- Email: [email protected]
Effects of long-term soil amendment with sewage sludges on soil humic acid thermal and molecular properties
Chemosphere 73 (2008) 1838–1844
Contents lists available at ScienceDirect
Chemosphere j o u r n a l h o m e p a g e : w w w . e l s e v i e r. c o m / l o c a t e / ch e m o s p h e r e
Effects of long-term soil amendment with sewage sludges on soil humic acid thermal and molecular properties José M. Fernández a,*, William C. Hockaday b, César Plaza a, Alfredo Polo a, Patrick G. Hatcher c a
Centro de Ciencias Medioambientales, Consejo Superior de Investigacio nes Científicas, Serrano 115 dpdo., 28006 Madrid, Spain Department of Earth Science, Rice University 6100 Main Street, Houston, TX 77005, United States c Department of Chemistry and Biochemistry, Old Dominion University 4541 Hampton Boulevard, Norfolk, VA 23529-0126, United States b
a r t i c l e
i n f o
Article history: Received 4 March 2008 Received in revised form 31 July 2008 Accepted 3 August 2008 Available online 20 September 2008 Keywords: Composting Thermal-drying Sewage sludge Humic acids 13 C NMR Thermal analysis
a b s t r a c t Sewage sludges are frequently used as soil amendments due to their high contents of organic matter and nutrients, particularly N and P. However, their effects upon the chemistry of soil humic acids, one of the main components of the soil organic matter, need to be more deeply studied in order to understand the relation between organic matter structure and beneficial soil properties. Two sewage sludges subjected to different types of pre-treatment (composted and thermally dried) with very different chemical compo sitions were applied for three consecutive years to an agricultural soil under long-term field study. Ther mal analysis (TG–DTG–DTA) and solid-state 13C NMR spectroscopy were used to compare molecular and structural properties of humic acids isolated from sewage sludges, and to determine changes in amended soils. Thermally dried sewage sludge humic acids showed an important presence of alkyl and O/N-alkyl compounds (70%) while composted sludge humic acids comprised 50% aromatic and carbonyl carbon. In spite of important differences in the initial chemical and thermal properties of the two types of sewage sludges, the chemical and thermal properties of the soil humic acids were quite similar to one another after 3 years of amendment. Long-term application of both sewage sludges resulted in 80–90% enrich ment in alkyl carbon and organic nitrogen contents of the soil humic acid fraction. © 2008 Elsevier Ltd. All rights reserved.
1. Introduction Animal manure and crop residues have traditionally been applied to soil as a means of maintaining and increasing the organic matter content. However, nowad ays their production can not satisfy the current soil requirements of cropland. This has moti vated the growing interest in agricultural applications of organic residues derived from a wide variety of human activities as a con venient alternative to their landfill disposal. In this context, sewage sludge (SS), the residual solid produced during wastewater treat ment, has been increasingly applied as an organic amendment pro viding substantial gains in soil fertility and crop yield due to its high content of organic matter and nutrients, particularly N and P (Hargreaves et al., 2008 and references therein). However, before being used, as with other organic waste mate rials, SS must be properly processed in order to obtain a stable organic matter. Phytotoxicity and other detrimental effects of the premature application of organic wastes have been extensively reported. Composting is one of the most common treatments that resemble natur al processes in soil, being an accelerated humifi cation process and has been widely studied for many decades * Corresponding author. Tel./fax: +34 91 4115301. E-mail address: jmfernan[email protected] (J. M. Fernández). 0045-6535/$ - see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2008.08.001
(Senesi, 1989; Senesi and Plaza, 2007). Another relatively new treatment method is thermal-drying, which consists of a short and intense heating of the SS, destroying pathogens, eliminating volatile chemicals and leading to a sanitized final product with good handling characteristics but lacking a maturation process (Tarrason et al., 2008; Fernández et al., 2007a). Although this pro cess has been increasingly introduced in municipal wastewater treatments plants, the effects of thermally-dried sewage sludge on soil properties and soil organic matter have been scarcely studied. Although benefic ial effects upon the physical (Navas et al., 1998; Sort and Alcañiz, 1999), chemical (Navas et al., 1998), and biological properties (Banerjee et al., 1997; Wong et al., 1998) have been widely investigated, little information is available on long-term effects induced by SS on soil humic substances, and particularly on humic acids (HAs), one of the most abundant and chemically active fractions of soil organic matter (Senesi, 1989; Reveille et al., 2003; Adani and Tambone, 2005). Due to the impor tant role that HAs play in the soil, information about their proper ties and transformations can be essential to understand the effects of SS on soil organic matter properties and evaluate the suitability of these materials as organic amendments. Among the numerous chemical and spectroscopic methods used to study HA structures and composition, thermal investiga tions are steadily gaining interest. However, most of the thermal
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
studies are still related to characterizations of HAs from soils (Gonet and Cieslewicz, 1998; Dell’Abate et al., 2002) or compost ing materials (Provenzano et al., 1998; Pietro and Paola, 2004), and do not focus on the evolution of HAs in soil organic matter resto ration experiments. The introduction of the use of this technique, which has advantages in terms of time and money, can assume an important role in this kind of study by providing highly valuable information. However, this innovative method still must be com pared with other reliable techniques. The use of solid-state 13C nuclear magnetic reson ance (NMR) spectroscopy has emerged as one of the most useful and powerful tools to obtain structural information of humic substances (Wilson, 1987; Hatcher et al., 2001; Smernik et al., 2003). The combination of cross polarization (CP) and magic angle spinning (MAS) is the most popular 13C NMR technique for the study of humic materials (Conte et al., 1997; Cook and Langford, 1998; Chefetz et al., 2002). However, this technique can under represent some structural units (e.g. fused aromatic moieties, alkyl groups with high degrees of molecular mobility, and carbonyl C) (Dria et al., 2002; Smernik et al., 2003). These studies, in combination with our own previous experiences, have suggested the use of the alternative Bloch decay (BD) technique in the analysis of C-rich HAs. Although this tech nique generally has longer requirements of time and lower signal to noise ratio than CP, BD provides the most accurate quantita tive distributions of structural units in humic acids (Smernik and Oades, 2000; Dria et al., 2002). In this study, solid-state 13C NMR and thermal analyses (TG, DTG and DTA) were performed in order to: (a) determine and com pare molecular and structural properties of the HA isolated from composted sewage sludge (CS) and thermally-dried sewage sludge (TS), and (b) investigate their contributions to the soil HA pool after three years, comparing results obtained with both techniques. 2. Materials and methods 2.1. Sewage sludges, soil and field experiment The CS sample (dry matter content, 435.8 g kg¡1) was col lected from a 3-month windrow-composted mixture of three SS originated from three municipal wastewater treatment plants in Madrid (Spain) metropolitan area. The TS sample (dry matter con tent, 846.4 g kg¡1) consisted of a SS dried by indirect convection with air heated to temperat ures between 380 °C and 450 °C in the wastewater treatment plant “SUR” in Madrid metropolitan area. The maximum temperat ure reached in the sludge during thermal drying was 75 °C. The field experiment was conducted in the experimental farm “La Higuerue la” located in Toledo (Spain). The site is characterized by a continental semiarid climate with an average annual rainfall of about 487 mm and an average annual temperature of 14 °C. The soil is a Calcic Luvisol (FAO-ISRIC, 1998) or Typic Haploxeralf (Soil Survey Staff, 2003), having a sandy loam texture (sand, 590 g kg¡1; silt, 220 g kg¡1; clay, 190 g kg¡1). Mean chemical properties of soil and both sewage sludges are included in Table 1. The experimental design consisted in four random blocks of nonirrigated soil plots (10 £ 3 m2) cropped with barley (Hord eum vulgare L.), either unamended (SO) or amended yearly over a 3-year period at a rate of 80 t ha¡1 with CS (CS80) or TS (TS80). Both SS were applied in mid-September, prior to barley planting in mid-October, and immediately incorporated into soil at a depth of 0–15 cm. The third year, after barley harvesting, in late June, four surface soil subsamples (Ap horizon, 0–15 cm depth) were collected randomly from each plot. Each soil subsample consisted of a mixture of 20 soil cores each of 3-cm diameter. A compos ite sample was then obtained for each treatment by mixing equal amounts (1 kg) of the four corresponding soil subsamples.
1839
Table 1 Main chemical properties (±standard errors) of unamended soil (SO), composted sewage sludge (CS) and thermally-dried sewage sludge (TS) used in the experi ment
pH (H2O) Electrical conductivity (dS m¡1) Total organic carbon (g kg¡1) Total extractable carbon (g kg¡1) Humic acids carbon (g kg¡1) Fulvic acids carbon (g kg¡1) C/N Total N (g kg¡1) P (g kg¡1) K (g kg¡1) a b
SO
CS
TS
5.7 ± 0.1 0.05 ± 0.01 7.2 ± 0.1 1.4 ± 0.2 0.7 ± 0.1 0.6 ± 0.1 8.0 ± 0.1 0.9 ± 0.1 0.09 ± 0.01a 0.20 ± 0.01a
7.1 ± 0.1 3.90 ± 0.01 181.0 ± 0.2 53.4 ± 0.2 21.0 ± 0.3 32.4 ± 0.3 7.6 ± 0.2 23.9 ± 0.1 13.90 ± 0.02b 5.02 ± 0.07b
7.0 ± 0.1 1.50 ± 0.02 296.0 ± 0.2 87.7 ± 0.3 24.8 ± 0.2 62.9 ± 0.3 8.3 ± 0.2 35.6 ± 0.1 13.43 ± 0.02b 4.29 ± 0.05b
Available content. Total content.
