasingly stringent regulations on import/export of unprocessed waste (HK EPD 2015). Therefore, it is vital and highly rewarding to establish recycling alternatives and broaden the market for recycled timber and plastic waste.

Recycling timber waste and plastic waste into wood-plastic composites (WPCs) is potentially an appealing technology because WPCs are sustainable green materials that contain plant fibres and thermoplastics (Ashori 2008a; Qiang et al. 2014). They demonstrate a higher strength-to-weight ratio than concrete as well as much higher strength and stiffness than plastics (Adhikary et al. 2009; Sommerhuber et al. 2016). These biocomposite products also offer advantages of light weight and improved properties of acoustic, impact and heat reformability in comparison to plastic products (Ashori 2008b; Bahari and Krause 2016). In view of the nature of hydrophilic timber and hydrophobic plastic as well as the presence of impurities in waste materials, previous studies employed chemical methods (e.g. use of coupling agents, cross-linking and acetylation of cellulose or grafting) to improve the fibre-polymer compatibility and composite properties by increasing the wetting and interfacial bonding in the matrix (Ashori and Nourbakhsh 2009; Cerbu and Motoc 2010; Franco-Marquès et al. 2011). Maleic anhydride-grafted polypropylene (MAPP) and maleic anhydride-grafted polyethylene (MAPE) were the most commonly used coupling agents to improve adhesion between polar wood flour and non-polar thermoplastics such as polypropylene (PP).

However, the durability of WPCs is one of the major concerns for their applications in construction field (El-Haggar and Kamel 2011), which may be susceptible to biological attack under outdoor conditions. For example, bacterial colonization and subsequent infection of the wood content of WPCs accelerated the deterioration in material properties (Bazant et al. 2014). Fungal attack caused a slight mass loss but severe compromise of mechanical performance, especially for high-wood-content WPCs (Sudára et al. 2013; Catto et al. 2016). In addition, algal decay damaged the surface and the bulk properties of WPCs, which also could cause disease outbreaks and economic losses (Kositchaiyong et al. 2013). The addition of anti-microbial agents could protect the WPCs against biological deterioration, but previous studies often investigated the performance of anti-microbial agents against a single type of microbes. In the natural environment, the effectiveness of anti-microbial agents for different biological deteriorations should be considered in parallel.

Moreover, the use of solvents may degrade crystalline structure and swell amorphous cellulose, which significantly increased the water absorption and thickness swelling (Behzad et al. 2012). The use of coating agents may be preferred, of which melamine formaldehyde (MF) resin showed high adhesiveness, hardness and scratch resistance on the surface of wood materials while providing excellent temperature and chemical resistance (Kim and Kim 2006; Kandelbauer and Widsten 2009). Although MF presents a certain degree of protection against bacteria and fungi, surface impurities on the WPCs still allow microbial growth. Therefore, various types of anti-microbial resistance agents (e.g. single or binary, inorganic or organic) can be incorporated into MF coating to produce synergistic protection via different mechanisms, for instance, disrupting cell wall, rupturing cell membrane, interfering respiratory enzymes or hindering DNA replication (Pathak and Gopal 2012; Zhou et al. 2012; Simonin and Richaume 2015).

To enhance the anti-microbial properties of recycled WPCs, this study aimed to (i) upcycle wood waste and plastic waste at varying mass ratios to produce WPCs and (ii) evaluate the effectiveness of MF coating and various anti-microbial agents against bacterial, fungal and algal decay. The mechanical properties of recycled WPCs were evaluated against standard requirements, and the anti-microbial performance was assessed in terms of bacterial, fungal and algal resistance.

Experimental methods

Waste properties and WPC production

Waste wood formworks (Masson pine) were collected from construction sites with the assistance of a local wood recycling company in Hong Kong Eco Park. The waste woods adhered with cement mortar were shredded into wood flour after removal of nails. The wood flour was sieved to <600 μm and dried at 60 °C to contain less than 2% moisture content before WPC production. Waste plastic was also collected from a local plastic recycling company in Hong Kong Eco Park (recycled white colour polypropylene (PP) with a melting flow rate of 7.5 g per 10 min at 230 °C under 2.16 kg according to ISO 1133 standard (2011) and a density of 0.930 g cm−3), where plastic waste was sorted by types, granulated and sieved to <4 mm. The WPCs were manufactured by using a co-rotating twin screw extruder (SHJ-20B, GIANT, Nanjing). The detailed manufacturing process was described in Supplementary Information (S1).

