Degradation Kinetics and Transformation Products of Levonorgestrel

Subscriber access provided by EDINBURGH UNIVERSITY LIBRARY | @ http://www.lib.ed.ac.uk

Agricultural and Environmental Chemistry

Degradation kinetics and transformation products of levonorgestrel and quinestrol in soils Tao Tang, Chenyang Ji, Zhenlan Xu, Changpeng Zhang, Meirong Zhao, Xueping Zhao, and Qiang Wang J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.8b04788 • Publication Date (Web): 22 Mar 2019 Downloaded from http://pubs.acs.org on March 25, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33

Journal of Agricultural and Food Chemistry

Degradation kinetics and transformation products of levonorgestrel and quinestrol in soils

Tao Tang†, Chenyang Ji‡, Zhenlan Xu†, Changpeng Zhang†, Meirong Zhao‡, Xueping Zhao†, Qiang Wang†,* † State

Key Laboratory Breeding Base for Zhejiang Sustainable Pest and Disease Control, Key Laboratory

for Pesticide Residue Detection of Ministry of Agriculture, Laboratory (Hangzhou) for Risk Assessment of Agricultural Products of Ministry of Agriculture, Institute of Quality and Standard for Agro-products, Zhejiang Academy of Agricultural Sciences, Hangzhou, 310021, Zhejiang, China

‡

Key Laboratory of Microbial Technology for Industrial Pollution Control of Zhejiang Province, College

of Environment, Zhejiang University of Technology, Hangzhou, Zhejiang 310032, China

*Corresponding author: NO.298 Desheng Road, Hangzhou 310021, People’s Republic of China Telephone number: 86-571-86404355 (O), 86-571-86404355 (FAX) Email: [email protected] 1

ACS Paragon Plus Environment


1

ABSTRACT

2

Levonorgestrel (LNG) and quinestrol (QUN) are typical endocrine disruptors that enter the

3

soil via sewage irrigation and sludge return. However, the fates of both compounds in soil are

4

not well understood. Laboratory microcosm studies were conducted to fill the gap of

5

understanding LNG and QUN behavior in soils. High values of goodness-of-fit indices (GFIs)

6

were obtained using the double-first-order in parallel (DFOP) model and the single first-order

7

(SFO) model to fit the degradation kinetics of LNG and QUN in soils, respectively. The

8

end-points (DT50 and DT90) of LNG and QUN were positive correlated with soil TOC. Soil

9

water content and temperature were observed to be critical factors in degradation of LNG and

10

QUN. The degradation rates of LNG and QUN were very slow under sterile and flooded

11

conditions, indicating the aerobic microbial degradation was dominant in the degradation of

12

LNG and QUN. Moreover, major transformation products were identified and biodegradation

13

pathways of LNG and QUN were proposed. The present study is expected to provide basic

14

information for ecological risk assessment of LNG and QUN in soil compartment.

15 16

KEYWORDS

17

Levonorgestrel; Quinestrol; Degradation; Model; Transformation product; Soil

2


Page 2 of 33

Page 3 of 33

18


INTRODUCTION

19

Steroid hormones, including estrogens, progestagens and androgens, are among the

20

environmental pollutants most widely investigated during the past 20 years. These substances

21

are defined as endocrine-disrupting chemicals (EDCs) by the US Environmental Protection

22

Agency (EPA) since the mid-1990s, which is associated with their adverse effects on humans,

23

animals and other nontarget organisms (e.g., feminization, masculinization, cryptorchidism

24

and reproductive dysfunction).1-4 Exogenous synthetic steroid hormones have generally been

25

reported to exhibit the higher affinities for binding to hormone receptors than natural steroid

26

hormones and thus great disruption potencies.5,6 Levonorgestrel (LNG) and quinestrol (QUN),

27

a long-acting synthetic progestogen and a long-acting synthetic estrogen, are usually used as

28

oral steroids for human contraceptive and hormone replacement therapy.7-9 The worldwide

29

consumption of LNG and QUN are not known; however, a few attempts have been made to

30

estimate synthetic progestins and estrogens (including LNG and QUN) consumption in some

31

European countries. For example, the annual consumption of synthetic progestins in

32

Switzerland and the Czech Republic is estimated to be approximately 495 and 2400 kg,

33

respectively.10,11 In recent years, some studies have reported on the adverse effects of LNG

34

and QUN on organisms after they enter the environment.12-14 Therefore, it is essential to

35

investigate their environmental behavior and whether they are safe for organisms.

