Introduction

Termites are devastating insect pests found worldwide, and they can cause serious damage to buildings, landscape trees, earthen dikes, and dams (Huang et al. 2006; Chouvenc et al. 2011). At present, chemical barriers are still the main method to prevent termites from entering houses (Lin 2015), considering the potential risk of chemical barriers to the environment and human health (Gentz 2009). Thus, it is necessary to search for more environmental friendly techniques for termite control (Benelli 2018). One of the most effective ways to prevent termites from damaging buildings is the use of physical barriers (Verma et al. 2009), which mainly includes sand, metal mesh, and PVC material (Lenz et al. 1997). The long-term and environmental friendly features of physical barriers are garnering increasing attention in termite control (Hu et al. 2006).

Diatomaceous earth (DE) is a natural insecticide widely used for the control of stored grain and urban pests because of its chemical stability and low toxicity (Quarles 1992; Al-Ghouti et al. 2003; Korunic 2013). Upon contact with DE, the wax layers of the epicuticle of insects are destroyed, leading to water loss and the death of the insects (Korunic 1998; Mewis and Ulrichs 1999; Mvumi et al. 2006). Previous studies have shown the significant effect of DE formulations in controlling storage grain pests (Kavallieratos et al. 2015). For example, after wheat and maize seeds were treated with 0.75% DE products (PyriSec® and Protect-It®) for 7 days, the mortality of the lesser grain borer Rhyzopertha dominica (F.) adults in seeds reached up to 90% (Athanassiou et al. 2007a). Similarly, Athanassiou et al. (2006) reported that the mortality of the larger grain borer Prostephanus truncatus (Horn) and R. dominica were 100% after wheat and maize were treated with 75-ppm DE products (DEA) for 14 days. DE products (Protect-It® and Dryacide®) also have significantly toxic effects on adults of the larger grain borer P. truncatus and the cowpea weevil Callosobruchus maculatus (F.) (Stathers et al. 2004). In addition, DEs can not only effectively cause mortality of pest adults but also suppress the progeny production (Athanassiou et al. 2007b; Athanassiou et al. 2008; Kavallieratos et al. 2018). However, there have been no reports about the application of DE in termite control.

The subterranean termite R. chinensis is widely distributed in China, especially in the Yangtze River basin as well as other places in North China (such as Beijing, Tianjin, and Shanxi) (Wei et al. 2007). The termite R. chinensis mainly threatens housing construction and forest trees (Huang et al. 2013). Currently, the control methods of R. chinensis in house mainly include spraying insecticide, spraying powders, and dropping baits (Li et al. 2010; Wei et al. 2010). Our objective of this study is to investigate how DE affects the mortality, penetrating behavior, tunneling behavior, and body surface characteristics of the subterranean termite R. chinensis.

Materials and methods

Diatomaceous earth

DE was purchased from Lingshou County of Hebei Province, China. The particle size of the DE ranged from 25 to 45 μm and comprised 99% amorphous SiO2. The DE was dried at 105 °C for 24 h, and then stored in a desiccator for the following experiments (Tsai et al. 2006).

Experimental termites

Four colonies of the subterranean termite R. chinensis were collected in Shizi Hill, Wuhan City, Hubei Province, China. These colonies were taken back to the laboratory and placed into a plastic box (40 × 20 × 20 cm) containing damp vermiculite. Experimental termites were reared at 25 ± 1 °C and 85 ± 5% relative humidity (RH) in the dark. Healthy adult workers were selected for subsequent experiments.

Determination of termite penetration behavior in dry DE of different thicknesses

Penetrations of termite through DE of different thicknesses were evaluated using an improved method as described by Gahlhoff and Koehler (2001). A total of 30 workers were placed in the bottom of a 9-cm-diameter plastic beaker, completely covered with moist vermiculite. Then, 1-, 2-, or 3-mm layers of dry DE were uniformly laid on the vermiculite layers (Fig. 1a). The same treatments for dry sand were set as a control (Fig. 1b). There were four replicates from four colonies for this part. The experimental setup was kept in the dark 25 ± 1 °C and 85 ± 5% RH. The number of emerging termites was recorded 1, 3, 6, and 24 h after the experiment began.

