**3.1 Soil physicochemical properties**

Soils in all transects were loamy with 15–20% clay at all depths, except transect 1, which had 10–11% clay (**Table 1**). The silt contents ranged from 38 to 44%, while the sand content was 35–45%. The control soil was also loamy with 15% clay, 15% silt and 69% sand. The lower clay content and lower total C (**Table 2**) beyond the 0–10 cm depth of transect 1 could be explained by the removal of OM-rich soil during harvesting of turf grass for sale as an instant lawn.

Soil pH (H2O) ranged from 4.6 to 5.4 at all depths in transect 1 and the control (4.9–5.1) and from 5.3 to 6.7 for all other transects, with pH 6.0–6.7 in the 0–10 cm depth (**Table 2**). Transect 1 and the control had lower pH than all other transects within the 0–30 cm depth, but there were no differences in the 30–40 cm depth. The total C was ≥9% in the 0–30 cm depth in transect 1 and in the top 30 cm of transects 2, 3, 4 and 5 ranged between 13 and 17%, while the control had 2% in the top 30 cm depth (**Table 2**). Transect 1 and the control had lower total C than all other transects, except the 0–10 cm depth, where all sludge-treated soils had similar levels. Total C rapidly decreased with depth beyond the 20–30 cm depth for all transects, with >2.5% at the 40–50 cm depth of transects 2, 3, 4 and 5. The available P was


### **Table 1.**

*Soil particle size distribution of 10–30 cm depths of the different transects (mean ± standard deviation).*


*Heavy Metals in Soils Following 50 Years of Sewage Sludge Application DOI: http://dx.doi.org/10.5772/intechopen.110009*

### **Table 2.**

*Mean pH, total C contents and available P of soils used in the study.*

>37 mg kg−1 in the 0–10 cm depth and declined with depth to 25–32 mg kg−1 in the top 30 cm of sludge-treated soil, except transect 4, where P increased from 30.8 to 39.6. The control had less than 22 mg P kg−1, which did not change with depth, while for the other transects, the levels rapidly declined from the 20–30 cm depth to the 30–40 cm depth.

The higher pH values in transects with higher total C indicated that the sludge had a liming effect, which could affect the mobility of the HMs and the availability of P as seen in **Table 2**. Sewage sludge in South Africa has been found to have pH ranging from 6.4 to 6.7 [24], and 50 years of continuous application of large amounts of such sludge could have limed the soil from pH 5.0 to 6.0–6.7. The available P in the sludge-treated soil could have originated from food additives, dish washing and laundry detergents, personal care products [25], and human urine, which contains about 0.03% P [26]. The highly available P in soils treated with sewage sludge was in agreement with [5], who found that double the recommended rate of sewage sludge application increased the available P over a 4-year period (**Table 3**).

### **3.2 Heavy metal concentrations in soil**

The HM concentrations in the sludge-treated soils than in the control, and their correlation with total C, suggested that the metals originated from the sludge. Variation in metal concentrations across transects treated with sewage sludge could be the result of clay content and slope position and non-uniform loading rates over the years. The increased HM concentration in soils on sewage sludge disposal land, compared with the control, was in agreement with [27]. The low Ni and the similar As between treated and control soil suggested that the long-term sludge application did not increase Ni and As levels, possibly due to low concentrations in the sludge. The high As even in control samples is an indication that local geology contributed significant quantities of As upon weathering of the rocks. The higher concentrations than the MPL of Cr (450 mg/kg), Zn (700 mg/kg), Pb (150 mg/kg) and Cd (5 mg/kg) posed a risk of leaching to ground water, toxicity to soil organisms and accumulation in plant tissue due to uptake. These risks depend on the mobility of these metals, which is affected by soil properties, including soil texture, organic matter content, pH and available P levels.