36

Like other pollutants, LNG and QUN are not efficiently removed by sewage treatment

37

plants and may be released into the environment.15,16 Soil residues of LNG and QUN mainly

38

result from the reuse of human and animal waste after treatment by sewage treatment plants,

39

including farmland irrigation with treated sewage and improvement in farmland fertility with

40

sludge. LNG has been found in sewage treatment plant effluents, surface waters and

41

groundwaters worldwide, including those in China.17 Its measured concentrations are

42

approximately 100 ng/L.18-21 LNG or hormone activity has also been measured to reach 3



43

hundreds of ng/kg in sediments and agricultural soils.22,23 QUN is the prodrug of

44

ethinylestradiol (EE2), and it is stored in body fat, subsequently released slowly and likely

45

converted into EE2.24 QUN is about three times as potent as EE2 due to its much longer

46

half-life.25 The sulfate and glucuronate conjugates of QUN and EE2 are formed in the kidney

47

and discharged into the environment through urine, and they can be disassembled by

48

arylsulfatase enzymes in the environment and converted back into the free compounds.26

49

Furthermore, the combination of LNG and QUN has been demonstrated to have a new

50

application as an efficient reagent for controlling rodents in grassland and plateau. 27

51

After entering soil, steroid hormones mainly undergo adsorption/desorption, leaching,

52

biodegradation or biotransformation. The degradation, migration, transformation and

53

photodegradation of natural estrogens, such as estrone (E1), 17α-estradiol (αE2),

54

17β-estradiol (βE2), and estriol (E3), and the synthetic estrogen EE2 in soil have been

55

investigated extensively.28-31 However, only a few studies about the adsorption or degradation

56

of LNG and QUN in soils were available when this work began.32,33 In view of the low-dose

57

effect of LNG and QUN as endocrine disruptors, knowledge of their degradation behavior is

58

urgently required because information on the biodegradation and biotransformation of LNG

59

and QUN is significant for accurate evaluation of their environmental risks.

60

To obtain this information, three models recommended by FOrum for the Co-ordination of

61

pesticide fate models and their USe (FOCUS),34 namely, single first-order (SFO), first-order

62

multicompartment (FOMC) and double-first-order in parallel (DFOP), were applied to fit the

63

observed degradation dynamics and to derive degradation end-points. Microorganisms play a

64

dominant role in pollutant degradation, and their activity in soil may be influenced by the

65

presence of water, temperature, organic carbon content and other factors.35 Therefore, the

66

degradation behavior of LNG and QUN under different conditions (different soil types,

67

temperatures, and water-soil ratios and sterilization treatment) was determined in five 4


Page 4 of 33

Page 5 of 33


68

contrasting soils from different areas of China. Additionally, gas chromatography-mass

69

spectrometry (GC-MS) was used to identify the transformation products of LNG and QUN in

70

soil, and the possible transformation pathways were tentatively postulated.

71

MATERIALS AND METHODS

72

Chemicals and Reagents. LNG and QUN (purity>99%) were obtained from

73

Sigma-Aldrich (Dorset, U.K.). Ammonium acetate, anhydrous calcium chloride, and acetic

74

acid were of analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd.

75

(Beijing, China). Methanol and acetonitrile were high-performance liquid chromatography

76

(HPLC)-grade chemicals from Fisher Scientific (Fair Lawn, NJ). Ultrapure water was

77

obtained in the laboratory using a Milli-Q water purification system (Millipore, Billerica, MA)

78

and had a resistivity greater than 18.2 MΩ cm for the preparation of samples and mobile

79

phases. All chemicals were used as received.