2.2. Humic acid isolation The HA fractions were isolated from CS, TS, CS80, TS80, and SO samples by conventional methods. The humic acid C contents of each sample are reported in detail in Fernández et al. (2007b). Prior to extraction of HA, soil and SS samples were air-dried and passed through a 2 mm sieve after removal of plant residues and stones from soils. Carbonates were removed from soil samples by mechanical stirring with 2 M H3PO4 for 30 min. The treatment was repeated three times, and then the samples were washed with dis tilled water until the suspension reached a pH 7. Carbonate-free soils and SS samples were first extracted with 0.1 M Na4P2O7 at pH 9.8 and then by 0.1 M NaOH, at room temperature (RT, about 293 K), and using a sample to extractant ratio of 1:10. Each extraction was repeated three times by mechanically shaking the mixture for 3 h, then centrifuging at 15 300g for 15 min, and filtering the superna tant through a Whatman no. 31 filter paper. The combined alka line supernatants were then acidified with HCl until reaching a pH 1, left standing for 24 h in a refrigerator to allow the complete precipitation of HA, centrifuged at 30 100g for 15 min, and then fil tered through a Whatman no. 31 filter paper. The HA precipitates were then purified by dissolution in 0.1 M NaOH, centrifugation at 30100 g, elimination of the residue, acidification of the alkaline supernatant with HCl–HF until pH 1, and standing for 12 h in a refrigerator. The purification procedure was repeated three times. The precipitated HAs were recovered with distilled water, dialyzed until free of Cl¡ ions, and finally freeze-dried. 2.3. Humic acid characterization The moisture content of HAs was measured by heating over night at 105 °C and ash content by combustion overnight at 550 °C. The C, H, N and S contents were determined in triplicate using a Fisons Instruments (Crawley, UK) elemental analyzer model EA 1108. Oxygen content was calculated by difference on an ash-free dry weight basis as: O% = 100 ¡ (C + H + N + S)%. Thermogravimetry (TG), differential thermogravimetry (DTG) and differential thermal analysis (DTA) were performed simul taneously using a Setaram (Caluire, France) SETSYS thermal ana lyzer equipped with the computer program Setsoft 1.6.4 (Setiram, Caluire, France) for data processing. Ten-milligram HS samples were isothermally heated from room temperature to 30 °C for 10 min and then to 1200 °C in a platinum crucible under a flow of air of 30 mL min¡1. The heating rate was set at 10 °C min¡1. Calcined alumina was used as reference material. Solid-state 13C NMR experiments were performed on a Bruker (Billerica, MA) DMX 300 NMR spectrometer. This spectrometer operates at a 13C frequency of 75.48 MHz. Samples were placed in
1840
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
a 4 mm (outside diameter) NMR rotor with a Kel-F cap (3M, Min neapolis, MN). Bloch decay spectra were acquired using a 20° 13C pulse, a recycle delay of 3 s, and an MAS frequency of 13 kHz. The recycle delay of 3 s was chosen because incrementing the delay to 4 s and 5 s gave no increase in signal intensity. The 3 s delay follow ing a 20° pulse is equivalent to a 150 second recycle delay accom panying a 90° 13C pulse (Ernst and Anderson, 1966). Approximately 22 500 scans were collected for each sample. The Kel-F rotor caps and probe components give rise to a significant 13C signal in the BD spectrum, which overlapped with the NMR signal of the sam ples. To correct for this, the empty-rotor spectrum was acquired separately and was subtracted from each BD spectrum before inte gration. A 40 Hz line broadening was applied when processing the NMR spectra, and peak areas were integrated according to the chemical shift regions defined in Table 4. 3. Results and discussion 3.1. Elemental analysis The elemental composition of SS HA differs according to their source and treatment process. The CS–HA (Table 2) features higher O/C and lower H/C and N/C ratios than TS–HA. These results sug gest that composting favours the formation of a more oxidized HA fraction than thermal drying. This is typical of a more mature organic material (Senesi et al., 1996). Simil ar results for composted and uncomposted SS are reported by other authors (Boyd et al., 1980; Gerasimowicz and Byler, 1985). Relative to SO–HA, CS–HA and TS–HA have lower or much lower C, O contents, and larger or much larger N and S contents. The high levels of N and S in SS–HAs may be ascribed to the incorporation of proteinaceous products and S-containing commercial surfactant residues in the HA frac tion (Senesi et al., 1996). In comparison with untreated SO–HA, the HA fraction of SS– amended soils has undergone partial incorporation of compounds from CS and TS into the soil HA fraction. This is evident by the sig nificant increases in C, H, N, and S content of CS80 and TS80 (Table 2). However, it is interesting to note that the original differences in CS–HA and TS–HA elemental composition are not manif ested in the soil HA fraction after 3 years of amendment. That is, CS80 and TS80 show equival ent C, H, O and S contents. The only difference is the slightly higher N content of TS80. 3.2. Thermal analysis Thermograms of HA samples are shown in Fig. 1 and main ther mal parameters are summarised in Table 3. The HA samples from TS reveal a significantly different thermal behaviour than CS–HA and SO–HA, which are fairly similar to one another. Meanwhile, HA samples from SS amended soils are quite simil ar. The primary fea tures of the differential thermograms are: (a) a small endothermic peak at low temperatures (90.0–101.8 °C), ascribed to reactions of dehydration; (b) an exothermic peak within the 338–341 °C range,
generally attributed to the degradation of polysaccharides, decar boxylation of acidic groups and dehydration of aliphatic alcohols (Flaig et al., 1975; Dell’Abate et al., 2002; Francioso et al., 2005); (c) an exothermic peak between 439 °C and 456 °C, possibly due to the decomposition of N compounds (e.g. proteinaceous material) and long chain hydrocarbons (Francioso et al., 2005); and (d) an exo thermic peak at high temperatures (539–560 °C), usually related to the combustion of aromatic structures and cleavage of C–C bonds (Peuravuori et al., 1999). Exothermic reactions, which determine weight losses at differ ent steps recorded on the TG curves, allow the quantitative com parison of the different fractions involved in the reactions (Table 3). For the first exothermic reaction, CS–HA shows a lower weight loss than TS–HA, which features the most important lost of weight. The DTG curves in this region also indicate rapid combustion of the TS–HA compounds at this range of temperatures in compari son with the other samples. These results suggest the presence of a large proportion of carbohydrates and aliphatic alcohols structures in TS–HA (Ricca et al., 1993). The exothermic peak at 439–456 °C, unique to the HA from SS–amended soils, is slightly higher for the TS80–HA than for CS80–HA. This is consistent with the persistence of proteinaceous and/or long chain N-alkyl surfactant structures present in the TS–HA (8.1% weight). This peak could also arise from N-aromatic compounds (e.g. melanoidins) formed by the condensa tion of reducing sugars and amines during the thermal drying pro cess (Stevenson, 1994). The last exothermic peak features slightly higher loss of weight for CS–HA (45.3%) in comparison with TS–HA (38.2%). However the biggest mass loss (53.1%) corresponds to SO–HA. Further, SO– and CS–HA show a higher decomposition rate than TS–HA. This could be attributed to the presence of aromat ics in CS–HA and SO–HA (Sheppard and Forgeron, 1987; Francioso et al., 2005). This is confirmed by 13C NMR, as discussed in Section 3.3. 3.3. Bloch decay 13C nuclear magnetic resonance analysis The description of the chemical nature of the organic C con tained in HA is more directly conferred by analysis of the BD 13C NMR spectra (Fig. 2). Spectra were divided into four regions and integrated as follows: 0–45 ppm assigned to alkyl C; 45–110 ppm assigned primarily to O-substituted alkyl C in carbohydrates, but also including methoxyl C and N-substituted alkyl C in protein; 110–160 ppm assigned to aryl C, including O-aryl and C- and Hsubstituted aryl C; and 160–220 ppm assigned to carbonyl C in car boxylic acids, ketones, esters, and amides (Table 4). Humic acids extracted from TS and CS show higher content of the alkyl and O/N-alkyl C and lower content of aromatic and car bonyl/amide C than SO–HA. These differences are more marked for 13 C NMR spectra of HA extracted from the TS, which is dominated by signals in the alkyl C region at d < 45 ppm. In particular, signals at d 6 20 ppm indicate methyl C in alkyl chains, and the signal near 25 ppm represents methylene C in alkyl chains. Meanwhile, the strong signal at 33 ppm can be attributed to rigid domains in
Table 2 Moisture, ash content, elemental composition and atomic ratios (± standard error) of humic acids (HA) isolated from the unamended soil (SO), composted sewage sludge (CS), thermally-dried sewage sludge (TS) and soils amended with CS or TS at a rate of 80 t ha¡1 y¡1 for 3 years (CS80 and TS80, respectively) Origin of HA sample
Moisture (g kg¡1)
Ash (g kg¡1)
C (g kg¡1)a
H (g kg¡1)a
N (g kg¡1)a
S (g kg¡1)a
O (g kg¡1)a
C/N (atomic ratios)
H/C (atomic ratios)
O/C (atomic ratios)
SO CS CS80 TS TS80
69.8 35.9 24.9 15.9 40.4
79.2 115.7 61.2 40.5 84.3
505 ± 6 460 ± 1 598 ± 0 503 ± 2 605 ± 0
53 ± 3 53 ± 1 67 ± 0 70 ± 3 72 ± 0
44 ± 1 85 ± 0 90 ± 0 83 ± 0 81 ± 0
4 ± 0 14 ± 0 16 ± 0 13 ± 0 15 ± 0
394 ± 9 389 ± 1 228 ± 0 330 ± 5 227 ± 0
13.3 6.3 7.8 7.1 8.8
1.3 1.4 1.4 1.7 1.4
0.6 0.6 0.3 0.5 0.3
a
On a moisture- and ash-free basis.