To determine the suitable wood-to-plastic ratio (W/P) of WPCs for fulfilling the standard requirement (GB/T 24137 2009), four W/P ratios at 3:7, 4:6, 5:5 and 6:4 by weight were tested (Table 1). The compatibility between wood and plastic was improved by using MAPP and MAPE as coupling agents, of which the effectiveness depended on its relative amount to wood, as the surface area of the wood flour increased with the wood ratio in the WPCs. The dosage of coupling agents was 8% by weight of wood content in WPCs with a W/P ratio of 5:5 in this study. Each composite formulation was tested in duplicates (Table 1).

Table 1 Anti-microbial agents for wood-plastic composites
Full size table

Physical properties and thermogravimetric analysis

Flexural strength and tensile strength of the WPCs were evaluated according to ASTM D1037 (2012). The fracture energy (G F) of samples was calculated from the flexural stress-deflection curve. The dimensional stability of the WPCs was examined in terms of thickness swelling and water absorption (ASTM D1037 2012). Besides, thermal conductivity of the WPCs was determined by Quick Thermal Conductivity Meter (QTM-500) in this study. In addition, the chemical components and thermal stability were evaluated by conducting thermogravimetric (TG) and derivative thermogravimetric (DTG) analysis of the WPCs from 100 to 600 °C at 10 °C min−1 with dry argon as stripping gas (Netzch TGA/DSC).

Microbial strains and anti-microbial agents

Three types of microorganisms (bacteria, fungi and algae) were employed to assess the anti-microbial performance of the WPCs. The freeze-dried purified bacteria Escherichia coli (E. coli) and fungus Coriolus versicolor (L. ex Fr.) Quel in slant medium were obtained from Micro-Culture Collection Center of Guangdong Institute of Microbiology, China. The media of bacteria and fungi were tryptone soya broth/agar and potato dextrose agar (PDA), respectively. Testing algae Chlorella vulgaris and the special nutrition medium (Blue Green-11) were purchased from the Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, China.

The WPCs with a W/P ratio of 5:5 (without coupling agents) were selected for the anti-microbial study. Two types of anti-microbial agents, i.e. 55% MF aqueous solution (Jining Huakai resin Co., Ltd., China) and 35% poly-diallyl-dimethyl-ammonium chloride aqueous solution (PolyDADMAC, Sigma-Aldrich), were coated onto the WPCs, respectively, in order to provide effective adhesion to the surfaces. The MF solution was smeared on the surface of the WPCs, which were then placed in the oven at 110 °C for 20 min for the solidification of MF, while PolyDADMAC smeared samples were dried at room temperature for 24 h. In addition, MF was investigated as an adhesive material for incorporation of binary anti-microbial agents by dispersing varying dosages of silver nitrate, copper nitrate, cetyltrimethylammonium bromide (CTAB, Sigma-Aldrich) and PolyDADMAC. The formulations are summarized in Table 1.

Anti-microbial tests

Anti-bacterial activity

The anti-bacterial activity of WPCs was measured in accordance with ISO standard (ISO 22196 2011). After preculturing, the E. coli suspension was used as the test inoculum where the concentration was between 2.5 × 105 and 10 × 105 cells per millilitre. Each test specimen (30 × 30 × 2 mm) was placed into a separate sterile Petri dish with the test surface up, and 0.4 mL of the test inoculum was dripped onto the test surface. A piece of film covered the test inoculum to smooth the surface. Petri dishes were incubated at a temperature of 35 ± 1 °C and a relative humidity of 90% for 24 ± 1 h. An amount of 50 μL suspension was recovered from the Petri dishes and smeared onto the tryptone soya agar. Viable E. coli was enumerated by performing tenfold serial dilutions of the tryptone soya broth in normal saline. The bactericidal rate (R) of every sample was calculated according to the following equation (Li et al. 2016):

$$ R=\left(\frac{B- A}{B}\right)\times 100 $$
(1)

where A was the number of colonies of the tested samples (CFU per mL) and B was the number of colonies of the controlled samples (CFU per mL).