80

Soil Samples. The tested soil samples included 4 farmland soils and 1 natural grassland

81

soil. The farmland soils were taken from Heilongjiang Province (HLJ), Beijing (BJ), Yunnan

82

Province (YN), and Guangxi Province (GX), China; the grassland soil was taken from Inner

83

Mongolia (NMG), China. Each soil sample was taken from the surface layer (0 to 20 cm).

84

After the soil samples were collected, they were spread evenly in a clean laboratory; the plant

85

residues, stones and other debris were removed; and the samples were air-dried and passed

86

through a 2 mm sieve. The treated soil samples were placed in plastic bags and stored in a

87

refrigerator at 4 °C for future use. Soil sterilization was accomplished by autoclaving soils at

88

120 °C under 300 kPa 3 times for 45 min at 3 d intervals. The properties of these soils have

89

determined in a previous study of ours.32

90

Soil Degradation. Degradation of LNG and QUN were determined using 140 g of soil

91

in a sterilized 250 mL beaker, and then distilled water was added to set the water-soil ratio to

92

1/5 (v/m). Subsequently, the beaker was sealed by permeable film and placed in a dark 5



93

incubator at 25 °C for 14 d of preincubation. In a small beaker, another dried soil (10 g) was

94

treated with 5 mL of methanol solution containing 60 μg/mL LNG or QUN. The soil was

95

stirred evenly with a glass rod and placed in a fume hood overnight. After the methanol was

96

completely evaporated, the sample was mixed thoroughly with the preincubated soil on a

97

shaker, and the initial concentrations of LNG or QUN were 2.0 μg/g. After water was added

98

up to the required water-soil ratio, The flasks were weighed periodically to check for water

99

loss, and deionized water was added to compensate for the water loss when necessary. At

100

different time intervals, certain amount of soil (equivalent of 10 g dry soil calculated on the

101

basis of soil water content) of each treatment was collected and transferred into a freezer

102

(-21 °C) to stop degradation.

103

Two soils (BJ and NMG soils) with significantly different parameters were used to assess

104

the appropriate degradation model of LNG and QUN. The degradation experiment was

105

performed in five different soils to examine how the degradation rates of LNG and QUN vary

106

with soil properties. The effects of temperature on the degradation of LNG and QUN were

107

determined by incubating them at 15 ± 2, 25 ± 2, 37 ± 2, and at 54 ± 2 °C with BJ soil. The

108

effects of the water-soil ratio on the degradation of LNG and QUN were determined by setting

109

the ratio at 1/10, 1/5, and 1/1 (flooded) with BJ soil. Additionally, abiotic degradation was

110

directly conducted by spiking a methanol solution of LNG or QUN into autoclaved BJ soil.

111

To monitor the transformation products (TPs), LNG and QUN were added separately to BJ

112

soil at an initial concentration of 20 μg/g. Sterile controls and matrix controls were also

113

performed. Flasks were sealed and incubated at 25 °C. After 7 d of incubation, 10 g of sample

114

was collected for measurement. All experiments were performed in triplicate.

115

Sample Extraction. A modified quick, easy, cheap, effective, rugged, and safe

116

(QuEchERS) method was used. After 10 g of soil sample (dry weight) was added to a 50 mL

117

screw-cap centrifuge tube, 5 mL of distilled water and 15 mL of acetonitrile (containing 1% 6


Page 6 of 33

Page 7 of 33


118

acetic acid) were added, and the solution was vortexed for 1 min and then centrifuged at 4000

119

rpm (1650 g) for 5 min. The extract was transferred to a 100 mL stoppered measuring

120

cylinder containing 1 g of sodium acetate and 4 g of anhydrous magnesium sulfate.