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
nant in SO–HA and CS–HA, showing the abundance of aromatic C (e.g. peak at 131 ppm.) and O-substituted aromatic C (e.g. peak at 151 ppm.) (Chefetz et al., 2002). These results agree with elemental
Soil
TG(%)
DTA( µV) 160
0 DTG
exo →
-20 -40
140 120 100 80
-2
60
-60
0
-1
straight alkyl chains, or to branched alkyl chains, such as R–CH and R–CH2 a substituted groups where RCOOH (Ricca et al., 2000). In contrast, the region between 110 and 160 ppm is predomi
1841
DTG(% min )
40 -80
20 DTA TG
-100 0
200
400
600
800
-4
0 -20
o
Temperature( C)
-40
120
60
-60
DTG
-20
100 80
140
0
-40
-2
20 DTA TG
-100 400
600
40 -80
-4
DTA TG
-100
-20
800
0
200
400
600
DTA( µV)
-40
100 80 60
-60
160
0
DTG
-20 -1
120
DTA( µV)
0
DTG(% min )
exo →
-20
140
TS80
TG(%)
160
DTG
-40
-2
20 DTA TG
-100 600 o
Temperature( C)
800
0 -20
140 120 100 80 60
-60
0
-2
40
40 -80
400
-4
-20
800
exo →
CS80
200
0
Temperature( C)
0
0
20
o
o
Temperature( C)
TG(%)
-2
-1
200
80
DTG(% min )
0
0
0
100
60
-60
40 -80
120
-1
exo →
-20
140
160
-1
DTG
DTA( µV)
0
DTG(% min )
0
TS
TG(%)
160
DTG(% min )
DTA( µV)
exo →
CS
TG(%)
-80
20
-4
DTA TG
-100 0
200
400
600
800
0
-4
-20
o
Temperature( C)
Fig. 1. TG, DTG and DTA thermograms of humic acids (HA) isolated from the unamended soil (SO), composted sewage sludge (CS), thermally-dried sewage sludge (TS) and soils amended with CS or TS at a rate of 80 t ha¡1 y¡1 for 3 years (CS80 and TS80, respectively).
1842
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
Table 3 Differential thermal analysis (DTA) peak temperatures and weight losses (%) of humic acids (HA) isolated from the unamended soil (SO), composted sewage sludge (CS), ther mally-dried sewage sludge (TS) and soils amended with CS or TS at a rate of 80 t ha¡1 y¡1 for 3 years (CS80 and TS80, respectively) Origin of HA sample
Peak temperat ure (°C)
Weight loss (%)
1st endotherm SO CS CS80 TS TS80
Peak temperature (°C)
Weight loss (%)
1st exotherm
92 93 109 88 92
10.9 12.5 7.5 7,6 9.2
341 338 336 338 338
Peak temperature (°C)
Weight loss (%)
2nd exotherm 31.8 33.1 28.0 42.6 31.3
Peak temperature (°C)
Weight loss (%)
3rd exotherm
– – 456 (sh) 448 434
– – 10.2 8.1 14.0
541 539 560 554 552
53.1 45.3 47.5 38.2 41.4
57
151
73
33
131
175
sh: shoulder.
SN/h 0.74
SN/h 0.86
57
17
33
CS
32 25
56
154
SO
SN/h 0.97
23
CS80
TS
SN/h 1.36
TS80
SN/h 1.03
250
200
150
100
50
0
ppm
-50
Fig. 2. Bloch Decay 13C NMR spectra of humic acids (HA) isolated from the unamended soil (SO), composted sewage sludge (CS), thermally-dried sewage sludge (TS) and soils amended with CS or TS at a rate of 80 t ha¡1 y¡1 for 3 years (CS80 and TS80, respectively). SN/h is the signal-to-noise ratio per hour of acquisition time.
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
Table 4 Bloch Decay 13C NMR relative peak area (%) for humic acids (HA) isolated from the unamended soil (SO), composted sewage sludge (CS), thermally-dried sewage sludge (TS) and soils amended with CS or TS at a rate of 80 t ha¡1 y¡1 for 3 years (CS80 and TS80, respectively) Origin of Aliphatic-C O/N-aliphatic-C Aromatic-C Carboxyl/amide-C/ HA sample (0 < d 6 (45 < d 6 110 (110 < d 6 ketonic-C 45 ppm) (%) ppm) (%) 160 ppm) (%) (160 < d 6 220 ppm) (%) SO CS CS80 TS TS80
16.8 22.0 32.6 38.5 30.1
25.8 27.7 20.8 30.8 28.7
34.5 28.5 28.6 14.3 23.8
23.0 21.7 17.9 16.4 17.4
and thermal analyses. The existence of a large proportion of alkyl and O/N alkyl structures in the TS–HA could be attributed to the lower degree of maturation (i.e. biological decomposition) of the TS organic matter. Similar results have been reported for raw SS in other studies (Smernik et al., 2003; Adani and Tambone, 2005). Meanwhile, the composting process to which CS has been submit ted has resulted in HA with more recalcitrant aromatic structures (i.e. thermally stable compounds) that are rather similar to those of SO–HA. After 3 years of consecutive additions, HA from soils amended with both SS clearly differ from those of the original unamended soil. In particular, with respect to the SO–HA, the proportion of alkyl C has increased and aromatic and carbonyl and/or amide C have decreased in both CS80–HA and TS80–HA. Meanwhile, the O/N-alkyl C decreased slightly in CS80–HA and increased slightly in TS80–HA (Table 4). The increase in alkyl C (13.3% in TS80–HA and 15.8% in CS80–HA) constitutes the most significant contribution of both SS to soil–HA. The TS80–HA spectrum in this region (0–45 ppm) is similar to that obtained for CS80–HA and features broad peaks at 17, 25 and 33 ppm instead of the well-resolved peaks showed in TS–HA spectrum. Several authors (Genevini et al., 2002; Adani and Tambono, 2005) maintain that alkyl structures in SS are relat ively more resistant to degradation in soils, whereas others report no differences in the decomposability of different C-types in some SS analyzed (Smernik et al., 2004). The incorporation of new alkyl C in CS80- and TS80–HA may be ascribed to the natural humification process in the soil. Dur ing decomposition, O-alkyl C has been postulated to be utilized by the microbial popul ation in the soil, resulting in a relative increase of aromatic and alkyl structures, which play a fundamental role in the structure of humic acids (Hatcher et al., 1985; Almendros et al., 1991; Cook and Langford, 1998; Chefetz et al., 2002). Although less important than the contribution of aliphatic-C from SS–HA to soil–HA, the incorporation of organic N has been also observed. To provide a more accurate measure of proteina ceous HA–C, the N-alkyl (45–60 ppm) and O-alkyl (60–90 ppm) spectral regions have been independently integrated. The HA from soils amended with CS and TS had slightly more N-alkyl C (6.1% and 8.2%, respectively), than the unamended SO–HA (5.2%). There fore, in addition to the high contribution of alkyl structure, the con tribution of N-substituted alkyl C from CS–HA, and especially from TS–HA, is evident. These results agree with those found in thermal analysis as well as with the increase of H/C and N/C ratios revealed in the elemental analysis. Similarities between TS80- and CS80–HA characteristics con trast with differences observed between TS- and CS–HA. This can be attributed to the maturation process in the soil during an intense microbial degradation of the most labile organic fractions present in the waste. Incubation experim ents show that microbial decom position of TS in these soils is especially rapid and extensive. This process has been monitored in incubation experiments described previously by Fernández et al. (2007a).