Anti-fungal activity

The inoculation of WPCs specimens was carried out according to the standard procedure (BS EN 113 1997). The purified fungi were transferred to Petri dishes containing PDA, which were kept in the incubator at 28 °C for 1 week until the culture medium was fully covered by the fungi. The cultured fungi were then transferred into Kolle flasks containing the culture medium, which was also incubated for 1 week at 28 °C. To prevent direct contact of the specimens (120 × 30 × 15 mm) with the culture medium, the specimens were mounted over two 3-mm platforms. The dishes containing the fungi and the specimens were stored in the incubator for 12 weeks at 28 °C and a relative humidity of 75%. At the end of fungi exposure, the specimens were carefully removed from the dishes and immersed in the 25% hydrogen peroxide solution for sterilization. The mycelium on the surface of WPCs was brushed off with a soft sponge before further analysis. Weight loss and flexural strength reduction were measured after exposure to fungi.

Anti-algal activity

The anti-algal activity was performed by observing the inactivating area of the testing algae around the specimens by using the agar overlay technique (Kositchaiyong et al. 2013). Initial C. vulgaris at 5 × 107 cells per millilitre was prepared in BG-11 soft agar (0.7% agar) and poured into a 9-cm-diameter Petri dish. A square test specimen (30 × 30 × 2 mm) was gently placed on the soft agar plate at the centre of the Petri dish under aseptic environment. Incubation was performed under a dark-light cycle of 12 h each for 28 days at 25 °C under fluorescent light with an illuminance of 2000 lx. Afterwards, the clear zones (i.e. growth inhibition) around the test specimens were photographically recorded and quantitatively analysed in terms of inhibition area of algae growth in square millimetre, as defined by the following equation:

$$ A=\pi {r}^2-{d}^2 $$
(2)

where A was the inhibition area of algae growth (mm2), r was the radius of the clear zone (mm), and d was the width of the test samples (mm).

Results and discussion

Recycled WPCs from waste timber and waste plastic

The flexural stress-deflection curves (Fig. 1) showed that the calculated fracture energy (G F) substantially decreased from 143.31 to 23.5 N mm−1 with an increase of wood content from 30 to 60%. However, the secant modulus of stress-deflection curve (slope between 0 and 30% of the peak stress) was promoted by increasing the wood content. This agreed with the rule of mixtures of composite materials because the bonding strength was mainly provided by plastic components and the wood flour possessed a greater stiffness compared to plastic (Afrifah et al. 2010). Unlike samples with a high content of PP, samples with 60% wood presented a relatively gentle decrease of flexural strength rather than linear and sharp decrease upon fracture point. This may imply the role of wood flour as fibres that elongate under stress. The addition of coupling agents significantly enhanced the fracture energy of the WPCs from 34.1 N mm−1 (without coupling agent) to 52.4 N mm−1 (MAPP) and 62.9 N mm−1 (MAPE). This was because anhydride groups of MAPP and MAPE could undergo an esterification reaction with wood fibres at the surface hydroxyl groups and polar sites of carbohydrates, which formed bridges of chemical bonds and promoted non-polar polymer adhesion (Ayrilmis et al. 2012; Migneault et al. 2015). Interestingly, compared with MAPP, the addition of MAPE enhanced the material toughness but compromised the stiffness of WPCs as shown by the smaller secant modulus in the stress-deflection curve. As shown in Fig. S1, the samples with 50% recycled wood successfully fulfilled the flexural strength requirement of 20 MPa as stipulated (GB/T 24137 2009). Thus, a W/P ratio of 5:5 (without coupling agents) was selected for the subsequent anti-microbial study. Besides, the products (Table S1) complied with the requirement of thickness swelling (0.5%), water absorption (2%) (GB/T 24137 2009) and qualified as lightweight and thermal-insulating materials according to the British Standard (BS EN 13986 2004).

Fig. 1
figure 1

Flexural stress-deflection curves of wood-plastic composites

Full size image

The thermogravimetric analysis of WPCs revealed a substantial weight loss at 270–370 °C (Fig. 2), where the magnitude was positively correlated with the wood content and it was indicative of wood decomposition (Wang et al. 2016b). The subsequent weight loss at 430–510 °C was attributed to polymer decomposition, and the addition of MAPP and MAPE showed negligible influence on the decomposition temperature (or the associated fire resistance). The weight loss in the temperature region of 430–510 °C was 71.4, 60.5 and 52.3% of total weight loss of the WPCs, corresponding to the PP content of 70, 60 and 50%, respectively. This indicated that plastic was completely decomposed at a temperature of 510 °C or above.