### *3.2.1 Chromium, zinc and copper*

The strong positive correlations of pH, soil C and available P with Cr, Zn and Cu indicated that the mobility of these heavy metals was significantly affected by all three factors. The higher Cr, Zn and Cu (**Figures 2**–**4**) concentrations in the top 30 cm of transects amended with sludge, which was in agreement with [27], could be a result of sorption and precipitation due to the increase in soil pH. Metal cations are known to favor sorption and precipitation at high pH levels, while at low pH, they become more available [27]. Although soil Cu concentrations were lower than the MPL, transects treated with sludge had significantly higher levels than the control at all depths. The lower pH and Cr, Zn and Cu concentrations in the top 30 cm of transect 1 than the others, and the concentrations of these HM and pH with depth, beyond the 20–30 cm depth, emphasize the importance of sorption of the metals on more negatively charged colloids at high pH. Repeated harvesting of turf grass and the soil attached to the roots explains the lower total C, P and HM especially Cr, Zn and Cu. Gao et al., [7] reported that mobility of Zn and Cu declined with increase in pH from pH 4.5 to 6.5, while Cr in CrO4 3− was not affected in soils enriched with OM. Formation of the hydrous chromium hydroxide (hydroxide of Cr3+), the species to


**Table 3.** *Mean EC of soils used in the study.*

*Heavy Metals in Soils Following 50 Years of Sewage Sludge Application DOI: http://dx.doi.org/10.5772/intechopen.110009*

### **Figure 2.**

*Distribution of total chromium concentrations for all transects. Error bars represent least significant differences (LSD) at p < 0.05. TMT and MPL for Cr are 350 and 450 mg kg−1, respectively [23].*

### **Figure 3.**

*Distribution of total zinc concentrations for all transects. Error bars represent least significant differences (LSD) at p < 0.05. TMT and MPL for Zn are 200 and 700 mg kg−1, respectively [23].*

which it rapidly converts under oxidizing soil conditions, could therefore have limited the mobility [16]. Whereas the effects of pH were evident, Cr and Zn concentrations were higher in transect 2 (>MPL) than in transect 5, which had higher pH, suggesting that other factors also contributed. Sorption to OM and clay minerals [27, 28] and formation of metal–organic matter complexes [27] could have resulted in the accumulation of Cr, Zn and Cu in the surface layers.

The accumulation of Cr in the surface layers (**Figure 2**) of the soil could be explained by complexation of the Cr3+ ion by OM, which was high in these layers, and

### **Figure 4.**

*Distribution of total copper concentrations for all transects. Error bars represent least significant differences (LSD) at p < 0.05. TMT and MPL for Cu are120 and 375 mg kg−1, respectively [23].*

chemisorption by silicate clays [27]. Copper and Zn tend to build up at the surface of contaminated soils with high OM, as a result of complex formation [27]. The lowest soil C and Cr, Zn concentrations in transect 1, while transect 2 had the highest than the other transects, together with the decline of these parameters with depth beyond the 30 cm depth, emphasized the close relationship of soil C with mobility of the metals. The high levels of these metals could have been toxic to soil microorganisms [29, 30], resulting in C sequestration in the soil and minimizing gaseous emissions. Any conditions that cause rapid oxidation of the >10% C in the top 30 cm depth of the soil could contribute significantly to CO2 emissions.

The high available P in soils treated with sewage sludge suggested that the mobility of Cr, Zn and Cu could have been limited through precipitation as metal phosphates [10], leading to the accumulation of these metals in the top 30 cm [10, 12]. The high Cr, Zn and Cu levels in all transects treated with sewage sludge, and their drastic decline at deeper layers beyond the 0–30 cm depth for all transects, were clearly associated with trends in available P levels. For example, transect 2 had the highest available P and soil Cr, Zn (both >MPL) and Cu than the other transects in the top 30 cm. The exception was in the 0–10 cm depth of transect 1, where Cr, Zn and Cu concentrations were equal to, or higher than, those in transects 4 and 5, yetthe available P levels were lower, further suggesting that the available P alone does not fully explain the variations in these metals, emphasizing the role of OM and soil pH.

Although the metals declined beyond the 0–30 cm depth, the concentrations of Cr (except transects 1 and 4), Zn and Cu (except transect 4) remained higher than those of the control to a depth of 40–50 cm, suggesting that significant mobility occurred. Higher levels of organic C and total Cr, Zn and Cu (except transect 4 for Cr and Cu) at 40–50 cm depth could possibly be a result of the formation of mobile metal-soluble organic matter complexes [8, 31]. The displacement of dissolved organic matter from sorption sites by a high level of available P could have enhanced the mobility of the organic matter together with complexed metals to deeper layers. In the 40–50 cm depths of transects treated with sludge, soil C was more than twice that

*Heavy Metals in Soils Following 50 Years of Sewage Sludge Application DOI: http://dx.doi.org/10.5772/intechopen.110009*

in the control, indicating mobility of C from the surface layers. Zhang and Zhang [11] reported that increased P application rates, to soil, resulted in elevated leaching of Zn and Cu, resulting in the formation of soluble Zn-dissolved organic matter complexes. Ashworth and Alloway [13] found that dissolved organic matter enhanced the mobility of Cu but not Zn. The findings were also contrary to [7] and [17].