Fig. 1
figure 1

Effect of different DE thickness levels on termite penetrating behavior. a Dry DE. b Dry sand

Full size image

Effect of DE with different moisture contents on termite mortality

DE with different moisture contents were calculated as the following formula: moisture contents (%) = weight of water ÷ weight of dry DE × 100% (Klute 1986). The 3-mm layers of DE with different moisture contents of dry, 10, and 25% (w:w) were laid into 9-cm-diameter Petri dishes before the experiment began, and then, 30 workers were placed in each Petri dish. The same treatments for 3-mm layers of sand were set as a control. There were four replicates from four colonies for this part. Deaths (immobile and unresponsive workers) were recorded 2 and 6 h after the experiment began. All other conditions were the same as for the experiment above. Mortality (%) was calculated as the following formula:

Mortality (%) = \( \frac{\mathrm{Number}\ \mathrm{of}\ \mathrm{dead}\ \mathrm{termites}}{\mathrm{Number}\ \mathrm{of}\ \mathrm{total}\ \mathrm{termites}} \) × 100%.

Effect of DE with different moisture contents on the tunneling behavior of termites

Tunneling behavior of termites through DE of different moisture contents were evaluated in testing device as described by Puche and Su (2001) with a few modifications. The testing device for the termite tunneling channel was made with a 2-mm-thick Plexiglas plate (Fig. 2). The device includes the termite source area (15 × 15 × 15 cm), tunneling area (30 × 8 × 0.15 cm), and food area (8 × 8 × 8 cm). Before the test, the 1.5-mm layers of DE of different moisture contents of 10, 25, and 50% (w:w) were laid on the tunneling area, and small pieces of wood were placed in the food area. A total of 100 workers were placed into the termite source area of the tunneling device. The same treatments for the 1.5-mm layers of sand were set as a control. There were four replicates from four colonies for this part. Tunneling distances of the termites were recorded 2, 6, 10, and 12 h after the experiment began. All other conditions were the same as for the experiments above.

Fig. 2
figure 2

Tunneling device of termites

Full size image

Scanning electron microscopic observation of termites treated with dry DE

A total of 30 workers were placed into 9-cm-diameter Petri dishes with 3-mm layers of dry DE. After exposure to DE for 1, 3, and 6 h, the workers were then placed in 2.5% glutaraldehyde solution for storage and fixing. The same treatments for 3-mm layers of sand were set as a control. There were four replicates from four colonies for this part. All workers used for scanning electron microscopic (SEM) observation were dehydrated in an ethanol series followed by critical point drying. The workers were mounted on stubs with conductive carbon adhesive tape followed by sputter coating with platinum. The samples were imaged using JEOL JSM-6390 LV SEM (JSM-6390, JEOL, Japan). All other conditions were the same as for the experiments above.

Statistical methods

All data were analyzed using one-way ANOVA by PASW Statistics 18.0 (IBM-SPSS Inc., Chicago, IL, USA), and significant differences were analyzed using Tukey’s HSD test (P < 0.05).

Results

Determination of termite penetration behavior in dry DE of different thicknesses

The number of emerging workers was determined 1 h after workers were placed in the bottom.

Workers could penetrate the 1-mm layers of dry DE and the three treatments with dry sand, but they could not penetrate the 2- and 3-mm layers of dry DE (Table 1). After 3 h, the average number of workers penetrating the 2- and 3-mm layers of dry DE was 1.5 and 0, respectively, which was significantly lower than that of the different sand layers (P < 0.05). The number of workers penetrating the 1-mm layers of DE was not significantly different from that of the sand treatments after 6 and 12 h (P > 0.05). However, the numbers of workers penetrating the 2- and 3-mm layers of DE were both significantly lower than that of the 1-mm layers of DE after 6 and 12 h (P < 0.05). During the whole experimental period, workers never penetrated the 3-mm layers of DE.

Table 1 Total number of emerging workers penetrating DE and sand layers of different thicknesses
Full size table

Effect of DE with different moisture contents on termite mortality

After the treatment with dry DE for 2 h, the mortality of worker reached 19.17%, which was significantly higher than the treatments with DE of 10 and 25% moisture content (P < 0.05) and the three sand treatments (P < 0.05) (Table 2). After 6 h, the mortality of workers were 100% in the treatment with dry DE, which was significantly higher than that of the DE with 10 and 25% moisture content (P < 0.05) and three sand treatments (P < 0.05).

Table 2 Effect of DE with different moisture contents on termite mortality
Full size table

Effect of DE with different moisture contents on the tunneling behavior of termites

The tunneling distances of workers in DE with 10, 25, and 50% moisture content were significantly shorter than those in the sand treatments with the same moisture content for 2 and 6 h (P < 0.05) (Table 3). The average tunneling distances of workers after 10 h in DE with 10, 25, and 50% moisture contents were 3.77, 2.48, and 2.6 cm, respectively, which were significantly shorter than in the three sand treatments (P < 0.05). The average tunneling distances of workers after 24 h in DE with 10, 25, and 50% moisture contents were 4.04, 2.48, and 2.6 cm, respectively, which were also significantly shorter than in the three sand treatments (P < 0.05).

