Sewage Sludge Ash Lightweight Aggregate
Sewage Sludge Ash Lightweight Aggregate
www.elsevier.com/locate/wasman
Abstract
In this study, sewage sludge ash (SSA), with similar characteristics to expansive clay, was used as the principal material and sewage
sludge (SS) as the admixture to sinter lightweight aggregate and to study the influences of raw material composition on pelletising, sin-
tering effect and aggregate properties. Results showed that both SS and SSA could be sintered to produce synthetic aggregates individ-
ually or mixed. Increasing the amount of SS would decrease the pelletising ratio. Under the consideration of energy saving, the mixture of
SSA was better for sintering normal weight aggregate. On the contrary, the mixture that added 20–30% of SS was more adequate to make
lightweight aggregates. Adding SS would enhance the oxidation–reduction reaction and lower the bulk density and sintering temperature
of aggregates to save energy. Sintering temperature affected the properties of sewage sludge ash lightweight aggregate (SSALA) more
than retention period did. Prolonging the retention period could improve bloating effect.
2006 Elsevier Ltd. All rights reserved.
0956-053X/$ - see front matter 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.wasman.2005.11.024
1454 I.-J. Chiou et al. / Waste Management 26 (2006) 1453–1461
Table 1
Chemical composition of sewage sludge ash (SSA) and expansive clay
Composition (%) SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O SO3 P2O5
SSA 63.31 15.38 6.81 1.80 1.03 0.70 1.51 1.02 7.02
Expansive clay 62.77 16.82 6.97 1.02 2.25 1.68 3.07 – –
Table 2
Chemical composition of SSALA (%)
Aggregate SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O SO3 P2O5
MA, 1150 C 47.92 12.24 12.75 0.96 0.42 0.68 2.48 <0.25 <0.23
MC, 1100 C 52.20 11.07 11.98 0.93 0.37 0.71 2.22 <0.25 1.32
I.-J. Chiou et al. / Waste Management 26 (2006) 1453–1461 1455
ignition loss (p < 0.001), and the raw material mixture was LI ¼ 0:476ðSS=SS þ SSAÞ 0:036t 0:560T ; R ¼ 0:84.
the next (p < 0.01), but the retention period (t) was ð2Þ
insignificant
Table 3
TCLP-results and concentrations of heavy metals in SS, SSA and SSALA
Materials Heavy metals Pb Cd Cr Cu Zn
Raw sludge Total weight (mg/kg) 0.28 ± 0.09 1.60 ± 0.13 9.51 ± 0.11 33.32 ± 1.28 176.57 ± 2.97
SSA Total weight (mg/kg) 1.20 ± 0.013 4.81 ± 0.03 21.96 ± 1.84 89.64 ± 3.77 567.3 ± 4.85
TCLP (mg/L) 0.017 ± 0.006 ND 0.10 ± 0.05 3.81 ± 0.13 –
SSALA TCLP (mg/L) SS/SS + SSA = 0%; 1150 C/10 min 0.088 ± 0.015 ND ND 0.437 ± 0.008 0.116 ± 0.020
TCLP (mg/L) SS/SS + SSA = 20%; 1100 C/10 min 0.098 ± 0.026 ND ND 0.539 ± 0.117 0.402 ± 0.003
Regulatory TCLP (mg/L) 5 1 5 15 –
SSA: sewage sludge ash; SSALA: sewage sludge ash lightweight aggregate; ND: not detected.
Fig. 3. SEM micrographs of SSALA (SS/(SS+SSA) = 0%, 200·): (a) 1050 C/10 min; (b) 1050 C/20 min; (c) 1100 C/10 min; (d) 1100 C/20 min; (e)
1150 C/10 min; (f) 1150 C/20 min.
I.-J. Chiou et al. / Waste Management 26 (2006) 1453–1461 1457
3.3.2. Chemical composition, TCLP and concentrations of and Fig. 6 shows that MD became sintered at 1050 C,
heavy metals and significantly expansive at 1100 C. Thus, it was con-
The chemical composition of sewage sludge ash light- cluded that increasing the amount of SS could effectively
weight aggregate (SSALA) is shown in Table 2. SSALA lower the sintering temperature, enhance the bloating
was mainly composed of SiO2, and Fe2O3, and then effect, and save energy.