Lead concentrations in the control ranged from 22 to 40 mg kg−1. The concentrations were higher in soils treated with sewage sludge than those of the control at all depths. Transect 1 had the highest Pb concentration in the 0–10 cm depth, while transect 2 had the highest at all other depths (**Figure 5**). The highest Pb concentration (203 mg kg−1) was in the 30–40 cm depth of transect 2. Whereas Pb remained high, with no clear trend with depth, it declined with depth for transect 1. Lead concentration was above the MPL (150 mg kg−1) in transects 1 (0–10 cm), 2 and 4 (10–20 cm). Nickel concentrations in the control were in the range 7–15 mg kg−1. The concentrations were higher in soils treated with sewage sludge than those of the control at all depths. Nickel accumulated in the top 30 cm and declined beyond that depth in all transects, except for transect 2 (**Figure 6**). Transect 2 had the highest concentration at all depths, with the greatest in the 40–50 cm depth. All transects had lower Ni than the MPL (200 mg kg−1) (**Figure 6**).

The strong positive correlations of Ni, and the weak correlation of Pb, with soil C and available P, and not with soil pH, indicated that the mobility of these heavy metals was only affected by the former two factors. Although metal cations are known to favor sorption and precipitation at high pH levels, while at low pH they become more available [27], there was no clear trend between soil pH and Pb and Ni concentrations in our study. The accumulation of Pb and Ni levels in the surface layers of most transects could be explained by sorption to organic matter and clay minerals [27, 28] and was in agreement with McBride [27]. The highest soil C (similar to transect 4), Ni and Pb concentrations were in transect 2, throughout the depth of the profile, while transect 3 had generally increasing soil C, Ni and Pb within the top 30 cm, and they all declined beyond the 30 cm depth, emphasizing the close relationship of these parameters. In transect 4, soil C declined beyond the 20–30 cm depth, while soil Ni and Pb remained high down to the 30–40 cm depth. These trends emphasized the close relationship between soil C and concentrations of Ni and Ni.

### **Figure 5.**

*Distribution of Pb concentrations for all transects. Error bars represent least significant differences (LSD) at p < 0.05. TMT and MPL for Pb are 100 and 150 mg kg−1, respectively [23].*

### **Figure 6.**

*Distribution of Ni concentrations for all transects. Error bars represent least significant differences (LSD) at p < 0.05. TMT and MPL for total Ni are 150 and 200 mg kg−1, respectively [23].*

The precipitation as metal phosphates [10] could also have contributed to moderating the mobility of Pb and N, which explains the accumulation of these metals in the top 30 cm of most treated transects [10, 12]. The available P and the total Pb and Ni were the highest in transects 1 (0–10 cm), 2 (10–30) and 4 (10–20 cm), which suggested the formation of Pb and Ni phosphate limited mobility. Solubility of Pb has been found to be reduced by the formation of insoluble Pb phosphate minerals [10, 32]. The accumulation of Pb in the surface soil could also have been toxic to soil microorganisms, resulting in sequestration of OM due to limited microbial degradation.

Contrary to the view that Pb tends to build up at the surface of contaminated soils with high organic matter, as a result of strong complex formation, with no downward movement [10, 27], the levels of Pb remained high beyond the 30 cm depth, which suggested that significant mobility had occurred. This was contrary to observations by [11]. The formation of mobile metal-soluble organic matter complexes could have resulted in the higher Pb, Ni and organic C in the 30–40 and 40–50 cm depths [8, 31]. This effect could have been enhanced by the displacement of dissolved organic matter from sorption sites by the high available P, resulting in the formation of soluble metaldissolved organic matter complexes. Zhang and Zhang [11] reported that increased P application rates to soil resulted in elevated leaching of Ni. Beyond 30 cm, soil C, Pb and Ni concentrations were higher than in the control, and in transect 2, Pb concentrations were even higher than in MPL, while Ni continued to increase with depth, suggesting that factors other than soil C and available P also contributed.