121

Subsequently, 15 mL of acetonitrile (containing 1% acetic acid) was added to the centrifuge

122

tube to repeat the extraction. The extracts were combined in a stoppered measuring cylinder,

123

which was manually shaken for 1 min and allowed to stand for 5 min. Then, 15 mL of the

124

upper extract was concentrated to near dryness under reduced pressure in a vacuum rotary

125

evaporator at 45 °C and dried under a stream of high-purity nitrogen gas. The residue was

126

reconstituted with 1 mL of HPLC-grade methanol and filtered using a 0.22 μm membrane

127

filter, followed by sample injection and analysis. The transformation product soil samples

128

were extracted the same way as described in the previous method. Preliminary experiments

129

showed that the recoveries of LNG and QUN from soils were higher than 85%, with relative

130

standard deviations of 4.8-7.6%, and the limits of quantification (LOQs) were lower than 9.6

131

μg/kg for the two compounds.

132

Analysis. The quantification of LNG and QUN were carried out on a Shimadzu HPLC

133

system consisting of two LC-20AT pumps, an SPD-M20A ultraviolet-visible (UV-vis)

134

detector, a CBM-20A system controller, a DGU-20A3 online degasser and an LC solution

135

workstation. Separation was performed on a Kromasil ODS C18 column (250 mm × 4.6 mm)

136

packed with 5 μm particles and connected to a guard cartridge (Kromasil Easy Guard

137

C18-6201). Isocratic elution was used, the mobile phase was acetonitrile-0.1% formic acid

138

(85/15, v/v) with a flow rate of 1 mL/min, the injection volume was 20 μL, and the detection

139

wavelength was 244 nm.

140

An Agilent 7890A GC equipped with an MSD 5975C MS detector and a7693 automatic

141

liquid sampler was used to identify potential transformation products. The instrument was

142

equipped with an HP-5MS capillary column (30 m × 250 μm, 0.25 μm film thickness). The 7



Page 8 of 33

143

oven temperature program was as follows: 50 °C for 2 min, followed by heating at 10 °C/min

144

to 300 °C, and then this temperature was held for 10 min. The carrier gas was high-purity

145

helium (99.999%), and the flow rate was 1.2 mL/min. The inlet temperature was 250 °C, and

146

the injection volume was 2 μL (splitless injection). The mass spectrometry operating

147

conditions were as follows: ionization mode EI, ionization energy 70 eV, ion source

148

temperature 230 °C, quadrupole temperature 150 °C, and transfer line temperature 290 °C.

149

The selected ion monitoring mode was used with a solvent delay of 5 min and a mass scan

150

range of 10 to 500 m/z.

151

Model Fitting and Data Analysis. Experimental data processing was conducted using

152

the method recommended by the FOCUS of the European Union to determine the optimal

153

degradation model.34 The amount of LNG and QUN not degraded on day 0 was set to 100%,

154

and the subsequent amount degraded was expressed as a percentage of the day 0 amount.

155

Statistical analysis and plotting of the experimental data were performed using SPSS 13.0,

156

Sigmaplot 11.0, and GraphPad Prism 5.04.

157

LNG and QUN degradation data were fitted with the SFO model, FOMC model and

158

DFOP model. The SFO (Eqs. 1 and 2), FOMC (Eqs. 3 and 4) and DFOP (Eq. 5) models can

159

be expressed using the following equations:

160

C = C0e-kt

(1)

161

DTx = ln (100/100-x)/k

(2)

162

C = C0 / (t/β + 1)α

(3)

163

DTx = β {[100 / (100-x)]1/α -1}

(4)

164

C = C1e-k1t + C2e-k2t

(5)

165

where C is the concentration percentage of compound remaining in the soil at a given time, C0

166

is the initial concentration of compound, k is the degradation rate of the compound in the soil,

167

x is the degradation percentage of the compound, α is the coefficient of variation for the 8


Page 9 of 33


168

degradation rate of the FOMC equation, and β is the location parameter. C1 is the initial

169

concentration percentage of the degraded compound that is not adsorbed or trapped by the soil,

170

and C2 is the initial concentration percentage of the degraded compound that is adsorbed or

171

trapped by the soil. The variables k1 and k2 are the degradation rates of the compounds in the

172

unabsorbed and the adsorbed phases, respectively.