Table 3 Effect of DE with different moisture contents on termite tunneling distance
Full size table

SEM observation of the body surface of termites treated with dry DE and sand

The SEM observation found that the worker body surfaces were impregnated by DE particles while no particles were found in the sand treatment (Fig. 3). After exposure to dry DE for 6 h, the workers all died with conspicuous abdomen shrinkage. However, the workers treated with sands were all alive, and their abdomens had no obvious shrinkage (Fig. 4).

Fig. 3
figure 3

SEM observation of the external surface of termites treated with dry DE and sand. Surface of the workers exposed to DE for 1 h (a), 3 h (b), and 5 h (c). Surface of the workers exposed to dry sand for 1 h (d), 3 h (e), and 5 h (f). DE particles (arrowheads) sticking to the surface of termites

Full size image
Fig. 4
figure 4

SEM observation of the whole body of termites treated with dry DE and sand for 6 h. a The whole body of workers treated with dry DE. b The whole body of workers treated with dry sand

Full size image

Discussion

DE as a physical barrier can prevent pest intrusion (Subramanyam and Roesli 2000). The present study showed that termites could not penetrate 3-mm layers of dry DE because the DE layers suppressed the termite tunneling behavior, and the termites died from the contact with DE during their movement. DE products can be used as grain protectants through grain surface treatments (Vardeman et al. 2007). Mvumi et al. (2006) found that adults of the maize weevil Sitophilus zeamais Motschulsky were able to move through 0.75 m of bulk maize admixed with DE product (Protect-It®) before dying, but most of the insects did not survive. In another test, the R. dominica adults were able to penetrate 30.5 cm of wheat admixed with DE product (Dryacide®), but adult survival and progeny production were significantly lower than in the untreated control (Vardeman et al. 2007). Thus, dry DE can effectively suppress the penetration behavior of the termites and stored-grain insect pests.

DE had a significant effect on the movement of insect pests. Vardeman et al. (2007) reported that the rate of movement of R. dominica adults was significantly slower on wheat treated with DE products (insecto™) than on untreated wheat (control). Mvumi et al. (2006) proved that S. zeamais adults could not migrate back onto maize treated with DE products (Protect-It®). Mohan and Fields (2002) found that DE had a significant effect on suppressing the adult movement of three species of stored-grain insect pests (S. oryzae, Tribolium castaneum (Herbst), and Cryptolestes ferrugineus (Stephens)). Similarly, our results found that DE with three moisture contents (10, 25, and 50%) could significantly suppress the tunneling behavior of R. chinensis (Table 3).

The bioactivity of insects treated with DE decreased with increased moisture content and relative humidity (Korunic et al. 1998). The mortality of the stored-product beetles P. truncatus adults treated with DE products with 15% moisture content (Protect-It™) was significantly lower than in the dry DE treatment (Fields and Korunic 2000). The mortality of the Mediterranean flour moth Ephestia kuehniella Zeller adults treated with DE product (Protect-It®) decreased with increased air humidity (Nielsen 1998). Or results showed that the mortality of termites was 100% after treatment with dry DE for 6 h, which was significantly higher than after treatment with DE with 10 and 25% moisture content, indicating that the moisture content may be closely related to the bioactivity of DE against termites. SEM observation found that a DE product (Insecto®) could attach to the whole body of S. zeamais adults and destroy its epidermis and sensory organs, leading to water lost and death (Malia et al. 2016). Additionally, we found that termites treated with dry DE exhibited obvious abdomen shrinkage and death.

Our study suggests that dry DE has ideal insecticidal activity and a significant preventative effect on the tunneling behavior of termites. During the whole experimental period, the workers of R. chinensis had never penetrated the 3-mm layers of DE, and treatment with dry DE for 6 h resulted in 100% mortality of workers. Thus, DE is expected to develop into an environmentally friendly physical technique for controlling termites inside houses. However, because the moisture level of DE can significantly affect the bioactivity of DE against termites, keeping DE dry after application is the subject of our future research. In addition, there are differences between the laboratory conditions of the performed experiments and field conditions of realistic applications. For example, large numbers of worker termites have big possibility to move through DE layer in the realistic conditions. Thus, the next research work about bioactivity of DE against termites should be performed under the realistic conditions.