### *3.2.2 Cadmium and arsenic*

Cadmium concentration in the control ranged 1–3 mg kg−1. The Cd was higher at all depths in soils treated with sewage sludge than the control (**Figure 7**). The 10–20 cm depth of transect 3 had the highest Cd concentration. Transect 2 had the highest Cd at all other depths. There was no decline of Cd concentration with depth for all transects. The Cd concentrations were at or above the MPL (5 mg kg−1) at all depths in all transects treated with sewage sludge, except the 0–10 cm depth of transect 3.

*Heavy Metals in Soils Following 50 Years of Sewage Sludge Application DOI: http://dx.doi.org/10.5772/intechopen.110009*

### **Figure 7.**

*Profile distribution of total Cd concentrations for all transects. Error bars represent least significant differences (LSD) at p < 0.05. TMT and MPL for Cd are 3 and 5 mg kg−1, respectively [23].*

All the layers of all transects had lower As than the MPL but exceeded the TMT limits in all transects. The control soil had a mean As of 5.6–7.4 mg kg−1. The concentrations of As in transects 1, 2 and 3 at all depths (treated with sewage sludge) were higher than in the control. Transect 1 had the highest concentration in the 0–10 (11.3) and 10–20 (13.6) depth, whereas transect 2 had the highest in the 30–40 (13.9) and 40–50 cm (14.4 mg kg−1) depth. While levels of As in transect 2 increased with depth, the concentrations remained constant in the other transects (**Figure 8**).

Correlation results suggested that soil pH and C, and not the available P, determined Cd mobility. At all depths, transect 1 had the least pH and Cd concentrations of transects treated with sludge, while the others had higher pH and Cd at all depths. The lower pH could be the result of sludge removal during the harvesting of turf grass at this transect. The mobility could have been related to sorption of the Cd2+ ion on the negative soil colloids at higher pH. Gao et al., [7] found that Cd mobility was found to decline with increase in pH from pH 4.5 to 6.5, in soils enriched with organic matter. In this case, the concentration of soil organic matter could therefore have affected the sorption. In the 0–10 cm depth, transects 1, 2 and 4 had the highest soil C and Cd concentrations, while at all other depths, transects 2, 3 and 4 had the highest soil C and Cd concentrations, which emphasized the importance of sorption of Cd to organic matter on mobility of Cd. Although Cd sorption is important, the weak sorption on organic matter, silicate clays and oxides at less than pH 7 [33] could have contributed the high total Cd (>MPL) beyond the 30 cm depth, which was in agreement with [34]. Formation of Cd-soluble organic matter complexes could also have resulted in the mobility and redistribution of the metal throughout the profile with greater risk for ground water pollution. The increased mobility also makes the element more available for plant uptake. The lack of reaction between Cd and P could explain the high concentrations of the metal at deeper layers.

The negative correlation between soil pH and As concentrations suggested that increasing pH increased the mobility of As with no effects of P and OM. Although As concentrations were below the MPL for all the transects, changes in soil pH affected

### **Figure 8.**

*Distribution of total arsenic concentrations for all transects. Error bars represent least significant differences (LSD) at p < 0.05. TMT and MPL for As are 2 and 20 mg kg−1 [23].*

its mobility and accumulation in parts of the soil profile. The relationship with pH was particularly clear where transects treated with sludge had higher As concentrations than the control in part of the profile (transects 1, 2 and 3). For example, in the 0–10, 10–20 and 20–30 cm depths, transect 1 had the lowest pH and the highest As concentration, while transects 4 and 5 had the highest pH and the lowest As (lower than that of the control). Although transects 4 and 5 also had elevated pH in the surface layer, their As concentrations were lower than those of the control, and the effects of pH was therefore not evident. The reason for the lower As in these two transects than the control could be that the As had leached beyond the 40–50 cm depth, possibly into ground water, due to the elevated pH. On the other hand, in transects 2 and 3, As concentrations increased with depth whereas pH decreased. The elevation of pH due to sludge accumulation in transects 2 and 3 could have increased the negative charge on soil colloids, making As, in anionic form, to be more mobile and accumulating at deeper layers, where it gets sorbed on soil colloids with pH-dependent charge. The increased mobility of As could increase its availability to plants.