173

SPSS 13.0 was used to conduct iterative calculations of the degradation data and

174

estimate the endpoint of degradation, and data describing the LNG and QUN degradation in

175

the soil were fit using GraphPad Prism software and the Levenberg-Marquardt method, which

176

is based on a nonlinear least squares algorithm. Additionally, the adjusted coefficient of

177

determination (R2adj) of each model and the associated model error value (model err) were

178

compared, and the goodness-of-fit indices (GFIs) describing how well each model fit the

179

LNG and QUN degradation in soil were evaluated. Each parameter can be calculated using

180

the following equations: R 2 adj = 1 − (1 − R 2 )

181

n −1 n − m −1

 (O − C ) = 1−  (O − O )

(6)

2

R2

182

i

i

i

a

i

(7)

2

i

err =

183

1

 2df ,

(Ci − Oi )2 i O 2 a

(8)

184

where C = model predicted value; O = measured concentration of LNG or QUN in the soil;

185

Οa = average of measured concentrations, df = degrees of freedom; α = 0.05; n = number of

186

samples measured; and m = number of parameters in the model. The calculated chi-square

187

values (χ2) of all models were smaller than χ2df, α; all three models passed validation, and thus,

188

the results are not listed in a table (SFO, χ27, 0.05 = 14.067; FOMC, χ26, 0.05 = 12.592; DFOP, χ25,

189

0.05

= 11.070). 9



190

RESULTS AND DISCUSSION

191

Degradation Model. SFO kinetics are described by a simple exponential equation with

192

only two parameters and assume that the number of molecules of the compound is small

193

relative to the number of degrading microorganisms. The FOMC model describes the soil as a

194

heterogeneous medium and divides the soil into a large number of subcompartments, each

195

with a different degradation rate constant.36 The DFOP model assumes that the compound is

196

initially distributed between solution and sorbed phases and that the degradation of the

197

compound in both phases is described by first-order kinetics37. The SFO, FOMC and DFOP

198

models were selected to fit the measured LNG and QUN degradation data due to their

199

simplicity and previous successful application in describing the degradation of organic

200

compounds. 37,38 Furthermore, the SFO and DFOP models have been used by the European

201

Food Safety Authority (EFSA) and the United States EPA in evaluating degradation kinetics

202

of compounds in soil. 39,40

203

The selection of the optimal model for LNG and QUN degradation in soil was based on

204

GFIs and degradation end-points (DT50 and DT90). The parameters of models fit to the

205

degradation profiles of LNG and QUN in BJ and NMG soils are presented in Table 1, and the

206

resulting curves and residual values are shown in Fig. S1 and Fig. S2. For LNG, although the

207

SFO and FOMC model fits resulted in adjusted R2 values ranging from 0.9679 to 0.9774, the

208

associated error percent was higher than 4.27. In contrast, the DFOP model fit the data with

209

an adjusted R2＞0.9879, and the error percent was only 1.20 in NMG soil. Furthermore, the

210

results also demonstrated that the DFOP model was more accurate in fitting the degradation

211

end-points than the SFO and FOMC models by comparing predicted and experimental values.

212

This result is in agreement with a study in which SFO and DFOP models were used to predict

213

the degradation kinetics of estrone-3-sulfate and estrone in agricultural soils; this study

214

concluded that the DFOP model resulted in the most accurate predictions.41 Overall the DFOP 10


Page 10 of 33

Page 11 of 33


215

model was found to be superior to the SFO and FOMC models for describing the LNG

216

degradation data in our study. For QUN, the SFO, FOMC and DFOP models fit the

217

degradation data well, with adjusted R2 values ranging from 0.9874 to 0.9949, and the

218

associated error percents were all lower than 3.05. Additionally, the SFO and FOMC models

219

give similar calculated end-point values that ranged, for instance, from 9.44 to 9.45 d for DT50

220

and 31.37 to 31.40 d for DT90 in BJ soil. These results were closer to experimental values than

221

results from the DFOP model. In the case of the DFOP model, the predicted end-point values

222

(12.10 d for DT50 and 25.36 d for DT90 in BJ soil) were slightly lower than the measured

223

values. When several models all fit degradation dynamics well, the model with the fewest

224

parameters is generally preferred because it reduces the complexity of data processing. 37

225

Therefore, the SFO model was selected to fit the degradation data of QUN in the subsequent

226

experiments.