1843
3.4. Agreement between TGA and NMR analysis As mentioned previously, there is a general agreement between the structural information revealed by TGA an NMR. For instance, HA samples with high aromatic carbon content also tend to have the greatest thermal stability, as indicated by greater mass loss during the 3rd exotherm. Therefore, to explore the potential for quantitative relationships between TGA and NMR data, linear regression analyses were performed upon the correlation between mass losses shown in Table 3 (after ash and water content cor rection) and the 13C NMR peak areas in Table 4. The correlation between the 1st exotherm of the TGA and the O-alkyl C deter mined by NMR had a correlation coef ficient (r) of 0.809. Similarly, the correlation between mass loss during the 3rd exotherm and aromatic carbon determined by NMR yielded an r value of 0.963. These results indicate that O/N-alkyl C content of the HA exam ined explains 65% of the variance in mass loss at 340 °C (P < 0.1), whereas aromatic C content explains 93% of the variance in mass loss at 550 °C (P < 0.01), thus suggesting a good agreement between TGA and NMR, especially for aromatic compounds. 4. Conclusions As a whole, this work shows the benefi ts of using thermogravi metric methods and NMR spectroscopy in the characterization of sewage sludge and their effects on soil–HA. Substantial differences have been observed in the chemical composition and structure of CS–HA and TS–HA. In particular, with respect to CS–HA, TS–HA fea tures prevalent alkyl structure and a greater content of O-alkyl (i.e. polysaccharide) and N-alkyl (i.e. proteinaceous) components, typ ical of a less mature organic material. The soil–HA shows a more aromatic character. After long-term application, the chemical and thermal properties of the soil HA fraction changed significantly. In general, soils amended with CS, and especially with TS exhibit enhancements in alkyl C and proteinaceous materials. Despite the notable differences between CS and TS, the different sludges pro duce a very similar HA fraction in the soil pointing to an “in situ” maturation process of the TS and highlighting the importance of the soil processes in determining the chemical composition of HA pool. In addition, the good agreement between the NMR and TGA results opens new possibilities for the use of thermal techniques in the study of organic matter evolution that should be object of further research. Acknowledgements J.M. Fernández is the recipient of a fellowship from the Conse jería de Educación de la Comunidad de Madrid partially funded by the European Social Fund. C. Plaza is a researcher of the Ramón y Cajal Program funded by the Spanish Ministry of Education and Science. We acknowledge the Ohio State University for the use of their research facilities. References Adani, F., Tambone, F., 2005. Long-term effect of sewage sludge application on soil humic acids. Chemosphere 60, 1214–1221. Almendros, G., Sanz, J., Gonzalez-Vila, F.J., Martin, F., 1991. Evidence for a polyalkyl nature of soil humin. Naturwissenschaften 78, 359–362. Banerjee, M.R., Burton, D.L., Depoe, S., 1997. Impact of sewage sludge application on soil biologic al characteristics. Agric. Ecosyst. Environ. 66, 241–249. Boyd, S.A., Sommers, L.E., Nelson, D.W., 1980. Changes in the humic acid fraction of soil resulting from sludge application. Soil Sci. Soc. Am. J. 44, 1179–1186. Chefetz, B., Tarchitzky, J., Deshmukh, A.P., Hatcher, P.G., Chen, Y., 2002. Structural characterization of soil organic matter and humic acids in particle-size frac tions of an agricultural soil. Soil Sci. Soc. Am. J. 66, 129–141. Conte, P., Piccolo, A., Van Lagen, B., Buurman, P., de Jager, P.A., 1997. Quantitative aspects of solid-state13C NMR spectra of humic substances from soils of volca nic systems. Geoderma 80, 327–338.
1844
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
Cook, R.L., Langford, C.H., 1998. Structural characterization of a fulvic acid and a humic acid using solid-state Ramp-CP-MAS13C nuclear magnetic resonance. Environ. Sci. Technol. 32, 719–725. Dell’Abate, M.T., Benedetti, A., Trinchera, A., Dazzi, C., 2002. Humic substances along the profile of two Typic Haploxwerert. Geoderma 107, 281–296. Dria, K.J., Sachleben, J.R., Hatcher, P.G., 2002. Solid-state Carbon-13 nuclear mag netic resonance of humic acids at high magnetic field strengths. J. Environ. Qual. 31, 393–401. Ernst, R.R., Anderson, W.A., 1966. Applications of Fourier transform spectroscopy to magnetic resonance. Rev. Sci. Instrum. 37, 93–102. FAO-ISRIC, 1998. WRB—World Reference Base for soil resources, World Soil Resources Report no. 84. FAO, Rome. Fernández, J.M., Plaza, C., Hernández, D., Polo, A., 2007a. Carbon mineralization in an arid soil amended with thermally-dried and composted sewage sludges. Geo derma 137, 497–503. Fernández, J.M., Plaza, C., Hernández, D., Polo, A., 2007b. Organic matter in degraded agricultural soils amended with composted and thermally dried sew age sludges. Sci. Total Environ. 378, 75–80. Francioso, O., Montecchio, D., Gioacchini, P., Ciavatta, C., 2005. Thermal analysis (TG-DTA) and isotopic characterization (13C–15N) of humic acids from different origins. Appl. Geochem. 20, 537–544. Flaig, W., Beutelspacher, H., Rietz, E., 1975. Chemical composition and physical properties of humic substances. In: Gieseking, J.E. (Ed.), In: In Soil Components, vol. 1. Springer-Verlag, Berlin, pp. 119–126. Genevini, P.L., Adani, F., Veeken, A., Nierop, G.J., Scaglia, B., Dijkema, C., 2002. Qualit ative modifications of humic acid-like and core-humic acid-like during high-rate compost ing of pig faeces amended with wheat straw. Soil Sci. Plant Nutr. 48 (2), 143–150. Geras imowicz, W.V., Byler, D.M., 1985. Carbon-13 CPMAS NMR and FTIR spectro scopic studies of humic acids. Soil Sci. 139, 270–285. Gonet, S.S., Cieslewicz, J., 1998. Differential thermal analysis of sedimentary humic acids in the light of their origin. Environ. Int. 24 (5/6), 629–636. Hatcher, P.G., Breger, I.A., Maciel, G.E., Szeveren yi, N.M., 1985. Geochemistry of humin. In: Hayes, M.H.B., MacCarthy, P., Malcolm, R.L., Swift, R.S. (Eds.), In Humic Substances H. Chichester, Wiley, pp. 275–302. Hatcher, P.G., Dria, K.J., Kim, S., Frazier, S.W., 2001. Modern analytical studies of humic substances. Soil Sci. 166, 770–794. Hargreaves, J.C., Adl, M.S., Warman, P.R., 2008. A review of the use of composted municipal solid waste in agriculture. Agric. Ecosyst. Environ. 123, 1–14. Navas, A., Bermúdez, F., Machín, J., 1998. Influence of sewage sludge application on physic al and chemical properties of Gypsisols. Geoderma 87, 123–135. Peuravuori, J., Paaso, N., Pihlafa, K., 1999. Kinetic study of the thermal degradation of lake aquatic humic matter by thermogravimetric analysis. Thermochim. Acta 325, 181–193. Pietro, M., Paola, C., 2004. Thermal analysis for the evalua tion of the organic matter evolution during municipal solid waste aerobic composting process. Thermo chim. Acta 413, 209–214.
Provenzano, M.R., Senesi, N., Miikki, V., 1998. Characterization of composts and humic acids from pulp and paper mill biosludges by DSC in association with FT-IR spectroscopy. J. Therm. Anal. 52, 1037–1046. Reveille, V., Mansuy, L., Jarde, E., Garnier-Sillam, E., 2003. Characterization of sew age sludge-derived organic matter: lipids and humic acid. Org. Geochem. 34, 615–627. Ricca, G., Federica, L., Astori, C., Gallo, R., 1993. Structural investigations of humic acid from leonardite by spectroscopic methods and thermal analysis. Geoderma 57, 263–274. Ricca, G., Severini, F., Di Silvestro, G., Yuan, C.M., Adani, F., 2000. Derivatization and structural studies by spectroscopic methods of humic acids from Leonardide. Geoderma 98, 115–125. Senesi, N., 1989. Composted material as organic fertilizer. Sci. Total Environ. (81/82), 521–542. Senesi, N., Miano, T.M., Brunetti, G., 1996. Humic-like substances in organic amend ments and effects on native soil humic substances. In: Piccolo, A. (Ed.), In: Humic Substances in Terrestrial Ecosystems, vol. 1. Elsevier, Amsterdam, pp. 531–593. Senesi, N., Plaza, C., 2007. The role of humification in recycling of organic wastes of vario us nature and sources as soil amendments. Clean Soil Air Water 35, 26–41. Sheppard, J.D., Forgeron, D.W., 1987. Differential thermogravimetry of peat frac tions. Fuel 66, 232–236. Smernik, R.J., Oades, J.M., 2000. The use of spin counting for determining quantita tion of solid state13C-NMR spectra of natural organic matter. 2. HF-treated soil fractions. Geoderma 96, 159–171. Smernik, R.J., Oliver, I.W., Merrington, G., 2003. Characterization of sewage sludge organic matter using solid-state carbon-13 nuclear magnetic resonance spec troscopy. J. Environ. Qual 32, 1516–1522. Smernik, R.J., Oliver, I.W., Mc Laughlin, M.J., 2004. Changes in the nature of sewage sludge organic matter during a twenty-one-month incubation. J. Environ. Qual. 33, 1924–1929. Soil Survey Staff., 2003. Keys to soil taxonomy, 9th ed. Natural Resources Conserva tion Service USDA, Washington. Sort, X., Alcañiz, J.M., 1999. Modifi cation of soil porosity after application of sewage sludge. Soil Till. Res. 49, 337–345. Stevenson, F.J., 1994. Humus Chemistry; Genesis. Composition, In: Reactions, 2nd ed. Wiley-Interscience, New York. Tarrason, D., Ojeda, G., Ortíz, O., Alcañiz, J.M., 2008. Differences on nitrogen avail ability in a soil amended with fresh, composted and thermally-dried sewage sludge. Biores our. Technol. 99 (2), 252–259. Wilson, M.A., 1987. NMR Techniques and Applications in Geochemistry and Soil Chemistry. Pergamon Press, Oxford. Wong, J.W.C., Lai, K.M., Fang, M., Ma, K.K., 1998. Effect of sewage sludge amend ment on soil microbial activity and nutrient mineralization. Environ. Int. 24 (8), 935–943.