Al2O3. The content of P2O5 apparently dropped to 1.32%
or lower during sintering. 3.3.4. Bulk density
TCLP results and concentrations of heavy metals in At 1100 C, with the increase in amount of SS, the bulk
SSALA are listed in Table 3. The SSALA toxicity charac- density of the aggregates dropped significantly. However,
teristic leaching procedure (TCLP) met the Taiwan Envi- when the temperature was increased to 1050 C, due to
ronmental Regulatory requirements. the sintering effect, the bulk density of aggregates only
slightly decreased, as shown in Fig. 7. At 1050 C, the
3.3.3. Sintering effect and bloating effect SSALA of MA and MB still had powdery film on the sur-
Fig. 3 demonstrates that MA became sintered and face. SSALA of MA would have the greatest bulk density
expansive at 1150 C. Fig. 4 showed that MB became sin- at 1100 C. However, the bulk density decreased to 1.35–
tered between 1100 and 1150 C, and its expansion was 1.39 g/cm3 at 1150 C; meanwhile, a glassy texture was
obvious at 1150 C. Fig. 5 indicates that at 1100 C, MC formed on the surface of SSALA, as shown in Fig. 8(a).
achieved the sintered and expansive stage simultaneously, When the temperature increased from 1050 to 1100 C,
Fig. 4. SEM micrographs of SSALA (SS/(SS+SSA) = 10%, 200·): (a) 1050 C/10 min; (b) 1050 C/20 min; (c) 1100 C/10 min; (d) 1100 C/20 min; (e)
1150 C/10 min; (f) 1150 C/20 min.
1458 I.-J. Chiou et al. / Waste Management 26 (2006) 1453–1461
the bulk density of SSALA of MB slightly decreased by of 1050 C/10 min and 1050 C/20 min, the bulk density of
3.4–7.9%. However, when the temperature went up to SSALA was 1.63–1.71 and 1.59–1.62 g/cm3, respectively.
1150 C, the bulk density dropped obviously (18.0– Under the sintering conditions of 1100 C/10 min and
33.3%). When the amount of SS added was more than 1100 C/20 min, respectively, the bulk density of SSALA
20%, such as MC and MD, under the sintering conditions was 0.78–0.82 and 0.76–0.83 g/cm3, which met the light-
Fig. 5. SEM micrographs of SSALA (SS/(SS+SSA) = 20%, 200·): (a) 1050 C/10 min; (b) 1050 C/20 min; (c) 1100 C/10 min; (d) 1100 C/20 min.
Fig. 6. SEM micrographs of SSALA (SS/(SS+SSA) = 30%, 200·): (a) 1050 C/10 min; (b) 1050 C/20 min; (c) 1100 C/10 min; (d) 1100 C/20 min.
I.-J. Chiou et al. / Waste Management 26 (2006) 1453–1461 1459
weight requirement (<1.6 g/cm3), as shown in Fig. 8(b). was the next (p < 0.001), but retention period (t) was
SSALA had a glassy surface and spherical shape. SSALA insignificant.
of MA and MC were dark brown and red brown, respec-
c ¼ 0:7316ðSS=SS þ SSAÞ 0:064t 0:602T ; R ¼ 0:81
tively, as shown in Fig. 8(a) and (b). Besides, at 1150 C,
SSALA of MC and MD would adhere to the kiln, resulting ð3Þ
in additional energy consumption, and would increase the SS supplied more carbide, which helped to process an
bulk density and weaken productivity. oxidation-reduction reaction with ferric oxide, and gener-
Therefore, the sintering conditions required for MA, ated CO2 and CO to enhance the expansion. So, if only
MB, MC and MD to become lightweight aggregates were SSA was used to sinter aggregate, such as MA, the light-
1150 C/10 min, 1150 C/10 min, 1100 C/10 min and weight characteristics were incurred mainly because pyro-
1050 C/10 min, respectively. For energy savings, MA, plasticity happening at high temperature would wrap the
MB, and MC and MD would be more suitable to be sin- gas. On the other hand, the lightweight characteristics of
tered as normal weight aggregate (>1.6 g/cm3), medium mixtures with SS, such as MB, MC and MD, were incurred
density aggregate (1.0–1.6 g/cm3) and low-density aggre- through organic matter decomposition and the oxidation–
gate (<1.0 g/cm3) separately. In addition, Fig. 7 demon- reduction reaction between organic matter and ferric oxide.
strates that the bulk density of SSALA sintered for 10
and 20 min were similar. This indicated that adding SS
3.3.5. Water absorption
would affect sintering temperature more significantly than
Table 4 shows that when the sintering temperature
retention period, and sintering for 10 min was sufficient.
reached 1050 C, the 24 h water absorptions of SSALA
However, increasing the retention period to 20 min would
of MA and MB, and MC and MD ranged from
slightly decrease the density.