227

Degradation in Different Soils. Several studies have shown that soil properties, especially

228

the total organic carbon (TOC), cation-exchange capacity (CEC), nitrogen (N) and clay

229

content, play a significant role in degradation.42,43 In this work, correlations between

230

end-points (DT50 and DT90) and soil properties were examined to evaluate the degradation

231

rates of LNG and QUN in different soils and to elucidate the factors principally responsible

232

for degradation rates. The DFOP model accurately described the degradation dynamics of

233

LNG in the five contrasting soils. The degradation rate of LNG was relatively fast in an initial

234

stage and then slowed (Fig. 1). DFOP model-derived DT50 values ranged from 6.77 d to 15.90

235

d, and the times required for 90% degradation ranged from 19.21 d to 32.32 d. LNG exhibited

236

its fastest and slowest degradation rates in BJ and YN soil, respectively. The DT50 and DT90

237

values were not directly related to soil CEC, clay and N, which was determined by their

238

correlation coefficients, R2, being lower than 0.306. Although the soil TOC produced the

239

highest R2 correlation coefficients of the soil parameters, the R2 correlation coefficients were 11



240

only 0.4756 and 0.4809 for DT50 and DT90, respectively (Table 2). This phenomenon might be

241

explained by our previous study that showed that strong adsorption of LNG (Koc＞946.49

242

L/kg) can decrease the degradation rate by preventing the availability of adsorbate to

243

microorganisms,32 and this phenomenon has also been confirmed by the other studies. 44

244

The SFO model accurately described the degradation dynamics of QUN in the test soils. In

245

contrast to LNG, QUN maintained an almost constant rate throughout its degradation in soil

246

(Fig. 2). SFO model-derived DT50 values ranged from 9.44 d to 14.72 d, and the time required

247

for 90% degradation ranged from 24.27 d to 32.32 d. The minimum DT50 value for

248

degradation of QUN was found for BJ soil and was slightly different from that for HLJ soil;

249

similarly, the shortest time required to degrade 90% of the QUN was found for HLJ soil. The

250

simple correlations between end points (DT50 and DT90) and soil properties were calculated,

251

and the results showed that because their R2 correlation coefficients were lower than 0.2740,

252

soil N and clay content had no significant effects on QUN degradation. However, the DT50

253

and DT90 values of QUN were strongly positive correlated with soil TOC, with R2 coefficients

254

ranging from 0.8074 to 0.9867, and CEC exhibited an R2 correlation coefficient that ranged

255

from 0.5971 to 0.9041 (Table 2). It should be noted that CEC strongly correlates with TOC.

256

The reason that degradation is correlated with CEC is just because organic matter has a high

257

CEC and that degradation primarily occurs in the organic matter domain. These results could

258

explain why the HLJ soil had the shortest degradation time among the five soils. Generally,

259

fast degradation occurs with high-organic carbon soils because of the high nutrient levels and

260

large microbial populations.42 The results of this study were consistent with previous research

261

that proved that the degradation rates of pharmaceuticals and personal care products increased

262

as the organic carbon content of soils increased. 45

263

Effect of Temperature on Degradation. During a typical growing season in China, soil

264

temperature can fluctuate between 10 and 50 °C depending on agricultural measures (e.g., 12


Page 12 of 33

Page 13 of 33


265

application of plastic film on soil) and soil depth. Accordingly, temperatures of 15, 25, 37 and

266

54 °C were selected for investigating the effect of temperature on LNG and QUN degradation

267

in soil. The degradation dynamics of LNG and QUN and their residues in BJ soil at different

268

temperatures after 28 d of incubation are shown in Fig. 3. At each tested temperature, the

269

DFOP model and the SFO model fit the degradation dynamics of LNG and QUN well,

270

respectively. After 28 d of incubation, the lowest residual LNG and QUN were found at 25 °C,

271

with values of 8.5% and 13.5%, followed by those at 37 °C, with values of 13.0% and 14.5%.