Contents lists available at ScienceDirect
Chemosphere j o u r n a l h o m e p a g e : w w w . e l s e v i e r. c o m / l o c a t e / ch e m o s p h e r e
Effects of long-term soil amendment with sewage sludges on soil humic acid thermal and molecular properties José M. Fernández a,*, William C. Hockaday b, César Plaza a, Alfredo Polo a, Patrick G. Hatcher c a
Centro de Ciencias Medioambientales, Consejo Superior de Investigacio nes Científicas, Serrano 115 dpdo., 28006 Madrid, Spain Department of Earth Science, Rice University 6100 Main Street, Houston, TX 77005, United States c Department of Chemistry and Biochemistry, Old Dominion University 4541 Hampton Boulevard, Norfolk, VA 23529-0126, United States b
a r t i c l e
i n f o
Article history: Received 4 March 2008 Received in revised form 31 July 2008 Accepted 3 August 2008 Available online 20 September 2008 Keywords: Composting Thermal-drying Sewage sludge Humic acids 13 C NMR Thermal analysis
a b s t r a c t Sewage sludges are frequently used as soil amendments due to their high contents of organic matter and nutrients, particularly N and P. However, their effects upon the chemistry of soil humic acids, one of the main components of the soil organic matter, need to be more deeply studied in order to understand the relation between organic matter structure and beneficial soil properties. Two sewage sludges subjected to different types of pre-treatment (composted and thermally dried) with very different chemical compo sitions were applied for three consecutive years to an agricultural soil under long-term field study. Ther mal analysis (TG–DTG–DTA) and solid-state 13C NMR spectroscopy were used to compare molecular and structural properties of humic acids isolated from sewage sludges, and to determine changes in amended soils. Thermally dried sewage sludge humic acids showed an important presence of alkyl and O/N-alkyl compounds (70%) while composted sludge humic acids comprised 50% aromatic and carbonyl carbon. In spite of important differences in the initial chemical and thermal properties of the two types of sewage sludges, the chemical and thermal properties of the soil humic acids were quite similar to one another after 3 years of amendment. Long-term application of both sewage sludges resulted in 80–90% enrich ment in alkyl carbon and organic nitrogen contents of the soil humic acid fraction. © 2008 Elsevier Ltd. All rights reserved.
1. Introduction Animal manure and crop residues have traditionally been applied to soil as a means of maintaining and increasing the organic matter content. However, nowad ays their production can not satisfy the current soil requirements of cropland. This has moti vated the growing interest in agricultural applications of organic residues derived from a wide variety of human activities as a con venient alternative to their landfill disposal. In this context, sewage sludge (SS), the residual solid produced during wastewater treat ment, has been increasingly applied as an organic amendment pro viding substantial gains in soil fertility and crop yield due to its high content of organic matter and nutrients, particularly N and P (Hargreaves et al., 2008 and references therein). However, before being used, as with other organic waste mate rials, SS must be properly processed in order to obtain a stable organic matter. Phytotoxicity and other detrimental effects of the premature application of organic wastes have been extensively reported. Composting is one of the most common treatments that resemble natur al processes in soil, being an accelerated humifi cation process and has been widely studied for many decades * Corresponding author. Tel./fax: +34 91 4115301. E-mail address: jmfernan[email protected] (J. M. Fernández). 0045-6535/$ - see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2008.08.001
(Senesi, 1989; Senesi and Plaza, 2007). Another relatively new treatment method is thermal-drying, which consists of a short and intense heating of the SS, destroying pathogens, eliminating volatile chemicals and leading to a sanitized final product with good handling characteristics but lacking a maturation process (Tarrason et al., 2008; Fernández et al., 2007a). Although this pro cess has been increasingly introduced in municipal wastewater treatments plants, the effects of thermally-dried sewage sludge on soil properties and soil organic matter have been scarcely studied. Although benefic ial effects upon the physical (Navas et al., 1998; Sort and Alcañiz, 1999), chemical (Navas et al., 1998), and biological properties (Banerjee et al., 1997; Wong et al., 1998) have been widely investigated, little information is available on long-term effects induced by SS on soil humic substances, and particularly on humic acids (HAs), one of the most abundant and chemically active fractions of soil organic matter (Senesi, 1989; Reveille et al., 2003; Adani and Tambone, 2005). Due to the impor tant role that HAs play in the soil, information about their proper ties and transformations can be essential to understand the effects of SS on soil organic matter properties and evaluate the suitability of these materials as organic amendments. Among the numerous chemical and spectroscopic methods used to study HA structures and composition, thermal investiga tions are steadily gaining interest. However, most of the thermal
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
studies are still related to characterizations of HAs from soils (Gonet and Cieslewicz, 1998; Dell’Abate et al., 2002) or compost ing materials (Provenzano et al., 1998; Pietro and Paola, 2004), and do not focus on the evolution of HAs in soil organic matter resto ration experiments. The introduction of the use of this technique, which has advantages in terms of time and money, can assume an important role in this kind of study by providing highly valuable information. However, this innovative method still must be com pared with other reliable techniques. The use of solid-state 13C nuclear magnetic reson ance (NMR) spectroscopy has emerged as one of the most useful and powerful tools to obtain structural information of humic substances (Wilson, 1987; Hatcher et al., 2001; Smernik et al., 2003). The combination of cross polarization (CP) and magic angle spinning (MAS) is the most popular 13C NMR technique for the study of humic materials (Conte et al., 1997; Cook and Langford, 1998; Chefetz et al., 2002). However, this technique can under represent some structural units (e.g. fused aromatic moieties, alkyl groups with high degrees of molecular mobility, and carbonyl C) (Dria et al., 2002; Smernik et al., 2003). These studies, in combination with our own previous experiences, have suggested the use of the alternative Bloch decay (BD) technique in the analysis of C-rich HAs. Although this tech nique generally has longer requirements of time and lower signal to noise ratio than CP, BD provides the most accurate quantita tive distributions of structural units in humic acids (Smernik and Oades, 2000; Dria et al., 2002). In this study, solid-state 13C NMR and thermal analyses (TG, DTG and DTA) were performed in order to: (a) determine and com pare molecular and structural properties of the HA isolated from composted sewage sludge (CS) and thermally-dried sewage sludge (TS), and (b) investigate their contributions to the soil HA pool after three years, comparing results obtained with both techniques. 2. Materials and methods 2.1. Sewage sludges, soil and field experiment The CS sample (dry matter content, 435.8 g kg¡1) was col lected from a 3-month windrow-composted mixture of three SS originated from three municipal wastewater treatment plants in Madrid (Spain) metropolitan area. The TS sample (dry matter con tent, 846.4 g kg¡1) consisted of a SS dried by indirect convection with air heated to temperat ures between 380 °C and 450 °C in the wastewater treatment plant “SUR” in Madrid metropolitan area. The maximum temperat ure reached in the sludge during thermal drying was 75 °C. The field experiment was conducted in the experimental farm “La Higuerue la” located in Toledo (Spain). The site is characterized by a continental semiarid climate with an average annual rainfall of about 487 mm and an average annual temperature of 14 °C. The soil is a Calcic Luvisol (FAO-ISRIC, 1998) or Typic Haploxeralf (Soil Survey Staff, 2003), having a sandy loam texture (sand, 590 g kg¡1; silt, 220 g kg¡1; clay, 190 g kg¡1). Mean chemical properties of soil and both sewage sludges are included in Table 1. The experimental design consisted in four random blocks of nonirrigated soil plots (10 £ 3 m2) cropped with barley (Hord eum vulgare L.), either unamended (SO) or amended yearly over a 3-year period at a rate of 80 t ha¡1 with CS (CS80) or TS (TS80). Both SS were applied in mid-September, prior to barley planting in mid-October, and immediately incorporated into soil at a depth of 0–15 cm. The third year, after barley harvesting, in late June, four surface soil subsamples (Ap horizon, 0–15 cm depth) were collected randomly from each plot. Each soil subsample consisted of a mixture of 20 soil cores each of 3-cm diameter. A compos ite sample was then obtained for each treatment by mixing equal amounts (1 kg) of the four corresponding soil subsamples.
1839
Table 1 Main chemical properties (±standard errors) of unamended soil (SO), composted sewage sludge (CS) and thermally-dried sewage sludge (TS) used in the experi ment
pH (H2O) Electrical conductivity (dS m¡1) Total organic carbon (g kg¡1) Total extractable carbon (g kg¡1) Humic acids carbon (g kg¡1) Fulvic acids carbon (g kg¡1) C/N Total N (g kg¡1) P (g kg¡1) K (g kg¡1) a b
SO
CS
TS
5.7 ± 0.1 0.05 ± 0.01 7.2 ± 0.1 1.4 ± 0.2 0.7 ± 0.1 0.6 ± 0.1 8.0 ± 0.1 0.9 ± 0.1 0.09 ± 0.01a 0.20 ± 0.01a
7.1 ± 0.1 3.90 ± 0.01 181.0 ± 0.2 53.4 ± 0.2 21.0 ± 0.3 32.4 ± 0.3 7.6 ± 0.2 23.9 ± 0.1 13.90 ± 0.02b 5.02 ± 0.07b
7.0 ± 0.1 1.50 ± 0.02 296.0 ± 0.2 87.7 ± 0.3 24.8 ± 0.2 62.9 ± 0.3 8.3 ± 0.2 35.6 ± 0.1 13.43 ± 0.02b 4.29 ± 0.05b
Available content. Total content.