Fig. 2
figure 2

TG (a) and DTG (b) of wood-plastic composites

Full size image

Anti-bacterial properties

As shown in Fig. 3a, the bactericidal rate increased with increasing dosage of MF coating on the WPCs, where the minimum inhibitory concentration (MIC) of MF was 20 g m−2 while the MIC of PolyDADMAC was 15 g m−2. The anti-bacterial activity of MF resin was related to the high emission rate of formaldehyde and volatile organic compounds (Kim and Kim 2006). On the other hand, PolyDADMAC was a cationic polymer biocide that physically disrupted the prokaryotic cell wall. The active quaternary ammonium compounds on the polymer could rapidly bind to the cellular envelope and cause the membranes to fragment and leak (McDonnell and Russell 1999). In view of the large molecular size of the PolyDADMAC that was impregnated onto the surface of WPCs, the potential concern of leaching and developing resistant strains was minimized. Considering the high cost of PolyDADMAC and environmental issue of MF (formaldehyde emission), a lower dosage of MF (12.5 g m−2) was used as an adhesive material for incorporating binary inorganic/organic anti-microbial agents.

Fig. 3
figure 3

Bactericidal rate of wood-plastic composites impregnated with various anti-microbial agents. a Coating of single agent. b Inorganic agents with MF coating. c Organic agents with MF coating

Full size image

Copper and silver at a low concentration present excellent anti-bacterial properties, where 0.5 wt% Cu (0.8 × 10−4 mol/g) achieved over 90% bactericidal rate and 0.1 wt% Ag (0.1 × 10−4 mol/g) led to 98% bactericidal rate. The MIC of Cu (1.5 wt%, 2.4 × 10−4 mol/g) was approximately 12 times higher than the MIC of Ag (0.2 wt%, 0.2 × 10−4 mol/g). A couple of mechanisms were involved in the metal-bacteria interactions. Silver ions could inhibit the bacterial growth by blocking its electron transport system, interact with bacterial cell DNA resulting in mutation and cell death, kill bacteria cells by rupturing the cell membrane and cell wall and destruct bacterial cell with free radicals (Akhigbe et al. 2014; Bellissima et al. 2014; Marx and Barillo 2014). As for copper ions, the absorption of Cu by bacteria resulted in the formation of cavities in the bacterial cell wall (Čík et al. 2001; Raffi et al. 2010). Although Ag presented superior anti-bacterial effectiveness, the economic concern may hinder its application at a large scale, as the price of silver compound was about 40 times higher than that of copper.

Figure 3c shows that only 0.2 wt% of PolyDADMAC on the base of MF accomplished over 97% bactericidal rate. In comparison, 0.2–0.6 wt% of CTAB on the base of MF showed no beneficial effect on anti-bacterial activity compared to 12.5 g m−2 MF coating alone. The MIC of CTAB in the MF matrix was found to be 0.8 wt%; thus, CTAB was slightly less effective than PolyDADMAC. Previous studies reported that CTAB may interfere with activities of respiratory enzymes and/or energy production of the bacteria, delaying bacterial growth as a result (Nakata et al. 2011; Jin et al. 2015). This study proved that MF was a suitable substrate for impregnating inorganic and organic anti-microbial agents on the surface of WPCs, and the binary agents presented more effective bactericidal performance.

Anti-fungal properties

Fungi were found to significantly compromise the durability of WPCs, especially for C. versicolor (white-rot fungus). The effectiveness of the previously mentioned anti-microbial agents against C. versicolor was investigated in terms of weight loss and strength reduction. As shown in Fig. 4a, the weight loss of the untreated WPCs was approximately 1.67% after 12 weeks, while the weight loss of WPCs impregnated with different anti-bacterial agents varied from 0.16 to 1.83%. It was important to note that MF coating by itself barely influenced fungal growth, while PolyDADMAC coating at a relatively low dosage of 15 g cm−2 effectively inhibited the fungal growth and resulted in the lowest weight loss. Besides, the weight loss of WPCs impregnated with Ag and CTAB on the base of 12.5 g cm−2 MF was apparently lower than the untreated one, which validated their anti-fungal activity. In contrast, Cu and PolyDADMAC in MF coating were ineffective against fungal decay, although they were found to be anti-bacterial (Fig. 3). Moreover, Fig. 4b reveals that the application of PolyDADMAC coating only (15 g cm−2) did not cause any strength reduction, while Ag (0.1 wt%) and CTAB (0.8 wt%) in MF coating also demonstrated good resistance to strength reduction, which was only 1.45 and 0.70%, respectively. This showed a notable improvement from the strength reduction of 4.81% in the untreated WPCs subject to fungal attack.