272

The slowest degradation of LNG and QUN was observed at 54 °C, where the degradation rate

273

constants were 35-70% of those at 37 °C (Table S1). In addition, variations in degradation

274

rate constants at different temperatures were high for LNG compared to QUN, which implies

275

that the degradation of LNG in soil was more sensitive to temperature than the degradation of

276

QUN. Temperature and degradation of chemicals in soil commonly share a direct relationship

277

to a certain degree as a result of increasing microbial activity with rising temperature.46

278

Several studies also observed rapid degradation of organic pollutants at temperatures from 10

279

to 40 °C.30,47 That increasing the temperature from 15 °C past 25 °C to 37 °C accelerated the

280

degradation of LNG and QUN in soil might indicate that biotic processes are dominant in the

281

degradation of LNG and QUN, since the soil microbial activity could be promoted by

282

elevated temperatures. However, when the temperature was increased further to 54 °C, the

283

degradation of LNG and QUN was inhibited, which might be related to the depressed

284

microbial activity at such an extreme temperature.48

285

Effect of Moisture on Degradation. The effect of moisture on the degradation of LNG

286

and QUN was investigated by using water-soil ratios of 1:10, 1:5 and 1:1, wherein the

287

water-soil ratio of 1:1 represented the flooded condition. The corresponding DT50 and DT90

288

values are listed in Table 3. The fastest degradation of LNG and QUN was found at the

289

water-soil ratio of 1:5, with DT50 values of 6.77 d and 9.44 d, which was followed in speed by 13



290

degradation at the water-soil ratio of 1:10. Degradation under the flooded condition

291

(water-soil ratio of 1:1) were the slowest, with DT50 values of 22.43 d and 23.74 d. This result

292

suggested that increasing soil water content enhanced the degradation rate for both LNG and

293

QUN by 30-57% when soil was not flooded, which is consistent with studies of Wang et al.49

294

and might be attributed to the increased dissolution of LNG and QUN at high soil water

295

content, which would enhance their bioavailability to soil microorganisms. Meanwhile,

296

adequate moisture content is necessary for the development and growth of microbes, favoring

297

the degradation of LNG and QUN. When soil is flooded (water-soil ratio of 1:1), representing

298

a typical anaerobic environment in agricultural production, the degradation of LNG and QUN

299

was extensively depressed, with DT50 values being 1.5-2.3 times longer as those at the

300

water-soil ratio of 1:5. This result indicated that water-soil ratio was a critical factor in the

301

degradation of LNG and QUN and that LNG and QUN were degraded rapidly under aerobic

302

conditions, which is in agreement with a study by Ying et al., who explored the degradation of

303

E1, E2, E3 and EE2 in soil and observed that the degradation rates of these steroid hormones

304

are faster under aerobic conditions than under flooded conditions.44

305

Degradation in Sterile and in Nonsterile Soil. The degradation dynamics of LNG and

306

QUN in sterile and nonsterile soils and the amounts of residual compounds after 28 d of

307

incubation are shown in Fig. 4. The degradation rates of LNG and QUN in sterile and

308

nonsterile soils were considerably different. After 28 d of incubation, the amounts of residual

309

LNG in sterile and nonsterile soils were 89.5% and 8.5%, respectively. The amounts of

310

residual QUN in sterile and nonsterile soils were 81.5% and 13.5%, respectively, which was

311

consistent with the work of Zhang et al. They reported a degradation half-life of

312

approximately 16 d, with 41.2% removal in nonsterile soil compared to 4.8% removal in

313

sterile soil after an incubation of 10 d.33 LNG and QUN loss was significantly slower in

314

sterile soil than that of nonsterile soil

(p

Degradation Kinetics and Transformation Products of Levonorgestrel

Recommend Documents