2.2. Humic acid isolation The HA fractions were isolated from CS, TS, CS80, TS80, and SO samples by conventional methods. The humic acid C contents of each sample are reported in detail in Fernández et al. (2007b). Prior to extraction of HA, soil and SS samples were air-dried and passed through a 2 mm sieve after removal of plant residues and stones from soils. Carbonates were removed from soil samples by mechanical stirring with 2 M H3PO4 for 30 min. The treatment was repeated three times, and then the samples were washed with dis tilled water until the suspension reached a pH 7. Carbonate-free soils and SS samples were first extracted with 0.1 M Na4P2O7 at pH 9.8 and then by 0.1 M NaOH, at room temperature (RT, about 293 K), and using a sample to extractant ratio of 1:10. Each extraction was repeated three times by mechanically shaking the mixture for 3 h, then centrifuging at 15 300g for 15 min, and filtering the superna tant through a Whatman no. 31 filter paper. The combined alka line supernatants were then acidified with HCl until reaching a pH 1, left standing for 24 h in a refrigerator to allow the complete precipitation of HA, centrifuged at 30 100g for 15 min, and then fil tered through a Whatman no. 31 filter paper. The HA precipitates were then purified by dissolution in 0.1 M NaOH, centrifugation at 30100 g, elimination of the residue, acidification of the alkaline supernatant with HCl–HF until pH 1, and standing for 12 h in a refrigerator. The purification procedure was repeated three times. The precipitated HAs were recovered with distilled water, dialyzed until free of Cl¡ ions, and finally freeze-dried. 2.3. Humic acid characterization The moisture content of HAs was measured by heating over night at 105 °C and ash content by combustion overnight at 550 °C. The C, H, N and S contents were determined in triplicate using a Fisons Instruments (Crawley, UK) elemental analyzer model EA 1108. Oxygen content was calculated by difference on an ash-free dry weight basis as: O% = 100 ¡ (C + H + N + S)%. Thermogravimetry (TG), differential thermogravimetry (DTG) and differential thermal analysis (DTA) were performed simul taneously using a Setaram (Caluire, France) SETSYS thermal ana lyzer equipped with the computer program Setsoft 1.6.4 (Setiram, Caluire, France) for data processing. Ten-milligram HS samples were isothermally heated from room temperature to 30 °C for 10 min and then to 1200 °C in a platinum crucible under a flow of air of 30 mL min¡1. The heating rate was set at 10 °C min¡1. Calcined alumina was used as reference material. Solid-state 13C NMR experiments were performed on a Bruker (Billerica, MA) DMX 300 NMR spectrometer. This spectrometer operates at a 13C frequency of 75.48 MHz. Samples were placed in
1840
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
a 4 mm (outside diameter) NMR rotor with a Kel-F cap (3M, Min neapolis, MN). Bloch decay spectra were acquired using a 20° 13C pulse, a recycle delay of 3 s, and an MAS frequency of 13 kHz. The recycle delay of 3 s was chosen because incrementing the delay to 4 s and 5 s gave no increase in signal intensity. The 3 s delay follow ing a 20° pulse is equivalent to a 150 second recycle delay accom panying a 90° 13C pulse (Ernst and Anderson, 1966). Approximately 22 500 scans were collected for each sample. The Kel-F rotor caps and probe components give rise to a significant 13C signal in the BD spectrum, which overlapped with the NMR signal of the sam ples. To correct for this, the empty-rotor spectrum was acquired separately and was subtracted from each BD spectrum before inte gration. A 40 Hz line broadening was applied when processing the NMR spectra, and peak areas were integrated according to the chemical shift regions defined in Table 4. 3. Results and discussion 3.1. Elemental analysis The elemental composition of SS HA differs according to their source and treatment process. The CS–HA (Table 2) features higher O/C and lower H/C and N/C ratios than TS–HA. These results sug gest that composting favours the formation of a more oxidized HA fraction than thermal drying. This is typical of a more mature organic material (Senesi et al., 1996). Simil ar results for composted and uncomposted SS are reported by other authors (Boyd et al., 1980; Gerasimowicz and Byler, 1985). Relative to SO–HA, CS–HA and TS–HA have lower or much lower C, O contents, and larger or much larger N and S contents. The high levels of N and S in SS–HAs may be ascribed to the incorporation of proteinaceous products and S-containing commercial surfactant residues in the HA frac tion (Senesi et al., 1996). In comparison with untreated SO–HA, the HA fraction of SS– amended soils has undergone partial incorporation of compounds from CS and TS into the soil HA fraction. This is evident by the sig nificant increases in C, H, N, and S content of CS80 and TS80 (Table 2). However, it is interesting to note that the original differences in CS–HA and TS–HA elemental composition are not manif ested in the soil HA fraction after 3 years of amendment. That is, CS80 and TS80 show equival ent C, H, O and S contents. The only difference is the slightly higher N content of TS80. 3.2. Thermal analysis Thermograms of HA samples are shown in Fig. 1 and main ther mal parameters are summarised in Table 3. The HA samples from TS reveal a significantly different thermal behaviour than CS–HA and SO–HA, which are fairly similar to one another. Meanwhile, HA samples from SS amended soils are quite simil ar. The primary fea tures of the differential thermograms are: (a) a small endothermic peak at low temperatures (90.0–101.8 °C), ascribed to reactions of dehydration; (b) an exothermic peak within the 338–341 °C range,
generally attributed to the degradation of polysaccharides, decar boxylation of acidic groups and dehydration of aliphatic alcohols (Flaig et al., 1975; Dell’Abate et al., 2002; Francioso et al., 2005); (c) an exothermic peak between 439 °C and 456 °C, possibly due to the decomposition of N compounds (e.g. proteinaceous material) and long chain hydrocarbons (Francioso et al., 2005); and (d) an exo thermic peak at high temperatures (539–560 °C), usually related to the combustion of aromatic structures and cleavage of C–C bonds (Peuravuori et al., 1999). Exothermic reactions, which determine weight losses at differ ent steps recorded on the TG curves, allow the quantitative com parison of the different fractions involved in the reactions (Table 3). For the first exothermic reaction, CS–HA shows a lower weight loss than TS–HA, which features the most important lost of weight. The DTG curves in this region also indicate rapid combustion of the TS–HA compounds at this range of temperatures in compari son with the other samples. These results suggest the presence of a large proportion of carbohydrates and aliphatic alcohols structures in TS–HA (Ricca et al., 1993). The exothermic peak at 439–456 °C, unique to the HA from SS–amended soils, is slightly higher for the TS80–HA than for CS80–HA. This is consistent with the persistence of proteinaceous and/or long chain N-alkyl surfactant structures present in the TS–HA (8.1% weight). This peak could also arise from N-aromatic compounds (e.g. melanoidins) formed by the condensa tion of reducing sugars and amines during the thermal drying pro cess (Stevenson, 1994). The last exothermic peak features slightly higher loss of weight for CS–HA (45.3%) in comparison with TS–HA (38.2%). However the biggest mass loss (53.1%) corresponds to SO–HA. Further, SO– and CS–HA show a higher decomposition rate than TS–HA. This could be attributed to the presence of aromat ics in CS–HA and SO–HA (Sheppard and Forgeron, 1987; Francioso et al., 2005). This is confirmed by 13C NMR, as discussed in Section 3.3. 3.3. Bloch decay 13C nuclear magnetic resonance analysis The description of the chemical nature of the organic C con tained in HA is more directly conferred by analysis of the BD 13C NMR spectra (Fig. 2). Spectra were divided into four regions and integrated as follows: 0–45 ppm assigned to alkyl C; 45–110 ppm assigned primarily to O-substituted alkyl C in carbohydrates, but also including methoxyl C and N-substituted alkyl C in protein; 110–160 ppm assigned to aryl C, including O-aryl and C- and Hsubstituted aryl C; and 160–220 ppm assigned to carbonyl C in car boxylic acids, ketones, esters, and amides (Table 4). Humic acids extracted from TS and CS show higher content of the alkyl and O/N-alkyl C and lower content of aromatic and car bonyl/amide C than SO–HA. These differences are more marked for 13 C NMR spectra of HA extracted from the TS, which is dominated by signals in the alkyl C region at d < 45 ppm. In particular, signals at d 6 20 ppm indicate methyl C in alkyl chains, and the signal near 25 ppm represents methylene C in alkyl chains. Meanwhile, the strong signal at 33 ppm can be attributed to rigid domains in
Table 2 Moisture, ash content, elemental composition and atomic ratios (± standard error) of humic acids (HA) isolated from the unamended soil (SO), composted sewage sludge (CS), thermally-dried sewage sludge (TS) and soils amended with CS or TS at a rate of 80 t ha¡1 y¡1 for 3 years (CS80 and TS80, respectively) Origin of HA sample
Moisture (g kg¡1)
Ash (g kg¡1)
C (g kg¡1)a
H (g kg¡1)a
N (g kg¡1)a
S (g kg¡1)a
O (g kg¡1)a
C/N (atomic ratios)
H/C (atomic ratios)
O/C (atomic ratios)
SO CS CS80 TS TS80
69.8 35.9 24.9 15.9 40.4
79.2 115.7 61.2 40.5 84.3
505 ± 6 460 ± 1 598 ± 0 503 ± 2 605 ± 0
53 ± 3 53 ± 1 67 ± 0 70 ± 3 72 ± 0
44 ± 1 85 ± 0 90 ± 0 83 ± 0 81 ± 0
4 ± 0 14 ± 0 16 ± 0 13 ± 0 15 ± 0
394 ± 9 389 ± 1 228 ± 0 330 ± 5 227 ± 0
13.3 6.3 7.8 7.1 8.8
1.3 1.4 1.4 1.7 1.4
0.6 0.6 0.3 0.5 0.3
a
On a moisture- and ash-free basis.