30.49% to 32.54% and from 9.08% to 12.46%, respec-
As shown in multiple regression Eq. (3), the raw mate-
tively. At 1100 C, the SSALA water absorption of
rial mixture was the most significant factor affecting the
MA and MB decreased, apparently because micro parti-
bulk density (p < 0.001), and sintering temperature (T)
cles started to tightly link. Meanwhile, a glassy texture
formed on the surface of SSALA, and thus prevented
3.0
its water absorption increase due to porosity. At
1150 C, SSALA of MA and MB have reached the sin-
1050?/10min 1100?/10min
1150?/10min 1050?/20min
tering stage, and made the water absorption decrease
2.4
1100?/20min 1150?/20min
to 1.65–2.84%. As shown in multiple regression Eq. (4),
sintering temperature (T) was the most significant factor
Bulk density (g/cm3)
Fig. 8. Appearance of SSALA (after sintering). (a) (SS/(SS + SSA) = 0%; 1150 C/10 min; (b) (SS/(SS + SSA) = 20%; 1100 C/10 min.
1460 I.-J. Chiou et al. / Waste Management 26 (2006) 1453–1461
Table 4
Performance of SSALA
SS/(SS+SSA) (wt.%) Properties of SSALA 1050 C 1100 C 1150 C
10 min 20 min 10 min 20 min 10 min 20 min
0% (MA) Ignition loss (%) 10.48 9.08 9.03 9.02 1.87 1.90
Bulk density (g/cm3) 1.81 1.82 2.09 2.12 1.39 1.35
Water absorption (%) 32.54 31.28 9.54 9.07 1.65 1.90
Point loading (kgf) 12.29 17.45 45.95 80.94 59.25 53.02
Soundness (MgSO4, %) 3.75 3.59 1.09 1.08 0.19 0.16
10% (MB) Ignition loss (%) 11.61 12.46 11.81 11.64 4.72 4.70
Bulk density (g/cm3) 1.78 1.77 1.72 1.63 1.46 1.18
Water absorption (%) 30.49 31.67 19.52 13.86 2.84 2.55
Point loading (kgf) 12.96 10.74 30.60 33.97 35.10 32.29
Soundness (MgSO4, %) 2.84 2.64 1.13 1.17 0.66 0.48
20% (MC) Ignition loss (%) 30.49 31.67 19.52 13.86 – –
Bulk density (g/cm3) 1.71 1.62 0.78 0.83 – –
Water absorption (%) 10.48 9.08 9.03 9.02 – –
Point loading (kgf) 96.87 95.10 26.28 22.82 – –
Soundness (MgSO4, %) 1.63 1.42 1.11 1.25 – –
30% (MD) Ignition loss (%) 32.54 31.28 9.54 9.07 – –
Bulk density (g/cm3) 1.63 1.59 0.82 0.76 – –
Water absorption (%) 11.61 12.46 11.81 11.64 – –
Point loading (kgf) 82.55 81.07 26.08 24.76 – –
Soundness (MgSO4, %) 1.58 1.56 1.34 1.26 – –
‘‘–’’ means adhering to kiln and cannot be detected.
Mix B
Mix C The failure point loadings of SSALA of MC and MD were
80 Mix D
95.10–96.87 and 81.07–82.55 kgf, respectively, because the
60 sintering reaction has happened at this moment, as shown
in Fig. 3. Fig. 5 also proves that sintering temperature had
40 a more significant effect on aggregate characteristics than
retention period. Besides, 10 min of retention period was
20
sufficient, but prolonging the retention period to 20 min
0 would lower the failure point loading of SSALA.
1050˚C 1100˚C 1150˚C
a 10 minutes 3.3.7. Frost resistance
120
The soundness test is supposed to simulate the effect of
ice crystallization in a saturated aggregate. The weight loss
Mix A
100 rate of aggregates, which were repeatedly soaked in a satu-
Failure point loading (kgf)
Mix B
Mix C
Mix D
rated solution of magnesium sulfate for 5 times, was far
80 lower than the regulatory value of 18%. The result is shown
in Table 4. The aggregates with bulk density lower than
60
1.0 g/cm3 had a weight loss rate between 1.11% and
40 1.34%. As shown in multiple regression Eq. (5), the sinter-
ing temperature (T) was the most significant factor to affect
20 the soundness (p < 0.001), and the raw material mixture
was the next (p < 0.05), but the retention period (t) was
0 1050˚C 1100˚C 1150˚C insignificant.
b 20 minutes
Sd ¼ 0:354ðSS=SS þ SSAÞ 0:037t 0:900T ; R ¼ 0:87
Fig. 9. Failure point loading of SSALA: (a) 10 min; (b) 20 min.
ð5Þ
3
To SSALA with a bulk density lower than 1.0 g/cm , the
was because the SSALA had not reached the sintering tem- weight loss was 1.11–1.34%. This explained that the
perature, and there was still powdery film on the surface. SSALA at an expansive stage still had excellent anti-chem-
The failure point loadings of SSALA of MA and MB were ical decomposition characteristic. At 1050 C, because MA
I.-J. Chiou et al. / Waste Management 26 (2006) 1453–1461 1461