Fig. 4
figure 4

Mass loss (a) and strength reduction (b) of anti-microbial agents impregnated wood-plastic composites after 12-week exposure to white-rot fungus

Full size image

The anti-fungal mechanism of PolyDADMAC on its own was similar to its anti-bacterial function (McDonnell and Russell 1999), while its large molecular size and high charge density appeared to impart superior anti-fungal performance in this study. Silver may kill fungal spores by destructing the membrane integrity and cause damage to DNA and proteins resulting in fungal cell death, as suggested by recent studies on silver nanoparticles (Krishnaraj et al. 2012; Funck et al. 2013). In contrast, the mechanism of anti-fungal action by CTAB did not seem to involve fungus cell lysis but rather the change of surface charge from negative to positive on fungal cells (Vieira and Ribeiro 2006). The WPCs impregnated with the previous three anti-microbial agents were regard as highly resistant to fungal decay, because their weight loss (0.16–0.58%) was significantly lower than those of untreated WPCs (~4.5%) and untreated wood (20–65%) (Silva et al. 2007; Ashori et al. 2013).

Anti-algal properties

Figure 5a–c shows that no clear zone could be identified around the untreated samples, whereas varying extents of light-coloured areas and clear zones were observed around the samples impregnated with various anti-microbial agents. The growth inhibition was related to the anti-algal efficacy of the chemical agents. Among them, PolyDADMAC and Cu in MF coating significantly inhibited the growth of algae, of which the inhibition areas were 1475 mm2 (15 g cm−2 PolyDADMAC only) and 1433 mm2 (1.5 wt% Cu in MF coating), respectively. In comparison, PolyDADMAC, CTAB and Ag in MF coating were less effective for suppressing the algal growth, although the latter two displayed good anti-fungal activity (Fig. 4). Therefore, this study clearly illustrated the importance of performing a holistic analysis of bacterial survival, fungal attack and algal growth, because the results of individual tests could lead to the selection of different anti-microbial agents.

Fig. 5
figure 5

Photographic records of clear zone test for wood-plastic composites impregnated with various anti-microbial agents. a Coating of single agent. b Inorganic agents with MF coating. c Organic agents with MF coating. d Durability of effective agents (14 and 28 days)

Full size image

The positively charged end of PolyDADMAC could attach onto the negatively charged algal cell wall and extend its hydrophobic ligand end to the exterior, rendering the surface of the algal cell hydrophobic and consequently inhibiting the growth of algae (Samarasinghe and Fernando 2015). Neither leaching nor depletion was detected from the PolyDADMAC-impregnated WPCs, which enabled its constant effectiveness in anti-algal performance in the longer term up to 28 days (Fig. 5d). Besides, because its anti-algal mechanism was enacted by hydrophobicity, the use of PolyDADMAC would not develop tolerance in algae.

Copper at high concentrations was toxic for algae growth and development (Han et al. 2008; Zou et al. 2015). The uptake of high content of Cu could severely disturb the synthesis of proteins for energy metabolism and photosynthesis process. As a result, Cu was shown to inhibit the growth and spore release of macroalgae by reducing the photosynthesis rate and chlorophyll concentration, inducing lipid oxidation and altering anti-oxidant enzyme activities (Zhu et al. 2011; Chen et al. 2016). However, the diffusion and leaching of Cu from the WPCs were a concern. As shown in Fig. 5d, the clear zone of Cu-impregnated WPCs became much smaller in the 28-day sample compared to the 3- and 7-day samples (Fig. 5b). Although the anti-algal activity of Cu was evident in the early stage, it was attenuated with increasing time as the concentration of free Cu ions decreased due to diffusion.

Conclusions

This study investigated the feasibility of upcycling timber and plastic wastes into anti-microbial WPCs. The mechanical strength and dimensional stability of WPCs satisfied the standard requirements, and the WPCs presented excellent thermal insulation for construction use. To enhance the durability of the WPCs, single/binary inorganic/organic anti-microbial agents were impregnated on the surface. All the anti-microbial agents illustrated excellent bactericidal effect, although the MIC varied. Nevertheless, the WPCs treated with PolyDADMAC, Ag and CTAB on the base of MF were highly resistant to fungal decay with minimal weight loss and strength reduction, while the other chemical agents were less effective. Besides, all the tested agents demonstrated certain resistance to the algal growth, but the effectiveness of PolyDADMAC and Cu was most distinctive, and the former exhibited more durable growth inhibition. Thus, PolyDADMAC on its own was the most effective anti-microbial agent for all the three types of biological deterioration considered in this study. These findings suggested a value-added engineering solution for waste management and bioeconomy development.