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
nant in SO–HA and CS–HA, showing the abundance of aromatic C (e.g. peak at 131 ppm.) and O-substituted aromatic C (e.g. peak at 151 ppm.) (Chefetz et al., 2002). These results agree with elemental
Soil
TG(%)
DTA( µV) 160
0 DTG
exo →
-20 -40
140 120 100 80
-2
60
-60
0
-1
straight alkyl chains, or to branched alkyl chains, such as R–CH and R–CH2 a substituted groups where RCOOH (Ricca et al., 2000). In contrast, the region between 110 and 160 ppm is predomi
1841
DTG(% min )
40 -80
20 DTA TG
-100 0
200
400
600
800
-4
0 -20
o
Temperature( C)
-40
120
60
-60
DTG
-20
100 80
140
0
-40
-2
20 DTA TG
-100 400
600
40 -80
-4
DTA TG
-100
-20
800
0
200
400
600
DTA( µV)
-40
100 80 60
-60
160
0
DTG
-20 -1
120
DTA( µV)
0
DTG(% min )
exo →
-20
140
TS80
TG(%)
160
DTG
-40
-2
20 DTA TG
-100 600 o
Temperature( C)
800
0 -20
140 120 100 80 60
-60
0
-2
40
40 -80
400
-4
-20
800
exo →
CS80
200
0
Temperature( C)
0
0
20
o
o
Temperature( C)
TG(%)
-2
-1
200
80
DTG(% min )
0
0
0
100
60
-60
40 -80
120
-1
exo →
-20
140
160
-1
DTG
DTA( µV)
0
DTG(% min )
0
TS
TG(%)
160
DTG(% min )
DTA( µV)
exo →
CS
TG(%)
-80
20
-4
DTA TG
-100 0
200
400
600
800
0
-4
-20
o
Temperature( C)
Fig. 1. TG, DTG and DTA thermograms of humic acids (HA) isolated from the unamended soil (SO), composted sewage sludge (CS), thermally-dried sewage sludge (TS) and soils amended with CS or TS at a rate of 80 t ha¡1 y¡1 for 3 years (CS80 and TS80, respectively).
1842
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
Table 3 Differential thermal analysis (DTA) peak temperatures and weight losses (%) of humic acids (HA) isolated from the unamended soil (SO), composted sewage sludge (CS), ther mally-dried sewage sludge (TS) and soils amended with CS or TS at a rate of 80 t ha¡1 y¡1 for 3 years (CS80 and TS80, respectively) Origin of HA sample
Peak temperat ure (°C)
Weight loss (%)
1st endotherm SO CS CS80 TS TS80
Peak temperature (°C)
Weight loss (%)
1st exotherm
92 93 109 88 92
10.9 12.5 7.5 7,6 9.2
341 338 336 338 338
Peak temperature (°C)
Weight loss (%)
2nd exotherm 31.8 33.1 28.0 42.6 31.3
Peak temperature (°C)
Weight loss (%)
3rd exotherm
– – 456 (sh) 448 434
– – 10.2 8.1 14.0
541 539 560 554 552
53.1 45.3 47.5 38.2 41.4
57
151
73
33
131
175
sh: shoulder.
SN/h 0.74
SN/h 0.86
57
17
33
CS
32 25
56
154
SO
SN/h 0.97
23
CS80
TS
SN/h 1.36
TS80
SN/h 1.03
250
200
150
100
50
0
ppm
-50
Fig. 2. Bloch Decay 13C NMR spectra of humic acids (HA) isolated from the unamended soil (SO), composted sewage sludge (CS), thermally-dried sewage sludge (TS) and soils amended with CS or TS at a rate of 80 t ha¡1 y¡1 for 3 years (CS80 and TS80, respectively). SN/h is the signal-to-noise ratio per hour of acquisition time.
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
Table 4 Bloch Decay 13C NMR relative peak area (%) for humic acids (HA) isolated from the unamended soil (SO), composted sewage sludge (CS), thermally-dried sewage sludge (TS) and soils amended with CS or TS at a rate of 80 t ha¡1 y¡1 for 3 years (CS80 and TS80, respectively) Origin of Aliphatic-C O/N-aliphatic-C Aromatic-C Carboxyl/amide-C/ HA sample (0 < d 6 (45 < d 6 110 (110 < d 6 ketonic-C 45 ppm) (%) ppm) (%) 160 ppm) (%) (160 < d 6 220 ppm) (%) SO CS CS80 TS TS80
16.8 22.0 32.6 38.5 30.1
25.8 27.7 20.8 30.8 28.7
34.5 28.5 28.6 14.3 23.8
23.0 21.7 17.9 16.4 17.4
and thermal analyses. The existence of a large proportion of alkyl and O/N alkyl structures in the TS–HA could be attributed to the lower degree of maturation (i.e. biological decomposition) of the TS organic matter. Similar results have been reported for raw SS in other studies (Smernik et al., 2003; Adani and Tambone, 2005). Meanwhile, the composting process to which CS has been submit ted has resulted in HA with more recalcitrant aromatic structures (i.e. thermally stable compounds) that are rather similar to those of SO–HA. After 3 years of consecutive additions, HA from soils amended with both SS clearly differ from those of the original unamended soil. In particular, with respect to the SO–HA, the proportion of alkyl C has increased and aromatic and carbonyl and/or amide C have decreased in both CS80–HA and TS80–HA. Meanwhile, the O/N-alkyl C decreased slightly in CS80–HA and increased slightly in TS80–HA (Table 4). The increase in alkyl C (13.3% in TS80–HA and 15.8% in CS80–HA) constitutes the most significant contribution of both SS to soil–HA. The TS80–HA spectrum in this region (0–45 ppm) is similar to that obtained for CS80–HA and features broad peaks at 17, 25 and 33 ppm instead of the well-resolved peaks showed in TS–HA spectrum. Several authors (Genevini et al., 2002; Adani and Tambono, 2005) maintain that alkyl structures in SS are relat ively more resistant to degradation in soils, whereas others report no differences in the decomposability of different C-types in some SS analyzed (Smernik et al., 2004). The incorporation of new alkyl C in CS80- and TS80–HA may be ascribed to the natural humification process in the soil. Dur ing decomposition, O-alkyl C has been postulated to be utilized by the microbial popul ation in the soil, resulting in a relative increase of aromatic and alkyl structures, which play a fundamental role in the structure of humic acids (Hatcher et al., 1985; Almendros et al., 1991; Cook and Langford, 1998; Chefetz et al., 2002). Although less important than the contribution of aliphatic-C from SS–HA to soil–HA, the incorporation of organic N has been also observed. To provide a more accurate measure of proteina ceous HA–C, the N-alkyl (45–60 ppm) and O-alkyl (60–90 ppm) spectral regions have been independently integrated. The HA from soils amended with CS and TS had slightly more N-alkyl C (6.1% and 8.2%, respectively), than the unamended SO–HA (5.2%). There fore, in addition to the high contribution of alkyl structure, the con tribution of N-substituted alkyl C from CS–HA, and especially from TS–HA, is evident. These results agree with those found in thermal analysis as well as with the increase of H/C and N/C ratios revealed in the elemental analysis. Similarities between TS80- and CS80–HA characteristics con trast with differences observed between TS- and CS–HA. This can be attributed to the maturation process in the soil during an intense microbial degradation of the most labile organic fractions present in the waste. Incubation experim ents show that microbial decom position of TS in these soils is especially rapid and extensive. This process has been monitored in incubation experiments described previously by Fernández et al. (2007a).
1843
3.4. Agreement between TGA and NMR analysis As mentioned previously, there is a general agreement between the structural information revealed by TGA an NMR. For instance, HA samples with high aromatic carbon content also tend to have the greatest thermal stability, as indicated by greater mass loss during the 3rd exotherm. Therefore, to explore the potential for quantitative relationships between TGA and NMR data, linear regression analyses were performed upon the correlation between mass losses shown in Table 3 (after ash and water content cor rection) and the 13C NMR peak areas in Table 4. The correlation between the 1st exotherm of the TGA and the O-alkyl C deter mined by NMR had a correlation coef ficient (r) of 0.809. Similarly, the correlation between mass loss during the 3rd exotherm and aromatic carbon determined by NMR yielded an r value of 0.963. These results indicate that O/N-alkyl C content of the HA exam ined explains 65% of the variance in mass loss at 340 °C (P < 0.1), whereas aromatic C content explains 93% of the variance in mass loss at 550 °C (P < 0.01), thus suggesting a good agreement between TGA and NMR, especially for aromatic compounds. 4. Conclusions As a whole, this work shows the benefi ts of using thermogravi metric methods and NMR spectroscopy in the characterization of sewage sludge and their effects on soil–HA. Substantial differences have been observed in the chemical composition and structure of CS–HA and TS–HA. In particular, with respect to CS–HA, TS–HA fea tures prevalent alkyl structure and a greater content of O-alkyl (i.e. polysaccharide) and N-alkyl (i.e. proteinaceous) components, typ ical of a less mature organic material. The soil–HA shows a more aromatic character. After long-term application, the chemical and thermal properties of the soil HA fraction changed significantly. In general, soils amended with CS, and especially with TS exhibit enhancements in alkyl C and proteinaceous materials. Despite the notable differences between CS and TS, the different sludges pro duce a very similar HA fraction in the soil pointing to an “in situ” maturation process of the TS and highlighting the importance of the soil processes in determining the chemical composition of HA pool. In addition, the good agreement between the NMR and TGA results opens new possibilities for the use of thermal techniques in the study of organic matter evolution that should be object of further research. Acknowledgements J.M. Fernández is the recipient of a fellowship from the Conse jería de Educación de la Comunidad de Madrid partially funded by the European Social Fund. C. Plaza is a researcher of the Ramón y Cajal Program funded by the Spanish Ministry of Education and Science. We acknowledge the Ohio State University for the use of their research facilities. References Adani, F., Tambone, F., 2005. Long-term effect of sewage sludge application on soil humic acids. Chemosphere 60, 1214–1221. Almendros, G., Sanz, J., Gonzalez-Vila, F.J., Martin, F., 1991. Evidence for a polyalkyl nature of soil humin. Naturwissenschaften 78, 359–362. Banerjee, M.R., Burton, D.L., Depoe, S., 1997. Impact of sewage sludge application on soil biologic al characteristics. Agric. Ecosyst. Environ. 66, 241–249. Boyd, S.A., Sommers, L.E., Nelson, D.W., 1980. Changes in the humic acid fraction of soil resulting from sludge application. Soil Sci. Soc. Am. J. 44, 1179–1186. Chefetz, B., Tarchitzky, J., Deshmukh, A.P., Hatcher, P.G., Chen, Y., 2002. Structural characterization of soil organic matter and humic acids in particle-size frac tions of an agricultural soil. Soil Sci. Soc. Am. J. 66, 129–141. Conte, P., Piccolo, A., Van Lagen, B., Buurman, P., de Jager, P.A., 1997. Quantitative aspects of solid-state13C NMR spectra of humic substances from soils of volca nic systems. Geoderma 80, 327–338.
1844
J.M. Fernández et al. / Chemosphere 73 (2008) 1838–1844
Cook, R.L., Langford, C.H., 1998. Structural characterization of a fulvic acid and a humic acid using solid-state Ramp-CP-MAS13C nuclear magnetic resonance. Environ. Sci. Technol. 32, 719–725. Dell’Abate, M.T., Benedetti, A., Trinchera, A., Dazzi, C., 2002. Humic substances along the profile of two Typic Haploxwerert. Geoderma 107, 281–296. Dria, K.J., Sachleben, J.R., Hatcher, P.G., 2002. Solid-state Carbon-13 nuclear mag netic resonance of humic acids at high magnetic field strengths. J. Environ. Qual. 31, 393–401. Ernst, R.R., Anderson, W.A., 1966. Applications of Fourier transform spectroscopy to magnetic resonance. Rev. Sci. Instrum. 37, 93–102. FAO-ISRIC, 1998. WRB—World Reference Base for soil resources, World Soil Resources Report no. 84. FAO, Rome. Fernández, J.M., Plaza, C., Hernández, D., Polo, A., 2007a. Carbon mineralization in an arid soil amended with thermally-dried and composted sewage sludges. Geo derma 137, 497–503. Fernández, J.M., Plaza, C., Hernández, D., Polo, A., 2007b. Organic matter in degraded agricultural soils amended with composted and thermally dried sew age sludges. Sci. Total Environ. 378, 75–80. Francioso, O., Montecchio, D., Gioacchini, P., Ciavatta, C., 2005. Thermal analysis (TG-DTA) and isotopic characterization (13C–15N) of humic acids from different origins. Appl. Geochem. 20, 537–544. Flaig, W., Beutelspacher, H., Rietz, E., 1975. Chemical composition and physical properties of humic substances. In: Gieseking, J.E. (Ed.), In: In Soil Components, vol. 1. Springer-Verlag, Berlin, pp. 119–126. Genevini, P.L., Adani, F., Veeken, A., Nierop, G.J., Scaglia, B., Dijkema, C., 2002. Qualit ative modifications of humic acid-like and core-humic acid-like during high-rate compost ing of pig faeces amended with wheat straw. Soil Sci. Plant Nutr. 48 (2), 143–150. Geras imowicz, W.V., Byler, D.M., 1985. Carbon-13 CPMAS NMR and FTIR spectro scopic studies of humic acids. Soil Sci. 139, 270–285. Gonet, S.S., Cieslewicz, J., 1998. Differential thermal analysis of sedimentary humic acids in the light of their origin. Environ. Int. 24 (5/6), 629–636. Hatcher, P.G., Breger, I.A., Maciel, G.E., Szeveren yi, N.M., 1985. Geochemistry of humin. In: Hayes, M.H.B., MacCarthy, P., Malcolm, R.L., Swift, R.S. (Eds.), In Humic Substances H. Chichester, Wiley, pp. 275–302. Hatcher, P.G., Dria, K.J., Kim, S., Frazier, S.W., 2001. Modern analytical studies of humic substances. Soil Sci. 166, 770–794. Hargreaves, J.C., Adl, M.S., Warman, P.R., 2008. A review of the use of composted municipal solid waste in agriculture. Agric. Ecosyst. Environ. 123, 1–14. Navas, A., Bermúdez, F., Machín, J., 1998. Influence of sewage sludge application on physic al and chemical properties of Gypsisols. Geoderma 87, 123–135. Peuravuori, J., Paaso, N., Pihlafa, K., 1999. Kinetic study of the thermal degradation of lake aquatic humic matter by thermogravimetric analysis. Thermochim. Acta 325, 181–193. Pietro, M., Paola, C., 2004. Thermal analysis for the evalua tion of the organic matter evolution during municipal solid waste aerobic composting process. Thermo chim. Acta 413, 209–214.
Provenzano, M.R., Senesi, N., Miikki, V., 1998. Characterization of composts and humic acids from pulp and paper mill biosludges by DSC in association with FT-IR spectroscopy. J. Therm. Anal. 52, 1037–1046. Reveille, V., Mansuy, L., Jarde, E., Garnier-Sillam, E., 2003. Characterization of sew age sludge-derived organic matter: lipids and humic acid. Org. Geochem. 34, 615–627. Ricca, G., Federica, L., Astori, C., Gallo, R., 1993. Structural investigations of humic acid from leonardite by spectroscopic methods and thermal analysis. Geoderma 57, 263–274. Ricca, G., Severini, F., Di Silvestro, G., Yuan, C.M., Adani, F., 2000. Derivatization and structural studies by spectroscopic methods of humic acids from Leonardide. Geoderma 98, 115–125. Senesi, N., 1989. Composted material as organic fertilizer. Sci. Total Environ. (81/82), 521–542. Senesi, N., Miano, T.M., Brunetti, G., 1996. Humic-like substances in organic amend ments and effects on native soil humic substances. In: Piccolo, A. (Ed.), In: Humic Substances in Terrestrial Ecosystems, vol. 1. Elsevier, Amsterdam, pp. 531–593. Senesi, N., Plaza, C., 2007. The role of humification in recycling of organic wastes of vario us nature and sources as soil amendments. Clean Soil Air Water 35, 26–41. Sheppard, J.D., Forgeron, D.W., 1987. Differential thermogravimetry of peat frac tions. Fuel 66, 232–236. Smernik, R.J., Oades, J.M., 2000. The use of spin counting for determining quantita tion of solid state13C-NMR spectra of natural organic matter. 2. HF-treated soil fractions. Geoderma 96, 159–171. Smernik, R.J., Oliver, I.W., Merrington, G., 2003. Characterization of sewage sludge organic matter using solid-state carbon-13 nuclear magnetic resonance spec troscopy. J. Environ. Qual 32, 1516–1522. Smernik, R.J., Oliver, I.W., Mc Laughlin, M.J., 2004. Changes in the nature of sewage sludge organic matter during a twenty-one-month incubation. J. Environ. Qual. 33, 1924–1929. Soil Survey Staff., 2003. Keys to soil taxonomy, 9th ed. Natural Resources Conserva tion Service USDA, Washington. Sort, X., Alcañiz, J.M., 1999. Modifi cation of soil porosity after application of sewage sludge. Soil Till. Res. 49, 337–345. Stevenson, F.J., 1994. Humus Chemistry; Genesis. Composition, In: Reactions, 2nd ed. Wiley-Interscience, New York. Tarrason, D., Ojeda, G., Ortíz, O., Alcañiz, J.M., 2008. Differences on nitrogen avail ability in a soil amended with fresh, composted and thermally-dried sewage sludge. Biores our. Technol. 99 (2), 252–259. Wilson, M.A., 1987. NMR Techniques and Applications in Geochemistry and Soil Chemistry. Pergamon Press, Oxford. Wong, J.W.C., Lai, K.M., Fang, M., Ma, K.K., 1998. Effect of sewage sludge amend ment on soil microbial activity and nutrient mineralization. Environ. Int. 24 (8), 935–943.